The Epistemology of Democracy [1 ed.] 1032317256, 9781032317250

Table of contents :
Cover
Half Title
Title Page
Copyright Page
Contents
List of Contributors
Introduction: What the Epistemology of Democracy Is All About
PART I: Democratic Pessimism
1. Sexy but Wrong: Diversity Theorem Defenses of Democracy
2. A Belated Failure: Condorcet in Contemporary Epistemic Conditions
3. Social Epistemic Miserliness: Populism against Democracy
4. Critical Thinking and Trusting Experts in Real-life Democracies
5. The Dangers of Disinformation
PART II: Democratic Optimism
6. The Politics of Resentment: Hope, Mistrust, and Polarization
7. Against the Individual Virtue Approach in the Epistemology of Democracy
8. Institutional Cynicism and Civic Virtue
9. Myside Bias in Individuals and Institutions
10. Listening for Epistemic Community
PART III: Democratic Realism
11. Sensemaking, Empathy, and Democracy
12. Political Skepticism, Bias, and Epistemic Colonization
13. Economic Inequalities and Epistemic Democracy
14. What Political Enemies Are for
15. Top-Down and Bottom-Up Solutions to the Problem of Political Ignorance
Index

Citation preview

THE EPISTEMOLOGY OF DEMOCRACY Edited by Hana Samaržija and Quassim Cassam

The Epistemology of Democracy

This is the first edited scholarly collection devoted solely to the epistemology of democracy. Its fifteen chapters, published here for the first time and written by an international team of leading researchers, will interest scholars and advanced students working in democratic theory, the harrowing crisis of democracy, political philosophy, social epistemology, and political epistemology. The volume is structured into three parts, each offering five chapters. The first part, Democratic Pessimism, covers the crisis of democracy, the rise of authoritarianism, public epistemic vices, misinformation and disinformation, civic ignorance, and the lacking quantitative case for democratic decision-making. The second part, Democratic Optimism, discusses the role of hope and positive emotions in rebuilding ­democracy, proposes solutions to myside bias, and criticizes dominant epistocratic approaches to forming political administrations. The third and final part, Democratic Realism, assesses whether we genuinely require emotional empathy to understand the perspectives of our political adversaries, discusses the democratic tension between mutual respect for others and a quest for social justice, and evaluates manifold top-down and ­bottom-up approaches to policy making. Hana Samaržija is a Ph.D. student in Philosophy at the University of ­Warwick. Her papers on countering epistemic injustice and seeking epistemically high-quality alternatives to democracy have been published in Social Epistemology and other academic journals as well as in the edited book The Philosophy of Fanaticism: Epistemic, Affective, and Political Dimensions (Routledge, 2022). Quassim Cassam is a Professor of Philosophy at the University of ­Warwick, an Honorary Fellow of Keble College, Oxford, and a Fellow of the British Academy.

The Epistemology of Democracy

Edited by Hana Samaržija and Quassim Cassam

First published 2023 by Routledge 605 Third Avenue, New York, NY 10158 and by Routledge 4 Park Square, Milton Park, Abingdon, Oxon, OX14 4RN Routledge is an imprint of the Taylor & Francis Group, an informa business © 2023 selection and editorial matter, Hana Samaržija and Quassim Cassam; individual chapters, the contributors. The right of Hana Samaržija 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. ISBN: 978-1-032-31725-0 (hbk) ISBN: 978-1-032-31726-7 (pbk) ISBN: 978-1-003-31100-3 (ebk) DOI: 10.4324/9781003311003 Typeset in Times New Roman by KnowledgeWorks Global Ltd.

Contents

List of Contributorsvii Introduction: What the Epistemology of Democracy Is All About

1

HANA SAMARŽIJA

PART I

Democratic Pessimism

15

1 Sexy but Wrong: Diversity Theorem Defenses of Democracy

17

JASON BRENNAN

2 A Belated Failure: Condorcet in Contemporary Epistemic Conditions

32

HANA SAMARŽIJA

3 Social Epistemic Miserliness: Populism against Democracy

51

NENAD MIŠČEVIĆ

4 Critical Thinking and Trusting Experts in Real-life Democracies

70

SNJEŽANA PRIJIĆ SAMARŽIJA

5 The Dangers of Disinformation

90

ÅSA WIKFORSS

PART II

Democratic Optimism

113

6 The Politics of Resentment: Hope, Mistrust, and Polarization

115

ALESSANDRA TANESINI

vi  Contents 7 Against the Individual Virtue Approach in the Epistemology of Democracy

135

MARKO LUKA ZUBČIĆ

8 Institutional Cynicism and Civic Virtue

152

IAN JAMES KIDD

9 Myside Bias in Individuals and Institutions

170

KEITH E. STANOVICH

10 Listening for Epistemic Community

195

HANNA KIRI GUNN

PART III

Democratic Realism

213

11 Sensemaking, Empathy, and Democracy

215

QUASSIM CASSAM

12 Political Skepticism, Bias, and Epistemic Colonization

233

MICHAEL P. LYNCH

13 Economic Inequalities and Epistemic Democracy

250

IVAN CEROVAC

14 What Political Enemies Are for

270

ROBERT B. TALISSE

15 Top-Down and Bottom-Up Solutions to the Problem of Political Ignorance

287

ILYA SOMIN

Index

316

List of Contributors

Jason Brennan is the Flanagan Family Professor at the McDonough School of Business, Georgetown University. He is the author of sixteen books, including Crack in the Ivory Tower: The Moral Mess of Higher Education (Oxford University Press, 2019). Quassim Cassam is a Professor of Philosophy at the University of Warwick, UK, an Honorary Fellow of Keble College, Oxford, and a Fellow of the British Academy.  Ivan Cerovac is an External Research Fellow at the University of Rijeka. He is the author of Epistemic Democracy and Political Legitimacy (Palgrave Macmillan, 2020) and John Stuart Mill and Epistemic Democracy (Rowman and Littlefield, 2022), and he writes and teaches on a range of topics in ethics and political philosophy, including political legitimacy, social justice, and democratic theory. Hanna Kiri Gunn is an Assistant Professor of Philosophy in the Cognitive and Information Sciences Department at the University of California, Merced. Her main area of research concerns the applied ethics and epistemology of epistemic communities and epistemic agency. Ian James Kidd teaches and researches philosophy at the University of Nottingham. His research interests include social and applied epistemology and moral and epistemic virtues and vices. Michael Patrick Lynch is a Board of Trustees Distinguished Professor of Philosophy at the University of Connecticut and the author of KnowIt-All Society and The Internet of Us, among other books. Nenad Miščević is a Full Professor of Philosophy at the University of Maribor and has, until recently, also been teaching at the Central European University in Budapest. He has worked on various philosophical subjects, emphasizing epistemology, philosophy of language, and political philosophy.

viii  List of Contributors Hana Samaržija is a Ph.D. student of Philosophy at the University of Warwick. Routledge and Social Epistemology have published her works on social epistemology, political epistemology, political polarization, and epistemic injustice. Snježana Prijić Samaržija is a Professor of Epistemology at the Faculty of Humanities and Social Sciences at the University of Rijeka. The main areas of her scientific interest are social philosophy, epistemology, and applied ethics. During her two rector’s mandates, she published her last book, Democracy and Truth: The Conflict Between Political and Epistemic Virtues (Mimesis International, 2018). Ilya Somin is a Professor of Law at George Mason University. He is the author of six books, including Free to Move: Foot Voting, Migration, and Political Freedom (Oxford University Press, rev. ed. 2021), Democracy and Political Ignorance: Why Smaller Government Is Smarter (Stanford University Press, 2nd ed. 2016), and The Grasping Hand: Kelo v. City of New London and the Limits of Eminent Domain (University of Chicago Press, rev. ed. 2016). Keith E. Stanovich is a Professor Emeritus of Applied Psychology at the University of Toronto. He is the author of The Bias That Divides Us (MIT Press, 2021) and has received the Thorndike Career Achievement Award from the American Psychological Association. Robert B. Talisse is W. Alton Jones Philosophy and Political Science Professor. at Vanderbilt University in Nashville, Tennessee. He specializes in democratic theory, emphasizing justice, citizenship, public deliberation, and political disagreement. Alessandra Tanesini is a Professor of Philosophy at Cardiff University. Her latest book is The Mismeasure of the Self: A Study in Vice Epistemology (Oxford University Press, 2021). Åsa Wikforss is a Professor of Theoretical Philosophy at Stockholm University and researches the intersection of epistemology, philosophy of mind, and political psychology. She is the author of two popular books and a member of the Swedish Royal Academy of Science and the Swedish Academy. Marko Luka Zubčić is a Postdoctoral Researcher at the Center for Advanced Studies South East Europe at the University of Rijeka and an associate lecturer at the Faculty for Humanities and Social Sciences in Rijeka. His research field is institutional epistemology, focusing on the economic and institutional design conditions of epistemic reliability in complex social systems.

Introduction What the Epistemology of Democracy Is All About Hana Samaržija

0.1 Introduction This volume is the first international publication devoted exclusively to the novel field of the epistemology of democracy. The contemporary significance of the epistemology of democracy is difficult to overstate. Democracy is at a crossroads, with recent volumes spelling out its procedural failures and the frequency of its collapses (Brennan 2016; Levitsky and Ziblatt 2018). Ever since venerated political philosophers first penned their classic theses, the wisdom of crowds appeared as the best way to resolve the problem of political governance (Mill 1982; Locke 1988; Rousseau 2003). A diverse group of informed citizens cognizant of their interests, seeking the common good while guarded by the division of powers and canceling each other’s biases, was guaranteed to arrive at the best possible conclusion about their joint problems (Goodin and Spiekermann 2018). However, these presuppositions have recently received a more critical reading, albeit long taken for granted. Profoundly disquieting decisions made by multi-million electorates cast doubt on the hitherto unchallenged and seemingly perennial belief in the wisdom of crowds (Cassam 2019). Deliberation between political parties essential to representative democracy, allegedly the most procedurally just political system we are acquainted with, repeatedly fails to deliver what many would regard as sensible policies (Ahlstrom-Vij 2012). In their contemporary epistemic conditions, liberal representative democracies depend primarily on their voters, who are often accused of being ignorant, ideologically biased, uninformed, and prejudiced (Delli Carpini and Keeter 1996; Delli Carpini 2005; Fricker 2006; Caplan 2007; Ahlstrom-Vij 2018). If those concerns are correct, irresponsible citizens cannot select responsive and epistemically responsible governments (Ahlstrom-Vij 2020). Accordingly, numerous elected politicians seem unconcerned by expert scientific warnings about oncoming recessions, the climate crisis potentially injurious to our life on Earth, and stealthy pandemics that have cost us millions of lives. It came as little wonder political epistemologists instinctively focused on probing democracy for its professed commitment to electoral equality and political DOI: 10.4324/9781003311003-1

2  Hana Samaržija inclusion. For the last fifteen years, scholars have been divided on the question of whether the solution to democracy’s struggles is fewer democratic processes, such as restricting the electorate to more knowledgeable citizens (Brennan 2020), or more inclusive democratic processes open to referenda and direct democracy (Landemore 2012, 2020). Others have argued that the resolution lies in educating the electorate by appealing to their agential interests or introducing monetary incentives for learning about cardinal democratic facts and procedures (Somin 2023). Methodologically, the epistemology of democracy is consciously interdisciplinary and eclectic (Goldman 2003). Knowing it has much to learn, it interacts with political studies and their more specialized branches, such as political behavior research and political psychology (Dalton and Klingemann 2009). It also borrows insights from the cognitive sciences, psychology, sociology, anthropology, probability theories, and mathematics. Consequently, this volume comprises chapters devoted to studying cognitive biases inherited from the cognitive sciences (Stanovich 2021a), assessments of civic knowledge based on painstaking reviews of decades of political behavior studies (Kuklinsky and Peyton 2007), portrayals of empathy that conjoin psychology and sociology with materialist philosophy (Marx 2000), and discussions of differential economic capital as an obstacle to democratic equality heavily involved with economics and welfare studies. Nonetheless, despite its refreshing interaction with other sciences, the epistemology of democracy remains profoundly philosophical in its fundamental methods, aims, language, and argumentation. Furthermore, not all epistemologists of democracy are equally committed to preserving or sustaining democracy (Talisse 2021). On the contrary, while some defend democracy for the supposedly beneficial epistemic effects of diversity in resolving complex problems and its fair maintenance of ideological pluralism, others repudiate the supposed wisdom of crowds, underline the significance of deference to experts (Ahlstrom-Vij 2013; Prijić-Samaržija 2017), and seek epistemically better justified alternatives to democracy that preserve its pledge to civic inclusion and equality. Some epistemologists of democracy remain impartial toward its objective epistemic value and instead inquire about how to understand people whose views we deem reprehensible properly and how economic inequality impacts ostensibly fair democratic processes. Even in its earliest steps, the epistemology of democracy tackles issues that concern us all. And that is why it merits our effort and attentiveness.

0.2 The Advent of the Epistemology of Democracy: A Very Short Story Traditional analytic epistemology has undergone several significant expansions throughout its lengthy and turbulent history. Initially, it

Introduction 3 presupposed a single, genderless, and disembodied epistemic agent who resided in something akin to a social vacuum. In such an asocial and apolitical environment, there was nothing to obstruct the acquisition of essential sensory and inferential knowledge (Prijić-Samaržija 2018, 3). The sole goals of early analytic epistemology’s abstruse epistemic agent were truth and knowledge. However, it was not long before numerous more contemporary epistemologists found the idea of truth and knowledge acquired by an asocial and disembodied epistemic agent excessively limiting. Traditional analytic epistemology’s first expansion into virtue epistemology replaced this truth monism with a plural account of epistemic values embodied in an agent’s epistemic or intellectual virtues (Battaly 2008). We could best describe epistemic virtues as dispositions, character traits, and modes of thinking that assist us in making proper assessments and acquiring, maintaining, and disseminating knowledge. Although Ernest Sosa drew the first sketch of a pluralistic account of the faculties capable of delivering true beliefs as the solution to the impasse between foundationalism and coherentism in his canonical article “The Raft and the Pyramid” (Sosa 1980), Linda Zagzebski finally set a plural account of epistemic values on solid ground in her now classic book Virtues of the Mind (Zagzebski 1996). As in every novel investigative field, virtue epistemologists immediately separated into two divergent camps. Virtue reliabilists argued that intellectual virtues amount to dependable innate gifts such as acute perception, a razor-sharp intelligence, and a staunch memory (Sosa 2001). In contrast, virtue responsibilists held that epistemic virtues are normative character traits we can dynamically work to acquire and cultivate. Both accounts of epistemic virtues comprised characteristics such as responsibility and caution with evidence, respect for epistemic authorities (Zagzebski 2012), conscientiousness with our work, earnest interest in learning, charity toward opposing viewpoints, epistemic justice toward marginalized knowers, and intellectual humility in the face of our inevitably limited knowledge (Hazlett 2012). Perceiving epistemic virtues as normative faculties capable of cultivation rather than innate genetic predicaments later became the prevailing outlook, although alternative approaches such as reliabilism remain (Battaly 2020). Subsequent work focused on epistemic vices, understood as character traits, attitudes, and ways of thinking that obstruct or otherwise impair knowledge acquisition, retention, and transmission. Quassim Cassam first named the field of vice epistemology in his paper “Vice Epistemology” (Cassam 2016) and developed it in his book Vices of the Mind: From the Social to the Political (Cassam 2019). It was further advanced in works by Heather Battaly, Ian James Kidd, Alessandra Tanesini, and others. Epistemic vices include both recognizable and proverbial everyday epistemic failings, such as dogmatism, closed-mindedness, and epistemic arrogance, or more obscure vices unfamiliar to the layperson, such as

4  Hana Samaržija Miranda Fricker’s testimonial and hermeneutical injustice, failures to ascribe vulnerable persons the trust they are due or to acknowledge their perspectives in our collective intellectual resources (Fricker 2007). Of course, vice epistemology did not arise out of the blue or develop solely as the abstract conceptual opposition to virtue epistemology: it was preceded by a lengthy and self-consciously international history of studying positive and negative character traits in different ancient, medieval, and renaissance traditions (Kidd, Battaly, and Cassam 2020, 3). However, we can grasp the most concrete link between intellectual virtues and vices in Fricker’s much discussed Epistemic Injustice: Power and the Ethics of Knowing, a book that essentially described two severe intellectual vices – testimonial and hermeneutical injustice – that the author endeavored to resolve by stimulating agents to develop the matching virtues of testimonial and hermeneutical justice. Most significantly, the legacy of virtue and vice epistemology finally brought epistemology down from its ivory tower and cradled it closer to our genuine experiential world. Owing largely to the work of Alvin Goldman, virtue and vice epistemologists have increasingly come to accept that philosophy ought to interact with the cognitive and social sciences to broaden its scope and significance (Goldman 2003). As a result, both virtue and vice epistemology have acquired a social dimension and established links with social epistemology, the study of the social conditions and obstacles to knowledge acquisition (Goldman 1990). Social epistemology dealt with questions such as the necessary conditions for trusting and distrusting others’ testimonies (Lackey 2010), fruitfully resolving ostensibly unending disagreements between scientific peers, the social environment’s influence on epistemic injustice toward stigmatized social groups (Fricker 2007), institutional rather than agential solutions to differential epistemic capital (Samaržija and Cerovac 2022), and collective epistemic agents such as smaller groups, entire democratic electorates, political parties, or institutions (Brady and Fricker 2016). Even as social epistemology branched out into more specialized research fields, it did not relinquish its interest in subjects such as the necessary and sufficient conditions for testimony to be reliable, the optimum stance to assume when two equally prominent academics fail to find common ground, and the realistic probability of relying on flawed individuals rather than institutions to abolish entrenched epistemic injustices (Fricker, Graham, Henderson, and Pedersen 2020). Several of social epistemology’s subjects branched into fecund and self-standing research fields. For instance, now we can speak of the established field of applied epistemology, which employs social epistemology’s conclusions to tackle concrete social problems, such as petrified disputes in the scientific community (Coady 2020), the self-­explanatorily titled ignorance studies committed to coping with the social side effects of disinformation in democracies dependent on informed electorates, and an entire intersectional field devoted to epistemic injustice that fruitfully

Introduction 5 interacts with feminist, racial, and post-colonial studies (Kidd, Medina, and Pohlhaus Jr. 2019). Several decades later, this interest in politically pertinent epistemic collectives, myriad alarming policy decisions, blatant climate inaction impervious to despondent scientific warnings, and the Internet’s development into a notoriously unregulated yet compelling platform for political and informal socialization incited the advent of political epistemology (Edenberg and Hannon 2021; Hannon and de Ridder 2021). Political epistemology is “a newly thriving field at the intersection of epistemology and political philosophy” (Hannon and de Ridder 2021, 1). It examines and studies its ancient predecessors, who tackled similar issues without an opportune umbrella term to encompass their views. Plato and Aristotle made significant contributions to what would today be known as political epistemology (Plato 2007; Aristotle 2013). Today, political epistemologists study political disagreement and polarization (Edenberg 2021; Iyengar 2021), the epistemic consequences of disinformation and misinformation (Gelfert 2021), epistemic virtues and vices in politics (Tanesini 2021), the Internet’s influence on political decision-making (Gunn 2021), political rationality and irrationality (Friedman 2021; Somin 2021), cognitive biases that stymie understanding (Stanovich 2021b), and the epistemic justification of democracy as a collective decision-making system (Prijić-Samaržija 2018). De Ridder and Hannon’s accessible Handbook of Political Epistemology trailed decades of arduous labor in introducing novel topics and innovating the present approaches to perennial political issues. Nevertheless, political epistemology’s concern for fallible and painlessly manipulable democratic processes branched political epistemology onto the novel, fruitful, and more specialized territory of the epistemology of democracy.

0.3  The Structure of the Volume This volume comprises three thematic sections, each representing a different conceptual, ethical, and epistemic attitude toward democracy’s epistemic value. The sections are not methodologically unified, as each author proffers their unique perspective and approach to the chosen problem. While some stay true to standard analytic philosophy’s rigorous deductive reasoning, intentionally bereft of enigmatic formulations, others conjoin analytic philosophy’s cutthroat clarity with a more continental historicism and stylistic eclecticism and speckle their papers with elegant passages rich with wry humor. This diversity of approaches contributes to the volume’s popular appeal and significance to the scientific community. Although the initial distribution among pessimists, optimists, and realists may appear provisional – as not all pessimists are equally committed to expert rule or other alternatives to democracy, and most optimists and realists are justifiably critical of democracy’s present state – it is valuable in navigating the volume and its elaborately nuanced contents. Whether

6  Hana Samaržija they advocate expert rule or dissimilar surrogates for democracy, all pessimists share distinct anxiety about the conceivable consequences of the emergent culture of ignorance and its correlation to social networking in obstructing present and future democratic processes. Likewise, optimists are cohesive in seeking the answer to an enhanced democracy in more genuinely democratic procedures: deliberation ameliorated with the competencies of empathizing and listening to adversaries or electoral processes ornated with more diverse participants. In the end, realists share a joint perspective on democracy’s presently dissatisfactory condition but do not venture to reform it. Instead, they focus on specific problems within democratic coexistence, such as the ostensible stipulation for empathy and compassion for our political rivals, or review varied assortments of pessimistic and optimistic proposals to amend democratic practices. This volume is also geographically inclusive in uniting political philosophers from all over Europe and the United States, underlining that a healthy governmental system is a subject essential to all of us. The first section, Democratic Pessimism, comprises five chapters that maintain a skeptical attitude toward democracy’s epistemic value and its quantitative defenses, question the professed epistemic value of self-reliant critical thinking and civic engagement, reconsider the negative effect of biases on democratic processes, or ponder upon the future of democracy in the context of the advent of social networking, filter bubbles, and echo chambers. The second section, Democratic Optimism, poses a formidable challenge to the first. While some authors hold that hopeful trust and honest communication, both eroded by polarization, constitute and enhance democracy’s epistemic value, others argue in favor of the epistemic benefits of diversity in resolving problems and attenuating myside bias or claim that an appropriate dose of cynicism toward institutions can fortify democratic engagement. The third and final section, Democratic Realism, maintains a neutral stance toward the perennial dilemma between truth and political values. Instead, the five included epistemologists argue that democracy encourages us to endeavor to make sense of viewpoints we deem reprehensible and maintain civil relationships with our political enemies, that truth and justice are not in conflict unless we make them so, and that economic inequalities in the informal political sphere can erode the political equality at the foundations of a healthy democracy.

0.4  Summary of Chapters 0.4.1  Democratic Pessimism Political epistemologist and philosopher of economics Jason Brennan opens the volume with a meticulous repudiation of the Hong-Page Theorem of Diversity over Expertise, frequently utilized to defend democratic or collective decision-making. In Brennan’s reading, the theorem

Introduction 7 endeavors to show that the miscellany of a decision-making group is a more significant predictor of the decision’s epistemic value than the reliability or the competence of individual epistemic agents. Brennan argues that the theorem’s ostensible proofs are questionable in most epistemic settings and that the proof itself is frivolous and mathematically unsubstantiated. He, therefore, determines that we should not engage the theorem in discussions concerning the epistemic value of democracy. Political and social epistemologist, and the primary editor of the volume, Hana Samaržija continues with the subject of rebutting theorems in ostensive favor of democratic decision-making. She carefully assesses Condorcet’s Jury Theorem, another mathematical proof often utilized in defenses of democracy, and its central claim that the probability a decision will be correct increases with the number of decision-makers, regardless of their expertise on the chosen topic. Samaržija argues that Condorcet’s three critical requirements expected from his imagined decision-makers – ­competence, independence, and sincerity – cannot outlive the contemporary epistemic circumstances of extreme political polarization, heightened enmity among the electorate, ambiguous algorithmic sorting on social networks, social epistemic structures such as epistemic bubbles and echo chambers, and strategic voting driven by a desire for material gain. She thus concludes that Condorcet’s Jury Theorem can no longer be sensibly employed as an epistemic justification of democracy. In the section’s third chapter, eclectic political philosopher and philosopher of mind Nenad Miščević tackles politically hazardous epistemic vices with substantial sociopolitical corollaries, such as closed-mindedness, epistemic arrogance, cowardice, dishonesty, and dogmatic epistemic self-affirmation. He then associates them with the advent of populism in global politics and endeavors to explicate them by commenting on cognitive miserliness, the systematic human intellectual failures liable for regular defects in cognitive processing. Miščević terminates his chapter with the stance that such miserliness-based vices are the most noteworthy systematic hindrance to fruitful political deliberation. Political, institutional, and social epistemologist Snježana Prijić-Samaržija proceeds with an innovative reading of critical thinking as a cognitive attitude that is only virtuous if it entails deference to objective epistemic authorities. Prijić-Samaržija acknowledges that social epistemology’s interdisciplinary perspective has allowed us to rethink the epistemic justification of democracy and the notion of critical thinking, which has become an educational platitude barren of meaning. She asserts that in the current culture of ignorance, replete with disinformation, misinformation, fake news, and conspiracy theories, people’s autonomously formed beliefs do not possess the epistemic value they would enjoy in idealized epistemic circumstances. Prijić-Samaržija then argues that the seemingly permanent animosity between critical thinking and regard for experts, founded primarily on ­modernity’s reverence for autonomous reflection and later postmodern critiques of expertise, posits

8  Hana Samaržija a particular threat within the current culture of ignorance and the ideology of each person’s unique judgment-based truth. As an alternative, she offers critical thinking based on justified deference to objective epistemic authorities as an essential constituent of democratic decision-making. Prijić-Samaržija closes her chapter by underlining the disquieting effects of identifying critical thinking with arbitrary and often misinformed or conspiratorial beliefs in the contemporary culture of ignorance. As the opening section’s final point, philosopher of mind and political epistemologist Åsa Wikforss closes the unit with the argument that the recently undependable information environment poses a grave threat to democratic practices. She recognizes that specific scholars hold that philosophers concerned about post-truth epistemic circumstances and their effect on democracy are erroneous in presuming that truth is essential to democracy. In their reading, truth is intrinsically an anti-democratic concept, as it does not hinge on the values and opinions of the masses. However, Wikforss underlines that such objections are misconstrued, as truth and knowledge play an indispensable role in sustaining democracies. Although democratic decision-making involves objective facts and values, diminishing the significance of expertise remains deleterious for democracy. To illustrate her point, she concludes the first section by discussing the damages that disinformation and misinformation pose to present liberal democratic societies. 0.4.2  Democratic Optimism Social, virtue, and vice epistemologist Alessandra Tanesini commences the section by analyzing a pertinent aspect of our emotionally charged political climate, a phenomenon she dubs “the politics of resentment.” This concept depicts a political perspective effectively articulated by the immediate attitude or moral emotion of resentment, a quality conspicuous in those who feel left behind or forlorn by their nation. Their resentment is an embittered response to an identified vitiated social status and to alleged or actual threats to the world which had once ensured their self-esteem. Epistemic agents engaged in the politics of resentment fear a further hierarchical collapse, frequently witness worsened economic conditions, and are losing privileges hitherto conferred to them due to their dominant ethnic and gender identities. Tanesini proffers an explanation for the affective reaction with resentment as a direct retort to the experience of losing some identity-defining hopes and emboldens scholars striving to examine the most significant hazards confronting democratic institutions to broaden their focus on the emotional rather than solely on the rational aspects of political life. In the second chapter, political and social epistemologist Marko Luka Zubčić proceeds by asserting that it has become philosophically prevalent to justify democracy by centering on the incidence of individual epistemic virtues within an electorate. In contrast, he claims that

Introduction 9 the epistemic value of institutional systems hinges on collective rather than agential virtues and endorses democracy by arguing it displays these collective epistemic virtues. Zubčić constructs his first case by establishing an explanatory structure based on institutional epistemology and the division of cognitive labor. As we can divulge none of these properties in individual epistemic agents, he maintains that systems based on agential expertise suffer from collective epistemic vice. Zubčić continues by arguing for Anderson’s experimentalist model of the epistemic justification of democracy as a paradigm of democracy satisfying the central requirements of institutional epistemic reliability. Finally, he criticizes present representative democracies that fail to exhibit his reading of group epistemic virtues. Prominent character epistemologist Ian James Kidd continues by diagnosing that philosophers and related theoreticians remain ambiguous about political cynicism. In his chapter, he establishes and supports a concept he dubs “institutional cynicism,” indicating it can promote a virtuous civic demeanor in present democracies. While Kidd acknowledges that specific forms of cynicism can be deleterious for healthy democratic practices, he asserts that institutional cynicism renders us more engaged citizens by augmenting our sensitivity to the defects of present political institutions. He, therefore, determines that institutional cynicism can contribute to our knowledge about how institutions genuinely operate without resulting in defeatism or despondency. Professor Emeritus, who pioneered myside bias in academic literature, Keith Stanovich proceeds by clarifying that it occurs when epistemic agents assess evidence and test hypotheses in a manner biased toward their prior opinions and attitudes. Myside bias does not discriminate, as we can divulge it in agents from all demographic groups. It is demonstrated even by highly educated and astute experts whom we expect to suppress their biases. Another way myside bias is an outlier bias is that it shows signs of very little domain generality and appears highly content-dependent. As intelligence and education do not seem to prevent mysided thinking and data assessments, elite research faculties studying divisive issues cannot be expected to attain objective conclusions unless they have employed a process of an adversarial collaboration. Stanovich criticizes ideological monocultures at universities for further preventing confrontational viewpoints. He concludes his chapter with the attitude that public trust in social science evidence will continue to deteriorate unless we employ measures to ensure that we are evaluating social issues using various frameworks. Lastly, social epistemologist Hanna Gunn argues that the hypothetical epistemological crisis suggests that, due to the hazard of heightening polarization, we might have lost the capability to engage effectively in democratic deliberation. A particularly perturbing concern is that democratic decision-making has become either futile or impossible, as it hinges on the quality of the interaction within democratic societies. Gunn notes that former social epistemological inquiries into public discourse traditionally attended to speakers, discussing

10  Hana Samaržija their dispositions, obligations, and relative social power. In comparison, she argues, there has been meager talk of listening and public discourse. According to Gunn, this chasm has grave effects as communication cannot occur without active and productive listening. She argues that numerous social epistemological analyses of public deliberation already presuppose we have listening obligations. Therefore, she asserts that we have a democratic duty to cultivate our listening competencies or skills as listeners. Given these competencies’ complex and somewhat ambiguous nature, Gunn argues that sustaining them requires both agential and collective actions that can be accomplished, among other pathways, through education. She claims her proposal builds on preceding examinations of the connection between classroom and liberal democratic competencies. As a final point, Gunn concludes that her proposal aligns with critical pedagogical work that adopts an analogous position on the relationship between the classroom structure and the quality of liberal democratic societies. 0.4.3  Democratic Realism The vice epistemologist, extremism scholar, and co-editor of the volume, Quassim Cassam opens the concluding section by assuming a critical stance against the view that emotional empathy is vital for democracy or an antidote to polarization. Empathy is a form of what Cassam calls “sensemaking,” but making sense of a political opponent’s beliefs and actions does not require one to adopt their perspective in a way that engages one’s emotions. According to Cassam, active listening rather than emotional empathy is indispensable for political sensemaking and democratic legitimacy. He notes that mutual understanding will not necessarily overcome political polarization since political adversaries might understand one another only too well. He cites Amos Oz’s view that some conflicts are actual and much worse than a misunderstanding. When socially or economically marginalized individuals appear blind to their interests, this might result from “false consciousness,” which Cassam defines as “a mode of consciousness that misrepresents socio-economic reality while also being determined by that reality.” A generalization about highly unequal societies is that they are often kept on an even keel by their ability to induce large numbers of socially and economically marginalized people to believe that the status quo works for them. However, historical observation rather than empathy reveals the truth of this generalization. Political epistemologist Michael Lynch proceeds in the same vein by conceding that epistemic agents inquisitive about politics frequently feel despondent about our failure to conceive a more just and equitable society. According to Lynch, the idea of political progress presupposes we know where we are heading and assumes we desire to approach genuine justice but lack knowledge on how to reach it. In his reading, the idea of political progress is inevitably tied to the much more

Introduction 11 complex question of whether there is truth in politics. He dubs the view we are incapable of comprehending political truths “political skepticism” and reviews several of its arguments, suggesting the reality of a specific form of epistemic corruption often called epistemic colonization. Lynch concludes by asserting that the skeptical threat should motivate political engagement and inclusion rather than quietism. On the other hand, political philosopher and epistemic democrat Ivan Cerovac tackles the influence of inequitable economic power on fair democratic procedures and their epistemic outcomes. Cerovac correctly remarks that equal electoral influence has been a vital political norm for over a century and is commonly deemed essential to democratic legitimacy. However, according to Cerovac, considerable imbalances in earnings, wealth, and capital ownership recurrently spill over from the economic and social spheres to the political arena. Therefore, his chapter catalogs and analyzes the processes that transform economic into political power, such as campaign donations, ownership of prominent media outlets, or lobbying to demarcate the political agenda or curb the scope of feasible political decisions. Cerovac elaborates his chapter by contrasting two manifestly epistemic approaches to economic inequality’s impact on democracy’s epistemic outcomes, the liberal epistemic and the egalitarian approach, and argues in favor of the latter. Ultimately, the chapter evaluates two comprehensive strategies to preserve democracy’s epistemic value. While the first approach strives to curb the mechanisms that relocate inequality from one sphere of life to another, the second advocates for a more egalitarian economic system and endeavors to mitigate present economic disparities. In the end, Cerovac observes that resolving the problem in suboptimal epistemic circumstances might necessitate harmonizing policies implemented from both approaches. In the penultimate chapter, political philosopher and epistemologist Robert Basil Talisse argues that democratic citizenship comprises two comprehensive moral directives. First, citizens must endeavor to pursue justice by engaging in the democratic process. Second, as all democratic citizens are political equals, they must remain responsible to one another. When discordant concerning politics, they must give their adversaries a fair hearing and acknowledge their equal entitlement to agential political beliefs. In real-world politics, however, treating agents whose views we deem reprehensible seems to concede something substantial to their opinions, which contradicts the first directive on working at justice. Therefore, politically engaged citizens face a dilemma. When it comes to tangible politics, why not treat one’s ideological adversaries as enemies rather than fellow citizens? Talisse’s chapter builds upon empirical research regarding belief polarization to suggest an epistemic underpinning for treating our political adversaries as equals. In short, to sustain the epistemic conditions under which one can pursue justice with their political allies, Tallise concludes they must also preserve civil relations with their rivals. Philosopher of law and

12  Hana Samaržija political epistemologist Ilya Somin concludes the volume by accepting the popular accord that pervasive voter ignorance and irrational assessments of political data are solemn hindrances to a healthy democracy. However, scholars differ in their approaches to the problem. “Top-down” strategies, such as expert rule and augmented deference to epistemic authorities, strive to diminish ignorance by giving way to more knowledgeable segments of an electorate. Conversely, “bottom-up” tactics, such as tutoring citizens and endorsing deliberative democracy, endeavor to heighten the common public’s political competence or proffer them enhanced inducements to make reasonable decisions than standard ballot box voting. Somin then reviews and criticizes an assortment of topdown and bottom-up strategies, determining that top-down approaches entail detrimental shortcomings that render our democratic environment even more malign. He concludes the volume by arguing for what he names “foot-voting” and the prospect of reimbursing voters to tackle their dearth of political knowledge.

References Ahlstrom-Vij, Kristoffer. 2012. “While Deliberative Democracy is (Still) Untenable.” Public Affairs Quarterly 26 (3): 199–220. Ahlstrom-Vij, Kristoffer. 2013. “Why We Cannot Rely on Ourselves for Epistemic Improvement.” Philosophical Issues 23: 276–296. Ahlstrom-Vij, Kristoffer. 2018. “Is Democracy an Option for the Realist?” Critical Review 30 (1–2): 1–12. Ahlstrom-Vij, Kristoffer. 2020. “The epistemic benefits of democracy: a critical assessment.” In The Routledge Handbook of Social Epistemology, edited by Miranda Fricker, Peter J. Graham, David Henderson, and Nikolaj J. L. L. Pedersen, 406–415. New York: Routledge. Aristotle. 2013. Politics. Translated by Carnes Lord. Chicago: University of Chicago Press. Battaly, Heather. 2008. “Virtue Epistemology.” Philosophy Compass 3 (4): 639–663. Battaly, Heather. 2020. The Routledge Handbook of Virtue Epistemology. New York: Routledge. Brady, Michael, and Fricker, Miranda. 2016. The Epistemic Life of Groups. Oxford, United Kingdom: Oxford University Press. Brennan, Jason. 2016. Against Democracy. Princeton: Princeton University Press. Brennan, Jason. 2020. “Epistemic democracy.” In The Routledge Handbook of Applied Epistemology, edited by David Coady, 88–101. New York: Routledge. Caplan, Bryan. 2007. The Myth of the Rational Voter: Why Democracies Choose Bad Policies. Princeton: Princeton University Press. Cassam, Quassim. 2016. “Vice Epistemology.” The Monist 99 (2): 159–180. Cassam, Quassim. 2019. Vices of the Mind: From the Social to the Political. Oxford, United Kingdom: Oxford University Press. Coady, David. 2020. The Routledge Handbook of Applied Epistemology. Oxford, United Kingdom: Oxford University Press.

Introduction 13 Dalton, Russel J., and Klingemann, Hans-Dieter. 2009. The Oxford Handbook of Political Behavior. Oxford, United Kingdom: Oxford University Press. Delli Carpini, Michael. 2005. “An Overview of the State of Citizens’ Knowledge About Politics.” In Communicating Politics: Engaging the Public in Democratic Life, edited by Mitchell S. McKinney, Lynda L. Kaid, Dianne G. Bystrom, and Diana B. Carlin, 27–40. New York: Peter Lang Publishing. Delli Carpini, Michael, and Keeter, Scott. 1996. What Americans Know about Politics and Why It Matters. New Haven: Yale University Press. Edenberg, Elizabeth. 2021. “Political disagreement: epistemic or civic peers?” In The Routledge Handbook of Political Epistemology, edited by Michael Hannon and Jeroen de Ridder, 123–133. New York: Routledge. Edenberg, Elizabeth, and Hannon, Michael. 2021. Political Epistemology. Oxford, United Kingdom: Oxford University Press. Fricker, Miranda. 2006. “Powerlessness and Social Interpretation.” Episteme 3 (1–2): 96–108. Fricker, Miranda. 2007. Epistemic Injustice: Power and the Ethics of Knowing. Oxford, United Kingdom: Oxford University Press. Fricker, Miranda, Graham, Peter J., Henderson, David, and Pedersen, Nicholas J. L. L. 2020. The Routledge Handbook of Social Epistemology. New York: Routledge. Friedman, Jeffrey. 2021. “Is political irrationality a myth?” In The Routledge Handbook of Political Epistemology, edited by Michael Hannon and Jeroen de Ridder, 263–274. New York: Routledge. Gelfert, Axel. 2021. “What is fake news?” In The Routledge Handbook of Political Epistemology, edited by Michael Hannon and Jeroen de Ridder, 171–181. New York: Routledge. Goldman, Alvin. 1990. Knowledge in a Social World. Oxford, United Kingdom: Oxford University Press. Goldman, Alvin. 2003. Liaisons: Philosophy Meets the Cognitive and Social Sciences. Massachusetts: MIT Press. Goodin, Robert, and Spiekermann, Kai. 2018. An Epistemic Theory of Democracy. Oxford, United Kingdom: Oxford University Press. Gunn, Hanna. 2021. “Filter bubbles, echo chambers, online communities.” In The Routledge Handbook of Political Epistemology, edited by Michael Hannon and Jeroen de Ridder, 192–203. New York: Routledge. Hannon, Michael, and de Ridder, Jeroen. 2021. The Routledge Handbook of Political Epistemology. New York: Routledge. Hazlett, Allan. 2012. “Higher-Order Epistemic Attitudes and Intellectual Humility.” Episteme 9 (3): 205–223. Iyengar, Shanto. 2021. “The polarization of American politics.” In The Routledge Handbook of Political Epistemology, edited by Michael Hannon and Jeroen de Ridder, 90–101. New York: Routledge. Kidd, Ian James, Battaly, Heather, and Cassam, Quassim. 2020. Vice Epistemology. London, United Kingdom: Routledge. Kidd, Ian James, Medina, José, and Pohlhaus Jr., Gaile. 2017. The Routledge Handbook of Epistemic Injustice. New York: Routledge. Kuklinsky, James H., and Peyton, Buddy. 2007. “Belief Systems and Political Decision-Making.” In The Oxford Handbook of Political Behavior, edited by Russel J. Dalton and Hans-Dieter Klingemann, 46–65. New York: Oxford University Press.

14  Hana Samaržija Lackey, Jennifer. 2010. Learning from Words: Testimony as a Source of Knowledge. Oxford, United Kingdom: Oxford University Press. Landemore, Hélène. 2012. Democratic Reason: Politics, Collective Intelligence, and the Rule of the Many. New Jersey: Princeton University Press. Landemore, Hélène. 2020. Open Democracy: Reinventing Popular Rule for the Twenty-First Century. New Jersey: Princeton University Press. Levitsky, Steven, and Ziblatt, Daniel. 2019. How Democracies Die. United States: Broadway Books. Locke, John. 1988. Locke: Two Treatises on Government. Cambridge, United Kingdom: Cambridge University Press. Marx, Karl. 2000. Karl Marx: Selected Writings. Oxford, United Kingdom: Oxford University Press. Mill, John Stuart. 1982. On Liberty. London, United Kingdom: Penguin Books Ltd. Plato. 2007. The Republic. Translated by Desmond Lee. London, United Kingdom. Penguin Books Ltd. Prijić-Samaržija, Snježana. 2017. “The Role of Experts in a Democratic DecisionMaking Process.” Ethics & Politics 19 (2): 229–246. Prijić-Samaržija, Snježana. 2018. Democracy and Truth: The Conflict Between Political and Epistemic Virtues. Udine: Mimesis International. Rousseau, Jean-Jacques. 2003. The Social Contract. Translated by Maurice Cranston. London, United Kingdom: Penguin Books Ltd. Samaržija, Hana, and Cerovac, Ivan. 2022. “The Institutional Preconditions of Epistemic Justice.” Social Epistemology 35 (6): 621–635. Somin, Ilya. 2021. “Is political ignorance rational?” In The Routledge Handbook of Political Epistemology, edited by Michael Hannon and Jeroen de Ridder, 241–253. New York: Routledge. Somin, Ilya. 2023. “Bottom-Up and Top-Down Solutions for Political Ignorance.” In The Epistemology of Democracy, edited by Hana Samaržija and Quassim Cassam. New York: Routledge. Sosa, Ernest. 1980. “The Raft and The Pyramid: Coherence Versus Foundations in the Theory of Knowledge.” Midwest Studies in Philosophy 5 (1): 3–26. Sosa, Ernest. 2001. “Goldman’s Reliabilism and Virtue Epistemology.” Philosophical Topics 29 (1–2): 383–400. Stanovich, Keith. 2021a. The Bias That Divides Us: The Science and Politics of Myside Thinking. Massachusetts: MIT Press. Stanovich, Keith. 2021b. “The irrational attempt to impute irrationality to one’s political opponents.” In The Routledge Handbook of Political Epistemology, edited by Michael Hannon and Jeroen de Ridder, 274–285. New York: Routledge. Talisse, Robert. 2021. Sustaining Democracy: What We Owe to the Other Side. Oxford, United Kingdom: Oxford University Press. Tanesini, Alessandra. 2021. “Virtues and vices in public and political debates.” In The Routledge Handbook of Political Epistemology, edited by Michael Hannon and Jeroen de Ridder, 325–336. New York: Routledge. Zagzebski, Linda Trinkaus. 1996. Virtues of the Mind: An Inquiry into the Nature of Virtue and the Ethical Foundations of Knowledge. Cambridge, United Kingdom: Cambridge University Press. Zagzebski, Linda Trinkaus. 2012. Epistemic Authority: A Theory of Trust, Authority, and Autonomy in Belief. Oxford: Oxford University Press.

Part I

Democratic Pessimism

1

Sexy but Wrong Diversity Theorem Defenses of Democracy Jason Brennan

Since Aristotle’s time, democratic theorists have argued that democracy or inclusive decision-making tends to be intelligent. They claim that two heads are better than one and many decision-makers are better than two. Aristotle says: The best man, then, must legislate, and we must pass laws, but these laws will have no authority when they miss the mark, though in other cases retaining it. But when the law cannot determine a point at all, or not well, should the one best man or all decide? According to our practice, assemblies meet, sit in judgment, deliberate, and determine, all of which relate to individual cases. Now any member of the group, taken separately, is certainly inferior to the wise man. But the state is made up of many individuals. And as a feast to which all guests contribute is better than a banquet furnished by a single man, a multitude is a better judge of many things than a single individual.1 Aristotle provides little evidence that these claims are valid, but many theorists find the sentiment appealing. (They think this despite, I take it, having gone to many lousy potlucks and seen extensive first-hand disconfirmation of Aristotle’s example.) Maybe Aristotle is right. Perhaps a large crowd beats a small group even if the individual people in the small crowd are more intelligent, more capable, or better informed. In 2004, Lu Hong and Scott Page published what appeared to be powerful proof of a remarkable theorem that could support this argument. The theorem claims that in collective decisions, increasing the cognitive diversity inside the group more strongly contributes to the group’s tendency to choose the correct answer than increasing the reliability or competence of individual decision-makers.2 At some margin, it is better to have a more diverse but less competent crowd than a less varied but brighter crowd. Diversity is an ascendant value in academia. You are not supposed to challenge diversity, and you are supposed to conclude diversity is good. Indeed, people can be ostracized, investigated, or fired for critiquing DOI: 10.4324/9781003311003-3

18  Jason Brennan diversity as a goal. The Hong-Page theorem seems to provide a robust theoretical backing for something people want to believe is true. So, it is unsurprising that this conclusion was widely lauded, and the paper received extensive uptake. Therefore, many prominent recent defenses of democracy rest upon it. For instance, Hélène Landemore uses it to argue that democracy always beats any form of decision-making by experts.3 Elizabeth Anderson argues it supports democratic decision-making, though she regards other models of collective decisions as superior.4 Scott Page himself uses the model to defend increased diversity and democratic decision-making in corporations.5 The 2004 paper has over 1635 citations as of April 2022, most of which appear to accept its conclusions and apply them to argue for increased diversity here or there. This chapter explains that we should not use the Hong-Page theorem to defend democracy. The proof of the Hong-Page theorem in the 2004 paper is trivial and question-begging. Hong and Page do not “prove” that diversity trumps ability in group decision-making; instead, they come closer to assuming it. Further, even if the proof were not trivial, applying the theorem to democracy is problematic because “diversity trumps ability” only when people behave in specific ways. Real-life democratic participants do not meet the behavioral requirements of the theorem. So, the theorem is not well-grounded a priori and cannot be applied to democracy a posteriori.

1.1  Problems with the Assumptions of the Proof Let’s examine the assumptions of their proof. Note carefully that these are the assumptions or premises of their evidence. 1 They assume that all agents in the collective decision have the same value function: the exact ranking/ordering of possible outcomes for better or worse.6 Every social scientific or physical model contains simplifying assumptions which render it an imperfect fit for the real world. That is not inherently a problem. For instance, in physics, we might model many fundamental particles as points though they may have some spatial extension. The ideal gas law describes gases with no molecular or atomic attraction toward one another, but in the real world, gases have some magnet even under low pressure. Gases are not genuinely ideal. Thus, one problem with modeling is assessing whether the simplifying assumptions assume something important, which renders the model impotent or irrelevant. Sometimes the simplifying assumptions are acceptable, and sometimes the things studied depart so far from the model that the model becomes useless. The Hong-Page model does not prove that citizens who deliberate together will come to share the same values. Instead, it assumes at

Sexy but Wrong 19 the outset that their values are the same. They agree on the problem, have the same values, and decide on what would count as a solution. Still, as we will see, the model presumes they have different problem-­ solving capabilities and differing degrees of reliability. However, because of the assumption of a lack of diversity in the decision-­ makers’ value function, the theorem is applicable only in exceptional cases where people share the same value function. In some cases, such as when a group of corporate leaders is trying to choose a business strategy to maximize profit, assuming everyone shares a value function seems appropriate. But in others, such as when diverse citizens with diverse values cast votes for political parties or candidates, it does not apply. Oddly, the theorem is often used to defend democratic decisions, but this is one of the cases where the theorem is least applicable. In the real world, agents making a collective political decision often possess different values and thus rank distinct states of affairs differently. They do not agree on what counts as a solution to the problem and usually do not agree on the issue. If a Marxist, a libertarian, a Rawlsian, and a conservative deliberate on tax policy, they have different ordinal rankings of states of affairs. 2 Hong and Page assume that the problem these agents are trying to solve is so tricky that no agent can solve the problem alone.7 In their general mathematical proof of their theorem, Hong and Page begin with the assumption that any agent in the group cannot solve the problem the group intends to solve. Accordingly, their proof concludes that many will outperform one because it simply assumes that one cannot solve the problem themselves. So, using the Hong-Page theorem to argue against the rule of one expert is question-begging because the proof presupposes no individual can solve a problem alone. Whether the Hong-Page model applies to any real-world decision is an open question. There may be some problems, such as fixing the leak behind the toilet, where an individual agent can find the solution alone. There may be others where they cannot. It is an empirical question, not a question to be settled a priori or by stipulation. 3 They assume that every agent in the decision-making process has one “heuristic” or method she uses to solve the problem.8 Thus, they assume if that agent uses that heuristic, she gets stuck on some answer or remains stuck until someone else helps her. This is, in part, a straightforward conclusion of their difficulty assumption and their assumption about agents’ abilities. Agents are described by having search rules that they use to find the answer to the group’s problem, and they are stipulated to be unable to solve the problem alone. Accordingly, on their own, when given numbers to try to map onto a value function, they find a local optimum but not necessarily the true optimum.

20  Jason Brennan Oddly, Hong and Page thus characterize their agents as lacking internal “cognitive diversity.” Each agent in their model has a particular problem-solving technique given their mental model and reliability. This part is where the proof starts to become problematic. The assumption that each agent has only one problem-solving method strongly partly trivializes the claim that two heads are better than one, that committees can outperform individuals, or that many heads are better than fewer. After all, we might imagine instead that each agent has a variety of heuristics and distinct mental models of the world. For instance, both natural and social scientists have diverse models which can be employed to describe the world. Individual high-ability people often have varied problem-solving methods and heuristics. When one strategy fails or is imperfect, they switch and try another. I, for one, have multiple modes and skill sets. But Hong and Page model their decision-makers, both the low- and high-ability agents, as having only one way of approaching a problem and being unable to switch methods to improve their work. This assumption thus reduces the value of what we might call “high ability” agents. The high-ability agents in their model are stipulated to have a highly reliable but imperfect heuristic or problem-solving method. When they get stuck, they can become unstuck when others help them. It is assumed that they cannot spontaneously adopt the search function of a different agent; they cannot switch to a distinct perspective on their own. Whether one finds this inappropriate or not depends partly on how one wants to characterize individual problem-solving agents in the real world. One might argue that we all have precisely one heuristic or method of solving a problem. When switching heuristics or procedures, we have just one bigger meta-level heuristic or problem-­ solving approach. The choice to use a hammer or a wrench, a sociological or an economic model, or quantum mechanics or general relativity collapses into one super-heuristic. But that seems implausible and question-begging. After all, what is the real difference between two agents with two methods and one with two approaches we call a “super-heuristic”? Their paper is sparse in describing agents’ individual decision-making functions, but it does not appear to accommodate internal diversity. After all, if, as they presume (see below), agents who get stuck can learn from other agents who have different methods, we could imagine individual agents instead spontaneously adopting those methods independently. Instead of a stuck agent needing another agent with a different approach to use a distinct heuristic, we can imagine the stuck agent himself using that same heuristic on

Sexy but Wrong 21 his own. The only reading that makes sense of their proof is that they are modeling agents as having relatively unique and straightforward methods rather than having the ability to switch strategies. If, on the contrary, they want to collapse the idea that switching among modes is a single complicated method, then their assumption that another agent can permanently save a stuck agent (see below) would be unwarranted as then individual agents’ search functions will already contain other agents’ heuristics. This last paragraph repeatedly mentions that if an agent gets stuck on a solution, another agent with a different method can always help them. Let’s examine this assumption next. 4 Hong and Page assume that whenever one agent gets stuck, there is “always” another agent who can improve upon the first by using a different heuristic. Here is the direct quotation from their paper:

{ }

Assumption 2 (Diversity). ∀x ∈ X \ x* , ∃θ ∈θ such that φ ( x ) ≠ x. This assumption is a simple way to capture the essence of diverse problem-solving approaches. When one agent gets stuck, another agent can always find an improvement due to a different approach.9 This assumption is part of what renders their proof trivial and question-begging. The authors stipulate that whenever one competent agent – an agent who also by stipulation uses only one method to solve a problem which she cannot solve alone and can only get unstuck with help from others – gets stuck, there is always another agent who can improve the situation by using a different approach, method, or heuristic. They do not prove this conclusion; they stipulate it as an assumption. Notice that this is what they use the word “diversity” to signify in their proof. The term “diversity” here does not stand for and is not analogous to what campus or corporate diversity officers mean by “diversity.” It does not signify diverse racial or demographic identity, diversity of life experiences, mental models, memories, categories and ideas, conceptions of reality, philosophies, or whatnot. It instead means “having a different heuristic or problem-solving method which can improve the solution to the problem.” It certainly does not signify diverse values because, on the contrary, they presume all agents have the same value function and values; their proof only applies to cases that lack value diversity. Thus, “diversity” in the “diversity trumps ability” theorem does not mean what most of us mean by “diversity.” It means having a different problem-solving heuristic or method. To prove that “diversity”

22  Jason Brennan helps group decision-making, Hong and Page assume rather than prove that there is always another person with another technique or heuristic that can improve upon whatever anyone else has done. In other words, diversity helps because they use the term “diversity” to refer to the stipulated existence of another agent with a different method that can shake the first agent out of their rut. (Note, on their behalf, that this need not be because the second agent has an overall more reliable method, but just a different one that uncovers something the first agent missed.) In academia and the corporate world, it is considered very sexy and enlightened to promote diversity as a value. There may indeed be good reasons for doing so. Sometimes variety is invoked because people believe increasing the diversity inside a group will improve the group. For instance, increased racial, ethnic, gender identity, class, or other demographic diversity inside a university may enhance the quality of research or help overcome biases. While this belief may be accurate, we should not invoke the Hong-Page theorem lightly in favor of this conclusion. The Hong-Page theorem describes a collection of agents working together on the same problem, while academia mainly comprises individual agents working separately on distinct issues. Even when academics write on the same subject, they work separately. Second, the theorem describes agents as having different mental models and methods. At best, demographic diversity can improve collective decision-making on the Hong-Page model if this demographic diversity leads to distinct viewpoints and methodological diversity. If we have fifty people who represent six hundred different demographic groups but all think alike – for instance, if they are all Rawlsians, critical race theorists, or libertarians – then the theorem does not apply to them. The kind of diversity that matters for it is cognitive, not demographic. Accordingly, a good illustration of what Hong and Page have in mind is this: Imagine a corporation needs to solve public relations problem. For instance, the public mistakenly thinks that GMO foods are unhealthy, but the corporation sells GMO foods. The Hong-Page theorem suggests (if we accept its assumptions) that bringing in a mix of marketing professionals with different skill sets, designers, engineers, philosophers, and others of different methodologies, ontologies, and ways of seeing the world and that problem would improve the group’s decisions. But suppose we instead have a group that perfectly matches the demographic diversity of the world as a whole, but they all have the same ideas and methods. The theorem would say this group lacks the right kind of diversity and does not apply. 5 Hong and Page assume all agents defer to other agents when those other agents can improve the group decision. They presume agents always recognize when someone else has produced a better solution/

Sexy but Wrong 23 improved upon the current best solution and then always defer to that improvement. Hong and Page are less explicit about this, but it is a necessary condition of their proof. After all, they have already described individual agents as having a single but imperfect method. They stipulate that there is always another agent with a distinct heuristic or procedure that the first agent can use to improve their decision. In short, their model involves individual imperfect agents learning from each other and seeing what others have to offer. If we want to map this onto real-world decision-making, this is a questionable assumption. Instead, we could imagine that when agents apply their search functions/methods/heuristics/whatnot, they get stuck at local optima but not the best overall solution. We can then imagine that there is always another agent with a method that could improve their group decision when combined with the previous agents’ work. But we could also imagine that agents refuse to learn from or listen to others. This point is not unmotivated or fanciful. In real-life group decision-­making, people do not always (and in politics, even often) defer to the cognitive value of diverse perspectives or methods. Sometimes they insist they are right despite overwhelming evidence to the contrary. Sometimes they fail to see the value in others’ approaches despite wanting to be open-minded. Sometimes when they try to work together, they misuse methods. Sometimes everyone conforms to the most popular or high-status practice, even if it is unreliable. They may react to diverse perspectives by polarizing and become even more rigid or extreme. There is, in fact, extensive literature in psychology, political science, and other fields examining how real-life groups or collectives react.10 Hong and Page’s proof presumes they react the way we want them to, but this presumption may be unwarranted in the real world. In the real world, it may be that increasing the diversity (as Hong and Page define it) inside a group impedes collective decision-making even though, in principle, it adds information and skill which could improve it, simply because group dynamics and political psychology are often dysfunctional. So, the Hong-Page theorem “shows” that group decision-making leads to improved decisions because they assume that agents learn from each other rather than react badly. This presumption frequently does not map onto real-world agents. We will return to this issue below. 6 They assume there is always a wide range of heuristics/methods we can use to solve a problem, rather than just one or a few. This assumption is less troubling than the others, but it may be false. Perhaps specific problems require or can be solved with only one method. 7 They assume that the agents are trying to solve one problem.

24  Jason Brennan The authors describe agents trying to solve one problem rather than balancing multiple issues simultaneously. Again, we can ask whether this assumption maps onto certain real-world decisions which friends of the Hong-Page theorem take to illuminate. Consider a simple problem. Suppose we have a simple business that sells only one product: plain donuts. Suppose they are considering one issue: whether lowering the price by 10 cents will increase profits. Here, the Hong-Page model of the decision problem might seem to apply. However, what if we consider, say, a democratic election? Here, it seems more plausible that agents – even if contrary to the fact these agents shared the same values – are trying to solve multiple problems at once rather than one big problem. Indeed, it is unclear whether they are even trying to solve a problem. Those are some, if not all, of the assumptions underlying the proof of the Hong-Page theorem. However, given all these assumptions, it becomes trivial rather than interesting that when agents are tasked with solving a problem, they can improve decision-making by adding more heads. It still takes mathematical proof involving computer simulations, and many critics think they go wrong there.11 Here is what the proof says: Assume that there is some problem everyone agrees on, both on what the problem is and what counts as a solution. Assume people have different ability levels in solving that problem, but no one can solve the problem alone. Assume that everyone accepts any improvement or better solution offered by anyone else. Assume that every individual uses the only method to solve the problem and can never switch approaches. They can only change practices or perspectives if prompted by a second agent. Assume that if anyone gets stuck, there is always someone out there with a new method that can improve upon the currently offered solution. (Remember, this is what Hong and Page call “diversity.”) As we add more and more people to the group, our chances of adding a person who will help (remember – we stipulated that these helpful people exist) become higher and higher. After all, Hong and Page stipulated that when we add people to the group, they can either improve the group’s decision or have no effect. Their proof establishes that diversity trumps ability by relying on question-begging assumptions. In their proof, diversity trumps ability as ability is insufficient, and diversity is stipulated to be a thing that always helps. We can learn some general lessons here. If we are trying to model decision-­makers, our assumptions determine whether “diversity” beats ability or vice versa. It depends on the nature of the problem we stipulate, what kinds of methods we prescribe that agents use, whether we specify that agents learn from each other, might disagree, or might become dumber upon interacting with each other, and whether we stipulate that they agree on what counts as a solution or not, and how much internal

Sexy but Wrong 25 “diversity” we prescribe agents have. A priori, we can concoct scenarios in which twelve agents outperform hundred (which includes the twelve) or vice versa. It depends on the assumptions we model about what agents know, how sophisticated they are, how they react to one another’s methods, and what methods the agents use. The more important question will be which of these infinite possible models we can construct best explains actual group decision-making in various contexts.

1.2  Behavioral Assumptions Above, I have primarily focused on describing the proof in plain English. My main goal was to establish the proof is mainly trivial as it more or less assumes what it means to prove. Throughout doing so, I flagged some of the behavioral assumptions of the proof and said I would return to them later to ask whether they were correct. For argument, let’s imagine the proof was other than it was. Let’s imagine that Hong and Page had a non-question-begging mathematical model showing that cognitively diverse agents working together on what they recognize is a shared problem would turn out to tend to benefit from one another’s cognitive diversity and, as a result, would as a group tend to become more reliable and proficient. Nevertheless, even if they had produced such a proof, it would be an open question whether actual agents behave the way needed for the proof to work. As an analogy, the previous pages argued that they have not even established, say, the equivalent of the ideal gas law for ideal gases. But now we need to ask, by analogy, whether the gases in democracy behave in clearly non-ideal ways. Page says their model works when decision-makers are reasonably well-informed and reasonably sophisticated, if not as sophisticated as experts. Page’s modest conclusion is that many diverse and good predictors are more successful than just a few excellent predictors.12 Page says in a lecture, “If we don’t get collective wisdom, it’s going to be because either people lack sophistication – that’s the garbage in, garbage out – or they lack diversity.” He adds that people need not just diverse information but diverse and good “models” or methods to interpret that information.13 He writes: “For democracy to work, people need good predictive models. And often, the problems may be too difficult or too complex for that to be the case.”14 This point is crucial because comprehensive empirical evidence collected over many decades shows that most voting citizens are unsophisticated and extensively ignorant. I will not belabor this point because I presume it is well-known now. In general, voters know very little basic political information, such as who their representatives are, what their representatives did in office, what they can do, what they propose to do, or the expected effects of their proposals. It’s not as though getting your neighbors to deliberate about politics is like getting a physicist, engineer,

26  Jason Brennan and chemist to discuss satellite repair. It’s more like getting three middle scholars to discuss how to build a quantum computer. Note another problematic behavioral issue when applying the theorem: The theorem is meant to defend cognitive diversity in group decision-­ making. As we saw above, the model involves problem-solvers trying to solve a singular problem with different methods. Still, they defer to others who use a different approach to improve the group’s current decision upon getting stuck. So, properly applied, this involves well-functioning group deliberation and decision-making by review. Think of a committee meeting going well. It does not apply to aggregative voting as done in elections. In an election, people show up and vote. They are not stuck on that date while trying to solve a problem and then reacting smartly to others’ superior methods. To apply this theorem to aggregative democracy and mass elections, we would need to show that when deciding how to vote, voters deliberate with one another, respond to each other’s methods, and so on in the right way. That is far from obvious. Indeed, it is likely to be false. This is another place where the theorem to defend democracy breaks down significantly. Let’s take a closer look at people’s behavior inside a democracy. In particular, consider two significant theories of voters voting and what they try to do when they vote. What we might call the “popular sovereignty” view holds that voters behave as follows: First, voters have a sense of their values and concerns. Second, they learn about how the world works, what politics can and cannot do, and what policies achieve what ends. Based on these values and information, they form an ideology or at least some set of somewhat coherent policy preferences. Here, their policy preferences reflect their goals, in the sense that they support those policies because they think implementing those policies will produce desired outcomes and goals. Third, they examine the candidates and parties on offer and vote for them based on shared ideology and policy preferences. Since everyone is doing this, the winning parties will thus tend to share the ideology or policy preferences of a large population segment. Indeed, to win, they must promote ideas that voters like. Fourth, as a result, the parties will implement the policies the voters want. Finally, the fifth, come the next election, voters will punish bad performance and reward good performance. On this model of voter behavior, elections convert popular opinion into policy. Elections are a means by which the government is made to do what many people – especially middle-ground voters – want the government to do. However, a large body of empirical work in political science challenges the claim that most voters behave that way. Instead, we have strong evidence that politics is not about policy for most citizens. Consider an illustrative example: Imagine you live and work in an environment where everyone says they believe that aliens created the Washington Monument. The belief is absurd, but everyone around you

Sexy but Wrong 27 thinks it. You cannot effortlessly move or work elsewhere. If you do not share this belief, they will tend to avoid you, ignore you, mock you, and mistreat you. So, if you want to form business partnerships, get a job, have romantic relationships, find friends, join clubs, and so on, you need to express that you share that belief. Further, suppose that people who exhibit strong credence in that belief – by wearing T-shirts or putting bumper stickers on the car – tend to get higher status. In this environment, you might well come to hold or express that belief because you want to consume the social benefits and avoid the social costs of nonconformity. The point of this example is that we can have social incentives to believe things or at least act as we believe them. Our reasons for belief are not always about tracking the truth. Now consider a popular sociological theory of religion. People face a problem of collective action and distrust. We need to cooperate to succeed, but we also know that others have the power and sometimes the incentive to free-ride on group efforts, renege on contracts, break the word, or take advantage of us. What we want, then, is some reliable signal that people are committed to the group’s welfare and to the norms that make cooperation possible. Shared religion partially solves this problem as religions require adherents to exhibit some mix of either expensive belief (adherents must claim to accept bizarre claims about supernatural metaphysics) or costly behaviors (devotees must participate in expensive rituals, avoid particular fun behaviors, modify their bodies, and so on). This works. A common religion facilitates trust inside the group, though it can sometimes demote trust between groups. To understand the value of signaling, consider non-doxastic forms that serve the same purpose. MS-13 gang members need to be able to work together and trust one another. But they face a problem: all of their members are violent criminals. It needs some mechanism to facilitate trust and cooperation. It needs to ensure members attend meetings, do their jobs inside the gang, contribute to the gang’s club goods, and do not steal from the unit itself. MS-13 solves the trust problem by requiring members to do horrible things – this screens out the insincere and uncommitted. It also requires them to display face tattoos and other external signals. Getting face tattoos is strong evidence one is committed to the gang. Further, because outsiders will no longer trust them, members have even stronger incentives to remain steadfast and in good standing inside their group. Each thinks, “I need to keep trust with the gang because no one else will trust me.” It works. Similar remarks apply to sports fandom.15 Sure, many people find sports fun to play and watch. But sports fandom also serves a signaling function. Wearing team clothing and loudly parroting absurd beliefs – such as that an apparent out was safe – signals commitment to one’s town. By publicly engaging in sports rituals and affirming silly beliefs, fans find it easier to make friends, make deals, and gain each other’s trust.

28  Jason Brennan A great deal of empirical evidence shows that for most citizens, picking a political party, voting for a candidate, and even affirming political beliefs is about trying to signal to other members of their identity group and their in-groups that they are good and loyal members of that group. As Kwame Anthony Appiah says, “People don’t vote for what they want. They vote for who they are.”16 For most voters, politics is less like picking a plumber to fix a pipe than waving a sports team’s banner. It’s less like trying to score baskets and more like wearing Air Jordans to impress others. For most Democratic voters, voting Democrat is the same thing as getting an MS-13 tattoo as a Southern evangelical claiming the Earth was created 6000 years ago is the same as an Eagles fan painting his face green and shouting obscenities. Because individual votes do not matter, individual voters are liberated and incentivized to use their political beliefs and behavior to promote non-political ends. They vote for other purposes. Now consider some bizarre empirical findings about voters: Most are poorly informed. They do not know what their preferred party has done in the past, stands for now, or plans to do in the future.17 Most voters are ideologically innocent; they do not have stable political beliefs and do not have much in the way of policy preferences, period.18 Most voters who appear to have fundamental ideologies do not; instead, they will parrot whatever their party is saying today. If their party changes policies, they will immediately switch to defending those without awareness that they have changed.19 For instance, before Trump, Republicans were hawkish and distrustful toward Russia, and Democrats were dovish, but after Trump, Republicans switched to having soft attitudes, and Democrats became hardline and conspiratorial. As Achen and Bartels say, one might think Americans have pretty “ingrained” views about Russia after the past century, but they don’t.20 Voters cluster their political beliefs around whatever the party happens to endorse. As psychologist Dan Kahan says, Whether humans are heating the Earth and concealed-carry laws increase crime, turn on wholly distinct bodies of evidence. There is no logical reason for positions on these two empirical issues – not to mention myriad others, including the safety of underground nuclear-­ waste disposal, the deterrent impact of the death penalty, the efficacy of invasive forms of surveillance to combat terrorism to cluster at all, much less form packages of beliefs that so firmly unite citizens of one set of outlooks and divide those of opposing ones. However, there is a psychological explanation […] That explanation is politically motivated reasoning.21 These issues are logically independent, yet if you take a stance on one point, we can predict what perspective you have on all the others.

Sexy but Wrong 29 Thus, it seems that people adopt whatever their party’s view is. They advocate policies because their party supports them; they do not choose a party because they share its policy ideas. Further, we have extensive evidence that citizens generally engage in motivated reasoning. They avoid, evade, and ignore evidence that contradicts whatever political beliefs they currently profess but then actively seek out evidence that reinforces those beliefs.22 These behaviors are hard to explain if citizens follow the popular sovereignty behavior model. If that model is correct, citizens are voting to intend to get the government to implement policies that will promote their goals and ends. The findings on ignorance and irrationality suggest that they are not very good at it. But the results of ideological innocence, parroting the party – including by switching positions overnight – make little sense, period, on that model. However, if citizens use politics for social purposes – as a form of social signaling meant to prove their commitment to their group – these behaviors make sense. If they attach themselves to parties for the social benefits of doing so, their ignorance, tribalism, and irrationality help them. Consider the sports or religion analogies again: Being a sports hooligan enables you to keep and make friends because it shows locals you are one of them. Firm commitments to religious rituals and beliefs do the same – it is a way of proving pro-sociality. Politics seems to function the same way. Indeed, voters are free to use politics this way because their votes do not matter much. The chances of a ballot breaking a tie or having any other effect on political outcomes are small. Political scientists and economists have long argued that this explains why voters are ignorant and irrational; they lack the incentive to behave better. But while this seems true, it also appears not fully understand the implications of voters’ perverse incentives. Because their votes do not matter, they are not only liberated to be ignorant or biased. Instead, it makes little sense for them to use their votes for the goal of promoting outcomes through policy, period. Instead, it makes more sense to use their political behavior for non-political purposes, such as signaling a commitment to their identity groups. Suppose the Hong-Page theorem is true (remember, we are ignoring the quality of the proof). In that case, it says that group decision-­making can be made more reliable by increasing the cognitive diversity of its members. But as we saw above, the proof does not posit that increasing cognitive diversity in any group decision always increases group reliability. Instead, the theorem describes group reliability growing because group members work on a common problem, want to solve the problem, and defer to one another’s better methods whenever the group gets stuck at a point where a principal solution exists. What we see when we look at actual voter behavior in elections, though, is that people engage in essentially expressive behavior, are not trying to solve a problem, are not doing much deliberation with one

30  Jason Brennan another, and are primarily innocent of ideology or mostly parrot whatever their thought leaders say today. They do not defer to better methods but instead try to arrive at a standard set of often absurd claims to prove fidelity to the group. There is little methodological diversity because signaling requires they prove loyalty by fixating on the same things. They do not take criticism well from each other and ignore and evade different ideas rather than defer to them. Voters are, in fact, tribalistic, pig-headed, conformist, cognitively biased, unsophisticated, and closeminded. When they vote, they are not trying to solve a problem but rather express their identities or engage in public signaling. If so, the Hong-Page theorem would not apply to such group “decisions.”23

Notes







1. Aristotle, The Politics and the Constitution of Athens, 2nd edition, trans. Stephen Everson, pp. 23–31 (New York: Cambridge University Press), 1285. 2. Lu Hong and Scott Page, “Groups of Diverse Problem Solvers Can Outperform Groups of High-Ability Problem Solvers,” PNAS 101 (2004): 16385–16389. 3. Hélène Landemore, Democratic Reason (Princeton University Press, 2012). 4. Elizabeth Anderson, “The Epistemology of Democracy,” Episteme 3 (2006): 8–22. 5. Scott Page, “Making the Difference: Applying a Logic of Diversity,” Academy of Management Perspectives 21 (2007): 6–20. 6. Hong and Page (2004, 16,386). They describe all decision-makers as trying to map the same “function V that maps a set of solutions into real numbers.” In their computational experiment, they again program all agents with the same value function. On page 16,387, they say, “For our analysis, we assume that all agents have the same v and that v has full support.” Here, “v” refers to the agent’s value function. 7. Hong and Page 2004, 16,387 offer as starting point that “The problem is difficult: no agent can always find the optimal solution.” Page confirms this interpretation is correct in Scott Page, The Difference (Princeton University Press, 2008), 159–165. 8. Hong and Page (2004, 16,387) describe each agent using a single search rule or mapping rule. 9. Hong and Page (2004). 10. Drew Westen, The Political Brain (Perseus Books: New York, United States, 2008), Jonathan Haidt, The Righteous Mind (Penguin Books, London: United Kingdom, 2013), Milton Lodge and Charles Taber, The Rationalizing Voter (Cambridge University Press: Cambridge, United Kingdom, 2013), Christopher Achen and Larry Bartels, Democracy for Realists (Princeton University Press: New Jersey, United States, 2016), Lilliana Mason, Uncivil Agreement (University of Chicago Press: Chicago, United States, 2017). 11. Abigail Thompson, “Does Diversity Trump Ability?,” Notices of the AMS, 69 (2014): 1024–1031. It argues that their computer simulation is flawed and their mathematical proof contains errors. 12. Page (2007, 346–347). Page (2007, 147) says, “The best problem solvers tend to be similar; therefore, a collection of the best problem solvers performs little better than any of them individually. A collection of random but intelligent problem-solvers tends to be diverse. This diversity allows them to be collectively better. Or to put it more provocatively, diversity trumps ability.”

Sexy but Wrong 31











13. Page (2012). 14. Page (2007, 345). 15. Haidt (2012). 16. https://www.washingtonpost.com/outlook/people-dont-vote-for-wantthey-want-they-vote-for-who-they-are/2018/08/30/f b5b7e44-abd7-11e88a0c-70b618c98d3c_story.html 17. Ilya Somin, Democracy and Political Ignorance (Stanford: Stanford University Press, 2013); Christopher Achen and Larry Bartels, Democracy for Realists (Princeton: Princeton University Press, 2016); Michael X. Delli-Carpini and Scott Keeter, What Americans Know about Politics and Why It Matters (New Haven: Yale University Press, 1996); Jason Brennan, Against Democracy (Princeton: Princeton University Press, 2016). 18. Donald Kinder and David Kalmoe, Neither Liberal Nor Conservative: Ideological Innocence in the American Public (Chicago: University of Chicago Press, 2017); Lillian Mason, Uncivil Agreement: How Politics Became Our Identity (Chicago: University of Chicago Press, 2017); Lillian Mason, “Ideologues without Issues: The Polarizing Consequences of Ideological Identities,” Public Opinion Quarterly 82: 280–301; Lilliana Mason and Julie Wronski, “One Tribe to Bind Them All: How Our Social Group Attachments Strengthen Partisanship,” Political Psychology 39 (2018): 257–277; Angus Campbell, Philip E Converse, Warren Miller, and Donald E. Stokes, The American Voter (New York: John Wiley, 1960). 19. Achen and Bartels (2016, 267–296); see also Gabriel S Lenz, “Learning and Opinion Change, Not Priming: Reconsidering the Priming Hypothesis,” American Journal of Political Science 53 (2009): 821–837; Gabriel S. Lenz, Follow the Leader? How Voters Respond to Politician’s Policies and Performance (Chicago: University of Chicago Press, 2012). 20. https://www.vox.com/policy-and-politics/2017/6/1/15515820/donald-trumpdemocracy-brexit-2016-election-europe 21. Dan Kahan, “The Politically Motivated Reasoning Paradigm, Part 1: What Political Motivated Reasoning Is and How to Measure It,” in Emerging Trends in the Social and Behavioral Sciences: An Interdisciplinary, Searchable, And Linkable Resource, 2016, https://onlinelibrary.wiley.com/ doi/abs/10.1002/9781118900772.etrds0417 22. Kahan (2016); Dan, Kahan, Ellen Peters, Eric Cantrell Dawson, and Paul Slovic. “Motivated Numeracy and Enlightened Self-Government,” Behavioral Public Policy 1 (2013): 54–86; Charles Taber and Milton R. Lodge, “Motivated Skepticism in the Evaluation of Political Beliefs,” American Journal of Political Science 50 (2006): 755–769; Drew Westen, Pavel Blagov, Keith Harenski, Clint Kilts, and Stephan Hamann, “The Neural Basis of Motivated Reasoning: An fMRI Study of Emotional Constraints on Political Judgment during the US Presidential Election of 2004,” The Journal of Cognitive Neuroscience 18 (2006): 1947–1958; Bryan Caplan, The Myth of the Rational Voter: Why Democracies Choose Bad Policies (Princeton: Princeton University Press, 2007); Dennis Choong, “Degrees of Rationality in Politics,” in The Oxford Handbook of Political Psychology, eds. David O. Sears and Jack S. Levy, pp. 96–129 (New York: Oxford University Press, 2013); Milton Lodge and Charles Taber, The Rationalizing Voter (New York: Cambridge University Press, 2013). 23. For example, see Jason Brennan, Against Democracy (Princeton University Press, 2016); Ilya Somin, Democracy and Political Ignorance (Stanford University Press, 2016).

2

A Belated Failure Condorcet in Contemporary Epistemic Conditions Hana Samaržija

2.1 Introduction It is outmoded news that democracy has misplaced a substantial portion of its former epistemic appeal, with dozens of publications enumerating its failures and collapses (Brennan 2017; Levitsky and Ziblatt 2019). Instead of governments present for their nations, democratic elections worldwide have generated epistemically irresponsible and unresponsive1 regimes that do not react to civic needs. The final blow to ventures to corroborate the epistemic justification of democracy with empirical facts came in 2016. Then we witnessed the successful referendum for Great Britain to depart from the European Union, an initiative based exclusively on and effortlessly refutable falsities and a misconstrued value of sovereignty as separation, and the United States’ infamous elections when Donald Trump scored a landslide victory against his adversary by relying on the ideological contempt between liberals and conservatives (Cassam 2019). Earlier, the first democratic elections in hitherto communist states did not yield epistemically responsible regimes eager to make the most of the apparent promises of the free market. Instead, they obtained corrupt and nepotist administrations fervid exclusively about manipulating the privatization of social resources. Suppose we wished to defend democracy by appealing to its epistemic virtues rather than just the political value of civic equality. In that case, it seems we would need to make an additional effort. Today’s representative democracies2 with parliamentary and presidential elections are contingent primarily on their voters’ epistemic capacities. However, political science’s findings indicate that electorates are uninformed about politically pertinent issues, ideologically biased, burdened by identity prejudice, and inept at establishing lasting preferences (Delli Carpini and Keeter 1996; Kuklinsky and Peyton 2009). Nonetheless, before any such research, the oldest persuasive quantitative epistemic defense of collective decision-making was Marquis Nicolas de Condorcet’s 1785 theorem, now famous as Condorcet’s Jury Theorem and commonly known as CJT. In the eighteenth DOI: 10.4324/9781003311003-4

A Belated Failure 33 century, Condorcet pioneered a mathematical approach to politics and parliamentarism. In short, CJT stipulates two challenging arguments. First, the outcome of a majority vote of a group of citizens, each of whom is sounder than random at selecting the correct option, is likelier to be epistemically valuable than any individual citizen’s judgment. Second, as the number of such voters nears infinity, the probability that their collective decision will be epistemically valuable approaches certainty (Goodin and Spiekermann 2018). However, CJT exposes its fictitious citizens to three arduous conditions. First, they must be competent. Their likelihood of making the correct choice must surpass randomness or the statistical value of p = 0.5. In the polar scenario, the majority vote of a group of citizens worse than chance would yield a disastrously incorrect decision. Second, Condorcet’s citizens must be independent. This kind of autonomy means that their decisions must be founded exclusively on their critical exploration of the available evidence rather than by acquiescing with other citizens or reflecting the choice of some dubitable authority. Third, they must be sincere. Condorcet’s genre of sincerity requires that voters select that option they genuinely deem epistemically and ethically best instead of voting for the lesser evil or a clientelist option that might bring them material gain. CJT’s unpretentious mathematical equation has endured centuries of philosophical progress and has become the quantitative cornerstone of democracy’s epistemic justification. The second alluring mathematical theorem favoring democracy arrived in 2004. It acquired intellectual fame as the Hong-Page theorem of diversity over ability. The theorem attempted to confirm that a group of citizens with diverse epistemic capacities and background knowledge would collectively produce an epistemically higher-quality decision than a single expert (Hong and Page 2004). Tragically, dutiful academic mathematicians contested its trivial proof, the equation’s impropriety for the desired conclusion, and the tendentious usage of lousy mathematics in the social sciences (Thompson 2014: 1024). On the other hand, CJT is still indisputably correct, so why it does not function in empirical life – as multimillion electorates seldom select the most ethical and epistemically responsible option – must be sought elsewhere. This chapter will argue that Condorcet’s requirements of competence, independence, and sincerity do not hold in the present epistemic conditions of the culture of ignorance, digital media’s obfuscating informational environment, epistemic bubbles and echo chambers, ideological polarization, and strategic voting for clientelist political options. In Section 2.2, I review more than four decades of political behavior studies and contemporary political epistemology, showing that average citizens are apathetic toward political topics, poorly acquainted with their representatives in administrative bodies, impassive about their country’s constitutional principles, and unable to maintain stable ideological

34  Hana Samaržija preferences. The tiny minority of citizens knowledgeable about politics, having absorbed substantial ideological beliefs, tends to perform complex cognitive strategies to retain their present preferences in the face of opposite information and exhibits ample myside bias (Stanovich 2021; Taber and Lodge 2006). It is symptomatic to mention that the same survey of civic knowledge about politics that political epistemology considers bitterly defeatist (Ahlstrom-Vij 2020: 406), Delli Carpini and Keeter’s study of the political ability of American citizens, is regarded as one of the more optimistic collections of empirical data in political behavior studies. From the perspective of political scientists, the finding that the best-informed fifth of citizens possesses only mediocre knowledge of politics muddles earlier and even more pessimistic projects (Kuklinsky and Peyton 2017: 55). In the eyes of the political epistemologist, the average citizen was always far removed from Condorcet’s ideal of the engaged citizen whose reflections are more reasonable than random. It is worth mentioning that Delli Carpini, writing alone, repeated their survey in 2005. He came across similar or worsened results, mainly due to the altered epistemic conditions of baffling new information sources (Delli Carpini 2005). Regardless, the competence requirement crumbles exclusively if all citizens err similarly so that their mistakes do not cancel each other out (Goodin and Spiekermann 2018: 4). Luckily for those renouncing CJT, the cultural prejudice we mechanically perpetuate guarantees that everyone will stumble in an equal way (Fricker 2006, 2007; Samaržija and Cerovac 2021). In Section 2.3, I assess the independence requirement. It will become evident that CJT’s independence cannot outlast decades of psychological work on the common knowledge effect. This tendency implies that the prevalent attitude in a group’s deliberation will not be the stance best supported by evidence. Instead, it will be the one most discussants hold before the debate begins (Gigone and Hastie 1993). Although it is doubtful whether there are hypothetical affairs where the independence requirement would be sustainable, it is especially endangered in the epistemic environment of social networks and algorithmic information sorting, which yields both epistemic bubbles and malign echo chambers (Nguyen 2018). Epistemic bubbles and echo chambers are two distinct social epistemic filters with dramatically different effects on our intellectual lives. For instance, although epistemic bubbles stem from restricted informational milieus, they do not prevent us from considering alternatives or accepting opposing attitudes. Contrarily, echo chambers are political pundits and ideological media’s active intention to ensnare their followers in a reality of alternative facts. In his seminal article, Nguyen suggested that echo chambers are problematic because their followers start perceiving their ideological adversaries as ethically defective (Nguyen 2018). I have added that the analysis of the extreme right’s communicational methods Nguyen bases his argument on also entails members

A Belated Failure 35 of echo chambers to commence perceiving strangers as epistemically deficient, naïve, and inclined to fall for liberal politicians’ superficial dishonesties (Samaržija 2023). Finally, the essential obstacle to the independence requirement is our general epistemic dependence, as we seldom receive our knowledge on our own. Our intellectual growth depends primarily on learning from others or interacting with society (Goldman 1990). Condorcet’s ideal of the citizen who acquires and elaborates on their political preferences without interacting with others emerges as an unattainable thought experiment. Section 2.4 tackles the sincerity requirement, which seems the least empirically suspect of CJT’s three normative demands. Even if they are neither competent nor independent, it appears as if citizens vote for the options that they believe will successfully handle urgent social problems, even if their choices are wrong. Nonetheless, research has established that voters will continue supporting candidates they perceive as deficient if they are emotionally attached to having once selected them (Swire-Thompson et al. 2019). Finally, loathing over party lines is so emotionally obliging that voters will continue reinforcing their select party even if they concede it is not the epistemically optimal political choice (Iyengar and Westwood 2014). Hence, even the most trivial requirement of Condorcet’s Jury Theorem cannot outlive the brutality of empirical facts. Finally, Section 2.5 evaluates the implications of the fact that collective decision-making and, thus, democracy may be epistemically indefensible. Before proceeding to a detailed rebuttal of CJT’s conditions, it is worth noting that repudiating the competence provision would render the theorem empirically nonfunctional. As we have already sketched, the soaring curve of the probability of reaching a perfectly correct decision by compiling voters sounder than random has its darker side. The sum of votes cast by people less competent than random will result in a descending curve leading to an entirely wrong democratic decision (Goodin and Spiekermann 2018: 51). Nonetheless, our goal here is more ambitious. By establishing that none of CJT’s normative requirements can survive the current epistemic conditions, we completely annul it as a quantitative defense of democracy. This finding enables us to open the space for discussions about different political models of creating a just, wise, and egalitarian society.

2.2  Condorcet’s Competence Requirement I will begin my analysis with the first requirement that ostensibly warrants CJT’s empirical accuracy, the competence prerequisite. In short, competence demands that citizens cast their votes more precisely than random or the statistical value of p = 0.5. The aggregated voices of such minimally competent voters steer their decision’s correctness to immaculate

36  Hana Samaržija certainty. Conversely, compiling people’s votes worse than random would yield an incorrect conclusion. The pronounced question is whether empirical voters can meet this ambitious requirement. Regardless of empirical studies of political behavior, apprehensions about civic epistemic capacities are as old as the foremost debates about epistemically optimal political systems. Probably the first and the most famous epistemic pessimist about democracy was Plato. In his Republic, he advocated the expert rule of a philosopher-king cognizant of the idea of good governance. John Stuart Mill partly inconsistently voiced similar doubts. Although he first claimed that even the most detrimental attitudes should circulate so that the free market would surface the epistemically most valuable beliefs, his discussion about parliamentary democracy urged for plural votes for epistemically better equipped voters (Mill 2007). In empirical data, the average voter’s ignorance is “one of the best-­ documented facts of modern politics” (Bartels 1996: 194). Bartels speaks exclusively about American citizens, the objects of his analysis of civic knowledge of politics before presidential elections. Nonetheless, he is also talking about American citizens before the advent of social media, which partially explains why today’s studies about the civic knowledge of politics are even more devastating. From his perspective of the late nineties, Bartels notes that “the greatest contribution of more than half a century of research in political science is documenting how the average citizen is a woeful approximation of the classic ideal of the informed democratic citizenry” (Bartels 1996: 195). To approach this ideal, the said citizen would have to be at least reasonably acquainted with the structure of their state – for instance, whether it is presidential, semi-­ presidential, or parliamentary, and how often elections are held – and lobby for their interests. However, decades of research have generated an image of the voter so ambivalent toward politics that they can scarcely cast a vote every four or five years and do so without knowing what they are choosing. The average citizen does not understand the symbolic division between the left and the right and does not care what it represents, so they are incapable of ideological self-identification and unchanging political preferences (Converse 1964). Suppose journalists interview them about their political values. In that case, they will promptly answer and then forget or replace those values with others (Zaller 1992). To make matters worse, that minuscule minority of citizens interested in politics tends to become so ideologically fraught that they will perform convoluted cognitive escapades to maintain their beliefs in the face of dissident evidence (Taber and Lodge 2006). 2.2.1  Civic Political Sophistication Bartels speaks about political scientists’ endeavors that officially began with Converse, who in 1964 published a seminal essay about the

A Belated Failure 37 American voters’ political knowledge (Converse 1964). In his original work, Converse relied on the concept of a “political belief system,” a steady ideological structure that enables citizens to identify with political parties, notable individuals, and pertinent policies. An ideological belief system is represented by the metaphorical divide between the progressive left and the conservative right (Converse 1990, 2000). By surveying American citizens with an assortment of open and closed-ended questions in 1956, 1958, and 1960, Converse concluded that a mere tenth of the electorate – twelve percent of them – genuinely manages to locate their values between the left and the right. Converse also examined whether they consistently exhibit progressive or conservative political attitudes, allowing the surveyed agents to evade explicit political self-identification that would block them from potentially valuable options and policies. Predictably, the answer was negative. In addition, examinees could not locate individual policies on the left and right ideological landscape. They casually selected mixtures of progressive and reactionary ­policies that canceled each other. Converse’s research’s essential upshot is that citizens do not understand what is liberal and conservative, making them both politically unsophisticated and challenging to inform. While, therefore, nine-tenths of the electorate is utterly ambivalent toward politics and its effects on their everyday lives, which they do not wish to engage with, twelve percent of citizens are remarkably interested, ideologically polarized, and politically literate. From today’s perspective, Converse’s study’s essential value is that he did not only measure political informedness, which may or may not be relevant for reliable civic conduct. Suppose someone, for instance, knows the name of their prime minister and all ministers. In that case, it does not automatically render them able to accurately assess their mandate’s ethical and epistemic successes. Converse measured citizens’ political sophistication: the capacity to grasp the political spectrum and employ this knowledge to form lasting political beliefs. Five decades later, political scientists have continued surveying civic political sophistication and informedness and have yielded results like Converse’s. Nevertheless, political scientists now mention Converse alongside three leading revisions of his conclusions: “the downbeat, the really downbeat, and the upbeat revision” (Kuklinsky and Peyton 2007: 46). Here, it is worth reiterating that the supposedly upbeat revision of Converse’s work is the survey by Delli Carpini and Keeter that political epistemologists deem a pessimistic epistemic disaster (Delli Carpini and Keeter 1996). Namely, political scientists have recognized that half the electorate can answer every other politically pertinent question as a promising result that renews faith in democracy. To understand how that could encourage optimism, we must first deal with the downbeat and the cynical revisions. As Kuklinsky and Peyton correctly underline, Converse did not accompany his empirical finding that only twelve

38  Hana Samaržija percent of citizens understand the landscape between the left and the right with a normative attitude that this minority of knowledgeable citizens should become the bearers of democracy (Kuklinsky and Peyton 2007: 49). Still, his findings entail fruitful implications for the sustainability of today’s form of democracy. Four decades after Converse’s seminal study, Taber and Lodge decided to deal with those twelve percent of particularly politically sophisticated citizens whom Converse had described as ideologues or partial ideologues (Taber and Lodge 2006). It concerned them whether their understanding of the ideological spectrum is beneficial or disadvantageous in democratic participation. By exploring their motives from the perspective of the budding branch of political psychology, the authors concluded that ideologically literate epistemic agents are at the same time excessively ideologically laden. Such citizens, restricted to their solidified stances, lose the capacity to select an option that is perhaps epistemically more advantageous but incompatible with their calcified ideological self-identification. Simply put, their political sophistication enables them to reverse contrary arguments, advocate alternative facts, and relativize data to preserve their beliefs in the face of new information. As social epistemology explored simultaneously, Taber and Lodge discovered that examinees who consider themselves thorough leftists or conservatives automatically accept arguments consistent with their current stances while employing resources to oppugn the legitimacy of conflicting information. In their words, “intensely partisan and politically astute respondents show an especially strong proclivity to rely on these processes” (Kuklinsky and Peyton 2007: 50). Although, according to Taber and Lodge, we might justify a mild dose of skepticism exhibited by an outstandingly knowledgeable agent, “skepticism becomes bias when it becomes unreasonably resistant to change and especially when it leads one to avoid information” (Taber and Lodge 2006: 22). So, if we link Converse to his younger colleagues, nine-tenths of citizens do not know enough about politics to make responsible decisions. That remaining erudite tenth reduces their knowledge to extreme ideological partiality and obedience to their chosen party. However, Zaller’s “really downbeat” revision proffered an even darker image of the average citizen’s epistemic capacities (Zaller 1992). By analyzing data collected in pre-election polls and articles published in the New York Times, Zaller showed that civic responses to surveys are not the product of stable beliefs and values but merely the first thing that pops into their minds after reading the daily news. Referring to Zaller, Bartels differentiated attitudes and preferences (Bartels 2003). According to Bartels, voters may have short-lived and associative attitudes, vulnerable to change, but not lasting and evidentially supported preferences. Far more dramatic in tone than Converse and Zaller, Bartels concluded that “popular rule is impossible but (…) citizens can exercise an intermittent, sometimes

A Belated Failure 39 random, and even perverse popular veto on the machinations of political elites” (Bartels 2002: 74). 2.2.2  Civic Political Informedness According to Kuklinsky and Peyton’s instructive review, we should seek these dismal numbers’ more hopeful counterparts in Converse’s work upbeat revision, Delli Carpini and Keeter’s (1996) survey of civic political informedness (Delli Carpini and Keeter 1996). Unlike Converse, who focused on the civic understanding of the ideological spectrum, Delli Carpini and Keeter tackled the interplay of political informedness and sophistication, so they also questioned citizens with quizzical queries about notable politicians and the rules of the game. Their inquiries encompassed a spectrum of knowledge about democratic institutions, issues, and procedures. For instance, they asked their examinees “how a bill becomes a law, or what rights are guaranteed by the US Constitution, (…) whether there is a federal budget deficit or surplus, or the percentage of Americans living in poverty” (Delli Carpini 2005: 29). The researchers, combining their results with data collected within five decades of pre-election polls, reported that “more than a small portion” of American citizens are moderately well acquainted with present political issues and socioeconomic indicators (Delli Carpini and Keeter 1996: 269). The exact value of this more significant portion, which seems unlike Converse’s twelve percent, has been the target of social epistemologist Ahlstrom-Vij’s sardonic commentary for over twenty years. In his review of the epistemic justification of democracy, Ahlstrom-Vij concludes that less than half the citizenry answering questions necessary for partaking in democratic practice is depressing (Ahlstrom-Vij 2020: 406). Likewise, according to Delli Carpini and Keeter, “many of the facts known by small percentages of the public seem critical for understanding – let alone acting in – the political world” (Delli Carpini and Keeter 1996: 101–102). Delli Carpini returned to their original research in 2005 and noted that studies of political sophistication and informedness keep reinforcing the image of an average voter as “woefully uninformed about political institutions and processes, substantive policies and socioeconomic conditions, and important political actors such as elected officials and political parties” (Delli Carpini 2005: 28). Finally, regardless of whether we prefer Converse’s initial research or its pessimistic and optimistic revisions, we will ultimately conclude that most citizens do not know the facts essential for orienting within the world of politics and cannot grasp the spatial metaphor of the left and the right. Moreover, they cannot position themselves ideologically and reduce their potential political sophistication by performing cognitive stunts to avoid the burden of changing their opinion. Nevertheless, for Condorcet’s competence requirement to crumble, all citizens must err

40  Hana Samaržija similarly. Otherwise, their mistakes might cancel each other out (Goodin and Spiekermann 2018). Luckily for rivals of justifying democracy with jury theorems, behavioral economists have spent decades enumerating the prejudices and heuristics we all use when evaluating data (Kahneman 2012). The concept of epistemic injustice signals another avenue of similar or same mistakes: due to implicit or implicit prejudice toward epistemic agents with marginalized identities – women and ethnic and sexual minorities – we often underestimate their credibility and cognitive capacities (Fricker 2006, 2007; Samaržija and Cerovac 2021). Since the ascribed lack of credibility to marginalized groups is now minutely documented, we can assume that most voters would, when assessing minority politicians or policies related to their rights, get it wrong the same way. Moreover, as Condorcet taught us, the aggregated votes of citizens wronger than random inevitably lead to an incorrect democratic decision. In the following section, I will show that genuine citizens do not only make countless mistakes but do so according to predictable and matching patterns.

2.3  Condorcet’s Independence Requirement First, we must establish what political independence would presume to disprove Condorcet’s independence requirement. According to Goodin and Spiekermann’s recent interpretation, Condorcet’s independent epistemic agents do not lean on “the same opinion leader, the same shared ideology or prejudice, the same shared psychological mechanisms, the same shared cues, on the same more fundamental shared properties, and on the same shared evidence, background information, or theories” (Goodin and Spiekermann 2018: 55). This inventory leads to the intuitive inference that there are no people who, when making political decisions, do not depend on “the same shared psychological mechanisms,” “the same background information,” or “the same shared ideology.” It is difficult to imagine an epistemic agent whose reasoning does not involve human cognitive capacities or who does not share background beliefs with concurring peers. Nevertheless, a comprehensive rebuttal of the independence provision will necessitate something more than mere intuition or our creative capabilities. 2.3.1  The Common Knowledge Effect The first empirically founded obstacle to the independence requirement is the psychological common knowledge effect. Since the nineties, the social sciences have acquiesced about the common knowledge effect’s epistemic consequences. During the deliberation of a group of epistemic agents, the group will not concur about the truest or the most empirically supported belief but about the view that most participants held

A Belated Failure 41 before the debate had even started (Gigone and Hastie 1993: 959). In their authentic work, the authors remarked that the discussants failed to realize that their arguments ended with consensus about the attitude that enjoyed the most support from the beginning. This fact is the most integral hindrance to the epistemic justification of the once fashionable concept of deliberative democracy. Deliberative democracy is not based on today’s political representation but on the discussions of interested and politically engaged citizens (Bohman 1998). As compelling as this concept might sound in theory, deliberation’s empirical fiascos to make epistemically valuable decisions in representative democracy’s deliberative bodies led to a gradual waning in interest in the concept. Today we can ascribe similar trendiness to the notions of epistemic (Cerovac 2021) and open (Landemore 2020) democracy. We establish deliberative bodies from the assumption that groups will make better decisions than individuals as they have compiled information from diverse sources. Gigone and Hastie tested such debates’ efficacy in the real world. Their group, tasked with deciding upon psychology students’ final grades, did not share information given to each debater but finally agreed on the judgments suggested to the highest number of participants before the experiment began (Gigone and Hastie 1993: 961). Suppose we generalize this finding to realistic voters’ collective political decision-making. In that case, discussions with others do not leave us independent. Instead, we consent to the stance ardently endorsed by most discussants. Given these conclusions, it is evident – if deliberative bodies automatically reduce to assemblies of blind sycophants – why the notion of deliberative democracy had progressively lost its philosophical charisma. Although the reasons behind such futility of information sharing are numerous, we might want to consider Sunstein’s suggestion that epistemic agents with unpopular stances are uneasy with sharing their knowledge with the group out of fear that their beliefs might render them ostracized (Sunstein 2015). 2.3.2  Epistemic Bubbles and Echo Chambers On the other hand, the independence of Condorcet’s agents is now imperiled by far more tempting hurdles than the common knowledge effect. According to Nguyen’s hugely popular analysis, “something in the flow of information has gone awry” (Nguyen 2018). Epistemic agents are exposed to radically filtered information that reinforces their current beliefs, regardless of how irrational they might be. Friends who share our values and social networks now foster two kinds of social epistemic structures that we must distinguish due to the divergence of their consequences. First, we can find ourselves in an epistemic bubble due to association with people who ideologically fit us and social networks’ often condemned tendency to offer us content equivalent to our previously

42  Hana Samaržija displayed preferences. In this social epistemic structure, outside voices are accidentally excluded due to omission. Within an epistemic bubble, due to the constant reiteration of the same information, we commence thinking those are the sole facts possessing any legitimacy. This effect, officially known as bootstrapped corroboration, is manifestly logically worthless. As Wittgenstein cynically remarked, if we spot the same title on every copy of some daily newspaper, the repetition of that header does not render it any more valid. A restatement would have to comprise additional evidence or an alternative form of supplementary epistemic value to make a claim more accurate. In the real world, it habitually happens that our friends, media we trust, and algorithmic sorting derive their information from the same sources, so we erroneously perceive the same data’s recurrence as additional evidence in favor of its truthfulness. Nevertheless, epistemic bubbles are consoling as we do not feel automatic antagonism against contrary information. Let us employ a simple example. Imagine an epistemic agent sincerely unsure about the safety of rapidly produced vaccines and whose friends and acquaintances exhibit similar concerns on social networks. After noticing dozens of recurrences of the same arguable article about the connection between quickly generated vaccines and lethal blood clots in vulnerable patients, our epistemic agent accepts these identical repetitions as added proof against vaccines. However, when scrolling down a reliable news site they seldom visit, our agent runs into a prominent doctor’s lecture about the safety of the newest inoculations. As they are not wary of dissimilar information, watching the speech alleviates our agent’s concerns about the safety of recent vaccines. In short, we can burst epistemic bubbles by exposing epistemic agents to impartial and dependable sources of scientific and non-scientific information (Samaržija 2023). In today’s epistemic conditions of constant virtual communication and exposure to various algorithms, we all belong to a lower or higher number of epistemic bubbles. Still, we can endeavor to rupture our pseudoscientific or prejudicial beliefs by exposing ourselves to impartial information sources. On the other hand, echo chambers are far more resilient. According to the leading definition, echo chambers are social epistemic structures where outside voices are deliberately excluded and depicted as ethically and epistemically defective (Nguyen 2018; Samaržija 2023). In his work, Nguyen holds that ideological leaders persuade their followers that their foes are morally questionable – dishonest and unreliable – and thus unworthy of their trust. Additionally, I have appended that, within an echo chamber’s narrative, ideological adversaries are illustrated as epistemically defective, regardless of whether it is due to their inherent stupidity or mere naivete (Samaržija 2023). Echo chambers’ established title reflects their central feature. All we can hear is the echo of our own and concurring agents’ attitudes. Unlike epistemic bubbles, which are the

A Belated Failure 43 random byproduct of our ideologically close friends and the functioning of social media algorithms, echo chambers are influential politicians’ or news sites’ purposive effort to ensnare their followers in a reality of alternative facts. Echo chambers lean on an epistemic effect formally known as preemptive evidencing. In simpler terms, ideological trailblazers and biased information sources preemptively convince their followers that the outside world wants to dispute their subversive truth out of malign intents or inherent idiocy. When the outside world genuinely does deny their reality, as it is empirically unsupported, biased, interwoven with prejudice, or oppressive toward disadvantaged social groups, agents entrapped within an echo chamber will perceive it as a verification of its narrative. After all, their ideological pundit had presaged them it would happen. Of course, although echo chambers are more frequent on the right, they can and do appear in liberal, green, and left-wing political communities. So, while bootstrapped corroboration makes epistemic agents believe that related information sources are unrelated and comprise additional epistemic value, members of an echo chamber are persuaded that unrelated information sources are associated with a complex conspiracy. While more radical echo chambers encompass cults and conspiracy theorists, we can find similar behavioral patterns in groups gathered around particular political ideas, lifestyles, or social goals. Let us revisit our past example of the vaccine skeptic whose antagonism toward inoculation is alleviated by interaction with a reliable contrary opinion. We can reflect upon what such a scenario would look like if we substituted the epistemic bubble with an echo chamber. Again, let us envision an epistemic agent apprehensive about hastily produced vaccines’ safety. However, this time our epistemic agent believes we are dealing with a conspiracy of global pharmaceutical industries, so patients would have to purchase costly drugs for ailments caused by injections. Our epistemic agent shares this conviction with intimate groups of virtual sycophants. At the same time, their ideological leader explicitly cautions them that their mainstream clinicians will endeavor to dissuade them but that such efforts are unworthy of their trust. Within their echo chamber’s structure, medicinal laypersons who have faith in vaccines are depicted as naïve morons incapable of penetrating the veil of fabrications woven by the pharmacological industry. In this case, our epistemic agent cannot exit their echo chamber by encountering impartial scientific facts, as they preventatively consider them a fallacy, nor by conversing with genuine medicinal experts as they perceive them as epistemically defective. Most social and political epistemologists working on echo chambers now agree that it is, in theory, only conceivable to exit such robust epistemic structures by renewing your trust in voices outside our subservient group and that such instances have seldom been empirically verified (Nguyen 2018; Samaržija 2023). Of course, although our examples of

44  Hana Samaržija epistemic bubbles and echo chambers dealt with inoculation and pharmacology, the political polarization distinctive of today’s partisan representative democracies is fecund soil for such epistemic structures (Boutyline and Willer 2016). Perhaps the most suggestive instances of political echo chambers come from the United States’ bipartisan presidential democracy, where political philosophers and political scientists have recorded conspicuous fear and loathing across party lines (Iyengar and Westwood 2014). In a related article, Nelson and Webster argue that epistemic bubbles are fictional, as empirical followers of digital media visit a broad spectrum of news sites and social networks, including those they fervently oppose. Nonetheless, such dissimilar reading material does not affect their political preferences (Nelson and Webster 2017). Still, even if epistemic bubbles are genuinely contentious, as such a media setting would rupture them, miscellaneous information sources are not a menace to echo chambers founded on profound distrust. Unlike Nelson and Webster, Sunstein has spent over a decade authenticating epistemic bubbles and portraying them as one of the causes of political polarization. This epistemic effect arises when agents endorsing some attitudes become highly sure of it (Sunstein 2011). Suppose we supplement the frequency of epistemic bubbles and echo chambers’ empirical support with the abovementioned phenomenon of epistemic injustice when agents of marginalized identities are deprived of the credibility their aptitudes deserve. It is justified to assume that citizens of modern representative democracies look nothing like Condorcet’s independent voters (Fricker 2006, 2007; Samaržija and Cerovac 2021). Quite the reverse, they employ the same heuristics, exhibit the same prejudice, passionately depend on the same ideological leaders, and frequently find themselves in the sturdier or the less sturdy structures of echo chambers and epistemic bubbles.

2.4  Condorcet’s Sincerity Requirement Finally, we are left with Condorcet’s sincerity prerequisite. Condorcet postulated that his competent and independent voters would also sincerely elect the political option they deem ethically and epistemically optimal. On the surface, this provision appears the most unsophisticated but, simultaneously, the most challenging to repudiate. Do epistemic agents not elect political options they genuinely regard as ethically and epistemically the finest choice? We could presume that advocates of all political options hold they shield them from some more significant peril, such as the advent of communist or nationalist terror or the gradual rise of some profoundly deplorable party. As for the electorate’s actual sincerity, sociological, psychological, and political studies of voting behavior disclose persistent patterns of pragmatic and insincere voting for the lesser evil, despite being cognizant of their epistemic and ethical defects,

A Belated Failure 45 we anticipate parties we are affectively bound to acquire us financial gain (Ilišin et al. 2018; Swire-Thompson et al. 2019). Regarding the question of voting for the lesser evil, Condorcet’s sincerity would have agents choose a politically hopeless option they deem ethically and epistemically best rather than calculate between two prominent parties likely to win the race. In a study cynically titled “They Might be a Liar, But They are My Liar,” Swire-Thompson and colleagues (2019: 1) tested whether an epistemic agent’s sentiments toward politicians and the likelihood of reelecting them would alter once they comprehended the candidates had purposely fed them disinformation. The authors oriented their examinees toward false information and rectified them. They then chronicled whether the surveyed agents would cease trusting their preferred candidate once they recognized they had defrauded them. Fortunately for our present argument, the outcome was negative. Even once they had cognized that they believed disinformation and accepted the amended fact, which is a considerable mental effort, the examinees remained equally eager to vote for their hitherto favored political option (Swire-Thompson et al. 2019: 6). Thus, although they had conceded that their political option does not represent the epistemic ideal and would perhaps not make decisions of pristine epistemic quality, the voters continued choosing it. Such a manifestly erroneous choice empirically invalidates Condorcet’s sincerity principle. However, we do not have to constrain ourselves to such contemporary research, fraught with the epistemic environment of partisan media and social networks. For instance, an alternative analysis of the electorate also points toward opting for a party whose epistemic privations we are more than aware of (Jost et al. 2004). According to Jost and his collaborators, voters ascribed the presiding party far more credibility than it had objectively merited out of mere inertia and dread of change. Such unconscious bolstering of the epistemically problematic status quo is best documented among marginalized social groups of poor economic standing, whose adaptive preferences render them even more enthusiastic about excusing their current political representatives (Jost et al. 2004: 908). Regarding insincere voting founded on mere material pragmatism, there is a well of pertinent information to be found in Ilišin and colleagues’ recent analysis of the Croatian youth’s3 political preferences. In the research, the number of examinees who reported affiliation with some political party statistically significantly surpassed those who stated that, in the Croatian political arena, there is a party whose ethical and epistemic character they sincerely trust (Ilišin et al. 2018: 64). As Croatia is, like many other post-communist states, still ascertaining how to relinquish nepotist and clientelist models of governance, it is not arduous to conclude what had incited the young examinees to join a party whose epistemic character they do not trust in. We can derive analogous inferences from Vuković’s study of Croatian local governments’ political

46  Hana Samaržija economy. He established that corrupt politicians, whose debased actions are empirically documented and publicly known, will more likely be reelected than their more ethical adversaries (Vuković 2017).

2.5 Conclusion In the closing part, we can abridge and reconsider our discussion. Condorcet’s Jury Theorem is an unpretentious quantitative proof advancing the argument that the collective decision-making of many epistemic agents sounder than random will produce an epistemically more valuable decision than any erudite individual agent. However, Condorcet encumbered his decision-makers with three formidable provisions. First, they must be more knowledgeable than random or go astray in ways that cancel each other out. Second, they must not be subject to the same ideological leader, the same information sources, the same background beliefs, or equal cognitive mechanisms (Goodin and Spiekermann 2018: 55). Third, they must elect that political option they sincerely deem the ethical and epistemic optimum. Unlike the Hong-Page theorem of diversity over ability, which was mathematically disproven shortly after its vastly popular advent (Thompson 2014), which some political epistemologists determinedly ignore (Landemore 2020), CJT’s mathematical proof has endured the test of time. It is still utilized as a quantitative argument favoring aggregative democratic decision-making. In the previous twenty pages, I have endeavored to prove that not even one of CJT’s requirements – competence, independence, or s­incerity – can be sustainable in today’s epistemic conditions. Regarding the competence requirement, citizens are satirically unacquainted with the essential facts necessary for partaking in democratic life. The median voter also ostensibly lacks the political sophistication required to locate their place on the ideological spectrum and understand the difference between progressive and conservative values (Converse 1964; Delli Carpini 2005). The small minority of citizens capable of articulating substantial ideological preferences or partisan self-identification misuses this political sophistication to execute convoluted cognitive strategies to uphold their beliefs in the face of conflicting empirical findings (Taber and Lodge 2006). Ultimately, epistemic injustice insinuates that most citizens will misstep similarly. Social epistemologists have spent decades detailing deficits of trust ascribed to marginalized or minority groups, which speaks of embedded stereotypes and prejudices in our collective epistemic resources (Fricker 2006, 2007). The independence requirement, whose description already indicates it is discernibly overly arduous, is easily refuted by the empirically verified common knowledge effect (Gigone and Hastie 1993: 691). The common knowledge effect discloses that the epistemically most valuable or best evidentially sustained stance will not perfunctorily prevail in a group

A Belated Failure 47 discussion. Conversely, the attitude that most discussants held before the debate had even started will retain the upper hand. Suppose that is not a satisfactory argument in today’s epistemic conditions of constant connection on social networks. In that case, epistemic agents are also imperiled by the alluring structures of epistemic bubbles and echo chambers. The first class of epistemic bubbles develops due to our selection of friends and trustworthy information sources. Regardless, they are effortlessly punctured by mere exposure to conflicting information. In contrast, echo chambers are ideological pundits and partisan media’s calculated efforts to ensnare their followers in a world of alternative facts (Nguyen 2018). Ultimately, the mentioned notion of epistemic injustice toward minority and disadvantaged social groups, due to which they are perceived as epistemically incompetent or unworthy of trust, denotes the electorate will approach topics interspersed with identity matters by erring in precisely the same way. As we have already remarked, the conclusive sincerity requirement seemingly appears the most challenging to repudiate due to its intuitiveness. After all, we can assume that voters will, even if they are mistaken, vote for that political option they sincerely deem the ethical and epistemic ideal. Nonetheless, empirical studies contradict this intuition. First, voters effectively attached to their selected political option will not decrease their affection for a favored politician after conceding they have calculatingly lied to them (Swire-Thompson et al. 2019: 6). Over a decade ago, political scientists recorded instinctive sustaining of the currently presiding political party out of apprehension of change, even if that ruling faction is disreputable for its epistemic failures (Jost et al. 2004: 908). According to sociologists, constituencies politically cultivated in states whose corruption renders them flawed democracies will elect and join parties whose ethical and epistemic character they do not confide in if they expect it will bring them economic advantages (Ilišin et al. 2018: 64). It turns out that robust emotional bonds or pragmatic behavior habitually make us cast insincere votes for options we identify as epistemically substandard or incapable of generating valuable decisions. However, this conclusion leads us to a series of open questions rather than a comprehensive answer. Suppose CJT’s requirements cannot survive current epistemic conditions, and the Hong-Page theorem of diversity over ability has been mathematically refuted. In that case, there is no epistemic justification of democratic decision-making more promising than the theoretical optimism of specific epistemologists. Nevertheless, even if it is epistemically substandard, democratic decision-making is undeniably still the fairest and the most inclusive tested procedure of political decision-making, unlike more epistocratic approaches that fare better in theory. Suppose we, as epistemologists resolutely committed to the necessity of democracy’s epistemic justification, adjudicate that the sole valid conclusion is cynical democratic pessimism (Ahlstrom-Vij 2020). In that

48  Hana Samaržija case, we are obliged to offer no less than a sketch of a decision-making procedure that would appease both our epistemic and ethical criteria. Fortunately, numerous social epistemologists, political epistemologists, and political economists have already begun reflecting upon alternative configurations of political decision-making. While some would restrict the electorate to those citizens who have satisfied minor epistemic criteria (Brennan 2011), others ponder upon divisions of epistemic labor between citizens and experts (Goldberg 2011) or hybrid models of political decision-­making. A fruitful epistemic renunciation of democratic decision-­making is not the culmination of our epistemological inquiry but merely the beginning of thorough work on cultivating a decision-making procedure fit for contemporary epistemic conditions.

Notes





1. Political theory and political philosophy utilize the concept of an “unresponsive government” to depict administrations whose policies do not answer to their citizens’ explicitly stated needs. The governments do not respond to evident civic requirements. 2. Parliamentary representative democracy’s institutions of the electorate’s formal equality within the election process, the freedom of choice and speech, and the absence of an autonomous presidential figure distinctive of presidential democracies are its fundamental ethical and political values, which merited its place as the ostensibly most just political configuration. 3. In Europe, the youth category, somewhat unintuitively, encompasses persons up to thirty years of age, so it is fallacious to imagine a surveyed group of high-school students and undergraduates.

References Ahlstrom-Vij, Kristoffer. 2020. “The Epistemic Benefits of Democracy: A Critical Assessment.” In The Routledge Handbook of Social Epistemology, edited by Miranda Fricker, Peter J. Graham, David Henderson, and Nikolaj J. L. L. Pedersen, 406–415. New York: Routledge. Bartels, Larry. 1996. “Uninformed Votes: Information Effects in Presidential Elections.” American Journal of Political Science 40 (1): 194–230. Bartels, Larry M. 2002. “Beyond the Running Tally: Partisan Bias in Political Perceptions.” Political Behaviour 24 (2): 117–150. Bohman, James. 1998. “The Coming of Age of Deliberative Democracy.” The Journal of Political Philosophy 6 (4): 400–425. Boutyline, Andrei, and Willer, Robb. 2016. “The Social Structure of Political Echo Chambers: Variations in Ideological Homophily in Online Networks.” Political Psychology 20 (20): 1–19. Brennan, Jason. 2011. “The Right to a Competent Electorate.” The Philosophical Quarterly 61 (245): 700–724. Brennan, Jason. 2017. Against Democracy. New Jersey, United States: Princeton University Press. Cassam, Quassim. 2019. Vices of the Mind: From the Intellectual to the Political. Oxford, United Kingdom: Oxford University Press.

A Belated Failure 49 Cerovac, Ivan. 2021. Epistemic Democracy and Political Legitimacy. Cham: Palgrave MacMillan. Converse, Phillip. 1964. “The Nature of Belief Systems in Mass Publics.” In Ideology and Discontent, edited by D. Apter, 206–261. New York: Simon and Schuster. Converse, Phillip. 1990. “Popular Representation and the Distribution of Information.” In Information and Democratic Processes, edited by J. Ferejohn and J. Kuklinsky, 369-388. Urbana: University of Illinois Press. Converse, Phillip. 2000. “Assessing the Capacity of Mass Electorates.” Annual Review of Political Science 3: 331–353. Delli Carpini, Michael. 2005. “An Overview of the State of Citizens’ Knowledge About Politics.” In Communicating Politics: Engaging the Public in Democratic Life, edited by Mitchell S. McKinney, Lynda L. Kaid, Dianne G. Bystrom, and Diana B. Carlin, 27–40. New York: Peter Lang Publishing. Delli Carpini, Michael, and Keeter, Scott. 1996. What Americans Know about Politics and Why It Matters. New Haven: Yale University Press. Fricker, Miranda. 2006. “Powerlessness and Social Interpretation.” Episteme 3 (1–2): 96–108. Fricker, Miranda. 2007. Epistemic Injustice: Power and the Ethics of Knowing. Oxford, United Kingdom: Oxford University Press. Gigone, Daniel, and Hastie, Reid. 1993. “The Common Knowledge Effect: Information Sharing and Group Judgement.” Journal of Personality and Social Psychology 65 (5): 959–974. Goldberg, Sandy. 2011. “The Division of Epistemic Labor.” Episteme 8 (1): 112–125. Goldman, Alvin I. 1990. Knowledge in a Social World. Oxford: Oxford University Press. Goodin, Robert, and Spiekermann, Kai. 2018. An Epistemic Theory of Democracy. Oxford, United Kingdom: Oxford University Press. Hong, Lu, and Page, Scott E. 2004. “Groups of Diverse Problem Solvers Can Outperform High-Ability Problem Solvers.” PNAS 101 (46): 16385–16389. Ilišin, Vlasta, Gvozdanović, Anja, and Potočnik, Dunja. 2018. “Contradictory Tendencies in the Political Culture of Croatian Youth: Unexpected Anomalies or an Expected Answer to the Social Crisis?” Journal of Youth Studies 21 (1): 51–71. Iyengar, Shanto, and Westwood, S. J. 2014. “Fear and Loathing across Party Lines: New Evidence on Group Polarization.” American Journal of Political Science 59 (3): 690–707. Jost, John T., Banaji, Maharzin R., and Nosek, Brian A. 2004. “A Decade of System Justification Theory: Accumulated Evidence of Conscious and Unconscious Bolstering of the Status Quo.” Political Psychology 25 (6): 881–919. Kahneman, Daniel. 2012. Thinking, Fast and Slow. London: Penguin Books. Kuklinsky, James H., and Peyton, Buddy. 2007. “Belief Systems and Political Decision Making.” In The Oxford Handbook of Political Behaviour, edited by Russel J. Dalton and Hans-Dieter Klingemann, 45–68. New York: Oxford University Press. Landemore, Helene. 2020. Open Democracy: Reinventing Popular Rule for the Twenty-First Century. New Jersey: Princeton University Press. Levitsky, Steven, and Ziblatt, Daniel. 2019. How Democracies Die. United States: Broadway Books.

50  Hana Samaržija Mill, John Stuart. 2007. On Liberty and the Subjection of Women. London, United Kingdom: Penguin Books. Nguyen, C. Thi. 2018. “Echo Chambers and Epistemic Bubbles.” Episteme 17 (2): 141–161. Samaržija, Hana. 2023. “The Epistemology of Fanaticism: Echo Chambers and Fanaticism.” In The Philosophy of Fanaticism, edited by L. Townsend, R. R. Tietjen, H. B. Schmid, and M. Staudigl. New York: Routledge. Samaržija, Hana, and Cerovac, Ivan. 2021. “The Institutional Preconditions of Epistemic Justice.” Social Epistemology 35 (6): 621–635. Stanovich, Keith. 2021. The Bias That Divides Us: The Science and Politics of Myside Thinking. Massachusetts, United States: MIT Press. Sunstein, Cass. 2011. Going to Extremes: How Like Minds Unite and Divide. New York: Oxford University Press. Sunstein, Cass. 2015. Wiser: Getting Beyond Groupthink to Make Groups Smarter. United States: Harvard Business Review Press. Taber, Charles S., and Lodge, Milton. 2006. “Motivated Skepticism in the Evaluation of Political Beliefs.” American Journal of Political Science 50 (3): 755–769. Thompson, Abigail. 2014. “Does Diversity Trump Ability: An Example of the Misuse of Mathematics in the Social Sciences.” Notices of the AMC 61 (9): 1024–1030. Vuković, Vuk. 2017. “The Political Economy of Local Government in Croatia: Winning Coalitions, Corruption, and Taxes.” University of Oxford: Department of Politics and International Relations. Zaller, John R. 1992. The Nature and Origin of Mass Opinion. Cambridge, United Kingdom: Cambridge University Press.

3

Social Epistemic Miserliness Populism against Democracy Nenad Miščević

3.1 Introduction Our subject will be populism’s struggle against reasonable democratic deliberation and the cognitive-epistemic forces enabling and sustaining it. This fray itself is a widespread phenomenon: like Chávez in Venezuela, elected leaders worldwide have subverted democratic institutions in Georgia, Hungary, Nicaragua, Peru, the Philippines, Poland, Russia, Sri Lanka, Turkey, and Ukraine. Democratic backsliding today begins at the ballot box. Democracy’s erosion is, for many, almost imperceptible. Let us deploy the example of Hungary. Its president, Viktor Orban, has spent several years leading a campaign against George Soros, a Hungarian Jew by birth who has contributed enormously to Hungary’s intellectually advantageous profile, directing, and financing the Central European University (CEU) in Budapest. Soros has been consistently critical of Orban’s populist politics, and the CEU has paid for this insolence by being exiled from Hungary. Orban’s populist campaign against Soros used explicitly anti-Semitic allusions to block free democratic deliberation exercised by CEU’s intellectuals and broadly propagated by CEU leadership. On May 19, 2022, Orban summarized his illiberal political program in twelve points while speaking at CPAC Budapest. He delivered a twelve-point “open-source” guide for conservatives looking to repeat his party’s recent electoral success. We shall also be pointing to other examples, particularly Donald Trump’s nationalist populism and Brexit propaganda. Overall, populism is the most pertinent present-day threat to democracy. And, if one does work on the epistemology of democracy, one should also examine stances and views that represent a menace to it. Therefore, the epistemology of populism and research into epistemic vices constitutive of it is relevant for the epistemology of democracy. In addition, we should investigate hazardous epistemic vices with significant social consequences, such as close-mindedness, sloth, intellectual arrogance, cowardice, dishonesty, epistemic injustice, and dogmatic intellectual self-affirmation. They all have individual and social incarnations: close-mindedness has DOI: 10.4324/9781003311003-5

52  Nenad Miščević historically been a characteristic of powerful secular and religious institutions, often accompanied by sloth, the unwillingness to inquire into epistemic alternatives. Academy has often suffered from intellectual arrogance and exhibited cowardice and dishonesty in more totalitarian times. This chapter will point to a general cognitive defect underlying epistemically vicious features. Here is the preview. The following Section 3.2 is dedicated to characterizing populism. We start with some recognizable examples and then turn to Jean Cohen’s (2019) concise definition of it, comparable to proposals by other prominent researchers investigating populism. It demonstrates that populism is a threat to democratic deliberation. Section 3.3 introduces three groups of cognitive phenomena denoting imperfect cognitive processing. We borrow from cognitive scientists the term “miserliness” for such processing and employ the phrase in the broadest possible sense for all systematic defects of cognitive processing. The three groups differ in their scope. First, there are imperfect methods of understanding and reasoning, the so-called “heuristics and biases” in general. Second, the distinct bias we are curious about is called “myside bias,” notorious for favoring the present convictions and attitudes of the relevant subject and their group. Third, the heuristics of stereotyping and the long route to polarization and radicalization often block democratic deliberation. Section 3.4 finally turns to the cognitive-epistemological grounding of populist thought and discourse, using the three groups delineated in the previous chapter to characterize the epistemology of populism and linking them to populism’s characteristics listed in Section 3.2. For instance, populism’s insistence on the contrast between “us,” the exemplary people, and “them,” alienated elites or foreigners, is tied to myside bias and aided by stereotyping. Epistemic vices support the political ones and vice versa in a vicious circle leading to increasingly anti-democratic results. In conclusion, we summarize our results and point to the necessity of divulging possible remedies to populist pathology. Let me conclude the introductory part by briefly delineating the theoretical framework of our project. It combines cognitive science and political epistemology. As the former has been well-known for almost a century, let me say a few words about the latter. Indeed, most chapters in the present volume belong to this research project and to its critical component. Analytic philosophy recently resurrected this project with a prolonged history. The much-needed resurgence was presumably provoked by catastrophically ideological recent events, including Trump’s escapades and the Brexit farce, belonging to the populist wave. As Michael Hannon and Jeroen de Ridder note in their 2021 volume, “Political epistemology is a newly thriving field at the intersection of epistemology and political philosophy, but it has old roots.” And they cite illustrious predecessors such as Plato, John Stuart Mill, Hannah Arendt, and John Rawls. Political epistemology is an exemplary case of engaged philosophy, looking to utilize the taxonomy developed by Jonathan Wolff

Social Epistemic Miserliness 53 (2019). As Wolff describes it, engaged philosophy starts from concrete social problems, in contrast to “applied” philosophy, which begins from a lofty theoretical framework. In applying philosophy, “the philosopher identifies relevant values, in the context of a problem, current facts, history, and contemporary alternatives. There is a certain amount of sifting and balancing to articulate the messy public debate’s moral dilemmas. Then the identification and evaluation of possible solutions, before making recommendations, may or may not affect actual policy” (Wolff 2019: 22). Political epistemology’s central element is the study of the cognitive or epistemic obstacles to correct epistemic functioning and a just organization of society, primarily in its present form. It addresses perilous epistemic vices, primarily those with momentous social consequences, such as close-mindedness, sloth, intellectual arrogance, cowardice, dishonesty, epistemic injustice, and dogmatic intellectual self-affirmation. Other parts study positive or realist aspects of political life – see Chapter 11 of this book on sense-making in democratic politics. A related epistemological problem is the status of political thought experiments, from Plato to contemporary contractualist thinkers, geared at identifying crucial positive features of political arrangement. It is helpful to apprehend the epistemological project’s levels of inquiry to locate the present chapter’s purpose. On the descriptive side, one has an ordinary understanding of politics embodied in everyday discourse. At this level, the central example of philosophical interest is Cassam’s pioneering work on epistemic vices, stressing and occasionally correcting regular dialogues about them. At the next level, there is the inclusion of cognitive research. The present chapter will discuss the epistemological consequences of cognitive psychological insights into typical human mental weaknesses. Many research programs and results at this level, such as social psychology and the study of practices like stereotyping, have tremendous political-epistemic consequences. We need an epistemology of such procedures to place the psychological result into a more philosophical context. The third level is philosophical, where critical political epistemology typically sets itself. This differentiation brings us to the normative side of the project. When confronted with the wealth of factual explanatory material on all three levels, philosophers must assume a normative stance and discuss and defend it. Which qualities and consequences, both epistemic and political, make a given practice abominable or unsatisfying, and which would make it normatively correct, appropriate, or even exemplary? Here, the critical epistemologist has a choice of several normative frameworks. The approach I assume in this chapter is based on virtue epistemology and, in the concrete case, focuses on epistemic vices to be avoided in the political domain. Vice epistemology appears appropriate to the issue we are interested in, the role of cognitive biases and the underlying cognitive miserliness in blocking democratic discourse. These tendencies seem to be a

54  Nenad Miščević clear example of epistemic vices. Once we have chosen a fitting normative framework, we can turn to a given practice’s concrete harms and benefits, with a particular interest in epistemic harms and epistemic benefits and their political aspects. In this chapter, we join other colleagues in discussing specific injuries to democratic practices stemming from cognitive imperfections and scrutinizing feasible potential remedies.

3.2  The Populist Threat to Democracy We can now turn to the populist threat. We start with a few examples and then turn to general characteristics. The American and British people have examples of Trump and Brexit in their countries. Here is a typical quote from Trump, manifesting the standard features of populist discourse: The US has become a dumping ground for everybody else’s problems. [Applause] Thank you. It’s true, and these are the best and the finest. When Mexico sends its people, they’re not sending their best. They’re not sending you. They’re not sending you. They’re sending people that have lots of problems, and they’re bringing those problems with us. They’re bringing drugs. They’re bringing crime. They’re rapists. And some, I assume, are good people. The quote starts with a stark contrast between us, the US, and the Mexicans. The Mexicans arriving in the US are not potential workers who desire to work and survive but narcos and criminals. And they are rapists. All of them? While it is not stated with a universal quantifier, the term “they” suggests “all of them.” Well, some are good people, Trump says, in conflict with the leading suggestions of his talk. Similarly, across Europe, populists claim that Muslim refugees are conquering Christian territories for a future Muslim empire. Populist news fills the media. One might think of the old description of the Fame, proposed by Virgil, who succinctly summarizes what she does in two verses: Things done relates, not done she feigns, and mingles truth with lies. Talk is her business, and her chief delight (IV, 189, Dryden’s translation) But what truth is here mingled with lies, and what lies do the populist media spread? Consider the features analyzed by researchers (J. Cohen, C. De La Torre) as populism’s characteristics, borrowing the list from Jean Cohen (2019): First, there is the appeal to “the people” and “popular sovereignty.” These phrases with little content are deployed to unify heterogeneous demands and grievances.

Social Epistemic Miserliness 55 Second, their pars pro toto logic extracts the “authentic people” from the rest of the population and is very apparent in the populist nationalistic rhetoric. Third, their discourse pits the people against elites: the political-­ economic and cultural “establishment” are cast as usurpers who corrupt, ignore, or distort the “authentic” people’s will. Fourth, we can identify the solid frontier of antagonism along the lines of a Schmittian friend/enemy conception of the political that identifies alien others who violate the people’s values. Fifth, there is unification, typically through strong identification with a leader claiming to embody the authentic people’s will and voice, incarnating their unity and identity. Let me add an illustration I like a lot: “Chavez is the people, the people are Chavez” has been a slogan of Venezuelan identification with their leader. Sixth and seventh, populism’s less prominent focus is on political representation’s symbolic dimensions and a performative leadership style that mimics the authentic people’s habitus (dress, speech, manners). Eighth, populism exhibits rather dramatic and rhetorical forms of argumentation linking talk about making the nation great to discourses about the restoration of honor, centrality, and political influence to the authentic people. Ninth, we can locate a focus on alleged crises, a national decline, and an orientation to the extraordinary dimensions of politics. Tenth, populism genuinely has a thin character: dependence on a host ideology for content and moral substance. (2019: 13) Here are some illustrations. Le Pen’s slogan “Nationalists of all countries, unite!” plays with the classical Marxist slogan “Proletarians of all countries, unite!” It abstracts from the fact that nationalists of different nations have conflicting interests (Ukrainian nationalists against Russian ones, Turkish nationalists against Kurdish ones, and so on), thus suggesting an impossible program. Or consider the “solution” strongly indicated in the Brexit campaign: “Let’s take back control.” Vote “Leave.” The unemployed British worker is to believe they will be “taking control” if they vote “Leave!” Currently, we can add to the list of such pseudo-­ solution proposals the establishment of a puppet pro-­Russian government in Ukraine to prevent the West from interfering with Russian interests. Such propositions are cognitively deficient, and we should scrutinize their defects. These examples bring us to the central epistemic puzzle of populism’s popularity: why do people opt for such an irrational, epistemically vicious, and anti-democratic interpretation of politics? Here, we shall join the authors who divulge a comprehensive explanation of human

56  Nenad Miščević cognitive mechanisms. For instance, Judith Glück, in her 2019 paper with the telling title “Wisdom vs. Populism and Polarization: Learning to Regulate Our Evolved Intuitions,” suggests that we should scrutinize cognitive heuristics and biases that “may” offer a good lens for understanding the increasing polarization currently happening in the most prosperous democracies of this planet (2019: 87). We now turn to detecting the cognitive weaknesses-vices underlying populist discourse and populist threats to democracy.

3.3  Cognitive Miserliness – A Brief Overview We shall divide these cognitive weaknesses into three categories: those characterized in an exclusive individualist way, those partially associated with social or group behavior, and social ones. The first category has been prominent since the very beginnings of cognitive research, while the second and third have appeared somewhat later. We need a general term to cover the three categories. Contrary to positive, epistemically virtuous rationality, we have decided to use “cognitive miser” and “cognitive miserliness” in this broadest possible sense. Here we thus apologize for using the terms perhaps somewhat more extensively than standard cognitive literature. We shall take the work of Keith Stanovich as our guiding source. In fact, we borrow his label of cognitive miserliness to portray the functioning of our mental apparatuses that produce biases of any kind (see Stanovich 2018). Cognitive miserliness goes rather far. It might be constitutive for the normal functioning of our cognitive system, particularly concerning reasoning and inference. Daniel Kahneman, one of the pioneers in these areas, has proposed his model of heuristics and biases as exemplifying such miserliness. We shall be using the term in the broadest possible sense for all systematic defects of cognitive processing. So, we shall start with a classical instance of miserly processing, namely heuristics and biases in ordinary reasoning. Then we shall pass to a broad bias, called “myside bias” in the literature, and end with the politically critical biased process of stereotyping. We shall argue that miserliness comprises all these defects. 3.3.1  The Classical Cognitive Biases and Heuristics Remember the classical example of cognitive bias offered by Daniel Kahneman, using the model of Italy and France competing in the 2006 final of the World Cup. Suppose the average fan is asked what they think about the probability of Italy winning. In that case, they will give a far different response than if they are asked what they think about France losing, even though both questions request their response to the very same event (2012: 354). The cognitive system responds differently to questions framed in terms of gain than equivalent ones depicted as losses.

Social Epistemic Miserliness 57 We might further acknowledge the relevance of framing by examining the given examples of populist discourse. Take Marakesh: to the average reader, the name sounds neutral. However, populists create an association with Christian martyrs, Franciscans killed by a Muslim ruler some seven centuries ago. The name is thus framed: whatever is done at this place, especially if it favors Muslims, must be harmful to us here. The same goes for Eric Zemmour’s latest tactics, the leader of the most potent populist party in France, launched on December 5, 2021. Zemmour proposed a “reconquest of the greatest country in the world” in front of 15,000 people and the name “Reconquest” for his party. The term refers to the so-called reconquest of Spain and part of France from the hands of Arabs in the Middle Ages. The negative attitude to the French of Arab origin and Muslim refugees is thus framed in terms of medieval “glorious victories,” all celebrated at the renowned medieval church of St. Denis, the sacred place of French culture and history. Framing is just one example of heuristics that generates biases. Kahneman and Tversky famously argued that our cognitive system exhibits many similar tendencies. They often supersede a more straightforward question for a difficult one, overestimate low probabilities, respond more strongly to losses than to gains (loss aversion), frame decision problems narrowly, in isolation from one another, and exaggerate emotional consistency. Kahneman and Amos Tversky have been analyzing the consequences of such experiments since the seventies, concluding that our mind is composed of two systems. System 2 is the conscious one. It is who we are: it is cautious but not very active. The other, System 1, is automatic, unconscious, and swift. He writes: The defining feature of System 2, in this story, is that its operations are effortful, and one of its main characteristics is laziness, a reluctance to invest more effort than is strictly necessary. Consequently, the thoughts and actions that System 2 believes it has chosen are often guided by the figure at the center of the story, System 1. However, there are vital tasks that only System 2 can perform because they require effort and acts of self-control in which the intuitions and impulses of System 1 are overcome. (2012: 33) System 1 does not reflect. It just proceeds automatically with little effort. Kahneman notes that “it sometimes answers easier questions than the one we asked it, and it has little understanding of logic and statistics.” (2012: 27) We have just listed some of the typical heuristics and biases that cognitive scientists argue are the work of System 1. This system sometimes

58  Nenad Miščević replaces a more straightforward question for a difficult one, overestimates low probabilities, manifests loss aversion, frames decision problems narrowly, in isolation from one another, and falls victim to the halo effect. And it reacts more strongly to losses than to gains. We could designate the tendency to supersede a more straightforward question for a difficult one in terms of laziness (as Kahneman does) or sloth, using a classical label for one of the central human vices. Nevertheless, Keith Stanovich has introduced the more precise term “cognitive miserliness” and explained that humans are cognitive misers due to their fundamental tendency to default to processing mechanisms of low computational expense. He quotes his colleagues saying that the rule that human beings seem to follow is to engage the brain only when all else fails – and usually not even then. All animals are under selection pressure to be as stupid as they can get away with. Thus, we shall retain his terms “cognitive miser” and “miserly processing” and use them in what follows. In other words, we shall accept his research’s central thesis: miserliness is the main characteristic of everyday reasoning and inference. And here, we shall be interested in vices derived from human miserliness. Let me briefly turn to one aspect of miserliness: System 1’s laziness or sloth. It might be seen as the general category of demotivating epistemic vice. It appears more extensive than epistemic insouciance (Cassam) or being a slacker (Battaly). Insouciance could be a matter of mere superficiality or could be motivated by haste, whereas sloth is a solid trait underlying various epistemic mistakes. Insouciance, being a slacker, and having other similar epistemic defects could result from one’s sloth. Sloth is more of an epistemic kind, encompassing more inertial (“passive”) and “self-assertive” individual and collective varieties, and both Cassam’s stance and posture types. But doesn’t genuine dogmatism require more than mere sloth, one might ask. However, it is not “mere”: it is a self-­ assertive sloth. Sloth is usually seen as a phenomenon of epistemic passivity, a lack of motivation to investigate. However, one can think of a more involved version of sloth, an active refusal to question and investigate. Understood in this broader sense, with a passive and an active component, sloth is the cornerstone of epistemic dogmatism. Dogmatism is probably responsible for blocking the central epistemic motivation of democratic discussion and deliberation. If we accept, following Mercier and Sperber (2017), that the average human is intellectually vigilant only when facing others from which one distances oneself and is non-vigilant with one’s ingroup, we might hypothesize that this is a further motive for social dogmatism, as alongside its ruinous political consequences. In this sense, a central demotivating vice would be fundamental for an entire system of epistemic vices. If this seems correct, we can appeal to the fact that sloth is an epistemic vice and conclude that it is a rudimentary epistemic vice. Snježana Prijić Samaržija has asked me in a discussion if dogmatism could be an epistemic virtue. Her question was

Social Epistemic Miserliness 59 inspired by reading Ahlstrom-Vij and his claim that, if we are consequentialists concerning truth, we should consider the issue of whether dogmatism can lead to or preserve truth an empirical question. I agree about the partly empirical character of the issue, but I believe we have enough empirical grounds to doubt its alleged advantages. So much about classical heuristics and biases. 3.3.2  Myside Bias and Stereotyping A distinctive miserly tactic we now turn to is what Stanovich, including in the present volume, calls “myside bias” (we shall remain with Stanovich as our guide for cognitive science’s views). The primary characterization of the bias is simple: it occurs when people evaluate evidence, generate proofs, and test hypotheses in a manner biased toward their prior opinions and attitudes. A typical illustration comes from a study showing that American examinees believed that a dangerous German vehicle in America was far more deserving of prohibition than an unsafe American vehicle in Germany. Nobody should ban our cars. This belief, however, does not extend to foreign automobiles. In work co-authored by West and Toplak, Stanovich rightly concludes that the study “demonstrated a sizable myside bias” (Stanovich et al. 2013: 262). If you need an illustration from the populist folklore, think of Trump’s incredible comment: “I think everyone is a threat to me” (Kellner 2016: 8). No wonder the authors add that the avoidance of myside bias is a component of the “multifarious concept of rational thought” (Stanovich et al. 2013: 263). We may add that prejudice is not only irrational but also is a paradigmatic epistemic vice. Otherwise, we can remain brief here and kindly ask the reader to study Chapter 9 of this book for further information. Stanovich and his collaborators note that political parties and ideologies have become present-day equivalents of tribes, and these “tribes” run roughshod over the nation’s cognitive life, positioning nonintellectual strategies for scoring points ahead of objective debate (2021: 125). No wonder the interview with Stanovich on the Wissenschaft & Komunikation blog1 and his Myside bias book feature as illustrations a photo of Trump and a poster inviting the viewers to vote for Brexit. The most critical further point is that myside bias trails membership in a group: Stanovich stresses the importance of group membership in “fostering mysided thinking” (2021: 47). Since the bias for the group attitude is politically essential and crucial for our present purposes, let me extend its name to “our-side bias”: it is merely myside bias extended to my group as an entity. One can assume that in the experiment we just commented on, the American subjects’ prejudicial thinking in favor of the American car is an example of such our-side bias. Stanovich talks about identity costs and group identity gains in this context (Ibid.). The famous injunction “Make America great

60  Nenad Miščević again” suggests that “we,” Americans, should make “our country” great again. One can take it almost as the political slogan of our-side bias, illustrating how populism goes well with typical dispositions associated with cognitively imperfect processing. We may end by pointing to stereotyping as an essential populist tactic. Traditionally, a stereotype has been defined as overgeneralized attributes related to the members of a social group, such as the reserved Englishperson or the geeky engineer. Here is a more comprehensive quote: Stereotypes are problematic because they are negative, inaccurate, and unfair – they would be part of the study of person perception more broadly if they weren’t. Regarding negativity, the data is precise, and we probably should acknowledge it more fully, as we generally do regarding prejudice. Although they can be positive, stereotypes are primarily negative. When asked to do so, we generate many more negative than positive stereotypes, and even expressing positive stereotypes is not seen positively. Consider how we might react to people who have claimed that African Americans have the positive traits of being athletic and musical. The problem, in part, is that if we express positive stereotypes, it is assumed that we hold the negative ones, too. (Stangor 2009: 2)

3.4 From Cognitive Miserliness to the Attractions of Populism We might then hypothesize that various aspects of cognitive miserliness, in the broadest sense, from ordinary sloth to our-side bias and stereotyping, will play a role in the populist blocking of democratic deliberation and that the siege will have a systematic nature. First, there must be a place for the most general cognitive activities of the miserly processing system, heuristics, and biases: the confirmation bias, the heuristics of superseding a more straightforward question for a difficult one, the strategy of responding more strongly to losses than to gains, and the exaggeration of emotional consistency (the halo effect). Next comes the specific feature of myside bias, which is reasonably fit for a role in contrasting “us” with “them,” assigning preferences to “our” views and a negative value for what comes from the “other.” Even more socially relevant are social heuristics and biases, concentrated upon the strategies of stereotyping. The process can begin with the typical myside bias and continue with stereotyping. I already quoted a fine summary of the latter method. It is easy to see the crucial role of stereotyping in central populist strategies. Suppose my biases and stereotyping come together in the interaction of the “us”-group and the “they”-group. They will typically lead

Social Epistemic Miserliness 61 to a polarization that can subsequently become less or more radical. The crucial phenomenon in this domain, both political and epistemic, is the dynamics of polarization. The epistemology of populism is therefore further characterized by increasing radicalization. Let us restate the epistemic characterization of the process. The process might start with simple myside bias on the side of the relevant group’s members, an ordinary, non-problematic type of bias. And the conflictual situation will lead to a strengthening of the group-centered tendency, with activities such as negative stereotyping as their main cognitive manifestations. These internal, cognitive-affective developments and external circumstances can increase polarization and radicalization. As we noted above, epistemic vices support the political ones and vice versa in a vicious circle leading to increasingly anti-democratic results. Thus, the threats to democratic deliberation grow and become dramatic, as is often noticed in the literature. So, let us start with classical cognitive biases and heuristics derived from around sloth or, to use Cassam’s term, insouciance. Both Cassam and Aberdein have noticed the relevance of insouciance, to stay with the word they use, for populism and populist argumentation. Aberdein notes that we may understand several factors to which the rise of populism has been attributed as arising from vices of poor argumentation, including arrogance, emulousness, and insouciance. Conversely, the virtues of argument, such as humility and good listening, offer some prospect of a constructive response to populism. Cassam, followed by Aberdein, links insouciance directly to populism by illustrating his account with a familiar quotation from the British politician Michael Gove, a reference often interpreted as an overt appeal to populism. Gove famously remarked that he thinks the people in this country are fatigued of experts: of organizations with acronyms saying they know what is best and consistently get it wrong. Aberdein comments that populist arguments are characteristically accusations of vice against the elite. Still, they are also the subject of charges of vice by other parties, so the roles of the critic and the target can be played by either populist arguer or their adversary. One can add that in populist argumentation, one finds a lot of appeal to pre-judgment, that is, a judgment made or maintained without proper regard to the evidence, first on the side of the arguer and then on the side of the listener(s). In our example from the journal “Demokracija,” this is illustrated by the claim that Muslims are aggressive and dangerous: they hate us as members of the Christian civilization. The inertial sloth will prompt the reader of such populist texts to accept the accusations against Muslim migrants without questioning them. This kind of self-assertive sloth will govern the practice of the author and editor of the populist press. We shall add a few more examples of the relevant biases below. We now pass to a brief analysis concerned with the other two categories of miserly thinking we noted, the myside bias and the stereotyping. So, let

62  Nenad Miščević us further document the connection between such cognitive factors and the anti-democratic political program of populism by taking a brief look at the features identified by researchers as characteristics of populism and examining their conceivable cognitive underpinnings. Please return to the features identified by Cohen and De La Torre as populism’s characteristics and look to their cognitive reinforcements. Stay with Cohen’s list of ten criteria to identify a movement, leader, or party as more or less populist. Recall the list: First, there is the appeal to “the people” and “popular sovereignty.” These phrases with little content are deployed to unify heterogeneous demands and grievances. Second, their pars pro toto logic extracts the “authentic people” from the rest of the population and is very apparent in the populist nationalistic rhetoric. Third, their discourse pits the people against elites: the political-­ economic and cultural “establishment” are cast as usurpers who corrupt, ignore, or distort the “authentic” people’s will. Fourth, we can identify the solid frontier of antagonism along the lines of a Schmittian friend/enemy conception of the political that identifies alien others who violate the people’s values. Note the immediate connection to myside bias and stereotyping with features two to four. Start with feature two. The “authentic people” carries the positive we-stereotype, while the caricatures of the elites and the enemy category carry the negative ones. It thus immediately introduces polarization. Next, the division pointed to in characterization number two advances our-side bias as a pertinent psychological source and supporting force. The people are “us,” and our-side bias is seen as a pro-people attitude: a good American has it for Americans, an excellent Slovenian for the Slovenian people. This elaboration helps us to comprehend characterization number one better. The words “the people” and “popular sovereignty” indeed have little content, as used in the populist discourse; theoreticians sometimes classify them as “empty signifiers.” This feature, however, nicely combines two dimensions of the use of the pronoun “we” and of our-side bias. The first is its emotional power: if person A belongs to my “people,” he is one of us, worthy of expressing our-side bias. The force of the bias supports A, supports A’s views and actions, and motivates me to join A in their efforts. The second is its plasticity: if the opposition is between Croats and Serbs, “the people” stand for national belonging: my people are people of the same nationality. If the opposition is overwhelmed by some other contrast, say Croats-and-Serbs versus Arab refugees, the “the people” will denote “us Christians” as against “the Muslims.” The our-side biases follow the line: the our-side

Social Epistemic Miserliness 63 preference for my co-nationals is replaced by our-side discrimination against Muslims. Here is a further example. A funny-looking poster by the populist “Alternative for Germany” (AfD) tells the viewer the following: “Burkas? No, thanks, we prefer bikinis. Germany, have confidence in yourself.” The accompanying illustration features three young white women wearing bikinis. Preferring bikinis to burkas can be seen as characterizing “our” culture in contrast to the traditional Muslim one, but what is the connection with Germany having confidence in itself? The plasticity of “us” connects the two desiderata, first, the aversion to Muslim prohibition, and second, the love of German nationality. The empty signifier “us” is doing its work rather well. So much about the first two features. Feature three adds another plastic signifier to the list, the word “elites.” Who counts as “we” and who counts as belonging to the “corrupt elite” depends on circumstances: the president of Hungary or Turkey belongs to the people, and his government can be seen both as his helpers or as the corrupt elite, depending on his immediate political needs. The ourside bias just follows the rulers designed as being “we.” The important anti-democratic feature of populism, its intolerance of pluralism, analyzed in detail by William Galston (2018) in his book with the telling title “The Populist Threat to Liberal Democracy,” is clearly in line with the solid our-side bias that we can recognize as characterizing it. Feature four takes us directly from our-side bias to stereotyping and polarization. We quoted social psychologists claiming that stereotypes “optimize the balance between the minimization of differences among people in the same group and the maximization of differences between ingroup and outgroup (or non-ingroup).” Prototypes define and prescribe the properties of group membership (perceptions, attitudes, feelings, behaviors) in such a way as to render the ingroup distinctive and high in entitativity. In the context of populist discourse, the crucial framework is the one of intergroup threat, when the we-group is presented as being threatened by the “them group.” For instance, Muslim refugees threaten the home working class, or Muslim refugee men imperil “our” white women with sexual violence. A dramatic example is the invention of the word “Rapefugees,” which suggests the rapist stereotype for the refugee population, of course, setting aside the fact that at least half of the people are women and children, and then projecting the halo effect of this nasty caricature of male refugees onto their entire large group. Social psychologists note that cognitive responses to intergroup threats include changes in perceptions of the outgroup, mentioning changes in stereotypes and dehumanization. Cognitive biases in intergroup perceptions should also be triggered or amplified by threats. For example, a threat may increase the occurrence of the ultimate attribution error (…), in which harmful

64  Nenad Miščević acts of the outgroup (and positive acts of members of the ingroup) are explained in terms of member characteristics. In contrast, positive outgroup acts (and hostile ingroup acts) are attributed to the situation. (Stephan et al. 2009: 50) This quote carries us to additional features characterizing populist strategies (we repeat the sections here to aid the reader): Fifth, there is unification, typically through strong identification with a leader claiming to embody the authentic people’s will and voice, incarnating their unity and identity. Sixth and seventh, populism’s less prominent focus is on political representation’s symbolic dimensions and a performative leadership style that mimics the authentic people’s habitus (dress, speech, manners). Eighth, populism exhibits rather dramatic and rhetorical forms of argumentation linking talk about making the nation great to discourses about the restoration of honor, centrality, and political influence to the authentic people. Of course, they all go with stereotyping: indeed, they require the existence of stereotypes of authentic people and their leaders. But also note that the features listed require other classical negative heuristics and biases: the confirmation bias helps the subject jump from one presented example (a dark-skinned foreigner who once attacked a homegirl at a New Year party at a public square transformed into “all dark-skinned foreigners”). Similarly, the heuristics of substituting a more straightforward question for a difficult one might lead the populist subjects to focus on the apparent characteristics of easily noticeable young male refugees, forgetting the predominance of women and children in the refugee population, the tactics which will not be available for Ukrainian refugees. We mentioned that naïve subjects respond more strongly to losses than gains: the populist condemnation of foreigners thus abstracts from the revenues for an aging Europe in terms of a young and capable workforce. Similarly, a further characteristic of miserly processing: it frames decision problems narrowly, in isolation from one another, and exaggerates emotional consistency (the halo effect). All these three features would be impossible without the halo effect heuristics and bias. This discussion brings us then to the penultimate and ultimate characteristics: Ninth, we can locate a focus on alleged crises, a national decline, and an orientation to the extraordinary dimensions of politics.

Social Epistemic Miserliness 65 We noted that the leading cognitive sub-system reacts more strongly to losses than gains. “We lost our ability to influence the economy, and Ukrainian politics” might appear more important than the gains from the peaceful coexistence of two neighboring countries; once the appearance is well-established, it might survive even in the face of economic threats from the EU and the US. Tenth, populism genuinely has a thin character: dependence on a host ideology for content and moral substance. Indeed, the nine crucial characteristics point to vital features of miserly processing in the term’s broadest sense; the last quality suggests a way of connecting the result to various ideologies that offer a more comprehensive framework. No strict rules govern the connection itself; take as an example the move performed by Maduro, supposed to be a leftist populist in favor of Putin’s Russia, which is not leftist even in their wildest dreams. It suffices that it is anti-US directed to gain the support of an allegedly leftist Venezuelan leader. So much about the cognitive mechanism supporting populist strategies. The reader might have noticed that some of the components of miserly processing can support other political options: stereotyping appears in all sorts of political approaches. For instance, it figures in the stereotyping of the “class enemy” in Stalinist ideologies or harmful stereotyping of women in any patriarchal ideology. Our characterization suggests that what is typical of populism is the combination of all the traits listed and the cooperation of various miserly strategies supporting each of them. With this brief analysis of the cognitive bases of the negative features of populism in mind, we can appreciate the diagnoses given by authors like Judith Glück (see her quoted 2019 paper). She notes that “dual-process theories (…) argue that people can reflect on their decisions and moral judgments in a relatively objective, unbiased way. However, this type of reflection requires time and mental effort, while intuitive judgments are fast and automatic” (2019: 87). And she reminds us that people do not always engage in conscious reflection and that a “goodenough decision” should be sufficient for most of the everyday situations. The process of antagonization, to recapitulate, can begin with ordinary myside bias. Of course, arrogance and emulousness then strengthen the attitudes and practices involved. Remember the stereotyping generalization “dark-skinned refugees are rapists.” It commences from the usual miserly heuristics and biases, jumping from one or two cases reported by newspapers to all dark-skinned refugees, contributing to the highly negative stereotype. The crucial phenomenon in this domain, both political and epistemic, is the ensuing dynamics of polarization, strongly supported by populist propagandists. The process thus goes from the vice of miserliness to the populists’ perverting of democratic deliberation.

66  Nenad Miščević Remember the previously mentioned Trump quote, claiming that Mexico sends people bringing drugs and crime and are rapists. It exemplifies most of the features we listed: negative framing, stressing of potential losses (our country is a “dumping ground for everybody else’s problems”) rather than potential gains, hasty overgeneralizations supported by negative halo effect, pointing to our-side bias and stereotyping, and an incentive to polarization. Of course, the process of populist manipulation needs a broader social-political framework in which it can take place. The framework required is the one in which the “people” can make political decisions: in the opposite situation, say of an aristocratic rule, the mobilization of the masses can hardly have similar political effects. The same with a solid one-party bureaucratic government, say in the classical Stalinist framework. Here, the march of the groups is controlled by a strict hierarchy of the ruling power, blocking in advance the spontaneous reaction of the “masses.” Similarly, on the opposite right-wing extreme, the classical dynamics of Nazi success can begin with populist dynamics but will soon end in a complex, hierarchized system. This phenomenon has attracted some attention in studies of right-wing populism; authors often talk of populist elements or components in the Hitler or Mussolini type systems without classifying the whole system as populist. In other words, what is needed for the populist scenario is a kind of proto-democratic framework where people can exercise intense pressure on the structures in power. Urbinati notes that the central tenets of populism “are embedded in the democratic universe of meanings and language” (Urbinati 2019: 95). Here is the quote in its broader context: Populism is a revolt against a pluralist structure of party relations in the name not of no-party or a “partyless democracy,” but of the power of “the part” that populism declares to be superior or that deserves supremacy because it is “good” part. This feature makes populism a form of factionalism that collides fatally with constitutional democracy, even if its central tenets are embedded in the democratic universe of meanings and language. (Urbinati 2019: 95) Populism thus perverts the proto-democratic context in which it usually appears. It is essential to keep this in mind to answer a broader and more vital question. Given that cognitive miserliness, with heuristics and biases, stereotyping, and polarization, is a general and all too human aspect of us human beings, why would it contribute to a particular political phenomenon like populism? In specific proto-democratic contexts, it contributes to circumstances of economic crises, unemployment, and the like, very extensively analyzed in the studies concerning the success of Trump, the Brexit, or the Latin American sociopolitical troubles. The

Social Epistemic Miserliness 67 typical structure suggested by populist rhetoric is one of a singularly active leader and very passive “people,” and it goes well with epistemic miserliness. However, this is a topic requiring a separate paper. The available literature on populism offers a range of answers that delimit the everyday context in which epistemic vices of miserliness are relevant and support populist practices. This chapter is just a brief introduction to this vast area, and much more work is being done and has yet to be done on the topic. To sum up, epistemic vices linked to miserly cognitive processing produce and support the populist caricature of democratic deliberation, threatening real democracy.

3.5 Conclusion: Populism and the Roles of Social Epistemic Miserliness Democratic deliberation essentially depends on the rationality of active deliberators and the rationality of their audience. We have looked at the phenomenon that is the most harrowing present-day threat to democracy, arising from the contemporary world’s proto-democratic framework, the danger of populism, and populist propaganda. Populists appeal to their “people,” co-national and co-religious, to advance their political objective of maintaining or increasing support for the leader. Cassam is right about stressing the goal or function of the given epistemic-political movement. See his book Conspiracy Theories (2019: 15). They usually do it by installing hatred of some specific group, typically foreigners or elite constituents, thereby developing polarization and possibly radicalization. We have examined the cognitive-epistemic roots of the whole process, concentrating on the social role of miserliness of ordinary human reasoning as the typical vice underlying the populist program. A virtue-theoretic, or, more precisely, vice-theoretic account of reason, can enhance our understanding of the phenomenon of populism and offer some lines of response. We looked at the cognitive-epistemic ground of the whole process and the role of cognitive miserliness in the populist efforts’ central features. We started with general cognitive parsimony, a kind of sloth, unwillingness to engage in reflection that also points to a dogmatism that might be responsible for blocking the central epistemic motivation for democratic deliberation. Suppose we accept that the average epistemic agent is intellectually vigilant only when interacting with agents they disagree with and non-vigilant with like-minded peers. In that case, this might be an additional motive for social dogmatism and its catastrophic political consequences. There might also be a structural perspective: as sloth is the opposite of curiosity, which is the central motivating virtue, we might perceive sloth as its opposition, a central demotivating epistemic vice, and the grounding pillar of the culture of ignorance. We noted at the beginning that we are here using the term “miserliness” in the broadest possible sense for all systematic

68  Nenad Miščević defects of cognitive processing, including “myside bias” in the literature and the politically crucial biases process stereotyping. We have counted all these defects as cases of miserliness. Next, we have sketched the road from general cognitive miserliness and heuristics and predilections characterizing miserly reasoning, through specific myside or our-side biases, to stereotyping and polarization, which can lead to radicalization, catastrophic for any democratic project. The exact negative contribution of human cognitive weaknesses to such an anti-populist project requires a clear political framework with enough democratic potential that then gets abused and “perverted” in the successful program of populist rule. This insight concludes our discussion on the harmful component of the critical social epistemology in this chapter. We started from the “bottom,” from concrete problems burdening a wide range of countries, and tried to sketch a broader, theoretically more affluent picture at a general level; the project thus belongs to what Wolff describes as the bottom-up or “engaged” philosophy. The next task awaiting the theoretician is at least as demanding as accounting for populist success, namely the assignment of identifying possible remedies. Some of them might be more local. For example, the anti-Semitic hysteria of Orban’s propaganda could be blocked by the awareness of the Hungarian belonging of the local Jews (and we could perhaps stop the anti-Soros propaganda by raising awareness of his contribution to Hungarian cultural, intellectual, and political life). Some may be more global; for instance, the Marrakesh advice about the refugee movement suggests strengthening international cooperation and global partnerships for safe, orderly, and regular migration or providing access to essential services for migrants. Some significant contributions by the authors represented in this volume – such as the ones by Ilya Somin and Robert Talisse, and Jason Brennan in his many works – point in an exciting direction. But the task is enormous, and this is what awaits us soon.

Note

1. See the blog at: https://a-g-i-l.de/the-virus-of-myside-bias-is-spreadingamong-cognitive-elites-interview-with-keith-e-stanovich/

References Cassam, Q. 2019 Conspiracy Theories. Cambridge: Polity Press. Cohen J. L. 2019. “Populism and the Politics of Resentment”. Jus Cogens 1: 5–39. Galston, W. 2018. The Populist Threat to Liberal Democracy. Yale: Yale University Press. Glück, J. 2019. “Wisdom vs. Populism and Polarization: Learning to Regulate Our Evolved Intuitions”, in Applying Wisdom to Contemporary World Problems, Sternberg, R. J.,·Nusbaum, H. C., and Glück, J. (eds.). Cham: Palgrave Macmillan. 81–110.

Social Epistemic Miserliness 69 Hannon, M. and De Ridder, J. (eds.) 2021. The Routledge Handbook of Political Epistemology. London: Routledge. Kahneman, D. 2012. Thinking Fast and Slow. London and New York: Penguin Books. Kellner, D. 2016. American Nightmare Donald Trump, Media Spectacle, and Authoritarian Populism. Brill: Sense Publishers. Mercier, H. and Sperber, D. 2017. The Enigma of Reason. Cambridge, MA: Harvard University Press. Stangor, C. 2009. “The Study of Stereotyping, Prejudice, and Discrimination Within Social Psychology: A Quick History of Theory and Research”, in Handbook of Prejudice, Stereotyping, and Discrimination, Nelson, T. D. (ed.). Oxfordshire: Psychology Press Taylor & Francis Group. Stanovich, K. E. 2018. “Miserliness in Human Cognition: The Interaction of Detection, Override and Mindware”. Thinking & Reasoning, 24(4): 423–444. Stanovich, K. E. 2021. The Bias That Divides Us: The Science and Politics of Myside Thinking. Cambridge, MA: Massachusetts Institute of Technology. Stanovich, K. E., West, R. F., and Toplak, M. E. 2013. “Myside Bias, Rational Thinking, and Intelligence”. Current Directions in Psychological Science, 22(4): 259–264. Stephan W. G., Ybarra, O., and Morrison K. R. 2009. “Intergroup Threat Theory”, in Handbook of Prejudice, Nelson T. (ed.). Mahwah, NJ: Lawrence Erlbaum. 43–59. Urbinati, N. 2019. “Antiestablishment and the Substitution of the Whole with One of Its Parts”, in Routledge Handbook of Global Populism, Carlos de la Torre (ed.). London: Routledge. 77–97. Wolff, J. 2019. “Method in Philosophy and Public Policy: Applied Philosophy versus Engaged Philosophy”, in The Routledge Handbook of Ethics and Public Policy, Lever, A. and Poama, A. (eds.). London: Routledge. 13–24.

4

Critical Thinking and Trusting Experts in Real-life Democracies Snježana Prijić Samaržija

4.1 Introduction As the epistemic justification of democracy did not start as a subject epistemology has traditionally dealt with, novel debates within political epistemology and the epistemology of democracy inevitably encounter conceptual and disciplinary ambiguities. Epistemology has undergone several genuinely significant expansions of its interests and topics in the last couple of decades. From standard analytical epistemology preoccupied with defining knowledge and its presuppositions, it first expanded to discussions about epistemologically pertinent cognitive processes such as perception, reasoning, memory, and testimony (Audi 1998; Huemer 2006). This expansion was accompanied by the value shift that substituted truth monism as the only genuine epistemic value with a pluralist account of epistemic values such as understanding, epistemic achievements, empirical adequacy, or the reliable usage of available evidence (Kvanvig 2005). However, the second essential expansion comprised virtue epistemology, which shifted the target of epistemic evaluations from the truth and epistemic justifiedness of a particular proposition to epistemic agents and their intellectual virtues (Greco 2002; Sosa 2007; Zagzebski 1996). The third and final extension of epistemic topics was the long-awaited legitimacy of social epistemology, which centered on epistemological investigations and evaluations of social processes, practices, and institutions. Social epistemology concerned itself with the epistemic features of interpersonal, collective, and institutional procedures (Goldman 1987, 2010, 2020). This opening brought attention to the interdisciplinary area and the relationship between epistemic and political or ethical virtues. The final pair of extensions denoted epistemology’s final departure from the starkly defined area of the beliefs held by individual, isolated, and idealized epistemic agents to a domain of real-world epistemic practices where the epistemic agent is always and inevitably situated in distinctive epistemic circumstances. Especially as they imbue epistemology with an entirely new and desirable element of applicableness or even engagement, we must understand these extensions toward DOI: 10.4324/9781003311003-6

Critical Thinking and Trusting Experts in Real-life Democracies 71 regulative epistemology as “(…) a response to perceived [perennial] deficiencies in people’s conduct, and thus is strongly practical and social, rather than just an interesting challenge for philosophy professors and smart students. This kind of epistemology aims to change the (social) world” (Roberts and Wood 2007: 21). This extended epistemological framework focused on the agency and virtues demonstrated by epistemic agents who form beliefs and make decisions in suboptimal socially formatted epistemic conditions. And this fact about socially determined nonideal epistemic circumstances where agents acquire, consider, and revise their opinions and conclusions is the necessary framework for questioning and resolving the epistemic justification of democracy. The general attitude long was that democracy is closest to the ideal conditions for making epistemically valuable decisions, as free and equal citizens – through fair deliberation, where they engage in independent and autonomous critical thinking – make decisions that solve their problems or that are true, correct, truth-sensitive, or truth conducive.1 Democracy’s epistemic legitimacy is founded upon the assumption that it is an optimal social system precisely because it guarantees equality in sovereign belief-formation and decision-making by ensuring fairness, equality, and freedom in making political choices. According to this argument, democracy is a system that enables us to reap the fruits of collective intelligence (the wisdom of crowds). It thus generates the epistemically most valuable beliefs and decisions. So, the first step in the epistemic justification of democracy is the claim that democracy ensures the epistemic circumstances for public critical thinking liberated from coercion to consent to authority opinions or remain docile to political elites. The second step emphasizes the political, ethical, and ostensibly epistemic inferiority of the alternative expert-based approach to decision-­making and the inferiority of individual intelligence to the collective. Large groups of people are superior at solving problems than individual – or even gathered – experts, regardless of their factual knowledge and specific training, as collective public intelligence stems from their random diversity: the natural and disorganized accumulation of perspectives, interpretations, evidence, and experiences. Ostensibly, these features guarantee more valuable decisions than those garnered by experts in the isolation of their proficiency in some domains. The deficit of experts’ epistemic position – individually but also as a collective of experts – is a chronic and unfixable dearth of diversity or the necessary number of different perspectives. Experts are limited by the facts of their high education, comparable material statuses, and belonging to the category of socially recognized intellectual authorities. In conclusion, regardless of its political benefits, the mere epistemic potential of free, critical, and redundant disagreement constitutes an essential element of the epistemic quality of democratic decisions (Brennan and Landemore 2022; Landemore 2013, 2020; Peter 2008, 2016). Before introducing the problematic nature of the

72  Snježana Prijić Samaržija ostensible opposition of the epistemic virtue of critical and free-thinking compared to the epistemic vice of obedience to experts, it would be wise to question whether people’s critical and free-thinking is always and in all social epistemic circumstances virtuous.

4.2 Critical Thinking and the Epistemic Circumstances of the Culture of Ignorance As critical thinking and intellectual independence genuinely seem like an epistemic agent’s epistemic virtues, our inquiry is whether that is indeed so and whether practicing these virtues is unconditional. Is it possible that in nonideal conversational contexts, people’s free and critical thinking frequently does not have the epistemic power it has in idealized circumstances? Is it possible that factual or existing democratic circumstances or real-life democracy (Talisse 2019) are not at the same time those epistemic circumstances that transform free and critical thinking into epistemically high-quality decisions? We do not learn nor cognize in epistemically ideal conditions. As a rule, epistemic agents are not strongly motivated to seek out facts and the truth, either out of epistemic immaturity because of which they are not aware of epistemic norms and values, out of an incorrect understanding of what objective epistemic quality and facts are and how they are achieved, or out of a mere lack of time and socially conditioned existential conditions that disable such ventures (Goldman 1990, 1991; Goldman and Cox 1996). We can describe such suboptimal epistemic circumstances as a state where factual ignorance is not infrequent, as low levels of informedness and a lack of political knowledge required for informed decision-making and belief-formation are widespread. Ample empirical findings corroborate that people lacking knowledge about political issues form their beliefs without relevant knowledge about the problem at hand (Achen and Bartels 2016; Delli Carpini and Keeter 1996; Somin 2015, 2020; Sunstein 2011, 2015). Understandably, most citizens are unskilled at many pertinent topics that decisions are made about as they are uneducated in many sophisticated areas. It is equally plausible to be moved primarily by random information, interests, and background beliefs. Genuine ignorance or the existence of social epistemic differences, including prejudices and stereotypes, is not unrecognized in diversity-based approaches. Still, biases are treated as resources or desirable and potent redundant disagreements. On the other hand, there are scholars unambiguously cognizant of deficient decision-making stemming from ignorance and biases who stress that such ignorance and irrationalities possess a rational foundation in the present conditions of democratic voting. The median voter, aware of the diminutive role of their vote for the final decision and the fact they cannot alter decisions they dislike, may lack the motivation to

Critical Thinking and Trusting Experts in Real-life Democracies 73 become adequately informed, meticulously check whether their information sources are reliable and coherent, and ponder upon whether their beliefs and decisions are biases. This phenomenon is called rational ignorance, while the indisposition to detect or control one’s biases is characterized as rational irrationality (Caplan 2007; Somin 2015, 2020). Democracy’s present social moment is commonly reduced to terms such as the culture of ignorance and the cult of amateurism, concepts that underline the radical suboptimality of our epistemic conditions, more broadly known as post-truth.2 However, the term “culture of ignorance” refers not only to rational and motivational ignorance (the lack of motivation to acquire relevant information), and solely to factual ignorance, the dearth of knowledge, or even the possession of wrong answers that are considered correct or whose correctness an agent is apathetic toward (Nottelmann 2016). The culture of ignorance refers primarily to ignorance of ignorance, normative ignorance (Peels and Blaauw 2016), or ignorance’s ideology as the epistemic circumstances where the norms and standards for deeming something intellectually virtuous agency have been altered and mutated. We speak of epistemic circumstances that radically deflate the value of truth or objective epistemic value and, consequently, display distrust toward expertise or resistance to epistemic authorities – precisely as their expertise rests upon ostensibly inexistent accurate epistemic value. The culture of ignorance refers to frequent practices of confidently autonomous belief-formation about questions we possess no relevant knowledge of but remain convinced that our judgments are the only relevant knowledge. Epistemic agents knowingly rest upon their “accidental” and disorganized knowledge, and pervasive stereotypes and prejudices are the sole conclusive beliefs justified by the fact they are theirs, formed in a free-spirited bout of critical thinking. They deliberately overlook and undervalue rational and responsible exploration and reflection, followed by an utter misunderstanding of the vitality of appropriate evidential material during belief-formation and decision-making: “(…) the culture of ignorance reflects an elevation of will over reason, the loss of a credible concept of objectivity, and a radical change in democratic epistemology” (DeNicola 2017: 9). Within the culture of ignorance, the epistemic value of truth disappears underneath the epistemic significance of “my own truth” founded on critical thinking and resistance to all guidance by authorities. We are speaking about the demand to legitimize a private, personal insight into the “genuine truth,” which includes the inability to assume epistemic responsibility for one’s attitudes and the consequences they might entail for others. Here it is crucial to note that the culture of – both factual and normative – ignorance is characterized by the fact that their understanding of autonomous and critical thinking is entirely cut off from epistemic quality related to objective epistemic value or truth-sensitivity and bound solely to the reality of creating your attitude. In the culture of ignorance,

74  Snježana Prijić Samaržija critical thinking refers to an ethical, political, and epistemic “right to believe,” that is, truthfully, each citizen’s right to be heard regardless of truth, her expertise, and the expertise of other epistemic authorities (DeNicola 2017; Nichols 2017). One of the “culture of ignorance’s” key constituents is the active cognitive and affective resistance toward experts’ opinions. This opposition genuinely reflects a self-­protective attitude regarding the right to an uninformed opinion or advocating for your theories and interests. The epistemic circumstances I have termed the culture of ignorance significantly determine the epistemic quality of the collective intelligence, which affects the attitude that democracy always generates epistemically valuable decisions or that critical thinking is always a virtue. The present epistemic environment is determined by a dearth of informedness about facts relevant for decision-making and a lack of knowledge about epistemic standards or norms related to justified or epistemically responsible belief-formation and belief-revision. The epistemic justification of democracy should not be based upon idealizing or misunderstanding the severity of how genuinely suboptimal our epistemic environment for decision-making and belief-formation is. Therefore, autonomous and critical thinking – which would be an epistemic virtue in ideal or better epistemic conditions lacking the culture of ignorance – demonstrates that it depends on the context where it is practiced. In the remainder of this chapter, I will attempt to explore two questions. The first question regards a conceptual clarification of what it means to think critically. More precisely, I will inquire about what renders critical thinking an epistemic virtue. The second question is what aspect of the definition of critical thinking severs it from guidance by epistemic authorities. The retorts to these questions should help us comprehend when critical thinking practiced by free and equal individuals genuinely is an epistemic virtue and an essential constituent of democratic decision-making and when it is not.

4.3  Critical Thinking and Intellectual Autonomy The usage of the term “critical thinking” in philosophy is frequently related to John Dewey, who defined critical or reflexive thinking as an active, persistent, and thorough assessment of a belief or an assumed form of knowledge in the contexts of the available evidential material supporting the view, but also what the said belief entails and which conclusions it can produce (Dewey 1933/1998). Dewey’s clarification of what is not critical thinking is instructive for our endeavor to understand it: a brisk acceptance of an idea that might be an answer or a solution to a question or the automatic suspension of a judgment at the slightest hint of doubt. Critiques from doctrinaire political or religious ideologues are not critical thinking just as much as deducing conclusions using an

Critical Thinking and Trusting Experts in Real-life Democracies 75 algorithm (Dewey 1933/1998). Genuine critical thinking is tightly bound to the ability to assess reasons, evidence, and arguments as good reasoners (Siegel 1988). In recent elaborations, critical thinking also comprises a “critical spirit,” which includes creativity and imaginative and observational aspects of cognition (Bailin and Siegel 2003). More and more frequently, authors link critical thinking to intellectual virtues and the intellectual character of an epistemic agent to be curious, open-minded, and attentive (Baehr 2020). Critical thinking is a vital element of all intellectual practices that can be described as epistemically responsible conduct or as “epistemically good mental activity” (Baehr 2020). For instance, love toward knowledge presupposes critical discussions that include meticulous attention to the speaker, articulating sound judgments, accepting objections, and replying to them. Intellectual firmness presupposes a crucial attitude as a cure against dogmatism and gullibility. Intellectual courage, vigilance, and humility aim to critically challenge different kinds of irrationality. In more detail, intellectual humility is the same as openness toward others’ critical attitudes. Finally, phronesis is nothing other than inclusive meta-reflection that assesses the justifiedness and consequences of possible conflicts between intellectual virtues (Roberts and Wood 2007). It is interesting to note that, in its generic sense of “good thinking” or a substrate for every intellectual virtue, critical thinking is often conceptually bound to epistemic autonomy or independence from another agent’s guidance. This bind is so sturdy that independence from epistemic authorities and even resistance toward intellectual regulation is commonly considered a crucial characteristic of critical thinking. Critical thinking – or the autonomous assessment and evaluation of the relevant evidential material and arguments – is thus contra-posed to the cognitive strategy of trusting experts and accepting their positions based on their skill set, as well as the epistemic attitude of skepticism or the cessation of looking for an answer (Huemer 2005). This feature is evident in manifestations of the culture of ignorance, such as recent movements against vaccination celebrating freedom festivals, conspiracy theories, and an excessively demeaning attitude toward institutional scientific authorities. We ought to seek sources of the internal conflict between critical thinking and credulity in experts precisely in conceptualizing critical thinking with intellectual autonomy. This tension is why examining the origins and presumptions included in or hidden by this abstract link is compelling. I will show that this conceptual connection is not accidental but dominates within two conflicting epistemological traditions: standard analytic epistemology and postmodern revisionist epistemology. Traditional analytic epistemology maintains the assumption of an idealized epistemic agent who cognizes and acts in non-social epistemic circumstances (e.g., Alston 1989; BonJour 1998; Gettier 1963; Nozick 1993; Plantinga 1993). This approach entirely disregards the social dimension

76  Snježana Prijić Samaržija of cognition. Namely, an epistemic agent practices rational and justified reasoning based on reliable epistemic dispositions and capacities. Their almost unlimited cognitive abilities endowed with full access to relevant evidence always lead to epistemically optimal outcomes. In this purely normative approach, epistemology has nothing to do with belief-­ formation processes that include social stereotypes, prejudice, and other invisible and visible derivatives of social power structures. However, we reason in nonideal epistemic circumstances where epistemic agents have personal cognitive and affective biases and limitations, which their social position can often explain. As I have mentioned, critical thinking or relying solely on your mental powers is nothing like a procedure involving an idealized epistemic agent and idealized conditions in such a nonideal and realistic epistemic environment. Conversely, in a nonideal epistemic environment, knowledge is often trumped by ignorance, and epistemic agents do not understand epistemic norms and standards and more frequently rely on a self-protective insistence on their attitudes and interests. The second assumption comes from the opposite clan of postmodernists and social constructivists who maintain a revisionist perspective on the epistemological tradition and negate the existence of truth or any other objective epistemic values. Quite the reverse, they endorse the end of epistemology and traditional epistemic values in favor of merely describing different conversational practices or practicing sociology of knowledge (e.g., Barnes and Bloor 1982; Latour and Woolgar 1986; Rorty 1979; Shapin 1994). Suppose there are no epistemic standards or norms for differentiating the epistemic values of different beliefs. These circumstances would render all views equally epistemically valuable, so there is no need to rely on experts and other epistemic authorities. The sole epistemic value that remains relevant is a socially critical attitude toward relationships of domination, and primarily the dominance of experts as privileged elites or symbols of social power (Foucault 1980). These presuppositions – (i) the presupposition of the circumstances of knowledge rather than ignorance and (ii) the presumption of the absence of any objective epistemic value – each bolsters the conceptual link between critical thinking and intellectual autonomy. In both traditions, for unrelated reasons, the ideal of epistemic self-reliance as critical thinking is counterposed to trust in experts or other epistemic authorities.

4.4 The Social Dimension of Cognition: Analytic and Postmodern Epistemology In the following chapter, I will devote more detail to some of the arguments that gave rise to these presuppositions at the root of linking critical thinking and epistemic autonomy. Plato, Descartes, Locke, and Kant claimed that the search for truth is closely related to epistemic autonomy and autonomous exploration, which always stressed the philosophical

Critical Thinking and Trusting Experts in Real-life Democracies 77 ideal of a self-reliant epistemic agent dependent solely on their intellectual forces (Plato 1973; Descartes 1967/1985; Locke 1690/1975; Kant 1785/1997). Critical thinking is conceptually bound with the endeavor to assess the available arguments and evidential material autonomously carefully, instead of mechanically accepting everything from others or epistemic authorities. Deference to others, including the pertinent expert, was considered an inferior epistemic procedure deficient in intellectual weight. Suppose there is an opportunity to make a personal judgment or become the obedient imitator of somebody else’s attitude. In that case, it seems evident that the latter represents an excessive example of intellectual weakness, gullibility, epistemic dependence, intellectual laziness, and incompetence. Moreover, we could even treat it as a morally problematic act of servility or humbleness. Here we ought to note that the foundational epistemic reason for this attitude stems from the mentioned presumption about the idealized epistemic agent, the theory’s traditional focus on their cognitive mechanisms, and a general epistemological restraint toward assessing the social sources of knowledge. We customarily considered trusting another informer’s testimony less reliable than relying on our observations. As a rule, other people and their testimonies are less reliable because there is always the open option of deceit. The said dishonesty can be either deliberate or accidental; we often cannot gauge the level of their competence for the transmitted judgments, they possess interests that might compel them to make us concede to their false attitudes, they can be capricious or malicious, or they might be burdened by emotional rationales we are entirely unaware of. These are reasons to question the trustworthiness of another epistemic agent’s attitudes and testimonies. Consequently, they are also reasons to prefer the practice of epistemic autonomy (E. Fricker 2006). On the other hand, on the side held by postmodern philosophers and social constructivists, the most prevalent attitude is that the epistemic ideals of objectivity, truth, and neutrality have no resilient theoretical basis. Truth is merely a social fabricate and a construct crafted by social elites rather than an objective epistemic value. This element of the revisionist epistemological presupposition regarding the social construction of objectivity and the epistemic significance of truth represents a theoretical underpinning for the attitude about the need to rely on your forces and reject domination by any authority, including experts. If there is no objective epistemic value, experts lose their statuses of epistemic authority and their beliefs, just like everyone else’s, merely reflect their interests and socio-political positions. In such cases, the desired ethical and political egalitarianism entails epistemology. The virtue of autonomy or relying solely on your reflections thus gains precedence over deference to experts, primarily due to the repulsion of the idea of being unjustifiably dominated by another person or that we must concede to the beliefs,

78  Snježana Prijić Samaržija desires, or attitudes of somebody else. This value system is equally true for our intellectual conduct and views on morals or politics (Foucault 1980, 1991). Both presuppositions – one about the socially isolated epistemic agent endowed with immaculate cognitive abilities and the other that there is no objective epistemic value – are, in a different manner, reflected in contemporary debates about the epistemic justification of democracy that derive epistemic value from just democratic procedures. Given that there is no objective epistemic value, a collective of sufficiently informed and independent individuals – although not perfect – will indeed arrive at the most epistemically valuable belief or decision. Both the assumption about the wisdom of crowds and Condorcet’s Jury Theorem contain the idealized attitude that a collective of people bearing different stances will generate higher epistemic quality than the stances of isolated individuals. In more recent debates about the epistemology of democracy – especially in discussions about deliberative democracy, but also about pop-cultural phenomena such as Wikipedia and opinion surveys – discussants frequently hold that the benefits which citizens lose in the epistemic value of an individual (expert) belief is compensated through amassing a diversity of dissimilar attitudes (Goodin and Spiekermann 2018). The ethically and politically valuable inclusion of large collectives and the justifiedness of not favoring anyone thus emphasize the epistemic value encompassed by different perspectives and evidential material (Landemore 2020; Peter 2008, 2016). At the same time, the authors assume that fairness and inclusivity will automatically meld into epistemic quality, which cannot be defined outside the belief-formation procedure and is not objective or external to the very social process. In short, it is commonly held that it is a better epistemic choice to practice a crowd’s autonomous (and cognitively biased) critical thinking than concede to expert opinion, as a collective’s participation and deliberation will yield epistemically higher quality solutions than an expert’s intellectual guidance. It appears, then, that the dominant understanding of the nature of critical thinking as intellectual self-reliance stems both from (i) a shift from the social dimension of cognition and blindness toward how epistemic circumstances affect the epistemic quality of decisions and (ii) overstating that very social dimension, which then negates the objective epistemic value of different beliefs. Both approaches lack a more complex comprehension of cognition’s social dimension, which would accept the epistemic agent’s social situatedness within specific epistemic circumstances but would not reduce the entire value of the cognitive agency to procedures or power relations. As an already well-established discipline, social epistemology has enabled us to take a more precise look at the social dimension of cognition and maintain a position between traditionalism (analytical epistemology) and reductionism (postmodern nihilistic epistemology)

Critical Thinking and Trusting Experts in Real-life Democracies 79 (M. Fricker 1998). This more vigilant analysis of the social dimension’s role in belief-formation unmasked traditional analytical epistemology’s ardent focus on the cognitive conditions where an entirely isolated individual epistemic agent can acquire knowledge. The epistemic agent is, in fact, not an asocial creature, and social circumstances are not wholly irrelevant to their belief-formation. While traditional analytic epistemology deliberately evaded the exploration of epistemic environments, movements such as postmodernism, the sociology of knowledge, and cultural studies gained popularity – given their hermetic wording, quite surprising – in intellectual circles specifically by emphasizing this deficit in conventional accounts of knowledge. Again, it is precisely analytical social epistemology – which did not ignore this popular albeit relevant epistemological trend but embraced it – that allowed us to comprehend their desire to overthrow epistemology’s normative dimension, and especially the objective epistemic value of a belief’s validity in favor of exploring the social conditions of belief-formation exclusively. Postmodernism’s overwhelming emphasis on knowledge’s social character led to an unacceptable reduction of opinions on the social conditions of their construction, demoted the epistemic agent to a mere function of power relations, and downgraded their beliefs to sociocultural constructs so that epistemology’s only task was to deconstruct or analyze these social conditions. Given that both traditions do not treat the social dimension of knowledge and the impact of epistemic circumstances on the quality of beliefs in an appropriate manner, social epistemology’s novel framework invites us to re-evaluate the conceptual link between critical thinking and intellectual autonomy, or the attitude that critical thinking is the complete opposite to deference to experts.

4.5  Critical Thinking and Epistemic Authorities Is intellectual deference to experts indeed an epistemic vice? I am commencing from, I assume, the universally acceptable presumption that experts are the best available individual guides to epistemic quality. Of course, experts do not possess truths. They do not have an objective and neutral insight into the posed questions, yet they enjoy more factual and normative knowledge than other epistemic agents. They are better at tracking truth, detecting erroneous beliefs, solving problems, or practicing epistemic virtues than the median epistemic agent. Experts not only possess more (highly specialized) factual knowledge and data but are also superior at structuring, contextualizing, and applying it. Moreover, experts are trained to use appropriate and reliable methods when researching, articulating arguments, resolving difficulties, and facing alternative solutions (Goldman 2001). In short, compared to other epistemic agents, they are closer to normative knowledge and farther

80  Snježana Prijić Samaržija removed from normative ignorance. They are undoubtfully closer to the side of learning and farther from conceding to ignorance. Consequently, regardless of the required insights into the epistemic reach of expert attitudes, it would be genuinely deleterious to question their general reliability in favor of the mindset that everyone can develop an equally valuable attitude by merely reflecting upon a given question.3 Conversely, non-experts are, by definition, epistemically dependent on experts to resolve their problems and aid them in the belief-formation process. Analyzing the social sources of knowledge has become entirely legitimate within social epistemology and the epistemology of testimony, unlike traditional epistemological discussions. Accordingly, while maintaining all the dilemmas regarding the reliability of accepting other people’s testimonies, it is a brute fact we are inevitably epistemically dependent on the testimonies of others, and in particular expert testimonies. It is another brute fact that most of our beliefs have been formed owing to other people and that our epistemic successes rest primarily on accepting expert opinions. Of course, it is necessary to define the conditions when we should embrace expert testimonies clearly, but merely rejecting or devaluing them in favor of relying solely on our epistemic forces would not bring us to truth but a circumstance of utter epistemic impoverishment (Reid 1785/1983). Second, experts are, like everyone else, susceptible to value-laden, ideological, theoretical, and other person presuppositions that can contaminate their approach to a specific question. However, they are not disproportionately more vulnerable to such influences than others. On the contrary, there are grounds to believe that experts are, due to their lengthy education, better epistemically equipped to reflect upon and recognize their limitations and deal with them in the best manner oriented toward divulging optimal solutions. While the critique of the myth of the neutral expert is not unfounded, its popularity far exceeds its justification. This stance’s broad support is certainly a motive to wonder to what extent the attitude about biased experts is precisely an attitude founded on a, perhaps justified, extra-epistemological fear of privileging elites. However, it is bolstered by the power-hunger of specific perilous anti-­ enlightenment currents instead of the unquestionable value of expertise. Either way, there does not seem to be a rational epistemic reason for excluding expert viewpoints. On the other hand, it becomes vital to make an even more precise distinction between the political and the epistemic domain when it comes to the tension between elitism and egalitarianism. While respecting the sentiments of injustice and resentment toward being dominated by others, we need to realize that in the epistemic domain, the plea for egalitarianism rapidly uncovers itself as intellectually irresponsible epistemic egoism. Zagzebski has clearly articulated the absurdity of epistemic egalitarianism: “We all admit that some persons have more inherited intelligence

Critical Thinking and Trusting Experts in Real-life Democracies 81 than others or have honed their intellectual skills to a greater degree than others, or have greater access to information in some domains, but we take for granted that these differences are not important for the way epistemologists approach epistemological issues, and they are not enough to ground authority in the domain of belief in an interesting and robust sense” (Zagzebski 2012: 6). In other words, there is something epistemically unjustified in the fact that we are all aware that some of us are better guides to truth but remain opposed to being dominated by them and prefer to retain personal beliefs that are likelier to be incorrect. Although it might be a reasonable demand from perspectives external to epistemology in highly particular circumstances, it is evident that such an epistemic decision is not intellectually responsible. The sole way to insist on your personal belief – because it is ours, regardless of how valuable it might be – is to embrace the epistemically revisionist or even nihilistic attitude that there are no such things as epistemically better or worse beliefs. Nonetheless, this assumption is far removed from being both a commonsensical attitude and a matter of coherent and acceptable argumentation. From the definition of critical thinking, it follows that debasing expertise or expert knowledge is uncritical thinking. Finally, it is ill-advised to unconditionally embrace the stance that the “wisdom of crowds” automatically transforms political value into epistemic. An aggregate of a collective’s stances and mere participation that comprises and assembles a multitude’s dissimilar beliefs does not possess an inherent feature that would guarantee epistemic quality (Goldman 1991, 1996, 1999, 2010). Aggregating many people’s stances is not an “invisible hand” that generates an epistemically high-quality belief if the starting attitudes are not adequately epistemically valuable. There is much empirical evidence in favor of a specific pessimism regarding the multitude’s stances precisely due to inadequate informedness, low cognitive capacities, and different motives that might give rise to unacceptable epistemic stances (Ahlstrom-Vij 2020; Brennan 2020). It has repeatedly been shown that ignorance entails a forceful meta-perspective. Those with less knowledge are less critical of their stances, competencies, and limitations. The capacity to form correct beliefs is disproportional to epistemic self-confidence and the readiness to revise one’s attitudes when confronted with different or more complex perspectives (Kruger and Dunning 1999). Besides, many social limitations – such as the so-called hegemony of common knowledge – also probe the epistemic quality of collective discussions. Namely, the individuals partaking in a debate are likely to form their beliefs independently and regardless of another agent’s influence based on the same data. Laypersons commonly share the same information that everyone can comprehend in the circumstances of ignorance. They are repeated throughout multiple shared debates. They gain an additional allure and significance they objectively do not possess.

82  Snježana Prijić Samaržija Finally, this lowest common denominator known and acceptable to all participants, including experts and non-experts, becomes the conclusive joint belief or judgment (Prelec, Seung, and McCoy 2017). Such an outcome is far removed from the epistemic quality we could have attained in a manner different from mere belief aggregation. In short, “the wisdom of crowds” – despite its immense democratic epistemic potential in idealized circumstances of comprehensive cognitive preparation and affective disinterest – ought to be evaluated in the light of our actual epistemic conditions. The wisdom of crowds will possess significant epistemic relevance only because all participants already flaunt epistemically high-quality beliefs, remain open to other epistemically valuable ideas, and are cognizant of their limitations. After all, classifying deference to epistemic authorities as an epistemic vice is unjustified if we care about epistemic quality in realistic epistemic circumstances. This resistance toward authorities stems from an uncritical transition of political and ethical motivations into the epistemic domain. We can mark it as political instrumentalism that sacrifices epistemic values to the political. It is further nourished by the epistemological limitations of both influential epistemological traditions. Given the necessity of optimally positing epistemic agency in the social context of factual epistemic circumstances, the remainder of my chapter will attempt to demonstrate that there is no need to stress the tension between these two epistemic strategies – critical thinking and deference to experts. Quite the reverse, I will try to show that trust in experts should not only be forcefully pitted against critical thinking but that it is a crucial element of a proper understanding of it.

4.6  Trusting Experts as Critical Thinking Epistemic autonomy or reliance on one’s intellectual strengths is a responsible epistemic agent’s indubitable virtue, but only if they reside in idealized epistemic circumstances. Practicing intellectual autonomy is not necessarily a virtue under the assumption of factual, normative, cognitive, and affective ignorance. Epistemic autonomy or reliance on one’s personal free and critical thinking is purely not an unconditional virtue. In specific epistemic circumstances, such as our real-life epistemic conditions, it is neither intellectually responsible behavior nor does it lead to truth. When defined as vigilant, systematic, reason-sensitive, and evidence-based reflection, critical thinking cannot be conceptually linked to intellectual autonomy and is understood as an unyielding exclusion of any form of intellectual guidance. While critical thinking includes intellectual independence, which refers to independent conscientiousness, vigilant understanding, questioning, and reflecting upon an issue, it can sometimes result in the epistemic decision to trust someone we judge to know more and better than we do. With careful reflection upon

Critical Thinking and Trusting Experts in Real-life Democracies 83 arguments, reasons, and evidential material, critical thinking does not necessarily preclude deference to experts. Intellectual autonomy and thus derived critical thinking have epistemic value solely if they include the possibility of intellectually autonomously deciding to trust experts. There can be no epistemically responsible critical thinking founded on insisting on your attitude, neglecting relevant evidence and reasons, and repudiating intellectual guidance which can support them. If we know less about the subject than some experts, our stance cannot possibly possess the same epistemic quality. It is also not epistemically responsible for being oblivious to your intellectual deficiencies and accepting the illusion of personal superiority regardless of comparatively worse knowledge and education. Quite the reverse, deference to experts who represent epistemic authorities in each domain is necessary for an epistemically responsible practice of epistemic autonomy. “(…) Epistemic authority does not refer only to experts (people who are reliable sources of information in some domain) but to the more complex situations in which it is a person in which beliefs (and testimonies) we trust that is better to believe her than me based on conscientious self-reflection” (Zagzebski 2012: 5) In short, it seems vital to properly understand the roles of the social context of epistemic agency and reality-based epistemic circumstances. It is equally essential to overcome the limitations of the traditionalism of the conventional analytic approach and postmodernity’s ardent reductionism, aiming to acquire a more complex and comprehensive view of epistemic virtues and vices. In this way, we gain a better understanding of the concept of critical thinking and intellectual autonomy that does not exclude deference to epistemic authorities. Thus understood, free and critical thinking can be an unconditional epistemic virtue in contemporary epistemic circumstances. Conversely, in opposition to guidance by epistemic leaders, free and critical thinking reveals itself as a vital constituent of the culture of ignorance. An epistemic justification of democracy that overlooks the culture of ignorance’s radically suboptimal epistemic circumstances and that displays a propensity to bar experts’ comparatively superior knowledge from decision-making processes in favor of collective intelligence rests on an entirely fallacious understanding of the virtue of critical thinking. Critical thinking practiced by free and equal individuals is an epistemic virtue and democratic decision-making’s essential constituent solely when it is open to the option of guidance by epistemic authorities.

4.7 Conclusion A vigilant re-evaluation of the virtue of critical thinking, autonomous reflection, and epistemic autonomy has shown that we must not define them as mere reliance on one’s intellectual capacities in opposition to trust in experts but as a demand for a heightened sensibility to the epistemic

84  Snježana Prijić Samaržija quality of the resulting beliefs. In reflection, we must not understand epistemic autonomy as epistemic egoism or egocentrism but as a space of individual judgment about attaining optimum views. Epistemic deference to others is not irresponsible if founded on a conscientious assessment of the belief’s final quality. We do not become less critical, independent, or autonomous if we decide that deference to experts is justified. We do not automatically turn more independent or autonomous by refusing to understand the definition of trust and epistemic dependence on experts. The essence of epistemic virtue – epistemic responsibility or the search for truth – is epistemic quality as the highest epistemic achievement, and not acceptance or non-acceptance to have our epistemic lives regulated by others, including legitimate epistemic authorities. Finally, we must keep in mind that the virtues of curiosity and learning are also grounded on epistemic deference to experts. The rejection of trust in experts shadows any openness to understanding and education. In opposition to the hyper-individualism of the quasi-solipsism offered by traditional analytic epistemology and postmodernity’s socio-political determinism or social fatalism, we must consider the novel idea of “independent dependence.” An epistemic agent to whom we can ascribe epistemic virtues is a person who wishes to be appropriately “regulated” by others while remaining an independent critical thinker and analyst. If critical thinking is an epistemic or intellectual virtue, it must incorporate elements of intellectual deference and humility without slipping into the epistemic vice of irresponsible dependence on others (Roberts and Wood 2007). As paradoxical as it may sound, the culture of ignorance stems primarily from an understanding of critical thinking as an unconditional practice of independent reflection or intellectual autonomy. Moreover, it appears as if it has been solidified that independent thinking that excludes deference to experts is always an epistemic virtue. In contrast, trust in experts or epistemic authorities is inevitably an intellectual deficit. It is difficult to imagine both traditional analytic epistemologists and postmodern thinkers perceiving a public display of the culture of ignorance as a proper practice of independent reflection and epistemic autonomy. When they spoke of the need for epistemic independence and intellectual self-reliance in conflict with authorities, they were undoubtfully not thinking about justifying ignorance and neglecting data and rational discussions or epistemic exchanges. However, it now appears justified to seek one of the culture of ignorance’s theoretical underpinnings – although undoubtedly not the only one, given various political and social factors – in insisting on each epistemic agent’s unconditional independence and epistemic autonomy. To conclude, the epistemic justification of democracy must reveal the culture of ignorance as comprehensive epistemic circumstances that do not contribute to the epistemic quality of the decisions generated through collective intelligence. In addition, we must appropriately comprehend

Critical Thinking and Trusting Experts in Real-life Democracies 85 the significance of expertise in practicing the virtue of critical thinking. When epistemically justifying democracy and decisions made employing democratic deliberation, we must perceive deference to expert knowledge as a crucial constituent of critical thinking and intellectual autonomy.

Notes





1. In this chapter, I use the concepts of “truth,” “truth-sensitivity,” or a procedure’s epistemic value of being “truth-conductive” as generic terms that refer primarily to the attitude there is objective epistemic value or epistemic quality. It is beyond my current task to discuss truth monism, pluralism, or epistemic values that comprise justifiedness, rationality, understanding, coherence with the available evidence, and problem-solving. I will also refrain from the debate on whether the pluralism of epistemic values jeopardizes the value of truth. It is customary to understand epistemic values in the sense of epistemic achievements articulated through the traditional epistemic value of truth (although an epistemic achievement genuinely comprises a far broader spectrum and includes all the mentioned epistemic accomplishments). The term “truth” in this chapter relates to the more comprehensive framework of virtue epistemology, which does not primarily bind truth to a proposition’s feature but to epistemic agency and motivation striving to attain true beliefs. 2. The culture of ignorance’s more common manifestations is various pseudoscientific movements, such as those against vaccination and teaching creationism at schools, conspiracy theories, and climate change deniers. Likewise, we must acknowledge a budding skepticism towards science with all the features of a new crisis of enlightenment. Skepticism towards science correlates with political ideologies, religious attitudes, moral beliefs, lifestyles, and fundamental scientific insights. Finally, there are open proclamations of anti-intellectualism, criticisms of “bookish” knowledge, and repudiations of institutional standards of expertise in favor of glorifying informal amateurism. Many social circumstances, such as social media, have contributed to this crisis through the phenomena of informational epistemic bubbles and more malicious echo chambers (Samaržija 2023). The culture of resentment towards evidence-based methodologies, which equalizes veridic and fake reports to guarantee the leveling of every interpretation’s epistemic value, is another paramount constituent of the phenomenon I have termed the culture of ignorance. 3. I must define specific limitations regarding the persons possessing an expert status. First, not all experts are “real” experts. There are persons in public positions that ought to be occupied by experts who do not possess the pertinent epistemic qualities. Such individuals are not objective but merely reputational experts. Second, an expert is never an expert concerning everything, as there is no such thing as a universal expert. An expert in one area is always a layperson in another. Third, the definition of an expert does not entail the privileged social position of a fundamental authority but exclusively derived authority based on trust in the area where there are knowledge-based results. According to our understanding, experts are not a privileged social elite and should not be conflated with political positions (Prijić-Samaržija 2017, 2018). However, there is no doubt that non-experts can generally recognize who is an objective expert in the field, whose expertise is merely reputational, and who is biased by their elite status, ideology, values, or like (Collins and Evans 2007; Goldman 2001; Guerrero, 2016; Somin 2020).

86  Snježana Prijić Samaržija

References Achen, Christopher H., and Bartels, Larry M. 2016. Democracy for Realists: Why Elections Do Not Produce Responsive Government. Princeton: Princeton University Press. Ahlstrom-Vij, Kristoffer. 2020. “The Epistemic Benefits of Democracy: A Critical Assessment.” In The Routledge Handbook of Social Epistemology, edited by Miranda Fricker, Peter J. Graham, David Henderson, and Nikolaj J. L. L. Pedersen, 406–415. New York: Routledge. Alston, William P. 1989.  Epistemic Justification: Essays in the Theory of Knowledge. Ithaca, NY: Cornell University Press. Audi, Robert. 1998. A Contemporary Introduction to the Theory of Knowledge. New York: Routledge. Baehr, J. 2020. “Intellectual Virtues, Critical Thinking, and the Aims of Education.” In The Routledge Handbook of Social Epistemology, edited by Miranda Fricker, Peter J. Graham, David Henderson, and Nikolaj J. L. L. Pedersen, 447–456. New York: Routledge. Bailin, Sharon, and Siegel, Harvey. 2003. “Critical Thinking.” In The Blackwell Guide to the Philosophy of Education, edited by Nigel Blake, Paul Smeyers, Richard D. Smith, and Paul Standish, 181–193. Malden, MA: Blackwell Publishing. Barnes, Barry, and Bloor, David. 1982. “Relativism, Rationalism, and the Sociology of Knowledge.” In Rationality and Relativism, edited by Martin Hollis and Stephen Lukes, 21–47. Cambridge, MA: MIT Press. BonJour, Laurence. 1998. In Defense of Pure Reason: A Rationalist Account of A Priori Justification. Cambridge: Cambridge University Press. Brennan, Jason. 2020. “Epistemic Democracy.” In The Routledge Handbook of Applied Epistemology, edited by David Coady and James Chase, 88–100. New York: Routledge. Brennan, Jason, and Landemore, Helene. 2022. Debating Democracy: Do We Need More or Less? Oxford: Oxford University Press. Caplan, Bryan. 2007. The Myth of the Rational Voter. Princeton: Princeton University Press. Collins, Harry, and Evans, Robert. 2007. Rethinking Expertise. Chicago: University of Chicago Press. Delli Carpini, Michael, and Keeter, Scott. 1996. What Americans Know about Politics and Why It Matters. New Haven, United States: Yale University Press. DeNicola, Daniel R. 2017. Understanding Ignorance. Cambridge, MA: MIT Press. Descartes, Rene. 1967/1985. The Philosophical Writings of Descartes, Vol. 1. Translated by John Cottingham, Robert Stoothoff, and Dugald Murdoch. New York: Cambridge University Press. Dewey, John. 1933/1998. How We Think: A Restatement of the Relation of Reflective Thinking to the Educative Process. Boston: Houghton Mifflin. Foucault, Michel. 1980. Power/Knowledge: Selected Interviews and Other Writings 1972-1977. Edited by Colin Gordon. New York: Pantheon. Foucault, Michel. 1991. Discourse and Truth: The Problematization of Parrhesia. Edited by Joseph Pearson. Evanston: Northwestern University Press.

Critical Thinking and Trusting Experts in Real-life Democracies 87 Fricker, Elizabeth. 2006. “Testimony and Epistemic Autonomy.” In The Epistemology of Testimony, edited by Jennifer Lackey and Ernest Sosa, 225–252. Oxford: Oxford University Press. Fricker, Miranda. 1998. “Rational Authority and Social Power: Towards a Truly Social Epistemology.” Proceedings of the Aristotelian Society 98(2): 159–177. Gettier, Edmund L. 1963. “Is Justified True Belief Knowledge?” Analysis 23(6): 121–123. Goldman, Alvin I. 1987. “Foundations of Social Epistemics.” Synthese 73(1): 109–144. Goldman, Alvin I. 1990. Knowledge in a Social World. Oxford, United Kingdom: Oxford University Press. Goldman, Alvin I. 1991. “Epistemic Paternalism: Communication Control in Law and Society.” Journal of Philosophy 88: 113–131. Goldman, Alvin I. 2001. “Experts: Which Ones Should You Trust?” Philosophy and Phenomenological Research 63: 85–109. Goldman, Alvin I. 2010. “Why Social Epistemology Is Real Epistemology?” In Epistemic Value, edited by Adrian Haddock, Allan Millar, and Duncan Pritchard, 1–28. Oxford: Oxford University Press. Goldman, Alvin I. 2020. “The What, Why, and How of Social Epistemology.” In The Routledge Handbook of Social Epistemology, edited by Miranda Fricker, Peter J. Graham, David Henderson, and Nikolaj J. L. L. Pedersen, 10–20. New York: Routledge. Goldman, Alvin I., and Cox, James C. 1996. “Speech, Truth, and the Free Market for Ideas.” Legal Theory 2: 1–32. Goodin, Robert E., and Spiekermann, Kai. 2018. An Epistemic Theory of Democracy. Oxford: Oxford University Press. Greco, John. 2002.  “Virtues in Epistemology.”  In Oxford Handbook of Epistemology, edited by Paul Moser, 287–312. New York: Oxford University Press. Guerrero, Alexander. 2016. “Living with Ignorance in a World of Experts.” In Perspectives on Ignorance from Moral and Social Philosophy, edited by Rik Peels, 156–185. London: Routledge. Huemer, Michael. 2005. “Is Critical Thinking Epistemically Responsible?” Metaphilosophy 36(4): 522–531. Huemer, Michael (ed.). 2006. Epistemology: Contemporary Readings. New York: Routledge. Kant, Immanuel. 1785/1997. Groundwork of the Metaphysics of Morals. Translated and edited by Mary Gregor. New York: Cambridge University Press. Kruger, Justin, and Dunning, David. 1999. “Unskilled and Unaware of It: How Difficulties in Recognizing One’s Incompetence Lead to Inflated SelfAssessments.” Journal of Personality and Social Psychology 77(6): 1121–1134. Kvanvig, Jonathan L. 2005. “Truth Is Not the Primary Epistemic Goal.” In Contemporary Debates in Epistemology, edited by Matthias Steup and Ernest Sosa, 285–295. Malden, MA: Blackwell. Landemore, Helene. 2013. Democratic Reason: Politics, Collective Intelligence, and the Rule of the Many. Princeton: Princeton University Press. Landemore, Helene. 2020. Open Democracy: Reinventing Popular Rule for Twenty-First Century. Princeton: Princeton University Press.

88  Snježana Prijić Samaržija Latour, Bruno, and Woolgar, Steve. 1986. Laboratory Life: The Construction of Scientific Facts. Princeton: Princeton University Press. Locke, John. 1690/1975. An Essay Concerning Human Understanding. Edited by Peter H Nidditch. Oxford: Oxford University Press. Nichols, Tom. 2017. The Death of Expertise. New York: Oxford University Press. Nottelmann, Nikolaj. 2016. “The Varieties of Ignorance.” In The Epistemic Dimensions of Ignorance, edited by Rik Peels and Martijn Blaauw, 33–56. Cambridge: Cambridge University Press. Nozick, Robert. 1993. The Nature of Rationality. Princeton: Princeton University Press. Peels, Rik, and Blaauw, Martijn (eds.). 2016. The Epistemic Dimensions of Ignorance. Cambridge: Cambridge University Press. Peter, Fabienne. 2008. Democratic Legitimacy. New York: Routledge. Peter, Fabienne. 2016. “The Epistemic Circumstances of Democracy.” In The Epistemic Life of Groups, edited by Michael Brady and Miranda Fricker, 133–149. Oxford: Oxford University Press. Plantinga, Alvin. 1993. Warrant: The Current Debate. Oxford: Oxford University Press. Plato, 1973. Theaetetus. Oxford: Oxford University Press. Prelec, Dražen, Seung, Sebastian H., and McCoy, John. 2017. “A Solution to the Single-Question Crowd Wisdom Problem.” Nature 541(7638): 532–535. Prijić-Samaržija, Snježana. 2017. “The Role of Experts in a Democratic DecisionMaking Process.” Ethics & Politics 19(2): 229–246. Prijić-Samaržija, Snježana. 2018. Democracy and Truth: The Conflict Between Political and Epistemic Virtues. Udine: Mimesis International. Reid, Thomas. 1785/1983. Essay on the Intellectual Powers of Man. In Thomas Reid’s Inquiry and Essays, edited by Ronald Beanblossom and Kieth Lehrer. Indianapolis: Hacket. Roberts, Robert, and Wood, Jay. 2007. Intellectual Virtues: An Essay in Regulative Epistemology. Oxford: Oxford University Press. Rorty, Richard. 1979. Philosophy and the Mirror of Nature. Princeton: Princeton University Press. Samaržija, Hana. 2023. “The Epistemology of Fanatism: Echo Chambers and Epistemic Bubbles”. In The Philosophy of Fanatism, edited by Leo Townsend, Hans Bernhard Schmid, Michael Staudigl, and Ruth Rebecca Tietjen. New York: Routledge. Shapin, Steven. 1994. A Social History of Truth. Chicago: University of Chicago Press. Siegel, Havey. 1988. Educating Reason: Rationality, Critical Thinking, and Education. New York: Routledge. Somin, Ilya. 2015. “Rational Ignorance.” In Routledge International Handbook of Ignorance Studies, edited by Matthias Gross and Linsey J. McGoey, 282–293. London: Routledge. Somin, Ilya. 2020. “Trust and Political Ignorance.” In  Trust: A Philosophical Approach, edited by Adriano Fabris, 153–170. Cham: Springer. Sosa, Ernest. 2007. A Virtue Epistemology. Oxford: Oxford University Press. Sunstein, Cass R. 2011. Going to Extremes: How Like Minds Unite and Divide. New York: Oxford University Press.

Critical Thinking and Trusting Experts in Real-life Democracies 89 Sunstein, Cass R. 2015. Wiser: Getting Beyond Groupthink to Make Groups Smarter. United States: Harvard Business Review Press. Talisse, Robert B. 2019. Overdoing Democracy: Why We Must Put Politics in Its Place. Oxford: Oxford University Press. Zagzebski, Linda Trinkaus. 1996. Virtues of the Mind: An Inquiry into the Nature of Virtue and the Ethical Foundations of Knowledge. Cambridge, United Kingdom: Cambridge University Press. Zagzebski, Linda Trinkaus. 2012. Epistemic Authority: A Theory of Trust, Authority, and Autonomy in Belief. Oxford: Oxford University Press.

5

The Dangers of Disinformation Åsa Wikforss

5.1 Introduction Democracy is in decline worldwide. In its Democracy Report 2022, the V-Dem Institute concludes that the level of democracy enjoyed by the average global citizen in 2021 is down to 1989 levels.1 The wave of democratization that ensued after the fall of the Soviet Union reached its peak in 2012 and is now down to the lowest levels in over twenty-five years. In 2021, a record number of nations were moving toward autocracies, harboring 36% of the world’s population. This includes six of the twenty-­ seven EU member states and three EU neighbors to the East. It is striking that the period of democratic backsliding, starting around 2010, coincides with the appearance of a new media landscape. The old ways of distributing information about current events that emerged during the 20th century, in step with the development of modern democracy, have been turned upside down. The new information technology has caused a severe loss of advertisement revenue for traditional media, which has led to the death of local journalism in particular. In contrast, countless other sources have emerged on the scene, claiming to spread information about current events – alternative news sources, public Facebook pages, blogs, and web pages run by political actors. In addition, some social media functions as a source of information (as when someone posts a video from a war scene) and as a distributor of the information provided by other sources. The result is that the traditional gatekeepers have been circumvented and that the choice of sources increasingly falls on the individual, leaving plenty of room for psychological biases, prior beliefs, and emotions to determine what content we consume. Add to this the fact that how things spread on social media, where we primarily consume our news, is determined by algorithms designed to feed us content that keeps us on the platforms. Indeed, it has been suggested that these algorithms are the new gatekeepers of the flow of information. They are not designed to prioritize reliable content, as the old gatekeepers were, but content that keeps social media users engaged. DOI: 10.4324/9781003311003-7

The Dangers of Disinformation 91 Whether this striking correlation between the decline of democracy and the emergence of a new information landscape is one of causation is a complex empirical question that we cannot determine without extensive empirical studies. It has been argued that explaining democratic decline requires appealing to factors such as demographic changes, increasing economic inequality, and failing welfare systems.2 This seems very plausible, as radical societal changes are rarely the result of one factor alone, and the harms done by disinformation need to be understood in a more extensive societal context. However, some scholars further suggest that focusing on the so-called post-truth era is a mistake. After all, they argue, what does truth have to do with democracy? In this chapter, I shall take on this challenge. I shall reflect on the role of truth, or knowledge, in democracy and examine how mis- and disinformation can harm democracy. I shall argue that there are reasons to take very seriously the hypothesis that the threats to knowledge resulting from the new media landscape also threaten democracy. Mis- and disinformation also play a central role in the ongoing autocratization process.3 The chapter is structured as follows. I begin by addressing the challenge that knowledge does not have much to do with democracy (Section 5.2) and the related suggestion that there is something profoundly undemocratic about the very idea of truth (Section 5.3). I then examine some central harms done by disinformation to democracy. After having clarified the main concepts of mis- and disinformation (Section 5.4), I consider two types of injuries: the use of disinformation to influence voters (Section 5.5) and the role played by disinformation in contributing to irresolvable disagreements (Section 5.6).

5.2  The Concept of Democracy “In a post-truth world, there is no democracy.” That is the headline of an opinion piece published in the Washington Post in 2018.4 The article is one of many on the same theme following the inauguration of Donald Trump as president in 2016. The widespread dissemination of disinformation and Trump’s transience regarding the truth gave cause for concern. There was talk of a truth crisis, and the peculiar term post-truth was named word of the year by the Oxford English Dictionary in the same year.5 The word was also defined in a way linked to democracy, reflecting the prevailing circumstances in which objective facts are less influential in shaping public opinion than appeals to emotion and personal belief. When French President Emmanuel Macron addressed a joint meeting of Congress in April 2018, he expressed this concern. Before continuing, he stressed that we must protect democracy and fight against fake news: “Without reason, without truth, there is no real democracy – because democracy is about true choices and rational decisions.”6

92  Åsa Wikforss As for Trump, we now know what was to come. He disseminated approximately 30,000 false or misleading statements during his four years as president. His presidency was characterized by constant attacks on media outlets that were undesirable to him, judges who ruled against him, and researchers who published uncomfortable scientific truths.7 The whole affair ended very much in keeping with the entire presidency with a lie so big and so harmful that it is uncertain whether American democracy will survive – that during the 2020 election, the Democrats engaged in widespread electoral fraud. At the same time, the question about the role of truth in democracy is complicated. Why should reason and truth be so central to democracy? Indeed, democracy is about the will of the people. These questions constitute the starting point of Johan Farkas and Jannick Schou’s book Post-Truth, Fake News and Democracy (2020), where they stake out their opposition to the perspective on democracy they believe was expressed by Macron in his address to Congress. The purpose of their book is to critically examine what they describe as the dominant narrative around democracy and the post-truth era and to explore the assumptions that are made vis-à-vis the relationship between democracy and truth. Among these is the premise that the concepts of reason, truth, reality, and democracy are essentially interconnected.8 Adherents to this school of thought, they argue, forget that democracy is about popular government and about giving people political influence in their day-to-day existence. Farkas and Schou suggest that the critical thing for a well-functioning democracy is not that it is based on reason and truth but that it enables various groups and political projects to make themselves heard. Democracy, they write, is about different visions for how society should be organized, it is about affect, emotions and feelings.9 Democracy is not just about reason and truth. It is possible to envisage a “truth state” whose primary purpose is to produce as much knowledge as possible and to make all decisions based on this knowledge without necessarily being a democracy. A state of this kind would be an out-andout technocracy or quite simply a dictatorship led by an omniscient despot (if there were any in existence). The truth can, therefore, never be sufficient for democracy precisely because a democratic form of government requires widespread influence. If Farkas and Schou object to the thesis that truth is sufficient for democracy, they are attacking a straw man. However, another more interesting assertion is that truth, knowledge, and reason are necessary for democracy – they are a prerequisite for democracy to work. This is probably what Macron is getting at when he says that without truth and reason, we have no democracy.10 To examine this statement, we must start by further clarifying the concept of democracy. According to Farkas and Schou, the debate around democracy and

The Dangers of Disinformation 93 truth stems from different perceptions of what characterizes a democracy: it is a contradiction between democracy in the sense of being a popular government on the one hand and liberal democracy with roots in the Enlightenment on the other. Looking at research in the field of political science, we quickly realize that things are more complex than that. Rather than just two concepts of democracy, there are several, and they are all based on the idea that democracy is – in some sense – about popular government. In its measurements of the development of democracy, the V-Dem Institute employs five different concepts of democracy. Two of these are fundamental: electoral democracy and liberal democracy.11 This relates to democracy as a decision-making process, that is, the opportunity to influence politics through free and fair elections with universal and equal suffrage. The concept is based on Robert Dahl’s ideas about the principle of political equality as lying at the foundation of democracy.12 This principle means that all participants should be regarded as equally qualified to participate in the political decision-making process, and respecting it requires that several criteria are met. This includes not just the requirement for universal and equal suffrage but also what Dahl refers to as enlightened understanding – that everyone has equal opportunities to obtain knowledge about relevant alternative policies and their likely consequences.13 Political equality is an ideal, and few decision-making processes live up to the ideal in full. Still, for a process to be regarded as democratic, it must meet the criteria for political equality to a high enough degree. In the case of groups the size of a nation, Dahl argues that this requires many institutions to be in place, including free and regular elections, freedom of association, freedom of expression, and freedom of the press.14 V-Dem uses the role of these institutions in its measurements of electoral democracy conditions in any given country. For example, they track the degree of state censorship, harassment of journalists, reliability of the electoral system, and academic freedom. The second concept of democracy used by V-Dem is the term Farkas and Schou oppose and frequently appears in the contemporary debate – that of liberal democracy. Liberal democracy encompasses more freedoms and rights than electoral democracy, including civil rights such as freedom from torture, freedom of movement (a liberty many countries deprived their citizens of during the COVID-19 pandemic), and freedom from slavery.15 The concept of liberal democracy also incorporates a functioning rule of law, the absence of corruption in public administration, legal restrictions on the executive, and so on. According to V-Dem’s measurements, Sweden has been at the top of the table since the 1970s regarding both metrics (electoral and liberal democracy). In comparison, a country like Hungary recorded a steep rise during the early 1990s before then seeing an almost equally precipitous fall in the 2010s.16

94  Åsa Wikforss Taking these two basic concepts of democracy into account, we can note two things immediately. First, as emphasized by Dahl, there is a direct link between electoral democracy and knowledge. Dahl argues in favor of the importance of an enlightened understanding of democracy by referring to the idea that the point of this form of government is that it makes it likelier that people get what they want, and this, in turn, presupposes knowledge of various kinds.17 In this way, knowledge becomes part of the most fundamental dimension of democracy – the opportunity to influence politics via the ballot box. I shall return to this below when discussing the harms that mis- and disinformation can inflict on democracy. Second, this makes it clear that the suggestion made by Farkas and Schou that there is a fundamental contradiction between democracy, in the sense of being a popular government on the one hand, and liberal democracy on the other, is misleading in potentially harmful ways. It means they ignore the concept of electoral democracy and the idea that majoritarianism without fundamental political rights does not respect the idea of political equality and is, therefore, not adequately described as a democratic process. No doubt, a liberal democracy will involve further rights that limit the people’s will in protecting minorities. It is a topic for political debate when and to what extent such limitations are justifiable.18 However, a system that involves decision procedures that do not respect the criteria of political equality, as outlined by Dahl, is not even an electoral democracy. The potential harm of simplifying the concept of democracy the way Farkas and Schou do is illustrated by the neologism of illiberal democracy employed by autocrats such as Victor Orban. It’s clever rhetorically since the concept recalls a democracy (which we all like) minus a few things only liberals believe in. In practice, it is about something completely ­different – about wanting to get rid of the foundations of both electoral and liberal democracy. What remains is an illusion of democracy where people go to the polls to cast their votes without the presence of democratic institutions, as is the case in Hungary. V-Dem’s designation of this type of society is electoral autocracy.

5.3  Is Truth Undemocratic? However, Farkas and Schou’s critique of how Macron and others discuss the post-truth era also seems to have different grounds. They worry that what is driving these discussions is a problematic view of truth and knowledge – a view that is undemocratic by its very nature – and they object to the way that truth is described as being something “out there”; as something that “is seen as having a necessary, essential and universal content.”19 Farkas and Schou want to replace this view of truth as objective with a statement that rejects the idea that there is a truth that transcends geography, epochs, and subjectivities (what they describe as truth

The Dangers of Disinformation 95 with a capital T). Instead, they stress that historically there have been many different truths resulting from social and political contradictions.20 This concern is common and is repeated by many (non-philosophers) involved in the debate around democracy and truth. For example, the political scientist John Keane warns against unreflective references to the truth and argues that the real discovery is that there is no objective truth and that what counts as true varies from place to place.21 He argues that we must question both truth and post-truth, seeing them as companions rather than opposing each other. According to Keane, those who claim we now need more truth and objectivity, including academics and journalists, fail in this regard and predicate their thinking on naive understandings of the truth. Democracy is our best defense against the abuse of power camouflaged as claims to truth and against the illusion of certainty. Keane argues it should remind us that no truths are apparent and that what counts as true is a matter of interpretation. He concludes by saying that democracy assumes no man or woman is good enough to claim that they know the truth and to control others permanently. It is perhaps a bit surprising, from the point of view of philosophy, that controversial ideas about the nature of knowledge and truth are employed to argue against the connection between truth and democracy. What is being opposed is not simply the idea that truth is crucial to democracy but also a particular outlook on truth and knowledge, a perspective considered dangerous to a democratic society. The objection is that defenders of the truth (Macron, fact-checking journalists, and researchers such as myself) have as a point of departure a problematic view of truth and knowledge that naively assumes that truth is objective, universal, and timeless. Moreover, it is held this view of truth is not only philosophically problematic but also fundamentally undemocratic. Now, it is difficult to determine precisely what concept of truth is being offered instead of the idea that truth is objective. Indeed, it needs to be clarified what notion of objectivity one is objecting to. After all, the talk of truth with a big T could be more illuminating. Let me reflect on what seems to be driving this type of reasoning. First, there is a mix-up between truth and certainty. This is basic epistemology. The fact that knowledge requires truth does not mean that knowledge entails certainty. That is to say, it is possible to know that p is based on evidence that does not entail the truth of p but only makes p (sufficiently) probable.22 In the case of knowledge about our society and the surrounding world, we seldom have proof, that is, the evidence so compelling that we cannot be mistaken. This means that we need to be open to the possibility that we are mistaken (even if the fact that one may be wrong is not in and of itself a reason to doubt a belief). After all, as the history of science shows, there are plenty of examples of how we have been wrong in situations where we have had good but inconclusive evidence. Second, most theories of truth taken

96  Åsa Wikforss seriously by philosophers (whether correspondence theories, coherence theories, etc.) accept that truth is an objective property in the following sense. Whether or not a statement or thought possesses this property is not up to us.23 For example, the statement “Whales are mammals” was true even when everyone believed that whales are fish. It is important to note that this notion of objective truth does not encourage dogmatism. On the contrary, because the truth is objective and not up to us, we have every reason to be humble and open to objections – meaning we may be wrong. On the other hand, if the truth were a subjective ­characteristic – if the truth were up to me as an individual – I would never have any cause to be humble and listen to objections.24 Moreover, the feeling that an objective conception of truth would be undemocratic is based on confusion. Keane puts this feeling into words when he writes that no person is good enough to claim that she, in particular, knows the truth and thereby has a permanent right to rule over others. However, this is mixing up two different things. The fact that some people are better placed to find out certain truths – for instance, people who are trained experts in a field of research – means that they (when they have sufficient evidence) are entitled to claim that they have knowledge. Their claim would be justified in light of the expertise they possess and the evidence they present. However, it does not follow that they are excused from arguing for their position. And this does not mean they are entitled to rule over others. In a democracy, such a right is not given by virtue of expertise but by virtue of elections. In a sense, of course, it is somewhat trifling to say the truth is undemocratic. Precisely because the truth is not up to us, it cannot be determined by a majority decision. Even if all citizens believed that carbon dioxide emissions did not cause climate change, it would remain true that those emissions impact the climate. Keane asserts that the truth assumes a despotic nature in politics and concludes that democracy must stand for a world beyond truth and post-truth.25 The claim that the truth is despotic comes from Hannah Arendt’s discussion of the totalitarian state. But she makes a different point – she does not argue that the authoritative nature of truth makes it incompatible with democracy, but instead that the despotic nature of truth poses a threat to totalitarian leaders. In her famous essay Truth and Politics (1967), she writes: “Truth, seen from the viewpoint of politics, has a despotic character. It is therefore hated by tyrants, who rightly fear the competition of a coercive force they cannot monopolize.”26 The fact that truth is something objective, independent of us and our opinions, means that it represents a threat to all of the world’s autocrats. They know that if the truth comes out – for instance, concerning their political actions and consequences – their power will be under threat. This is why they use all means possible to try and control what can be controlled: people’s access to the truth. Democracy does not stand for a world beyond truth, as suggested by Keane, but for a world

The Dangers of Disinformation 97 where truth has a genuine opportunity to come out and be used as a basis for political decisions. Moreover, we should note that the reasoning itself has it backward. Let’s assume that democracy was incompatible with truth being an objective characteristic. Does it follow that we should give up on the claim that truth is objective? Obviously not. If it were the case that democracy was incompatible with the nature of truth, this would clearly not constitute an argument against a particular theory about the nature of truth. The philosophical question is not determined by our desires, just as our desires do not define scientific questions.27 Next, I shall develop the idea that knowledge is essential to electoral democracy by examining the threats posed by disinformation to democracy. I will start by clarifying the central concept of disinformation.

5.4  What Is Disinformation? Claire Wardle has compiled a lexicon for the post-truth era, Information Disorder: The Essential Glossary.28 “Disinformation” is defined as false information that is intentionally created and disseminated to cause harm. This definition provides a helpful starting point as it permits us to distinguish unintentional errors from disinformation. However, the description is overly narrow. First, disinformation is created and disseminated for various reasons, not just to cause harm. For instance, it is well known that a large part of the fake news spread in connection with the US presidential election in 2016 was produced by entrepreneurial teenagers in North Macedonia who found that it was a good way of making advertisement revenue. Second, disinformation does not always have explicit content that is false, which is a requirement of Wardle’s definition. Propagandists throughout the ages have always known that the best propaganda insidiously mixes truth and falsehoods. The important (and dangerous) thing about disinformation is that it makes the recipient believe something false, and it is perfectly possible to do so by telling the truth. This may sound strange, but it is a commonplace. An example discussed in the philosophy of language is how we use assumptions about communication, such as Grice’s relevance maxims, to implicitly communicate a false message. If I am asked whether I’m coming to the party tonight and I reply that I have to work, the implication is clear – I’m not coming – and if I go to the party, I would rightly be accused of having misled my interlocutor. Don Fallis (2015) defines disinformation as better suited to capturing this, employing the notion of misleading information. This information is likely to create false beliefs (but need not do so). Disinformation is then defined as misleading information that has the function of misleading someone. This, Fallis stresses, is a nonaccidental feature of disinformation – it is intended or designed to be tricky. It

98  Åsa Wikforss should be clear that this definition, as Fallis also notes, allows for true information to be used to disinform, that is, if the true information is designed to cause false beliefs. Misinformation, then, is simply misleading information where there is no intention to cause false belief. A complication in the current information landscape is that disinformation is typically spread on social media by people who take it to be true. Is it still disinformation, then? I believe it is because the content spread is designed to mislead, even if the user who unintentionally spreads it cannot be accused of disinforming anyone. To some extent, of course, this is a mere terminological question. However, I think there are reasons to stick to this classification. Thus, it has been shown that intentionally deceptive information has some distinctive content features, such as partisan bias, the use of negative emotions to provoke fear or anger, low levels of verifiability, long and sensational headlines, and the use of informal language.29 These are features that explain why disinformation tends to go viral and potentially harms democracy in ways that misinformation usually is not. In other words, the category of disinformation, as defined here, is useful for research. That true information can be used to disinform is of some importance when discussing the harms to democracy, since a central propaganda tool consists precisely of a certain way of disinforming by telling the truth – what I have previously referred to as the false narrative.30 This is about cherry picking, which involves making a skewed, misleading selection of facts, taking them out of context, and communicating an overall false message. For example, the false narrative is commonly used to undermine confidence in the democratic institutions of a country. By opting to consistently spread nothing but information about everything negative in the country, about what works poorly, and altogether avoid informing people about what is good, the message is spread that the country is on the verge of a breakdown. The individual claims are true, while the overall statement is false, which poses serious difficulties regarding standard fact-checking efforts.31 In Fallis’s sense, this is disinformation since the information is such that it is probable that it will cause false beliefs in the recipient and that it is intentionally designed to fulfill this purpose. And here, too, assumptions of relevance play a role in communicating the false message since the audience expects the selection of facts to be relevant to the topic of the communication – the state of society. I shall return to this type of disinformation below. We should perhaps stress that propaganda is not only about disinformation. At the bottom, propaganda is about manipulation, about intentional efforts to get us to react and act in ways that suit the sender’s agenda. This can be achieved by influencing our factual beliefs or manipulating our emotions and desires. And it is typically done without providing the subject with proper reasons, evidence, and arguments. Marlin (2013) defines propaganda along these lines: “the organized attempt to

The Dangers of Disinformation 99 affect belief or action or inculcate attitudes in a large audience in ways that circumvent an individual’s adequately informed, rational, reflective judgment.”32 Disinformation, thus, is a tool for propaganda but not the only one. For instance, in the current era, it has been noted that memes and jokes are frequently used to promote right-wing perspectives and radicalize people. In what follows, however, I shall focus on the dangers of disinformation, particularly regarding voter influence and factual disagreements.

5.5  Voter Influence In an electoral democracy, voters have two important tasks: to demand responsibility from their representatives for policies pursued (a retrospective task) and to issue a mandate by electing new representatives for the next period (a forward-looking task). Both tasks require knowledge and the ability to process political information. To hold politicians accountable, we must, for instance, know how they have done their jobs, what they have undertaken, and the consequences of which, but we also need knowledge about at which political level a decision was made. When it comes to issuing mandates, there is a body of knowledge that is crucial: this includes knowledge about which political alternatives are available to choose from, what the different parties stand for, their outlooks on society, and how they wish to change it. Of course, how we vote depends not only on what we believe about the world but also on what we value, our ideological beliefs, and our emotions. However, how we vote is not independent of what we believe about the world, so the ideal of the enlightened citizen plays a central role in the theory of democracy. And if we can influence society’s development in a way that is in line with our preferences, we must have knowledge about relevant facts. As Carpini and Keeter (1996), put it, political knowledge is the currency of citizenship – it allows us to exchange our preferences for political power.33 And just as with real money, they emphasize, this currency is unfairly distributed in society. Whether you obtain the required knowledge depends not only on individual ability but also on how society is organized, for example, whether everyone has equal opportunities to receive a good education and access reliable sources of information. This relates to how disinformation can be used to influence voting behavior. An example that has received widespread attention in recent years is the Brexit referendum held in the UK. The journalist Carole Cadwalladr visited her hometown in Wales to find out why there had been such a significant vote in favor of Brexit in that area.34 She was confused, given that in recent years the area had benefitted from a large influx of EU funding to regenerate a run-down mining community: a new sports hall, a new bridge, a research center, railway stations, and so on. When she interviewed local community members, they all referred to one thing

100  Åsa Wikforss as decisive in guiding their vote. They feared that Turkey was on the cusp of EU accession and that more than one million immigrants would overrun the UK. The village was home to almost zero immigrants, but they were nonetheless worried. The claim that Turkey was on the fast track to EU membership was false, and Cadwalladr wondered where it had come from. The answer was Facebook. Using political adverts on Facebook, the Leave campaign circumvented laws governing how much money can be spent in an electoral campaign in the UK and customized intentionally false messages designed to influence voters. Detailed focus group work and digital data collection were used to find the right messages to appeal to voters in low-income areas, such as Cadwalladr’s hometown. Two messages were selected (both false) that seemed to stick even with voters who were uninterested in politics: withdrawal from the EU would deliver an extra £350 million a week to the National Health Service, and Turkey was about to become an EU member. A total of more than one billion Facebook adverts were deployed at a meager unit cost.35 An opinion poll carried out shortly before the referendum showed that 47% of voters believed the claim about the NHS to be true (just 39% believed it to be false), and in studies carried out after the vote, 20% of voters gave this as their main reason for voting in favor of Brexit.36 A fateful choice for the UK with incalculable consequences for the union and Europe as a whole may have been decided based on disinformation. If it were true that leaving the EU would deliver an extra £350 million to the National Health Service, this would be a reason to vote against EU membership (one of many reasons, of course, and there may be more substantial reasons to remain). Still, if the statement is false, it does not constitute such a reason. The manipulation of voter preferences is not just an attempt to make it difficult for voters to carry out their civic duties – it also risks leading to poorer decisions. There is much to say about what “poorer” means in this context, but a decision made based on false assumptions will always be worse because it was not made based on good reasons. If such a decision brings about something good, it is nothing more than pure good fortune.37 Russia’s attempts to influence the 2016 American election have been well documented.38 They involved a plethora of initiatives. This ranged from hacking into Hillary Clinton’s emails to spreading disinformation via fake Facebook accounts and using Facebook groups to organize demonstrations on controversial issues (such as race, religion, and guns). The fundamental goal was to deepen divisions in American society and help secure the election of Trump. At the heart of it all was the Internet Research Agency based in St Petersburg, a troll farm funded by the Kremlin, where they worked systematically to produce disinformation. Personnel based there were set productivity goals. If the number of messages hostile toward Clinton was not high enough, the staff were subjected to harsh criticism. In the summer of 2018, Robert Mueller filed charges

The Dangers of Disinformation 101 against thirteen Russian citizens and three Russian firms, including the Internet Research Agency, for conspiring against the USA. Russia’s activities were described in detail in the twenty-nine-page indictment. Among other things, it was noted that on 27 July 2016, Russian hackers sought access to Clinton’s email server. Trump had publicly urged them to do just that on the same day.39 Even those who confine themselves to a very minimalist concept of democracy ought to be worried about these possibilities of manipulating the will of the people. One might object that as long as people are able to vote as they please without any external coercion, everything is fine. Indeed, it is up to everyone to elect whomever they like, regardless of what happens to be true. This may sound seductive, but things are more complicated than that. Suppose that you are trying to persuade me to come with you to a new bar in Stockholm’s Old Town. I’m not keen, but you tempt me with the fact it has the best locally brewed beer in the city, and since I like beer, I come along. Now I do want to visit that bar in the Old Town. However, it transpires that you lied to me – the venue is a wine bar without a pint glass in sight. In one sense, I did what I wanted (you didn’t force me to go there), but in another sense, I didn’t do what I wanted – you lured me there under false pretenses. My true will is arguably how I would have acted had I owned all the relevant facts. Admittedly, it is hard to say how I would have acted had you not deceived me into believing the bar served good beer. Perhaps I would have come along anyway simply because you were so persistent. In the same way, it is difficult to say how people would have voted had they been better informed.40 Political scientists usually emphasize how difficult it is to answer these hypothetical questions precisely because how we vote depends on values and emotions. For instance, it would be naive to believe that if everyone had been in possession of all the facts, then Trump would have lost massively. But no one denies that what we believe about society and politics is one of the factors that determine how we vote, so disinformation can determine elections.41 We should also note that how much disinformation can influence electoral outcomes depends on the electoral system. In this perspective, the US system is particularly vulnerable, given that the states hand out electors following the principle of “winner takes it all.” It may be enough to move a few thousand votes in a (populous) state where the election is close to determining the presidency.42 Similarly, the effects are more significant in a referendum where there are two simple options, in or out, yes or no, and where factual beliefs arguably play a more prominent role than traditional party loyalties (as seems to have been the case in the EU referendum). In multi-party, proportional systems of the type that Sweden exemplifies, where the number of representatives for a party in parliament reflects the number of votes in the general election, it is harder to move large parts of the population.

102  Åsa Wikforss In this context, one should view the dangers of the false narrative mentioned above. Unlike individual pieces of fake news, it has the potential to cause significant shifts in people’s worldviews and determine what goes on the agenda in public debates in a way that has the potential to change the voting landscape fundamentally. Indeed, the activities of the Swedish Populist Party, The Swedish Democrats, provide a good illustration of this. The party has its roots in the neo-Nazi movements from the 1990s, and for a long time, it was well below the threshold required to be part of the Swedish Parliament, 4%. This all changed with the growth of social media and alternative news channels. In 2010, the party made it across the threshold for the first time (5.7%), and in 2014 it had grown to a sizable party (12.9%). This growth continued in 2018 (17.53%), and in the last election, 2022, they became the biggest of the four parties on the conservative side of Swedish politics, with 20.5% of the votes. The Swedish Democrats have systematically used alternative media and social media to spread the narrative of Sweden as a country in decline. Recently, it was also exposed that they use tax money to run troll factories, employing some of Sweden’s most significant political Facebook pages. The method is simple: by making a highly skewed selection of facts, mainly focused on crimes committed by immigrants, they have managed to communicate the overall message that Sweden is a country on the verge of a system collapse.43 This, it should perhaps be noted, is disinformation, plain and simple. While Sweden has severe gang violence problems, it is by no means a country on the verge of collapse. Indeed, a recent study found that Sweden was one of the most well-functioning countries in the world.44 The fact-checking site EU vs. Disinfo, which monitors the impact of Russian disinformation on Europe, has identified many messages of this kind – what they refer to as meta-narratives with the purpose of disinforming. These relate to an overarching message communicated through text and images that can be adapted to suit different target audiences. They observe that pro-Russian sources promote five meta-narratives to attack democratic institutions in the west, spreading mistrust and getting people to lose faith in democratic processes. Three of these narratives are familiar features of right-wing populist movements. The first relates to an evil elite who do not care about the (true) people and who have taken over power and are now concealing the truth about critical issues such as migration. The second is about a threat posed to values – about a decadent, morally rotten western world governed by political correctness and feminism. The third is that the nation faces systemic collapse and civil war as a result of overly excessive immigration. The Swedish Democrats had skillfully combined these three narratives to paint the picture of a fallen society, an imagined golden past before the elite ruined everything by allowing too many immigrants, and a current state of crisis with rampant crime and multicultural chaos. The story is

The Dangers of Disinformation 103 very similar to that communicated by Sputnik, the Russian news agency. Charlotte Wagnsson, at the Swedish Defense University, describes how Russia has systematically used “antagonistic narrative strategies” to weaken Swedish democracy. A central one is the tale of Swedish decline, which she summarizes as follows: On the whole, Sputnik continually places Sweden in a declining spiral with few possibilities for betterment. The narrative is temporally and factually selective in terms of appropriated events with an apparent inclination towards the negative to construct the required presentation of a “Sweden in decline”. The main plot pictures Sweden, once a prosperous, thriving country with admirable ethical and moral values, now as a politically shattered weak state, experiencing political disorder and social chaos, in what can best be described as a state of failure.45 This false narrative set the agenda for the political debates leading up to the Swedish Election in September 2022 – the year that the Swedish Democrats became one of the biggest right-wing populist parties in Europe.

5.6  Dangerous Disagreements Jennifer Kavanagh and Michael D. Rich have characterized our current era as one of truth decay.46 They define periods of truth decay in terms of four trends: a blurring of opinion and fact in media, an increasing volume of personal experience over fact, declining trust in formerly respected sources of factual information, and increasing disagreement about facts and interpretation of data. Therefore, what they provide is more valuable than the vague talk of a post-truth society and the sweeping statement that public opinion is driven more by emotions and feelings than facts. The focus is not on the truth as such but on societal trends that impact the dissemination of knowledge and the ability to unite around a common view of basic facts. The fourth trend, increasing disagreements, is plausibly seen as a result of the other three and is closely connected with the new high-choice information environment. The blurring of opinion and fact in the new media landscape, the increasing volume of personal experience and anecdotal evidence on social media, and declining trust in reliable sources of factual information (combined with unwavering trust in unreliable sources) will lead to increasing disagreements. Of course, that there are growing disagreements need not be a sign of trouble. A democratic society is naturally built on the idea that there are disagreements of various kinds, and it is well equipped to handle these through public debates and political compromises. It is only if differences of opinion become so extreme and

104  Åsa Wikforss profound that genuine discussion is no longer possible and compromises become unthinkable that democracy is in trouble. When discussing disagreements in democracy, naturally, we need to distinguish between two types of differences in opinion: dispute relating to values and disagreement relating to factual beliefs. Disagreement in a society is not just about facts and interpretation of data, as mentioned by Kavanagh and Rich, but also about what is worth striving for. In general, all political disagreements stem either from a disagreement on facts or a disagreement on values (or both). Value disagreements are usually harder to bridge via discussion and argumentation. There is no generally accepted method for determining the truth of a value statement (if indeed these can be true at all), and there are no value experts in the sense that there are scientific experts. This is one of the reasons why the freedoms and rights in any democratic society are such central pillars. People have different ideas about what constitutes the good life, and society permits them to essentially live their lives in the way they choose. This pluralism represents the core of modern democracy.47 While value disagreements are unavoidable, they can be accommodated by democracy (that is, as long as the disagreement is not one about the system as such). Indeed, this is arguably one of the greatest strengths of democracy: it allows us to disagree about values in a civilized way and jointly negotiate our future, find compromises, and live together despite our disagreements. However, this presupposes that we do not also end up with intractable factual disagreements. As Hélène Landemore (2013) emphasizes, political disagreements are often based on disagreements about the facts. If the latter type of disagreement becomes too great within a society, then the political disagreement risks becoming unmanageable and poor decisions are made (in the sense that they are based on false factual beliefs).48 Concerning factual issues, disagreement can stem from an uncertain knowledge base where the evidence is incomplete or contradictory, as in the early stages of the COVID-19 pandemic. In such cases, disagreement is not a sign of a dysfunctional climate of discourse or irrationality. On the contrary, in such cases, disagreement is a sign of good health, and debate is absolutely necessary. This is how knowledge emerges, and we can gradually transition from the uncertain to the well-founded and reliable. We must be willing to listen to each other’s arguments and accept that there is uncertainty and adjust our degree of certainty according to how strong the evidence is.49 However, when the available evidence strongly supports a given proposition, when there is a significant expert agreement, continued factual disagreement will have other sources and be detrimental to democracy. Dan Kahan has argued that people today are more polarised on issues related to socially relevant facts (e.g., gun control or the climate) than on classic political value issues.50 He speaks of a form of fact polarisation, characterized as disagreement on factual issues (admitting empirical

The Dangers of Disinformation 105 evidence) that cannot be bridged with further information, even though there is a sizeable expert consensus. The relevant facts tend to be policy-­ relevant, for instance, facts having to do with climate change, gun control, or vaccines. What explains fact polarization? One hypothesis is that people are knowledge resistant, and they refuse to accept evidence that contradicts the factual beliefs that have come to be vital to them.51 This is also Kahan’s hypothesis, referring to what he labels politically motivated reasoning. Motivated reasoning is a form of reasoning driven by values – desires – in place of evidence. In somewhat simplified terms, you believe what you want to believe instead of having reasonable grounds. It is thus a type of irrationality, and it can affect us all. It is not about deliberately setting out not to accept facts but the intervention of subconscious psychological mechanisms. When this happens, we use skewed reasoning to protect the threatened belief at any cost rather than seeking the truth. For instance, we weigh the evidence for our position as greater than the evidence against it.52 In terms of politically motivated reasoning, Kahan asserts it is driven by the desire to protect opinions that have become identity markers for our ideological groups. In a famous experiment, he demonstrates that people on both sides of the political spectrum – conservative or liberal – tend to read tables of data according to their political identities when the tables concern relevant policy facts. Conservatives tend to get the numbers wrong when the table shows that gun control reduces lethal violence, while liberals similarly muddle their numbers when it shows that gun control increases deadly violence. Yet both groups perform following their mathematical skills when engaging with factual issues lacking political charge (e.g., whether a particular skin lotion counteracts eczema). These results, Kahan argues, cannot be explained as a result of science comprehension deficits since the data show that numeracy skills strengthened motivated reasoning rather than protected against it. Consequently, the underlying cause is motivated reasoning. Politically motivated reasoning quite clearly represents a challenge to a democratic society. When people accept factual beliefs based on political affiliation instead of evidence, it is no longer possible to bridge disagreements using reason and argument. The more factual beliefs become markers of political identity, the harder this becomes. The upshot is that factual contentions increasingly behave like value disagreements – they become intractable and cannot be overcome by appealing to evidence. In a society where affective polarization is intense, where people regard the outgroup with animosity and have low feelings of trust toward it, the danger of politically motivated reasoning increases.53 However, another source of intractable factual disagreements derives from the new media landscape and the flow of disinformation. How we assess the evidence we receive always depends on our background knowledge. Moreover, as stressed by Tappin, Pennycook, and Rand (2020),

106  Åsa Wikforss political identities co-vary with prior beliefs in a politically polarized media landscape.54 If a person has been fed a large amount of unreliable information, the background beliefs formed may make it (subjectively) rational for this person to dismiss available knowledge. The result will be a form of intractable factual disagreement along the political fault lines. Attempts to overcome a dispute the usual way, by appealing to evidence and arguments, will then easily fail, even without either side engaging in irrational reasoning. After all, given what this person already believes, the evidence presented can be dismissed as of low evidential value and can be done so perfectly rationally.55 The polarized and unreliable media landscape means we increasingly risk ending up with intractable factual disagreements that Landemore warns against. Democracy needs to be rooted in a shared reality – otherwise, the upshot is a form of intractable political disagreement that stymies public debate, strengthens affective polarization, and counteracts political compromises. And when democracy is radically uprooted from a shared reality, the field of play is left open to democracy’s enemies. In August 2022, president Joseph Biden delivered a warning about the fate of democracy. As The New York Times reports, Biden underscored deep rifts in American society that make it an almost ungovernable moment in the nation’s history: “Not only do Americans diverge sharply over important issues like abortion, immigration, and the economy, they see the world in fundamentally different and incompatible ways.”56 There is, however, one thing that Americans do agree on, that American democracy is in danger. In a Quinnipiac University poll from August 2022, 69% of Democrats and 69% of Republicans say that democracy is in danger of collapse.

5.7  Concluding Remarks I have argued that democracy depends on truth and knowledge and that there are strong reasons to suspect that the current democratic backsliding results from the radical transformation in the information landscape. My argument has proceeded in three steps. First, I have argued that there is an essential link between electoral democracy and knowledge. The claim that we can disconnect truth and democracy put forth by Farkas and Schou, among others, is based on a simplistic understanding of the nature of democracy and ignores the essential role enlightened understanding plays in electoral democracy. Second, the suggestion that there is something undemocratic about the very idea of objective truth is based on flawed reasoning. Third, to further elucidate the importance of knowledge to electoral democracy, I have examined some of the harms caused by disinformation to the democratic society. Disinformation is employed to manipulate voter preferences, tamper with their political power, and disconnect the collective decision-making procedure of democracy from

The Dangers of Disinformation 107 the factual knowledge needed to solve societal problems. Moreover, disinformation causes increasing factual disagreements, resulting in unmanageable political disputes and increasing affective polarization. Currently, initiatives are taken in Europe and elsewhere to strengthen our defenses against disinformation. The challenges are obvious – how can this be done while preserving the very freedoms that are an essential part of democracy? Experts disagree on the answer, but one thing can be said with some certainty: unless something is done about our current, poisoned information landscape, it will be challenging to stop the democratic decline.

Notes









1. V-Dem stands for Varieties of Democracy. They have a unique approach to measuring democracy, employing “a multidimensional and disaggregated dataset that reflects the complexity of the concept of democracy as a system of rule that goes beyond the simple presence of elections.” https://­ v-dem.net 2. See, for instance, Eatwell and Goodwin (2018). 3. Parts of Sections 5.2–5.5 are based on my Swedish book, Därför Demokrati. Om kunskapen och folkstyret (“The Case for Democracy. On Knowledge and the Rule of the Many”), 2021. 4. Rubin (2018). 5. Steve Tesich first used it in an article in 1992. According to Tesich, the US media failed to scrutinize the Iran–Contra scandal, which taught politicians that they did not need to care about the truth. A dictator would rub their hands together in glee, writes Tesich, given that it appears we have voluntarily decided to live in some sort of post-truth world. 6. Macron (2018). 7. See Kessler et al. (2021). 8. Farkas and Schou (2020, p. 5). 9. Ibid., p. 7. 10. It is the irony of fate that Macron delivered his address at the Capitol, which barely two years later was stormed by furious crowds fuelled by wild conspiracy theories and a president who had spread the lie that Democrats had stolen the 2020 election. 11. The other three are egalitarian democracy, participatory democracy, and deliberative democracy. These are not independent of the first two concepts of democracy, instead describing dimensions of democracy that allow it to work better. 12. Dahl (1989, Chapter 8) and Dahl (1998, Chapter 4). 13. For a discussion of this issue, see Dahl (1989, pp. 163–175). 14. As Dahl stresses, this is not a matter of arbitrarily stating that these elements feature in a particular definition of democracy that he likes. Instead, his point is that all these things constitute necessary prerequisites in purely empirical terms for democracy to exist. 15. Whether it is right to restrict these freedoms under certain conditions, such as a pandemic, is a complex matter that I will not discuss here. From the point of view of democracy, the important thing is that such restrictions have a time limit. V-Dem reports that autocratization was sped up during the pandemic since authoritarian leaders across the globe used it as an excuse for permanent restrictions on fundamental freedoms and rights.

108  Åsa Wikforss 16. See V-Dem Country Graph, https://www.v-dem.net/data_analysis/CountryGraph/ 17. Dahl (1989, pp. 111–112). 18. Rosenfeld (2019) describes how there has historically been an opposition between, on the one hand, populist conceptions of democracy, downplaying the role of truth as well as of rights, and more liberal conceptions of democracy on the other. 19. Farkas and Schou (2020, p. 52). 20. Ibid., p. 9. 21. Keane (2018). 22. This is called a fallibilist conception of knowledge. Although fallibilism is not undisputed, it has come to be very widely endorsed within epistemology. 23. According to correspondence theories, truth consists of correspondence or agreement between a statement (a proposition p) and the world. In contrast, according to the coherence theory, a statement is true if and only if it coheres (in a sense to be specified) with the totality of accepted statements. Although coherence theories are standardly taken to be problematic since they lead to some form of relativism, making the truth of a statement relative to different belief systems, it still allows for a distinction between a person believing that p and p being true – after all, that the belief has the property of cohering with the more extensive system is not up to the individual. Someone like Keane could try to employ a version of the coherence theory to argue for the claim that for specific belief systems, at a point in time, the proposition Whales are fish was true. But they would then need to address well-known philosophical challenges to this type of relativism. See, for instance, Boghossian (2007). 24. I discuss this further in my book Alternativa Fakta (“Alternative Facts”), 2017, Chapter 2, Fri Tanke förlag. 25. Keane (2018, p. 13). 26. Admittedly, Arendt is also concerned that experts’ factual knowledge may have an oppressive impact and prevent open debate. There are concerns about democracies becoming increasingly epistocratic, but it is a bit difficult to understand that factual knowledge is oppressive in any problematic way – knowledge should affect the debate. Otherwise, the discussion is not truth-seeking. 27. The same applies to the question of the objectivity of values. Some political scientists have argued that democracy is incompatible with value objectivism (see, for example, Lewin 1990). Whether value objectivism is true does not depend on what we happen to think about its relationship with democracy. 28. Claire Wardle is a social media expert and one of the founders of First Draft, a leading non-profit organization researching disinformation and how to counteract it. See https://firstdraftnews.org 29. See Damstra et al. (2021). 30. See Wikforss (2018). 31. Of course, its proponents are usually more than happy to mix in plenty of explicitly false statements when it comes to propaganda. The false narrative, therefore, interacts with fake news in a dangerous manner: once you have swallowed the story about the country’s imminent demise, you are more receptive to fake news. 32. Malin (2013, p. 12). 33. Delli Carpini and Keeter (1996, pp. 8, 11). 34. Cadwalladr (2020). She also describes her findings in a TED talk that is well worth watching: Facebook’s Role in Brexit – and the Threat to Democracy, TED 2019.

The Dangers of Disinformation 109 35. Further details about this are provided in a piece by Dominic Cummings, one of the most influential strategists in the Leave campaign, where he describes how they worked to reverse opinions away from Remain, which had been prevalent ahead of the referendum. For a link to Cummings’ article, see Cadwalladr (2020). 36. This example is discussed in Goodin and Spiekermann (2018, p. 339). 37. In 2018, we learned more about the ability to use targeted political messaging and how this had been exploited during the American election in 2016. It transpired that a data analysis company, Cambridge Analytica, had harvested millions of Facebook profiles before the election and built models of individual American voters that could then be used to design targeted advertising campaigns. The company was owned by billionaire Robert Mercer and its Vice President at the time was Donald Trump’s adviser Steve Bannon. Among other things, they used different types of personality tests where people consented to collect their data. This made it possible to tailor political messages to exploit individuals’ disparate vulnerabilities, fears, and anxieties. See New York Times summary, Cambridge Analytica and Facebook: The Scandal and the Fallout So Far, Nicholas Confessore, 4 April 2018. 38. See, for example, US Senate (2019). 39. It is not entirely straightforward to determine in quantitative terms what the consequences of the Russian campaign were on the electoral outcome. This issue is explored in-depth in the book Cyberwar: How Russian Hackers and Trolls Helped Elect a President – What We Don’t, Can’t, and Do Know, by Kathleen Hall Jamieson (2018). According to her, it is very likely the Russian attacks decided the election. 40. This issue is discussed in K. Ahlström-Vij (2021). 41. See Arnold (2012) for evidence that voters’ level of knowledge can play a decisive role in the outcome of elections. Arnold’s study of elections in 27 democracies demonstrates that electoral results would probably have been different had voters had more politically relevant information. 42. According to Gunther et al. (2018), there is evidence that this is precisely what happened in the 2016 election. 43. The efforts have been strengthened by international actors who take an interest in seeing a liberal democracy like Sweden fail. See Rapacioli (2018) and Becker (2019). Rapacioli, the editor of the English-language news site The Local in Stockholm, wrote his book after having noticed that whenever they reported on adverse events in Sweden, it was suddenly the subject of global spread, but that the same thing did not happen when reporting on adverse events that occurred in other countries such as Denmark or the Netherlands. Rapacioli concluded that this related to a deliberate and politically motivated campaign of disinformation. 44. See Strömbäck (2022). 45. Wagnsson (2021, p. 244). 46. Kavanagh and Rich (2018). According to Kavanagh and Rich, other periods of truth decay were during the 1920s and 1930s with the rise of the new evening press and radio, as well as the 1960s and 1970s, which were characterized by extensive political propaganda, including in relation to the Vietnam war. The most recent period is our present-day – the 2010s and 2020s. 47. See Weale (2018). 48. Landemore (2013). 49. See Angner (2020).

110  Åsa Wikforss 50. Kahan et al. (2017, pp. 54–86). 51. For recent research on the topic, see our research program Knowledge Resistance: Causes, Consequences, and Cures (funded by Riksbankens Jubileumsfond). See also our joint volume on the subject, Strömbäck et al. (2022). 52. A few different things can trigger motivated reasoning. It can relate to conflicts of interest (you would very much like to believe that wine is good for you), fears (you would prefer not to stick a needle in your child), or to worldview (climate deniers often have a more hierarchical worldview as opposed to egalitarian). For a good overview, see Hornsey and Fielding (2017). 53. According to recent assessments, affective polarization in the USA has reached maximum levels, as measured by standard methods (see V-Dem Democracy Report 2022). Note that here, too, disinformation may play a role, as when disinformation about the other side, the outgroup, is prevalent. 54. As Tappin et al. note, there is an essential confound in the paradigmatic experiments testing for knowledge resistance, as designed by Kahan’s team. See also Glüer and Wikforss (2022). 55. It may be that, ultimately, there is irrationality in the subject’s information history, as when she trusts a source, she knows to be unreliable. For a discussion, see Glüer and Wikforss (2022). 56. Baker and Hounshell (2022).

References Ahlström-Vij, K. 2021. Do We Live in a Post-Truth Era? Political Studies, June 2021. doi: 10.1177/00323217211026427 Angner, E. 2020. Epistemic Humility – Knowing Your Limits in a Pandemic, Behavioral Scientist, 13 April 2020. Arnold, J.R. 2012. The Electoral Consequences of Voter Ignorance, Electoral Studies, 31(4). Baker, P. & Hounshell, B. 2022. ‘Parties’ Divergent Realities Challenge Biden’s Defense of Democracy, The New York Times, 2 September 2022. Becker, J. 2019. The Global Machine Behind the Rise of Right-Wing Extremism, New York Times, 10 August 2019. Boghossian, P. 2007. Fear of Knowledge. Oxford University Press. Cadwalladr, C. 2020. If You’re Not Terrified about Facebook, You Haven’t Been Paying Attention, The Guardian, 26 July 2020. Confessore, N. 2018. Cambridge Analytica and Facebook: The Scandal and the Fallout So Far, The New York Times, 4 April 2018. Dahl, R. 1989. Democracy and Its Critics. New Haven: Yale University Press. Dahl, R. 1998. On Democracy. New Haven: Yale University Press. Damstra, A., Boomgarden, H.G., Broda, E., Lindgren, E., Strömbäck, J., Tsfati, Y. & Vliegenthart R. 2021. What Does Fake Look Like? A Review of the Literature on Intentional Deception in the News and on Social Media, Journalism Studies, 22: 1947–1963. Delli Carpini, M.X. & Keeter, S. 1996. What Americans Know About Politics and Why It Matters. New Haven: Yale University Press.

The Dangers of Disinformation 111 Eatwell R. & Goodwin, M. 2018. National Populism: The Revolt Against Liberal Democracy. New York: Penguin Random House. Fallis, D. 2015. What Is Disinformation?, Library Trends 63. Farkas, J. & Schou, J. 2020. Post-Truth, Fake News and Democracy. London: Routledge. Glüer, K. & Wikforss, Å. 2022. What is Knowledge Resistance? In Knowledge Resistance in High-Choice Information Environments (eds. J. Strömbäck, Å. Wikforss, H. Oscarsson, T. Lindholm & K. Glüer). Routledge: London, United Kingdom. Goodin, R.E. & Spiekermann, K. 2018. An Epistemic Theory of Democracy. Oxford: Oxford University Press. Gunther, R., Beck, P. & Nisbet, E. 2018. Fake News May Have Contributed to Trump’s Victory, The Conversation, 15 February 2018. Hall Jamieson, K. 2018. How Russian Hackers and Trolls Helped Elect a President – What We Don’t, Can’t, and Do Know. Oxford: Oxford University Press. Hornsey, M.J. & Fielding, K.S. 2017. Attitude Roots and Jiu Jitsu Persuasion: Understanding and Overcoming the Motivated Rejection of Science, American Psychologist, 72(5). http://dx.doi.org/10.1037/a0040437 Kahan, D., Peters, E., Cantrell Dawson, C. & Slovic, P. 2017. Motivated Numeracy and Enlightened Self-Government, Behavioral Public Policy, 1 May 2017. Kavanagh, J. & Rich, M. 2018. Truth Decay: An Initial Exploration of the Diminishing Role of Facts and Analysis in American Public Life. RAND Corporation: Santa Monica, United States. Keane, J. 2018. Post-Truth Politics and Why the Antidote Isn’t Simply “FactChecking” and truth, The Conversation, 23 March 2018. Kessler, G., Rizzo, S. & Kelly, M. 2021. Trump’s False and Misleading Claims Total 30,573 over Four Years, The Washington Post, 24 January 2021. Landemore, H. 2013. Democratic Reason: Politics, Collective Intelligence, and the Rule of the Many. Princeton: Princeton University Press. Lewin, L. 1990. Upptäckten av framtiden: en lärobok i politisk idéhistoria. Stockholm: Norstedts Juridik. Macron, E. 2018. Speech of the President of the Republic, Emmanuel Macron, Before the Congress of the United States of America, 25 April 2018. Marlin, R. 2013. Propaganda and the Ethics of Persuasion. Peterborough: Broadview Press. Rapacioli, P. 2018. Good Sweden, Bad Sweden. The Use and Abuse of Swedish Values in a Post-Truth World. Stockholm: Volante. Rosenfeld, S. 2019. Democracy and Truth: A Short History. Philadelphia: University of Pennsylvania Press. Rubin, J. 2018. In a Post-Truth World, There Is No Democracy, The Washington Post, 1 May 2018. Strömbäck, J., Wikforss, Å., Oscarsson, H., Lindholm, T. & Glüer, K. (eds.). 2022. What Is Knowledge Resistance? In Knowledge Resistance in High-Choice Information Environments. Routledge. Strömbäck, S. 2022. Beyond the Political Framing Battles – Sweden in an International Comparison, Futurion, 7 September 2022.

112  Åsa Wikforss Tappin, B.M., Pennycook, G. & Rand, D.G. (2020). Thinking Clearly about Causal Inferences of Politically Motivated Reasoning: Why Paradigmatic Study Designs Often Undermine Causal Inference, Current Opinion in Behavioral Sciences, 34, 81–87. Tesich, S. 1992. A Government of Lies, The Nation, 13 June 1992. US Senate. 2019. Report on Russian Active Measures Campaigns and Interference in the 2016 US Election. Volume 2: Russia’s Use of Social Media with Additional Views. Washington DC: United States Senate Select Committee on Intelligence. Wagnsson, C. 2021. A Framework for Analyzing Antagonistic Narrative Strategies: A Russian Tale of Swedish Decline, Media, War & Conflict, 14(2). Weale, A. 2018. The Will of the People: A Modern Myth. Cambridge: Polity Press. Wikforss, Å. 2017. Alternativa fakta. Stockholm: Fri Tanke. Wikforss, Å. 2018. Politikens falska berättelser låter sig inte faktakollas, Dagens Nyheter, 6 July 2018. Wikforss, Å. 2022. Därför demokrati (with Mårten Wikforss). Stockholm: Fri Tanke.

Part II

Democratic Optimism

6

The Politics of Resentment Hope, Mistrust, and Polarization Alessandra Tanesini

6.1 Introduction Citizens in several Western democracies have become increasingly polarized (Sunstein, 2009). This phenomenon is exemplified in Britain by the rancor associated with the Brexit vote and in the USA by angry and resentful diatribes connected to Trump-style politics. Elsewhere populism is also on the rise bringing in its trail acrimonious divisions (Betz, 1993; Eatwell & Goodwin, 2018). Arguably, a distinctive feature of current polarized attitudes is their affective character. For instance, Liliana Mason (2018) has argued that genuine disagreements of opinion are not now necessarily more pronounced than in the past. In the USA, at least the average voter of either Democrat or Republican leanings is likely to hold reasonably centrist views. However, at present, people in opposite camps dislike and despise each other vehemently. They are, for example, unlikely to form friendships across these divides (Iyengar et al., 2019). In this chapter, I explore one aspect of this affectively charged political climate: “the politics of resentment” (Cramer, 2016; Hochschild, 2016; Reicher & Ulusahin, 2020). This expression is meant to capture a political outlook or perspective shaped by the reactive attitude, or moral emotion, of resentment. Such an outlook is characteristic of those who feel “left behind” or experience themselves as having become “strangers in their own land” (Hochschild, 2016). Their resentment is a bitter response to a perceived loss of status and to apparent (or real) threats to the social world within which they occupy a position that secures their self-respect. Those who engage in the politics of resentment fear slipping down the social ladder, often see their economic conditions deteriorate, and experience a loss of some entitlements previously conferred to them in virtue of their dominant ethnic or gender identity. The chapter has two main aims. The first is to offer an account of the emotional outlook characteristic of the politics of resentment and to explain its connection to cruel and wilful hopes. The second is to encourage scholars who wish to understand and address some of the main dangers facing Western democratic institutions to focus also on the emotional DOI: 10.4324/9781003311003-9

116  Alessandra Tanesini dimension of political life rather than exclusively on its more cognitive components. Our current problems are not solely or primarily caused by ignorance, disregard for the truth, and disagreements in belief. Instead, I argue that the promotion of cruel and wilful hopes is one source of the corrosion of liberal democratic institutions.1 The chapter consists of five sections. In Section 6.2, I describe the dynamics of the politics of resentment and present some evidence of its pervasiveness. In Section 6.3, I briefly explain how the politics of resentment is connected to seemingly paradoxical political behaviors that run contrary to the economic interests of those who engage in them. These behaviors are not fully explained by invoking voters’ ignorance caused by the consumption of partisan and ideological news sources or by voters’ motivated ignorance that would be a product of cultural identity-­ protective cognition (Kahan et al., 2007). Instead, I show that a complete analysis rationalizes these seemingly paradoxical behaviors in terms of political outlooks informed by resentment. The political behaviors that might seem paradoxical make sense if they are seen as measures designed to restore one’s social status. In Section 6.4, I explore the metaphor of standing in a queue while others cut to the front of the line, which captures the narrative animating the politics of resentment (Hochschild, 2016). I deploy Tajfel and Turner’s (1979) social identity theory to explain the meaning of this metaphor. In Section 6.5, I show that cruel and wilful hopes are an essential component of the politics of resentment. In Section 6.6, I conclude by advancing an ameliorative proposal based on the power of university education to instill hopes and values apt to reduce defensiveness. I also briefly answer some of the concerns raised by critics of the politics of hope (Lindroth & Sinevaara-Niskanen, 2019; Warren, 2015).

6.2  The Politics of Resentment: Polarization and Mistrust The expression “the politics of resentment” is widely used to refer to views and sentiments that have been fomented and exploited by nationalist and populist politicians in Europe, Asia, and the Americas. This is a global phenomenon exemplified in the 1990s by the rise of the National Front in France, the Northern League in Italy, and the Freedom Party in Austria (Betz, 1993). The same expression has been adopted to describe the attitudes that have led to Brexit (Eatwell & Goodwin, 2018; Reicher & Ulusahin, 2020), the rise of the Tea Party, and the subsequent election of Trump in the USA (Cramer, 2016; Hochschild, 2016), but also the success of Modi in India and Bolsonaro in Brazil (Eatwell & Goodwin, 2018).2 Whilst there undoubtedly are significant differences between these cases due to varied economic circumstances, democratic traditions, and institutions, there are also commonalities. One prominent similarity is the ability of these parties and movements to capitalize on the resentment

The Politics of Resentment 117 felt by some sectors of the population toward the government and other elements of the so-called establishment, but also toward members of different social groups that are perceived as the undeserving beneficiaries of special treatment. In Italy, for example, the Northern League relies on the belief widespread among inhabitants of the affluent and industrialized North that their hard-earned cash is stolen when taxes exacted by a parasite-like state are used to benefit politicians in Rome and citizens living in the South (Betz, 1993, p. 418; Ivaldi et al., 2017, p. 365). This belief is also associated with the conviction that politicians and Southerners are greedy, or lazy and incompetent, and, therefore, undeserving of these benefits. Despite some differences, these attitudes are remarkably similar to those found by Katherine Cramer (2016) in rural Wisconsin. Cramer’s interviewees do not live in affluent regions, but like Northern Italians, they self-identify as hard-working and self-reliant people. They resent their taxes being spent on politicians and workers in the public sector who, in their view, do not work hard enough to merit their income and benefits. Their resentment also targets welfare recipients as undeserving and lacking grit and determination. This is the same rhetoric, distinguishing the deserving from the undeserving poor, that features prominently in the welfare reforms proposed in the UK by the Cameron government before the Brexit referendum (Hoggett et al., 2013).3 This brief discussion of some expressions of the politics of resentment highlights three main features. First, those whose politics is infused with resentment share some social identity that is often geographically and ethnically circumscribed. This is an identity that, in their view, entitles them to privileges for which others do not qualify. Hence, for instance, the moniker of British jobs for British workers deployed to appeal to Leave supporters before Brexit (cf. Shabi, 2017).4 In some cases, resentful citizens also think of themselves as hard-working, freedom-loving, and self-reliant (Betz, 1993; Cramer, 2016; Eatwell & Goodwin, 2018; Hochschild, 2016; Ivaldi et al., 2017). These moral traits would ground their special entitlements and distinguish them from the undeserving. Second, these individuals’ political outlooks and understanding pivot on grievances. They conceive of themselves as the victims of some injustice. This conviction that some wrong has been done to them dominates their understanding of the current political situation (Cramer, 2016; Reicher & Ulusahin, 2020). Equally dominant, though, is the affective tone of their political outlook. This tone is that of the negative moral emotions of anger and resentment. Both anger and resentment are moral emotions involving appraisals of situations as circumstances in which some moral expectations have been violated. The angry (and resentful) person evaluates her condition as one in which some specific individuals have wronged or slighted her by depriving her of something she could legitimately have expected (Martin, 2014, pp. 122–123).5 Anger and resentment are manifested in confrontational behavior directed at the targets of

118  Alessandra Tanesini resentment singled out as individuals one dislikes, despises, or even hates because one holds them responsible for the injustice to which one has allegedly been subjected.6 Third, resentment has two targets. The first are members of the establishment, including elected representatives and public employees. The second target comprises members of social groups perceived to be the underserving beneficiaries of special treatment in welfare, jobs, and access to education. Often the negative evaluation of these individuals is based on moral assessments of their characters and belief systems. Hence, they are perceived as unwilling to work hard enough or as having views incompatible with Western ideals of freedom and equality.7 To fully appreciate the connections of the politics of resentment to affective polarization and distrust of political institutions, however, we must examine the relationship of anger to resentment. Whilst resentment is closely related to anger, it can also possess undertones of bitterness that anger often lacks. Those whose political outlook is dominated by resentment are not just angry. They are also embittered because they think they are being wrongly judged for being angry. They are frustrated and feel powerless because they experience the moral disapproval of the so-called liberal elites as unwarranted and an insult additional to the initial anger-provoking injury (Walker, 2006, p. 108).8 This experience also gives rise to feelings of being misunderstood, of having become strangers in one’s own land (Hochschild, 2016). Hence, the resentment characteristic of a politics based on grievances is associated with feelings of powerlessness, alienation, and separation. In this regard, political anger as a motivation for action is fundamentally different from political resentment (Cherry, 2021). Anger is compatible with dialogue, communicative engagement, requests for apologies, and demands that the wrong or slight be addressed (Tanesini, 2021). But bitter resentment makes engagement feel pointless and foments hostility toward those who, in one’s view, have wrongly judged one’s remonstrances to be unwarranted and against whose judgment one seemingly has no recourse. If this is right, the politics of resentment are a component of an unfolding dynamics that lead to affective polarization. The dynamic starts with a loss of status, entitlements, or benefits (or with fear of such a loss) on the part of sections of the population that share some social identity such as ethnicity, religion, gender, or geographical location. This actual or feared loss is perceived as unfair and thus engenders anger directed toward those that one holds responsible for the injustice. But socially powerful constituencies, such as the media, public officials, or other elites, make it plain that they view one’s complaints as illegitimate and as reflecting poorly on one’s moral character because they are evidence of bigotry, sexism, racism, or xenophobia. Further, one sees oneself as powerless in

The Politics of Resentment 119 the face of these judgments and responds to them with resentment. The other side perceives the resentment as evidence of the correctness of the initial assessment and thus responds to the resentment with contempt. Hilary Clinton’s unfortunate remark describing Trump’s supporters as a “basket of deplorables” exemplifies these dynamics as it typifies the kind of moralizing attitude that identifies the other resentful side as deserving of contempt.9 What starts as anger and moral disapproval – which are emotions that are consistent with engagement and communication – develops into emotional stances of avoidance. It is possible to be angry with someone and be prepared to forgive them if they redress the injury that they have caused. It is also possible to morally disapprove of a person while seeing them as capable of moral improvement. Bitter resentment is instead incompatible with engagement. The person who resents in this way thinks there is no current possibility that the two injustices they have been subjected to (the initial loss or its threat and the judgment that anger is unwarranted) are addressed. It is this sense of the finality of the injustices that grounds the bitterness and the resulting lack of engagement. The contempt experienced by the other side is also incompatible with engagement. It is a negative global moral emotion that evaluates someone as being morally beyond the pale (Bell, 2013). This interlocking of resentment and contempt is what is often described as affective polarization in a society where different constituencies dislike, despise, or even hate each other and avoid interacting with each other as much as possible. The link between the politics of resentment and mistrust of current democratic institutions runs deep but is complex. On the one hand, those whose political outlooks are shaped by resentment blame members of the establishment, including elected representatives and public officials, for giving unfair advantages to undeserving individuals. They also resent these same elites for calling them bigoted, xenophobes, or sexists while depriving them of what they are owed. Therefore, resentful individuals have lost trust in current liberal democratic institutions, which they perceive as working to further the interests of others who are less deserving than they are. On the other hand, those politicians who run the wave of resentment – be it Bossi, Le Pen, or Trump – always present themselves as democrats. Whilst this might be a ruse, it is not clear that these politicians are necessarily in favor of authoritarianism. They are, however, not liberals. They do not see the point of democratic institutions as enabling the peaceful coexistence of groups of citizens with varied opinions and value systems. Rather, they think of the role of democratic government as embodying the will of the people and as offering stability and security. In their view, democratic institutions should give voice to the opinions and values of relatively homogeneous settled populations in societies where the flux of migrants and refugees is severely restricted and within which those

120  Alessandra Tanesini who are admitted must adopt the values and customs of the host culture (Eatwell & Goodwin, 2018). It thus would be wrong to assert that populism is corrosive of democracy in all its incarnations. Instead, its target is liberal democracy which it seeks to replace with “illiberal democracy” (Zakaria, 1997).

6.3  The Paradox of Political Resentment Often commentators express surprise at the political behaviors of those whose politics is colored by resentment. In the aftermath of the Brexit referendum, for example, some journalists were baffled that the constituencies and regions in which the Leave vote was at its strongest were precisely those that stood out to lose out economically. Thus, for example, a reporter for The Guardian pointed out that Ebbw Vale, a Welsh town in the Valleys region which in 2016 was the region with the highest rate of economic deprivation in the whole of north-western Europe, is “a town with almost no immigrants that voted to get immigrants out. A town that has been showered with EU cash that no longer wants to be part of the EU. [Here,] [t] here’s a sense of injustice that is far greater than the sum of the facts” (Cadwalladr, 2016). In the article, the reporter also interviews a resident who appears to justify her vote by expressing the belief that “we put in [the EU] more money than we get out.” Whilst it is unclear what the journalist meant when she wrote that the sense of injustice exceeded the sum of the facts, one is led to two possible but related interpretations of her words. The inhabitants of Ebbw Vale felt such a deep sense of injustice at their relative economic deprivation and loss of status following the closure of heavy industry that they were easily misled into believing false facts about the EU or that they voted for change and against the Tory government – because they found the status quo unbearable – and rationalized their vote with false claims about their economic relation to the European Union. Either way, the article is intended to generate a sense of puzzlement: how could these people vote in a manner that runs contrary to their interests? They must have been persuaded by propaganda or, if not, must be blind and wilfully ignorant of the facts staring at them in the face. This puzzlement closely resembles that which motivated Arlie Hochschild (2016) to interview some Louisiana supporters of the Tea Party whose lives were blighted by pollution but nonetheless opposed environmental regulations. She concluded that these individuals believed they had a stark choice between good jobs or clean air. They would have liked clean air, but they preferred jobs. Hence, they voted for candidates who promised to abolish the Environmental Protection Agency. Hochschild also shows that the dichotomy between jobs and clean air is false. It is possible to have both by stimulating other sectors of the economy that have been displaced by the polluting industries (Hochschild, 2016, pp. 258–260).

The Politics of Resentment 121 Thus, in Hochschild’s explanation, the Tea Party supporters are good people who have been, at least partly, misled by politicians. These explanations for why people living in poverty voted to leave the European Union and victims of environmental disasters are against environmental legislation are not wholly without merit. Yet, they fail to explain the connection between these political behaviors and resentment. Why would a strong sense of being a victim of injustice make one more susceptible to propaganda? Why would it make one more likely to ignore facts pertinent to one’s economic well-being? Concerning Brexit, at least, there is empirical evidence that voters were not motivated by a false belief that Brexit would make them or the country economically better off. On the contrary, a YouGov poll conducted in 2017 showed that 61% of Leave voters said that even significant damage to the UK economy was a price worth paying to get out of the EU (Smith, 2017). Whilst Hochschild’s Tea Party supporters might genuinely believe that they cannot have both jobs and clean air, it is also clear that their commitment to reducing the size of government is rooted in a belief in the value of self-reliance and other aspects of their political outlook. Therefore, we have reasons to suspect that their alleged ignorance cannot fully explain voters’ seemingly paradoxical political behavior. Nevertheless, even if these voters are not ignorant of the relevant facts, there might be a sense in which they choose to ignore them. That is, they are least very selective about which facts they bring to bear in justifying their voting behavior. The claim quoted above by the resident of Ebbw Vale justifying her vote to leave the EU is illuminating and bears repeating: “we put in [the EU] more money than we get out.” There are at least two plausible readings of this claim depending on the referent of “we.” The claim is false when “we” refers to Ebbw Vale or Wales. The claim is true, though, when “we” refers to the whole of the UK, which before its exit was a net contributor to the European Union (Office for National Statistics, 2019). The Guardian journalist appears to have understood her interviewee as intending the first reading and thus takes her to be either ignorant or irrational in her anger. But it is equally plausible to understand her interviewee as making the true claim that the United Kingdom was a net contributor to the European Union. According to this interpretation, the Ebbw Vale’s resident implicitly relies on her temporarily salient identification as a United Kingdom citizen rather than on her more stable identities as Welsh or a resident of the Valleys as a motivation for her vote.10 Such an interpretation would also chime with analyses that show that commitment to the sovereignty of the United Kingdom was one of the motivations of some Leave voters (Eatwell & Goodwin, 2018, ch. 2). Still, one might wonder why a resident of an area of Wales with high rates of economic deprivation would think that the fact that the UK is a net contributor to the EU is good reason to vote Leave. It is not

122  Alessandra Tanesini plausible to attribute to her the belief that extra money would come to Wales as a result. Welsh residents typically think the that UK economy is lopsided, and most government investments target the affluent SouthEast of England (Gray & Barford, 2018; McCann & Ortega-Argilés, 2021). A more plausible interpretation focuses instead on the temporary salience for this voter of her identification as a UK citizen (additional to her Welsh identity) in the context of the Brexit vote. There is good empirical evidence that when people strongly identify with a social group, especially if that identity is perceived as being under threat, they evaluate the evidence on identity-relevant topics in deeply biased ways (Kahan, 2017; Kahan et al., 2007). They are, for example, strongly motivated to discount evidence that is not congruent with behavior affirming the salient social identity. Thus, since standing as a proud and independent people is an essential component of English and British identity, the Leave vote might have been a way of expressing these identities. It would thus not be surprising if Leave supporters justified their votes by engaging in reasoning motivated by the need to affirm their identities as citizens of the United Kingdom, relying exclusively on evidence that supports expressions of independence, and ignoring or discounting evidence indicating that leaving the European Union would be economically damaging.11 Similar attitudes might also explain why supporters of other national populist parties in Europe usually favor leaving the European Union (Ivaldi et al., 2017). I do not doubt that propaganda and motivated reasoning to protect cultural and social identities have played a role in the explanation of Brexit. They might also partly explain the attitudes of Tea Party supporters. The latter might be discounting evidence that one can have good jobs and clean air because their identities as hard-working and self-­reliant people are invested in supporting unfettered free enterprise.12 These explanations are, however, incomplete. Why would a resident of Ebbw Vale care about UK sovereignty in the context of Brexit? Why would she wish for more control to be given to the British Government? After all, Welsh voters typically feel that Westminster ignores their interests. In addition, the most prominent members of the Leave campaign were Tory politicians, while Labour is the vastly dominant political party in Wales, especially in the South Welsh Valleys.13 There is little doubt that those who felt “left behind” and were resentful of politicians and migrants largely voted in favor of leaving the European Union. It is plausible that the rationalizations of their votes exemplify the patterns characteristic of motivated reasoning. But the appeal to motivated ignorance is incomplete since it cannot explain the patterns of social identification that find their expression in Tea Party support and in some constituencies of the Leave vote. To understand these phenomena, we must return to the resentment that animates these constituencies.

The Politics of Resentment 123

6.4  Queue Jumping I have argued in Section 6.2 that the politics of resentment is initiated by anger at a perceived injustice that mutates into resentment when individuals belonging to some powerful social groups appear to dismiss the initial perception of injustice. I have also begun to explain the nature of the apparent unfairness in terms of the alleged special treatment afforded by state agencies to members of some social groups that are perceived as undeserving by those who are angry because of these government’s initiatives. In this section, I further investigate the nature of this perceived initial injustice to understand the outlook that rationalizes welfare provision for recent migrants or the unemployed and initiatives seemingly designed to mitigate discrimination against women and people of color as being unfair. I am guided in my analysis by a metaphor adopted by Hochschild (2016) to illustrate the sentiments of the Tea Party supporters, whose alienation makes them feel as if they are strangers in their own land. This metaphor is meant to capture the narrative or “deep story” that informs the political outlook of her interviewees. It also resonates with the descriptions used in Britain by those who think of themselves as being “left behind.” The metaphor represents groups of people standing patiently in line, waiting for their turn, only to see others (blacks, foreign migrants, refugees) who should be behind them in the queue, cutting in front because of the government’s assistance. The unfairness that provokes anger is thus conceptualized as a failure to respect the order in which individuals as members of social groups should have access to benefits and resources. That is, the metaphor illustrates a view of society as being stratified by social groups where members of some groups are entitled to claim some goods (e.g., housing and jobs) before others who should gain access to them only once those ahead in the queue have been served. Thus, if recent migrants secure good jobs, they are perceived as stealing these positions from long-standing citizens who should have the first pick. One can invoke different principles to justify the ranking of citizens according to their entitlement to accessing goods and services before others. Some principles, such as prior contribution to the state’s finances or permanent residence in the country, are, without a doubt, defensible (Miller, 2016). But other principles, such as giving precedence to men over women or whites over black and brown people, would be wholly inappropriate. I would be astonished, however, if any white man or woman who experiences the success of some person of color as unfair would ever explain the perceived unfairness in explicitly racist terms. After all, as I argued in Section 6.2, their resentment is importantly motivated by their perception that they are unfairly labeled bigots, sexists, or racists. Indeed, left-leaning individuals who do not experience the threat of a loss of status are quick to judge those who do as holding morally

124  Alessandra Tanesini reprehensible views. In reality, at least some of these claims about lost but deserved entitlements are, even if perhaps ultimately mistaken, not unreasonable or bigoted. That said, there is also little doubt that often the experience of unfairness is justified by relying on false stereotypes about people who belong to marginalized groups (Fein & Spencer, 1997). These stereotypes are often myths that supply some appearance of moral legitimacy to dominant groups’ sense of entitlement to superiority (Reicher & Ulusahin, 2020; Tajfel, 2001). For example, a job candidate who resents the success of a black applicant would not justify their sense of unfairness by claiming that whites should be given preference. But they would presume that the black candidate got the job not because of their superior ability or qualifications but because of positive discrimination that would allow them to cut in front of the line rather than take their place in the order of merit determined by candidates’ track records and CVs. According to social identity theory, these patterns of outgroup disparagement and ingroup favoritism, combined with the adoption of stereotyping to legitimize the status quo, are what we should expect in situations of social competition. These are circumstances in which social hierarchies between social groups whose boundaries are impermeable are widely believed to be unstable and illegitimate (Rubin & Hewstone, 2004; Tajfel & Turner, 1979).14 In these circumstances, the social identity of members of the dominant group becomes highly salient to them. This identity is a positive and essential plank of a person’s self-esteem (Martiny & Rubin, 2016). The desire to preserve this sense of self-worth, which is perceived as being under threat, leads to the adoption of defensive strategies of identity management (Fein & Spencer, 1997; Jordan et al., 2005; Rubin & Hewstone, 2004, p. 824). These involve discrimination and prejudice directed at members of marginalized groups who are perceived as threatening to replace the previously dominant group. These features of social identity theory explain some important aspects of the profiles of the Tea Party supporters interviewed by Hochschild (2016) and of at least some Leave voters. Contrary to widespread belief, Leave voters were more commonly found among the affluent and the older working class who did not think of themselves as struggling to make ends meet, as well as a smaller group of genuinely economically deprived individuals, such as some of the inhabitants of the South Wales Valleys (Swales, 2016). Similarly, many of Hochschild’s interviewees were white-collar workers with a reasonably comfortable lifestyle that identified “up” with the wealthy, presumably because of their whiteness, optimism, and enterprise (Hochschild, 2016, p. 217).15 A pattern of upward identification also partially explains the success of the Leave campaign among Welsh voters living in poverty. The strategies adopted by Leave politicians made salient to these voters their dominant identities as UK native citizens while promoting the perception that such dominance was under threat (Eatwell & Goodwin, 2018, ch. 2).

The Politics of Resentment 125 Resentful Tea Party supporters and Leave voters know that they are not near the top of the social order, but they believe that they share some of the characteristics of the wealthy and powerful. Some think of themselves as self-reliant or hard-working. They are all proud of their national identities and see themselves as the true representatives of their nations. Because of their circumstances, these same individuals also know that their social status could easily slip down. It is because they are under threat of seeing those who represent the bottom rung of the ladder overtaking them that they feel first angry and subsequently resentful.16 They mourn the loss of forms of deference and courtesy that they previously enjoyed. To summarize, I have argued in this section that the metaphor of standing in line only to witness others jump unfairly to the front of the queue captures the nature of the perceived injustice that causes the initial anger that mutates into resentment once the individuals who feel victimized are the targets of moral disapproval. The metaphor is apt because it makes explicit the hierarchical conception of society at the root of the politics of resentment. Once we attend to what the metaphor illustrates, we can appreciate that while some of the ranking principles invoked by those who feel unfairly demoted are defensible, others are not. Social identity theory predicts that individuals rely on stereotypes to legitimize prejudicial rankings. The reasons offered by those whose political outlook pivots on resentment exemplify this pattern. I hasten to add that these considerations alone do not entail that affluent liberals are immediately justified in their negative moral judgments directed at those who are aggrieved. Instead, one is tempted to conclude that moral disapproval in this instance is hypocritical since were these liberal-minded citizens to find themselves in a position where their sense of superiority is threatened, they might also engage in derogatory behavior in the service of self-enhancement. If this is true, they might well lack the standing required to be entitled to blame others for their shortcomings (Wallace, 2010).

6.5  Cruel, Magic, and Wilful Hopes Two further aspects of the metaphor of queueing for one’s turn reward scrutiny. First, the metaphor suggests that what citizens hope to achieve is not immediately available to them but requires patience. The metaphor also intimates a kind of competition since even when there is a queue where everyone is eventually served, some individuals always get what they want sooner than others. Second, the metaphor also suggests that one can sustain one’s hope for success by measuring one’s progress in the queue. Even when one has not been able yet to satisfy their desires, they can remain optimistic if they are getting closer to the front of the line. In this section, I argue that to understand why those whose politics

126  Alessandra Tanesini is colored by resentment think that their situation is aptly described by the metaphor of going backward within a queue, we need to look at the nature of the hopes that animate them. These include cruel hopes fostered by an economic system that turns citizens into consumers and wilful hopes promoted by divisive government policies. Hochschild (2016, pp. 136–137) explicitly connects the metaphor of the queue with the fulfillment of the so-called American Dream for wealth, status, and respect. She notes that her interviewees’ anger and resentment are rooted in the feeling that they are not making any progress toward fulfilling their hopes. No matter how hard they work, they are no closer to the front of the line. Instead, they are moving backward because others who were once behind them can now be seen ahead. Many of the aspirations of Hochschild’s interviewees are for the kind of goods that facilitate a good life. Still, others associated with the dream of making it are cruel because they trap people into a life of misery (Berlant, 2011). For example, hopes of wealth, fame, and success are obstacles to the flourishing of those who harbor them. The person whose life plans are determined by these hopes measures achievement primarily in terms of promotions, salary raises, and acquiring the newest luxury item. They work hard to gain these goods. Their hard work creates more highly paid jobs they do not occupy, wealth they do not have, and material goods they do not possess. The longer and harder they work, the more goods become available that they strive to achieve. Those caught in these dynamics are kept in perpetual misery because no matter how much they have achieved, there is always something they lack. Further, even though every promotion or salary rise brings some immediate pleasure, these individuals are always dissatisfied because there is invariably another promotion or salary increase that is not within their grasp. Hence, they never experience themselves as getting closer to the front of the line where the Dream is fulfilled. On the contrary, they might feel they are slipping backward despite their hard work if they live in a society where inequality is increasing. This feeling that their social standing is being diminished provokes the kind of defensiveness predicted by social identity theory and expressed by anger and resentment. Those British citizens who feel they are going backward rather than forward, no matter how hard they try, often have more prosaic hopes than their US counterparts. They tend primarily to desire material goods that facilitate a good life: healthcare, affordable housing, and a decent job. Hoping for these things is not cruel since these goods promote rather than obstruct human flourishing. Nevertheless, the way they hope for these things is being distorted by the divisiveness encouraged by many political messages. Since 2008, and before the recent pandemic, Conservative and coalition UK governments introduced austerity measures that have cut welfare provisions (Poinasamy, 2013). To gain support for these policies, then Prime Minister David Cameron in 2012 also initiated a sustained

The Politics of Resentment 127 rhetorical campaign that sought to convince people that many recipients of the welfare provisions to be cut did not deserve help (Hoggett et al., 2013). The campaign focused on the claim that hard-working families should not be worse off than those who were unwilling to work. The strategy was successful as it deflected public attention away from government actions toward the behavior of people on benefits.17 The success of the government strategy, though, did not directly consist in making people living in poverty believe that the loss in benefits targeted only the undeserving. These same people of modest income were already distrusting governments and thus likely to be skeptical about the claims made by politicians. Further, they would have had firsthand experience of the effects of austerity measures on their prospects and those of friends and family. Whilst the affluent might have, on the politicians’ saying so, believed in the existence of large numbers of fraudulent benefit claimants, those whose income was supplemented by benefits would have been more distrusting and less likely to be unaware of the reality on the ground. Instead, the rhetorical campaign succeeded in gaining the approval of native British people of modest means for welfare cuts that harmed them because it supplied these individuals with the discursive ammunition necessary to cope with the social competition created by austerity measures combined with social norms proscribing xenophobia, racism, and sexism. These circumstances made social identities salient and highlighted the threats to the dominant status of some people who considered themselves entitled because of their native status and hard work. This triggering of defensive strategies of identity management helps to explain the increased stereotyping and derogatory behavior toward marginalized groups, such as the disabled or recent migrants, that occurred in the aftermath of Cameron’s rhetorical campaign (Hoggett et al., 2013, pp. 569–570). One way of describing some of the effects of the government’s rhetoric is as creating the conditions in which some people are put in a position where they can only strive to fulfill their legitimate hopes wilfully. Wilfulness in hoping occurs when one’s aspirations are pursued single-­ mindedly while treating other people as collateral damage or as mere means to one’s ends (McGeer, 2004). Victoria McGeer (2004, p. 116) links wilful hoping to fear and an insecure sense of self. In her view, those who, for whatever reason, are deprived of recognition of their worth as agents often wholly invest their self-esteem into fulfilling some hoped-for ends. For this reason, they are prepared to trample over other people’s interests in the pursuit of these hopes. Even this brief characterization helps to see the connection between wilful hoping and the circumstances of those whose political outlook is colored by resentment. These are citizens whose social groups once enjoyed public recognition through deference and special privileges. Subsequent social changes have created situations of social competition

128  Alessandra Tanesini in which some privileges are widely believed to be morally illegitimate. In these conditions, members of the once-dominant groups tend to re-­ legitimize their dominance by stereotyping members of marginalized groups, which are thus portrayed as meriting a subordinate status. We should understand the divisiveness of UK government policies within this social psychological context. These policies give some citizens the moral discursive means required to justify in their own eyes the wilful pursuit of legitimate hopes for good jobs and comfortable housing. Stereotyping refugees, migrants, black and brown people, or single mothers legitimizes their treatment as collateral damage in the struggle to get to the front of the queue.18

6.6  Good Hopes In this chapter, I offered a diagnosis of some features of polarization, focusing on its affective character. I indicated that the politics of resentment, fomented by populist politicians in Europe, the USA, and elsewhere, have the power to corrode liberal democratic institutions. I analyzed the nature of this kind of resentment using Hochschild’s (2016) metaphor of a queue as a guide to explore different aspects of the resentment felt by those who feel alienated and left behind. In the previous section, I highlighted another emotive feature of the politics of resentment: the connection of resentment to bad hoping. Those who are resentful are experiencing a loss of social status because the dominance of some of their social identities is now generally believed to be illegitimate. The resulting real or feared threat to self-esteem triggers ingroup favoritism and stereotyping of outgroups. These discriminatory behaviors are also manifest in the way in which people pursue their hopes and aspirations. When people find themselves in conditions of social competition, their hoping often becomes wilful and thus harmful to their fellow citizens. In this section, I conclude this chapter with some brief remarks on ameliorative proposals. In the first instance, it should be possible to make progress toward blocking the mutation of anger into resentment if affluent liberal-minded citizens avoid derogatory expressions of moral disapproval directed toward those who feel left behind. These expressions are not warranted for at least two reasons. First, some anger-eliciting circumstances can be reasonably seen as genuinely unfair. Second, these affluent citizens are sometimes hypocritical since they would behave in similar ways were they to find themselves in the situation of their less wealthy counterparts. Here, however, I would like to propose an ameliorative proposal of a different kind. I suggest that broadening access to university education can contribute to the reduction of the politics of resentment.19 Failure to make inroads with people with a university degree is a common feature of populist movements in the US, UK, and continental

The Politics of Resentment 129 Europe make. This is especially true of those who have subsequently developed careers in health, education, welfare, and the creative industries (Eatwell & Goodwin, 2018, ch. 1). This relation between college education and imperviousness to populism and the politics of resentment is not explained by wealth. Some supporters of populism are affluent, and many do not think of themselves as struggling financially. It is also not explained by ignorance and intelligence. There is no reason to believe that supporters of populism are significantly less informed or able than their university-educated counterparts. Instead, empirical research points to socialization into liberal values as the main driver of this significant difference in outlook (Stubager, 2008). This research suggests that universities are institutions populated by lecturers with liberal values. Students learn the same values from their teachers and adopt them partly out of a motivation to fit in with their social group.20 They retain these values after they complete their education because values tend to develop during formative years and remain relatively stable in adulthood. These same values of tolerance, commitment to justice, and belief in equality immunize against a tendency to become wilful in the manner of one’s hoping when facing circumstances of social competition. In other words, liberal values can scaffold a person’s sense of self-worth, making one feel more secure (Jenssen & Engesbak, 1994). These considerations highlight the connection between education and the cultivation of the art of good hoping (Gross, 2019). Universities are liberal environments that offer opportunities for frequent endorsement of values of self-­transcendence, such as benevolence and tolerance.21 The affirmation of these values has been shown to reduce defensiveness (Critcher & Dunning, 2015; Kim & McGill, 2018). Hence, if McGeer (2004) is right to trace the genesis of wilful hoping to insecurity about the worth of the self, then a reduction in defensiveness should help improve how one pursues one’s hopes. I would like to conclude by making one concession and briefly answering an objection. First, I concede that addressing the increasing economic inequalities in the USA and Europe is a crucial step to ameliorate affective polarization. Cultivating good hoping by way of being socialized into university environments is by itself clearly inadequate to address the issues that generate the politics of resentment. Second, one might object that the focus on avoiding defensiveness in conditions of social competition and reorienting the manner of one’s hoping via socialization in an educational setting falls right into the trap of the rhetoric of political hope (Lindroth & Sinevaara-Niskanen, 2019). Focusing on the future fulfillment of hope would distract from the present need for political action. In response, I submit that hoping well involves acting rather than waiting. It does not consist in promising future rewards and incentivizing current passivity. In addition, by throwing light on those current political circumstances that encourage citizens to adopt cruel hopes and to pursue their aspirations wilfully, in this chapter, I have offered support for some of the

130  Alessandra Tanesini claims made by hope skeptics (Warren, 2015). However, recognizing the damage that bad political hopes can inflict on people should not lead us to give up on hope but find ways of hoping better.

Notes









1. There are others. These include increased wealth inequalities but also the creation of some international institutions, such as the World Bank, whose officials are appointed rather than elected (Eatwell & Goodwin, 2018). 2. As Ivaldi et al. (2017) point out, both left- and right-wing populist movements exist in Europe that tap into the same dynamics of resentment. In this chapter, I rely on examples that exemplify the populism of the right. I do not intend to imply that these are the only kind. 3. Several right-wing populist parties in Europe are pro-welfare but wish to restrict access to benefits to settled populations speaking the national language (Eatwell & Goodwin, 2018, ch. 2). 4. It is worth remembering, however, that the slogan was coined much earlier by Gordon Brown when in 2009, he was the Prime Minister in a Labour Government and endorsed by left-wing trade unions (Summers, 2009). 5. Resentment and anger are moral emotions because they involve moral evaluations of situations (Ben-Ze’ev, 2002). 6. Various authors identify the kind of resentment at play in the politics of resentment as the emotion labeled by Nietzsche as ressentiment. See, for example, Katsafanas (2022). While there are similarities between ressentiment and the feeling involved in the politics of resentment, the vital link between dominant social identity and resentment is missed when focusing on ressentiment as a getting even of the weak against the strong. 7. European populists often make this latter claim to justify Islamophobia (Eatwell & Goodwin, 2018, ch. 2). 8. For an account of bitterness as helpless anger, see Cremaldi and Kwong (2022). 9. Clinton used the expression she came to regret during a speech at a gala event (Jacobs, 2016). 10. The inhabitants of the South Wales Valleys strongly identify as Welsh and as residents of the Valleys (Rutter & Cartier, 2018, pp. 44, 211). In England, those who voted for Leave also tended to identify more strongly as English (Swales, 2016, p. 7). 11. For evidence that supporters of Leave and Remain engaged in motivated reasoning about their perceptions of how the British economy is faring after Brexit, see Sorace and Hobolt (2020). 12. See Kahan (2013) for evidence that motivated reasoning explains why Republican voters often deny that anthropogenic climate change is real. 13. That said, the Tory Prime Minister at the time campaigned to remain. 14. Boundaries are impermeable when a person cannot quickly move between social groups. Racial, nationality, or gender groupings are examples of impenetrable social groups. 15. Eatwell and Goodwin (2018, ch. 1) also observe that the prominent supporters of right-wing nationalism are not necessarily working class but belong to the once privileged skilled and semi-skilled workers whose social status is slipping down. 16. See Baldwin (1998, pp. 218–219) on black people representing the bottom rung of the social ladder.

The Politics of Resentment 131 17. It also sparked controversial TV programs such as Channel 4 Benefit Streets, which ran two series aired in 2014 and 2015. 18. Populism is also successful because it brings new hopes that one’s rightful place in a progressing queue is restored. One of the roots of its success is its ability to regain optimism in those who experience the sense of impotence associated with bitter resentment. 19. Broadening access requires a host of policy initiatives to make university education affordable to people from every socioeconomic background. 20. This process is not one of indoctrination but, at its best, relies on students’ ability to reflect critically. 21. See Schwartz et al. (2012) for categorizing these as values of self-­ transcendence.

References Baldwin, J. (1998). Collected Essays (T. Morrison Ed.). New York: Library of America. Bell, M. (2013). Hard Feelings: The Moral Psychology of Contempt. New York: Oxford University Press. Ben-Ze’ev, A. (2002). Are envy, anger, and resentment moral emotions? Philosophical Explorations, 5(2), 148–154. doi:10.1080/10002002058538728. Berlant, L. G. (2011). Cruel Optimism. Durham: Duke University Press. Betz, H.-G. (1993). The new politics of resentment: radical right-wing populist parties in Western Europe. Comparative Politics, 25(4), 413–427. doi:10.2307/422034. Cadwalladr, C. (2016, 25 June). View from Wales: town showered with EU cash votes to leave EU. The Guardian. Retrieved from https://www.theguardian.com/ uk-news/2016/jun/25/view-wales-town-showered-eu-cash-votes-leave-ebbw-vale Cherry, M. (2021). The Case for Rage: Why Anger Is Essential to Anti-Racist Struggle. New York: Oxford University Press. Cramer, K. J. (2016). The Politics of Resentment: Rural Consciousness in Wisconsin and the Rise of Scott Walker. Chicago and London: University of Chicago Press. Cremaldi, A., & Kwong, J. M. C. (2022). Bitterness without hope. Journal of Social Philosophy. doi:10.1111/josp.12455. Critcher, C. R., & Dunning, D. (2015). Self-affirmations provide a broader perspective on self-threat. Personality and Social Psychology Bulletin, 41(1), 3–18. doi:10.1177/0146167214554956. Eatwell, R., & Goodwin, M. J. (2018). National Populism: The Revolt Against Liberal Democracy. London: Penguin Books. Fein, S., & Spencer, S. J. (1997). Prejudice as self-image maintenance: Affirming the self through derogating others. Journal of Personality and Social Psychology, 73(1), 31–44. doi:10.1037/0022-3514.73.1.31. Gray, M., & Barford, A. (2018). The depths of the cuts: the uneven geography of local government austerity. Cambridge Journal of Regions, Economy, and Society, 11(3), 541–563. doi:10.1093/cjres/rsy019. Gross, J. (2019). Practices of hope: care, narrative, and cultural democracy. International Journal of Cultural Policy, 1–15. doi:10.1080/10286632.2019. 1702032.

132  Alessandra Tanesini Hochschild, A. R. (2016). Strangers in Their Own Land: Anger and Mourning on the American Right. New York: New Press. Hoggett, P., Wilkinson, H. E. N., & Beedell, P. (2013). Fairness and the politics of resentment. Journal of Social Policy, 42(3), 567–585. doi:10.1017/ s0047279413000056. Ivaldi, G., Lanzone, M. E., & Woods, D. (2017). Varieties of populism across a left-right spectrum: the case of the Front National, the Northern League, Podemos, and Five Star Movement. Swiss Political Science Review, 23(4), 354–376. doi:10.1111/spsr.12278. Iyengar, S., Lelkes, Y., Levendusky, M., Malhotra, N., & Westwood, S. J. (2019). The origins and consequences of affective polarization in the United States. Annual Review of Political Science, 22(1), 129–146. doi:10.1146/annurevpolisci-051117-073034. Jacobs, B. (2016, 10 September). Hillary Clinton calls half of Trump supporters bigoted ‘deplorables.’ The Guardian . Retrieved from https://www.theguardian.com/ us-news/2016/sep/10/hillary-clinton-trump-supporters--bigoted-deplorables Jenssen, A. T., & Engesbak, H. (1994). The many faces of education: why are people with lower education more hostile towards immigrants than people with higher education? Scandinavian Journal of Educational Research, 38(1), 33–50. doi:10.1080/0031383940380103. Jordan, C. H., Spencer, S. J., & Zanna, M. P. (2005). Types of high self-esteem and prejudice: how implicit self-esteem relates to ethnic discrimination among high explicit self-esteem individuals. Personality and Social Psychology Bulletin, 31(5), 693–702. doi:10.1177/0146167204271580. Kahan, D. M. (2013). Ideology, motivated reasoning, and cognitive reflection. Judgment and Decision Making, 8(4), 407–424. Kahan, D. M. (2017). Misconceptions, misinformation, and the logic of identityprotective cognition. Cultural Cognition Project Working Paper Series No. 164, 1–9. doi:10.2139/ssrn.2973067. Kahan, D. M., Braman, D., Gastil, J., Slovic, P., & Mertz, C. K. (2007). Culture and identity-protective cognition: explaining the white-male effect in risk perception. Journal of Empirical Legal Studies, 4(3), 465–505. Katsafanas, P. (2022). Group fanaticism and narratives of ressentiment. In L. Townsend, R. R. Tietjen, H. B. Schmid, & M. Staudigl (Eds.), The Philosophy of Fanaticism: Epistemic, Affective, and Political Dimensions (pp. 157–183). New York, NY: Routledge. Kim, S., & McGill, A. L. (2018). Helping others by first affirming the self: when self-affirmation reduces ego-defensive downplaying of others’ misfortunes. Personality and Social Psychology Bulletin, 44(3), 345–358. doi:10.1177/0146167217741311. Lindroth, M., & Sinevaara-Niskanen, H. (2019). Politics of hope. Globalizations, 16(5), 644–648. doi:10.1080/14747731.2018.1560694. Martin, A. M. (2014). How We Hope: A Moral Psychology. Princeton: Princeton University Press. Martiny, S. E., & Rubin, M. (2016). Towards a clearer understanding of social identity theory’s self-esteem hypothesis. In S. McKeown, R. Haji, & N. Ferguson (Eds.), Understanding Peace and Conflict Through Social Identity Theory: Contemporary Global Perspectives (pp. 19–32). New York: Springer. Mason, L. (2018). Uncivil Agreement. Chicago: Chicago University Press.

The Politics of Resentment 133 McCann, P., & Ortega-Argilés, R. (2021). The UK ‘geography of discontent’: narratives, Brexit and inter-regional ‘leveling up.’ Cambridge Journal of Regions, Economy and Society, 14(3), 545–564. doi:10.1093/cjres/rsab017. McGeer, V. (2004). The art of good hope. The Annals of the American Academy of Political and Social Science, 592, 100–127. Miller, D. (2016). Strangers in Our Midst: The Political Philosophy of Immigration. Cambridge, MA: Harvard University Press. Office for National Statistics (2019, 30 September). The UK contribution to the EU budget. Retrieved from https://www.ons.gov.uk/economy/governmentpublic sectorandtaxes/publicsectorfinance/articles/theukcontributiontotheeu budget/2017-10-31 Poinasamy, K. (2013). The true cost of austerity and inequality: UK case study. Retrieved from https://www-cdn.oxfam.org/s3fs-public/file_attachments/cstrue-cost-austerity-inequality-uk-120913-en_0.pdf Reicher, S., & Ulusahin, Y. (2020). Resentment and redemption: on the mobilisation of dominant group victimhood. In J. Vollhardt (Ed.), The Social Psychology of Collective Victimhood (pp. 275–294). New York, NY: Oxford University Press. Rubin, M., & Hewstone, M. (2004). Social identity, system justification, and social dominance: commentary on Reicher, Jost et al., and Sidanius et al. Political Psychology, 25(6), 823–844. doi:10.1111/j.1467-9221.2004.00400.x. Rutter, J., & Cartier, R. (2018). National Conversation on Immigration. British Future and HOPE not hate. http://www.britishfuture.org/wp-content/uploads/ 2018/09/Final-report.National-Conversation.17.9.18.pdf Schwartz, S. H., Cieciuch, J., Vecchione, M., Davidov, E., Fischer, R., Beierlein, C., … Konty, M. (2012). Refining the theory of basic individual values. Journal of Personality and Social Psychology, 103(4), 663–688. doi:10.1037/a0029393. Shabi, R. (2017, 8 September). ‘British jobs for British workers’ is back, a fascist incursion into mainstream politics. The Guardian. Retrieved from https://www. theguardian.com/commentisfree/2017/sep/08/british-jobs-workers-fascistgovernment-brexit-plan Smith, M. (2017, 1 August). The ‘extremists’ on both sides of the Brexit debate. Retrieved from https://yougov.co.uk/topics/politics/articles-reports/2017/08/01/ britain-nation-brexit-extremists Sorace, M., & Hobolt, S. B. (2020). A tale of two peoples: motivated reasoning in the aftermath of the Brexit Vote. Political Science Research and Methods, 9(4), 675–692. doi:10.1017/psrm.2020.50. Stubager, R. (2008). Education effects on authoritarian-libertarian values: a question of socialization. The British Journal of Sociology, 59(2), 327–350. doi:10.1111/j.1468-4446.2008.00196.x. Summers, D. (2009, 30 January). Brown stands by British jobs for British workers remark. The Guardian. Retrieved from https://www.theguardian.com/ politics/2009/jan/30/brown-british-jobs-workers Sunstein, C. R. (2009). Going to Extremes: How Like Minds Unite and Divide. Oxford and New York: Oxford University Press. Swales, K. (2016). Understanding the Leave vote. Retrieved from https://natcen. ac.uk/media/1319222/natcen_brexplanations-report-final-web2.pdf Tajfel, H. (2001). Social stereotypes and social groups. In M. A. Hogg & D. Abrams (Eds.), Intergroup Relations: Essential Readings (pp. 132–145). Hove: Psychology Press.

134  Alessandra Tanesini Tajfel, H., & Turner, J. C. (1979). An integrative theory of intergroup conflict. In W. G. Austin & S. Worchel (Eds.), The Social Psychology of Intergroup Relations (pp. 33–37). Monterey, CA: Brooks/Cole. Tanesini, A. (2021). Passionate speech: on the uses and abuses of anger in public debate. Royal Institute of Philosophy Supplement, 89, 153–176. doi:10.1017/ s1358246121000047. Walker, M. U. (2006). Moral Repair: Reconstructing Moral Relations after Wrongdoing. Cambridge: Cambridge University Press. Wallace, R. J. (2010). Hypocrisy, moral address, and the equal standing of persons. Philosophy & Public Affairs, 38(4), 307–341. doi:10.1111/j.1088-4963.2010.01195.x. Warren, C. L. (2015). Black nihilism and the politics of hope. CR: The New Centennial Review, 15(1), 215–248. doi:10.14321/crnewcentrevi.15.1.0215. Zakaria, F. (1997). The rise of illiberal democracy. Foreign Affairs, 76(6), 22–43.

7

Against the Individual Virtue Approach in the Epistemology of Democracy Marko Luka Zubčić

7.1  The Individual Virtue Approach Recently, a significant amount of research into the epistemology of democracy has concentrated on individual epistemic virtues while forgoing the systemic analysis paradigmatically advanced by institutional epistemology (IE). In part, this trend appears to be rooted in an epistemic panic that emerged from inaction to the climate crisis, the proliferation of misinformation on social media, the election of Donald Trump, and the Brexit referendum. It is frequently inspired by the (delayed) apprehension of the epistemic suboptimality of median epistemic agents derived from empirical research on voters’ lack of information and cognitive biases. The relevant normative projects derived from this focus vary. The much more theoretically sophisticated strand aims to distribute individual epistemic virtues in the population – for instance, by advocating for a political epistemology of overcoming intellectual vices (Cassam 2019), pursuing epistemic perfectionism by fostering a form of “epistemic capabilities” in individuals (Talisse 2009), or by cultivating psychological features conducive to epistemically virtuous social deliberation (Tanesini 2021). For some more extreme political epistemologists, if the citizens will not internalize the proper ethos and become subjects that philosophers can recognize as epistemically virtuous, there is always the option to regulate their political and epistemic activity to at least limit their more dangerous excesses. Contemporary “epistocrats” tend to choose this route – restricting or downgrading franchises to reduce the influence of the uninformed and irrational public on political decision-­ making (Brennan 2016). I will group these works in political epistemology under the Individual Virtue Approach (IVA) to epistemic democracy and subsume their diverse articulations under the following central thesis, constituting their inevitable shared underlying commitment.

DOI: 10.4324/9781003311003-10

136  Marko Luka Zubčić The IVA Thesis The epistemic powers of representative electoral democracy, or any governance system, are constitutively determined by individual epistemic virtues of citizens and fundamentally eroded by individual epistemic vices of citizens. This chapter shows that IVA wrongly models the epistemic reliability of governance systems and democracy, prescribes epistemically vicious governance systems, and fails to account for the epistemic deficits of realworld representative electoral democracies. I will define democracy as a system of universally inclusive self-­ governance of free and equal persons. I will leave the term deliberately broad – it thus includes a diversity of possible democratic systems but most relevantly representative electoral democracy. However, I will refer to representative electoral democracy, particularly its current governing families under approximately liberal constitutions and variably open or captured markets, explicitly as representative electoral democracy (RED). The plan of the chapter is as follows. In Section 7.2, I argue that IVA incorrectly models epistemic reliability in dealing with governance problems. Namely, IVA shows that individually epistemically virtuous agents are more likely to solve governance problems. This premise holds whether governance problems are simple or complex problems can be reliably solved by delegating problem-solving to individually virtuous agents. By building on research on the division of cognitive labor, New Diversity Theory, and, more broadly, governance studies, I will show that neither assertion holds under scrutiny. As they are more frequently referred to in planning and governance studies, simple or tame problems have repeating constitutive features that can be reliably acted upon by people with appropriate cognitive history and epistemic character. However, it is uncontroversial that governance problems are not paradigmatically tame. If issues are complex, they require diversity to be solved, if they can even be. Introducing complexity entails introducing constitutive epistemic agent suboptimality, under which diversity becomes the minimal condition of the possibility of epistemic success. Diversity, further­more, constitutively requires individually epistemically vicious agents. In Section 7.3, I argue that democracy is epistemically justified by its systemic properties, not individual epistemic virtue. By building on Elizabeth Anderson’s experimentalism, I argue that epistemic powers of democracy in IE are a result of systemic features of democracy which model epistemic reliability in solving complex problems appropriately – namely, its abilities (1) to harvest unpredictably distributed information and feedback on policies and (2) to protect diversity in its deliberative and decision-making bodies (3) under restraint by a constitutional liberal order. I will refer to this as the systemic account of the epistemic powers of democracy.

Against the Individual Virtue Approach 137 Lastly, in Section 7.4, I trace the implications of the epistemic reliability in dealing with complex problems for institutional and policy designs supported by IVA – namely, showing that both the universal distribution of individual epistemic virtues and the delegation of epistemic labor of governance to individually epistemically virtuous agents constitutes a collective epistemic vice – and provide brief remarks on the epistemic deficiencies of representative electoral democracy, notwithstanding its comparative epistemic advantage to any system derived from IVA.

7.2 Varieties of Epistemic Reliability in Institutional Epistemology IVA implies a wrong model of governance problems and their attendant conditions of epistemic reliability. It follows from IVA Thesis – the epistemic powers of representative electoral democracy are constitutively determined by individual epistemic virtues of citizens and fundamentally eroded by individual epistemic vices of citizens – that the individually epistemically virtuous citizens can or are more likely to solve governance problems both simple (or “tame,” as I will refer to them, following conventional governance studies vocabulary) and complex. Epistemic powers minimally include reliable problem-­solving abilities – being more likely to give correct answers to questions (Goldman 1983). Thus, either of the following two propositions must stand for IVA to hold. First, the proposition that governance problems are tame problems and, therefore, can be reduced to repeating constitutive features which can be reliably recognized by agents with appropriate individual epistemic virtue and acted upon to solve these problems reliably – I will call this the Simple Governance Thesis (SG).1 Secondly, the proposition that when solving complex problems, the best groups will outperform all other groups – I will call this Problem-Solving a Tournament Thesis (PT). However, neither of these two propositions holds under scrutiny. First, tame problems are the paradigmatic epistemic tasks only agents with appropriate individual epistemic virtues can solve. Epistemic virtue broadly denotes a property that renders an individual or a collective more likely to attain knowledge, while epistemic vice indicates the opposite. In the most fundamental sense, an epistemically virtuous individual agent is the one who will pursue the “best-up-to-now” strategy in the search for knowledge. For this chapter, we could delineate thick and thin individual epistemic virtue – the former requires both epistemic character (the agent loves the truth and revises their beliefs following the evidence) and cognitive history (education and history of practice), while the latter requires cognitive history and minimal epistemic character reducible exclusively to reliable willingness to derive correct actions from this mental history when appropriate (in other words, a medical doctor who does not want

138  Marko Luka Zubčić to help the patient even though they could is not epistemically virtuous under this conception of thin epistemic virtue). Tame problems can be reduced to repeating constitutive features, and solving them is a matter of following the instructions derived from the historical record of successful solutions. Only agents with the cognitive history required to recognize the constitutive features and act appropriately can solve the tame problem. An epistemically reliable process for solving tame problems delegates the epistemic labor to the individually epistemically virtuous agents. I will refer to this as “simple reliability.” Notably, as Neil Levy and Mark Alfano have shown in detail, this account of simple reliability does not require thick virtue. A significant number of epistemic practices of arguably greatest relevance for human society, namely, practices of “cumulative culture,” which condition human adaptation and allow for each inherited behavioral repertoire to serve as a basis for future innovation, rely on over-imitation, conformist bias, and prestige bias (Levy and Alfano 2020). Indeed, chimps exhibit more individual epistemic virtue than humans – they experiment while imitating to improve the received behavioral model. In contrast, children imitating adults do not experiment to improve and even resist upgrading their received models even when shown that specific steps could be discarded. Moreover, humans’ tendency to over-imitate appears to increase with age. However, it is precisely this individual epistemic vice of humans that conditions the development of cumulative culture (Levy and Alfano 2020). This account presents simple reliability theorists with a particular conundrum concerning justification. These individually epistemically vicious practices appear to be epistemically reliable in a relevant real-world sense (agents can count on being knowledge-conducive) and thus can serve as a basis for justification. However, these processes need not have input beliefs justified by anything else, as they are derived from this same reliable (but vice-based) process. I will not focus here on this rather curious finding. For this chapter, it is sufficient to posit that even in cases in which the problem is such that a member of the group knows the solution, this need not (and frequently is not) be a result of their open-­ mindedness, curiosity, love of truth, or other thick individual epistemic virtues. It is, however, a result of thin epistemic virtue – an appropriate cognitive history acted upon appropriately. However, governance problems are not paradigmatically tame – the relevant governance problems cannot be reduced to repeating constitutive features. As has been relatively uncontroversial in policy and planning studies, each governance problem is unique because it is defined by specific contingent mutually influencing and potentially unknown factors unique to a particular social situation. For instance, as Elinor Ostrom shows, there is no panacea to problems of collective resource governance – each social situation requires its specific configuration of rules and mechanisms for policy learning (Ostrom 2005). For instance,

Against the Individual Virtue Approach 139 the ceteris paribus formulation of an economic law under idealized assumptions is of limited use for actual financial problems (Gaus 2008). Furthermore, the implementation of any policy will always encounter an abundance of unknown and unpredictable variables, which may lead to considerably worse consequences than the problem which the policy attempted to solve (Gaus 2016) – indeed, it is the critical feature of governance that when it comes to policy, “the real benefits usually are not the ones we expected, and the real perils are not the ones we feared” (Tanner 1996, 272). Furthermore, it is impossible to determine a priori which cognitive history should be favored for recognizing and acting upon the supposed repeating features of the totality of social and governance problems. Indeed, it is impossible to determine this a priori for any social and management problem. What kind of knowledge is required to decide whether to have a vaccine mandate or how to reduce crime (if we were to consider these as tame problems)? Knowledge about representatives and legislative procedures (the “empirical substance of politics”) (Ahlstrom-Vij 2021), moral, situated, technical, strategic, or scientific knowledge? Accordingly, it is impossible to determine a priori which evaluative standard should be used to assess the totality of policies or any given policy – should we exclusively favor “fairness, efficiency, civic virtue, wealth maximization, individual responsibility, security” (Gaus 2008, 16)? Crucially, as Rittel and Webber have shown, some governance problems can be wicked – they cannot be formulated, cannot be tested. The consequences of the “least bad” attempts to resolve them cannot be assessed because it is impossible to determine when the waves of repercussions end and how the effects are linked in a complex social system. Fundamentally, wicked problems have no solution which can be recognized as the solution according to the evaluative standards of all involved (Rittel and Webber 1973). Indeed, the complexity of governance problems is so apparent to anybody willing to honestly look at the real social world that the burden of proof is on those who would claim that governance problems are paradigmatically tame. The Simple Governance Thesis is wrong. Second, IVA theorists might agree that, of course, governance problems are complex. Nevertheless, according to IVA supporters, solving complex issues should require the delegation of epistemic labor to individually epistemically virtuous agents.2 While this proposition might sound intuitive, it does not hold up under closer epistemological analysis. When dealing with complex issues with unique and unknown constitutive features, “no action is guaranteed to succeed or fail, and no history determines an optimal action with certainty” (Mayo-Wilson et al. 2011, 662). In other words, in complex situations, epistemic agents bet. Two fundamental types of complex problems exist: those with a solution and those without (Zubčić 2022). In institutional epistemology, the former is referred to as “learning” problems (Mayo-Wilson et al. 2011). These

140  Marko Luka Zubčić problems have a globally optimal solution, but one that is unknown to all agents. However, the reliable process for reaching this solution requires a diversity of search strategies tested and contested for long enough to avoid getting stuck at suboptimal lock-ins – if we do not know the answer, we can find it only if we hedge our bets. Call this “discovery reliability.” Notice the difference between simple and discovery reliability. In simple reliability, the solution already exists, and some people have it. The task of an institutional system (or a collective) is to streamline the input of those who have the solution. Nobody has the answer to discovery reliability, and the group needs to test and contest different proposals to discover the solution. The instruction to include diversity thus follows from the definition of the problem. Diversity entails protecting investigators pursuing the best-up-tonow and the currently inferior or random3 strategy (Mayo-Wilson et al. 2011). Collective epistemic virtue in learning problems is thus irreducible to individual epistemic virtue since individually epistemically virtuous bet only on the best-up-to-now theory, as instructed by their shared cognitive history (in case of thin epistemic virtue) and epistemic character (in case of thick epistemic virtue). When the problem is complex, the epistemic nature and cognitive history of the best are insufficient to solve it – on the contrary, they render those who share them vulnerable to getting stuck at the same suboptimal lock-in (Kitcher 1990; Page 2008; Weisberg and Muldoon 2009; Zollman 2010; Mayo-Wilson et al. 2011; Smart 2018). Diversity, on the other hand, protects from such lock-ins precisely as it entails a sufficient difference in the cognitive histories and epistemic characters among the members of the problem-solving group for agents not to settle for an inferior theory too soon and, hopefully, to “unstick” each other from their diverse lock-ins. Note that this does not mean this will happen, nor does it suggest, as Hong-Page theorem optimistically models it, that the diverse investigators will build on each other’s lock-ins – all it means is that diversity makes it possible that they avoid suboptimal lock-ins. Critically, for a discovery of a solution to be possible when dealing with learning problems, the group must include members who have different cognitive histories and epistemic characters – and, thus, individually epistemically vicious agents. Research in the division of mental labor and, broadly, collective search tasks provide ample evidence that when the epistemic landscapes have multiple local optima and a single global optimum, as they do in cases of learning problems (as opposed to exclusively single global optimum, which defines the tame problems), individual uninformedness, dogmatism, laziness, biases, and other forms of apparent individual vices contribute to the epistemic success of the group. These individual epistemic vices protect the group from locking in too soon on chance successes of a flawed theory and allow agents to test and contest different solutions (including sufficiently challenging the

Against the Individual Virtue Approach 141 best-up-to-now strategy) and explore diverse and distant strategies in the search for knowledge – more precisely, these individual vices are a function of diversity (Kitcher 1990; Zollman 2010; Mercier and Sperber 2011; Smart 2018; Levy and Alfano 2020). Indeed, individual epistemic vices are mere normative diversity4 – the vicious agents search the landscape under a different set of rules than the individualist epistemologist prescribes. Note here that this does not entail some caricature view that all individual epistemic vices in all situations are collectively epistemically beneficial, nor that all individual epistemic virtues are collectively epistemically detrimental. Quite the opposite, it entails that in solving learning problems, both can entail either positive or negative consequences. An a priori determination of the best distribution of virtues and vices across the totality of situations is impossible. Crucially, improving on any given individual epistemic virtue does not increase the epistemic power of a population or a system. Without diversity, agents may find the global optimum because they are lucky. This outcome, however, only further strengthens the case for diversity as the minimal condition of epistemic reliability in learning problems – as both John Stuart Mill and Miranda Fricker, most prominently, argued, if free and open disagreement is not protected, true propositions cannot be justified, and thus cannot become knowledge (Mill 2015; Fricker 2015; see also Kelly 2006). Therefore, if we follow only the best-up-to-now strategy in learning problems, even if it shows to be correct, this is a case of blind epistemic luck and not of a reliable process. Nevertheless, in learning problems, the diversity is transient (Zollman 2010) – it collapses into a single strategy when the global optimum, the actual solution, is found. This characteristic also requires at least an overlap in evaluative standards (or at least some of their relevant demands) of otherwise diverse investigators – they can all agree on the global peak once it is found. Learning problems may then become tame problems – if the group discovers the solution, they might also discover the constitutive features. Scientific issues, once solved, can become technical problems. This further points to a relevant addition to simple reliability, namely, that it is conditioned on discovery reliability – if the group reduces the diversity too soon and mistakes a suboptimal lock-in for a solution, they will continue to use the same flawed strategy to solve future problems with constitutive features they recognize as similar or identical. However, learning problems in governance rarely (or never) become tame because the constitutive features of social, political, and economic difficulties rarely (or never) have repeating elements.5 Moreover, it is seldom evident when the learning problem is solved in actual governance. Indeed, as I will expand on in the next section, and as Elizabeth Anderson rightly emphasizes, therefore feedback mechanisms and protection of disagreement after the decision has been made are fundamental to the epistemic powers of democracy – they allow for the item to return to the

142  Marko Luka Zubčić agenda if the population signals that the problem persists, or the solution is suboptimal. In complex problems, all agents are epistemically suboptimal – their knowledge and character are insufficient to solve the problem. Thus, IE analysis assumes what many IVA theorists have recently discovered. Diversity does not guarantee they will find the solution because no process can ensure that a group of epistemically suboptimal agents will solve a problem. It does, however, make it possible. The situation complicates further when dealing with complex problems that do not have a solution – wicked problems. In both learning and wicked problems, the critical comparative standard of epistemic reliability is how able the system is at avoiding error. In learning problems, the iteration of this negative reliability constitutes positive reliability – ­conditions that make it possible to prevent error with time leading to truth. In wicked problems, however, only negative reliability is possible. Wicked problems can be best described in “complex evaluative spaces” (CES). Gerald Gaus and Fred D’Agostino’s work in New Diversity Theory argues for the distinct epistemic value of free diversity for this type of problem space (D’Agostino 2009; Gaus 2016, 2018). This approach describes the epistemic situation through modeling the search for “peaks” in a “landscape.” CES is defined by “rugged landscapes” in which peaks are scattered randomly, meaning that the closeness of points in the landscape does not account for the similarity in their epistemic value. In other words, the landscape is not “smooth” because the agents are searching for a solution to keep going to arrive at the peak. CES is also defined by deep evaluative pluralism, meaning that an object in a domain may be assessed according to a plurality of evaluative standards, which are interdependent “in the sense that modifying an object in a way which changes its value against one standard may also change its value against another standard” (D’Agostino 2009, 104). Thus, no object version can score the highest according to all evaluative standards. How should boundedly rational agents search such landscapes? “Myopic” search – for instance, all agents following exclusively best-up-to-now theory – will get locked in at a suboptimal peak. “Exhaustive” search – searching the entire landscape – is inefficient. We might decompose the object into sub-objects, problem to sub-problems, and assign a team to work with those search routines, looking for their local peaks – thus running a “parallel” search. CES, however, are not highly decomposable – the sum of local peaks found in a parallel search need not represent a global peak. CES exhibits interdependencies – a local peak according to M1 is a valley according to M2. If M1 and M2 went their separate ways and returned to report to each other, they would find that their respective search results exhibit very low commensurability, if not outright incommensurability. In such situations, parallel search is inconclusive “because there is no natural ‘equilibrium’ into which it can rightly settle” (D’Agostino 2009, 112).

Against the Individual Virtue Approach 143 This is so because the parallel search does not sufficiently partition the total problem space. By allowing the agents to explore the landscape according to their evaluative standards and thus performing a “diversified” search, we increase the group’s effectiveness while decreasing their chances of a lock-in – they will search the landscape according to an actual (and not exhaustive) variety of evaluative standards. They will not all get stuck in the same suboptimal search routine. They might “cross paths” and discover peaks they share – in other words, they might encounter unexpected learning sub-problems “sprinkled” through the landscape. In total CES, however, only the diversification of evaluative standards and consequent adverse selection (shallow consensus on search routines that lead to suboptimal lock-ins) exhibit weak epistemic reliability. When the problem is such that the solution cannot be recognized, the best we can do is not choose a single way of solving it and try to identify how diverse attempts have failed. The only reliable process available for dealing with wicked problems is thus negative – not betting on a single strategy but providing conditions for diversifying methods in the search for knowledge. I will refer to this variety of epistemic reliability as “epistemic immunity.” Critically, discovery reliability is conditioned on epistemic immunity in social epistemic systems because the latter protects the pool of diverse investigators, which provides diversity for dealing with learning problems. In a social epistemic system lacking epistemic immunity, the diversity is already “normalized” according to some controversial evaluative standards (Gaus 2016) and thus insufficient for reliable problem-solving in learning problems. The key takeaway is that complexity and diversity are not incidentally but conceptually linked, as are diversity and individual epistemic vice. Complex problems, by definition, require diversity, and diversity, by definition, requires individual epistemic vice. These conceptual implications had to be made explicit to show that IVA cannot stand in any governance system. Instead, the opposite stands – the epistemic powers in governance systems are not constituted by individual epistemic virtue. To rely on individual epistemic virtue is to model an unrealistically simple social epistemic system. To conclude, complex problems are not reliably solved by delegating problem-solving to individually virtuous agents. Problemsolving as a Tournament Thesis is then also wrong.

7.3  The Epistemic Powers of Democracy If SG and PT propositions do not stand, the epistemic powers of a governance system cannot be based on agents’ epistemic character or cognitive history. Indeed, they are not. Instead, the epistemology of governance systems, and thus of democracy, rests precisely on the systemic properties which mitigate and manage the constitutive epistemic suboptimality

144  Marko Luka Zubčić of epistemic agents faced with complex problems. In a democracy, these systemic properties are threefold – I will refer to them as “core epistemic functions.” They restate Elizabeth Anderson’s fundamental argument on the epistemic powers of democracy. In her seminal paper “Epistemology of Democracy,” Elizabeth Anderson (2006) argues that the epistemic powers of democracy cannot be accounted for by either Jury Theorem or Diversity Trumps Ability (DTA). Jury Theorem6 is unsatisfactory for institutional epistemology (IE) because it models agents as minimally more likely to be correct than wrong, while the founding insight of IE was precisely that epistemic agents are constitutively epistemically suboptimal (Hayek 1945; Anderson 2006; Zubčić 2021), and thus minimally more likely to be wrong than right. The Jury Theorem’s model then provides the opposite result – it leads to a worse aggregative outcome. However, as Anderson notes, the Jury Theorem is an unsatisfactory model of democracy because it fails to account for the diversity required when problems are complex and needs feedback on policies. Surely, an important part of the case for the epistemic merits of democracy rests on its ability to pool this asymmetrically distributed information about the effects of problems and policies to devise solutions that are responsive to everyone’s concerns. We, therefore, need a model of democracy in which its epistemic success is a product of its ability to take advantage of the epistemic diversity of individuals. (Anderson 2006, 11) Furthermore, the Jury Theorem models agents independent of one another – they do not deliberate together, a collective epistemic virtue for epistemically suboptimal agents. Both votes and talk – both imperfect but robust – condition agents’ abilities to come up with solutions in unpredictable ways.7 DTA’s core contribution to IE is that it provides further – admittedly, one of the most popular as well as too epistemologically optimistic8 – modeling support for the claim that diversity, and the introduction of randomness in the search for knowledge, is a collective epistemic virtue when the problem is complex or, more precisely, agents are suboptimal. However, for Anderson, DTA cannot model epistemic powers of democracy because it fails to capture the critical systemic feature of liberal democracy, which renders it more likely to attain knowledge, solve problems or learn – namely, feedback required for a revision of erroneous commitment. Therefore, experimentalism best models the epistemic powers of democracy, argues Anderson. The iterative process of policy and feedback, test and contest, and the protection of disagreement before, during, and after the decision is made provides liberal democracy its epistemic powers. Anderson is concerned exclusively with representative electoral democracy, which, she argues, derives its epistemic powers

Against the Individual Virtue Approach 145 from its ability to satisfy the experimentalist core epistemic functions comparatively better than other forms of governance. Thus, democracy is epistemically robust because it aims to harvest distributed expertise9 dispersed in the population – feedback on policies and the unpredictably distributed unique and relevant information relevant for policy-making. Democracy harvests distributed expertise through its liberal constitution, which allows for freedom of speech and association, allowing for disagreement before and after the policy decision. But, critically, representative electoral democracy (RED) harvests the feedback through periodic elections on different levels of government. Note that distributed expertise differs from concentrated expertise of cognitive history. It moreover requires only minimal individual epistemic features of agents – being able to report the truth. The harvest does not entail that those public members who share relevant information and feedback exhibit epistemic virtue in any substantial sense. They may have unique information without being the “best” reasoner in the population (for instance, due to their social position, which allows them unique insights into some parts of the causal chain in policy implementation in a complex social world, or simply because they are on the receiving end of a policy). They may share the information for non-epistemic reasons (for instance, simply because it is in their interest to change a policy that they find harmful). The unpredictability of the distribution of the unique information is critical here – the relevance of input cannot be determined by education, professional track record, or other similar social markers of epistemic competence. Instead, relevant input can “hide” among people who tend not to pass the philosopher’s dubious standards of individual epistemic virtue – people who are “low-skilled” and less formally educated in areas relating to policy-making or science, and who may exhibit close-mindedness, gullibility, dogmatism, laziness, or any other of the individual epistemic vices (Cassam 2017). As in free speech, so with elections, universal inclusion is the function of diversity. There is no a priori road map to total governance problems, and universal inclusion is a robust principle for harvesting unpredictable distributed expertise. I will later argue that elections of REDs imperfectly satisfy this function. Second, democracy is epistemically robust because it protects diversity in its institutional deliberative and decision-making body (in representative electoral democracies, the parliament) as well as in deliberation and decision-making “in the wild” (through the liberal constitution). Diversity protects from suboptimal lock-ins which haunt the homogeneous groups of the “best and brightest.” As I will also argue later, RED imperfectly satisfies this function but indeed outperforms giving power to the virtuous. Third, both first functions are quite obviously conditioned by constitutional liberal order. If there is no freedom from oppression and poverty10 (Gaus 2016; Zubčić 2022), neither harvest nor diversity function can be

146  Marko Luka Zubčić satisfied because the pool for either harvesting unpredictably distributed information or feedback, or maintaining diversity in the decision-­making body, is critically limited in an illiberal order. Indeed, an illiberal order is an epistemic catastrophe. If democratic decision-making overrides the liberal constitution, democracy is no longer epistemically reliable (see Gaus 2019 for detailed arguments on the priority of liberal order to reliable self-governance). Relevant collective governance problems that can be solved are learning difficulties. The first two core epistemic functions mirror discovery reliability – they protect diversity and harvest feedback and unique information required for testing and contesting strategies. The third core epistemic function reflects epistemic immunity – it protects free diversity. The first two core functions (discovery reliability) are conditioned by the third (epistemic immunity). Thus, the epistemic powers of democracy are predicated on three core epistemic functions: (1) harvest of unpredictably distributed information and feedback on policies, (2) protection of diversity in its deliberative and decision-making bodies, and (3) restraint by a constitutional liberal order. These are, notably, systemic properties – collective epistemic virtues. To argue against the epistemic powers of democracy, one must show how it fails to satisfy these functions.

7.4 Epistemic Capture and the True Vice of Representative Electoral Democracy While IVA supporters are numerous, and a separate chapter would be required to overview their contributions and particularities of their accounts (many of which are highly insightful and relevant within the bounds of individualist epistemology), the spectrum of IVA can be roughly boiled down to two institutional design types derived from its fundamental commitment. On one end, IVA may lead to instruction to design policies that aim to distribute individual epistemic virtues in the population universally. Conversely, IVA may lead to delegating the epistemic labor of governance to the members of the population deemed individually epistemically virtuous or at least giving significantly more weight to their input. These political programs aim to be compatible with representative electoral democracy, but primarily for non-epistemological reasons, and they tend to feature a solid tilt toward third-party governance. It follows from the analysis of reliability under complexity in Section 7.2 that both designs constitute a collective epistemic vice – they mistake epistemic immunity and discovery reliability for simple reliability and thus provide institutional conditions that render the population less likely to attain knowledge and solve problems or learn. I will refer to this specific type of collective epistemic vice – namely, epistemically unjustified reduction of relevant diversity in solving complex issues – as

Against the Individual Virtue Approach 147 “epistemic capture.” Commitment to IVA is thus committed to epistemic capture. However, epistemic capture is not a property only of some imaginary technocratic, scholocratic, and perfectionist systems – it is very much a collective epistemic vice present in contemporary representative electoral democracies (RED). Real-world democracies vary significantly along many axes, and the comprehensive comparative analysis of their epistemic features requires a separate paper, or more likely a book. I will focus briefly on three interrelated epistemic deficiencies that plague them. First, real-world democracies frequently fail to protect free diversity. Secondly, they are commonly run by socioeconomic elites and thus lack diversity in deliberative and decision-making bodies. Thirdly, elections, the bedrock of RED, fail to properly harvest feedback because of unprotected free diversity, epistemic capture by socioeconomic elites, and lack of subsidiarity. The most egregious epistemic defect of many actual democracies is that they fail to protect free diversity because they fail to protect the population either from oppression or poverty. The United States of America is one of the leading examples of this fundamental failure – while formally a liberal democracy, its systemic racism, exclusionary socioeconomic inequalities, and lack of social security severely compromise the most relevant diversity to be protected. Furthermore, one of the most repugnant violations of freedom of the 21st century has been the treatment of refugees and migrants by the European Union and its member states. Many REDs suffer from epistemic capture by violation of free diversity. They moreover suffer from epistemic capture by socioeconomic elites. For one thing, our parliaments consist of the well-off who went to specific schools, have been groomed to exhibit specific manners of behavior, and have a specific (limited) insight into the world’s workings. This elite bias, quite straightforwardly, reduces diversity in governing bodies. However, affluence influences elections in actual democracies – in the United States in particular, Martin Gilens and Benjamin Page’s work shows that the outcomes of elections depend on the wealthiest 10 percent (Gilens and Page 2014). “Representative democracy as we know it, at least in the United States, is more akin to rule by economic elites – plutocracy, some might say – than to the rule of the people” (Landemore 2020, 28). While, in principle, open and constrained mainly by liberal constitutions, contemporary REDs tend to be, in effect, much more a system of governance by competing socioeconomic elites than a system of self-governance. Lastly, the power of partisan elites in less free democracies may compromise their epistemic performance through media, legal system, markets, and administrative state – however, it is conditioned by elections. In actual democracies, elections tend to be a robust but flawed mechanism for feedback on policy, particularly under conditions of violated free diversity and elite capture of governance. For one thing, they are not

148  Marko Luka Zubčić resistant to hijacking by tyrants and cartels. Nevertheless, more importantly, they are not sensitive to mapping the polity to the problem – in other words, highly centralized 1-person-1-vote electoral systems lack subsidiarity (Weyl 2019). While upgraded institutional mechanisms for feedback on policy and diversification of deliberative and decision-making bodies must be and are currently tested, RED satisfies core epistemic functions comparatively significantly better than IVA systems. If RED takes place within a liberal constitution, a system of social security, and implemented anti-trust laws, and if the financing of parties and political activity is well regulated, free citizen association and free media can protect public deliberation and influence policy-making to a relevant degree, and voters can punish bad governments in elections. While the varieties of technocracy and scholocracy (including those compatible with representative electoral democracy) may contain some mechanisms for the pooling of information from the population (they might have learned that such means are epistemically instrumental from epistemic democrats and liberals), they are built on the assumption that some form of individual epistemic virtue is necessary to allow participation in self-governance on epistemological grounds. This assumption is wrong, and the systems built on it are inevitably epistemically vicious.

7.5 Conclusion Individual epistemic virtue is mainly irrelevant to the epistemic powers of democracy. Indeed, the founding problem of IE is how to design epistemically powerful institutions in the face of constitutive epistemic suboptimality of individual epistemic agents. Democracy, for its part, is epistemically robust because it aims to harvest feedback and unique information dispersed in the population and to protect diversity in the deliberative and decision-making body. Policy developers and institutional designers ought to examine mechanisms that better fulfill these epistemic functions than the mechanisms of actual representative electoral democracies. On the other hand, any policy-making or institutional design informed by IVA Thesis in any of its incarnations would have epistemically detrimental or catastrophic consequences. Given the epistemic panic plaguing the academic, media, and policy classes, epistemic capture masked as an expert rule or epistemic perfectionism presents the most significant contemporary threat to our search for knowledge.

Notes

1. Jeffrey Friedman has recently provided a detailed analysis of the commitment of both technocrats and populists to governance as a process of solving simple problems with manifest solutions (Friedman 2020). While I

Against the Individual Virtue Approach 149















find his account partly precise, convincing, and worth further discussion, it would require an additional paper to offer a clear and comprehensive analysis. 2. IVA theorists might argue that the entire delegation of epistemic labor of governance to individually epistemically virtuous agents cannot be justified politically or morally. Thus, a softer, quasi-scholocratic approach should be favored. However, I am concerned exclusively with the instrumentalist epistemological aspects of institutional design – namely, an institutional design that is more likely to lead to knowledge, problem-solving, or learning. 3. This feature has also been already discovered through cumulative culture: there is considerable evidence that randomizing for protection against bias is a hidden function of culturally transmitted religious practices of consulting augurs for relevant social decisions, such as where to plant crops or hunt, which is itself an epistemically reliable practice which has to be described by reference to the individual epistemic vice (Levy and Alfano 2020, 903). 4. Normative diversity is where cognitive history and epistemic character meet – a particular cognitive history conditions agents to focus on different aspects of the problem, recognize further evidence, and follow additional abductions. 5. Strategies in learning governance problems may be partially decomposable into tame problems. For instance, if the population decides to test a vaccine mandate, administering the vaccines would constitute a tame problem. 6. According to Condorcet’s Jury Theorem, if group members are minimally more likely to be correct than wrong, adding more members to the group increases the group’s probability of making the right decision. 7. The classical criticism of deliberation is that deliberative groups in the real world consistently fall prey to epistemic defects. Thus deliberative ­activities do not spontaneously satisfy the ideal conception of deliberation (Ahlstrom-Vij 2019). However, these “low-hanging fruit” objections miss the constitutive value of deliberation, as argued for by pragmatists (Brandom 2001; Talisse 2009) and recently prominently by Mercier and Sperber (2011) – without the exchange in the space of reasons, there are no epistemic agents at all. While this is more than enough to commit to free social deliberation, Ostrom also reports that in cases of collective resource governance, groups engaged in “cheap talk” perform significantly better than their independent counterparts (Ostrom 2005). That said, I do not argue here in favor of any particular strand of deliberative democracy alone but consider talk, votes, and prices to be the elementary, robust, but imperfect mechanisms for harvesting collective intelligence, and I believe a viable system makes use of all three in their diverse variants. I do not believe we should choose one exclusively – I think this would make the system less reliable. 8. Roughly, Diversity Trumps Ability Theorem states that if (1) the agents are sufficiently diverse that they do not get stuck at the same local optimum, (2) the problem is complex, and thus no agents have the solution, and (3) the agents can build on each others’ local optimums (the problematically optimistic assumption), then the diverse group will outperform the group of the best agents (Page 2008). 9. While insufficient in learning problems, some knowledge and character are necessary because some agents must also pursue the best-up-to-now strategy. In governance problems, this at least includes situated, scientific, civic,

150  Marko Luka Zubčić and moral knowledge – reducing the relevant input in problems in complex social systems a priori to exclusively scientific, or for that matter to situated or moral, is unjustified and frankly strange. 10. Zubčić (2022) argues in detail how the protection of free diversity necessitates the defense of freedom from poverty.

References Ahlstrom-Vij, Kristoffer. 2019. “The Epistemic Benefits of Democracy: A Critical Assessment.” In The Routledge Handbook of Social Epistemology, edited by Miranda Fricker, Peter J. Graham, David Henderson, and Nikolaj J.L.L. Pedersen. New York and London: Routledge. Ahlstrom-Vij, Kristoffer. 2021. “Do We Live in a ‘Post-Truth’ Era?” Political Studies, 1–17. https://journals.sagepub.com/doi/10.1177/00323217211026427 Anderson, Elizabeth. 2006. “The Epistemology of Democracy.” Episteme: A Journal of Social Epistemology 3, no. 1–2: 8–22. Brandom, Robert. 2001. Making It Explicit: Reasoning, Representing, and Discursive Commitment. Cambridge and London: Harvard University Press. Brennan, Jason. 2016. Against Democracy. Princeton: Princeton University Press. Cassam, Quassim. 2017. “Vice Epistemology.” The Monist 99, no. 2: 159–180. Cassam, Quassim. 2019. Vices of the Mind: From the Intellectual to the Political. Oxford: Oxford University Press. D’Agostino, Fred. 2009. “From Organization to Division of Cognitive Labor.” Politics, Philosophy & Economics 8, no. 1: 101–129. Fricker, Miranda. 2015. “Epistemic Contribution as a Central Human Capability.” In The Equal Society: Essays on Equality in Theory and Practice, edited by George Hull, 73–91. London: Lexington Books. Friedman, Jeffrey. 2020. Power without Knowledge: A Critique of Technocracy. Oxford: Oxford University Press. Gaus, Gerald. 2008. “Is the Public Incompetent? Compared to Whom? About What?” Critical Review 20: 291–311. Gaus, Gerald. 2016. The Tyranny of the Ideal: Justice in a Diverse Society. New Jersey: Princeton University Press. Gaus, Gerald. 2018. “The Complexity of a Diverse Moral Order.” The Georgetown Journal of Law and Public Policy 16: 645–679. Gaus, Gerald. 2019. “What Might Democratic Self-Governance in a Complex Social World Look Like?” San Diego Law Review 967. Gilens, Martin and Page, Benjamin. 2014. “Testing Theories of American Politics: Elites, Interest Groups, and Average Citizens.” Perspectives on Politics 12, no. 3: 564–581. Goldman, Alvin. 1983. “Epistemology and the Theory of Problem Solving.” Synthese 55, no. 1: 21–48. Hayek, Friedrich. A. 1945. “The Use of Knowledge in Society.” The American Economic Review 35, no. 4: 519–530. Kelly, Paul. 2006. “Liberalism and Epistemic Diversity: Mill’s Sceptical Legacy.” Episteme 3, no. 3: 248–265. Kitcher, Philip. 1990. “The Division of Cognitive Labor.” The Journal of Philosophy 87, no. 1: 5–22.

Against the Individual Virtue Approach 151 Landemore, Hélène. 2020. Open Democracy: Reinventing Popular Rule for the 21st Century. Princeton: Princeton University Press. Levy, Neil, and Alfano, Mark. 2020. “Knowledge from Vice: Deeply Social Epistemology.” Mind 129, no. 515: 887–915. Mayo-Wilson, Conor, Zollman, Kevin, and Danks, David. 2011. “The Independence Thesis: When Individual and Social Epistemology Diverge.” Philosophy of Science 78, no. 4: 653–677. Mercier, Hugo and Sperber, Dan. 2011. “Why Do Humans Reason? Arguments for an Argumentative Theory.” Behavioral and Brain Sciences 34, no. 2: 57–111. Mill, John Stuart. 2015. “On Liberty.” In Utilitarianism and Other Essays, edited by Mark Philp and Frederick Rosen, 5–112. Oxford: Oxford University Press. Ostrom, Elinor. 2005. Understanding Institutional Diversity. Princeton and Oxford: Princeton University Press. Page, Scott. 2008. The Difference: How the Power of Diversity Creates Better Groups, Schools, Firms, and Societies. Princeton: Princeton University Press. Rittel, Horst and Webber, Melvin. 1973. “Dilemmas in a General Theory of Planning.” Policy Sciences 4: 155–169. Smart, Paul. 2018. “Mandevillian Intelligence.” Synthese 195: 4169–4200. Talisse, Robert. 2009. Democracy and Moral Conflict. Cambridge: Cambridge University Press. Tanesini, Alessandra. 2021. “Virtues and Vices in Public and Political Debate.” In The Routledge Handbook of Political Epistemology, edited by Michael Hannon and Jeroen de Ridder, 325–336. New York and London: Routledge. Tanner, Edward. 1996. Why Things Bite Back. London: Fourth Estate. Weisberg, Michael and Muldoon, Ryan. 2009. “Epistemic Landscapes and the Division of Cognitive Labour.” Philosophy of Science 76, no. 2: 225–252. Weyl, Glenn. 2019. “The Political Philosophy of RadicalxChange.” Accessed May 19, 2022. https://blog.radicalxchange.org/blog/posts/2019-12-30-gqx4th/ Zollman, Kevin. 2010. “The Epistemic Benefit of Transient Diversity.” Erkenntnis 72, no. 1: 17–35. Zubčić, Marko-Luka. 2021. “Ignorance, Norms and Instrumental Pluralism: Hayekian Institutional Epistemology.” Synthese 198, no. 6: 5529–5545. Zubčić, Marko-Luka. 2022. “Disagreement without Discovery and the Epistemological Argument for Freedom from Poverty.” Synthese 200, no. 1: 1–19.

8

Institutional Cynicism and Civic Virtue Ian James Kidd

The power of accurate observation is commonly called cynicism by those who have not got it. (George Bernard Shaw)

8.1 Introduction Scholars are divided on the relationship between cynicism and political life. Jeffrey Goldfarb spoke for many when he claimed, in The Cynical Society, that public cynicism is “the single most present challenge facing American democracy today.” Cynicism, he worried, “dominates the assumptions of our political and cultural life” and has become “confused” with the ideals of “democratic deliberation and political wisdom.” If Goldfarb is right, endemic public cynicism is destructive because it “promotes acceptance of the existing order of things” (Goldfarb 1991: 2, 30). Worries about cynicism and democracy recur in the work of Patrick Deneen. In Democratic Faith, he warns us not to be tempted by “the retreat into easy optimism or the temptation to a kind of democratic cynicism or despair,” each of which constitutes threat to democratic governance (Deneen 2009: 12). Other voices offer more encouraging views on cynicism and its relation to democratic politics, often with a historical stance on forms of cynicism. Historian Sharon Stanley argues that “cynicism inevitably constitutes an ineradicable element of democracy” that can, with care, be “managed and mobilized in ways that are perfectly hospitable to the continued, healthy functioning of democratic life” (Stanley 2012: 181ff). She rightly distinguishes cynicism from other forms of disillusionment and invites us to consider differing kinds of cynicism. The classical Cynics used nature as the standard to attack corrupting social conventions – the stance of Diogenes of Sinope, the “man in the tub” (Hard 2012). It is different from the kinds of cynicism that emerge after the Enlightenment (Shea 2010). Stanley uses this fact of historic pluralism to challenge a “presumed connection between cynicism and political withdrawal”: DOI: 10.4324/9781003311003-11

Institutional Cynicism and Civic Virtue 153 It is virtually taken for granted among contemporary commentators, inside and outside of academia, that cynicism corrodes and destroys the civic dispositions necessary for a healthy democracy. Contrary to all such readings, I argue that cynicism is an ineradicable and constitutive component of democracy. Yet, this fact need not spell the collapse of democracies worthy of our allegiance. (Stanley 2012: 179) So, on the one hand, we can sometimes put cynicism to work in democratic cultures. It may be part of a rational response to the morally crumpled reality of political life. On the other hand, as every scholar tells us, there are many meanings and senses to the term cynicism. Some may indeed contribute to processes that lead to the “collapse of democracy.” Others, though, might help sustain a commitment to democratic governance. A cynic could, but need not, be an enemy of democracy. Such is the optimistic spirit of Helen Small’s tellingly titled book, The Function of Cynicism at the Present Time. She argues for a pluralist account: certain cynicisms have a “curative property” if they become “functional” in serving more complex “aims” – aims that can separate functional cynicism from the “more casualized and corrosive” kinds (Small 2020: 226). One lesson of these accounts is that many things can get called cynicism and affect democratic governance and public culture (Citrin 1974; Eisinger 2000). In this chapter, I describe and endorse what I call institutional cynicism and suggest it can feature within kinds of virtuous civic stances in democratic societies. Put in rough detail, institutional cynicism is a critical appreciation that an institution’s norms, practices, and ethos tend to be corrupting – that is, liable to damage the moral character of those working within the institution. As I experience the institution as corrupting, I will be distrustful, skeptical, and critical of how it is organized and tends to operate. After all, if it is corrupting, something is deeply wrong with it. Other things might be wrong with it, too, but the concept of institutional cynicism is not meant to capture everything that can be wrong with an institution. Institutional cynicism of this sort will not be an entirely happy disposition, but that should not stand against it. Appraisal of dispositions is a matter of determining their necessity rather than their niceness: cynicism might be bitter medicine, but one we need to take. I accept that other forms of cynicism can be as destructive and as anti-democratic as critics insist. But being cynical involves, among other things, being careful about the cynical outlook one adopts. Any disposition, taken to an extreme, will cause us problems. With cynicism, everything depends on the forms it takes, its manifestations in interpersonal interactions and political life, and its systematic effect on our overall civic conduct (Mazella 2007). There are three compelling issues I do not engage in. First, if the institutional cynicism I describe is a virtue, is it a burdened virtue? These

154  Ian James Kidd dispositions help navigate hostile environments while jeopardizing or compromising their bearers’ flourishing (cf. Tessman 2005). Second, I do not consider the idea that an institution may be served if they contain a few exemplary institutional cynics. It may be that, though everyone needs some cynicism, it is enough that a few of us have the virtue. A third issue is the other virtues or dispositions that support institutional cynicism. Candidates might include truthfulness and a kind of civic hope (Snow 2018; Williams 2002: 206f). If institutional cynicism is a virtue, then it would be one of the civic virtues. Like most modern writers, I define that term in an Aristotelian sense (Curzer 2012; Vaccarezza and Croce 2021): (a) a civic virtue is a stable, robust disposition of thought, action, and feeling, which (b) enables productive forms of political relationships and that is (c) animated by a conception of the public good (see Dagger 1997; Edyvane 2013). Within these general terms, there is scope for variety or disagreement: between liberal and conservative theorists about which dispositions should count as virtues and about the character of the political good (a rich discussion of these issues is McPherson 2022). Such complexities also attend to the neglected topic of civic or political vices (Button 2016).

8.2  Character and Corruption My account of cynicism begins with a familiar conviction: many political institutions are corrupting in the sense that engaging with or immersion in them is an active source of moral damage (Card 1996; Tessman 2005). Specifically, corrupting institutions cause damage to our moral character. Of course, there are other senses in which we might describe institutions as corrupting, but the character-centered is what concerns me. Critics often express the corrupting effects by using toxicological or environmental metaphors – we easily talk of institution environments being poisonous, polluted, or toxic. Such metaphors invite us to think and talk about those environments as containing things that are harmful if one is exposed to them for too long (Tirrell 2021). The concept of corruption in the character-centered sense was once central to moral and political philosophy in a way that it is no longer. Granted, it has not gone away, but a different meaning of corruption has become dominant (Philp 2015). Using a term with various senses is fine as long as the meaning is clear. In this section, I explain the character-centered sense of corruption and will start with its value. The first is its role in understanding the etiology of individual-level vices and character failings. If we worry about vice, then we should be interested in how we develop them – the origins and causes of our failings of character, or what José Medina calls their “sociogenesis” (Medina 2012: 30). These aetiologies are complicated – we explore psychological and developmental and interpersonal and structural factors that are empirically complex (Dillon 2012; Kidd 2022).

Institutional Cynicism and Civic Virtue 155 A second function of the concept of character corruption is to guide ameliorative projects. If we worry about vice, we want to do something about them, which might mean trying to reduce their incidence or severity or repairing the damage they cause. Quassim Cassam calls this “the project of vice-reduction” (Cassam 2018: 186). Sometimes, there might be nothing we can do about collective patterns of viciousness, but that must be a conclusion we arrive at, not a pessimistic conviction we start from (even if some of us are persuaded by more pessimistic, misanthropic stances on human life – see Cooper 2018 and Kidd 2021a). The third value of the concept of corruption is that it can help us think about the moral and epistemic agency. However defined, our ability and willingness to think, act, and feel well depends to a massive degree on the quality of the social environment. Granted, this shouldn’t come to eclipse individual agency: hyper-individualism is a problem just as much as hyper-structuralism (Dillon 2012: 89-90f). Tracing the dialectics of corruption is a way to seek something more sensible. I gave fuller accounts of my character-focused history of corruption in earlier work and will not rehearse all the details (see Kidd 2020a, 2021b, 2022). The core claim is the crucial role in the ongoing dynamics of individual character development of corrupting social conditions. By corrupting, I mean liable to cause moral damage in one or both of these ways: (a) the conditions tend to facilitate the extirpation or erosion of virtues and other excellences of character, and (b) the development and exercise of vices and other failings of character. Character corruption is a dynamic process – or a set of strategies – usually contemporaneous with various counter-corrupting factors. One might be subjected to all sorts of corrupting pressures and temptations. Still, one can also seek ways of pushing back – for instance, affirming positive values, altering one’s habits, seeking supportive or reparative interactions with virtuous exemplars, or seeking different emplacements within the corrupting conditions to shield oneself as much as possible from their damaging efforts. Doubtless, other strategies can be identified or even imagined. If we are sensitive to corrupting forces, we must develop the abilities that help us do that work (this is one connection to the later discussion of cynicism). Corruption analysts should also recognize that people are differentially susceptible to corrupting forces. We have different anxieties, aims, degrees of tolerance of moral tension, different structures of values, and different kinds of responses to temptations or threats. Some people seem unaffected by specific corrupting forces – unmoved by the allure of wealth, say, or insensitive to certain forms of social pressure. Much of this reflects profound differences in psychology and temperament. We should also reckon with the fact that the range of corruptors – the features of a social environment that have corrupting effects – are diverse. This often makes identifying them tricky, but there is often a consensus about the corruptors and their consequences. Corruptors may

156  Ian James Kidd include norms, practices, operational expectations, incentive systems, objectives, hierarchical structures, and institutional cultures (agonistic, competitive, etc.). All these and more could facilitate the erosion of virtue and the worsening of vice. Some corrupt our ideals. Some work to entrench bad habits. Some acclimatize us to morally dubious compromises. Some strain our values. Some distort the developmental trajectory of our character. Crucially for the later discussion of cynicism, corruptors can be accidental and unwanted features of our institutions. After all, most institutions are products of organic patterns of higgledy-piggledy development with all the roughness. No one sat down and designed the social world, even if some groups work hard to shape and reshape specific parts of it. Unfortunately, other corruptors are intentionally built into institutions to corrupt individuals in ways that render them exploitable. Consider cases of politicians who alter institutions to encourage invidious behaviors – weakening the standards of public life, disbanding watchdogs and systems of monitoring, and so on (see, e.g., Feldman and Eichenthal 2013). Corrupted people often try to corrupt systems to their advantage and corrupt others. For this reason, our criticism ought to be directed, in most cases, to corrupting conditions and not to corrupt individuals. The exceptions include (a) people who are plausibly complicit in their corruption – those who buy into corrupting pressures for the sake of their enhancement (call these Faustian cases) – and (b) people who work hard to create more corrupting conditions. But I leave these cases aside. I need to elaborate on the corruption mechanisms before moving on to institutional cynicism.

8.3  Corrupting Conditions Many of us experience political institutions as corrupting, and it is easy to see how this could lead to a kind of cynicism. It needs some spelling out, though. I am interested in how political institutions’ arrangements and operations can corrupt the characters of those embedded in them. When embedded in an institution, one is subjected to its values and goals; these can be internalized or, at the least, influence our habits and dispositions. Our sense of identity – our tolerance for moral tension, our ideals, our standards of everyday moral and epistemic conduct – could all be shaped by the internal realities of the institution. Over time, one can become inured to making the uncomfortable compromises often integral to institutional life. Therefore, being a part of the institution starts to become corrupting – one is subjected to moral damage that might begin to mark one’s character. In some cases, this awareness of potentially or actually corrupting effects can motivate someone to stay away from the institution – or to get out of it. Anecdotally, friends of mine have left governmental institutions because, sadly, they felt themselves becoming worse people – “It was doing bad things to me” was a common lament.

Institutional Cynicism and Civic Virtue 157 This is a specific conception of corruption, and I argue it can connect to a particular kind of cynicism. Other types of cynicism are available in political theory and postmodernism (cf. Sloterdijk 1983), not all of which relate to moral damage. At the same time, some political theorists do gesture to character-based kinds of corruption. Susan Rose-Ackerman and Bonnie Palifka, for instance, ask, “how the basic structure of the public and private sectors produces or suppresses corruption” (Rose-Ackerman and Palifka 2016: 37). One answer is that structures contain corruptors that tend to damage and distort the moral character of those embedded within the institutions. Other political theorists are skeptical about institutional cultures being morally corrupting. For instance, Jason Brennan and Peter Jaworski argue tenaciously against claims that the market is morally corrupting (cf. Brennan and Jaworski 2015: Part 3). I want to note two points of their discussion. First, we must provide evidence supporting the corrupting effects of political institutions, drawing on history, sociology, ethnography, and other disciplines (Chubb, Forstenzer, and Kidd 2021). Second, Brennan and Jaworski propose that we distinguish the claims that structures cause people to have worse characters and that they reveal the lousy character people had all along (Brennan and Jaworski 2015: 91f). I think this is too limited: there are other ways corruptors can have harmful effects on our moral character: other modes of corruption. I believe there are at least five, indicating a more comprehensive set of options than “causing” and “revealing.” 1 Acquisition – a corrupter leads to the acquisition of a new vice not previously present in a person’s character. 2 Activation – a corrupter activates a dormant vice, one present in someone’s character but not actively shaping their thoughts, actions, and feelings. 3 Intensification – a corrupter increases the strength of a vice. A once weak form of dogmatism becomes an intense kind of ultra-dogmatism. 4 Propagation – a corrupter increases the scope of vice and how it affects a person’s thinking, action, and feeling. A once-localized arrogance starts to spread, ever more widely, across more aspects of one’s conduct. 5 Stabilization – a corrupter increases the stability of vice, the extent to which it is resistant to disruption. These modes of corruption refer to ways that a person gets corrupted. With them in mind, we can also distinguish sets of sub-claims: a corruptor may, for instance, facilitate a vice or a whole cluster of vice, or one corruptor might activate a vice, and then other corruptors intensify and stabilize it, and so on. In such cases, it will become hard to sustain any clear distinction between causing and revealing certain vices.

158  Ian James Kidd Here are some of the general kinds of corruptors that we might look for in our institutional environments (see Kidd 2020a: §4.4): a b c d e f g h i j

Absence of exemplars of virtue Derogation of exemplars of virtue Presence of exemplars of vice Valorization of exemplars of vice Rebranding of vices-as-virtues Rebranding of virtues-as-vices Increasing the exercise costs of virtue Decreasing the scope for the exercise of virtue Increasing the rewards of vice Decreasing the penalties for vice

Consider examples of features of British political culture that critics have argued acted as corruptors. The first is certain cultures of secrecy. A certain amount of confidentiality is necessary for the running of a government, if not the running of our relationships. The Blair premiership, into its second and third terms, was increasingly accused of being secretive in more problematic ways. One critic pointed to a “combination of a genuine need for confidentiality, a siege mentality, and habitual caution.” Collectively these created a culture of secrecy that “tended to reinforce the walls of a closed world impervious both to diverse options and the consequences of its own actions” (Rhoades 2011: 287). Such cultures enabled Blair’s inner circle to conceal the considerations, evidence, and reasoning that informed their decisions – and this included their monumental decision to support the US-UK invasion of Iraq in the Second Gulf War. Unfortunately, that culture of secrecy served to insulate Blair’s government from critical debate and rival perspectives. Why would this be corrupting? Cultures of secrecy entrench the conditions that feed individual-level closed-mindedness and dogmatism (Battaly 2018). Those epistemic vices are scaffolded by social environments that structurally remove or marginalize alternative epistemic options. Secrecy also lets people conceal their epistemic procedures – which ones they used, how well they did them, what problems they ran into, and so on. Outside agents do not reliably know which options were consulted, what one thought of them, or how accurately or fairly we assessed them. In these ways and others, cultures of secrecy are corrupting. Of course, such cultures can be problematic for other reasons, too. They may contribute to institutional opacity: the tendency of social institutions to become resistant to understanding and epistemic appraisal (Carel and Kidd 2022). A second famous feature of the Blair premiership was its increasing reliance on “spin,” the artful manipulation of political information to present governments in a positive light. Governments dominated by spin

Institutional Cynicism and Civic Virtue 159 often acquire an ethos of performative superficiality – where a sharp distinction emerges between looking good and doing good. The priority is given to the appearance rather than the actual performance. For Jill Johnson, Labour’s head of communications until January 1996, the party’s reliance on spin doctors was feeding “a view that politics is and should be run secretively from the centre” and was “slowly dripping poison into the body politic” (in Foley and Foley 2000: 197). In May 1999, Johnson nicely criticized New Labour’s alarming “fusion of policy with spin”: After two years, the outstanding feature of this government is that its overwhelming desire to appease Middle England has led it to conceive policies not so much to solve a problem as to conjure an appearance. As image-making becomes progressively blurred with policy development, politicians move from objective reality to a virtual world of their creation. (in Foley and Foley 2000: 197) One legacy of this privileging is the phenomenon of bullshit, and “posttruth” politics and the entrenchment into the political life of a vice Quassim Cassam named epistemic insouciance (Cassam 2018: ch. 4). Institutional ethoi of performative superficiality tend to corrupt in at least two ways. First, privileging positive appearance over actual positive performance will encourage us to assign less weight to procedural virtues. These are dispositions necessary that enable us to perform repeatable practices correctly. Examples include diligence, carefulness, meticulousness, and thoroughness. Unexciting as they sound, procedural virtues are vital to the everyday practical operations of life, including running big political institutions like the civil service. Of course, exercising these procedural virtues requires us to resist the temptations to cut corners, rush, skim over the fine print, and skip the other repetitive and tedious aspects of life. But what drives us to exercise procedural virtues is a sense that it does matter that we do well. Conscientiousness matters. This sense is apt to be eroded by institutions whose ethos is one of performative superficiality. Institutional life often involves tasks and activities that tend to be repetitive, unglamorous, and unexciting – but utterly essential. A good institution should encourage virtuously procedural conduct, requiring an ethos of procedural conscientiousness. Second, a person can internalize the attitudes and values of an ethos of performative superficiality. They will start to refocus their energy on effectively presenting as doing good rather than doing well. This can encourage them to develop and exercise the vices of manipulation – ­dishonesty, deceit, and manipulativeness in various forms. I think this happens for two reasons: one, the performatively superficial focus less on actually doing well, so they will tend to start doing poorly, and – as a

160  Ian James Kidd further consequence – they will need to begin concealing their deficiencies and their cause. The manipulative person will therefore find themselves starting to rely increasingly on patterns of manipulativeness to keep up appearances. They will “spin,” lie, bullshit, deceive, divulge good times to “bury bad news,” omit relevant context when giving news, constantly move the goalposts to manufacture appearances of success, cunningly use implicature, and so on. The Blair premiership was accused of this (Barnet and Gaber 2001; Jeffries and Walker 2017: ch. 3). Against a thin defense that all governments do this, two writers argue that “spin … did not start in the Blair years, but it became more pronounced and widely used at this time” until it “evolved a sense in its own right” (Jeffries and Walker 2017: 41). I am sure that cultures of secrecy and institutional ethoi of performative superficiality are corrupting in other ways. The cultures and ethos could corrupt other vices, or the vices they encourage could, in turn, facilitate the exercise and development of other vices in appalling vice-­ cascades. This would have to be determined by careful empirical study, not a priori. We should also investigate how specific psychological profiles of political actors interact with these cultures and ethoi. I suspect the dispositionally arrogant are more vulnerable to corruption by cultures of secrecy and ethoi of performative superficiality. Arrogant people feel less need to be accountable to others and more entitled to exempt themselves from norms incumbent on members of social communities (see Roberts and Woods 2007: 244f; Tanesini 2021: §5.2). An expectation that one will conform to expected standards is rejected – proper accountability and performative conscientiousness are for the little people. I hope these examples are sufficient to indicate what the corruptors built into political institutions can look like. This account of institutional corruption should also resonate with many readers’ perceptions of the harmful effects institutions can have on those who work within them. Moreover, the account points to a further reason that corrupting political institutions are bad: they damage public trust in those institutions. If so, we are well on the way to some kind of cynicism.

8.4  Institutional Cynicism Imagine someone who wants to be politically engaged but perceives the relevant institutions as liable to corrupt their moral character. They might suppose that their engagements will require them to cultivate particular new virtues or dispositions, ones apt to help them navigate an environment full of potent corruptors. Some virtues regulate our experience and ways of engaging with our environments. These could be a dedicated set of virtues – Emily Sullivan and Mark Alfano identify classes of virtues aimed at monitoring, adjusting, and restricting social-­epistemic networks (Sullivan and Alfano 2020: §8.4). Or particular virtues could

Institutional Cynicism and Civic Virtue 161 have structural monitoring among their functions if developed in the right ways. José Medina argues that humility, curiosity, and open-­ mindedness can function to help members of oppressed groups cope in hostile social environments (Medina 2012: 30–48). Or the virtues may be collective virtues – ones possessed by collectives of people rather than individual agents – which may include a form of virtuous solidarity (cf. Battaly 2022; Byerly and Byerly 2016). I suggest that a certain kind of cynicism should be among the virtues salient to those engaged with corrupting political institutions. This sets me against those who regard cynicism as antipolitical. Politicians often condemn cynical attitudes and outlooks. Obama contrasted a “politics of hope” with a “politics of cynicism.” London Mayor Sadiq Khan once contrasted “optimism and hard work” with “cynicism, lethargy, and fatalism.” The Democratic politician, Alexandra Ocasio-Cortez, says “the biggest hurdle … our communities have is cynicism.” Those on the political right make similar claims. Sir Roger Scruton condemns kinds of cynicism – fostered, he says, by the ruination of “old institutions” and traditional certainties – which he thinks have left us “empowered, but without a destination” (Scruton 2019: 134f). Of course, different kinds of cynicism are being invoked, and these writers do not always explain what they mean by the term. I take it that cynicism in its most general sense has two components: (a) a sense that the professed motivations of a person or institution are different from their operative ones – the ones directing their behavior – and (b) a conviction that the operative ones, if identified, are likely to be morally inferior. “Morally inferior” need not mean evil or invidious, just lesser in moral quality. Think of the signs in hotel bathrooms that invite guests to reuse their towels – the professed motivation is ecological sustainability, but the operative motivation is to reduce laundry costs. Of course, wanting to reduce costs is not a morally bad motivation, but it is lesser than environmental concern. Indeed, many ecological ethicists and activists accused the practice of “greenwashing” of feeding cynicism (Bowen 2014). The conjunction of (a) and (b) generates the epistemic and affective aspects characteristic of cynicism. These include policies of suspiciousness, distrust, and a striving to identify the “true” motivations behind an action. This is at the core of cynicism, and its various kinds are built upon it. It differs from the account of cynicism offered by Samantha Vice: Cynicism is a b c d e

a stance of disengagement, characterized by distrust, contempt, skepticism (to differing degrees), adopted toward humans, their institutions, and values, adopted as a response to a belief that humans are motivated only by self-interest, or more generally, that human beings are of little worth.

162  Ian James Kidd We can question these components and their connection to one another. The cynic need not disengage entirely from the world and, in many cases, may be unable to, even if they want to. The cynic may be forced into complex and perhaps awkward patterns of engagement and disengagement or be compelled to maintain existing arrangements but in new and revised ways – more watchful, less trusting, marked by new attitudes of vigilance or alertness. Indeed, continuing one’s engagement with the institution could help to keep cynical dispositions sharp and well-trained. Distrust, contempt, and skepticism might also be contingent features of specific ways of being cynical. I’m pretty cynical, but as far as I can tell am not contemptuous of the people and institutions of which I am cynical (on cynicism and contempt, cf. Bell 2013: §6.6). We should also resist building into conceptions of cynicism contentious axiological claims about the worth of human beings, or controversial anthropological claims. We ought to distinguish three kinds of cynicism, determined by the object of cynical attitudes and the sorts of behavior expressive of them. First, agential cynicism is directed at individuals. We suspect or expect or suppose them to operate with different and lesser operative motivations and values than the ones they publicly profess (e.g., Citrin 1974). I may be cynical about colleagues if they constantly talk about their commitment to collegiality rather than shunt arduous duties onto the precariously employed junior staff. Of course, cynical perceptions ought to be well-supported by evidence, and we should also guard against the risk of our perceptions and judgments being distorted by prejudices. The second is institutional cynicism, directed at a specific political institution’s practices, norms, and values. This is familiar from everyday life: many of us feel cynical toward the government, businesses, the police, and so on. In most cases, our cynicism is directed toward those with the power to shape an institution’s ethos and practical life – a university Vice-Chancellor or executives of a Fortune 500 company. This is consistent with a cynicism directed at the institution itself. Institutions have values or goals manifested in collective practices; these reliably bring about specific results that serve the interests of some groups at the expense of other attractions and groups (Fricker 2020). Institutional structures and ethoi also survive radical personnel changes. Vice-Chancellors come and go, but the university’s working features persist. The third kind of cynicism is anthropological cynicism. It involves some more or less systematic account of our essential nature or essence as human beings. A conviction that we are essentially selfish creatures, “all the way down,” that human nature precludes genuine altruism is a classic sort of anthropological cynicism, of a kind often attributed (rightly or wrongly) to Machiavelli and Hobbes (Agger et al. 1961). I’m skeptical about anthropological cynicism: conceptions of human nature can be projective rather than descriptive; the appeals to the sciences are often questionable and mangled at worst. One attraction of anthropological

Institutional Cynicism and Civic Virtue 163 cynicism is that it licenses the “bad faith” so well described by JeanPaul Sartre. It is more accurate to judge that human nature is deeply “dappled,” shot through with streaks of selfishness, selflessness, and complicated capacities for virtue and vice. Myopic fixation on one set of aspects – darker, selfish, violent – is a failure to be avoided, not a clearsighted revelation of our true nature (Dupré 2001; Midgley 2002). I’m not concerned with exploring agential or anthropological cynicism in further detail. I will only mention that their relationships to institutional cynicism – my focus in the rest of this chapter – are complicated. Some cynics roll together agential and structural cynicism: specific agents work hard to corrupt certain institutions, often producing unscrupulous individuals. We probably cannot sharply distinguish agential and institutional cynicism. Moreover, a deep sort of anthropological cynicism can be the foundation for agential and institutional cynicism. If one supposes that human beings are, by nature, essentially radically selfish, it makes sense to presume that individuals and the institutions they create will reflect that selfishness at some deep level. Such deep cynicism is attractive but not my concern here. I suggest that a certain kind of institutional cynicism can encourage critical epistemic attitudes toward political institutions. I may direct cynical attitudes toward political institutions’ practices, arrangements, strategies, and goals. When an institution declares it is doing A for the sake of reason B, I wonder what its operative values are (Why do businesses start to put up rainbows during Pride month? Why do big cosmetics companies support breast cancer initiatives? Why are these universities suddenly getting into food security?) Of course, many things can feed institutional cynicism – commercial motivations, desires to enhance reputation by playing into a trending issue, the conclusions of focus groups attempts to attract new audiences and markets. The list includes many factors from the broader literature on corporate and political corruption. I want to connect institutional cynicism to the corrupting effects of institutions in the final section.

8.5  Cynicism and Corruption A key motivator for institutional cynicism is the recognition that an institution is corrupting: its internal environment contains corruptors apt to cause us moral damage. If so, one recognizes the institution likely has many corrupted agents. Of course, people who are already, to some degree, morally corrupt might also be attracted to those institutions (Lord Acton’s remark that “power tends to corrupt” has been amended to “power tends to attract the corruptible”). One is also likely to realize the risk to oneself of engaging with the institution. “If you step into the fire,” explained a friend who works in the national government, “you’re going to get burned.” Institutional cynicism is sustained by experience and can

164  Ian James Kidd be explained and justified. Cynics of this sort could easily describe specific corruptors and point out cases of severe moral damage done to people (including themselves) while also elaborating on the mechanisms of corruption. Indeed, those who work within institutions naturally grumble about them, and doing so can serve essential epistemic-evaluative functions (see Norlock 2018). Here we should distinguish two problems with morally corrupting institutions. The first is their tendencies to cause moral damage to those embedded within them – and maybe also to those who must engage with them in other ways. The crucial second point is that corrupting ­institutions – and the corrupted people they attract and create – will erode public trust in the rules and standards of political life. The damage to individual political actors is one harm; another is the broader damage done to public confidence in the political system (Uslaner 2015). At this point, critics of cynicism worry that adding cynicism into the mix worsens things: the public sees corrupted politicians in a corrupting political system, feels the first wave of alienation and distrust, then becomes cynical in ways that intensify all this. I agree this destructive cynicism exists, but I also think there are positive kinds of cynicism. Institutional cynicism is a case in point, even if, for sure, it can deteriorate into kinds of dogmatism. Cynicism should be seen as a risky quality – a bitter pill that we must take only in careful measures. Institutional cynicism can, if unregulated, become dogmatic and jaundiced and entrap us within closed outlooks that feed fatalism, if not despair. In other cases, cynicism can take self-serving forms, “a kind of cunning complicity in the status quo” (Sennett 1992: 16). For these reasons, institutional cynicism cannot be, by itself, a civic virtue. It can be part of a broader civic stance if accompanied by other attitudes, values, and commitments. These unaccompanied cynicisms are rightly criticized by Arnett and Arneson (1999) and Kanter and Mirvis (1989). Speaking chemically, cynicism is a volatile quality but with valuable properties if mixed in proper measures with other components. Institutional cynicism of the sort I have sketched differs in at least three ways from the broader kind of cynicism described by Vice. To start with, the institutional cynic is not confined to types of disengagement – their practical life includes much more discerning kinds of disengagement and engagement. In some cases, one might be unable to disengage from an institution (quitting one’s job isn’t an option for everyone). A cynic may remain engaged with certain aspects of their institution but adroitly avoid other kinds of engagement – perhaps ones they suspect would be profoundly corrupting. Or the cynic may decide to engage in new ways with their institution: maybe a policy of acting in ways that generate kinds of what José Medina calls epistemic friction – actions that “disrupt” epistemic habits in ways that are “disruptive and reenergizing” (Medina 2012: 224). This sort of practice, of course, has venerable

Institutional Cynicism and Civic Virtue 165 precedent in those provocative philosophical acts and rhetoric of the classical Cynics (Desmond 2014; Navia 1998). The “man in the tub” Diogenes of Sinope attacked complacency and hypocrisy to try and “replac[e] false values with those which would … enable human beings to fulfill their true nature” (Hard 2012: ix). A second difference is that, for my institutional cynic, the epistemic and affective attitudes are aimed at the institution. Such attitudes can include cautiousness, distrust, reticence, suspiciousness, watchfulness, and, doubtless for some, hate and contempt (cf. Eisinger 2000). The object, though, is institutional norms, practices, and structures. I distrust my university’s many claims to support equality and diversity (many of its working standards and expectations say otherwise). I doubt when oil and gas companies talk of their investment in green energy and suspect they are greenwashing. But this does not automatically extend to the individuals: that would be some further development of my cynical stance. Furthermore, institutional cynicism need not rely on claims about human nature, about which one can be agnostic. Insisting that the appraisal of collective human life must “go anthropological” is problematic. If we push down into our underlying natures, we risk occluding institutional structures, which tend to serve the interests of those on the winning side of those structures (cf. Kidd 2020b). The third feature of institutional cynicism is that it is consistent with what we might call philanthropic sentiment. Vice, recall, suggested that a cynic thinks human beings are motivationally dominated by self-interest and, also, of little worth: The cynical stance prevents the cynic from interpreting people in any way other than self-interested or corrupt, and seeing people like this is to see them in ways unmoved by faith, hope, and charity […] At least part of the harmfulness of cynicism is that the cynical attitude is destructive of the goods constitutive of a flourishing human life and undermines moral experience and progression. The conclusion must be that we cannot value cynicism if we value morality and meaningful life with others. (Vice 2011: 178–179) An institutional cynic, however, needn’t see people as ruled by self-­ interest. Or, rather, the cynic may accept that many, if not most people are, some or most of the time, driven by various corrupted motivations. But this is because their social environment has corrupted them. If one removed those corrupting pressures, we might reveal a much richer moral psychology. Human beings are of great moral worth, which is why we should lament the moral damage done to them by corrupting environments. Indeed, because we are of value, normatively loaded talk of “corruption” or “moral damage” is intelligible to us. In his book Everybody

166  Ian James Kidd Knows, William Chaloupka says that cynicism is a condition of “lost belief” – of lost trust or confidence in the ideals and aspirations constitutive of civic life (Chaloupka 1999: xiv). Cynicism may be the loss of a belief, of course, or it might, instead, be something subtler – like the loss of confidence in the hospitability of our social world to those ideals or a loss of certain styles of believing (a cynic might continue to believe in the American Dream or ideals of equality and liberty for all, but find that belief is now strained or tested, rather than sanguine and secure, pushing what Deneen (2009) calls their “democratic faith”). Moreover, an institutional cynic should rue that entrenched corruptors cause moral damage and are destructive of the goods of human life. That explains why this kind of cynic wants to cause epistemic friction and try to change things if and where they can. Certain cynics might submit to fatalistic surrender, but that is not an automatic effect of institutional cynicism. Helen Small sees cynicism as a basis for a “tactical” approach to institutions (Small 2020: 154f). At one point, Vice offers this worry: We cannot value cynicism if we value humanity’s ongoing, small steps towards progress. I have also argued that cynicism is inimical to other goods – relationships, trust, a delight in others and the world. (Vice 2011: 179) An institutional cynic will reply that we can never make moral progress without their cynicism about political institutions. We cannot create or sustain human goods without awareness of how they are rendered fragile by the corrupting realities of the world. The cynicism of the sort described here reveals that fragility but does not revel in it, nor does it entail a pessimistic sense that we could ever do nothing about it. Cynicism reveals uncomfortable realities but, in that sense, also exposes them as uncomfortable – thus, as something that one might want to change, if possible. Institutional cynicism is not necessarily anti-democratic, even if it helps erode particular untenable or naïve faith in democratic institutions. An institutional cynic has a more critically lucid, skeptical, and savvy stance which can help us better and more engaged citizens. My cynic does not lack faith but has shaken off naïve belief and is thus more likely to have a better sense of the status of the democratic institutions of their society (and I leave aside the interesting question of how institutional cynicism might function or be needed in other kinds of the political ­system – see Steinmüller and Brandtstädter 2016). I suggest that institutional cynicism should be seen as an essential component of a civic stance. It serves crucial, if regrettable, functions in a world whose institutions are full of corruptors. However, it must be accompanied by attitudes, commitments, and values that help prevent it from developing into the sort of caustic, corroding cynicism that worries

Institutional Cynicism and Civic Virtue 167 critics. Describing that civic stance and identifying its other components would be formidable. Still, we will only take up one if we accept that certain kinds of cynicism can play valuable roles in political life. If some commentators are correct, then the future of democratic societies will depend on our ability to operationalize certain kinds of cynicism.

Acknowledgments I am grateful to the editors for their invitation to contribute to the volume and for their generous and constructive comments. I also learned much from discussions with Taylor Matthews and an audience at the University of Bristol.

References Agger, Robert E., Marshall N. Goldstein, and Stanley A. Pearl (1961) ‘Political Cynicism: Measurement and Meaning,’ The Journal of Politics 23.3: 477–506. Arnett, Ronald C. and Pat Arneson (1999) Dialogic Civility in a Cynical Age: Community, Hope, and Interpersonal Relationships (Albany: State University of New York Press). Barnet, S. and I. Gaber (2001) Westminster Tales: The Twenty-First Century Crisis in Political Journalism (London: Continuum). Battaly, Heather (2018) ‘Closed-mindedness and Dogmatism,’ Episteme  15.3: 261–282. Battaly, Heather (2022) ‘Solidarity: Virtue or Vice?’ Mark Alfano, Colin Klein, Jeroen de Ridder (eds.),  Social Virtue Epistemology  (New York: Routledge), 303–324. Bell, Malacaster (2013) Hard Feelings:  The Moral Psychology of Contempt (Oxford: Oxford University Press). Bowen, Frances (2014) After Greenwashing: Symbolic Corporate Environmentalism and Society (Cambridge: Cambridge University Press). Brennan, Jason F. and Peter Jaworski (2015) Markets without Limits: Moral Virtues and Commercial Interests (London: Routledge). Button, Mark E. (2016) Political Vices (Oxford: Oxford University Press). Byerly, Ryan T., and Meghan Byerly (2016) ‘Collective Virtue,’ Journal of Value Inquiry 50.1: 33–50. Card, Claudia (1996) The Unnatural Lottery: Character and Moral Luck (Philadelphia: Temple University Press). Carel, Havi and Ian James Kidd (2022) ‘Institutional Opacity, Epistemic Vulnerability, and Institutional Testimonial Justice,’ International Journal of Philosophical Studies 29.4: 473–496. Cassam, Quassim (2018) Vices of the Mind: From the Intellectual to the Political (Oxford: Oxford University Press). Chaloupka, William (1999) Everybody Knows: Cynicism in America (Minneapolis: University of Minnesota Press). Chubb, Jennifer, Joshua Forstenzer, and Ian James Kidd (2021) ‘Epistemic Corruption and the Research Excellence Framework,’ Theory and Research in Education 19.2: 148–167.

168  Ian James Kidd Citrin, Jack (1974) ‘Comment: The Political Relevance of Trust in Government,’ American Political Science Review 68.3: 973–988. Cooper, David E. (2018) Animals and Misanthropy (London: Routledge). Curzer, Howard (2012) Aristotle and the Virtues (Oxford: Oxford University Press). Dagger, Richard (1997) Civic Virtues, Rights, Citizenship, and Republication Liberalism (Oxford: Oxford University Press). Deneen, Patrick (2009) Democratic Faith (Princeton: Princeton Press). Desmond, William (2014) Cynicism (London: Routledge). Dillon, Robin (2012) ‘Critical Character Theory: Towards a Feminist Theory of ‘Vice’’, Sharon Crasnow and Anita Superson (eds.), Out from the Shadows: Analytical Feminist Contributions to Traditional Philosophy (New York: Oxford University Press), 83–114. Dupré, John (2001) Human Nature and the Limits of Science (Oxford: Oxford University Press). Edyvane, Derek (2013) Civic Virtue and the Sovereignty of Evil (London: Routledge). Eisinger, Robert M. (2000) ‘Questioning Cynicism,’ Society 37: 55–60. Feldman, Daniel L. and David R. Eichenthal (2013) The Art of the Watchdog: Fighting Fraud, Waste, Abuse, and Corruption in Government (Albany, NY: SUNY Albany Press). Foley, Michael Stewart and Frances Lesley Foley (2000) The British Presidency (Manchester: Manchester University Press). Fricker, Miranda (2020) ‘Institutional Epistemic Vice: The Case of Inferential Inertia,’ Ian James Kidd, Heather Battaly, and Quassim Cassam (eds.), Vice Epistemology (New York: Routledge), 89–107. Goldfarb, Jeffrey C. (1991) The Cynical Society: The Culture of Politics and the Politics of Culture in American Life (Chicago: University of Chicago Press). Hard, Robin (2012) Diogenes the Cynic: Sayings and Anecdotes with Other Popular Moralists (Oxford: Oxford University Press). Jeffries, Lesley and Brian Walker (2017) Keywords in the Press: The New Labour Years (London: Bloomsbury). Kanter, Donald L. and Philip H. Mirvis (1989) The Cynical Americans: Living and Working in an Age of Discontent and Disillusion (San Francisco: Jossey-Bass). Kidd, Ian James (2020a) ‘Epistemic Corruption and Social Oppression,’ Ian James Kidd, Heather Battaly, and Quassim Cassam (eds.), Vice Epistemology: Theory and Practice, (New York: Routledge), 69–87. Kidd, Ian James (2020b) ‘Humankind, Human Nature, and Misanthropy,’ Metascience 29: 505–508. Kidd, Ian James (2021a) ‘Varieties of Philosophical Misanthropy,’ Journal of Philosophical Research 46: 27–44. Kidd, Ian James (2021b) ‘Epistemic Corruption and Political Institutions,’ Michael Hannon and Jeroen de Ridder (eds.), The Routledge Handbook to Political Epistemology (New York: Routledge), 347–358. Kidd, Ian James (2022) ‘From Vice Epistemology to Critical Character Epistemology,’ Mark Alfano, Colin Klein, and Jeroen de Ridder (eds.), Social Virtue Epistemology (New York: Routledge), 84–102. Mazella, David (2007) The Making of Modern Cynicism (Charlottesville, VA: University of Virginia Press). McPherson, David (2022) The Virtues of Limits (Oxford: Oxford University Press).

Institutional Cynicism and Civic Virtue 169 Medina, Jose (2012) The Epistemology of Resistance: Gender and Racial Oppression, Epistemic Injustice, and Resistant Imaginations (Oxford: Oxford University Press). Midgley, Mary (2002) Beast and Man: The Roots of Human Nature (London: Routledge). Navia, Luis E. (1998) Diogenes of Sinope: The Man in the Tub (Westport, CT: Greenwood Press). Norlock, Kathryn (2018) ‘Can’t Complain,’ Journal of Moral Philosophy 15.2: 117–135. Philp, Mark (2015) ‘The Definition of Political Corruption,’ Paul M. Heywood (ed.), The Routledge Handbook to Political Corruption (New York: Routledge), 17–29. Rhoades, R.A.W. (2011) Everyday Life in British Government (Oxford: Oxford University Press). Roberts, Robert C. and W. Jay Woods (2007) Intellectual Virtues: An Essay in Regulative Epistemology (Oxford: Oxford University Press). Rose-Ackerman, Susan and Bonnie J. Palifka (2016) Corruption and Government: Causes, Consequences, and Reform (Cambridge: Cambridge University Press). Scruton, Roger (2019) A Political Philosophy: Arguments for Conservatism (London: Bloomsbury). Sennett, Richard (1992) ‘The Decline of Public Discourse,’ Contemporary Sociology 21.2: 15–17. Shea, Louisa (2010) The Cynic Enlightenment: Diogenes in the Salon (Baltimore, MD: John Hopkins University Press). Sloterdijk, Peter (1983) Critique of Cynical Reason, trans. M. Eldred (Minneapolis: University of Minnesota Press). Small, Helen (2020) The Function of Cynicism at the Present Time (Oxford: Oxford University Press). Snow, Nancy (2018) ‘Hope as a Democratic Civic Virtue,’ Metaphilosophy 49.3: 407–427. Stanley, Sharon A. (2012) The French Enlightenment and the Emergence of Modern Cynicism (Cambridge: Cambridge University Press). Steinmüller, Hans and Susanne Brandtstädter (2016) Irony, Cynicism and the Chinese State (London: Routledge). Sullivan, Emily and Mark Alfano (2020) ‘Vectors of Epistemic Insecurity,’ Ian James Kidd, Heather Battaly, and Quassim Cassam (eds.), Vice Epistemology: Theory and Practice, (New York: Routledge), 148–164. Tanesini, Alessandra (2021) The Mismeasure of the Self: A Study in Vice Epistemology (Oxford: Oxford University Press). Tessman, Lisa (2005) Burdened Virtues: Virtue Ethics for Liberatory Struggles (Oxford: Oxford University Press). Tirrell, Lynne (2021) ‘Discursive Epidemiology: Two Models,’ Aristotelian Society Supplementary 95.1: 115–142. Uslaner, Eric M. (2015) ‘The Consequences of Corruption,’ Paul M. Heywood (ed.), The Routledge Handbook to Political Corruption (New York: Routledge), 199–211. Vaccarezza, Maria Silvia and Michel Croce (2021) ‘Civility in the Post-truth Age: An Aristotelian Account,’ Humana.Mente 39: 127–150. Vice, Samantha (2011) ‘Cynicism and Morality,’ Ethical Theory and Moral Practice 14: 169–184. Williams, Bernard (2002) Truth and Truthfulness: An Essay in Genealogy (Princeton: Princeton University Press).

9

Myside Bias in Individuals and Institutions Keith E. Stanovich

9.1 Introduction Belief polarization has numerous diverse causes and can be approached scientifically from many distinct levels of analysis. As I approach the issue from the perspective of a cognitive psychologist, this chapter will focus on myside bias, the primary psychological contributor to our society’s failure to achieve belief convergence on numerous critical issues. Myside bias occurs when people evaluate evidence, generate evidence, and test hypotheses in a manner biased toward their own prior beliefs, opinions, and attitudes. Myside bias occurs in a wide variety of judgment domains. People in all demographic groups exhibit it, and it is displayed even by expert reasoners, the highly educated, and the highly intelligent. It has been demonstrated in studies across a variety of disciplines, including cognitive psychology (Edwards & Smith 1996; Toplak & Stanovich 2003), social psychology (Ditto et al. 2019a), political science (Taber & Lodge 2006), behavioral economics (Babcock et al. 1995), legal studies (Kahan, Hoffman, et al. 2012), cognitive neuroscience (Westen et al. 2006), and in the informal reasoning literature (Kuhn & Modrek 2018). Myside bias has been found to surface at every stage of information processing. That is, studies have shown a tendency toward a biased pursuit of evidence, biased evaluation of data, biased assimilation of evidence, biased memory of outcomes, and biased evidence generation (Bolsen & Palm 2020; Clark et al. 2019; Ditto et al. 2019a; Epley & Gilovich 2016; Hart et al. 2009; Mercier & Sperber 2017). That is the good news – we know a lot about myside bias. The bad news is that it is one of the strangest cognitive biases, unlike the dozens of others examined. For the first several decades of work in the heuristics and biases tradition (Kahneman & Tversky 1973; Tversky & Kahneman 1974), from the 1970s to the 1990s, myside bias (often termed confirmation bias, see Mercier 2017) was treated as merely another on a growing list of biases (anchoring bias, hindsight bias, availability bias, and many others), and researchers assumed that it would act like other biases: DOI: 10.4324/9781003311003-12

Myside Bias in Individuals and Institutions 171 that it would be correlated with the same individual difference variables, it would show the same degree of domain generality, and its status as a non-normative response tendency would be equally secure. None of these expectations have been realized. In this chapter, I will argue that the strange properties of myside bias have implications for understanding its role in creating belief polarization.

9.2  An Empirically Strange Bias When Richard West and I began studying individual differences in cognitive biases in the 1990s, one of the first consistent results from our early studies was that the biases tended to correlate with each other (Sá et al. 1999; Stanovich & West 1997, 1998a, 1998b, 2000). The correlations were usually relatively modest, but, then again, they derived from tasks measured with just a few items and hence of relatively low reliability. Another consistent observation in our earliest research was that almost every cognitive bias was correlated with intelligence as measured with various cognitive ability indicators. Individual differences in most cognitive biases were also predicted by several well-studied thinking dispositions, such as actively open-minded thinking and the need for cognition. These early indications that the tendency to override various cognitive biases correlated with individual differences in cognitive ability and thinking dispositions have been replicated in other research studies (see Stanovich 2021 for citations). This finding has held for some of the most well-studied biases in the Kahneman and Tversky tradition (Kahneman 2011; Tversky & Kahneman 1974): anchoring biases, framing biases, hindsight bias, overconfidence bias, outcome bias, conjunction fallacies, representativeness errors, the gambler’s fallacy, probability matching, base-rate neglect, sample-size neglect, ratio bias, covariation detection errors, pseudo-diagnostic effects, and many others. There is no doubt that, based on prior work, the evident expectation is that any new cognitive bias studied will show the same correlations with individual difference variables. However, it turns out that myside bias is not predictable from standard cognitive and behavioral functioning measures. The degree of myside bias is not correlated with intelligence (Klaczynski 1997; Klaczynski & Lavallee 2005; Klaczynski & Robinson 2000; MacPherson & Stanovich 2007; Perkins et al. 1991; Sanchez & Dunning 2021; Stanovich & West 2007, 2008; Toplak & Stanovich 2003). General intelligence’s failure to attenuate myside bias extends to variables highly related to intelligence, such as numeracy, scientific thinking, reflectivity, and general knowledge (Drummond & Fischhoff 2019; Kahan 2013; Kahan, Peters, et al. 2012; Kahan et al. 2017; Van Boven et al. 2019). Converging with these results is the literature in political science showing that various indicators of cognitive sophistication, such as educational level, knowledge level, and political awareness, do not attenuate partisan

172  Keith E. Stanovich bias and can often increase it. For example, Joslyn and Haider-Markel (2014) found that highly educated partisans disagreed more about policy-­ relevant facts than less educated partisans. Jones (2019) found that political perceptions about policy-relevant conditions, such as the state of the economy, were more polarized among the more informed and politically aware partisans. Numerous measures of cognitive sophistication show that cognitive elites display more polarization on various political issues (Drummond & Fischhoff 2017; Ehret et al. 2017; Hamilton 2011; Henry & Napier 2017; Kahan & Stanovich 2016; Kraft et al. 2015; Lupia et al. 2007; Sarathchandra et al. 2018; Yudkin et al. 2019). In summary, well-controlled laboratory studies of myside bias converge with survey research and polling data showing that intelligence and education do not inoculate against myside tendencies. As Ditto et al. (2019b) note, “What if bias is not the sole province of the simpleminded?….A growing body of research suggests that greater cognitive sophistication and expertise often predicts greater levels of political bias, not less….Cognitive sophistication may allow people to more skillfully argue for their preferred conclusions, thus improving their ability to convince ­others – and themselves – that their beliefs are correct” (2019b: 312). From a perspective on individual differences, myside bias displays other curious tendencies. Other biases in the literature display correlations with intelligence and thinking dispositions related to rational thinking, such as actively open-minded thinking and the need for cognition (see Stanovich 2021 for citations). Despite these consistent findings involving almost every other cognitive bias, myside bias has failed to correlate with relevant thinking dispositions in the same manner that it has failed to correlate with intelligence (Clements & Munro 2021; Eichmeier & Stenhouse 2019; Guay & Johnston 2021; Kahan 2013; Kahan & Corbin 2016; Kahan et al. 2017; Macpherson & Stanovich 2007; Stanovich & West 2007; Stenhouse et al. 2018; Toplak & Stanovich 2003). Even personality dispositions that seem most directly related to evading myside bias fail to attenuate it. For e­ xample, Simas et al. (2020) suggested that empathy (or lack thereof) would be a pivotal mechanism in developing political polarization, partisan bias, and ideological conflict. However, two studies found that the differences in empathic concern did not predict the degree of partisan bias in evaluating a contentious public event. High empathic concern did not attenuate the degree of affective polarization among partisans. Simas et al. (2020) explain their findings by positing that empathy is biased toward one’s ingroup and thus does not provide protection against myside bias. A reasonably extensive literature has emerged on whether there are ideological and partisan differences in myside bias. These results have converged with the literature failing to find predictable differences in the degree of myside bias. Ditto et al. (2019a) meta-analyzed forty-one experimental studies of partisan differences in myside bias that involved over

Myside Bias in Individuals and Institutions 173 12,000 subjects. After amalgamating all of these studies and comparing an overall metric of myside bias, Ditto and colleagues concluded that partisan bias in these studies was quite similar for liberals and conservatives (see Sanchez & Dunning 2021; Washburn & Skitka 2018). Another way myside bias is an outlier bias is that, in most cases, it shows very little domain generality and appears very content-dependent. Individuals who display high myside bias on one issue do not necessarily show high myside bias on another unrelated issue (Tetlock 1986; Toner et al. 2013; Toplak & Stanovich 2003). These results are unlike other biases, such as framing effects, where other investigators and we obtain reliabilities in the range of .60–.70 across a dozen or so different items (Bruine de Bruin et al. 2007; Stanovich et al. 2016). In the literature, most biases have a substantial degree of domain generality (Stanovich 2021) but not myside bias. Individual difference variables do not predict the degree of exhibited myside bias, but one variable that does is the strength of the subject’s opinion on that specific issue (Gugerty et al. 2021 Stanovich & West 2008), a finding that has been reported in many studies (Bolsen & Palm 2020; Druckman 2012; Edwards & Smith 1996; Houston & Fazio 1989; Taber & Lodge 2006). In short, the level of myside bias displayed on a particular issue in a specific paradigm is highly content-dependent.

9.3  Normative Complications Myside bias is an outlier bias in another essential way. It is easy to show that they lead to suboptimal decisions for most of the other biases in the literature (anchoring biases, framing effects, base-rate neglect, and many others). In contrast, despite all the damage that myside bias does to our social and political discourse, it is shockingly challenging to show that, for an individual, it is a thinking error. In determining what to believe, myside bias operates by weighting new evidence more highly when it is consistent with prior beliefs and less highly when it contradicts a previous conviction. This tendency seems wrong, but it is not. Many formal analyses and arguments in the philosophy of science have shown that in most situations that resemble real life, it is rational to use your prior belief to evaluate new evidence (Alloy & Tabachnik 1984; Evans et al. 1993; Kornblith 1993). It is even rational for scientists to do this in the research process (Koehler 1993; Tappin et al. 2020). It is rational because people (and scientists) are not presented with information of perfect reliability (Hahn & Harris 2014). The degree of reliability is something that we must assess. A vital component of that reliability involves estimating the credibility of the information or new data source. For example, it is perfectly reasonable for a scientist to use prior knowledge of a question to evaluate the credibility of new data (Bovens & Hartmann 2003; Gentzkow & Shapiro 2006; Hahn &

174  Keith E. Stanovich Harris 2014; Olsson 2013). Scientists do this all the time, and it is rational. They use the discrepancy between the data they expect, given their prior hypothesis, and the actual data observed to estimate the credibility of the new data source (O’Connor & Weatherall 2018). The more significant the discrepancy, the more surprising the evidence is, and the more a scientist will question the source and thus reduce the weight given the new evidence. This cognitive strategy is sometimes called knowledge projection (see Stanovich 1999, 2021), and what is intriguing is that it is rational for a layperson to use it, too, if their prior belief represents actual knowledge (an evidence-based prior) and not just an unsupported desire for something to be true. What turns this situation into one of inappropriate myside bias is when a person uses not a belief that prior evidence leads them to think is true but instead projects an initial belief the person wants to be true despite inadequate evidence that it is correct by using a conviction-based prior (see Stanovich 2021). The term conviction conveys that these beliefs are often accompanied by emotional commitment and ego preoccupation (Abelson 1988). They can sometimes derive from protected values or partisan stances. The problematic kinds of myside bias derive from people projecting convictions, rather than evidence-based beliefs, onto new evidence they receive. That is how we end up with a society that seemingly cannot agree on empirically demonstrable facts. All arguments in favor of the normative appropriateness of myside bias given previously have concerned epistemic rationality only. However, there is a further set of arguments in favor of myside bias being instrumentally rational because of the social benefits of that kind of thinking. The social benefits of myside reasoning have been explored by many others (Clark & Winegard 2020; Clark et al. 2019; Greene 2013; Haidt 2012: Kahan 2013, 2015; Kahan et al. 2017; Mercier & Sperber 2017; Sloman & Fernbach 2017; Tetlock 2002; Van Bavel & Pereira 2018) and thus will not be pursued here other than to note that they complement the epistemic analysis in showing that it is difficult to ascertain, on a net-net basis, that mysided processing is non-normative. Thus, for all the reasons discussed in this section, myside bias is a different kind of bias – and it requires a different kind of theoretical explanation than traditional tasks in the heuristics and biases literature (Kahneman 2011; Stanovich 2011).

9.4  A Theoretical Alternative: Memetics In the literature, the default theoretical stance about myside bias tends to see it as process driven. The findings discussed above indicate that this default may require a reset. If it is indeed a process-based bias, those processes seem to be unpredictable from the most well-studied individual

Myside Bias in Individuals and Institutions 175 difference variables in psychology: intelligence and thinking dispositions such as actively open-minded thinking and need for cognition. Instead, opinion strength explains more variance in myside bias than psychological process indicators. We need an alternative conceptualization where myside bias is viewed as a content-based effect and not an individual difference trait. Models that focus on the properties of acquired beliefs, such as memetic theory (Blackmore 1999; Dennett 1995, 2017; Stanovich 2004, 2021), provide more suitable frameworks for studying myside bias. The critical question becomes not “How do people acquire beliefs?” (the tradition in social and cognitive psychology) but instead, “How do beliefs acquire people?” To avoid the most troublesome kind of myside bias (projecting beliefs that are not evidence-based), we need to distance ourselves from our convictions. It may help conceive our beliefs as memes with their own interests. We treat beliefs as possessions (see Abelson 1986) when we think that we have thought our way to these beliefs and that the beliefs are serving us. What Dennett (2017) calls the meme’s eye view leads us to question both assumptions (that we have thought our way to our beliefs and that they are serving our personal goals). Memes want to replicate whether they are beneficial for us or not, and they do not care how they get into a host – whether they get in through conscious thought or are simply an unconscious fit to innate psychological dispositions. In short, acquiring essential beliefs (convictions) without reflection is possible. There are many psychological examples where people acquire declarative knowledge, behavioral proclivities, and decision-making styles from innate propensities and (largely unconscious) social learning. For example, Haidt (2012) invokes this model to explain moral beliefs and behavior. The model is also applicable to the case of myside bias (see Stanovich 2021). The convictions driving your myside bias are partly caused by your biological makeup and partly by social learning from parents, peers, and schools. The convictions that determine your side when you think in a mysided fashion often do not stem from rational thought. People will feel less ownership of their beliefs when they realize they did not consciously reason their way to them. When a belief is held less like a possession, it is less likely to be projected to new evidence inappropriately. Recall that the problematic kind of myside bias (see Stanovich 2021 for a fuller discussion) is the kind that results when a person projects a conviction-­ based desired belief as a prior probability rather than a prior probability that has resulted from the rational processing of previous evidence. If we understand where convictions come from (our temperaments and social experience), we might be able to develop a more depersonalized stance toward our beliefs and thus avoid the problematic types of ­myside bias.

176  Keith E. Stanovich

9.5 The Peculiar Properties of Myside Bias Create an Epistemic Crisis in Universities The bias blind spot is a crucial meta-bias demonstrated in a paper by Pronin, Lin, and Ross (2002). They found that people thought various motivational biases were much more prevalent in others than themselves, a much-replicated finding (Pronin 2007; Scopelliti et al. 2015). In two studies, my research group (see West et al. 2012) showed that there is a blind spot regarding most of the classic cognitive biases in the literature – people think that most of these biases are more characteristic of others than themselves. We found positive correlations between the blind spots and cognitive sophistication – more cognitively skilled people were more prone to the bias blind spot. However, this makes sense because most cognitive biases in the heuristics and biases literature are negatively correlated with cognitive ability – more intelligent people are less biased (Stanovich 1999, 2011; Stanovich & West 1998a; Stanovich et al. 2016). Therefore, it would make sense for intelligent people to say that they are less biased than others – because they are! However, one bias – myside bias – sets a trap for the cognitively sophisticated. Regarding most biases, they are used to thinking that they are less biased. However, myside thinking about your political beliefs represents an outlier bias where this is not true (Drummond & Fischhoff 2019; Kahan, Peters, et al. 2012; Kahan et al. 2017; Sanchez & Dunning 2021; Stanovich 2021; Stanovich & West 2008; Van Boven et al. 2019). This disparity may lead to a particularly intense bias blind spot among cognitive elites. Specifically, they may be prone to think that traits such as intelligence (which they have) and experiences like education (which they also have in abundance) provide them with very generalizable inoculations against biased thinking. If you are a person of high intelligence, have lots of education, and are strongly committed to an ideological viewpoint, you will be especially prone to think that you thought your way to your opinions. You will be even less likely than the average person to know that you derived your beliefs from the social groups around you. The beliefs comported with your temperament and innate psychological propensities (see Haidt 2012; Stanovich 2021). There is, in fact, a group of people who tick all of these boxes: people who are highly intelligent, highly educated, and strongly committed to an ideological viewpoint. That group happens to be the group of social scientists who study politicized topics! The university professoriate is overwhelmingly left/liberal. This demographic fact has been demonstrated in numerous studies (Abrams 2016; Bikales & Goodman 2020; Ellis 2020; Horowitz et al. 2018; Jussim 2021; Kaufmann 2020; Langbert 2018; Langbert & Stevens 2020; Lukianoff & Haidt 2018; Peters et al. 2020). The trend is ubiquitous in the social sciences (sociology, political science, and like), and it is particularly solid

Myside Bias in Individuals and Institutions 177 in psychology, the source of many studies on politicized topics (Buss & von Hippel 2018; Cardiff & Klein 2005; Ceci & Williams 2018; Clark & Winegard 2020; Duarte et al. 2015). Ideological and partisan beliefs are known to lead to the unwarranted projection of prior attitudes on the evidence concerning a variety of issues, such as sexuality, morality, the psychological effects of poverty, family structures, crime, childcare, productivity, marriage, incentives, discipline techniques, educational practices, and many comparable subjects where distal political attitudes are intertwined with people’s beliefs on specific issues. It should be clear why scientific institutions with such an ideological imbalance cannot produce accurate research on these topics. Science overcomes the myside bias of individual scientists by immersing them in a system of checks and balances – where other scientists with differing biases are there to critique and correct. The bias of investigator A might not be shared by investigator B, who will then look at A’s results with a skeptical eye. Likewise, when investigator B presents a result, investigator A tends to be critical and look at it skeptically. However, what can ruin this scientific error detection and cross-checking process should be obvious. What ruins it is when all investigators share precisely the same bias. Unfortunately, the field of psychology is in just this situation concerning political ideology. The pool of investigators is politically homogeneous. Thus, we cannot rest assured that our science has enough variability to objectively approach politically charged topics like those mentioned above.

9.6  Psychology’s Self-Correction Problem The previously discussed Ditto et al. (2019a) findings highlight the danger of an academic elite thinking that they can investigate incendiary political topics on which they have strong feelings without compromising their research by myside bias. The Ditto et al. (2019a) findings show that the ideology of the cognitive elite is no less prone to myside bias than the political ideologies of the citizens that academics oppose. Nevertheless, because of their cognitive ability and educational backgrounds, society’s cognitive elites will think that their evidence processing is less driven by myside bias than their fellow citizens. As a result, we have ceased being a self-correcting science regarding specific topics. For years, it has been known that various types of racism scales1 used in psychology do not measure the construct correctly (Agadjanian et al. 2021; Carmines et al. 2011; Carney & Enos 2019; Reyna 2018; Wright et al. 2021; Zigerell 2015). Many of them literally build in correlations between prejudice and conservative views. Early versions of these scales included items on policy issues such as affirmative action, crime prevention, busing to achieve school integration, or attitudes

178  Keith E. Stanovich toward welfare reform and then scored any deviation from liberal orthodoxy as a racist response. Even endorsing the view that hard work leads to success for many people in America will get a higher score on a “symbolic racism” scale. The social science monoculture repeatedly yields the same embarrassing sequence time and again. We set out to study a negative trait concept (prejudice, dogmatism, authoritarianism, intolerance, close-­m indedness – the list is long). The traits studied are highly valenced – with one end of the trait continuum being good and the other being bad. Then the scale items are constructed like the racism scales discussed above – deliberately building in conservative social policy to define the negative construct. The scale is then used to associate conservatism with negative traits for a decade. Hundreds of articles are produced. The New York Times articles about the relevant research are written to show its liberal readers that research psychologists (yes, science!) have confirmed the reader’s view that liberals are indeed psychologically superior people – doing better on all the tests that scientists have constructed to measure whether people are open-minded, tolerant, and fair. The flaws in these scales were pointed out as long ago as the 1980s (Sniderman & Tetlock 1986). The decades-long failure to correct such deficiencies in these ideologically slanted scales undermines public confidence in psychology – as it should. Of course, there is indeed some self-correction in social science. After about a decade (or maybe two), a few researchers begin to probe whether there may have been theoretical confusion in a particular trait concept. Subsequent research often shows that the proposed trait was something different, or perhaps that the negative aspects can be found on either side of the ideological spectrum.2 For example, Conway et al. (2018) designed an authoritarianism scale on which liberals score higher than conservatives. They simply took old items that had disadvantaged conservatives and substituted content that disadvantaged liberals. The old item “Our country will be great if we honor the ways of our forefathers, do what the authorities tell us to do, and get rid of the ‘rotten apples’ who are ruining everything” was changed to “Our country will be great if we honor the ways of progressive thinking, do what the best liberal authorities tell us to do, and get rid of the religious and conservative ‘rotten apples’ who are ruining everything.” After the change, liberals scored higher on “authoritarianism” for the same reason that made the old scales correlate with conservatism – the content of the questionnaire targeted their views specifically.3 However, the fact that the scale eventually gets corrected, and the psychological construct eventually gets clarified, should not be viewed as necessarily flattering to psychology. Simply saying that corrections are eventually made obscures that the errors are always made in one direction (like at your local grocery, where things “ring up wrong” in the overcharge direction much more often than the reverse). The initial

Myside Bias in Individuals and Institutions 179 conclusion is that conservatives have higher levels of bad psychological traits. I had done this myself – in the 1990s, when, with colleagues, I constructed a questionnaire measuring actively open-minded thinking (AOT). One essential processing style tapped by the AOT concept is the subject’s willingness to revise beliefs based on evidence. Our early scales, first constructed decades ago, had several items to tap this processing style. However, my colleague Maggie Toplak and I discovered (Stanovich & Toplak 2019) that there is no generic belief revision tendency. Belief revision needs to be tapped with content because the specific belief determines how much people are willing to revise. In the mid-1990s, our items were biased against religious (and conservative) subjects as initially written. No doubt, if the correlations had come out in the other direction, we would have been quicker to notice a problem, as those of us constructing the items were all secularists. Cherry-picking scale items to embarrass our enemies is a seemingly irresistible tendency in psychology, as demonstrated during the recent pandemic. As is now well known, the media leaped to label as a conspiracy theory the idea that the virus might have originated in a lab in Wuhan because most mainstream media disliked the administration associated with that idea. Of course, labeling an alternative hypothesis as a conspiracy was deeply unscientific. There were still many viable virus origination theories at the time in 2020 when the media started their conspiracy mantra. In science, especially at the borderline of the unknown, we do not label every hypothesis other than the one with the highest Bayesian prior to be a conspiracy theory. In 2021, the media was embarrassed by this earlier behavior because, as often happens in science, the probability distribution across the viable theories shifted. It was well-publicized that even fact-checkers had made the error of calling everything but the primary hypothesis a conspiracy theory (Jilani 2021; Taibbi 2021; Tufekci 2021). Perhaps such media bias should be expected in the present environment but seeing fellow social science researchers doing the same thing was mortifying. Several studies of Covid-19 misinformation and Covid-19 conspiracy theory beliefs had items in their scales that labeled belief in the laboratory origin of the virus as a conspiracy theory.4 Before we knew anything with confidence about the origins of the virus that caused Covid-19, social scientists also jumped on the partisan bandwagon when they should have been the first to point out that minority hypotheses should not be labeled conspiracy theories at the beginning of an investigation. For some years, studies of conspiracy beliefs have been plagued by item selection bias. A few conspiracy theories are prevalent on the left; others are on the right; many have no association with ideology.5 It is thus trivially easy to select conspiracy theories disproportionally so that there will be ideological correlations in one direction or the other. However,

180  Keith E. Stanovich such correlations would not represent facts about people’s underlying psychological structure. They would merely be sampling artifacts. Kahan (2015) has shown that the heavy reliance of scientific knowledge tests on items involving belief in climate change and evolutionary origins has built correlations between liberalism and scientific knowledge into such measures. Notably, his research has demonstrated that removing belief in human-caused climate change and evolutionary origin items from scientific knowledge scales reduces the correlation between scientific knowledge and liberalism. Eliminating these items also makes the remaining test more valid because responses on climate and evolution items are expressive responses signaling group allegiance rather than responses that reflect actual science knowledge (Kahan & Stanovich 2016). All studies of the “who is more knowledgeable” type in the political domain are at risk of being compromised by such item selection effects. Over the years, Democrats in the United States have called themselves the “party of science” – and they are regarding climate science and belief in the evolutionary origins of humans. Nevertheless, regarding topics like the heritability of intelligence and sex differences, the Democrats suddenly become the “party of science denial” (Stanovich 2021). Whoever controls the selection of items will find it irresistible not to bias the selection according to their notion of what knowledge is essential – choosing items that fellow tribe members pass easily and that are opaque to their political enemies and not seeing the possibility of someone else doing exactly the opposite. It is distressingly easy to expose the ignorance of a group we do not like if we control the selection of the items.

9.7 Academia’s Perverse Response to the Scientific Problem of an Ideological Monoculture: Doubling Down How has social science responded to the problems inherent in being an institutionalized ideological monoculture? How has it sought to bolster confidence in its scientific conclusions? Incredibly, the response has been to double down on insisting that, when defining good thinking, only they are the authorities. You must answer the questions posed by these researchers correctly, but you must now also affirm that the creators of the tests are the ultimate arbiters of what constitutes good thinking. I am referring to the increasing popularity of so-called science trust or “faith in science” scales (I am guilty of authoring one of these scales myself!). On such questionnaires, the respondent is often asked whether they trust universities, the media, or the results of scientific research on pressing social issues. When the respondent answers that they do not trust university research very much, their epistemic abilities are deemed inferior. They are called science deniers, or people who do not “follow the science.”

Myside Bias in Individuals and Institutions 181 However, most of these “trust in experts” measures used in behavioral science have methodological problems. The researchers employing them view low-trust subjects as epistemically defective in their failure to rely strongly on expert opinion when forming their beliefs. Investigators consider more acceptance of information from experts such as these as better. Indeed, maximum acceptance (answering “complete acceptance” on the scale) is explicitly deemed optimal in the statistical analysis of such measures. How times have changed. In the 1960s and the 1970s, it was viewed as progressive to display skepticism toward all claims of expertise. Making people more skeptical toward government officials, journalists, and universities was viewed as progressive because, by doing so, we thought we were moving toward a more accurate worldview. It was thought then that truth was obscured by the self-serving interests of precisely the groups listed on current “expert acceptance” questionnaires! Nevertheless, it is viewed as an epistemological defect when conservatives currently evince more skepticism on these scales. Actually, no one knows how much trust in institutions is optimal, so these measures cannot possibly have valid scoring protocols. Related to these “trust in experts” scales are the “trust in science” scales in the psychological literature (or their complement, so-called anti-scientific attitude scales). I have constructed such a scale but now consider that it is a conceptual error and, as a measure, will be prone to misuse. When you ask a subject to respond to an item such as “science is the best method of acquiring knowledge,” you might as well ask the subject whether they have received higher education. The social benefit of attending a university is learning that you are supposed to endorse items like this. Every person with a BA knows it is a good thing to “follow the science,” as we have often heard during the pandemic. The same BA equips you to criticize your fellow citizens who do not know that “trust the science” is a code word used by university-educated elites. For these reasons, I have removed the anti-science attitudes subtest from my lab’s omnibus measure of rational thinking, the Comprehensive Assessment of Rational Thinking (Stanovich et al. 2016). It does not provide a clean and unbiased measurement of that construct. Suppose we want to examine people’s attitudes toward scientific evidence. In that case, we must take a domain-specific belief that a person has on a scientific matter, present them with contradictory evidence, and see how they assimilate that contradictory evidence (as some studies have done). You cannot just ask people whether they “follow the science” on a questionnaire. That is the equivalent of constructing a test and giving half the respondents the answer sheet. It would not be an independent finding when those with the answer sheet do better. Such scales measure nothing more than whether the respondent is a member of the tribe that designed the test.

182  Keith E. Stanovich My scale was far from the most misguided of this type. These questionnaires can get quite aggressive in what they require assent to if one is to avoid the label “anti-science.” For example, one scale (Farias et al. 2013) requires the subjects to affirm propositions such as “We can only rationally believe in what is scientifically provable,” “Science tells us everything there is to know about what reality consists of,” and “All the tasks human beings face are soluble by science,” “Science is the most valuable part of human culture.” The above is a quite an uncompromising set of beliefs to have to endorse so as not to end up in the “low faith in science” group in an experiment! One can be appropriately calibrated to scientific evidence without enthusiastically affirming statements like these, which seem to claim Promethean status for science. In addition to this problem of overblown notions of what belief in science entails, the social sciences employ definitions saturated with their own myside bias. Consider a study (Feygina et al. 2010) that attempted to link the conservative worldview with “the denial of environmental realities.” Subjects were presented with the following item: If things continue their present course, we will soon experience a major environmental catastrophe. If the subject did not agree with this statement, they were scored as denying ecological realities. The term denial implies that what is being denied is a descriptive fact. However, without a clear description of what “soon” means in this statement, what “major” means, or what “catastrophe” means, the statement itself is not a fact – and so labeling one set of respondents as science deniers based on an item like this reflects little more than the ideological position of the study’s authors. This tendency to conflate liberal responses with the correct answer (or ethical response, fair response, scientific response, or open-minded response) is particularly prevalent in social psychology and personality psychology subareas. Studies often label any legitimate policy difference with liberalism as an intellectual or personality defect (dogmatism or authoritarianism or racism or prejudice, or science denial). In one utterly typical study (Azevedo & Jost 2021), the aggressive label “social dominance orientation” is used to describe anyone who does not endorse both identity politics (emphasizing groups when thinking about justice) and the new meaning of equity (equality of outcomes). A subject who does not support the item “group equality should be our ideal” is scored in the direction of having a social dominance orientation (wanting to maintain the dominant group in a hierarchy). A conservative individual (or an old-style Democrat) who values equality of opportunity and focuses on the individual will naturally score higher in social dominance orientation than a left-wing advocate of groupbased identity politics who focuses on equality among groups. The entire construct of social dominance represents a form of doubling down on the bet that one’s beliefs are correct. The concept assumes that the world should be interpreted through the lens of identity politics and scores in the negative direction any response that strays from that

Myside Bias in Individuals and Institutions 183 worldview. In contrast, if a subject denies that group performance is the measure of fairness and thinks instead that fairness is a construct that applies at the individual level, they will be said to have a social dominance orientation – even though group outcomes are not even salient in the subject’s framework. The subject’s fairness concepts are ignored, and the experimenter’s framework is instead imposed upon them. The study then defines “skepticism about science” with just two items. The first, “We believe too often in science, and not enough in faith and feelings,” builds into the scale a direct conflict between religious faith and science that many subjects might not actually experience, thus inflating correlations with religiosity. The second, “When it comes to fundamental questions, scientific facts don’t help very much,” is even more interesting. If you think that the essential things in life are marriage, family, raising children with good values, and being a good neighbor – and thus answer that you agree on this item, you will get a higher score on this science skepticism scale than a person who believes that the essential things in life are climate change and green technology. Neither of these items shows that conservative subjects are anti-science, but they ensure that conservatism/religiosity will be correlated with the misleading construct that names the scale: “science skepticism.” Cognitive elites often use “fact-checking” to double down on their insistence that adherence to the norms of the institutions that they control are the only arbiters of truth. If you do not accept the conclusions of the fact-checkers, you are not “following the science.” Academic researchers in the social sciences seem oblivious to an implication from research on myside bias (Stanovich 2021) – that a primary source of bias is the selection of items to fact-check in the first place. More problematic than inaccuracies in the fact checks themselves is the automatic myside bias that will trigger choosing one proposition over another for checking amongst a population of thousands (Uscinski & Butler 2013). Unfortunately, fact-checkers have become just another player in the unhinged partisan cacophony of our politics. Progressive academics populate many leading organizations in universities, others are run by, and some are connected to Democratic donors in the United States. You cannot expect such entities to win respect among the general population when they have such ideological connections and do not fully instantiate inclusive adversarial collaboration (see below). Fact-checking is particularly prone to myside bias in the political domain. The slipups that occur always seem to favor the ideological proclivities of the liberal media outlets that sponsor the fact checks. As previously noted, academic research groups immediately began including items that classified belief in a lab origin for the Covid-19 virus as a conspiracy belief in their studies. Scholars and commentators talking about the possibility of a lab origin were censored on social media (Taibbi 2021). A New York Times reporter (a science reporter no less)

184  Keith E. Stanovich said the lab origin hypothesis had “racist origins” (Jilani 2021). The idea of a lab origin was labeled false and a “debunked” idea by fact-checking websites (Taibbi 2021). However, in May 2021, the fact-checking website Politifact issued a retraction to their September 2020 assessment that a lab origin was a “pants on fire” claim. Likewise, fact-checking websites quickly refuted the Trump administration’s claims that a vaccine would be available in 2020 (Tierney 2021). Of course, we now know the vaccine rollout was in December 2020. Regarding many Covid-19 issues, these organizations had no business treating ongoing scientific disputes (origins of the virus, the efficacy of lockdowns) as if they were a matter of established “fact” they could check. They were, as Tufekci (2021) phrased it in an essay, “checking facts even if you can’t.” In an ongoing scientific dispute with a dominant hypothesis warranting a 60% Bayesian prior, a minority hypothesis with 20% credence does not become a “conspiracy theory,” and those advocating for it are not making a “pants on fire” claim. Unfortunately, this was characteristic of fact-checking organizations and many social science researchers studying the spread of misinformation throughout the pandemic. They were too quick to double down by insisting that adherence to their approach to these complex pandemic issues was a sign of epistemic rationality.

9.8  Restoring Epistemic Legitimacy to the Social Sciences I wrote a book on myside bias (Stanovich 2021) that discussed in detail the difficulty each of us has in checking our tendencies to evaluate and generate evidence in a manner that favors our pre-existing opinion. The remedy for our society-wide epistemic crisis will not be any quick fix at the individual level. The ultimate reform must be at the level of our institutions (Rauch 2021) – precisely the institutions (media and universities) that have lost their status as neutral adjudicators of truth claims in recent years. The answer cannot be to tell the populace to turn more strongly to the same institutions that have been failing us. You cannot do that unless you change the institutions themselves. Academics pile up more and more studies of the psychological “deficiencies” of the voters who do not support the Democrats or who voted for Brexit. They pile up more conclusions on all the pressing issues of the day (immigration, crime, inequality, race relations) using research teams without any representation outside the left/liberal-progressive consensus. We – universities, social science departments, my tribe – have sorted by temperament, values, and culture into a monolithic intellectual edifice that has long ceased to be a neutral adjudicator of fraught social issues. We create tests to reward and celebrate the intellectual characteristics we define ourselves by and skewer those we deplore. The broader population no longer trusts us. Thus, in the last ten to twenty years, we have created

Myside Bias in Individuals and Institutions 185 another layer of tests to find our critics guilty more elementally: they are anti-science and do not “trust expertise.” How do you get a low score on our new meta-tests? Answer: Say you do not trust us. I have sat with university faculty and joked about how doctors think they can regulate themselves, but we psychologists know that it is implausible that they will do it objectively. Nevertheless, we turn around and object when there is disbelief in our self-regulating ability. We have an epistemic crisis because cognitive elites have become so self-referential that they no longer command the respect of the rest of the populace. We have been cleansing disciplines of ideologically dissident voices for thirty years now with relentless efficiency. We have attempted to define the beliefs of our political enemies as pathologies. If you say that hard work will lead to African Americans’ success, you will display “symbolic racism.” Belief in the equality of opportunity for individuals combined with skepticism of government-enforced equality of outcomes for groups becomes a “social dominance orientation.” If you do not believe that “All the tasks human beings face are soluble by science” or that science does not answer many of the most important questions, you will be labeled anti-science or a science skeptic. If you do not believe there will soon be a “climate catastrophe,” you will be labeled a science denier.6 As a result, public trust in us is sinking. It will not reverse, and it should not reverse – until we take measures to ensure that we are triangulating social issues using various frameworks.

9.9 Adversarial Collaboration and the Ideological Monoculture Institutions, administrators, and faculty seem unconcerned about the public’s plummeting trust in universities. Most outsiders, though, see the monoculture as a bug. If academia wanted to fix the bug, it would turn intensely to mechanisms such as adversarial collaboration, which is well described on the website of the Adversarial Collaboration Project of the University of Pennsylvania (see Clark & Tetlock 2022). Adversarial collaboration seeks to broaden the frameworks within research groups by encouraging disagreeing scholars to work together. Researchers from opposing perspectives design methods that both sides agree to constitute a fair test and jointly publish the results. Both sides participate in interpreting the findings and conclusions based on pre-agreed criteria. Adversarial collaborations prevent researchers from designing studies likely to support a predetermined hypothesis and dismiss unexpected results. Most importantly, findings based on adversarial collaborations can be fairly presented to consumers of scientific information as proper consensus conclusions and not outcomes determined by one side’s success in shutting the other out. There is a significant obstacle, however. It is not certain that, in the future, universities will have enough conservative scholars to function in

186  Keith E. Stanovich the needed adversarial collaborations. The diversity statements that candidates for faculty positions must now write are a significant impediment to increasing intellectual diversity in academia. A candidate will not advance their chance of attaining a faculty position unless they affirm belief in the tenets of progressive identity politics and pledge allegiance to its many terms and concepts without getting too picky about their lack of operational definition (diversity, systemic racism, white privilege, inclusion, equity). Such statements function like ideological loyalty oaths (Jussim 2019; McBrayer, 2022; Rozado 2019; Small 2021; Thompson 2019). You will not be hired if you do not endorse the current shibboleths of identity politics. One wonders whether, in the future, there will be enough intellectual diversity left in academia to make actual adversarial collaboration possible.

Notes





1. Contemporary scales go by a variety of names. The most common labels are racial resentment, symbolic racism, and modern racism (Carmines et al. 2011; Henry & Sears 2002). 2. Alternatively, the history of such scales starts out being unidimensionally negative on one end (high authoritarianism is always worse). However, the concept morphs into something resembling a cognitive style – where extremes on either end look suboptimal, and the wisest response seems somewhere in the middle. For example, authoritarianism morphs into security concerns (Hibbing 2020) or a fixed versus fluid worldview ­(Hetherington & Weiler 2018). 3. Costello et al. (2021) present results on a more psychometrically sound leftwing authoritarianism scale than that investigated by Conway – one with a more thoroughly established construct validity. 4. See Gligorić et al. (2021) and Teovanović et al. (2021), but there are many other examples. 5. If a scale includes a broad sampling of items, the correlation with ideology should not be that large (Enders & Uscinski 2021; Oliver & Wood 2014; Stanovich et al. 2016). The conspiracy belief subtest of our Comprehensive Assessment of Rational Thinking (CART) sampled twenty-four different conspiracy theories. 6. Singal (2018) parodied our field’s flaws when he posited the Jesse Singal Authoritarianism scale consisting of three items cherry-picked to expose liberals in the same way that the academic literature targets conservatives: “In certain cases, it might be acceptable to curtail people’s constitutional rights to stop them from spreading climate-change denialism”; “The government needs to do a much more comprehensive job monitoring ­Christian-oriented far-right terrorism”; “Some people want to act like the causes of racism are complicated, but they aren’t”: “Racists are moral failures, and that’s that” – a scale which would target liberals as the authoritarians.

References Abelson, R. P. (1986). Beliefs are like possessions. Journal of the Theory of Social Behaviour, 16, 223–250. Abelson, R. P. (1988). Conviction. American Psychologist, 43, 267–275.

Myside Bias in Individuals and Institutions 187 Abrams, S. (2016, January 9). Professors moved left in the 1990s, but the rest of the country did not. Heterodox Academy. https://heterodoxacademy.org/ professors-moved-left-but-country-did-not/ Agadjanian, A., Carey, J. M., Horiuchi, Y., & Ryan, T. J. (2021). Disfavor or favor? Assessing the valence of white Americans’ racial attitudes. SSRN. https://ssrn.com/abstract=3701331 Alloy, L. B., & Tabachnik, N. (1984). Assessment of covariation by humans and animals: The joint influence of prior expectations and current situational information. Psychological Review, 91, 112–149. Azevedo, F., & Jost, J. T. (2021). The ideological basis of anti-scientific attitudes: Effects of authoritarianism, conservatism, religiosity, social dominance, and system justification. Group Processes & Intergroup Relations, 24(4), 518–549. Babcock, L., Loewenstein, G., Issacharoff, S., & Camerer, C. (1995). Biased judgments of fairness in bargaining. The American Economic Review, 85, 1337–1343. Bikales, J., & Goodman, J. (2020, March, 5). A plurality of surveyed Harvard faculty support Warren in the presidential race. Harvard Crimson. https://www.­t hecrimson.com/article/2020/3/3/faculty-support-warrenpresident/#disqus_thread Blackmore, S. (1999). The Meme Machine. New York: Oxford University Press. Bolsen, T., & Palm, R. (2020). Motivated reasoning and political decision-­making. In W. Thompson (Ed.), Oxford Research Encyclopedia, Politics. doi:10.1093/ acrefore/9780190228637.013.923. https://oxfordre.com/politics/politics/view/ 10.1093/acrefore/9780190228637.001.0001/acrefore-9780190228637-e-923 Bovens, L., & Hartmann, P. (2003). Bayesian Epistemology. Oxford: Oxford University Press. Bruine de Bruin, W., Parker, A. M., & Fischhoff, B. (2007). Individual differences in adult decision-making competence. Journal of Personality and Social Psychology, 92, 938–956. Buss, D. M., & von Hippel, W. (2018). Psychological barriers to evolutionary psychology: Ideological bias and coalitional adaptations. Archives of Scientific Psychology, 6, 148–158. Cardiff, C. F., & Klein, D. B. (2005). Faculty partisan affiliations in all disciplines: A voter-registration study. Critical Review, 17(3–4), 237–255. Carmines, E. G., Sniderman, P. M., & Easter, B. C. (2011). On the meaning, measurement, and implications of racial resentment. The American Academy of Political and Social Science Annals, 634(1), 98–116. doi:10.1177/0002716210387499 Carney, R. K., & Enos, R. (2019). Conservatism, just world belief, and racism: An experimental investigation of the attitudes measured by modern racism scales. 2017 NYU CESS Experiments Conference. Working Paper under review. http://www.rileycarney.com/research Ceci, S. J., & Williams, W. M. (2018). Who decides what acceptable speech on campus is? Why restricting free speech is not the answer. Perspectives on Psychological Science, 13, 299–323. Clark, C. J., Liu, B. S., Winegard, B. M., & Ditto, P. H. (2019). Tribalism is human nature. Current Directions in Psychological Science, 28, 587–592. Clark, C. J., & Tetlock, P. E. (2022). Adversarial collaboration: The next science reform. In C. L. Frisby, R. E. Redding, W. T. O’Donohue, & S. O. Lilienfeld (Eds.), Political Bias in Psychology: Nature, Scope, and Solutions. New York: Springer.

188  Keith E. Stanovich Clark, C. J., & Winegard, B. M. (2020). Tribalism in war and peace: The nature and evolution of ideological epistemology and its significance for modern social science. Psychological Inquiry, 31, 1–22. Clements, Z. A., & Munro, G. D. (2021). Biases and their impact on opinions of transgender bathroom usage. Journal of Applied Social Psychology. doi:10.1111/ jasp.12741 Conway, L. G., Houck, S. C., Gornick, L. J., & Repke, M. A. (2018). Finding the Loch Ness monster: Left-wing authoritarianism in the United States. Political Psychology, 39, 1049–1067. Costello, T. H., Bowes, S. M., Stevens, S. T., Waldman, I. D., Tasimi, A., & Lilienfeld, S. O. (2021). Clarifying the structure and nature of left-wing authoritarianism. Journal of Personality and Social Psychology, 122, 135–170. Dennett, D. C. (1995). Darwin’s Dangerous Idea: Evolution and the Meanings of Life. New York: Simon & Schuster. Dennett, D. C. (2017). From Bacteria to Bach and Back. New York: Norton. Ditto, P., Liu, B., Clark, C., Wojcik, S., Chen, E., Grady, R., Celniker, J., & Zinger, J. (2019a). At least bias is bipartisan: A meta-analytic comparison of partisan bias in liberals and conservatives. Perspectives on Psychological Science, 14, 273–291. Ditto, P., Liu, B., Clark, C., Wojcik, S., Chen, E., Grady, R., Celniker, J., & Zinger, J. (2019b). Partisan bias and its discontents. Perspectives on Psychological Science, 14, 304–316. Druckman, J. N. (2012). The politics of motivation. Critical Review, 24(2), 199–216. Drummond, C., & Fischhoff, B. (2017). Individuals with greater science literacy and education have more polarized beliefs on controversial science topics. Proceedings of the National Academy of Sciences, 114(36), 9587. Drummond, C., & Fischhoff, B. (2019). Does “putting on your thinking cap” reduce myside bias in evaluating scientific evidence? Thinking & Reasoning, 25, 477–505. Duarte, J. L., Crawford, J. T., Stern, C., Haidt, J., Jussim, L., & Tetlock, P. E. (2015). Political diversity will improve social psychological science. Behavioral and Brain Sciences, 38, e130. doi:10.1017/S0140525X14000430 Edwards, K., & Smith, E. E. (1996). A disconfirmation bias in the evaluation of arguments. Journal of Personality and Social Psychology, 71, 5–24. Ehret, P. J., Sparks, A. C., & Sherman, D. K. (2017). Support for environmental protection: An integration of ideological-consistency and information-deficit models. Environmental Politics, 26, 253–277. Eichmeier, A., & Stenhouse, N. (2019). Differences that don’t make much difference: Party asymmetry in open-minded cognitive styles has little relationship to information processing behavior. Research & Politics, 6(3), 2053168019872045. doi:10.1177/2053168019872045 Ellis, J. M. (2020). The Breakdown of Higher Education. New York: Encounter. Enders, A. M., & Uscinski, J. E. (2021). Are misinformation, anti-scientific claims, and conspiracy theories for political extremists? Group Processes & Intergroup Relations, 24(4), 583–605. Epley, N., & Gilovich, T. (2016). The mechanics of motivated reasoning. Journal of Economic Perspectives, 30(3), 133–140.

Myside Bias in Individuals and Institutions 189 Evans, J. St. B. T., Over, D. E., & Manktelow, K. (1993). Reasoning, decision making, and rationality. Cognition, 49, 165–187. Farias, M., Newheiser, A.-K., Kahane, G., & de Toledo, Z. (2013). Scientific faith: Belief in science increases in the face of stress and existential anxiety. Journal of Experimental Social Psychology, 49(6), 1210–1213. Feygina, I., Jost, J. T., & Goldsmith, R. E. (2010). System justification, the denial of global warming, and the possibility of “system-sanctioned change.” Personality and Social Psychology Bulletin, 36(3), 326–338. Gentzkow, M., & Shapiro, J. (2006). Media bias and reputation. Journal of Political Economy, 114, 280–316. Gligorić, V., da Silva, M. M., Eker, S., van Hoek, N., Nieuwenhuijzen, E., Popova, U., & Zeighami, G. (2021). The usual suspects: How psychological motives and thinking styles predict the endorsement of well-known COVID-19 conspiracy beliefs. Applied Cognitive Psychology. doi:10.1002/acp.3844 Greene, J. D. (2013). Moral Tribes. New York: Penguin Press. Guay, B., & Johnston, C. (2021). Ideological asymmetries and the determinants of politically motivated reasoning. American Journal of Political Science, 65. Gugerty, L., Shreeves, M., & Dumessa, N. (2021). Biased belief updating in causal reasoning about COVID-19. Journal of Experimental Psychology: Applied, 27, 695–721. Hahn, U., & Harris, A. J. L. (2014). What does it mean to be biased: Motivated reasoning and rationality. In B. H. Ross (Ed.), Psychology of Learning and Motivation (Vol. 61, pp. 41–102). New York: Academic Press. Haidt, J. (2012). The Righteous Mind. New York: Pantheon Books. Hamilton, L. C. (2011). Education, politics, and opinions about climate change evidence for interaction effects. Climatic Change, 104, 231–242. Hart, W., Albarracin, D., Eagly, A. H., Brechan, I., Lindberg, M. J., & Merrill, L. (2009). Feeling validated versus being correct: A meta-analysis of selective exposure to information. Psychological Bulletin, 135, 555–588. Henry, P. J., & Napier, J. L. (2017). Education is related to greater ideological prejudice. Public Opinion Quarterly, 81, 930–942. Henry, P. J., & Sears, D. O. (2002). The symbolic racism 2000 scale. Political Psychology, 23, 253–283. Hetherington, M., & Weiler, J. (2018). Prius or Pickup? New York: Houghton Mifflin. Hibbing, J. R. (2020). The Securitarian Personality. New York: Oxford University Press. Horowitz, M., Haynor, A., & Kickham, K. (2018). Sociology’s sacred victims and the politics of knowledge: Moral foundations theory and disciplinary controversies. The American Sociologist, 49, 459–495. Houston, D. A., & Fazio, R. H. (1989). Biased processing as a function of attitude accessibility: Making objective judgments subjectively. Social Cognition, 7, 51–66. Jilani, Z. (2021, May 27). American journalists shielded China and erased the Wuhan lab leak theory. Newsweek. https://www.newsweek.com/americanjournalists-shielded-china-erased-wuhan-lab-leak-theory-opinion-1595400 Jones, P. E. (2019). Partisanship, political awareness, and retrospective evaluations, 1956–2016. Political Behavior. doi:10.1007/s11109-019-09543-y

190  Keith E. Stanovich Joslyn, M. R., & Haider-Markel, D. P. (2014). Who knows best? Education, partisanship, and contested facts. Politics & Policy, 42, 919–947. Jussim, L. (2019, February 24). My diversity, equity, and inclusion statement. Quillette. https://quillette.com/2019/02/24/my-diversity-equity-and-inclusion-statement/ Jussim, L. (2021, March 25). How social norms create a culture of censorship, part 2. Psychology Today. https://www.psychologytoday.com/us/blog/rabbleKahan, D. M. (2013). Ideology, motivated reasoning, and cognitive reflection. Judgment and Decision Making, 8, 407–424. Kahan, D. M. (2015). Climate-science communication and the measurement problem. Political Psychology, 36, 1–43. Kahan, D. M., & Corbin, J. C. (2016). A note on the perverse effects of actively open-minded thinking on climate-change polarization. Research & Politics, 3(4), 1–5. doi:10.1177/2053168016676705 Kahan, D. M., Hoffman, D. A., Braman, D., Evans, D., & Rachlinski, J. J. (2012). “They Saw a Protest”: Cognitive illiberalism and the speech-conduct distinction. Stanford Law Review, 64(4), 851–906. Kahan, D. M., Peters, E., Dawson, E., & Slovic, P. (2017). Motivated numeracy and enlightened self-government. Behavioural Public Policy, 1, 54–86. Kahan, D. M., Peters, E., Wittlin, M., Slovic, P., Ouellette, L., Braman, D., & Mandel, G. (2012). The polarizing impact of science literacy and numeracy on perceived climate change risks. Nature Climate Change, 2, 732–735. Kahan, D. M., & Stanovich, K. E. (2016, September 14). Rationality and belief in human evolution. Annenberg Public Policy Center Working Paper No. 5. https://ssrn.com/abstract=2838668 Kahneman, D. (2011). Thinking, Fast and Slow. New York: Farrar, Straus & Giroux. Kahneman, D., & Tversky, A. (1973). On the psychology of prediction. Psychological Review, 80, 237–251. Kaufmann, E. (2020, September 1). The denial of cancel culture. Quillette. https://quillette.com/2020/09/01/the-denial-of-cancel-culture/ Klaczynski, P. A. (1997). Bias in adolescents’ everyday reasoning and its relationship with intellectual ability, personal theories, and self-serving motivation. Developmental Psychology, 33, 273–283. Klaczynski, P. A., & Lavallee, K. L. (2005). Domain-specific identity, epistemic regulation, and intellectual ability as predictors of belief-based reasoning: A dual-process perspective. Journal of Experimental Child Psychology, 92, 1–24. Klaczynski, P. A., & Robinson, B. (2000). Personal theories, intellectual ability, and epistemological beliefs: Adult age differences in everyday reasoning tasks. Psychology and Aging, 15, 400–416. Koehler, J. J. (1993). The influence of prior beliefs on scientific judgments of evidence quality. Organizational Behavior and Human Decision Processes, 56, 28–55. Kornblith, H. (1993). Inductive Inference and Its Natural Ground. Cambridge, MA: MIT University Press. Kraft, P. W., Lodge, M., & Taber, C. S. (2015). Why people “Don’t trust the evidence”: Motivated reasoning and scientific beliefs. The American Academy of Political and Social Science Annals, 658(1), 121–133. Kuhn, D., & Modrek, A. (2018). Do reasoning limitations undermine discourse? Thinking & Reasoning, 24, 97–116.

Myside Bias in Individuals and Institutions 191 Langbert, M. (2018). Homogenous: The political affiliations of elite liberal arts college faculty. Academic Questions, 31, 186–197. Langbert, M., & Stevens, S. (2020). Partisan registration and contributions of faculty in flagship colleges. National Association of Scholars. https://www.nas.org/blogs/ article/partisan-registration-and-contributions-of-faculty-in-flagship-colleges Lukianoff, G., & Haidt, J. (2018). The Coddling of the American Mind. New York: Penguin. Lupia, A., Levine, A. S., Menning, J. O., & Sin, G. (2007). Were Bush tax cut supporters “simply ignorant?” A second look at conservatives and liberals in “Homer Gets a Tax Cut.” Perspectives on Politics, 5, 773–784. Macpherson, R., & Stanovich, K. E. (2007). Cognitive ability, thinking dispositions, and instructional set as predictors of critical thinking. Learning and Individual Differences, 17, 115–127. McBrayer, J. P. (2022, May 23). Diversity statements are the new faith statements. Inside Higher Ed. https://www.insidehighered.com/views/2022/05/23/ diversity-statements-are-new-faith-statements-opinion Mercier, H. (2017). Confirmation bias - myside bias. In R. Pohl (Ed.), Cognitive Illusions (2nd Ed.) (pp. 99–114). New York: Routledge. Mercier, H., & Sperber, D. (2017). The Enigma of Reason. Cambridge, MA: Harvard University Press. O’Connor, C., & Weatherall, J. O. (2018). Scientific polarization. European Journal for Philosophy of Science, 8, 855–875. Oliver, J. E., & Wood, T. (2014). Conspiracy theories and the paranoid style(s) of mass opinion. American Journal of Political Science, 58, 952–966. Olsson, E. J. (2013). A Bayesian simulation model of group deliberation and polarization. In F. Zenker (Ed.), Bayesian Argumentation (pp. 113–133). Netherlands: Springer. Perkins, D. N., Farady, M., & Bushey, B. (1991). Everyday reasoning and the roots of intelligence. In J. Voss, D. Perkins, & J. Segal (Eds.), Informal Reasoning and Education (pp. 83–105). Hillsdale, NJ: Erlbaum. Peters, U., Honeycutt, N., De Block, A., & Jussim, L. (2020). Ideological diversity, hostility, and discrimination in philosophy. Philosophical Psychology, 33(4), 511–548. Pronin, E. (2007). Perception and misperception of bias in human judgment. Trends in Cognitive Sciences, 11, 37–43. Pronin, E., Lin, D. Y., & Ross, L. (2002). The bias blind spot: Perceptions of bias in self versus others. Personality and Social Psychology Bulletin, 28, 369–381. Rauch, J. (2021). The Constitution of Knowledge: A Defense of Truth. Washington, DC: Brookings. Reyna, C. (2018). Scale creation, use, and misuse: How politics undermines measurement. In J. T. Crawford & L. Jussim (Eds.), The Politics of Social Psychology (pp. 81–98). New York: Routledge. Rozado, D. (2019, August 5). What do universities mean when they talk about diversity? A computational language model quantifies. Heterodox Academy. https://heterodoxacademy.org/diversity-what-do-universities-mean/ Sá, W., West, R. F., & Stanovich, K. E. (1999). The domain specificity and generality of belief bias: Searching for a generalizable critical thinking skill. Journal of Educational Psychology, 91, 497–510.

192  Keith E. Stanovich Sanchez, C., & Dunning, D. (2021). Cognitive and emotional correlates of belief in political misinformation: Who endorses partisan misbeliefs? Emotion, 21, 1091–1102. Sarathchandra, D., Navin, M. C., Largent, M. A., & McCright, A. M. (2018). A survey instrument for measuring vaccine acceptance. Preventive Medicine, 109, 1–7. Scopelliti, I., Morewedge, C. K., McCormick, E., Min, H. L., Lebrecht, S., & Kassam, K. S. (2015). Bias blind spot: Structure, measurement, and consequences. Management Science, 61, 2468–2486. Simas, E. N., Clifford, S., & Kirkland, J. H. (2020). How empathic concern fuels political polarization. American Political Science Review, 114, 258–269. Singal, J. (2018, July 15). How social science might be misunderstanding conservatives. New York Magazine. https://nymag.com/intelligencer/2018/07/howsocial-science-might-be-misunderstanding-conservatives.html Sloman, S., & Fernbach, P. M. (2017). The Knowledge Illusion. New York: Riverhead Books. Small, A. (2021, June 30). Diversity, equity, and inclusion statements: Compassion filter or ideological test? Heterodox Academy. https://heterodoxacademy. org/blog/diversity-equity-and-inclusion-statements-compassion-filter-orideological-test/ Sniderman, P. M., & Tetlock, P. E. (1986). Symbolic racism: Problems of motive attribution in political analysis. Journal of Social Issues, 42, 129–150. Stanovich, K. E. (1999). Who Is Rational? Studies of Individual Differences in Reasoning. Mahwah, NJ: Erlbaum. Stanovich, K. E. (2004). The Robot’s Rebellion: Finding Meaning in the Age of Darwin. Chicago: University of Chicago Press. Stanovich, K. E. (2011). Rationality and the Reflective Mind. New York: Oxford University Press. Stanovich, K. E. (2021). The Bias That Divides Us: The Science and Politics of Myside Thinking. Cambridge, MA: MIT Press. Stanovich, K. E., & Toplak, M. E. (2019). The need for intellectual diversity in psychological science: Our own studies of actively open-minded thinking as a case study. Cognition, 187, 156–166. Stanovich, K. E., & West, R. F. (1997). Reasoning independently of prior belief and individual differences in actively open-minded thinking. Journal of Educational Psychology, 89, 342–357. Stanovich, K. E., & West, R. F. (1998a). Individual differences in rational thought. Journal of Experimental Psychology: General, 127, 161–188. Stanovich, K. E., & West, R. F. (1998b). Who uses base rates and P(D/~H)? An analysis of individual differences. Memory & Cognition, 26, 161–179. Stanovich, K. E., & West, R. F. (2000). Individual differences in reasoning: Implications for the rationality debate? Behavioral and Brain Sciences, 23, 645–726. Stanovich, K. E., & West, R. F. (2007). Natural myside bias is independent of cognitive ability. Thinking & Reasoning, 13, 225–247. Stanovich, K. E., & West, R. F. (2008). On the failure of intelligence to predict myside bias and one-sided bias. Thinking & Reasoning, 14, 129–167. Stanovich, K. E., West, R. F., & Toplak, M. E. (2016). The Rationality Quotient: Toward a Test of Rational Thinking. Cambridge, MA: MIT Press.

Myside Bias in Individuals and Institutions 193 Stenhouse, N., Myers, T. A., Vraga, E. K., Kotcher, J. E., Beall, L., & Maibach, E. W. (2018). The potential role of actively open-minded thinking in preventing motivated reasoning about controversial science. Journal of Environmental Psychology, 57, 17–24. Taber, C. S., & Lodge, M. (2006). Motivated skepticism in the evaluation of political beliefs. American Journal of Political Science, 50, 755–769. Taibbi, M. (2021, May 24). “Fact-Checking” takes another beating. Substack. https://taibbi.substack.com/p/fact-checking-takes-another-beating Tappin, B. M., Pennycook, G., & Rand, D. G. (2020). Thinking clearly about causal inferences of politically motivated reasoning. Current Opinion in Behavioral Sciences, 34, 81–87. Teovanović, P., Lukić, P., Zupan, Z., Lazić, A., Ninković, M., & Žeželj, I. (2021). Irrational beliefs differentially predict adherence to guidelines and pseudoscientific practices during the COVID-19 pandemic. Applied Cognitive Psychology, 35(2), 486–496. Tetlock, P. E. (1986). A value pluralism model of ideological reasoning. Journal of Personality and Social Psychology, 50, 819–827. Tetlock, P. E. (2002). Social functionalist frameworks for judgment and choice: Intuitive politicians, theologians, and prosecutors. Psychological Review, 109, 451–471. Thompson, A. (2019, December 19). The university’s new loyalty oath: Required ‘diversity and inclusion’ statements amount to a political litmus test for hiring. Wall Street Journal. https://www.wsj.com/articles/the-universitys-new-loyaltyoath-11576799749 Tierney, J. (2021, May 17). This article is “partly false” City Journal. https://www. city-journal.org/facebook-and-its-fact-checkers-spread-misinformation Toner, K., Leary, M. R., Asher, M. W., & Jongman-Sereno, K. P. (2013). Feeling superior is a bipartisan issue: Extremity (not direction) of political views predicts perceived belief superiority. Psychological Science, 24, 2454–2462. Toplak, M. E., & Stanovich, K. E. (2003). Associations between myside bias on an informal reasoning task and amount of post-secondary education. Applied Cognitive Psychology, 17, 851–860. Tufekci, Z. (2021, May 25). Checking facts even if one can’t: This isn’t any way to fight “misinformation.” Substack. https://www.theinsight.org/p/checking-factseven-if-one-cant Tversky, A., & Kahneman, D. (1974). Judgment under uncertainty: Heuristics and biases. Science, 185, 1124–1131. Uscinski, J. E., & Butler, R. W. (2013). The epistemology of fact-checking. Critical Review, 25(2), 162–180. Van Bavel, J. J., & Pereira, A. (2018). The partisan brain: An identity-based model of political belief. Trends in Cognitive Sciences, 22(3), 213–224. Van Boven, L., Ramos, J., Montal-Rosenberg, R., Kogut, T., Sherman, D. K., & Slovic, P. (2019). It depends: Partisan evaluation of conditional probability importance. Cognition, 188, 51–63. Washburn, A. N., & Skitka, L. J. (2018). Science denial across the political divide: Liberals and conservatives are similarly motivated to deny attitudeinconsistent science. Social Psychological and Personality Science, 9, 972–980. West, R. F., Meserve, R. J., & Stanovich, K. E. (2012). Cognitive sophistication does not attenuate the bias blind spot. Journal of Personality and Social Psychology, 103, 506–519.

194  Keith E. Stanovich Westen, D., Blagov, P., Kilts, C., & Hamann, S. (2006). Neural bases of motivated reasoning: An fMRI study of emotional constraints on partisan political judgment in the 2004 US Presidential Election. Journal of Cognitive Neuroscience, 18, 1947–1958. Wright, J., Goldberg, Z., Cheung, I., & Esses, V. (2021). Clarifying the meaning of symbolic racism. PsyArXiv: https://psyarxiv.com/5et3b/ Yudkin, D., Hawkins, S., & Dixon, T. (2019). The perception gap: How false impressions are pulling Americans apart. More in Common. https://psyarxiv. com/r3h5q/ Zigerell, L. J. (2015). Distinguishing racism from ideology: A methodological inquiry. Political Research Quarterly, 68(3), 521–536. doi:10.1177/106591 2915586631

10 Listening for Epistemic Community Hanna Kiri Gunn

10.1 Introduction Of the various headline-worthy crises we presently face, the epistemic crisis might be construed as a Cardinal Crisis interweaved through all the rest. Under this banner, one can find a broad concern for how we have lost, or are losing, the ability to collectively engage in meaningful and productive deliberations on issues that affect us all. One more specific potential loss is the very success of democratic decision-making. Such success depends on the nature and quality of communication within democratic societies. Quality communication, in turn, depends on both the quality of speech and the quality of listening. And while much attention in social epistemology and philosophy of language has been paid to the ways others may silence us, little attention has been given to our duties to listen to one another.1 In this chapter, I argue that as members of an epistemic community, we are obliged to develop listening competencies in ourselves and others. One of the detriments of our current political climate is a failure to meet these obligations. For instance, widespread polarization and the phenomenon of “stupidification” lead to and exemplify listening failures. I propose that the inability or unwillingness to listen undermines our capacity to generate and maintain the kind of healthy epistemic community that we desire and is arguably required for the sustainability of a democratic society. Our epistemic crisis, thus, is not just a crisis of democracy but a skillful or competent epistemic community. To cultivate a healthy epistemic community, we ought to recognize and act on social obligations to foster the capacities needed for good listening. A solution to the present epistemic crisis must incorporate some means for promoting the skill of listening and the competencies that underlie that skill. Crucially, therefore, the obligations to listen do not stem from purely epistemic goods – that is, the values of effective belief formation and deliberation – but rather from the distinctly social-epistemic values of the epistemic agency and healthy epistemic community. DOI: 10.4324/9781003311003-13

196  Hanna Kiri Gunn To be clear, my goal here is not to specify, in detail, our obligations around listening practices. Instead, the goal is to argue that we have such duties and that future work in social epistemology ought to progress in trying to spell these out. Further, as I highlight, many theorists working in related areas – for example, those working on intellectual humility and corresponding virtues and politically oriented philosophers of education – are already implicitly committed to the value of listening, especially within the context of a democratic society.

10.2  Polarization, Stupidification, and Epistemic Exclusion The epistemological crisis has been interpreted as a crisis of epistemic and communicative dimensions with significant consequences on liberal democratic political systems. One undertheorized area, however, remains that of the normative aspects of listening and its role in maintaining healthy epistemic communities. This first section aims to motivate the idea that listening failures are essential to the epistemological crisis. The intuitive idea is that a range of polarization effects2 indicate that there are widespread epistemic and communicative failures that we can understand, in part, as failures of listening competencies. A crucial part of this failure is an inability to overcome animosity in the face of disagreement. Animosity, as it appears, clogs up ears just as well as it does empathy. As an attempt at generalization, polarization is about doxastic changes in individuals or groups that occur either over time intervals or because of an intervention, many of which are taken to be (though need not be) unjustified changes in belief. Polarization effects take a range of potential subjects, including but not limited to the views of an individual, a group’s views across time, and the belief set of significant collectives like the political ideology of political parties. Polarization research has become especially salient for those interested in making sense of the epistemological crisis as it intersects with political partisanship. One reason for this, as I understand the connection, involves the correlation between polarization effects and partisan antipathy.3 The substance of this antagonism is notably both doxastic and affective. Doxastically, the kind of partisan antipathy regularly portrayed in the context of polarization research results in group-think, unwarranted deference to the beliefs of one’s partisan allies, and the downright dismissal of what is believed by one’s partisan peers. As an affective phenomenon, partisan antipathy can be seen in survey results in the United States, indicating that, over the past three decades, in particular, there has been a sharp rise in perceptions of partisan rivals as evil, immoral, and even unsuitable for marrying one’s children.4 Polarized partisans, in short, do not seem to desire much to do with one another. This phenomenon is problematic at least because this

Listening for Epistemic Community 197 fracturing of an ideologically divided but still-united democratic community causes individuals to opt out of political and social activities.5 More robust than this, the ingroup dynamics that seem to co-occur with pervasive polarization involve excluding outgroup members as people whose opinions and beliefs are allowed to influence one’s own. A further effect that appears to occur (at least in part) due to polarization is stupidification that devalues knowledge and, by extension, those who possess it. Lisa Heldke (2006) introduces and defines this process of “stupidification.” To stupidify a body of knowledge is to take it to lack several kinds of value, including practical, aesthetic, and moral significance. Stupidification also results in stupid knowing: a stupid knower knows about stupidified things. Heldke’s target is the stupidification of the knowledge of rural people by big-city dwelling metrocentrists. As she explains, the knowledge rural people have “aren’t the sorts of things anyone would want to know, let alone anything anyone would need to know, or would be enriched by knowing” (2006, 152, emphasis in original). Stupid knowing is different from ignorance: it is not because someone lacks the belief that they are a stupid knower. Instead, having that knowledge makes one a stupid knower. This ability to cast others as irrational and stupid because of what they know is an extraordinarily successful method of ostracization. Heldke’s definition of stupidification describes a process by which epistemic control is exercised over others. A change in the perceived value of some knowledge system can make it permissible to ignore, or even spurn, those who believe it. Significantly, a consequence of this process of devaluing knowledge is that it warrants, both epistemically and morally, ignoring and discounting those ideas from serious consideration. For a democratic community, this is antithetical to a fundamental commitment to the equality of ideas, culture, and practice. The critical point, for our purposes, is that part of the harm of stupidification and polarization effects is that they encourage us to become resistant to worldviews that might provide opportunities for reflection and improvement. As Heldke nicely illustrates, a vital feature of the stupid knowing designation is that a stupid knower cannot be a “sophisticated” knower (2006, 161). For this reason, stupidification warrants ignoring them – refusing to listen to them – in our deliberations and conversations, treating what they know not as valuable but, at best, as a curiosity. Now, stupidification and polarization are not the same things, they do not need to coincide, and neither would appear to necessitate the other’s existence. It is also implausible that all kinds of political disagreement entail stupidification. However, in our actual case, there are reasons to think that polarization is causing us to see one another as

198  Hanna Kiri Gunn stupid knowers due to the entrenchment of partisan attitudes about the political “other.” The relationship between stupidification and polarization that is of interest here concerns this “othering” effect that we can call “epistemic estrangement.” By epistemic estrangement, I refer to a process where individuals within the same epistemic community gradually become alienated from other community members due to failing to value, respect, and understand one another’s beliefs. These may be attitudes about particular policies, what constitutes the good life, or differing perspectives about risks and threats faced by the community. Epistemic estrangement can occur because of polarization effects, such as spread or ideological polarization, and it may occur because of stupidification. The motivation for calling this consequence “epistemic estrangement” is that it highlights the social-epistemic implications for the community. One of the significant barriers facing public discourse in democratic societies is not merely that people disagree about matters of importance. Indeed, this is very often a political or epistemic community’s virtue. It is, instead, how disagreement manifests in the community and how it enables or disables creative political decision-making. What appears to be taking place as a side effect of a range of polarization effects is an affective shift in the perception of those we are meant to treat as political equals. Stupidification predicts such an outcome. If one is a stupid knower, one is morally and epistemically deficient. It is hard not to see stupidification as almost a boiler-plate political strategy in 2022. I have written previously about what I call the “incredulous reaction,” a mixture of surprise and concern at other people for failing to adhere to epistemic norms like trusting perceptual evidence, for example, from photographs about crowd sizes (Gunn 2021a). Sometimes other people’s epistemic choices or commitments strike us as so misguided that they must be wilful, thus the desire to punish those who seem to flout good epistemic conduct. What I have called the “norm of exclusion” validates the incredulous reaction: this norm warrants excluding the testimony of those who one takes to be irrational. But this is problematic if we treat it as indefeasible – if we take epistemic principles that aim at truth to trump all other values that seek different epistemic goods. The risk of a norm that recommends not listening to those one takes to be irrational is that it can establish a default attitude of exclusion from epistemic – and thus, political – life for people we assume to be unreasonable. Research into polarization and partisan divisions reveals that many partisans believe their differently partisan peers are not trustworthy, reliable, or rational others – they are evil and wildly misguided about the facts. Such a norm rejects treating others with the respect and tolerance that democratic commitment entails: our default setting ought to be treating others as reasonable equals.

Listening for Epistemic Community 199

10.3 Epistemic Community, Epistemic Agency, and Listening What norms should guide citizens of democratic societies amid what many consider an “epistemic crisis”? And how should we conceive of the crisis? We have come to refer to a cluster of epistemic and communicative ailments as a crisis by analogy to the climate crisis. (It is striking how rich climate metaphors and analogies have been for contemporary social epistemologists.) Our epistemic environment, by analogy, is in a state of emergency – our “canaries in the coal mine” range from outlandish theories about space lasers igniting wildfires (Branson-Potts et al. 2021) to butterfly sanctuaries serving as sites for human smuggling (BBC 2022). However, the framing of the crisis part of “climate crisis” has shifted: we increasingly hear more about climate “tipping points” and spiraling, as-yet-uncontrollable “domino effects” that would put an abrupt end to our attempts to get the upper hand and stabilize climate systems. These are distinct views: one proposes we are in the midst of an ongoing emergency, and the other suggests that the ongoing crisis is a dire warm-up before we hit a point of no return. We must consider that the epistemological crisis may merely pose a “tipping point.” We may be nearing the edge of a cultural cliff in our epistemic environment from which we cannot climb back up. Put less metaphorically, without interventions that can meaningfully push back against the ways that we are actively trained to be partisans incapable of meaningful, respectful, and edifying public discourse, we may lose out on the goods that healthy public discourse promises altogether. How many generations of failing to cultivate the dispositions, competencies, and attitudes that support a healthy epistemic community does it take until the prospect of a healthy epistemic community is all but gone? What are the analogous tipping-point risks we arguably face, then? From the perspective of someone who typically thinks in the social epistemology space about issues of epistemic agency as understood above, the entrenchment of the range of “polarization” harms is a particularly salient tipping-point risk. If we are in such an epistemic crisis, either an ongoing emergency or facing some epistemic tipping point(s), what should we do to either halt or prevent it? Many authors writing on the epistemic crisis have focused on our doxastic practices: how are people forming beliefs? Are they reading the viral stories or merely sharing them online? Do people know how to verify a source, to engage in lateral reading? Listening has simply not had as much theoretical attention as a part-solution to the epistemic crisis as protecting speech has. It is, of course, imperative to safeguard speech. But speech without an audience is empty – it lacks the force to move anyone or thing. Moreover, an audience who perceives you as a stupid knower is at least as helpful as not having one and potentially much more dangerous.

200  Hanna Kiri Gunn Speech without an audience fails in another way that overlaps with speaking to an audience who takes you as a stupid knower. Alongside a lack of uptake for what has been said, the speaker fails to get any of the other goods from taking the floor: being understood, being shown respect by being attended to, being included in an ongoing conversation where others have been allowed their say or the promise of an informative discussion with a sincere interlocutor. Our epistemic agency is deeply intertwined with listening: one’s capacity and willingness to listen well to others and others’ willingness and capacity to listen to oneself. To have the ability to speak to one’s opinions, values, or knowledge without being heard is not to be included in epistemic life. To talk without being listened to in a democratic society is to be sequestered and ineffectual. Indeed, it is not to be treated as an equal. It is much easier to understand the value of listening if we conceive of it through the lens of social-epistemic goods and within the context of an epistemic community. An underappreciated social-epistemic value is that of the health of the epistemic community itself. We should understand this in terms of the epistemic agency of the community’s members. The role of listening, then, is critical for the health of the epistemic community, given its significance for the epistemic agency of its members. In the next section, I argue that much analytic epistemic theorizing is implicitly committed to this model of understanding epistemic communities and listening duties. Still, explicit and intentional theorizing committed to the value of the health of the epistemic community is needed. Let me now say more about what I mean by “epistemic community” and “social-epistemic agency.” I understand the epistemic environment as a complex system constituted by, at a minimum, a combination of actual people (their attitudes, dispositions, skills), social networks, and technology. An epistemic community is a group of people embedded in an epistemic environment and united at the level of joint commitment to some fundamental epistemic values or aims (Gunn 2020). All political societies are also, in a foundational sense, epistemic ones but not the reverse. The epistemic character of a political society is evident in liberal democratic systems where democratic processes revolve around testimony, persuasion, and collective deliberation, each of which is a kind of social-epistemic practice. The health of the epistemic community can be seen as a part of the common good of political society. By attending to and promoting the health of the epistemic community, we meet what I propose is an obligation that promotes a common interest in both epistemic and political well-being. At the risk of asserting a controversial premise, being a community member is not a static role: communities only exist because their members maintain them. The particular character of a community, its sustainability, and its resistance to radical change or welcoming of it depend on the nature of that specific community. Or so I assume for this discussion.

Listening for Epistemic Community 201 Epistemic communities, then, are groups of people brought together owing to a “joint commitment” in Margaret Gilbert’s sense (Gilbert 2013; Gunn 2020).6 The function of joint commitment is that it ratchets up the normative stakes by making us accountable to other agents who share the responsibility. If we want not to be bound by it, we must ask to be released or violate our commitment. When we think about epistemic communities, we think about groups of people who jointly commit epistemically. Epistemic communities enable or disable the epistemic agency of their members. They do this because epistemic agency, as I understand it, is a competency developed through training by and with other people and letting one participate in epistemic life more-or-less well. Put differently, the epistemic agency is a relational skill or a relational ability. It is relational in two ways. First, it is causally relational in that competencies have to be acquired – one needs to learn from others how to do things like responsibly form beliefs, do research, and effectively teach others. Second, it is constitutively relational because our ability to bring about changes in the epistemic environment depends on other people too. For example, whether or not people are willing to learn from me in a classroom determines whether I can occupy the epistemic role of a teacher. In general, to be an epistemic agent is to be the kind of creature that can bring about epistemic changes in their epistemic environment, which requires being able to make epistemic changes in oneself and others.7 Epistemic agency, or the ability to meaningfully participate in epistemic life, is (like other sorts of agency) good for human beings. Listening is an essential part of epistemic agency. We are often able to bring about epistemic changes by being listened to. It is through listening to others that we can improve our worldview, to subject our beliefs and convictions to reasonable challenges. It is essential, then, that individuals can be listened to by competent others and have the same ability to support the other members of their epistemic community. An epistemic crisis threatens the epistemic environment by affecting any constitutive parts: via social networks, via new technology, via undermining attempts to exercise epistemic agency or develop some competency it requires. It is probably reasonably clear, then, why threats to epistemic agency register as threats to what, simply by analogy, we can call “political agency.” If one cannot exercise epistemic agency, say, by having their say about whether the government should build a bridge, they are excluded from the political project too.

10.4 Civility Approaches to Public Discourse and Obligations to Listen In this way of thinking about epistemic communities, one of the opportunities in times of epistemic crisis is that it reveals that joint commitments crop up functionally in practice, whether in theory or not. When

202  Hanna Kiri Gunn we are considering solutions for the epistemic situation, this framing of epistemic communities as collectives grounded in joint commitments helps to keep in focus the (social-) fact that we are responsible, individually and collectively (as the distribution of social power allows), for the current and future character of the epistemic community. There are many takes on the causes of the epistemic crisis. A common but arguably naive one is that the internet is responsible for the current state of polarization effects at the individual and collective levels. However, consider the following passage – which will be contextualized below – commenting on the perceived state of public discourse in the United States, First, the claim is made that our political and social discourse is highly polarized, uncivil, and unproductive. Second, our political and social leaders are more interested in following public opinion than imagining bold, new solutions to old problems. Third, our communal life is increasingly fragmented into increasingly smaller and more highly politicized units. Fourth, a breakdown in public mores, tolerance, and self-restraint is lamented, and a blurring of the boundaries between private and public discourse is felt to influence dramatically and negatively every aspect of our public culture. This diagnosis of the state of public discourse is almost two decades old. It comes from Judith Rodin and Stephen P. Steinberg’s (2003) introductory chapter, “Incivility and Public Discourse,” of their edited book, Public Discourse in America. Despite the proliferation of arguments that the internet – and social media in particular – have brought about the epistemic crisis, this quote comes from a pre-social media age. Indeed, 2003 was the year both 4chan and The Pirate Bay first went live, the year that Second Life was released, and the year the iTunes store opened. The moral of this is not to propose that the internet and social media have no impact on the present state of public discourse or our epistemic lives. Instead, the goal is to contextualize the current situation in a slightly longer history. This history helps us understand how public discourse came to be the way it is and develop a sense of recurrent problems. Given that many of the contributors to Rodin and Steinberg’s (2003) book are reflecting on research from the 1990s, we can assert with some confidence that these barriers to public discourse have been blocking the path for at least three decades – with the internet arriving at a time to exaggerate and make them worse. Recent theorizing about public discourse has continued to share the framing of civility and incivility. However, there has been a sharp rise in treating the problem as an issue of epistemic responsibility. More specifically, often a case of the virtues and vices of epistemic agents or the dispositions of epistemic agents. As illustrated below, many existing theories diagnosing epistemically harmful propensities like intellectual arrogance and positive accounts of virtues like intellectual humility

Listening for Epistemic Community 203 entail normative claims about listening skills. Despite this, and while many may mention listening and its centrality to epistemic life, there are as of yet few accounts of listening in its own right, a point that has not gone unnoticed by others also calling for further theoretical attention to listening for social justice (e.g., Beausoleil 2021). One influential view in the intellectual humility literature is offered by Whitcomb et al. (2017). They propose that intellectual humility involves developing a particular dispositional profile to properly understand one’s epistemic strengths and weaknesses. Whitcomb et al. are engaged in an interdisciplinary research project involving psychological and philosophical analysis of intellectual humility. They, therefore, include predictions about what possessing the virtue of intellectual humility will entail. And prediction eleven is that the intellectually humble have an increased “propensity to consider alternative ideas, to listen to the views of others, and to spend more time trying to understand someone with whom he disagrees” (2017, 13). Balancing some notion of intellectual deference and intellectual arrogance is present in most accounts of intellectual humility. Whitcomb et al. are committed, then, to the idea that becoming intellectually humble is valuable and that one of its consequences is to modify our ability to listen to one another. Intellectual humility is proposed to be a virtue for managing intellectual confidence by Ian James Kidd (2016). Humility “requires a variety of practices of confidence calibration,” Kidd writes, and these include “listening, objecting, querying” as processes of learning how to manage one’s confidence (2016, 397). Confidence calibration is required for both managing under- and overconfidence. One of the outcomes of this is that the humble person is “willing to argue with and listen to others” because they recognize that doing so maintains a virtuous degree of confidence and affords others the respect that comes from contributing to others’ intellectual life (2016, 401). Alessandra Tanesini (2016) explicitly proposes that audiences have responsibilities toward speakers lest they display the vice of “haughtiness.” In particular, “the haughty tend not to listen to objections or not to take them as seriously as they deserve to be taken” (2016, 81). Michael P. Lynch (2018) argues that “epistemic arrogance” is the attitude of thinking that one has nothing to learn from others. Lynch’s motivating examples for this account of epistemic arrogance include the Obnoxious Uncle (who refuses to listen to others on topics he takes himself to be an expert on), the Dogmatic Listener (who is not listening but merely waiting to speak), and the Mansplainer (who may learn from listening to others but is incapable of seeing this as a result of listening to them). Tanesini and Lynch are both committed to the idea that our dispositions to listen are a significant part of one’s epistemic agency and that it is incumbent on us to develop such tendencies toward, for example, humility and away from arrogance.

204  Hanna Kiri Gunn In Know-It-All Society (2019), Lynch writes: “Democracies need their citizens to have convictions, for an apathetic electorate is not an electorate at all. Yet democracies also need their citizens to listen to one another’s convictions, to engage in political give-and-take” (14). This quote highlights an important issue that has not yet been discussed directly: listening duties should not be misunderstood as duties to believe or support what is heard. For example, the harm of the polarization-stupidification dynamic is not that there is a failure to “come together” in the sense of everyone “agreeing that p is true.” The harm, instead, is that in failing to listen to someone else, we undermine their epistemic agency by preemptively and unjustifiably refusing to let their testimony have any bearing on our point of view. This, of course, can damage one’s epistemic agency by failing to allow for the improvement of one’s set of beliefs. Up to this point, I have defended the claim that we have listening duties. Still, by the end of the remaining discussion, I will also have made some gestures toward how we might need to do so to support a healthy epistemic community. I will argue that, while we can productively make sense of duties to listen to others to bring about particular moral ends like respect and epistemic ends like understanding, we ought to focus first on what duties there are to develop and maintain listening competence as a proper part of one’s epistemic agency. There is, I think, a reasonable proposal that for the collective health of the epistemic community, we need to invest in the joint project of edifying epistemic agency for individuals and that such a project requires thinking seriously about how to cultivate the kind of listening competence that would resist harms like intellectual arrogance, intellectual deference, and, of course, the stupidification of others. To do so, I begin the next section by discussing a potential objection that listening to others in moments of polarizing epistemic crisis risks merely ingraining polarization and its consequences.

10.5 A Virtue of Reciprocity Requires Investing in Listening Obligations In Sustaining Democracy: What We Owe to the Other Side (2021), Robert B. Talisse presents the “democrat’s dilemma.” The democrat’s dilemma arises because citizens of democracies face two requirements: first, to pursue justice and, second, to acknowledge one another as political equals. These requirements conflict when we do not perceive other members of our political society as equals because of their objectionable views. The democrat’s dilemma is that it can seem like treating other community members as equals fails to pursue justice (because of their objectionable opinions that the government could make into policy). Still, we undermine the second requirement by failing to treat them as political equals (and thus giving their views a hearing).

Listening for Epistemic Community 205 The positive proposal that Talisse offers includes a discussion of three virtues of democratic citizenship: public-mindedness, reciprocity, and transparency (2021, 30–33). The virtue of reciprocity entails that we owe one another “the disposition to formulate one’s political views in a way that reflects a willingness not merely to allow others to speak but to give them a hearing. We manifest the virtue of reciprocity when the views and concerns of others contribute to our own thinking” (2021, 32, emphasis added). Talisse argues that we do need to be able to remain open to our views being improved by others, thus resisting epistemic arrogance or dogmatism, but that does not require abandoning conviction in one’s beliefs, thus giving over to intellectual deference. The aim of giving others a hearing is to do one’s due diligence in forming, updating, and modifying one’s views. He argues, then, that there are distinct harms from failing to listen appropriately in an epistemic crisis constituted by polarization effects: listening poorly, listening too much, and listening too much to specific others – including our political allies – exacerbate polarization effects. Thus, he argues, we ought to resist arguments that we must “love our political enemies” to resist polarization effects.8 It strikes me as fair to be skeptical about proposals that we merely need to put aside our differences, come together, and talk it out until we see one another as equals and can return to a more civil political culture. In part, this is because such desires are nostalgic and thus crave a past that did not exist – placing the current moment in a long history of research into public discourse helps to reveal this. More to the point, though, listening duties need not be about responsibilities to engage in attending to each other (and that risks “overdoing” democracy (Talisse 2019)). Listening duties can also be about how we listen when we do. Further, once we conceive of listening as one competence of epistemic agency, listening duties may aim at edifying this competence. We need to consider what skills underlie the virtue of reciprocity and what competencies individuals need to have mastery of to become open-minded, intellectually humble, and responsibly selective listeners. My main contention is this. To “deliver the goods” via the virtue of reciprocity, we must already have developed into the hearing epistemic agents that the virtue demands. Competence with listening in circumstances of disagreement, perhaps inimical conflict, requires work. Being able to deliver on the virtue of citizenship presupposes that one is a particular kind of epistemic agent, but this requires active investment in developing one’s epistemic competencies. In order to get reciprocity, we need to invest in the development of epistemic agency – in particular, in ways that encourage listening that treats others as epistemic and political peers.

206  Hanna Kiri Gunn

10.6 Investing in Listening for the Epistemic Community: Education and Skill Development I have proposed that flourishing epistemic communities support the development, maintenance, and exercise of the epistemic agency of their members. Epistemic crises can threaten epistemic agency in various ways, depending on the situation’s causes. I have committed to the idea that we should take the health of our epistemic community to be an essential epistemic good (which also has tremendous instrumental value) and, in the political domain, to be a part of the common good. The role of listening duties I have proposed is that they are essential to enabling epistemic agency and a healthy epistemic community. I agree with Aristotle’s general framework of good listening developed by Suzanne Rice (2011). According to Rice’s analysis, Aristotle would not have taken there to be essential listening virtues. Instead, what is required to listen virtuously will depend on the specific instance of listening – listening in teaching is vitally different from listening in politics and listening to a baby’s cry. Hence, the proposal that we should first consider what it takes to be a competent listener and what our obligations may be, collectively, for developing this competence in ourselves and others. To some extent, and given the practical end that such collective listening obligations seek to support, it strikes one that empirical research into listening skills may be a crucial descriptive constraint on normative accounts of listening. Still, we must work out such theoretical questions in the longer run. But whatever the correct account of listening ends up being, the critical point, for our purposes, is that reflection on our obligations to listen points to our epistemic interdependence, and our broader obligations to promote the bases for a healthy epistemic community. While it is often not the central focus of much philosophical theorizing, we can find this line of thinking in the work of various influential epistemologists, feminist philosophers, philosophers of education, and educational theorists. W. K. Clifford, for instance, argues that our duty to believe following the epistemic norms is crucial for the cross-generational health of the epistemic community. Metaphorically, he describes our epistemic practices as, “an heirloom which every succeeding generation inherits as a precious deposit and a sacred trust to be handed on to the next one, not unchanged but enlarged and purified…An awful privilege, and an awful responsibility, that we should help to create the world in which posterity will live” (1999, 74). The title of this section is adapted from a similar claim made by Lorraine Code (1987) in a chapter on epistemic community, that to be a participant in a society is to be at the same time a “conserver and modifier of practices” and that “Practices can be created and preserved only by their practitioners; they are neither self-generating nor self-sustaining” (193).

Listening for Epistemic Community 207 Of course, promoting good epistemic practices and community is not just a matter of following personal obligations to listen or reason well. There are also collective obligations to ensure that our primary epistemic institutions – our educational system, for instance – properly support a healthy epistemic community. Thus, for political pedagogical theorists like Paulo-Freire and bell hooks, there is no clear division between our epistemic lives more generally and our political lives. Concerning the development of the competencies that underlie political life, educational practice is critical. Further, both hooks and Freire directly confront issues of political marginalization and thus aim to present alternative models of education that aim to create a genuinely democratic social structure in the classroom. In an exchange about creating community in classrooms, bell hooks and philosopher Ron Scapp converge on the idea that teachers and students are all “subjects in history” who cannot drop their identities off at the door. hooks argues that “[Teachers] must intervene to alter the existing pedagogical structure and to teach students how to listen, how to hear one another” (1994, 150, emphasis in original). Scapp develops this idea by proposing that it “doesn’t mean to listen uncritically or that class can be open so that anything someone else says is taken as true, but it means taking seriously what someone says” (ibid). The connection between pedagogy and political culture is central to much of John Dewey’s writing on education. Of particular importance for this discussion is that Dewey presented a theory of listening, distinguishing between “straight-line” or “one-way” listening and “transactional listening-in-conversation” (Waks 2011). We can grasp the difference by contrasting the kind of listening required in a lecture where a professor expects students to merely transcribe the professor’s thoughts on a topic and a dialogue-based seminar where students and teachers collectively work to understand the issue together. As Waks explains, Dewey took transactional listening to enable “cooperative friendship” that itself “generates both the means and the end of democracy as a form of social life” (2011, 198). Thus, listening practices for Dewey feed directly not only into our skills for engaging in dialogue but also affect the ways that we relate, morally, to one another. Similar notions are present in the educational theory of J. F. Herbart, a contemporary of Dewey. Andrea English (2011) argues that educating for critical listening was an essential role for teachers in Herbart’s theory. In modeling critical listening practices, teachers can help students cultivate what Herbart called an “inner censor,” an intellectually humbling moral disposition. I have suggested that quality listening supports the health of an epistemic community. What might this look like? We might find one such example in the Nordic models of early childhood education. In the early 1990s, all Nordic countries committed to including a democratic structure

208  Hanna Kiri Gunn in early childhood education that emphasizes the role of the child as an active participant in democratic processes at school (Broström 2019). This model of early childhood education is committed to the idea of the “child’s perspective” that entails perceiving children as contributors to their democratic society.9,10 The Nordic models employ the concept of “Bildung,” which denotes active learning intended to develop moral and epistemic character, including the ability to listen and argue with others in a co-determining fashion. Co-determination in this context means that children are treated as having a voice that will be included in shared decision-making and that they ought to extend the same right to others (children or teachers). Early childhood educators working within these models are encouraged to perceive children as competent participants who bring their values and opinions to school. Part of the educator’s responsibility is to allow the children to explore societal issues from their own perspectives, supporting their exploration in such a way that it develops their acquisition of both knowledge and norms. As Broström (2019) explains, the democratic-based curriculum requires preschool teachers to “listen to children and challenge them to reflect and express their thoughts and actions and take the initiative themselves.” With such an approach, teachers can model skillful listening and work to establish a norm of listening inside and outside the classroom. The core ideas about the connection between the skills developed in educational settings and their consequences on political culture are correct. Our political community is not separable but depends on our broader epistemic community. The current epistemic crisis, often framed as a political problem, is more fundamentally a crisis of the epistemic community. A healthy epistemic community requires the promotion of epistemic agency. Epistemic agency, in turn, requires the ability to speak and that others listen. Polarization and stupidification often lead to epistemic marginalization, refusals to listen, and thus the frustration of epistemic agency. If we want to promote the health of the epistemic community, we need to understand more fully what our obligations to listen are. But we also cannot ignore the practices and institutions that foster the competencies necessary for listening well and that we, as members of the epistemic community, must design and support to secure ­stability for the democratic structure.

Notes

1. I have in mind work on illocutionary silencing (Langton 1993) alongside rival theories for this phenomenon, including, for example, perlocutionary frustration (Bird 2002) and discursive injustice (Kukla 2014). Relatedly, a broad range of work analyzes the impact of prejudices in our epistemic and communicative lives that focus on speakers. These include, for example, how prejudice can cause systematic failures to perceive speakers’

Listening for Epistemic Community 209







­credibility accurately (Fricker 2007) and how prejudice can cause frequent failures of misinterpretation and bring about self-censorship in speakers (Dotson 2011; Medina 2012). 2. See Gunn (2021b) for a discussion of polarization, including the disambiguation of filter bubbles and echo chambers. Many other measurable effects are also recognized as polarization effects in the broader social science literature. Bramson et al. (2016), for example, explain nine kinds. These include, for example, “spread,” which refers to the range of attitudes in a population on a specific topic. To say that a population is polarized in the sense of spread implies a considerable distance between the most extreme attitudes at the “poles” within the group. 3. See Pew Research Center (2022). 4. McBrayer (2021) provides an extended discussion on group conformity biases and post-truth political culture, including research on partisan antipathy. 5. See Talisse (2019) for an extended analysis of how the partisan political culture in the United States undermines “civic friendship.” 6. A joint commitment involves all parties personally committing to some action (“I will do x”) and committing to others to undertake that action together (“I will do x with you”). 7. These two metaphysical commitments for epistemic agency make this a version of the relational theories of autonomy (see Mackenzie & Stoljar 2000, 7–8). 8. See his discussion in Chapter 3, “The Polarization Dynamic,” and §3.1 on rejecting the idea we must love our political enemies. 9. In Denmark, for example, early childhood education policies require that preschool children have “co-determination, joint responsibility, understanding, and experience with democracy. Furthermore, preschool should contribute to developing children’s autonomy and abilities to participate in binding social communities” (Broström 2019). 10. To provide a non-Nordic example, one German preschool has taken on an explicitly deliberative-democratic model in their school that exemplifies many of the commitments of the Nordic model. Dolli Einstein Haus characterizes itself as a “democracy nursery” the children at this center co-­ created a constitution specifying their rights at the center, and all decisions about naps and meal times are decided through deliberative, democratic means (Oltermann 2017). This strategy directly targets the children’s ability to listen and be listened to, developing in them an expectation that this is how collective decisions are made.

References BBC. 2022. “Texas butterfly center closes after QAnon threats.” BBC, https:// www.bbc.com/news/world-us-canada-60254327. Accessed: 2/6/2022. Beausoleil, Emily. 2021. “Listening obliquely: Listening as norm and strategy for structural justice.” Contemporary Political Theory, 20(1): 23–47. Bird, Alexander. 2002. “Illocutionary silencing.” Pacific Philosophical Quarterly, 83(1): 1–15. Bramson, Aaron, Grim, Patrick, Singer, Daniel J., Fisher, Steven, Berger, William, Sack, Graham, and Flocken, Carissa. 2016. “Disambiguation of social polarization concepts and measures.” The Journal of Mathematical Sociology, 40: 80–111.

210  Hanna Kiri Gunn Branson-Potts, Hailey, Serna, Joseph, and Alejandra Reyes-Velarde. 2021. “How wildfires became ripe areas for right-wing conspiracy theories.” Los Angeles Times, https://www.latimes.com/california/story/2021-01-30/a-space-laser-did-it-gop-congressman-had-out-there-theory-on-deadly-california-wildfire. Accessed: 2/6/2022. Broström, Stig. 2019. “Te Whāriki and the Nordic Model: Comments on Te Whāriki from a Norwegian and Danish Perspective.” In Alexandra C. Gunn and Joce Nuttall (eds.), Weaving Te Whāriki: Aotearoa New Zealand’s Early Childhood Curriculum Framework in Theory and Practice. 3rd edition. Wellington, New Zealand: NZCER Press. Clifford, William K. 1999. “The Ethics of Belief.” In Tim Madigan (ed.), The Ethics of Belief and Other Essays. Prometheus Books: New York, United States. pp. 70–96. Code, Lorraine. 1987. Epistemic Responsibility. Hanover (NH): University Press of New England. Dotson, Kristie. 2011. “Tracking epistemic violence, tracking practices of silencing.” Hypatia, 26(2): 236–257. English, Andrea. 2011. “Critical listening and the dialogic aspect of moral education: J.F. Herbart’s concept of the teacher as moral guide.” Educational Theory, 61(2): 171–189. Fricker, M. (2007). Epistemic Injustice: Power and the Ethics of Knowing. Oxford: Oxford University Press. Gilbert, Margaret. 2013. Joint Commitment: How We Make the Social World. New York (NY): Oxford University Press. Gunn, Hanna Kiri. 2020. “How should we build epistemic community?” Journal of Speculative Philosophy, 34(4): 561–581. Gunn, Hanna Kiri. 2021a. “Exclusion and epistemic community.” Revue Internationale De Philosophie, 297(3): 73–96. Gunn, Hanna Kiri. 2021b. “Filter Bubbles, Echo Chambers, Online Communities.” In Michael Hannon and Jeroen de Ridder (eds.), The Routledge Handbook of Political Epistemology. New York (NY): Routledge. Heldke, Lisa. 2006. “Farming made her stupid”. Hypatia, 21(3): 151–165. hooks, bell. 1994. Teaching to Transgress: Education as the Practice of Freedom. New York (NY): Routledge. Kidd, Ian James. 2016. “Intellectual humility, confidence, and argumentation.” Topoi, 35(2): 395–402. Kukla, Rebecca. 2014. “Performative force, convention, and discursive injustice.” Hypatia, 29(2): 440–457. Langton, Rae. 1993. “Speech acts and unspeakable acts.” Philosophy and Public Affairs, 22(4): 293–330. Lynch, Michael Patrick. 2018. “Arrogance, truth and public discourse.” Episteme, 15(3): 283–296. Lynch, Michael Patrick. 2019. Know-it-All Society: Truth and Arrogance in Political Culture. New York (NY): WW Norton. Mackenzie, Catriona, and Natalie Stoljar. 2000. Relational Autonomy: Feminist Perspectives on Autonomy, Agency, and the Social Self. New York (NY): Oxford University Press. McBrayer, Justin P. (2021). Beyond Fake News. New York (NY): Routledge. Medina, José. (2012). “Hermeneutical injustice and polyphonic contextualism: Social silences and shared hermeneutical responsibilities.” Social Epistemology, 26(2): 201–220.

Listening for Epistemic Community 211 Oltermann, Philip. 2017. “Put to the vote: German nursery where children make the decisions.” The Guardian, https://www.theguardian.com/world/2017/aug/11/ german-nursery-children-make-decisions-vote-dolli-einstein-haus. Accessed: 2/6/2022. Pew Research Center. 2022, August. “As Partisan Hostility Grows, Signs of Frustration with the Two-Party System.” Rice, Suzanne. 2011. “Toward an Aristotelian conception of good listening.” Educational Theory, 61(2): 141–154. Rodin, Judith, and Stephen P. Steinberg. 2003. Public Discourse in America: Conversation and Community in the Twenty-First Century. Philadelphia: University of Pennsylvania Press. Talisse, Robert B. 2019. Overdoing Democracy: Why We Must Put Politics in Its Place. New York (NY): Oxford University Press. Talisse, Robert B. 2021. Sustaining Democracy: What We Owe to the Other Side. New York (NY): Oxford University Press. Tanesini, Alessandra. 2016. “‘Calm down, dear’: Intellectual arrogance, silencing, and ignorance.” Aristotelian Society Supplementary Volume, 90(1): 71–92. Waks, Leonard J. 2011. “John Dewey on listening and friendship in school and society.” Educational Theory, 61(2): 191–206. Whitcomb, Dennis, Battaly, Heather, Baehr, Jason, and Daniel HowardSnyder. 2017. “Intellectual humility: Owning our limitations.” Philosophy and Phenomenological Research, 94(3): 509–539.

Part III

Democratic Realism

11 Sensemaking, Empathy, and Democracy Quassim Cassam

11.1  The Great Paradox In Strangers in Their Own Land, the sociologist Arlie Russell Hochschild introduces readers to a man called Mike Schaff, who exemplifies what she describes as the Great Paradox.1 Schaff was a victim of a vast environmental disaster in Louisiana, the appearance of a sinkhole that stretched over thirty-seven acres and devoured everything in its wake. A lightly regulated drilling company was to blame. Still, Schaff “hailed government deregulation of all sorts, as well as drastic cuts in government spending – including that for environmental protection” (2018: 5). This is the essence of the Great Paradox: “great pollution and great resistance to regulating polluters” (2018: 21). How could victims of environmental pollution not favor ecological protection? How can the poor oppose more government help for the poor? How can a state that was “one of the most vulnerable to volatile weather be a center of climate denial” (2018: 23)? Hochschild spent years in Louisiana, listening to people like Schaff and empathizing with them in search of answers. From Hochschild’s liberal standpoint, and presumably that of many of her readers, the Great Paradox is also a great surprise. Her project is, therefore, an exercise in what others have called sensemaking, “the making of sense” (Weick 1995: 4).2 We find efforts at sensemaking “whenever the current state of the world is perceived to be different from the expected state of the world” (Weick, Sutcliffe & Obstfeld 2005: 414). Schaff’s is one such world. One would expect victims of environmental disasters resulting from weak environmental regulation to favor stronger regulation. This is not what Hochschild found. How is this to be explained? What would a satisfactory explanation look like? Is it necessary to empathize with people like Schaff to understand them? These are among the questions to be addressed below. Many of Hochschild’s subjects voted for Donald Trump in the 2016 US presidential election. Liberals, surprised by his victory, tried to explain it and work out what it meant. This was another exercise in sensemaking. In these cases, there is an event E, and the objective is to make sense of E after the event. Sensemaking is retrospective.3 It is not about predicting DOI: 10.4324/9781003311003-15

216  Quassim Cassam the future but understanding the past. It is not confined to making sense of events, as making sense of an event like Trump’s win requires one to make sense of the attitudes that led many who would not benefit economically from a Trump presidency to vote for him. Because sensemaking is a response to a surprise, and what is surprising to one person might not be surprising to another, there may be different views about when sensemaking is called for. Conservatives who expected Trump to win and people like Schaff to vote for him saw no need for sensemaking. Completely different events and different attitudes elicit their sensemaking. Hochschild sees empathy as the key to sensemaking. Explaining the Great Paradox requires understanding “how life feels to people on the right – that is, the emotion that underlies politics” (2018: xi). To understand people on the right, she had to imagine herself in their shoes, that is, to empathize. To “know others from the inside, to see reality through their eyes,” it is necessary to “cross the empathy wall” (2018: 5). The ultimate objective was not just to better understand her subjects but to see if it was possible to “make common cause on some issues” (2018: xiii). Empathy, as Hochschild understands it, is an antidote to polarization. Empathetic understanding of other people is also seen as vital for democracy.4 However, there are many different accounts of empathy.5 On one view, it is “the activity of imaginatively adopting another person’s perspective in a way that somehow engages the emotions of the one doing the imaginative work” (Bailey 2022: 52). Others represent it as a bloodless exercise in reading the mind of another. The former is sometimes called “emotional empathy,” while the latter is described as “cognitive empathy.”6 It is open to question whether emotional empathy is necessary for sensemaking. On the face of it, it is possible to understand why Hochschild’s subjects see the world as they do without imaginatively adopting their perspective in Bailey’s sense. For example, Marxists may regard the Great Paradox as illustrating the power of ideology or as a compelling illustration of the phenomenon of false consciousness, a mode of consciousness that misrepresents socio-economic reality while also being determined by that reality.7 This socio-structural explanation of the Great Paradox does not require emotional empathy with people like Mike Schaff. Indeed, for all her talk of empathy, it is not obvious that Hochschild’s insights result from emotional empathy with her subjects.8 What explains the appeal of the idea that empathy is required for sensemaking and the key to a healthy democracy? Aside from the hope that emotional empathy might be an antidote to polarization, there is also the suggestion that the objective of sensemaking is personal understanding and that personal understanding requires empathy. On this view, sensemaking involves understanding people in a distinctive manner. Specifically, it involves relating to other people as individuals and engaging with their subjectivity. Because socio-structural responses to the Great Paradox fail to do this, they do not deliver personal understanding or Verstehen,

Sensemaking, Empathy, and Democracy 217 as it is sometimes called, of individuals like Schaff. However, it is unclear that a particular form of empathetic personal understanding is required to resolve the Great Paradox. The fact that some victims of pollution are against greater regulation of polluters reflects their ideology, but we do not need empathy to understand their ideologies or the emotions to which they give rise. We have their words and deeds to go on. The discussion below will proceed as follow: Section 11.2 will argue that emotional empathy has little to contribute to sensemaking in the political domain. Others have written about the barriers to emotional empathy in general, but the political realm is one in which these barriers are especially challenging to overcome. It is easy to exaggerate the role of empathy in listening exercises such as Hochschild’s. On the face of it, it is possible to listen to someone without empathizing with them. Section 11.3 will criticize arguments for the view that empathy is vital for democracy. Section 11.4 will discuss the relative merits of empathy and socio-structural approaches to political understanding. A case will be made for downplaying the role of empathy and avoiding empathy fetishism. There are problems with the notion of false consciousness, but it remains illuminating when explaining the Great Paradox.

11.2  Sensemaking and Empathy We can understand the notion of empathy in many different ways. It will be understood here as made up of two components identified by Olivia Bailey. First, empathizing is a form of imaginative perspective-taking: it “necessarily involves using one’s imagination to ‘transport’ oneself, such that one considers the other’s situation as though one were occupying the other’s position. So, for instance, when we try to imagine how things are for a recent widower empathetically, we might imagine having just lost a spouse or other loved one” (2022: 52). The second feature of empathy is that it is emotionally charged imaginative perspective-taking: In certain critical respects, the emotional experience of the one who empathizes closely resembles the emotional experience of the target of empathy. An admittedly metaphorical but apt way of thinking about how the emotions are implicated in empathy is to conceive of the empathizer as encountering their imaginative recreation of the other’s situation through the same emotional lens as the target of empathy. The widower apprehends his loss through the lens of grief. We as the widower’s empathizers also allow our thoughts to be directed in ways characteristic of grief. The isomorphism between this empathetic experience and the original grief of the widower strongly recommends the conclusion that when we empathize, we do not merely imagine feeling some emotion. Instead, we do not merely imagine that we are feeling some emotion. (Bailey 2022: 53)

218  Quassim Cassam The two components of empathy are separable. Emotionally charged perspective-taking is emotional empathy. Affectless perspective-taking is cognitive empathy.9 It is quite possible to imagine being in the shoes of a recently bereaved widower and understand the grief that that position entails without experiencing the same type of emotion as the widower. In these cases, the widower’s emotion is intelligible but not mirrored by the empathizer. Emotionally charged perspective-taking is essential in many personal relationships, including relationships with close friends and family. Empathizing with a person is a way of engaging with them emotionally. This is much easier to achieve with people with whom one has a close personal relationship than with casual acquaintances or total strangers. Emotional empathy is psychologically demanding, and there are psychological limits to the number of people with whom a person can empathize.10 Empathizing with someone means engaging with their subjectivity. In the widower’s case, one engages with the subjectivity of a unique individual rather than with recently bereaved widowers in general. As Gregory Currie puts it, “we think of empathy as an intimate, feeling-based understanding of another’s inner life” (2011: 82). A person without such an understanding of anyone else’s inner life is seriously impoverished. A person who has, or even claims to have, a feeling-­ based understanding of the inner lives of scores of people is a freak or a charlatan. What is the role of emotional empathy in political sensemaking? When the current state of political reality is different from its expected state, for example, when someone like Trump is elected President, there is sensemaking to be done. Thus, one might ask how so many white working-­ class voters could come out in favor of such an unlikely candidate. This is a question for political science, sociology, and other disciplines but the accounts that these disciplines offer are impersonal. They are not, and cannot be, based on emotionally charged perspective-taking or a feeling-based understanding of the inner lives of millions. It is tempting to think this difficulty can be circumvented by a selective emotional engagement with representative citizens whose perspectives can be generalized. This is a way to think about Hochschild’s investigation. She formed relationships, even friendships, with individuals in Louisiana who exemplified the Great Paradox. By getting into these individuals’ heads, she extracted valuable insights about what would otherwise be an extremely puzzling phenomenon. On reflection, however, it is unclear that she needed emotional empathy; she could have arrived at the same conclusions without empathy for her interlocutors. One of her interlocutors spoke of her aversion to regulation and learning to live without it. She wanted clean air and water, but “sometimes you had to do without what you wanted. You couldn’t have both the oil industry and clean lakes, she thought, and if you had to choose, you had

Sensemaking, Empathy, and Democracy 219 to choose oil” (2018: 177). You had to choose oil for economic and political reasons. Regulation was seen as being at odds with capitalism and the American Dream, and several of Hochschild’s subjects were explicit in their commitment to both. One told her that she was “so for capitalism and free enterprise” and that the “environmentalists want to stop the American Dream to protect the endangered toad” (2018: 122). Regulation puts power in the hands of the federal government, but “the federal government was taking money from the workers and giving it to the idle. It was taking from people of good character and giving to people of bad character” (2018: 144). It has no business regulating people’s lives, especially if climate change, to which environmental regulation is supposed to be a response, is a “bunch of hooey” (2018: 48). It is difficult to imagine a more self-consciously ideological explanation of the Great Paradox. Hochschild’s subjects are victims of pollution who are resistant to regulating polluters because doing so would conflict with their ideological commitments. The latter are matters of principle; they see pollution as a price worth paying for capitalism and the American Dream. In other words, Hochschild’s subjects value keeping the government out of their affairs more than they value clean air and water. When ideologically committed victims of pollution oppose greater regulation of polluters, they are simply being ideologically consistent. Furthermore, there is no need for empathy to grasp their ideological commitments and values, and their implications for environmental regulation. It is enough to listen to what they say about their reasons and motives. No imaginative adoption of their perspective is required. One simply needs to take them at their word, as Hochschild does. Many questions remain unanswered. One might wonder why Hochschild’s subjects are so hostile to the federal government and why their values are as they are. When they talk about the federal government giving money to people of bad character, who do they have in mind? What is the role of race in their thinking about these matters? People’s reasons for their political beliefs and preferences may be rationalizations rather than their actual motives. Empathizing with them will not reveal whether they are rationalizing or speaking in code when they object to welfare payments going to people of bad character.11 If one is mystified by another person’s political beliefs and tries to make sense of them by empathy, one is unlikely to succeed. Fully empathizing with someone else means envisaging oneself in their situation with their beliefs and other psychological characteristics rather than one’s own. Thus, the fact that one is mystified by their beliefs might make it impossible for one fully to empathize with them. Tasked with empathizing with a racist or someone else with alien values, one may have no idea how to begin.12 However, this is not the end of the story. Hochschild is not merely concerned with her subjects’ political beliefs but with “how life feels to people on the right” (2018: xi). Perhaps the point at which empathy comes into its

220  Quassim Cassam own concerning the Great Paradox is the point at which, to understand her subjects, she needs to understand their emotions. She sees herself as trying to understand “the hopes, fear, pride, shame, resentment, and anxiety” (2018: 135) in the lives of those she talked with. Their emotions include resentment about the liberal perception of people like them as backward, racist, sexist, homophobic, and overweight. This is one factor that makes them feel like strangers in their own land. Another source of resentment is the feeling of being held back while immigrants, black people, and refugees cut in line ahead of them with the federal government’s help. Without emotional empathy, how can their emotions make sense to an outsider? In Bailey’s example, empathizing with the widower’s grief by imagining the loss of a loved one presents no great challenge for anyone capable of experiencing the normal range of human emotions. The imagined loss of a loved one, like the actual loss of a loved one, “may be felt as a tightness in the throat or hollowness in the stomach” (2022: 53). This isomorphism between the empathetic experience and the grief of the widower is the basis of the notion that when we empathize, we do not merely imagine that we are feeling some emotion but experience it. Empathizing with Hochschild’s subjects is a different matter. One might be incapable of viewing help for minorities through the lens of resentment if one can only see such support in a positive light. Someone repelled by racism cannot experience resentment about federal government help for people of bad character if they suspect that “person of bad character” is a covert racial epithet. Isomorphism between the putative empathetic experience and the original resentment of Hochschild’s subjects might be unattainable because of a fundamental difference in political outlook and values. The difficulty is that the emotions in these cases are political emotions and that empathizing with someone else’s political emotions requires what might be called political empathy rather than plain human sympathy.13 Unlike the widower’s grief, political emotions are ones whose source is ideological and take as their object abstract matters of political principle, such as the relative merits of capitalism and regulation. Empathizing with such emotions requires the imaginative adoption of the political perspective that underpins them, but such political empathy may prove impossible for those who find the perspective in question wrong-headed and repellent.14 For example, a committed environmentalist might be able to understand, in the abstract, why someone who is in love with free market capitalism would resent government regulation of the environment. Still, it would be unsurprising that the environmentalist cannot feel any such emotion. Their emotional experience cannot resemble that of the person who resents environmentalists because they are killing the American Dream to save the toad. However, the environmentalist’s lack of empathy for Hochschild’s subjects need not prevent them from understanding

Sensemaking, Empathy, and Democracy 221 the latter’s perspective on regulation. Since their perspective flows from their ideological commitments, how they feel about environmental regulation and the federal government is not a mystery. It is not empathy but dialogue that reveals the political outlook of Hochschild’s subjects and removes any sense that their opposition to environmental legislation is paradoxical. It is possible to make sense of them without feeling what they feel, even if some form of cognitive empathy is required.15 Knowing how someone feels about something by inference from their politics is different from knowing how they feel by feeling what they feel. Environmentalists may lack an intimate, feeling-based understanding of the inner lives of Hochschild’s subjects. In this sense, they lack an essential form of personal understanding of these subjects, but no such understanding is required for political sensemaking. Unsurprisingly, the latter requires political rather than personal understanding. There is more about the distinction between political and personal understanding in Section 11.4, but one obvious reason for not making political understanding depend on personal understanding is that it is only possible to have a unique understanding of, or emotional empathy with, people one knows personally and knows well. The number of such people is tiny when compared to the number of people of whom one seeks, and perhaps achieves, political understanding. This makes it unlikely that personal understanding is the key to political understanding. On the final page of her book, Hochschild speaks of the need to find new ways to “get acquainted across our differences” (2018: 266).16 She recommends, among other things, high school domestic exchange programs for which “students could prepare by learning active listening and epistemology” (2018: 266) as well as history and civics. This reference to active listening reveals more about Hochschild’s methodology than talk of empathy. Active listening is attentive, compassionate, unhurried, nonjudgmental, and unaggressive. Active listeners use respectful questioning and non-verbal cues to demonstrate their interest in what the speaker is saying. They do not interrupt and they verify their understanding through paraphrasing the speaker’s message. This is a fair summary of how Hochschild proceeds. Her emphasis on the “deep story” of her interlocutors is particularly compelling. Their deep story focuses on relationships between social groups in America. Like any conscientious active listener, Hochschild reconstructs her subjects’ deep stories and tests them “to see if they thought it fit their experiences” (2018: 135). They did. According to one of her subjects, she had succeeded in reading his mind. Active listening is sometimes described as empathetic, but this sense of “empathy” has little to do with emotionally charged perspective-taking. Political sensemaking requires one to be a good active listener but does not require the imaginative adoption of the speaker’s perspective. When it comes to active listening and sensemaking, we should resist the temptation to exaggerate the role of empathy.

222  Quassim Cassam

11.3  Empathy and Democracy Why would anyone think that empathy is vital for democracy? Two arguments in favor of this view are the argument from democratic legitimacy and the argument from polarization. Jason Stanley proposes a version of the first argument in How Propaganda Works.17 It is based on Du Bois’s account of the political system of the American South during the period after the Civil War.18 According to Du Bois, as Stanley reads him, the South’s laws lacked democratic legitimacy for two reasons. The obvious one is that Blacks were not allowed to participate in the making of laws that applied to them. A less apparent reason is that “those who created the laws did not have empathy for some of those subject to them, namely, their Black fellow citizens” (2015: 101). This meant that “the laws were crafted in such a way that did not reflect respect for the viewpoints of Black citizens” (2015: 101). Lacking respect, the laws also lacked democratic legitimacy. Stanley understands empathy as cognitive empathy, the capacity to imagine oneself “as someone in the situation of the other” (2015: 102). This mental capacity “underlies the capacity to give the perspectives of our fellow citizens equal weight” (ibid.). By implication, it is a precondition of democratic legitimacy. A democratic culture is “one in which everyone has a say in the policies and laws that apply to them” (2015: 16). It is also one in which, when proposing a policy, policymakers “imagine being someone subject to that policy” (2015: 102). The “someone” in this formulation implies an impartial stance. This is different from Bailey’s conception of empathy since it does not involve imagining oneself in the shoes of a specific other and is not a piece of emotionally charged perspective-taking. Nevertheless, it is empathy as Stanley understands it. Stanley’s argument fails. Democratic legitimacy requires that policymakers appreciate the impact of their policies on their fellow citizens, especially their negative impact on specific groups of citizens. It also requires that policymakers are not indifferent to such effects and properly consider them in formulating their proposals. However, as Stanley notes, “to gain an appreciation of the fact that others would be negatively affected by a policy I support, I do not need to be able to occupy their perspective, even in an impartial manner” (2015: 103).19 For example, a childless minister might be incapable of imagining being someone with a child, but this does not prevent them from appreciating the impact on low-income families with children of a policy to reduce levels of child support. Much depends on what counts as “appreciating” the fact that others would be negatively impacted by a policy one supports. It might seem that without empathy, the childless minister can only have an intellectual appreciation of what his policy means for people with children rather than a “real” appreciation. The minister might understand that some

Sensemaking, Empathy, and Democracy 223 people will end up worse off because of his policies but also be indifferent. Empathizing with those affected makes it harder not to care. Indeed, there is the view that empathy is “itself a way of caring” (Bailey 2022: 51). Furthermore, the form of empathy that is most likely to make vivid the impact of a policy to cut child support is not the relatively bloodless and impartial empathy that Stanley describes but full-blown emotional empathy that enables the minister to feel the economic pain of those affected by his policies. The picture of a government minister who is indifferent to the impact of his policies on sections of the population is not attractive, and it is not implausible that policies that display such indifference lack democratic legitimacy. However, the necessary remedy is not empathy. Even if empathizing with affected groups is not feasible, as it might not be, it is reasonable to expect lawmakers to show compassion for people affected by their decisions and to be willing to listen to them. As Bloom points out, compassion and empathy are not the same things. Compassion is “simply caring for people, wanting them to thrive” (2018: 50). It is “more diffuse than empathy” (2018: 40) and does require one to mirror anyone else’s feelings: “it is weird to talk about having empathy for the millions of victims of malaria, say, but perfectly normal to say that you are concerned about them or feel compassion for them” (2018: 40–41). Compassion, rather than empathy, is the antidote to indifference. Listening matters because, in a democracy, people who will be affected by a law or policy deserve a hearing. This was Du Bois’s point about the South. As he puts it: “it is pitiable that frantic efforts must be made at critical times to get lawmakers in some States even to listen to the respectful presentation of the black man’s side of a current controversy” (1994: 89). The listening that is at issue here is active. Active listeners engage with the arguments presented to them. They take them, and the people who put forward these arguments, seriously rather than dismissively. However, as argued above, active listening does not require empathy. The active listener tries to make sense of opposing points of view, especially when they are surprised by them. They are, in this sense, engaged in sensemaking. It is compassionate sensemaking on the part of policymakers rather than empathy that is needed for democratic legitimacy. The argument from polarization has a different take on the link between democracy and empathy. The idea is that excessive polarization threatens democracy and that the antidote is empathy. The notion that polarization is a threat to democracy is a familiar one. In polarized societies, neither side in political disputes sees their political opponents’ views as legitimate. Political adversaries “often regard each other as immoral, stupid, lazy, and even threats to each other’s way of life” (Hannon 2020: 597). As people become more polarized, they become more antagonistic and less willing to compromise. Eventually, democratic institutions such as elections and an independent judiciary are threatened as the process

224  Quassim Cassam accelerates. Polarization in the US-led Donald Trump’s supporters to use force to attempt to overturn the result of the 2020 Presidential election. In the United Kingdom, bitter arguments about Brexit led some sections of the British press to employ the Nazi tactic of representing uncompliant judges as the enemy within. This sort of behavior is no basis for a healthy democracy. In response to this concern, Hannon outlines a version of deliberative democracy that is “partly grounded in empathetic understanding” (2020: 592). Deliberative democrats take the exchange of reasons for preferring specific outcomes or believing certain facts to be central to decision-­ making. The exchange of reasons is only possible if people understand each other, and the relevant form of understanding is what Hannon calls “empathetic understanding” (2020: 597). This requires a willingness to listen to other people, including one’s political adversaries. More than this, “it requires the ability to ‘take up’ another person’s perspective. We must be able to see the other person’s point of view” (2020: 598). We must be able to “reenact or imitate the thought processes of others” (2020: 598). Only if we do that are we likely to find common ground with our fellow citizens and a basis for compromise. That is why democracies should “encourage citizens to understand others empathetically” (2020: 602). Democracies that fail to do this risk falling apart under the pressure of polarization. This argument suffers from some of the same defects as those put forward by Stanley and Hochschild. One theme that unifies these arguments is the transition from calls for people to listen to their political adversaries to calls for people to empathize with their adversaries. It is possible to listen without empathizing, and empathy is in many cases neither necessary nor possible. As noted above, it is difficult to empathize with people whose views are repellent, but this does not mean that it is not possible to understand their views. Understanding does not have to be empathetic. For example, many people who stormed the Capitol on January 6, 2021 believed that the 2020 election had been stolen and had a story about how the so-called steal had happened. It is perfectly straightforward to listen to these views and understand them without empathizing with them, or the people whose views they are. Empathizing with the rioters requires a sympathetic identification that many would find impossible and unnecessary. Sympathetic identification with a recently bereaved widower is one thing. Sympathetic identification with the Capitol rioters is another. Its emphasis on the need for mutual understanding, and the role of empathy in securing such understanding, raises another question about the argument from polarization: how far is political polarization the result of a lack of understanding? Liberals in America understood the motives and beliefs of the Capitol rioters only too well. There was no lack of understanding to be remedied by empathy. The problem was that there were deep and irreconcilable differences between the two sides. As Amos

Sensemaking, Empathy, and Democracy 225 Oz observes concerning the Israeli/Palestinian dispute, “some conflicts are real” and “much worse than a misunderstanding” (2012: 8). Even if it were possible for the Capitol rioters to use their imagination to take up the perspective of their liberal critics, it is wishful thinking to suppose that this would have made them more likely to arrive at a compromise with them. In the same way, liberals who empathize with rioters enough to see their point of view may feel more rather than less hostile to them due to this exercise. Empathic understanding could end up exacerbating polarization.20 It might seem that the position at which we have arrived is an uncomfortable one. On the one hand, polarization has been represented as a threat to democracy. On the other hand, many democracies are becoming more polarized. How, in that case, is democracy in these countries possible? How is democracy in America possible? Three responses suggest themselves. The first is to question the idea that polarization is incompatible with democracy. The second is to accept this idea but argue that the degree of polarization in America has not yet reached a critical level and is still compatible with democracy. The third would be to question the assumption that countries like America are democracies. Thus, one might agree with Stanley that America is a democracy in name only and that the language of democracy is used to mask a thoroughly undemocratic reality.21 The extent of polarization is not the only basis for this view. A democratic culture is one in which all citizens have an equal say in the policies and laws that apply to them. It would be difficult to argue that this is true in a country like America, where voter suppression is rife, and a person’s wealth determines access to political power. These are problems to which empathy is not the solution.

11.4  Political Understanding In his General Psychopathology, Karl Jaspers distinguishes between explanation and understanding. By explanation, he means causal explanation: “We find by repeated experience that a number of phenomena are regularly linked together, and on this basis we explain causally” (1997: 301). Based on the observation of events, experiments, and the collection of examples, “we attempt to formulate rules. At a higher level, we establish laws, and in physics and chemistry, we have to a certain extent reached the ideal, which is the expression of causal laws in mathematical equations” (1997: 302). Understanding is different since it pertains to what Jaspers calls “meaningful psychic connections” (1997: 301). In this context, “psychic” means “psychological.” Understanding is personal. When we attempt to understand another person, we immerse ourselves in their psychology and try to “understand genetically by empathy how one psychic event emerges from another” (1997: 301). For example, we understand by empathy, rather than by experiment, that “attacked

226  Quassim Cassam people become angry and spring to the defense, cheated persons grow suspicious” (1997: 302). Jaspers subscribes to a form of what Christoph Hoerl calls epistemic particularism about understanding.22 This is the view that understanding “is achieved (if it is achieved) directly upon confrontation with a particular case” (Hoerl 2013: 108). Understanding “is not achieved by bringing certain facts under general laws established through repeated observation” (2013: 108). The role of understanding is to make something “visible to our experience” (Jaspers 1997: 312). It can only play this role “because it deals specifically with connections between elements of a person’s conscious life” (2013: 109). It is possible to immerse oneself in another person’s psychic situation because “there is something it is like to be in that situation” (2013: 109). Thus, understanding is particularist, it is directed at the mental life of another person, and it engages with the subjectivity of the other by empathy rather than through the application of general laws. Given the distinction between explanation and understanding, one way to approach the Great Paradox is to look for an explanation. This is the approach of theorists who view the paradox through the lens of false consciousness, a mode of consciousness that misrepresents socio-economic reality while also being determined by that reality. In capitalist societies, the socio-economic reality that this form of consciousness misrepresents includes the reality that the socio-economic status quo serves the political and economic interests of the ruling class but not the working class. For the latter to acquiesce in such a system, they need to misperceive their interests and identify with capitalism and free enterprise.23 Their identification with capitalism is manifested by, for example, their pro-capitalist ideology. This ideology masks key features of the socio-economic reality by which it is “determined.” Whether or not Hochschild’s subjects are “working class,” there is little doubt that many people in Louisiana, including some who spoke to her, do not do well economically or socially. According to her data, Louisiana ranked forty-ninth out of fifty states in human development and last in overall health. Only eight out of ten Louisianans have graduated from high school, and only 7 percent have graduate or professional degrees.24 Yet many of these people describe themselves as “so for capitalism and free enterprise.” From a false consciousness perspective, this is a classic case of the disenfranchised and marginalized being blind to their interests. Their blindness is not a mystery since it is explained by a generalization about highly unequal societies: what keeps them on an even keel is their ability to induce large numbers of socially and economically disadvantaged people to believe that the status quo works for them. The basis of this generalization is not empathy but a historical observation. Paraphrasing Jaspers rather than Marx, we find by repeated experience that gross inequality and working-class loyalty to the system often go

Sensemaking, Empathy, and Democracy 227 together. Those who have the least to gain from the established order are among its most enthusiastic defenders. This explains the Great Paradox in general terms by referring to the law about how unequal societies work. Hochschild objects to this explanation because it lacks a complete understanding of the role of emotion in politics: Many liberal analysts …. have tended to focus on economic self-­ interest. It was a focus on this that led me …. to carry the Great Paradox like a suitcase on my journey through Louisiana. Why, I’d repeatedly asked myself, with so many problems, was there so much disdain for federal money to alleviate them? These were questions that spoke heavily to economic self-interest. And while economic self-interest is never entirely absent, what I discovered was the profound importance of emotional self-interest – a giddy release from the feeling of being a stranger in one’s own land. (2018: 228) False consciousness explanations suppose that people who do not act in accordance with their own economic interests must be misperceiving them. However, someone can see clearly that taxes to fund higher welfare payments would be economically beneficial to them personally and still be against this policy because they hate big government or resent welfare payments going to the idle and undeserving.25 Such a person does not fail to grasp their economic interests. It is just that other things matter to them more. To understand what those other things are and why they matter to them, we need to understand their emotions and engage with their subjectivity as individuals rather than resort to a generalization about unequal societies. Political understanding, an understanding of people’s political choices, is ultimately a form of personal understanding. On this account, the challenge is to understand the Great Paradox and explain it. To understand it, we need to understand individual people, where the relevant notion of understanding is the one described by Jaspers. There is something right about this, but it calls for the false consciousness approach to the Great Paradox to be modified rather than abandoned. It is undoubtedly a mistake to suppose that people who do not benefit economically from capitalism are always unaware of this fact or blind to their economic interests. However, most are not as clear-eyed as Hochschild’s ideologically driven subjects, who understand that they must choose between oil interests and clean lakes or between having more money and having low taxes for the rich. As for those who are cleareyed about these matters, it is not enough to point out that they have strong feelings about the size of the government or taxes. The question is: why do they feel the way they do? Why is their loyalty to the oil industry stronger than their desire for clean lakes or their hatred of the government stronger than their commitment to their financial well-being? These

228  Quassim Cassam emotions and preferences express their ideology, but the obvious question is: why do they have such an ideology? One would not be asking these questions if their political emotions and preferences made perfect sense. These questions are pressing precisely to the extent that, from an “objective” standpoint, the marginalized ought not to care more about the interests of large corporations than about the gradual destruction of the physical environment in which they and their children will be living. By implication, this way of putting things distinguishes a person’s real interests from their actual preferences or emotions. Large corporations have a real interest in minimal regulation, but Hochschild’s subjects do not. Their real interest lies elsewhere. This is not merely an issue of economic self-interest but of overall well-­ being. Whatever their feelings, it is, in fact, worse for people to be living in a highly polluted but unregulated environment than in a regulated but unpolluted environment. If they fail to see this, the most plausible explanation is one in terms of their false consciousness. Similarly, if false consciousness has a socio-structural explanation, so does the Great Paradox. Liberal critics will almost certainly object to talking about people’s “real” interests, as distinct from their preferences.26 Some may feel that there is no “objective” standpoint from which we can distinguish real interests from actual preferences, but this is a mistake. The issues are far too complex to be satisfactorily dealt with here. Still, one way to allow actual preferences and real interests to come apart is to insist that there is such a thing as the human good or a good life for a human being and that a person’s real interests are at least partly a reflection of the human good.27 Living in an environment that is not dreadfully polluted is part of a good life for a human being. Whether they realize it or not, Hochschild’s subjects have a genuine interest in living in such an environment. They are victims of false consciousness if they fail to realize it and care more about protecting oil interests. Their consciousness is not false because they misrepresent socio-economic reality but because their priorities are skewed relative to a plausible vision of the human good. It is striking how little empathy reveals about these fundamental matters. Empathizing with Hochschild’s subjects, or understanding them in Jaspers’ sense, will reveal their priorities but not that their priorities are skewed. Confrontation with a particular case might reveal a specific individual’s worldview and the sources of that worldview in their life story. Still, it will not connect their worldview with that of others in the same situation. It will not explain how their view comes to be shared by many people in the same social position or the role of socio-structural factors in manufacturing consent to pollution or inequality. The particularist orientation of understanding is attractive on a human level but too narrowly focused to be a practical exercise in sensemaking on a macro level. As noted above, the number of people in whose mental life one can immerse oneself is far lower than the number whose views and preferences need

Sensemaking, Empathy, and Democracy 229 to be understood. These considerations all point to the need for a more general or generalist approach to the Paradox. In Jaspers’ terminology, they point to the need for an explanation. The thesis that the best explanation is one in terms of false consciousness has not been defended here. It has only been put forward as a potentially satisfactory explanatory approach to sensemaking. Whether it is actually a satisfactory approach cannot be settled here. However, the shape of the approach is more important for present purposes than its details. In contrast with epistemic particularism, with its emphasis on empathy and confrontation with particular cases, the false consciousness approach to the Great Paradox is based on observation rather than empathy. It is generalist rather than particularist in its orientation. Rather than focusing on individuals, it focuses on groups or classes of people, defined by reference to their social location. These are the basic units of explanation, and the mode of explanation is functional or causal. When people in the same situation have the same perverse preferences, the perversity of their priorities is explained by reference to their function of maintaining the status quo. A response to this line of thinking would be to say that there is no need to choose between empathy and understanding or between false consciousness and emotional empathy as tools for dissolving the Great Paradox. Why can’t empathizers do their thing while social scientists analyze the paradox in terms of general laws? The short answer is that there is nothing wrong with the two approaches running in parallel, but there is a deeper issue about the nature of political understanding. Hochschild offers a new model of political understanding, which says that the best way to make sense of the coexistence of significant pollution and great resistance to regulating polluters is to empathize with the emotions of people like Mike Schaff. The attractions of this approach are apparent, but so, now, are its limitations. The political, as distinct from personal, understanding that it delivers is relatively shallow. We should not fetishize empathy when making sense of people’s political preferences. Individual psychology is no substitute for social science.

Notes 1. Hochschild (2018). 2. On sensemaking as a response to a surprise, see Louis (1980). 3. The retrospective nature of sensemaking is emphasized in Weick, Sutcliffe and Obstfeld (2005). 4. See Hannon (2020). 5. Coplan lists no fewer than seven mental processes, or states described as empathy. See Coplan (2011): 4. 6. See Bloom (2018: 17) for more on the distinction between cognitive and emotional empathy. 7. See Marx and Engels (1970) and Meyerson (1991) for a valuable account of the Marxist theory of false consciousness.

230  Quassim Cassam













8. I leave it open whether Hochschild’s project or socio-structural explanations of the Great Paradox require cognitive empathy. Henceforth, unless otherwise indicated, by “empathy,” I mean emotional empathy. Thanks to Hana Samaržija for urging me to be more explicit about this. 9. “In claiming that empathy (in my sense) involves both emotion and perspective-taking, I do not mean to deny that some forms of perspective-­ taking are affectless” (Bailey 2022: 52, n. 6). 10. See Hannon (2020: 604). 11. Could “people of bad character” be a coded racial epithet, like “welfare recipient”? There is more about such epithets in Gilens (1996). 12. One could try to “bracket” one’s own beliefs and attitudes, as Hannon recommends (2020: 598), but some attitudes are impossible to bracket without undermining one’s identity. As Bailey notes (2022: 52), there is a difference between imagining being in someone else’s position with one’s character, history, and physical features intact and imagining being in their shoes with their character, history, and physical characteristics. I cannot imagine being in the shoes of a virulent racist with his character and values. Imagining being in his position with my character and values is unlikely to cast much light on the racist’s inner life. In this case, it seems that empathy is either impossible or useless. 13. As Martha Nussbaum notes, “all societies are full of emotions” (2013: 1). While some of these emotions have little to do with political principles or public culture, “others are different: they take as their object the nation, the nation’s goals, its institutions and leaders, its geography, and one’s fellow citizens as fellow inhabitants of a common public space” (2013: 2). These are examples of political emotions. 14 As Sharon Krause points out, “there is nothing in perspective-taking, construed as a purely intellectual act, that effectively moves us to think beyond the limits of our personal convictions” (1998: 162). The same goes for emotionally charged perspective-taking. 15. See below on whether even cognitive empathy is required. 16. This is her response to a magazine article by Frank Rich in which he writes that for all Hochschild’s “fond acceptance of her new Louisiana pals, and for all her generosity in viewing them as virtually untainted by racism, it’s not clear what such noble efforts yielded beyond a book, many happy memories of cultural tourism, and confirmation that nothing will change any time soon. Her Louisianans will keep voting for candidates who will sabotage their health and their children’s education; they will not be deterred by an empathic Berkeley visitor, let alone Democratic politicians” (Rich 2017). Rich is right. 17. Stanley (2015, chapter 3). 18. Du Bois (1994). 19. Stanley credits Sharon Krause with this insight. See Krause (1998: 162– 165). Another consideration is that it might not be possible “to imaginatively place oneself in the situation of others who have had dramatically different life experiences” (Stanley 2015: 103). See Paul (2015). 20. As Hannon concedes. See Hannon (2020: 599). 21. Stanley (2015: 13). 22. Hoerl (2013). 23. This is one of Meyerson’s two dimensions of false consciousness. She describes “twin states of mind” as involving false consciousness: “first, the rationalizations of members of the ruling class, their inaccurate conception

Sensemaking, Empathy, and Democracy 231





of their motives, and, second, the blindness of the workers to their interests, their identification with the capitalist system …. It is the rulers who benefit from both mistakes” (1991: 8). 24. Hochschild (2018: 9). 25. Naturally, such people do not think of themselves as undeserving. It is only other people who are freeloaders. 26. According to Isaiah Berlin, “it is one thing to say that I know what is good for X, while he himself does not; and even to ignore his wishes for its – and his – sake; and a very different one to say that he has eo ipso chosen it” (1969: 133). From a false consciousness perspective, there is no question of Hochschild’s subjects somehow choosing stricter environmental regulation, despite their protestations to the contrary. The critical point, which Berlin does not deny, is that people do not always know what is good for them. 27. Inspired by Aristotle, Philippa Foot remarks that “the idea of the human good is deeply problematic” but that “for all the diversities of human life, it is possible to give some quite general account of human necessities, that is, of what is quite generally needed for human good” (Foot 2001: 43). These human necessities include clean air and clean water but, whatever Hochschild’s subjects might think, not unregulated oil production. The idea of basing an account of false consciousness on an Aristotle-inspired account of the human good deserves more detailed consideration than it can be given here.

References Bailey, O. (2022), ‘Empathy and the Value of Humane Understanding,’ Philosophy and Phenomenological Research, 104: 50–65. Berlin, I. (1969), Four Essays on Liberty (Oxford: Oxford University Press). Bloom, P. (2018), Against Empathy: The Case for Rational Compassion (London: Vintage). Coplan, A. (2011), ‘Understanding Empathy: Its Features and Effects,’ in A. Coplan & P. Goldie (eds.) Empathy: Philosophical and Psychological Perspectives (Oxford: Oxford University Press): 3–18. Currie, G. (2011), ‘Empathy for Objects,’ in A. Coplan & P. Goldie (eds.) Empathy: Philosophical and Psychological Perspectives (Oxford: Oxford University Press): 82–95. Du Bois, W. E. B. (1994), The Souls of Black Folk (New York: Dover). Foot, P. (2001), Natural Goodness (Oxford: Clarendon Press). Gilens, M. (1996), ‘“Race Coding” and White Opposition to Welfare,’ American Political Science Review 90 (3): 593–604. Hannon, M. (2020), ‘Empathetic Understanding and Deliberative Democracy,’ Philosophy and Phenomenological Research, 101: 591–611. Hochschild, A. R. (2018), Strangers in Their Own Land: Anger and Mourning on the American Right (New York: The New Press). Hoerl, C. (2013), ‘Jaspers on Explaining and Understanding in Psychiatry,’ in G. Stanghellini & T. Fuchs (eds.) One Century of Karl Jaspers’ General Psychopathology (Oxford: Oxford University Press): 107–120. Jaspers, K. (1997), General Psychopathology, volume 1, trans. J. Hoenig & M. Hamilton (Baltimore: Johns Hopkins University Press).

232  Quassim Cassam Krause, S. (1998), Civic Passions: Moral Sentiment and Democratic Deliberation (Princeton: Princeton University Press). Louis, M. R. (1980), ‘Surprise and Sense Making: What Newcomers Experience in Entering Unfamiliar Organizational Settings,’ Administrative Science Quarterly, 25: 226–251. Marx, K. & Engels, F. (1970), The German Ideology, ed. C. J. Arthur (London: Lawrence & Wishart). Meyerson, D. (1991), False Consciousness (Oxford: Clarendon Press). Nussbaum, M. (2013), Political Emotions: Why Love Matters for Justice (Cambridge, MA: Harvard University Press). Oz, A. (2012), How to Cure a Fanatic (London: Vintage Books). Paul, L. A. (2015), Transformative Experience (Oxford: Oxford University Press). Rich, F. (2017), ‘No Sympathy for the Hillbilly,’ New York Magazine (March 19, 2017). Stanley, J. (2015), How Propaganda Works (Princeton: Princeton University Press). Weick, K. E. (1995), Sensemaking in Organizations (London: Sage Publications). Weick, K. E., Sutcliffe, K. M. & Obstfeld, D. (2005), ‘Organizing and the Process of Sensemaking,’ Organization Science, 16: 409–421.

12 Political Skepticism, Bias, and Epistemic Colonization Michael P. Lynch

12.1  Political Skepticism Anyone interested in politics has, at one time or another, despaired at the lack of progress that we seem to make toward a more just, equitable, or peaceful society. Such despair, and the cynicism that accompanies it, most frequently arises from observing recurrent obstacles to progress such as racism, religious violence, or the inherent selfishness and greed of human beings – hindrances that recur so often they can make the political task seem downright Sisyphean. A central assumption behind the idea that political progress is possible is that we know where we should be heading. From a comprehensive perspective, that is not too hard – we head toward the horizon of justice. The hard part is knowing the details of the way. That, it seems, is very difficult indeed. So difficult that some think we should acknowledge it is impossible. That is, we are incapable of knowing what is true in politics. Call this view political skepticism. The threat that political skepticism poses is not to the possibility of truth in politics but its utility as a political concept.1 As I will be using the term here, skepticism about X – whether X is the external world, or other minds, morality, or politics – is the view that, to some degree or other, we are incapable of knowing about X – the truth about it eludes our grasp. That is distinct from the view that there are no political truths to know. To those with an ear tuned to the practical, that may sound like a distinction without a difference, but it is not. To cite just one: the idea that there are no political truths has the consequence that there are also no political falsehoods, and hence no such thing as political mistakes. On the other hand, the political skeptic believes we are constantly making political mistakes – making poor or unjustified political judgments. That is a view that has been used to motivate a variety of political standpoints. But it can also motivate the rejection of the significance of political knowledge and appeals to it in democratic theory and practice. As such, no inquiry into the epistemology of democracy can ignore it. Before we start, however, let me say what I mean by political knowledge. Sometimes political scientists talk about political knowledge as DOI: 10.4324/9781003311003-16

234  Michael P. Lynch knowledge of how a political system works or its elements. For instance, how many branches of government there are or who the president is (Somin 2013).2 There is nothing wrong with labeling this kind of knowledge as political knowledge, but judgments like that are not what primarily concerns us here. We are interested in those judgments that figure in public political discourse and constitute such discourse. Moreover, as everyone knows, the range of such judgments is extensive. As Orwell once lamented, “There is no such thing as keeping out of politics … all issues are political issues.”3 And Orwell’s sentiment is certainly in keeping with our times when everything from COVID to climate change is politicized. To accommodate his point while avoiding any foolhardy attempt to give a conceptual analysis of something so slippery, I will assume that political knowledge is knowledge involving judgments with political meaning. A judgment has such meaning when it carries distinctive political associations for a community in a context.4 These will include the actions it is thought to engender (voting, protesting, and liking), the convictions and values it is thought to express, and whose side is perceived as “winning” if the judgment is widely accepted. Thus, whether one’s knowledge counts as political depends on the context and community one is in. What is apolitical knowledge in one context (knowledge of vaccines, for example) can suddenly become political when the context shifts or the issues change. If so, then skepticism about political knowledge concerns whether we can know which of the judgments that have political meaning for us are true. In what follows, I will review several lines of skeptical argument directed against the possibility of political knowledge so understood, aiming to show that one of them – concerning the reality of a conceptual corruption often called epistemic colonization – is far more plausible than the others. I conclude by offering some thoughts on how this skeptical threat, while profound, should motivate us not toward quietism but a more vigorous and inclusive democratic politics.

12.2  Skepticism from Disagreement The ancient academic skeptic Carneades, visiting Rome to appeal on behalf of the Athenians, spent his time lecturing – one day in favor of virtue, the next against it – and endeavoring to demonstrate that there were always good arguments on either side of any consequential question. Unsurprisingly, this did not go down well with the Roman authorities. Cato the Younger is said to have granted the Athenian requests just to get Carneades and his dangerous rhetoric out of town and out of range of tender Roman ears (Laursen and Paganni 2015: 7). The story reminds us of two essential facts about skepticism. It is irritating to political partisans, and it usually begins from the observation that evidence does not definitively settle specific questions. If the former

Political Skepticism, Bias, and Epistemic Colonization 235 fact lends skepticism to some of its appeal, the latter reminds us of the oldest argument in its support. Crudely put. Wherever there is pervasive disagreement in judgment, we should conclude that we do not know which judgments are true and which are not.5 Curiously, the same premise has been used throughout history to support a very different conclusion: that judgments about morality are only relatively true or false. The relativist, unlike the skeptic, thinks we can know, but only what is relative to our perspective. That is appealing in the same manner that Mom’s assurance that her children’s picture is appealing – but it is not precisely transparent what it means. On the other hand, the skeptic offers straightforward counsel: admit that you do not know. Yet the clarity here only goes so far, with the ancient skeptics divided into two schools. The Academics, of which Carneades was a chief representative, held that skeptical arguments demonstrated we merely do not know anything. Followers of Pyrrho gave even more radical counsel than the Academics: we should not just refrain from saying we know. We should refrain from believing – even believing the proposition that we do not know anything. On this latter view, the point of skepticism was to free yourself from commitment – to allow yourself to be carried along on the gentle seas of appearance. The sentiment was that without dogmatic commitment, one would be happier. Whether one can genuinely be happier while believing in nothing is an intriguing question. However, we need not settle that to address the argument from disagreement as it might apply to politics. The first premise would seem true: that there is widespread and seemingly intractable political disagreement. Indeed, some (notably Schmitt 2007/1932) have asserted this is almost accurate. In their view, part of what makes a judgment political is that it is the subject of a certain kind of “us vs. them” conflict. No dissent, no politics. In any event, the first premise is plausible, even if inadequate by itself to raise skeptical doubts. Unsurprisingly, advocates of these sorts of arguments (whether about politics or something else) typically rest substantial weight not only on how widespread political disagreements are but also on the kind of disagreements in play. This claim brings us to the second premise – that we cannot “step outside” our political judgments and check to see which are true. How we regard this will depend in part on how it is defended. And it needs defense because it isn’t obviously true. One kind of defense would appeal to general concerns about how humans acquire knowledge. Perhaps, as George Berkeley thought, we can never know what judgments are true because we can never check and see without making another judgment – and that one, too, would need to be checked for truth. Or perhaps, as we will examine in greater detail in a moment, we can appeal to general facts about humans’ penchant for bias. But if we are defending the second premise with general facts about

236  Michael P. Lynch how bad at comprehending stuff human beings are, then those facts will presumably undermine not just political knowledge but all knowledge. We will have an argument on our hands that will be more destructive than we might have wanted. Nevertheless – and even if they pale at accepting a general skepticism about all knowledge – proponents of this argument face a further problem. Historical, scientific, and economic judgments can all become political in a reasonably intuitive sense. As the pandemic has taught us, almost any judgment, even causal judgments about the physical world, such as that mask-wearing lowers the chance of infection, can become the subject of political debate and so take on political meaning in a sense defined above. Furthermore, that presents the supporter of the argument with a dilemma: since it seems that almost any judgment can have political meaning in some context or other, either we accept a general and deeply implausible skepticism according to which we cannot know whether almost any proposition is true (including whether the earth is flat or climate change is real), or our knowledge of their truth is relative to a context (sometimes we know them, sometimes we do not), or we must restrict our skepticism to normative political judgments. Of these, I suggest the last is the most plausible, but it remains unclear how that restriction would go. Our general political views are entangled with one another – with normative political judgments being supported by, and in return, supporting non-normative political judgments, it will not always be clear how we could be skeptical about one kind of judgment without being skeptical about the other. Second, and more importantly, we have yet to be given a reason to think that normative political judgments are inherently unknowable. They would be if, for example, we held, with political emotivism, that such judgments are not truth-apt in the first place. But that is not the view we are considering, as we noted at the outset. Moreover, the mere fact that it is difficult to be objective, neutral, or serene about normative political judgments does not mean that it is impossible to know whether any of them are true. Hence, the general argument from disagreement so far canvassed is not effective at motivating a particular skepticism about political knowledge. However, appeals to a disagreement can motivate a more limited, if also more unsettling, political skepticism. It is possible to see irreconcilable political disagreement undermining not all possible political knowledge a priori, but as an empirical fact that undermines our access to a particular kind of political knowledge concerning the state’s legitimacy.6 This kind of skepticism has been attributed to Nietzsche by Shaw (2007). Such an attribution can sound surprising. Nietzsche, after all, is often identified with relativism, not skepticism – as rejecting the idea of objective political truth as opposed to arguing that we cannot know what it is. But Shaw reads Nietzsche as arguing for a different approach. As she puts it, “Nietzsche’s most distinctive argument for political skepticism…

Political Skepticism, Bias, and Epistemic Colonization 237 rests on the claim that even if we can assume that there are knowable normative truths, secular societies will have a tremendous problem in making those truths effective in political life” (Ibid. 9). The political effectiveness Shaw has in mind concerns the role such truths can play in building consensus. The idea is this: modern secular states require normative consensus in order to be, or at least appear to be, legitimate. Such consensus requires, at the very least, agreement on certain general principles concerning the governance of society and its fundamental values. In a democracy, these might concern principles about voting, representation, and the extent of state power. Shaw’s Nietzsche is willing to grant that some people may be in a place to know these principles. Still, most people are not equipped to do so, and appeals to reason will not effectively settle a pervasive disagreement about them. A consensus of that sort has typically only been achieved by appeal to myth – to religion. However, that path is cut off by the secular state. As a result, in the absence of socially dominant myths or religious authority, the secular state is unlikely to achieve the consensus over grounding political truths that it needs to legitimize its exercise of power. The best it can do is to utilize its power to coerce people into accepting its coercion as justified. The Nietzschean skeptical argument is both weaker and more robust than the other arguments we just canvassed, invoking disagreement. It is weaker in the sense that it rests on at least three assumptions beyond the scope of skepticism itself: (1) that the chief usefulness of normative political judgments lies in its role in consensus-building; (2) that state legitimacy rests on such consensus; and (3) that religion is a stable mechanism for building it. Of these, the second seems plausible; the third – given the long history of religious strife – much less so. The first we will have the occasion to examine below. Regardless, the core skeptical argument here is also more potent in a distinct sense and can be considered separate from the above points. Unlike the previous versions, that core argument is not against the possibility of political knowledge but against whether enough people will ever have it to make such knowledge useful. It is, we might say, a defense of a lived political skepticism – and thus, were it combined with an unambiguous defense of its central skeptical hypothesis, would present a formidable challenge to the utility of the very idea concepts like truth and knowledge in politics.

12.3  Liberalism and Who Knows Better Before we move on and discuss a revised and updated Nietzschean approach to the problem, I want to pause and deflect a possible misunderstanding. So far, I have been representing skepticism about political knowledge in a way that may give the impression that it is a purely abstract philosophical problem with little real-world application. Nevertheless, political theorists have employed political skepticism to motivate all

238  Michael P. Lynch sorts of political positions, sometimes with significant (if often unfortunate) real-world consequences. I will briefly highlight three examples. Perhaps the most notorious use of skepticism goes back to Montaigne, who famously used Pyrrhonist arguments to shed doubt on dogmatic religious and political movements in 16th century France. At the same time, it was being torn apart by both plague and civil war. Montaigne was so struck by these arguments that he dropped out of politics altogether, resigning from his political post and counseling that dogmatic “zeal works wonders when it strengthens our tendency to hatred… but it never makes one fly toward goodness”7: 495). Montaigne seemed to draw the political lesson that we should be less engaged and less committed politically, as such commitments are precisely what motivates political violence. Skepticism here is used to justify political quietism. As we noted above, the Pyrrhonist arguments are entirely general, which Montaigne was keen to emphasize. But a more focused political skepticism was wielded during the 20th century by thinkers with a very different political agenda – one which sees liberty as the central political value. For example, we find Friedrich Hayek defending individualism by arguing that: The basic fact is that it is impossible for any man to survey more than a limited field, to be aware of the urgency of more than a limited number of needs. …This is the fundamental fact on which the whole philosophy of individualism is based. It does not assume, as is often asserted, that man is egoistic, selfish, or ought to be. It merely starts from the indisputable fact that the limits of our powers of imagination make it impossible to include in our scale of values more than a sector of the needs of the whole society…. This recognition of the individual as the ultimate judge of his ends, the judgment that, as far as possible, his views ought to govern his actions, comprises the essence of the individualist position. (Hayek 1944/2007: 102) Hayek’s “basic fact” supporting individualism is a central skeptical tenet. Roughly put, human life is so complicated and individuals so different that humans cannot make reasonable judgments about what people very different from themselves need and want. As a result, Hayek wants to conclude that attempts to centrally plan economies around views about what is in “everyone’s interests” are bound to fail. Such plans presuppose planners know what other people want and need. Ironically, where Hayek uses a form of political skepticism to motivate individualism and views opposed to central planning, others have used skepticism to motivate the rule of experts. Economist Bryan Caplan’s argument that voters in a democracy are systematically irrational, for example, can be taken that way (Caplan 2007). Where Hayek

Political Skepticism, Bias, and Epistemic Colonization 239 sometimes seems to assume that human beings are at least experts on what is in their self-interest, Caplan claims that human beings are terrible reasoners and are subject to frequent “rational irrationality.” When the outcomes of holding a false or unjustified judgment are not obvious, people may be prone to hold the judgment despite the evidence because doing so may have identity-confirming or other psychological benefits. Caplan’s argument is concentrated on economics: he is concerned with showing that people embrace “irrational” views, which most mainstream economists reject. This sentiment is what he means by saying that voters in a democracy make terrible decisions. They make terrible economic decisions that often run against their longterm self-interest. A tempting conclusion – perhaps Caplan’s, perhaps not – is that, at least when it comes to economic matters, we should often favor government by expertise over democratically arrived decisions. Roughly and crudely put, his view is that economists know best about economics. I do not accept either of these arguments. Both invoke skeptical premises but try to restrict the skeptical impact – Hayek by assuming that people can at least know their own interests and Caplan by assuming that experts can be free of the biases that bind ordinary people. However, my point here is not to assess either view but to use them to signal the critical role that skepticism can play in motivating actual political policy – even dreadful policy. Yet looking at both arguments together raises an obvious concern. What if everyone – both ordinary voters and experts – were flawed and biased in their politically relevant reasoning? Moreover, what if this failure was not confined to purely normative political judgments but extended to any judgment with political meaning, whether normative or not? If so, then Shaw’s Nietzschean argument sketched above would seem to gain a grip: even if there could be some political knowledge, disagreement about who knows what might be so pervasive that such knowledge, even when possessed, would be useless.

12.4  The Argument from Bias What I will call the argument from bias begins with the following two thoughts: (a) humans are naturally biased when making politically meaningful judgments, and (b) they are terrible at spotting bias in themselves. There is a vast and ever-growing body of research that would seem to suggest that both thoughts are true. Many psychologists today agree, for example, that our cognitive infrastructure is divided between considered, reflective reasoning and intuitive, automatic thinking. Given that we encounter a superabundance of sensory information moment-tomoment basis and that reflective reasoning processes are laborious, we could not get on with our lives if our brains did not quickly process most of the information with which we are regularly confronted.8 Thus, our

240  Michael P. Lynch brains regularly take “shortcuts” in service of the aim of fast and efficient processing.9 Yet what is efficient is not always true. In the context of social categorizing, the fact that we regularly take cognitive shortcuts when forming judgments about people means we overemphasize outgroup differences and ingroup similarities. But it also means that racist, sexist, or otherwise discriminatory associations that have permeated the more extensive social context can distort how we conceive of the categories themselves. To take a concrete example, consider Payne’s (2001) studies, in which subjects identified guns faster and misidentified tools as guns more often when primed with non-white faces than when primed with white faces. The widely accepted explanation for this result is that the subjects were more likely to associate blackness with danger than they were to associate whiteness with danger. Moreover, a wealth of other research corroborates these and similar points. None of it is a surprise. We are flawed, biased reasoners made of very crooked timber indeed. However, evidence suggests that we are unreliable at telling the difference between judgments formed by reason and those based on bias.10 In particular, many people exhibit what is known as a “bias blind spot,” an inability to recognize when biases inform their judgments. Thus, as Pronin and Schmidt (2013) discuss, people are prone to deny that their political or other ideological judgments are formed due to partisan or ideological alignment, insisting that their judgments were reached instead based on sound reasoning. The existence of bias blind spots should also come as no surprise, as we have been aware for quite some time of the Dunning-Kruger effect, whereby people tend to both generally overestimate their abilities and rate their capabilities as higher than those of their peers. Assuming this tendency carries over to the problem of bias detection. Not only is it the case that we regularly overestimate our ability to detect when our thinking is biased but also, the worse we are at detecting this, the better we think we are. Suffice it to say: the problem of pervasive bias is one we cannot ignore. It also seems relevant to whether democracies can expect enough of their citizens, expert or otherwise, to have political knowledge to counter the Nietzschean worry. One reason to think so is straightforward: we expect our political judgments to be biased. Indeed, a bias toward one’s partisan community seems almost inevitable. No one witnessing today’s polarized age of misinformation would think such bias is not rampant and dangerous to political knowledge. This fact seems an obvious motivation for political skepticism.11 The argument might go like this: if I know the politically consequential proposition that p, then I must be able to rule out the possibility that my judgment that p is the result of bias. Nevertheless, I cannot rule that out precisely because I may be suffering from a biased blind spot. So, I do

Political Skepticism, Bias, and Epistemic Colonization 241 not know whether any politically meaningful proposition is true. We do not know what we think we know when it comes to what is politically meaningful to us.12 The argument from bias has some plausibility, partly because it employs a skeptical possibility that seems all too real, unlike appeals to evil demons or brains-in-vats. However, it fails because, like those other, more famous philosophical arguments, it tries to deduce a universal conclusion a priori. This fact sets it up for failure. First, one might take exception to it by rejecting the premises. One can reject the first premise, for example, on the externalist grounds that one can know without having to rule out possibilities that you do not. One can sometimes know that p without having any thoughts about how you formed your judgment that p. And one might question the second premise by noting that we often CAN identify our biases in individual cases – and likewise, we can sometimes rule out the possibility that we are biased in a way that undermines knowledge. The empirical evidence we cited above is consistent with the fact that we do not always suffer from blind spots, and sometimes we can tell when we are not. An even more basic problem with the argument is its assumption that bias consistently undermines knowledge. As Kelly and McGrath have recently noted, that is not always clear. To cite their example: the fact that it is in a person of color’s self-interest to judge that persons of color should not be targets of discrimination should hardly count against her being said to know that people of color should not be discriminated against (Kelly and McGrath 2022: 31). Bias does not always undermine knowledge and may sometimes aid acquisition. As Rini has pointed out, being partisan and biased in one’s data acquisition toward reliable sources can increase one’s knowledge in an environment saturated with fake news and misinformation by preventing your judgments from being falsely undermined (Rini 2017). So, while bias often undermines knowledge, it is not an omnipotent evil demon. It is not a cultural and sociological reason why some individuals suffer from it in some contexts more than others. Insofar as it ignores these reasons, the argument from bias is not just unpersuasive, and it seems naïve. I now turn to considerations concerning bias that can motivate a more realistic skepticism about our political judgments.

12.5  The Argument from Epistemic Colonization Thus far, we have observed it is surprisingly difficult for the political skeptic to mount arguments that are both extensive enough to target a variety of political judgments yet not so general as to support a global skepticism. I now want to turn to a third argument that would seem better suited to achieve that balance. It also concerns a kind of bias – but a kind of systemic bias.

242  Michael P. Lynch Systemic bias is usually understood as occurring whenever a system or institution consistently produces outcomes that favor one group or set of individuals over others. These products can include beliefs and judgments. Thus, a legal system is systemically biased when it generates legal judgments that favor whites over black and brown people. A financial system is when it consistently produces financial judgments that do the same. It is well-known that liberal democracies and their institutions are rife with systemic bias – what we might call institutional bias. Institutional bias is best seen as corruption – a rot that eats away at institutions and produces unjust outcomes. It is not all that difficult to spot, but it is challenging to stop – because those who benefit from those systems (whites) tend to be those in power. But there is another kind of systemic bias that lies farther below the surface and is accordingly more challenging to even expose. This kind of bias operates not at the level of institutional systems but in conceptual systems. It comes in different forms, but since it involves or is the result of kinds of systemic bias, and bias of that sort is a kind of corruption, I will call it epistemic corruption. A conceptual system is epistemically corrupt when its concepts are consistently used to produce unjustified and false judgments on some range of subjects that favor one group of people over another.13 So understood, epistemic corruption is an element in other forms of systemic corruption since it works at the level of how people think, and how people think affects how they act. The corruption of institutional bias is at least partly a result of epistemic corruption. Until recently, analytic philosophers have not been focused on the problem of systemic bias in our conceptual frameworks or the power dynamics involved in epistemic practices generally. The intellectual traditions most focused on the problem have been feminist, queer, and decolonial theory. What this literature has repeatedly shown us that, as with other forms of systemic corruption and bias, there are several different cultural mechanisms at work that help to explain how and why we may suffer from epistemic corruption. Here, I will focus on just one, which I will call, following Franz Fanon and philosophers Nelson Maldonado-Torres, Walter Mignolo, and Lewis Gordon, epistemic colonization.14 Epistemic colonialization occurs when one group, through systems of education, literary norms, and legal power, comes to impose a particular set of concepts and norms of rationality on another group. Over the last five hundred years, the literal imposition of what Fanon called “Greco-Latin” or Eurocentric standards and concepts on other cultures has had that effect. However, the impact extends to the colonizers as well, since assumptions about what philosophies, political ideas, and histories are essential to understand – and which can be ignored – impact not just what the colonized believe but what the

Political Skepticism, Bias, and Epistemic Colonization 243 colonizers themselves take for granted. Moreover, as Gordon notes, the point extends even further: Fanon observed in Black Skin, White Masks that this kind of colonization is radical because it enchained what people think and how they think. We could say the same thing about norms and normative thought – namely, our understanding of what it means to be good, to do things right, and to make the world such that is, upon reflection, the best we could and ought to achieve. (Gordon 2021: 34) Gordon’s point is that epistemic colonization constrains what we think is morally and politically possible. As another white man once said to me about racism when, many years ago, I was teaching in Mississippi, “that is just how it is.” He meant that racism was an inevitable product of how humans think and could not change. This lack of hope is the political outcome of epistemic colonization, affecting, as Fanon well knew, the colonizer and the colonized. Epistemic colonization does not just impose concepts. Like colonization generally, it also takes resources from the colonized, in this case conceptual resources. It can prevent people from having the conceptual tools necessary to recognize or understand their experiences. This phenomenon was familiar to philosophers of the race from Anna Julia Cooper to Du Bois and Fanon and has been discussed widely under the label of “the epistemology of ignorance”; more recently, it has been called “hermeneutical injustice” by Fricker.15 This conduct results in epistemic corruption because lacking concepts like “sexual harassment” or even “institutional bias” can not only lead people to believe that there is no sexism or racism in contexts where there is, but also lead to the result that Gordon emphasizes above: a limiting of the political imagination and hope. You cannot imagine how things could get better if you don’t understand how bad they are in the first place. Epistemic corruption, in whatever form it takes, can motivate a kind of political skepticism. If we cannot rule out that our present political judgments are the result, or even just partly the result, of epistemic corruption, we may doubt whether they are fully justified – and hence whether they qualify as knowledge. We should take this kind of political skepticism seriously. It is, after all, based on a simple thought: human judgments are always shaped by the concepts those judgments employ. And our concepts are often shaped by processes some of which are beyond our control. As Wittgenstein emphasized, no one makes their concepts – we inherit them from the community in which we live and the political ideologies that normatively govern our actions and thoughts. Moreover, the history of humanity, and the history of colonization, strongly suggest that epistemic corruption

244  Michael P. Lynch happens to the oppressed and the oppressors alike. It would be astonishing if our political concepts were not at least partly corrupted. Yet while I consider epistemic corruption as a reason to take political skepticism seriously, I do not think that the kind of skepticism it motivates is unhealthy. Nor is it particularly friendly to any of the three kinds of skeptical politics I canvassed above: Montaigne’s quietism, Hayek’s liberation individualism, or Cohen’s epistemic autocracy. Nor, as I will argue, should it cause us to give up on the goal of democratic consensus or the ideal of truth in democracy. Indeed, I think it should cause us to take that idea seriously. Let us start by examining the kind of skepticism that the problem of epistemic corruption would seem to motivate, what that kind of skepticism might be used to justify, and what it cannot justify. The first thing to observe is that epistemic corruption comes in degrees. It is not an all-or-nothing matter; thus, the skeptical challenge it presents is neither. The degree to which forces like racial ignorance or colonization corrupt our concepts will vary from time and place and individual. Moreover, in almost every case, it is implausible that an individual’s concepts are so corrupted that every judgment someone makes about politics will be false or unjustified. Just as judicial corruption due to institutional bias can nonetheless leave many legal judgments untouched, so too with many political judgments that might result from a corrupted conceptual system. So the kind of skepticism motivated by the problem of epistemic corruption is a mediated one, according to which we must acknowledge that many of our political judgments – including perhaps our Lockean intuitions about the property and Rawlsian intuitions about justice – may be less justified than we thought. Further, it may even warrant a slightly stronger conclusion, namely, that our political judgments may always be, at best, only partially justified. Seen from the perspective of real-politic, this is an unsurprising result. After all, as anyone engaged in actual political decision-making knows, politics always requires making high-stakes decisions in limited information states. We never know enough to be anywhere close to certainty on most of our everyday political decisions because, for example, we lack insight into candidates’ true motivations, we cannot predict all the consequences of our policies, and we do not know what other states actors might do in the future, and so on. So, the fact that on top of all the other reasons, we must add the probability that we are victims of epistemic corruption should not cause us to be more skeptical in total about our political judgments but be skeptical in a different way. It tells us that we must be more skeptical of our ways of thinking. Being skeptical about our own patterns of thought can be a good thing from the point of view of political knowledge. That is because awareness of epistemic corruption should encourage those who benefit from it to

Political Skepticism, Bias, and Epistemic Colonization 245 adopt a more intellectually humble attitude toward politics. To be intellectually humble, in a sense I intend here, is to (i) own one’s intellectual limitations16 and (ii) be willing to learn from other people’s testimony and experience – to be willing to revise one’s judgments and attitudes considering evidence supplied by others. Thus, intellectual humility is both a self-regarding and an other-regarding attitude. It concerns seeing oneself as a limited cognitive being, capable of being affected by bias and epistemic corruption. Still, it also means being motivated to listen to others – that is, to think that one alone cannot know it all. As Fanon might have said, white people have a hard time being intellectually humble around black people (Fanon 1952). Humility, not to mention reason, as Fanon reminded us, takes flight whenever blacks enter the room. However, that is precisely why it is essential from the standpoint of democratic political action. As Gordon notes, what epistemic colonization teaches us about democracy is that one must act “based on what one cannot know and despite what one thinks one is” (Gordon 2021: 55). Awareness of epistemic corruption can be liberatory precisely because it can help encourage more democratic attitudes.17 From this point of view, it would be a mistake to think that such awareness should cause us instead to investigate politics or dig in on Hayekian individualism. That would seriously miss the point. The point of revealing the corruption of any form – legal, financial, or epistemic – is liberatory, to blow the whistle and alert us to harms and injustices. Shrugging one’s shoulders and slinking off to the “arms of learned virgins” (as Montaigne famously put it) or accepting a kind of individualism that exemplifies the very essence of colonizing thinking (as Hayek essentially does) is just entirely at odds with the spirit of decolonizing, the spirit of uncorrupting our political thinking. Nor does the problem of epistemic corruption add any comfort to those like Caplan. They wish to point out that ordinary individual voters are frequently massively wrong at coming to the correct answer (even where the “correct answer” means, as it indeed does not always, “what is in their best economic interest”). That may be so, but the problem of epistemic corruption provides little comfort for the position that we should give more power to elite decision-makers. That is because epistemic corruption, where it happens, corrupts both the ordinary and the elite alike. Where a mechanism for such corruption like colonization is at work, all of us can fall into adopting expectations, norms, and concepts that put, for instance, white men in a better position than black or brown women or men. What epistemic corruption does suggest is that Nietzschean worry is a serious one – that deep political disagreement is unlikely to be resolved by appeal to individual reason alone. Such disagreement tends to concern issues of power between groups: between whites, for example, and people of color. By limiting our imagination, epistemic corruption due to colonization makes it harder to reach justified resolutions to disagreements via

246  Michael P. Lynch the rational consensus of the individuals involved. Of course, “harder” is not “impossible,” but it means that if we want to combat epistemic corruption, focusing just on the individual epistemic agent is a mistake. We need more democratic social-epistemic practices and institutions – practices that correct for, and not just replicate, the effects of epistemic corruption. I do not assert that constructing and promoting such practices will bring us consensus in the face of disagreement. Rorty, following Dewey, hoped that a commitment to democratic ideals like solidarity might play that role (Rorty 1989). However, the politics of this century are repeating the lesson of the last one: namely, that consensus – or more pointedly, the consensus amongst the privileged – is often best achieved by appeals to fascist and nationalist myths of racial superiority. In the face of fascism and fake news, appeals to truth often fail in building consensus. Crucially, however, the utility of political truths does not lie solely in their use in consensus-building. What we should take away from the rise of disinformation and global fascism is that the pursuit of truth in politics is best seen not as a guarantee of consensus but as a protective bulwark against anti-democratic forces, and it is for that purpose that combatting epistemic colonization via the construction of better social-epistemic practices is crucial. How to do this, of course, is a delicate further question. I conjecture that it would involve developing, first, more inclusive practices. As Landemore has persuasively argued, “inclusive deliberation, that is deliberation involving all the members of a given group, has greater epistemic properties than less-inclusive deliberation, involving only a few members of a group” (Landemore 2013: 90ff). Second, to develop better practices, we need to embody attitudes like intellectual humility as regulative ideas within them. A social practice embodies an attitude as a regulative ideal just when the activities constitutive of that practice are guided by the idea that participants ought to adopt that attitude (Lynch 2019). Developing these ideas is a task beyond the scope of our discussion. However, the practices I have in mind are those crucial for the expansion of public epistemic trust and include those embedded in historical and scientific inquiry (archival techniques, experimental replication, peer review); journalistic standards (using more than one source); dialogue techniques (having empathy, giving everyone a chance to speak, listening); and legal investigation (appealing to reliable evidential techniques, examining the motivation of witnesses). An overlooked but obvious fact is that these sorts of practices exist because we recognize that our epistemic assessments are often flawed (Allen and Lynch 2022). Biases are hard to spot – that is why they are biases, and appeal to informational checks and balances is a way to compensate for that fact (Lynch 2019). In following practices like these, we strive to play by rules meant to correct our fallibility, compensate for our biases, and implicitly encourage us to

Political Skepticism, Bias, and Epistemic Colonization 247 take a more reflective view. By giving and asking for reasons that emerge from such practices, we sustain and participate in the democratic space of reasons itself – a space that remains open to dialogue even when we are not. I began this chapter by noting that political skepticism does not challenge the existence of political truth so much as its relevance – simply because if we cannot know the truth, we are, as Rorty might have said, “aiming at a target we cannot hit” (Rorty 2021). I urge a different conclusion: far from undermining the relevance of political truth, awareness of the mechanisms of epistemic corruption should encourage us to adopt a more mitigated, engaged skepticism toward ourselves and our practices. This modestly skeptical and intellectually humble attitude is predicated on taking truth more rather than less seriously. What epistemic corruption teaches us is a lesson that, like most important lessons in life, we need to learn again and again. Politics is an uncertain business, and if we want to figure out our best collective path forward, we need to build social-epistemic institutions and practices that can help us correct our all too human flaws. The struggle against the forces of epistemic colonization is the struggle for democracy.18

Notes 1. In many of his works across his long career, Rorty (1979, 2021) argued that truth was either too metaphysically vague or too barren a concept to use as a political ideal. In this sense, he could be considered sympathetic to the political skeptic. However, as I see it, Rorty’s ultimate view is not skepticism about our capacity to know political truths but the concept of truth (in or out of politics) itself. Since I have taken up Rorty’s arguments elsewhere (Lynch 2004, 2012), I will set them aside here. 2. This usage has some of its roots in Zaller’s The Nature and Origins of Mass Opinion (Zaller 1992). 3. Orwell (2009: 282). 4. Political meaning, so defined, is a species of what Lessig (1995) and Haslanger (2014) call social meaning. 5. As I have noted elsewhere (Lynch 2004), we can employ comparable arguments concerning disagreement to motivate relativism about various kinds of truths. However, here, I will concentrate on skepticism. 6. Another approach, often associated with Berlin (1958) (Two Concepts of Liberty), centers on the thought that specific political values are incommensurable – unable to be evaluated along the exact dimension of evaluation. If, for instance, there is no way to settle the classic debate of whether liberty or equality is paramount, one might believe this leads to general political skepticism. Nevertheless, while the debate about whether this form of value pluralism persists, there is no consensus that it leads to skepticism rather than relativism about political values. 7. Montaigne (2003). 8. See Kahneman (2011). 9. See Kahneman (2011, especially 2011: 105). 10. Kornblith (2012).

248  Michael P. Lynch 11. As we have already noted, Caplan (2007) arguably employs a similar argument: see Kornblith (2012) and Saul (2013) for explicit discussions of the epistemic impact of bias. The present argument draws on Allen and Lynch (2022). 12. One might complain that the argument establishes too much – like the argument from disagreement, it may seem to undermine all knowledge, not just political knowledge. See Allen and Lynch (2022) for discussion. In any event, there are plenty of other reasons to reject the argument. 13. Kidd (2019) has recently and usefully appealed to the idea of epistemic corruption to help analyze the effects of oppression and education. Kidd’s focus, however, is on a kind of corruption that inhibits the development of intellectual virtues and encourages intellectual vices in individuals. 14. Maldonado-Torres (2008), Mignolo and Walsh (2018), and Gordon (2021). 15. See, for example, Sullivan and Tuana (2007) and Fricker (2007). Medina’s summary is an excellent overview of the connection between these works of literature (Medina 2017). 16. See Whitcomb et al. (2015) and M. Lynch (2019). 17. Here I am influenced by Heather Battaly’s important paper on this topic (Battaly 2020). 18. For help with this paper, I thank the editors, a reviewer, Teresa Allen, Nimra Asif, Eric Berg, and Lewis Gordon.

References Allen, Teresa, and Michael Patrick Lynch. 2022. “Can We Be Reasonable? Bias, Skepticism and Public Discourse.” In Reason, Bias and Inquiry, by Nathan Ballantyne and David Dunning, 82–106. Oxford: Oxford University Press. Battaly, Heather. 2020. “Can Humility be a Liberatory Virtue?” In The Routledge Handbook of Philosophy of Humility, by M. Alfano, A. Tanaseni and M. P. Lynch. London: Routledge. ———. 1958. Two Concepts of Liberty. Oxford: Clarendon Press. Caplan, Bryan. 2007. The Myth of the Rational Voter: Why Democracies Choose Bad Policies. Princeton: Princeton University Press. Fanon, Franz. 1952. Black Skin, White Masks. New York: Grove Press. Fricker, Miranda. 2007. Epistemic Injustice. Oxford: Clarendon Press. Gordon, Lewis. 2021. Freedom, Justice and Decolonization. London: Routledge. Haslanger, Sally. 2014. “Social Meaning and Philosophical Method.” Proceedings and Addresses of the American Philosophical Association 88: 16–37. Hayek, F. A. 1944/2007. The Collected Works of F.A. Hayek, Vol II: The Road to Serfdom. Chicago: University of Chicago Press. Kahneman, Daniel. 2011. Thinking Fast and Slow. New York: Farrar, Straus and Giroux. Kelly, Thomas, and Sarah McGrath. 2022. “Bias: Some Conceptual Geography.” In Reason, Bias and Inquiry, by Nathan Ballentyne and David Dunning, 11–33. Oxford: Oxford University Press. Kidd, Ian. 2019. “Epistemic Corruption and Education.” Episteme 16 (2): 220–235. Kornblith, Hilary. 2012. On Reflection. Oxford: Oxford University Press. Landemore, Helen. 2013. Democratic Reason. Princeton: Princeton University Press.

Political Skepticism, Bias, and Epistemic Colonization 249 Laursen, J.L., and G. Paganni. 2015. Skepticism and Political Thought in the Seventeenth and Eighteenth Centuries. Toronto: University of Toronto Press. Lessig, Lawrence. 1995. “The Regulation of Social Meaning.” University of Chicago Law Review 62 (3): 943–1045. Lynch, Michael. 2019. Know-it-All Society: Truth and Arrogance in Political Culture. New York: Norton Liveright. Lynch, Michael P. 2012. In Praise of Reason: Why Rationality Matters for Democracy. Cambridge: MIT Press. ———. 2004. True to Life: Why Truth Matters. Cambridge: MIT Press. Maldonado-Torres, Nelson. 2008. Against War. Durham: Duke University Press. Medina, Jose. 2017. “Epistemic Injustice and the Epistemologies of Ignorance.” In The Routledge Companion to the Philosophy of Race, by Linda Alcoff, Luvell Anderson, and Paul Taylor, 247–260. New York: Routledge. Mignolo, Walter, and Catherine Walsh. 2018. On Decoloniality. Durham: Duke University Press. Montaigne, Michel de. 2003. The Complete Essays. Edited by M. A. Screech. London: Penguin. Orwell, George. 2009. Facing Unpleasant Facts: Narrative Essays. New York: Mariner Books. Payne, B. K. 2001. Prejudice and Perception: The Role of Automatic and Controlled Processes in Misperceiving a Weapon. Journal of Personality and Social Psychology 81 (2): 181–192. Pronin, E., and K. Schmidt. 2013. Claims and Denials of Bias and Their Implications for Policy. In The Behavioral Foundations of Public Policy, by E. Shafir, 195–216. Princeton, NJ: Princeton University Press. Rini, Regina. 2017. “Fake News and Partisan Epistemology.” Kenney Institute of Ethics Journal 27 (Number 2 Supplement): 43–64. Rorty, Richard. 1989. Contingency, Irony and Solidarity. Cambridge: Cambridge University Press. ———. 1979. Philosophy and the Mirror of Nature. Princeton: Princeton University Press. ———. 2021. Pragmatism as Anti-authoritarianism. Cambridge: Harvard University Press. Saul, Jennifer. 2013. “Skepticism and Implicit Bias.” Disputatio 5 (37): 243–263. Schmitt, Carl. 2007/1932. The Concept of the Political. Chicago: University of Chicago. Shaw, Tasmin. 2007. Nietzsche’s Political Skepticism. Princeton: Princeton University Press. Somin, Ilya. 2013. Democracy and Political Ignorance. Berkeley: Stanford Law Books. Sullivan, Shannon, and Nancy Tuana. 2007. Race and the Epistemologies of Ignorance. New York: SUNY Press Whitcomb, D., H. Battaly, J. Baehr, and D. Howard-Snyder. 2015. Intellectual Humility: Owning our Limitations. Philosophy and Phenomenological Research 94 (3), 509–539. Zaller, John. 1992. The Nature and Origin of Mass Opinion. Cambridge: Cambridge University Press.

13 Economic Inequalities and Epistemic Democracy Ivan Cerovac

13.1 Introduction Most contemporary political epistemologists appreciate and endorse democracy as a collective decision-making procedure with legitimacy-­ generating potential. While some hail it for its procedural qualities and internalizing the values of freedom and equality, others focus on its instrumental qualities and the epistemic significance of political outcomes it produces, and some incorporate both approaches. Most of these strategies address democracy as an idealized procedure within an equally idealized social environment. However, democratic processes occur in real societies subject to many extraneous and non-ideal social and economic conditions. Economic inequalities frequently negatively affect the qualities that had initially made democracy so appealing. This chapter aims to demonstrate how substantial social and economic disparities reduce democracy’s legitimacy-generating potential and analyze what can be done to preserve it. The chapter proceeds as follows. In Section 13.2, I briefly address the qualities a decision-making procedure must fulfill to be able to generate legitimate political decisions. Writing from the standpoint of an epistemic democrat, I indicate that these qualities can be both moral (aiming for equality in the decision-making process) and epistemic (aiming for the quality of outcomes, or at least for the procedural quality of the deliberative process). I end the first part by underlining a severe concern: significant inequalities in income, wealth, and capital ownership often spill over from the economic and social spheres to the political arena. The considerable effect these inequalities have on decision-making might diminish democratic procedures’ moral and epistemic qualities, impairing their ability to produce legitimate political decisions. Section 13.3. summarizes various spillover mechanisms and approaches wealthier citizens can use to acquire more-than-equal political influence. These might include both actions used to obtain disproportionate impact in the decision-­making process and conduct that constrains the scope of the decisions that democratic governments can produce. Section 13.4. DOI: 10.4324/9781003311003-17

Economic Inequalities and Epistemic Democracy 251 discusses how we can evaluate economic inequalities’ impact on democratic legitimacy. I analyze two distinctly epistemic approaches, liberal epistemic and egalitarian epistemic positions, and provide arguments for endorsing the latter. Finally, in Section 13.5. I sketch two approaches epistemic egalitarianism can assume to remedy the spillover problem. First, we can try to regulate (or even disable) the mechanisms that allow affluent citizens to translate economic and social power into political influence. Second, we can leave the spillover mechanisms unregulated and instead try to diminish existing social and economic inequalities.

13.2 Epistemic Democracy and the Informal Political Sphere Epistemic approaches to political legitimacy assume various forms and endorse democratic and non-democratic political systems.1 The common ground is represented by these approaches’ epistemic standards when evaluating various decision-making procedures. As we can utilize different instrumental and purely procedural epistemic norms, and as meeting the epistemic standard can be the sole requirement or just one of the many requirements a decision-making process must fulfill, I shall simplify the debate by focusing on the standard account of epistemic democracy. This account builds upon Rawls’ liberal principle of legitimacy and argues that a decision-making procedure must be acceptable to all reasonable (or qualified) citizens to have legitimacy-generating potential (Rawls 1993, 137).2 Since the unequal distribution of political influence introduces the invidious comparisons objection (Estlund 2008, 36–37), decision-­making must allocate political power equally while simultaneously organizing the decision-making process in an epistemically optimal way: one compatible with an equal allocation of political influence. Numerous contemporary authors writing in this tradition perceive democracy as a decision-making procedure with legitimacy-generating potential. It distributes political power equally among all (qualified) citizens, meeting the moral criterion (Cohen 1989, Beitz 1990, Rawls 1993, Christiano 2008). It also ostensibly outperforms other fair decision-­ making procedures in placating the epistemic standard. Namely, it exceeds other suitable practices in producing decisions of substantial epistemic quality (Estlund 2008, see also Marti 2006, Landemore 2013, Goodin and Spiekermann 2018, Cerovac 2020) and in fostering deliberation and epistemic virtues (Peter 2011, 2021). Epistemic democracy thus provides two essential criteria by which we can evaluate decision-making procedures: a moral and an epistemic criterion (Estlund 1997).3 However, decision-making procedures occur in the real world, and their ability to placate the two criteria can be significantly affected by the economic and social circumstances in which decision-­ making ensues. While some procedures might meet the two criteria under

252  Ivan Cerovac ideal conditions and within strictly controlled environments, their ability to treat citizens as equals and secure the decent quality of deliberation and the quality of resulting political outputs might be jeopardized when applied to realistic circumstances. To better understand this phenomenon, we must introduce the distinction between the formal and informal political spheres (Estlund 2008, 189, see Walzer 1983, Stacey 2010, Radnitz 2011). The former includes elections, parliamentary debates, voting in the parliament, decision-making in the executive bodies of the government, as well as many other processes conducted under strict rules and regulations. The latter incorporates political speeches (typically held outside formal institutions), propaganda, public debates and writing to the media, lobbying, citizen rallies and protests, and activities such as producing and distributing visual or literary artworks that spread political messages. Strict egalitarian norms such as the appeal for equal political influence in the decision-making process typically apply within the formal political sphere (for instance, the one-person, one-vote principle4). Despite substantial difficulties, we can apply these norms within the formal political sphere. The decision-making process in this sphere can mirror the ideal democratic procedure that meets both the moral and the epistemic criterion. However, trying to implement strict egalitarian norms requiring the equality of political influence5 in the informal political sphere comprises many difficulties. Namely, the free exercise of civic liberal rights, including personal and economic liberties, can result in citizens having profoundly unequal political influence in the informal political sphere. Insisting on the equality of political impact in the informal political sphere would thus entail a systematic denial of some rights and liberties. Such a claim might give rise to the infamous leveling-down objection (Christiano 2008) when citizens end up with an equal but lower level of well-being than they would have had the inequality been preserved.6 Contemporary discussions thus differentiate between formalist approaches, where the equality of political influence is preserved only in the formal political sphere, and substantivist or egalitarian approaches, where the plea for equal political power applies to the formal and the informal political sphere. Note again that the debate is not about the distribution of resources in general but the distribution of capabilities for exercising political influence. Substantivist approaches do not necessarily object to general inequalities in the economic or social sphere. They locate the problem in the spillover of inequalities from economic and social spheres into the informal political sphere and subsequently from the informal into the formal political sphere (Walzer 1983, see also Sunstein 1994, Robeyns 2019). Substantivist approaches rest on the spillover thesis – their arguments make sense only if there are good reasons to believe that considerable inequalities in income and wealth can cause disparities in political influence, which undermine the moral value of democratic procedures.

Economic Inequalities and Epistemic Democracy 253 Additionally, epistemic substantivists endeavor to demonstrate that vast inequality in the economic and social spheres might also undermine democratic procedures’ epistemic value, either by reducing the epistemic value of the deliberative process itself or by decreasing the quality of political decisions produced using democratic systems.7 The following section of the chapter discusses and analyzes various mechanisms the rich can use to translate financial power into political power.

13.3  Spillover Mechanisms Political influence can manifest itself in various spheres of life and at different stages of the collective decision-making process. Many authors have analyzed and written extensively on these mechanisms (Beitz 1990, Rosenstone and Hansen 1996, Knight and Johnson 1997, Lijphart 1997, Christiano 2010, Campante 2011, Christiano 2012, Machin 2013). First, before the decision-making process even begins in its traditional form, citizens can use financial power to fund campaigns influencing public opinion. Second, money can support political campaigns and activities shaping voting behavior during the electoral process. Such conduct might affect the elections’ outcome and motivate politicians and political parties to embrace values and policies that will help them draw additional funding. Third, financial power can also affect the feasibility of some democratically chosen aims by making them counterproductive and obsolete and practically vetoing some proposals. Since there are considerable inequalities in the distribution of income, wealth, and ownership of productive assets, the concern is that, by using these mechanisms, financial imbalances might and often do cause considerable inequalities of political influence in the informal political sphere. 13.3.1  Influencing the Public Opinion Civic opinions and preferences regarding most political issues are typically formed and modified far before the formal deliberative decision-­ making process takes place. Apart from inputs within their intimate sphere (e.g., family and friends), citizens receive and are called to evaluate much information coming from the informal political sphere, including mainstream media and social networks, citizen rallies, public debates, promotional materials, and even works of art. These inputs typically affect which political issues people consider relevant and thus determine what citizens think about and their attitudes and preferences regarding those issues. For example, the effects of mainstream media (TV and newspapers) on civic political opinions have been extensively studied (Entman 1989, Livingstone and Lunt 1994, Jacobs and Shapiro 2000). Although these effects have substantively decreased in the past thirty years, mainstream media still represents an essential source of information for many

254  Ivan Cerovac citizens (Gurevitch, Coleman, and Blumler 2009). Social networks profoundly influence civic opinions (Bond et al. 2012, Marichal 2016), contributing to the spread of false news (Klein and Wueller 2017), ideological segregation (Flaxman et al. 2016), and epistemic and political polarization (Kubin and von Sikorski 2021). Despite the hope that they might level the playfield, they too have proven to be susceptible to financial power, from paid advertising to bots and fake accounts. As affluent citizens can invest disproportionate amounts of money in promoting their political views, they can exercise political influence to a far greater extent than the average citizen. This influence can “change the ideological climate and what is perceived as sound evidence” (Robeyns 2019, 256). Citizens can use their money to fund research groups and think tanks that produce arguments supporting the funders’ economic, social, and political views. These arguments can later be interpreted as sound evidence and disseminated using mainstream media, social networks, or even privately funded non-­governmental organizations engaged in public advocacy.8 Financial power can thus influence public opinion even before some political issues have appeared on the agenda. It can significantly affect which topics will be discussed publicly and become the object of collective decision-making. The spillover from the economic to the political sphere is not always intentional. Of course, money is sometimes used to advocate for a particular law or policy and sometimes to support a more general political aim or even a political ideology. However, it seems that financial power can turn into political influence even when one lacks the intention to do so. For example, the owner of a publishing house can decide to publish only novels where the protagonists exhibit some personal and social virtues (for instance, independence and self-reliance, or selflessness and care for others). These novels might carry political messages or affect how people think about politics, even if the author or the publisher did not conceive them as politically relevant works. Although the reasons for the publisher’s decision might have been purely personal or aesthetic, having the power to decide which novels will be published gives one disproportionate political influence. However (as discussed later in the chapter), sometimes equalizing political influence entails actions that so severely affect civic fundamental rights and liberties that no well-intended liberal (or even egalitarian) thinker would consider such actions justified (Estlund 2008). For now, it suffices to say that, even before something becomes a public issue and even before it is put on the political agenda, inequalities in income and wealth enable some citizens to have a disproportionate influence on others. 13.3.2  Funding Political Campaigns The focus now turns from influencing public opinion in general (where the political struggle need not be yet mirrored in the formal political

Economic Inequalities and Epistemic Democracy 255 process) to impacting the results of a collective decision-making process once the relevant issues have been put on the agenda. Campaign contributions are probably the most direct example of the influence transfer from the economic to the informal political sphere. Of course, contributions do not directly influence the electoral process – votes are what win elections. However, political campaigns consist of various actions that aim to influence civic voting behavior, and none of these actions are cheap or easy to perform. Running for office and acquiring civic support (as well as the support of fellow party members and other politicians) requires the means to present and explain one’s views (along with reasons and arguments supporting those views) to others. In large communities, such a task can only be completed by using paid advertisements, organizing public rallies, appearing in the media, and employing public relations experts. Vast economic inequalities imply that these means are not equally available to all citizens.9 Those having more income and wealth will, other things equal, “have access to better means for conveying their political message to the public than poor candidates” (Cerovac 2020, 215). Of course, most contributions politicians receive for their campaigns are donated by other citizens. However, wealthier citizens can contribute more, thus exercising more significant political influence in the informal political sphere. Consequently, politicians will be motivated to shape the political aims and the public policies they are advocating to meet better the preferences of those who can donate significantly more (Campante 2011).10 Since the affluent are not a homogeneous group, the preferences the politicians try to meet need not be similar (and can be, and often are, conflicting and incompatible). This fact does not seem to alleviate the original problem, as political influence in the informal political sphere remains unequally distributed. Is unequal political influence concerning campaign contributions genuinely a significant threat to political equality? Christiano (2012, see also Robeyns 2019) explains why the affluent have more significant capabilities to donate money to political campaigns. Due to the decreasing marginal utility of money, rich people are left with far more money than the poor when they meet their basic needs. Donating money to political campaigns thus does not represent a relevant loss of utility. This explanation is supported by empirical evidence indicating that most people who contribute to political campaigns are “concentrated at the upper echelons of the distribution of income” (Christiano 2012, 245, see also Ansolabehere, de Figueiredo, and Snyder 2003). Besides, recent research suggests that representative bodies tend to systematically neglect the perspectives and interests of the citizens ranked in the lowest two-thirds of the income scale (Bartels 2016). Gilens and Page present similar (or even more worrisome) results. They argue that economic elites “have substantial independent impacts on US government policy, while average citizens and mass-based interest groups have little or no independent influence”

256  Ivan Cerovac (Gilens and Page 2014, 564). This spillover from the economic to the political sphere is seldom unintentional. Citizens can fund political campaigns because they believe that this action will contribute to the realization of their political aims and values.11 Even those who might have some non-political motives (for instance, helping a good friend who happens to be a candidate) understand that their actions represent the exercise of political influence in the informal political sphere. On the one hand, since campaign contributions represent a straightforward and clear exercise of political impact, many countries regulate them. On the other hand, since donating money to political campaigns is often understood as exercising the freedom of speech, regulation tends to balance these two values: the equality of political influence and freedom of speech. I raise some doubts regarding this approach later in the chapter. 13.3.3  Imposing Structural Constraints Finally, citizens with disproportionate financial power can exert substantial influence on political decisions without exercising unequal political impact in the decision-making process. Namely, excessively affluent citizens (and especially those owning productive assets, as well as those controlling financial capital) can influence the feasibility conditions by exercising their property rights in a way that makes some political aims complex to achieve (Christiano 2010, 201). In other words, citizens controlling capital can sometimes “subordinate decisions and actions of the state” to their own financial decisions (Cohen 1989, 28). For example, suppose the democratic government aims to increase the minimum wage to reduce inequalities and improve the well-being of the worst-off citizens. In that case, owners of firms respond by laying off workers, thereby reducing the well-being of the worst-off. The government’s minimum wage policy thus becomes self-defeating. Similarly, the government’s approach to introducing more progressive taxes to fund healthcare or education can be met with capitalists’ decisions not to invest as much as before (since they expect their gains will be taxed away), and this might, in turn, reduce economic growth and decrease the funds that the government can spend on healthcare or education.12 Wealthy citizens’ ability to render policies they disagree with self-­ defeating can significantly limit the scope of the democratic government’s mandate. Citizens can exercise this political influence directly in the form of a threat or a warning. For example, in a recent annual report, Meta, the company owning both Facebook and Instagram, indicated that it is considering shutting down its services in Europe if it cannot keep transferring user data back to the United States. The warning came as a response to the legal opinion issued by the European Court of Justice, following which the current practices do not adequately protect the privacy of European citizens.13 In many cases, however, political influence

Economic Inequalities and Epistemic Democracy 257 is exercised indirectly. The government expects capital owners to act in a certain way (one that will make the policy self-defeating) and is thus motivated not to implement policies that might result in capitalists exercising their property rights. As with campaign contributions, democratic states are again facing a dilemma. They must choose between the free exercise of property rights, on the one hand, and political equality in the informal political sphere, on the other hand. Some insights into this debate are presented in the remainder of the chapter. For now, it suffices to say that the unequal distribution of wealth and ownership over productive assets can generate inequalities in influence over collective decisions.

13.4 How Do Political Inequalities Endanger Democratic Legitimacy? This section aims to indicate how each spillover mechanism reduces the moral and the epistemic value of democracy, thus endangering its legitimacy-generating potential. The task is to demonstrate that democracy, characterized by substantial economic inequalities that spill over to the political sphere, fails to be a decision-making procedure that meets the liberal principle of legitimacy, that is, a process that can be endorsed by all reasonable (or qualified) citizens as free and equal (Rawls 1993). Traditionally, epistemic arguments against wealth-induced inequalities in political influence can go along two (often intertwined) lines. First, considerable disparities in political power can reduce the intrinsic (procedural) quality of political deliberation. Discussion and the exchange of reasons and arguments help us not only make correct, efficient, and just political decisions but also learn from deliberation and other people’s views.14 In such well-ordered deliberation, “no power except for the forceless force of the better argument” is exercised (Habermas 1990, 185). However, suppose one’s political influence in collective deliberation depends partly on one’s income and wealth (as it often does, considering that money helps convey a political message using media campaigns, public relations experts, and rallies). In that case, decision-making procedures fall short of the deliberative ideal. Additionally, since money for political campaigns typically comes from contributors, it will urge elected representatives to vote for laws and policies that advance donors’ interests, not for political decisions supported by the best reasons (Beitz 1990). Second, inequalities in political influence can simultaneously impair the instrumental epistemic value of democratic deliberation. When a small group of wealthy citizens can disproportionately influence public opinion, affect the decision-making process, or put structural constraints on democratic decisions, we can expect that politics will be shaped to disproportionately address perspectives and promote the interests of this small group. Recall Bartels’ (2016) thesis that votes in the US Senate show little to no responsiveness to the perspectives and interests of the

258  Ivan Cerovac citizens ranked in the lowest two-thirds of the income scale. This fact suggests that the government’s policies will likely miss the interests of the majority and the common good. Suppose inequalities in income and wealth tend to spill over from the economic to the informal political sphere (as described earlier in the chapter) and thus produce disparities in the distribution of political influence. In that case, one might be tempted to conclude that, since all citizens should have an equal political impact, the system enabling these spillovers fails to meet the liberal principle of legitimacy. However, it seems that the principle (at least as Rawls intended it15) does not directly call for equality in the decision-making process. It simply calls for equality in an earlier stage, when all reasonable citizens are free and equal to deliberate on the decision-making procedure they can all endorse. This procedure does not have to be egalitarian – if all reasonable citizens can support some non-egalitarian decision-making procedure (for example, one that epistemically outperforms other procedures16), we can say that such a procedure meets the liberal principle of legitimacy (Estlund 2008). However, this implies that we must publicly justify all deviations from equality, and all reasonable citizens must be able to see why (and how) such variations improve the procedure’s epistemic qualities. Can inequalities in the informal political sphere (caused by the spillover of political influence from the economic to the political sphere) be justified in such a way? David Estlund (2000) gives a persuasive epistemic argument conditionally supporting inequalities in the informal political sphere. He characterizes this take as a liberal epistemic position and differentiates it from a libertarian position since the rationale for disparities in the informal political sphere is not moral (i.e., grounded in property rights) but epistemic. Estlund makes a clear distinction between political input and political influence. The latter is a zero-sum game, where an increase in one person’s political influence implies that another person’s political influence is reduced. The former is an approximate non-zero-sum game, where we can increase one person’s political input without reducing some other person’s input. If, for example, civic political input is measured in letters written to their political representatives, and one starts to write two additional letters every year, one’s political input will increase, but that will not reduce the number of letters sent by other citizens.17 Estlund’s central presumption is that more political input generally increases the procedure’s epistemic quality. While democracy and coin-flipping distribute political influence equally among all citizens, we hold that democracy epistemically outperforms coin-flipping because it gives citizens more significant political input. Therefore, we should increase civic political intake to improve the decision-making procedure’s epistemic qualities. Of course, since inequalities in political input produce epistemically harmful inequalities in political influence, we must balance the

Economic Inequalities and Epistemic Democracy 259 two, keeping the total level of political input as high as possible and the distribution of political power as egalitarian as possible. However, how can we do that? Estlund follows Rawls’ difference principle (Rawls 1999, 65–68) and applies it to the informal political sphere, arguing that allowing some inequalities in political input leads to a greater level of total political intake. The state then must regulate imbalances in political input using progressive taxation to reduce disparities in political influence and increase the political intake of the worst-off. In practice, this means that the state should allow campaign contributions (although they introduce inequalities in the informal political sphere) but should tax them and use the acquired funds to increase the political input of citizens who do not (and cannot) donate money for political campaigns. Estlund believes this will increase the total level of political input and improve the political intake of the worst-off while simultaneously keeping the inequalities in political influence within acceptable boundaries. There are reasons to doubt whether such regulation will improve the epistemic quality of political outputs. First, while the message space of politics (where political views can be broadcasted and displayed) is ample and leaves enough room for new inputs, individuals have limited cognitive space to process these messages. Christiano calls this phenomenon socially induced cognitive scarcity (Christiano 2010, 248). Since citizens cannot be realistically expected to use more than one hour a day to consider public and political issues, the total level of political input is somewhat limited. Of course, political candidates can spend donations to hire better public relations experts, fight for prime time during television broadcasts, or outbid paid advertising spaces on social media. However, in the end, this turns out to be a fight for political influence. In other words, after a certain point, an increase in one’s political input might be epistemically meaningless unless accompanied by a rise in one’s political influence. The worry is that, since an agent’s cognitive space is limited, it can be dominated by those with disproportionate political impact. Second, there are doubts about whether we can apply the difference principle to the distribution of political influence. Some argue that it might be appropriate when the distribution of income and wealth is in question as the endorsed inequality might have motivational effects and thus increase the total amount of resources being redistributed. However, the same does not apply in the informal political sphere. We can achieve the increase in the total amount of political input without introducing political inequalities since the resources invested in producing political input (for instance, money to fund political campaigns) comes from the motivational effect of disparities in the economy, not the political sphere (Cerovac 2020). Finally, liberals might argue that the epistemic difference principle fails to meet the principle of reciprocity. Namely, the difference principle is not introduced to regulate political liberties,

260  Ivan Cerovac and the equal fundamental liberties principle, as well as the fair equality of opportunity principle, has lexical priority over the difference principle. By allowing inequalities in political influence, the liberal epistemic position fails to protect the fair value of political liberties (Rawls 1999). A more egalitarian approach (the egalitarian epistemic position) builds upon Estlund’s claim that increases in political input are epistemically rewarding but does not resort to justifying inequalities in political influence to increase the total level of political input. Unlike strict egalitarianism, this position is not concerned solely with political power equality but with the quality of political decisions. However, unlike Estlund’s view, it holds that we can benefit from increased levels of political input without having to introduce epistemically detrimental effects of unequal distribution of political influence. To achieve this aim, the egalitarian epistemic position cannot simply restrict or forbid campaign contributions and other instances where economic power spills over to the informal political sphere. While this would be an egalitarian move, it would disregard the epistemic importance of civic political input. To meet high epistemic standards, the egalitarian epistemic position must increase civic political intake while equally distributing political influence. This provision requires giving the state a prominent role in securing that all citizens share a sufficiently high level of roughly equal political input.

13.5 Egalitarian Epistemic Position and the Role of the State The spillover of inequalities from the economic sphere to the informal political sphere raises legitimacy concerns because inegalitarian systems cannot meet the liberal principle of legitimacy, that is, they cannot be endorsed by all reasonable (qualified) citizens. Additionally, since higher levels of civic political input improve decision-making procedures’ epistemic qualities, not all policies that distribute political influence equally among all citizens can be reasonably endorsed. Coin-flipping, queen for a day,18 and aggregative democracy give all citizens an equal chance to influence the final decision. However, we reject these procedures because they fail to properly harness and utilize civic epistemic contributions. Therefore, the solution is not to equalize civic political influence at the cost of significant reductions in political input. Although imposing substantial restrictions or forbidding many activities in the informal political sphere might balance civic political power, this would have devastating effects on the procedure’s epistemic value. The state must divulge forms of regulation that simultaneously give all citizens an equal level of political influence and keep their political input relatively high.19 Two broad approaches can be used to tackle the spillover problem. First, we can focus on stopping the spillover process and implementing regulations to insulate the political sphere from the harmful effects of

Economic Inequalities and Epistemic Democracy 261 background inequalities. For example, we can introduce laws prohibiting the use of private wealth in political campaigns or raise public funds that will finance political campaigns. This approach allows vast inequalities in income and wealth and targets wealthy citizens’ mechanisms to translate their economic power into political influence. Second, we can focus on reducing the disparities in income and wealth, particularly in control over productive assets. For example, we can argue for a more egalitarian distribution in the economic sphere, insisting on more progressive taxation or even advocating for economic alternatives to capitalism, such as property-owning democracy or liberal socialism. Unlike the former approach, this one is not concerned with the spillover process. If we remove considerable inequalities in income and wealth, there will be nothing to spill over from one sphere to the other. Both approaches have some advantages and disadvantages, and I briefly evaluate them in the rest of the chapter. 13.5.1  Procedural Reform The first approach attempts to tackle the spillover problem from the standpoint of non-ideal theory. Some procedural proposals addressed here are already implemented in the national legislation of many countries, and the central issue becomes not what procedural elements should be introduced but to which extent they should be applied. Charles Beitz categorizes these procedural elements as “ceilings” and “floors” (Beitz 1990, 193–194). “Ceilings” represent limits set on private donations to political campaigns. Since campaign contributions are often regarded as an exercise of the freedom of speech (and sometimes the freedom of association), most countries allow private donations. However, many specify the maximal amount that can be legally donated (Gulzar, Rueda, and Ruiz 2021). “Floors” include state subsidies for political activities. In most cases, party subsidies are when the state funds political parties and their activities. However, some philosophers (Ackerman and Ayres 2004) have suggested that funds should go directly to citizens in the form of vouchers that can be spent to (anonymously) support a candidate or a political organization. While the first model tries to reduce inequalities in political influence by limiting wealthy civic political input, the second model is more focused on increasing the political intake of each citizen and does not address the problems related to inequalities in political influence.20 Since the egalitarian epistemic approach aspires to abolish inequalities in political influence while promoting high levels of civic political input, it must combine “ceilings” and “floors” to achieve its aims. However, while similar strategies are already used in many countries worldwide, the egalitarian epistemic approach would have to put “ceilings” extremely low while simultaneously rising the “floors” to reasonably high levels.

262  Ivan Cerovac This approach implies that the state should subsidize vouchers of considerable monetary value (for instance, Ackerman suggests that each adult citizen should receive a coupon of 50 “Patriot dollars”) and, at the same time, severely limit private donations. Ideally, the space between the two thresholds would be minimal or non-existent, implying that all citizens have equal political influence and political input. We can raise several objections against this proposal, and I consider only the two most significant. First, the egalitarian epistemic approach severely limits the freedom of expression, practically blocking any private donations apart from those made through vouchers. Citizens’ freedom of expression manifests itself in many forms and using personal funds to support one of the candidates or political parties seems like a straightforward form of political expression (Anderson 2000). However, many freedoms must be regulated to be meaningfully utilized and enjoyed. Citizens value the freedom of speech in collective deliberation because it enables all participants to specify their interests and perspectives, thus enabling them to participate as free and equal in the decision-making process (Dworkin 200021). Additionally, some restrictions on the freedom of speech are required to keep collective deliberation sufficiently inclusive and insulate it from the harmful effects of money, thus protecting its epistemic value (Bennett 2020). The very reasons we value the freedom of speech in collective deliberation raise the justification of some restrictions regarding its use. This fact does not call for the censorship of any political view, as the intention is not to strip political influence from some individuals or groups but to distribute political power equally among all citizens. Second, the egalitarian epistemic approach might seem to be vulnerable to a form of leveling-down objection (Estlund 2008). If we prioritize equality of political influence, we might end up endorsing egalitarian models of distribution that support deficient levels of political input, so low that the citizens will be worse off than in some other, non-egalitarian model of distribution. While this can be a valid concern in the economic sphere, we can (and should) use state support to keep the levels of political input sufficiently high (Cerovac 2020, Blau 2021). Since inequalities in the informal political sphere are not the only (and probably not even the primary) source of civic motivation to be more productive in the economic sphere, we can avoid the leveling-down objection by using public funds to ensure high levels of civic input, while simultaneously keeping the economy competitive by allowing inequalities in the economic sphere. The procedural reform proposal and the former two objections raised against it focus on campaign contributions as the central (and maybe only) source of inequalities in the informal political sphere. However, this interpretation lacks and neglects numerous disparities in the informal political sphere caused by other spillover mechanisms. Even if we introduce vouchers for funding political campaigns and impose severe

Economic Inequalities and Epistemic Democracy 263 restrictions on private campaign contributions, many inequalities will remain unaddressed. Citizens possessing publishing and media houses, citizens donating money to think tanks and non-governmental organizations, and those owning giant firms, factories, and other productive assets will still have disproportionate political influence. Introducing procedural reforms regulating political campaigns and decision-making might not be enough to eradicate inequalities in the informal political sphere. 13.5.2  Social Structure Reform Campaign contributions are just one of many mechanisms prosperous citizens can use to spill their economic power into the political sphere. Owning media outlets and publishing houses on the one hand and productive capital on the other can significantly increase one’s political influence. While contemporary discussions typically address some extreme cases (for example, billionaires who own social networks such as Facebook or Twitter, newspapers such as The New York Times or The Washington Post, or giant firms such as Walmart, Amazon, or Apple), we must consider that simply owning a small publishing house, a local newspaper, or a small hotel that employs twenty people gives you disproportionate political influence in the informal political sphere. Implying that the state should constantly supervise how citizens utilize their property, intervening whenever there is an instance of spillover, seems a hazardous, illiberal suggestion that opens the door to a range of slippery slope arguments (Estlund 2008, 187–189). Additionally, even if we want more state control, some doubt remains that the state will ever be able to stop the spillover between the two spheres (Robeyns 2019, 256). An alternative approach (one addressing the problem from the standpoint of ideal theory) recedes from procedural reforms that aim to stop the spillover and instead focus on reforms changing the economic and social structure. The main thought is that, since there will be no significant inequalities in income and wealth, the state will not have to block or regulate spillover mechanisms. This approach is endorsed by John Rawls, who, just like Robeyns, doubts the state’s ability to block spillover mechanisms. He indicates that “welfare-state capitalism rejects the fair value of political liberties” and “permits enormous inequalities in the ownership of real property (productive assets and natural resources) so that the control of the economy and much of political life rests in few hands” (Rawls 2001, 137–138, emphasis added). He argues for a property-owning democracy or liberal socialisms, systems characterized by a broad dispersal of (both human and non-human) capital, where every citizen controls roughly equal amounts of money. While this approach seems to offer a more comprehensive solution, it might be targeted by Estlund’s leveling-down objection. Namely, unlike the “ceilings” and “floors” system discussed earlier,

264  Ivan Cerovac this approach entails substantial reforms in the economic and social structure, making somewhat justified the worry that the total level of political input might be reduced.

13.6 Conclusion Considering the harmful effect inequalities in the informal political sphere have on the moral and the epistemic qualities of democratic procedures (and consequently on their legitimacy-generating potential), there is a strong need for reforms that can remedy political inequalities. However, the shortcomings of both approaches indicate no simple solution. While the two approaches represent substantively different projects for reducing political inequalities, applying other methods, and even having quite different goals, their drawbacks suggest that combining efficient elements from both might help temporarily alleviate the problem and render it less severe.

Notes









1. For a comprehensive overview of different epistemic approaches to political legitimacy, see Peter (2011) and Cerovac (2020). Also, for an excellent overview of theories discussing the epistemic qualities of democracy (rather than its legitimacy-generating potential), see Prijić-Samaržija (2018). 2. We can find similar requirements in many other contemporary authors as well. For example, Joshua Cohen argues for a “social order in which the justification of the terms of association proceeds through a public agreement among equal citizens” (Cohen 1989, 30). 3. Numerous authors without a desire to be categorized as epistemic democrats endorse similar criteria. For example, Thomas Christiano writes that a political system should be designed to “make collective decisions that aim for the common good and justice [non-procedural, epistemic criterion] in a way that treats all citizens as free and equal participants [procedural, moral criterion]” (Christiano 2012, 241; see also Cerovac 2020, 53–61). 4. It is a phrase typically accredited to British trade unionist George Howell, who used it in the 1880s. 5. Of course, the egalitarian call is not for equal political influence but equal opportunity for exercising political impact. It is up to citizens to decide whether to use their capability to influence the decision-making process. For more information, see Estlund (2000, 128). 6. The leveling-down objection undermines strict egalitarianism by negating one of its critical arguments with the claim that we should eliminate equality if it implies reducing the well-being of those better-off to the level of the worst-off. It invites us to imagine two alternative states: S1 and S2. In S1, everyone is equally well-off, while in S2, everyone is better off than in S1, but some are better-off than others. Strict egalitarians would have to favor S1 and argue for policies that equalize citizens’ well-being, even if that indicates reducing everyone’s well-being, including the well-being of the worst-off. When applied to political epistemology, this objection implies that strict egalitarians would have to insist on equal levels of citizens’ political input, even if that entails substantively reducing each citizen’s level of

Economic Inequalities and Epistemic Democracy 265











political input and negating some essential rights and liberties. For more information about the leveling-down objection, see Parfit (2002) and Christiano (2008), and for its application in political epistemology, see Estlund (2000). 7. An extensive debate on the epistemic value of collective deliberation significantly surpasses the scope of this chapter. We can find arguments in favor of this value in Landermore (2013 and 2021) and Talisse (2021), as well as in a few chapters within the Democratic Optimism section of this volume. Additionally, collective deliberation in controlled and inclusive environments can help alleviate the harmful effects of epistemic injustice (Samaržija and Cerovac, 2021). For arguments questioning or rejecting this value, see Ahlstrom-Vij (2012) and Brennan (2016), as well as Chapter 2 of this book within this volume’s Democratic Pessimism section. However, although the article assumes that public deliberation has some epistemic value, we can effortlessly adapt most arguments to demonstrate the detrimental effect of economic and social inequalities on the epistemic import of aggregative procedures. Using money to influence public opinions, fund political campaigns, and impose structural constraints harms aggregative and deliberative democratic processes. 8. We can find a comprehensive account of how private funds can affect and even dictate academic research (for example, in economics or political science) in Biglaiser (2002) and Stedman Jones (2012). 9. Joshua Cohen, for example, argues that “economic resources provide a material basis for organized political action.” Poor groups thus face profound organizational and political disabilities. See Cohen (1989, 29). 10. Politicians, of course, believe that campaign contributions are only one of the means that helps win elections. Ultimately, their political messages must appeal to the citizens, not only those who can contribute most to politicians’ campaigns. However, they must also consider shaping their political messages to draw the most votes and attract as much funding as possible. Campaign contributions thus lead to “an endogenous wealth bias in the political process since the decisive agent whose preferences will prevail in equilibrium will be wealthier than the median” (Campante 2011, 646–647). 11. There might be numerous motivations for donating money to support a particular political candidate or a political party. However, the main reason seems to be that donors agree with the aims and values (or concrete policies) the candidate or the party is advocating (Magleby, Goodliffe, and Olsen 2018). 12. Another notable example focuses on situations where the government aims to reduce pollution by limiting the emissions of greenhouse gases. This decision can be met with industries’ decision to move the production process to another country, thus increasing unemployment and foiling some other aims the democratic government wanted to achieve. These wellknown examples, as well as a few others, can be found in Cohen (1989), Christiano (2010), and Robeyns (2019). 13. For additional information, see Schechner and Glazer (2020) and Shead (2022). 14. The inequality of political influence does not always have to affect collective deliberation negatively. For example, Mill argues for his plural voting proposal because (among other things) he holds it will prevent class legislation and thus motivate different groups to deliberate within the Parliament (Mill 1977; see also Cerovac 2022). However, Mill also argues against pledges and campaign promises, as well as for (limited) public funding of

266  Ivan Cerovac











political campaigns, because he is aware that money-driven political campaigns can seriously impair the quality of deliberation. For other accounts that emphasize the educative effects of democratic deliberation, see, for example, Peter (2011). 15. Rawls’ liberal principle of legitimacy simply requires that all citizens, as free and equal, can “endorse the society’s fundamental political arrangements” (Wenar 2021). 16. Imagine flipping a perfect or semi-perfect coin (or having a group of experts or a wise AI), which tends to produce correct political decisions far more often than any other decision-making procedure. Additionally, suppose that the coin can do so publicly, that is, all reasonable citizens can agree that it outperforms all other procedures. A procedure in which all political decisions are made by such a coin (or a group of experts, or wise AI) would be able to meet the liberal principle of legitimacy. However, most (or all) citizens would have no political influence whatsoever (Cerovac 2020). 17. Of course, increasing one’s political input does not necessarily increase one’s political influence. If my political input increases and the input of all other citizens remains the same, my political influence will be increased, and other citizens’ political influence will be reduced. However, if my political input increases, but this increase is accompanied by a proportionate rise in other citizens’ political input, my political influence will remain the same. Finally, I can simultaneously increase my political input and reduce my political impact, provided that other citizens’ political input increases disproportionately (i.e., to a greater extent than mine). 18. Queen for a day is another procedure Estlund (2008) introduced to demonstrate that procedural fairness is not the only relevant feature of decision-­ making procedures. Every (adult) citizen can be randomly selected for one day to make all the pertinent political decisions. Estlund intends to show that a decision-making procedure can give all citizens an equal chance to influence the final decision (since every citizen has an equal opportunity to be randomly selected as a ruler) and still be epistemically lacking. A similar argument could be made for aggregative democracy: although all citizens have an equal chance to influence the final decision, their political input is reduced just to their votes, and thus the procedure fails to utilize civic contributions properly. 19. This view, particularly its sufficientarian aspects focused on keeping the level of political input reasonably high, is recently defended by Adrian Blau (2021). 20. However, the aspiration is that the substantial increase in (lower and middle class) civic political input produces epistemic benefits and will improve the representation of civic interests in the political process. For additional information, including doubts about whether this is the case, see Pevnick (2016). 21. John Rawls endorses a somewhat similar position when he argues that all citizens should have a “fair opportunity to take part in and to influence the political process” (Rawls 1999, 224). For an excellent analysis of this position and a new perspective on the conflict between freedom of speech and equality of political influence, see Dawood (2013).

References Ackerman, Bruce, and I. Ayres. 2004. Voting with Dollars: A New Paradigm for Campaign Finance. New Haven: Yale University Press.

Economic Inequalities and Epistemic Democracy 267 Ahlstrom-Vij, Kristoffer. 2012. “Why Deliberative Democracy Is (Still) Untenable?” Public Affairs Quarterly 26(3): 199–220. Anderson, Annelise. 2000. “Political Money: The New Prohibition.” In Political Money: Deregulating American Politics, edited by A. Anderson, 171–185. Stanford: Hoover Institution Press. Ansolabehere, Stephen, J. M. de Figueiredo, and J. M. Snyder. 2003. “Why Is There so Little Money in US Politics?” Journal of Economic Perspectives 17(1): 105–130. Bartels, Larry M. 2016. Unequal Democracy: The Political Economy of the New Gilded Age. Princeton: Princeton University Press. Beitz, Charles R. 1990. Political Equality: An Essay in Democratic Theory. Princeton: Princeton University Press. Bennett, Michael. 2020. “An Epistemic Argument for an Egalitarian Public Sphere.” Episteme 2020: 1–18. Biglaiser, Glen. 2002. “The Internationalization of Chicago’s Economics in Latin America.” Economic Development and Cultural Change 50(2): 269–286. Blau, Adrian. 2021. “Political Equality and Political Sufficiency.” Moral Philosophy and Politics. https://doi.org/10.1515/mopp-2020-0059 Bond, Robert M. et al. 2012. “A 61-Million-Person Experiment in Social Influence and Political Mobilization.” Nature 489(7415): 295–298. Brennan, Jason. 2016. Against Democracy. Princeton: Princeton University Press. Campante, Filipe R. 2011. “Redistribution in a Model of Voting and Campaign Contributions.” Journal of Public Economics 95(7): 646–656. Cerovac, Ivan. 2020. Epistemic Democracy and Political Legitimacy. Cham: Palgrave MacMillan. Cerovac, Ivan. 2022. John Stuart Mill and Epistemic Democracy. Lanham: Lexington Books. Christiano, Thomas. 2008. The Constitution of Equality. Oxford: Oxford University Press. Christiano, Thomas. 2010. “The Uneasy Relationship between Democracy and Capital.” Social Philosophy and Policy 27(1): 195–217. Christiano, Thomas. 2012. “Money in Politics.” In The Oxford Handbook of Political Philosophy, edited by David Estlund, 241–257. Oxford: Oxford University Press. Cohen, Joshua. 1989. “The Economic Basis of Deliberative Democracy.” Social Philosophy and Policy 6(2): 25–50. Dawood, Yasmin. 2013. “Democracy and the Freedom of Speech: Rethinking the Conflict between Liberty and Equality.” Canadian Journal of Law and Jurisprudence 26(2): 293–311. Dworkin, Ronald. 2000. Sovereign Virtue: The Theory and Practice of Equality. Cambridge: Harvard University Press. Entman, Robert M. 1989. “How the Media Affect What People Think: An Information Processing Approach.” The Journal of Politics 51(2): 347–370. Estlund, David. 1997. “Beyond Fairness and Deliberation: The Epistemic Dimension of Democratic Authority.” In Deliberative Democracy: Essays on Reason and Politics, edited by J. Bohman and W. Rehg, 173–204. Cambridge: The MIT Press. Estlund, David. 2000. “Political Quality.” Social Philosophy and Policy 17(1): 127–160.

268  Ivan Cerovac Estlund, David. 2008. Democratic Authority: A Philosophical Framework. Princeton: Princeton University Press. Flaxman, Seth, S. Goel, and J. M. Rao. 2016. “Filter Bubbles, Echo Chambers and Online News Consumption.” Public Opinion Quarterly 80(S1): 298–320. Gilens, Martin, and B. I. Page. 2014. “Testing Theories of American Politics: Elites, Interest Groups, and Average Citizens.” Perspectives on Politics 12(3): 564–581. Goodin, Robert E., and K. Spiekermann. 2018. An Epistemic Theory of Democracy. Oxford: Oxford University Press. Gulzar, Saad, M. R. Rueda, and N. A. Ruiz. 2021. “Do Campaign Contribution Limits Curb the Influence of Money in Politics?” American Journal of Political Science 66(4): 932–946. Gurevitch, Michael, S. Coleman, and J. G. Blumler. 2009. Political Communication—Old and New Media Relationships. The Annals of the American Academy of Political and Social Science 625: 164–181. Habermas, Jürgen. 1990. The New Conservatism. Cambridge: MIT Press. Jacobs, Lawrence R., and R. Y. Shapiro. 2000. Politicians Do Not Pander: Political Manipulation and the Loss of Democratic Responsiveness. Chicago: University of Chicago Press. Klein, David O., and J. R. Wueller. 2017. “Fake News: A Legal Perspective.” Journal of Internet Law 20(10): 5–13. Knight, Jack, and J. Johnson. 1997. “What Sort of Political Equality Does Deliberative Democracy Require?” In Deliberative Democracy: Essays on Reason and Politics, edited by J. Bohman, and W. Rehg, 279–319. Cambridge: The MIT Press. Kubin, Emily, and C. von Sikorski. 2021. “The Role of (Social) Media in Political Polarization: A Systematic Review.” Annals of the International Communication Association 45(3): 188–206. Landemore, Hélène. 2013. Democratic Reason: Politics, Collective Intelligence, and the Rule of the Many. Princeton: Princeton University Press. Landemore, Hélène. 2021. “An Epistemic Argument for Democracy.” In The Routledge Handbook of Political Epistemology, edited by M. Hannon and J. de Ridder, 363–373. New York: Routledge. Lijphart, Arend. 1997. “Unequal Participation: Democracy’s Unresolved Dilemma.” American Political Science Review 91(1): 1–14. Livingstone, Sonia, and P. Lunt. 1994. Talk on Television Audience Participation and Public Debate. London: Routledge. Machin, Dean J. 2013. “ Political Inequality and the ‘Super-Rich’: Their Money or (Some of) Their Political Rights.” Res Publica 19(2): 121–139. Magleby, David B., J. Goodliffe, and J. A. Olsen. 2018. Who Donates in Campaigns? The Importance of Message, Messenger, Medium, and Structure. Cambridge: Cambridge University Press. Marichal, José. 2016. Facebook Democracy: The Architecture of Disclosure and the Threat to Public Life. London: Routledge. Marti, Jose L. 2006. “The Epistemic Conception of Deliberative Democracy Defended: Reasons, Rightness and Equal Political Autonomy.” In Deliberative Democracy and Its Discontents, edited by J. Marti and S. Besson, 27–56. Hampshire: Ashgate Publishing.

Economic Inequalities and Epistemic Democracy 269 Mill, John Stuart. 1977. “Considerations of Representative Government.” In Collected Works of John Stuart Mill, Vol. 19, edited by John M. Robson, 371– 578. Toronto: University of Toronto Press. Parfit, Derek. 2002. “Equality or Priority?” In The Ideal of Equality, edited by M. Clayton and A. Williams, 81–125. New York: Palgrave Macmillan. Peter, Fabienne. 2011. Democratic Legitimacy. New York: Routledge. Peter, Fabienne. 2021. “Truth and Uncertainty in Political Justification.” In Political Epistemology, edited by E. Edenberg and M. Hannon, 64–75. Oxford: Oxford University Press. Pevnick, Ryan. 2016. “Does the Egalitarian Rationale for Campaign Finance Reform Succeed?” Philosophy and Public Affairs 44(1): 46–76. Prijić-Samaržija, Snježana. 2018. Democracy and Truth: The Conflict Between Political and Epistemic Virtues. Milano: Mimesis International. Radnitz, Scott. 2011. “Informal Politics and the State.” Comparative Politics 43(3): 351–371. Rawls, John. 1993. Political Liberalism. New York: Columbia University Press. Rawls, John. 1999. A Theory of Justice. Cambridge: Harvard University Press. Rawls, John. 2001. Justice as Fairness: A Restatement. Cambridge: Harvard University Press. Robeyns, Ingrid. 2019. “What, If Anything, Is Wrong with Extreme Wealth.” Journal of Human Development and Capabilities 20(3): 251–266. Rosenstone, Steven J., and J. M. Hansen. 1996. Mobilization, Participation, and Democracy in America. New York: Longman. Samaržija, Hana, and I. Cerovac. 2021. “The Institutional Preconditions of Epistemic Justice.” Social Epistemology 35(6): 621–635. Schechner, Sam, and E. Glazer. 2020. “Ireland to Order Facebook to Stop Sending User Data to the US.” The Wall Street Journal, September 9th, 2020. https://www.wsj.com/articles/ireland-to-order-facebook-to-stop-sendinguser-data-to-u-s-11599671980 Shead, Sam. 2022. “Meta Says It May Shut Down Facebook and Instagram in Europe over the Data-Sharing Dispute.” CNBC, February 7th, 2022. https:// www.cnbc.com/2022/02/07/meta-threatens-to-shut-down-facebook-and-instagram-in-europe.html Stacey, Jeffrey. 2010. Integrating Europe: Informal Politics and Institutional Change. Oxford: Oxford University Press. Stedman Jones, Daniel. 2012. Masters of the Universe. Hayek, Friedman, and the Birth of Neoliberal Politics. Princeton: Princeton University Press. Sunstein, Cass R. 1994. “Political Equality and Unintended Consequences.” Columbia Law Review 94: 1390–1414. Talisse, Robert B. 2021. “A Pragmatist Epistemic Argument for Democracy.” In The Routledge Handbook of Political Epistemology, edited by M. Hannon and J. de Ridder, 384–394. New York: Routledge. Walzer, Michael. 1983. Spheres of Justice. New York: Basic Books. Wenar, Leif. 2021. “John Rawls.” In: The Stanford Encyclopedia of Philosophy (Summer 2021 Edition), edited by E. N. Zalta. https://plato.stanford.edu/ archives/sum2021/entries/rawls

14 What Political Enemies Are for Robert B. Talisse

14.1 Introduction This chapter explores a moral conflict at the heart of democratic citizenship. It is the conflict between pursuing justice and treating one’s political opponents with the kind of respect that is appropriate among citizens. When the stakes are high in a political decision, we are bound to perceive our opponents as not merely on the other side of the issue but on the unjust side. Accordingly, treating our foes as fellow citizens feels like a capitulation to injustice. Why not forgo the niceties of citizenship and treat opponents as mere obstacles to justice? Why not play democracy to win? Standard accounts in democratic theory ground the requirement to respect opponents in the equal standing of all citizens. The idea is that, except for those who embrace ideas that are unambiguously beyond the pale, citizens are owed recognition, even when we regard them as mistaken about justice. While I do not repudiate this view, it strikes me as insufficient. The duty to recognize the equal standing of one’s fellow citizens is a pro tanto requirement. The question is why one should uphold civic relations with opponents when doing so appears to impede the pursuit of justice. The dynamics of polarization supply a compelling reason why we must uphold civic relations with at least some of our political enemies. Specifically, I claim that we must sustain democracy with our enemies to maintain democratic ties with our allies properly. Although that will be my account of what our political enemies are for, it will become clear that the argument is not crudely instrumental. I will argue that we have moral reasons for sustaining democracy with our enemies. I proceed as follows. In Section 14.2, I spell out the moral conflict sketched above, what I will call the democrat’s dilemma. I will show that the difficulty arises from our sincere effort to meet our civic duties; the issue is inherent to citizenship. In Section 14.3, I present an account of polarization’s threat to democracy. After disambiguating two polarization phenomena, I will show how polarization destabilizes our alliances. DOI: 10.4324/9781003311003-18

What Political Enemies Are for 271 This illustration provides the basis for my response to the democrat’s dilemma. However, having a moral reason to sustain democratic relations with political enemies does not mean it is effortlessly done. Section 14.4 discusses difficulties with respecting those whose politics one despises.

14.2  The Democrat’s Dilemma Let us begin with a commonplace: Democracy is many things. It is a form of government, a kind of constitution, a system of institutions, a decision procedure, a series of activities, and more. At a more profound level, though, democracy is the ideal of self-government among equals. The familiar institutions of democracy – free elections, majority rule, public offices open to all, and the like – reflect the pursuit of a self-governing society of equals. In a democracy, the people are not mere subjects of the government. They are equitable participants in government. They are citizens. Democracy thus is a dignifying proposal. The claim is that a decent, stable, and relatively just society is possible without royals, overlords, and bosses. Democracy contends that people can govern themselves. So understood, democracy responds to a moral problem. Government in any form exercises power over those it governs. In a democracy, this power is shared equally among the citizens. However, one upshot of their equality is that citizens are likely to disagree over how their government should deploy political power. When democracy enacts policy, some citizens will be forced to comply with opposing decisions. Thus, the moral problem is: how can one be subjected to political power without subordination? Democracy answers that even when citizens find themselves on the losing side of a vote, they retain their equal status because they remain participants. In the wake of political defeat, they need not withdraw, resign, or go quietly. They are entitled to continue rebuking, campaigning, and organizing on behalf of their favored ideas and against the standing policies. Accordingly, even when subjected to decisions they oppose, they remain citizens rather than subordinates, each with an equal say and the social standing to contest prevailing choices and policies. The democratic government thus deploys power over citizens without rendering them mere objects; therein lies democracy’s legitimacy. Democracy’s reconciliation of exercises of power with the equal standing of those subjected to it entails that citizenship is a moral office. It involves processes by which citizens collectively direct their government to exert power over themselves, even though they disagree on how that power should be deployed. As exercising power among equals is a morally solemn enterprise, citizenship is an ethically weighty office. Unsurprisingly, citizenship comes with responsibilities, many of which are not strictly legal. Although only citizens could perform the criminal

272  Robert B. Talisse acts of treason or sedition, there also are ways citizens can behave that render them morally blameworthy, reckless, admirable, or decent. For example, it is commonly thought that voting is typically praiseworthy, while not casting a vote is blameworthy. We take a similar stance toward being informed citizens; we believe citizens should follow the news and responsibly form political opinions. Ignoring politics altogether is a mild dereliction, while deliberately consuming blatantly unreliable political information and forming political opinions irresponsibly are considered blameworthy, perhaps seriously so. We hence have a concept of civic duty. I cannot here present a complete account of our civic duties.1 However, what has been said suggests that citizenship involves at least two moral requirements. First, as citizens are members of a self-governing community, they must bear responsibility for their political order. This requirement implies that they must deploy their share of political power in ways that reflect their best understanding of what justice requires. Citizens who vote unreflectively or only based on a crude assessment of their narrow interests are reproachable. In taking responsibility for their political order, citizens are called upon to consider the public good and not merely their interests. In short, citizens are under a moral duty to use their share of political power to pursue justice. A second requirement is that citizens are also members of a community of political equals. Thus, when acting as citizens, they must acknowledge that their fellow citizens are their political equals – lateral partners in self-government, each entitled to an equal political say. In this way, citizens are responsible for one another. This tie means that, when it comes to politics, they owe one another a certain kind of regard, what we might call civic respect. On many accounts, civic respect requires citizens to consult with one another, try to understand each other’s perspectives, and maybe even reason together, even when they vehemently disagree. Sometimes civic respect is also cast as a collection of moral and intellectual character traits, such as open-mindedness, humility, curiosity, and generosity. Important questions loom. However, I need not attempt to define the precise contours of civic respect here. It is sufficient to note that specific modes of political engagement are unfitting among democratic citizens precisely because they express a fundamental disregard for political equality. Current disagreements over practices like de-platforming and canceling demonstrate broad endorsement of the claim that we generally do owe one another civic respect; the debates concern the limits of that requirement. The point of civic respect is tethered to the conception of democratic legitimacy with which this section began. In a democracy, the government must never brutely impose political power on citizens; it must be directed by processes in which each citizen gets an equal say. Accordingly, democratic citizens who find themselves on the losing end of a vote must have been afforded a chance to appeal to their peers. Moreover, even after the

What Political Enemies Are for 273 votes are counted, those on the losing side remain entitled to challenge, question, critique, and protest the prevailing result. They are not reduced to mere subjects that must quietly comply with the dominant view, as their fellow citizens are bound to permit them to object and hear their objections. The democrat’s dilemma arises because these two requirements can clash. In instances of disagreement over political questions that strike us as normatively weighty, we are bound to see our political opponents as not merely wrong about the issue but in the wrong. We are prone to see them as not simply incorrect about what justice requires but as on the side of injustice. We must take it that if they prevail, our society will move further from justice. Accordingly, democratic citizenship invokes an oddly conflicted moral posture. When the political stakes are high, treating our opponents as political equals feels like a concession to them and a corresponding betrayal of justice. In expressing civic respect, we fall short of the requirement to pursue justice. Moreover, pursuing justice seems to recommend that we seek to shut down the opposition, break civil relations with them, and treat them as mere obstacles rather than democratic partners. That is not all. We can strain our political alliances by expressing civic respect for our opponents. Insisting that those on the other side are our political equals entitled to an equal say – and perhaps also a hearing – typically sounds to our allies like a betrayal, a confession that our allyship is at best half-hearted. However, one must join a chorus to have a compelling democratic voice. When expressing civic regard dilutes the potency of our political coalitions, it counteracts our responsibility to pursue justice. Amid this clash, a citizen can plausibly ask why she should bother expressing civic respect for her political enemies. She could affirm that, in general, she is required to show civic respect to fellow citizens, but she could add that in specific cases, showing civic respect detracts from the pursuit of justice. Noting that she is also under the requirement to take responsibility for her political order, she wonders why her commitment to respect her fellow citizens who have demonstrated poor and possibly corrupt political judgment should trump justice. Why shouldn’t she simply suspend democratic relations with her enemies, disregard their perspective, and do whatever she can to secure a political outcome consistent with justice? That is the democrat’s dilemma. Before moving on, a few clarificatory points are in order. First, the difficulty emerges within the office of democratic citizenship. It is not the product of a democratic deficit on the part of the citizen. Instead, it arises from striving to live up to what democracy demands. Second, the dilemma concerns conflicting moral requirements of citizenship. The citizens in the grip of the conflict do not deny that they are subject to two ethical requirements; instead, they seek

274  Robert B. Talisse a reason why civic respect ought to be upheld at the expense of pursuing justice. Accordingly, the dilemma does not address cases where citizens advocate for more radical departures from democratic norms. Our envisioned citizen is not proposing that her political enemies ought to be disenfranchised or prosecuted; instead, she is pointing to a moral reason internal to the office of citizenship for breaking off civil relations with them. Third, the dilemma does not assume that citizens must demonstrate civic respect no matter what. The difficulty does not regard persons who may hold the status of democratic citizenship but who nonetheless espouse grossly anti-democratic doctrines. The question of what stance an authentically democratic citizen should take toward democratically divested citizens is essential, but it differs from the issue at hand. The democrat’s dilemma rests on the observation that political disagreements among democratic citizens are often disagreements about what justice requires; it embeds the premise that political opinions are incompatible with justice, which responsible democratic citizens may espouse. One might seek to dispel the democrat’s dilemma by arguing that in a democracy, the pursuit of justice requires one to uphold civic respect. This maneuver invokes the thought that the justice of a political decision is partly a function of the fairness of the process by which it was made. The argument continues that upholding civic respect is required for procedurally just collective decisions. The democratic pursuit of justice is thus constrained by the requirement to respect the equality of one’s fellow citizens. The dilemma dissolves. Although I am sympathetic to this general line of argument, I see it as insufficiently attentive to the perspective of the democratic citizen. The democrat’s dilemma is not an abstract exercise in reconciling two conflicting directives. It is a conflict from the standpoint of the engaged democratic citizen. It is worth trying to address the citizens who find themselves in the grip of the dilemma rather than dismissing them out of hand. Here things stand: the democrat’s dilemma shows that democratic citizenship is morally conflicted. We are directed to acknowledge as our equal democratic partners those we cannot help but see as agents of injustice. Under political circumstances where the stakes are high and urgent, there is a moral case for suspending democratic relations with our foes. Under such conditions, why not abandon civic respect and pursue justice? Why attempt to sustain democratic relations with political enemies? The dynamics of political polarization supply a compelling moral reason why we must uphold civic respect.

14.3  The Polarization Dynamic Polarization looms large in prevalent diagnoses of political dysfunction. Ezra Klein calls it the “master story” of what’s “awry” in American democracy (2020, xix). However, those who lament polarization rarely

What Political Enemies Are for 275 offer a precise account of what polarization is. If polarization is our “master story,” we need details. First, we must distinguish two phenomena: political polarization and belief polarization. When people lament polarization, they usually talk about political polarization. Political polarization is a metric of the ideological distance between opposed political groups. When it is pronounced, the common ground dissolves. Stalemate, deadlock, and frustration result. Although the idea of political polarization is familiar, “ideological distance” remains vague. It is helpful to distinguish three ways of construing it. We can regard them as distinct sites where ideological fissures emerge. One construal looks at the official doctrines of the opposing groups. In the case of political parties, we look to platforms: Two parties are politically polarized to the degree that their political agendas are opposed or, in some other sense, incompatible. Refer to this species of political polarization as party polarization. Party polarization is endemic in a democracy. Indeed, it can be politically healthy, as it makes citizens’ electoral options salient. However, when it becomes excessive, party polarization can impede government functions. A second construal looks at party leaders and officials. It gauges the extent of unanimity within the party. Opposed groups are highly polarized when their respective constituents include few moderates or bridge-builders. As it focuses on leaders and officials, call this elite polarization. Elite polarization involves the shunning of moderates. It typically is accompanied by the attitude that cooperation with the other side constitutes disloyalty. When elite polarization is pronounced, parties valorize purity. A vernacular for ridiculing moderates develops.2 With moderates sidelined, the hardliners take control. The result is a deadlock accompanied by cross-partisan animosity among elites. A third understanding of ideological distance examines the attitudes prevalent among ordinary citizens who identify with a party.3 Widespread polarization occurs when rank-and-file citizens embrace intensely negative attitudes and dispositions toward those perceived to be politically dissimilar from themselves. As it centers on the attitudes citizens take toward perceived opponents, widespread polarization does not need to track actual policy disputes (Levendusky and Malhorta 2016). Instead, citizens are polarized in this sense simply by virtue of the animosity they harbor for perceived rivals. The data are striking. Widespread polarization has escalated significantly in the United States over the past four decades, even though divides among citizens over policies have either remained stable or eased (Iyengar and Krupenkin 2018; Mason 2018b). Importantly, citizens not only report dislike for affiliates of the opposing party but also see them as

276  Robert B. Talisse untrustworthy, unpatriotic, unintelligent, and dangerous. Furthermore, this animosity is directed at fellow citizens, not only the opposing side’s candidates, officials, and spokespersons (Iyengar, Sood, and Lelkes 2012). Moreover, animosity is generalized. Citizens do not only dislike perceived rivals’ politics; they also find their nonpolitical behavior disagreeable. The clothes they wear, the vehicles they drive, the food they eat, their modes of entertainment, and more all become triggers of contempt (McConnell et al. 2017; Hetherington and Weiler, 2018).4 Widespread polarization expands partisanship into a “mega-identity” or lifestyle (Mason 2018a, 14). Even though widespread polarization has escalated in the absence of commensurate policy divisions, citizens believe that their political differences are significant and intensifying (Bougher 2017).5 They attribute sharply opposing – and radical – views to rivals. These projections are generally inaccurate (Ahler 2014; Ahler and Sood 2018). Heightened levels of popular polarization help explain the pervasiveness of elite and party polarization.6 Elite polarization is incentivized when citizens intensely dislike those outside their political tribe. After all, politicians and party leaders seek to win elections, and they do this by extracting political behavior from their base (Lee 2016). Animosity, distrust, and contempt are motivationally potent (Iyengar and Krupenkin 2018, 215). Accordingly, when party-affiliated citizens manifest intense animosity toward the other side, candidates and party leaders do well to amplify hostility toward their opposition. This, in turn, rewards parties that adopt platforms that punctuate opposition to the rival party’s agenda. Widespread polarization makes for easy campaigning: demonize the opposition, valorize intransigence, and stoke animosity. This feeds back to citizens, fueling adverse effects toward partisan rivals. The three sites of political polarization thus reinforce each other. These circumstances are familiar. Our next question is why widespread polarization is so popular. To understand this, we need to look at belief polarization. Belief polarization is a doxastic shift that occurs within like-minded groups.7 Iterated interactions among like-minded people result in each person adopting a more extreme formulation of their shared view. When we surround ourselves only with others who reinforce our ideas, we tend not only to become more confident in their correctness; we also adopt more exaggerated formulations of them. Interaction with like-minded others transforms us into more extreme versions of ourselves. This shift involves an alteration in our grasp of the basis of our beliefs. We come to overestimate the weight of supporting evidence. We also become more dogmatic, unreceptive to counterevidence, and resistant to correction. We more readily dismiss detractors as benighted. We become less inclined to listen to them and more prone to interrupt when speaking (Sunstein 2009, 44; Westfall et al. 2015).

What Political Enemies Are for 277 Typically, our like-minded associates are similarly situated. Group members hence fuel one another’s escalation. As they adopt the shift toward extremity, they become more disposed to act in concert. Moreover, as the shift escalates confidence, they grow more inclined to engage together in risky behavior (Stoner 1968). All the while, their views of outsiders grow more intensely negative. Extremification is only part of the phenomenon. As groups embrace extremity, members begin to withhold information and hide preferences that deviate from perceived group expectations (Sunstein 2019, 84). Accordingly, groups also tend to become more alike in ways beyond their initial like-mindedness (Hogg 2001). They also become more invested in being alike. Our more extreme selves are also more conformist. This homogenization is unsurprising. Belief polarization intensifies adverse attitudes regarding outsiders, leading groups to fixate on the borders between insiders and outsiders. They adopt increasingly exacting standards for authentic membership. As qualifications for good-standing membership intensify, the group develops means for detecting poseurs. Standards for authenticity thus come to encompass behaviors and attitudes that extend beyond the group’s defining ideas. Compliance with these broader expectations becomes a way of expressing one’s membership in the group. Accordingly, belief polarized groups tend to be highly susceptible to the Black Sheep Effect – their negative attitudes toward perceived apostates are more intense than those they take toward outright foes (Marques et al. 1988). Further, as groups become more conformist, they also become more reliant on centralized standard-setters to establish the markers of authentic membership. Belief polarization hence renders groups more hierarchical and less internally democratic. Thus, as conformity pressures intensify, even slight deviations from the group’s expectations amplify into serious infractions. Belief polarized groups are prone to factionalize; they, therefore, tend to shrink into smaller cohorts of hardliners. Thus far, we have seen that belief polarization involves two shifts: doxastic extremification and group homogenization. That is the core of the phenomenon. Nevertheless, our picture remains incomplete. We need to look at how belief polarization works. Belief polarization is pervasive.8 It has been studied for more than six decades and found within groups of all kinds. It is robust across differences of nationality, race, gender, religion, economic status, and level of education.9 This prompts the question of how belief polarization works. Two intuitive views suggest themselves, an informational and a comparison account. The former holds that belief polarization results from the informational filters in like-minded groups. The latter contends that it occurs because group members strive to be seen by peers as authentic; they amplify their commitment to being slightly above the perceived mean. As the members are recalibrating simultaneously, the group itself

278  Robert B. Talisse shifts toward extremity. We cannot examine these accounts here.10 Suffice it to say that neither can be the whole story. Neither information exchange (Myers et al. 1980) nor ingroup comparison (Baron et al. 1996) is necessary to initiate belief polarization. A better account holds that belief polarization is driven by group-­ affiliated corroboration of one’s views. We shift toward extremity when we feel that a group we identify with embraces a belief or attitude that we espouse. We need not hear new evidence, nor need we compare ourselves to other group members. Instead, realizing that one’s view is popular among one’s identity group is sufficient. The corroboration view treats belief polarization as centrally a matter of group affinity. However, it recognizes that exchanging information and performing social comparisons are often ways of situating oneself within a group; hence, the corroboration view need not deny that the other accounts capture ways of initiating belief polarization. The corroborationist holds that the mechanism driving the phenomenon has to do with the positive effect that results from affirmation from one’s self-­ ascribed identity group. This explains why extremity shifts are more severe when group membership is primed (Abrams et al. 1990) and less pronounced among like-minded subjects who do not regard themselves as sharing a group identity (Le 2007). The view holds that corroboration from our peers makes us feel good about our beliefs, and this feeling of affirmation in our self-ascribed identity leads us to shift toward extremity (Hogg 2001). Thus, belief polarization is not only a phenomenon of extremification and homogenization but also a process by which group identity is made salient and centered. Notably, the relevant corroboration can come by way of highly indirect channels. For example, presenting a liberal subject with a chart showing that liberals widely oppose genetically modified food can prompt belief polarization; similarly, exposure to a poll showing that conservatives strongly favor a particular military action can produce an extremity shift in a conservative already favorably disposed to that action (Baron et al. 1996, 558). Moreover, the environmental prompts need not be verbal, overt, or literal; they can be merely implicit signals to group members that some belief is prevalent among them. Hats, pins, campaign signs, logos, gestures, songs, and the like are potential initiators of belief polarization for those who share that identity (Baron et al. 1996, 559). Note a connection between belief polarization and partisan sorting (Bishop 2009). As the United States has grown more diverse in various respects, citizens’ local spaces have become more politically homogeneous (Chen and Rodden 2013; Tam Cho et al. 2013). Schools, workplaces, occupations, congregations, and neighborhoods are sorted according to partisan allegiance (Iyengar and Westwood 2015; Mason 2015). Consequently, our daily routines tend to place us in social interaction

What Political Enemies Are for 279 only with those who share our political identity. Even when these interactions are not about politics, our partisan identities have grown so central to our overall sense of ourselves that our political commitments are corroborated. Sorted social spaces expose us to belief polarization, which encourages further sorting. With these details in place, we can see that the problem of polarization lies in the interaction of belief and political polarization. These two phenomena work together to exacerbate divisions, reward divisiveness, and erode our capacity for responsible democratic citizenship. To see how this works, recall that as we extremify, we also come to adopt increasingly hostile stances toward those perceived to be different. As we shift, their views look disfigured and irrational. Those who hold such views strike us as ignorant, naïve, and malign. We see them as a radicalized monolith, unified around the most extreme versions of the views that oppose our own and unworthy of civic respect. As a result, we are more thoroughly embedded within our camp, reinforcing tendencies to regard outsiders as untrustworthy, unpatriotic, unintelligent, threatening, and treacherous (Pew Research Center 2019). Ironically, as we regard our opponents in these ways, we come to fit the description we ascribe to them: we grow more intensely insular, distrustful, close-minded, and tribal. Once belief polarization has escalated, even amicable interactions with our political opponents tend to further our extremity (Bougher 2017; Bail et al. 2018); attempts to “reach across the aisle” backfire (Nyhan and Reifler 2010). Often, they are subject to the same forces. A tragically self-fulfilling projection results: each side shifts toward their opposition’s most unflattering caricature. Thus, the polarization dynamic. Belief polarization escalates widespread polarization, leading us to regard our political rivals as threats to neutralize. Civic respect begins to look Pollyannaish, if not positively complicit. Allies hive together. This tendency initiates further belief polarization, which produces more popular polarization. As was noted earlier, the deepening of widespread polarization activates party and elite polarization. Officeholders are released from electoral pressures to deliver tangible results – they can win reelection simply by stoking their base’s contempt for the opposing side (Puglisi and Snyder 2011; Sood and Iyengar 2016). This fuels further popular and belief polarization. And on it goes: we embrace exaggerated conceptions of our differences; we over-ascribe extreme ideas and vices to opponents; we attribute to them an implausible degree of internal unanimity; and we lose the ability to track the reasons they offer for their ideas. Eventually, we see them as divested from democratic citizenship and unworthy of civic respect. Politicians strategize accordingly, which fuels further belief polarization. However, that is not all. Ultimately the polarization dynamic constricts our sense of the range of permissible political disagreement among democratic citizens. It leads us to see political opponents as irredeemably

280  Robert B. Talisse beyond the pale. It fosters the profoundly anti-democratic stance that democracy is possible only among those who agree about politics. Now for the crucial upshot. The idea that democracy is possible only among those who agree about politics is not only anti-democratic; it is also unsustainable. To see why, recall that belief polarization involves both extremification and homogenization. As coalitions grow more uniform, they become increasingly resolute in enforcing conformity. Thus, belief polarization causes groups to become less tolerant of dissension within their ranks and more prone to expel deviating members. Consequently, belief polarization also erodes our ability to navigate disputes among our allies. The endeavor to sustain democratic relations only among our political allies is doomed: belief polarized coalitions splinter and shrink. When we think of toxic political divides, we fix on disputes among liberals and conservatives, Democrats and Republicans. We thereby overlook that there are significant political disputes within partisan coalitions. Consider the nascent rifts between religious and libertarian-leaning Republicans over abortion or between liberal and progressive Democrats over criminal justice. These, too, are disputes over justice; there is no reason to think that, once engaged, they would not be volatile. A progressive prison abolitionist may vote for the same candidates as a liberal criminal justice reformer, but each must see the other as advocating for injustice. Return to the democrat’s dilemma. We were envisioning citizens who take their democratic responsibilities seriously. They want to use their share of political power to advance justice, and they recognize the general duty to show their fellow citizens civic respect. However, there are political circumstances where expressing civic respect for their opponents counteracts their advocacy for justice. They wonder why they should bother engaging with their foes at all. Why not simply disregard them and work with allies to secure justice? The polarization dynamic supplies a compelling reply. When we dismiss our enemies and break off civic interactions with them, the phenomenon of belief polarization does not simply dissolve. Instead, it turns inward, toxifying our relations with our political allies by escalating conformity pressures and dismantling our capacity to navigate disagreements with them. This shift, in turn, causes our coalitions to splinter and shrink, rendering them less politically efficacious. Hence, the response to the citizen caught in the democrat’s dilemma is that when we break off democratic relations with our enemies, we endanger our capacity to sustain democratic ties with our allies. Yet we must cultivate and maintain solid political coalitions to pursue justice effectively. Thus, we need to support democratic relations with our political enemies if we are to seek justice. The argument may seem objectionably instrumental. It claims that unless we uphold civic respect for our enemies, we will lose the capacity

What Political Enemies Are for 281 to maintain the kinds of political relationships necessary for pursuing justice. This suggests that expressing civic respect for our enemies is crudely strategic. That is hardly respect at all. Although I concede that the argument is instrumental, I deny it is objectionably so. First, upholding democratic relations with political enemies is itself moral: we have a moral reason to maintain the social conditions under which our efforts to secure justice could succeed. Second, recall that the argument is addressed to citizens who already are disposed to regard their political opponents as democratically divested or otherwise undeserving of civic respect. I endorse a relatively straightforward moral view according to which citizens as such are owed civic respect simply by virtue of the fact that political decisions almost always involve the exercise of coercive power among equals. The audience I aim to address at present is likely to find that kind of argument unmoving. Accordingly, the argument I have proposed casts engagements with political enemies as occasions for building and maintaining stronger alliances. Note also that the argument calls for expressing civic respect for one’s foes; it does not say that one must simply proceed as if one respected them. Given the damage to democracy that the polarization dynamic has already caused, the crudely strategic reason for upholding civic respect might be the best we could do. I would say, then, that the argument has a proleptic flavor: it gives citizens a moral reason to proceed as if their political enemies were deserving of respect; the hope is that by proceeding that way, citizens will eventually develop authentic civic respect for their enemies.

14.4  Managing Belief Polarization Thus far, I have argued that the polarization dynamic provides a compelling moral case for upholding democratic relations with political enemies, even when we regard them as on the side of injustice. In calling the argument “compelling,” I am assessing its philosophical quality, not its practical force. Upholding democratic relations with our political enemies is difficult, even under the best political circumstances. Moreover, present political circumstances are far from ideal. Specifically, the argument above may be compelling for citizens confronting the democrat’s dilemma, wondering why they should not suspend democratic relations with their enemies. However, many citizens in contemporary democracies are well past this point. They are already in the grip of the polarization dynamic; they have already written off their enemies as undeserving of civic respect. It will not do, then, simply to recommend that citizens attempt to engage with foes as if the polarization dynamic were not already in effect. As noted above, some data show that this strategy backfires (Bail et al. 2018). After all, there is a difference between what one must do to prevent an outcome and what one must do to reverse it. What can be done if the way forward does not lie in

282  Robert B. Talisse introducing more peace, love, and understanding into political interactions with our foes? The first thing to observe is that belief polarization is a byproduct of active democratic citizenship; the polarization dynamic emerges from the normal functioning of democracy. It thus cannot be fixed or eliminated but only managed. Secondly, we must acknowledge our vulnerability to belief polarization. We need to recognize that, to some degree, our political thinking and relations with our fellow citizens are shaped by those dynamics. This method is not a “both sides” maneuver but rather a reasonable inference from well-established results. When we think of polarization and extremity, we quickly see it in our opponents. However, the idea that we are not vulnerable to the same forces is naïve. So, recognizing that the problem lies within us is the first step. Nevertheless, what must we do? Recall that as we shift toward extremity, our views look more obviously correct; thus, our conception of the scope of reasonable disagreement shrinks. Accordingly, one way we operationalize the realization that our political thinking is very likely impacted by polarization is to create occasions for reminding ourselves that our political thinking is not beyond reasonable criticism. To be clear, this does not require us to adopt the stance of the Millian fallibilist, who holds that beliefs are never quite the entire truth and are perpetually in danger of becoming dead dogma. Instead, it calls for the more modest stance that we are epistemically improvable. This is consistent with the thinking that our political beliefs are correct as they are. It involves only the recognition that our articulation, appreciation, and grasp of them could be deepened, sharpened, and refined. Acknowledging the possibility of reasonable criticism does not require divestment from our commitments but only recognizing that we could do better by them by exposing ourselves to occasions where we could improve our command of them. Managing belief polarization starts with self-criticism among allies. Under more modest belief polarization conditions, we could introduce robust “Devil’s Advocate” norms into our alliances. These norms encourage allies to subject our shared commitments to reasonable criticism. However, we cannot again attempt to manage belief polarization by proceeding as we would if it had not already heavily taken effect. The worry is that in especially active political coalitions, conformity pressures have escalated to such a degree that Devil’s Advocacy leads only to the ostracism of the internal critic. The task is to introduce mechanisms of self-criticism without escalating conformity pressures among our allies. Given existing levels of belief polarization, the prospects seem bleak. The task of managing belief polarization thus begins neither with our cross-partisan relations nor our alliances. What then? My suggestion is counterintuitive: We need solitude if we seek to manage belief polarization. Recall earlier points about partisan sorting and the corroboration view of

What Political Enemies Are for 283 belief polarization. We are surrounded by partisan prompts and overt and implicit calls to our allegiance. Partisan rifts are omnipresent, as are sites for signaling allyship. In addition, you may be familiar with arguments claiming that social media echo chambers distort our politics. On the corroboration view of belief polarization, partisan sorting turns our physical environments into echo chambers; they encourage the same distortions. The idea, then, is that we must occasionally withdraw if neither our allies nor our opponents can function as reasonable critics of our political ideas. In another work, I argued that citizens must engage in cooperative activities where political affiliation is simply beside the point (Talisse 2019). The idea here is different. Managing belief polarization calls for us to occasionally engage in politics alone, away from the pressures of allies and foes. It is not that we must withdraw from politics; instead, we need to engage in a kind of solitary political reflection. Typically, we figure out where we stand politically by looking to our allies. Polarization management requires that we sometimes engage in political reflection in a detached context from the polarization dynamic. I am calling not to withdraw from politics but for occasional civic solitude. The problem is that our social worlds are already colonized by partisanship. Consequently, civic solitude calls for distance. Citizens need space to grapple with political ideas that are not prepackaged in the political fissures of the moment. Accordingly, managing belief polarization involves expanding our political imagination by reflecting in ways that remind us that the spectrum of opinion available to democratic citizens is broader and more profound than what figures into the day’s politics. By encountering a more expansive palate of political ideas, democratic citizens can more accurately position their commitments and rivalries; they more reliably distinguish between the unavoidable disputes among citizens and unbridgeable gaps between democratic ideals and those of anti-democracy. The details of civic solitude cannot be spelled out here. If we are submerged in the polarization dynamic, we must extricate ourselves from the morass. This will involve exposing oneself to political ideas alien to the prevailing categories and debates, ideas that are not immediately translatable into the idiom of the moment’s politics. Perhaps this is achieved by reading ancient political theory, immersing oneself in the politics of a foreign democracy, or studying the political art of 19th century America. It does not matter much, so long as the process is undertaken in a way that permits the material to remain detached from the day’s politics. We are accustomed to thinking that democracy is about action. A Google search of the phrase “this is what democracy looks like” instantly returns thousands of images of masses gathered in public space to express a shared political message. To be clear, this is what democracy looks like. Citizens must be active participants in self-government. However, citizens must also be reflective. We have seen that prominent modes of democratic action activate forces that unravel our capacities for political

284  Robert B. Talisse reflection. Without those capacities, our political action might feel even more urgent and necessary, but it is also more likely to be directed by distortions that ultimately contribute to democratic dysfunctions.

Notes





1. For further details, see Talisse (2021, Ch 1). 2. In the United States, the terms “RINO” (“Republican in name only”) and “neoliberal” (a professed liberal who nonetheless serves corporate interests) serve this purpose. 3. This phenomenon is frequently called affective polarization, but this term is not ideal in the present context for reasons that will become clear below. See Iyengar et al. (2019) for a review. 4. It may come as no surprise, then, that in the United States, widespread disapproval of inter-partisan marriage is now more pronounced than disapproval of inter-faith and inter-racial marriage (Iyengar and Westwood 2015, 691). This animosity is good because co-partisanship is the most reliable predictor of long-term relationship success among those paired on online dating platforms (Huber and Malhotra 2017; Iyengar and Konitzer 2017). 5. See the “perception gap” data presented in Beyond Conflict (2020) and More in Common: https://perceptiongap.us/. 6. Alternatively, at least the public expressions associated with these forms of polarization perform that function. The parties and party members in the United States are not as divided over political policies. 7. The phenomenon is generally called “group polarization.” In the present context, this more common name is misleading. I am distinguishing political and belief polarization, both of which have to do with groups. Also, I should note that I am using the word “doxastic” broadly. 8. Hence, Lamm and Myers (1978, 146), “Seldom in the history of social psychology has a nonobvious phenomenon been so firmly grounded in data from various cultures and dependent measures.” 9. The appendix in Sunstein (2009) summarizes the most important findings. 10. For more detail, see Talisse (2021, Ch 3).

References Abrams, D., M. Wetherell, S. Cochrane, M. A. Hogg, and J. C. Turner. 1990. “Knowing What to Think by Knowing Who You Are: Self-categorization and the Nature of Norm Formation, Conformity, and Group Polarization.” British Journal of Social Psychology 29.2: 97–119. Ahler, Douglas J. 2014. “Self-Fulfilling Misperceptions of Public Polarization.” Journal of Politics 76: 607–620. Ahler, Douglas J., and Gaurav Sood. 2018. “The Parties in Our Heads: Misperception about Party Composition and Their Consequences.” Journal of Politics 80: 964–981. Bail, Christopher A., Lisa P. Argyle, Taylor W. Brown, John P. Bumpus, Haohan Chen, M. B. Fallin Hunzaker, Jaemin Lee, Marcus Mann, Friedolin Merhout, and Alexander Volfovsky. 2018. “Exposure to Opposing Views on Social Media Can Increase Political Polarization.” Proceedings of the National Academy of Sciences 115.37: 9216–9221.

What Political Enemies Are for 285 Baron, Robert S., Sieg I. Hoppe, Chaun Feng Kao, Bethany Brunsman, Barbara Linneweh, and Diana Rogers. 1996. “Social Corroboration and Opinion Extremity.” Journal of Experimental Social Psychology 32: 537–560. Beyond Conflict. 2020. America’s Divided Mind. https://beyondconflictint.org/ americas-divided-mind/. Accessed February 15, 2022. Bishop, Bill. 2009. The Big Short. New York: Mariner Books. Bougher, Lori D. 2017. “The Correlates of Discord: Identity, Issue Alignment, and Political Hostility in Polarized America.” Political Behavior 39: 731–762. Chen, Jowei, and Jonathan Rodden. 2013. “Unintentional Gerrymandering: Political Geography and Electoral Bias in Legislatures.” Quarterly Journal of Political Science 8: 239–269. Hetherington, Marc, and Jonathan Weiler. 2018. Prius or Pickup? Boston: Houghton Mifflin Harcourt. Hogg, Michael A. 2001. “A Social Identity Theory of Leadership.” Personality and Social Psychology Review 5: 184–200. Huber, Gregory A., and Neil Malhotra. 2017. “Political Homophily in Social Relationships: Evidence from Online Dating Behavior.” The Journal of Politics 79: 269–283. Iyengar, Shanto, and Tobias Konitzer. 2017. “The Moderating Effects of Marriage across Party Lines.” Working paper. https://pdfs.semanticscholar. org/a55b/50f3de44529ee301c662aa42fb244e4ab992.pdf. Accessed May 10, 2022. Iyengar, Shanto, and Masha Krupenkin. 2018. “The Strengthening of Partisan Affect.” Advances in Political Psychology 39: 201–218. Iyengar, Shanto, Yphtach Lelkes, Matthew Levendusky, Neil Malhotra, and Sean J. Westwood. 2019. “The Origins and Consequences of Affective Polarization in the United States.” Annual Review of Political Science 22: 129–146. Iyengar, Shanto, Gaurav Sood, and Yphach Lelkes. 2012. “Affect, Not Ideology: A Social Identity Perspective on Polarization.” Public Opinion Quarterly 76: 405–431. Iyengar, Shanto, and Sean J. Westwood. 2015. “Fear and Loathing across Party Lines: New Evidence on Group Polarization.” American Journal of Political Science 59: 690–707. Klein, Ezra. 2020. Why We Are Polarized. New York: Avid Reader Press. Lamm, Helmut, and David Myers. 1978. “Group-Induced Polarization of Attitudes and Behavior.” Advances in Experimental Social Psychology 11: 145–187. Le, Eun-Ju. 2007. “Deindividuation Effects on Group Polarization in ComputerMediated Communication: The Role of Group Identification, Public-SelfAwareness, and Perceived Argument Quality.” Journal of Communication 57.2: 385–403. Lee, Frances E. 2016. Insecure Majorities: Congress and the Perpetual Campaign. Chicago: University of Chicago Press. Levendusky, Matthew, and Neil Malhorta. 2016. “Does Media Coverage of Partisan Polarization Affect Politics Attitudes?” Political Communication 33: 283–301. Marques, José M., Vincent Y. Yzerbyt, and Jacques-Philippe Leyens. 1988. “The ‘Black Sheep Effect’: Extremity of Judgments towards Ingroup Members as a Function of Group Identification.” European Journal of Social Psychology 18: 1–16.

286  Robert B. Talisse Mason, Lilliana. 2015. “‘I Disrespectfully Agree’: The Differential Effects of Partisan Sorting on Social and Issue Polarization.” American Journal of Political Science 59: 128–145. Mason, Lilliana. 2018a. Uncivil Agreement. Chicago: University of Chicago Press. Mason, Lilliana. 2018b. “Ideologues without Issues: The Polarizing Consequences of Ideological Identities.” Public Opinion Quarterly 82: 280–301. McConnell, Christopher, Yotam Margalit, Neil Halhorta, and Matthew Levendusky. 2017. “The Economic Consequences of Partisanship in a Polarized Era.” American Journal of Political Science 62: 5–18. Myers, D. G., J. B. Bruggink. R. C. Kersting, and B. A. Schlosser. 1980. “Does Learning Others’ Opinions Change One's Opinion?” Personality and Social Psychology Bulletin 6: 253–260. Nyhan, Brendan, and Jason Reifler. 2010. “When Corrections Fail: The Persistence of Political Misperceptions.” Political Behavior 32: 303–330. Pew Research Center. 2019. “Partisan Antipathy: More Intense, More Personal.” https://www.pewresearch.org/politics/2019/10/10/partisan-antipathy-moreintense-more-personal/. Accessed May 19, 2022. Puglisi, Riccardo, and James M. Snyder. 2011. “Newspaper Coverage of Scandals.” Journal of Politics 73: 931–950. Sood, Guarav, and Shanto Iyengar. 2016. “Coming to Dislike Your Opponents: The Polarizing Impact of Political Campaigns.” Working paper. https://www. gsood.com/research/papers/ComingToDislike.pdf. Accessed May 17, 2022 Stoner, J. A. 1968. “Risky and Cautious Shifts in Group Decisions: The Influence of Widely Held Values.” Journal of Experimental Social Psychology 4.4: 442–459. Sunstein, Cass R. 2009. Going to Extremes: How Like Minds Unite and Divide. New York: Oxford University Press. Sunstein, Cass R. 2019. Conformity. New York: NYU Press. Talisse, Robert B. 2019. Overdoing Democracy. New York: Oxford University Press. Talisse, Robert B. 2021. Sustaining Democracy. New York: Oxford University Press. Tam Cho, Wendy K., James G. Gimpel, and Iris S. Hui. 2013. “Voter Migration and the Geographic Sorting of the American Electorate.” Annals of the Association of American Geographers 103: 856–870. Westfall, Jacob, Leaf Van Boven, John R. Chambers, and Charles M. Judd. 2015. “Perceiving Political Polarization in the United States: Party Identity Strength and Attitude Extremity Exacerbate the Perceived Partisan Divide.” Perspectives on Psychological Science 10: 145–58.

15 Top-Down and Bottom-Up Solutions to the Problem of Political Ignorance Ilya Somin

15.1 Introduction There is broad, though not universal, agreement that widespread voter ignorance and irrational evaluations of evidence are significant threats to democracy. The rise of authoritarian-leaning right-wing populist movements in the United States and many European countries has accentuated the significance of this menace, as has the role of public ignorance during the Covid-19 pandemic and in fostering the common yet incorrect belief that the 2020 election was “stolen” from Donald Trump. The problem is not a new one, however. Nor is it confined to one country, the supporters of one or a few specific politicians, or one side of the political spectrum. While there is increasing awareness of the danger, there is deep disagreement over strategies for mitigating it. “Top-down” approaches, such as epistocracy and lodging more authority in the hands of experts, seek to alleviate ignorance by consolidating more political power in the hands of the more knowledgeable segments of the population. Another version of this approach would give expert officials more power to control the flow of information to the public, thereby limiting the influence of “misinformation,” particularly that spread on the internet and social media. By contrast, “bottom-up” approaches seek to either raise the political competence of the public or empower ordinary people in ways that give them better incentives to make reasonable decisions than ballot-box voting does. Examples of bottom-up strategies include increasing voter knowledge through education, various sortition proposals, paying voters to increase their understanding of politics, and shifting decisions to institutional frameworks where citizens can “vote with their feet.” This chapter surveys and critiques a range of both top-down and bottom-up approaches. In the process, I build on my previous work assessing feasible strategies for addressing the problem of political ignorance. I tentatively conclude that top-down strategies have severe flaws and are likely to make things worse rather than better. While we should not categorically reject them, there is a solid basis for skepticism about DOI: 10.4324/9781003311003-19

288  Ilya Somin their desirability. Bottom-up strategies have significant limitations of their own. But they have greater upside potential. Expanding foot voting opportunities is more promising than any other currently available option. Paying voters to increase their knowledge also deserves more significant consideration than it has received. Section 15.2 briefly surveys the problem of political ignorance, which includes low political knowledge levels and highly biased assessments of the information voters do learn. Both issues primarily stem from ultimately rational individual voter behavior that can result in detrimental collective outcomes. Section 15.3 outlines and critiques several top-down approaches to addressing political ignorance, including epistocracy, the delegation of power to experts insulated from political pressure, and government control of information flows. These strategies each have unique defects. But they also have systematic ones that are likely inherent in any top-down approach. Those include flawed incentives of decision-makers, limitations on their knowledge, and how public ignorance inhibits effective monitoring of the very officials empowered to make up for it. We should not altogether reject top-down strategies. But their severe limitations are cause for wariness. Section 15.4 considers bottom-up strategies, including using education to increase political knowledge, sortition mechanisms, paying voters to increase their knowledge, and shifting decisions to frameworks where people can “vote with their feet” and thereby have more substantial incentives to seek relevant information than ballot-box voters do. None of these strategies are perfect, and some have flaws similar to those that bedevil top-down strategies. But, overall, they are more promising, especially in the case of expanding opportunities for foot voting. Paying voters to increase their knowledge is also a potentially promising approach that deserves more significant consideration.

15.2  The Problem of Political Ignorance Political ignorance is a two-level problem.1 Most of the voting public is rationally ignorant – knowing little or nothing about the issues, parties, and candidates at stake in elections. In addition, most voters have little incentive to evaluate the information they learn objectively. They instead assess those facts in a highly biased way, overvaluing those that reinforce their preexisting views and undervaluing or even ignoring the rest. Voters have strong incentives to be “rationally ignorant” because there is so little chance that their votes will make a difference to the outcome. In an American presidential election, the odds that a given vote will make a decisive difference to the result is approximately 1 in 60 million, or even less.2 In a situation where there is little or no benefit to acquiring additional knowledge, it is often perfectly rational for individuals to remain largely ignorant about the issues at hand.3 Survey data show that

Top-Down and Bottom-Up Solutions to Political Ignorance 289 voters often lack even rudimentary knowledge about the candidates and policies at issue in any given election.4 They also have little incentive to analyze whatever information they learn logically and unbiasedly. On the contrary, voters have incentives to fall prey to what economist Bryan Caplan calls “rational irrationality”: when there are few or no negative consequences to error, it is rational to make little or no effort to control one’s biases.5 Thus, citizens routinely overvalue evidence supporting their preexisting views while downplaying or ignoring anything that cuts the other way.6 These tendencies toward biased evaluations of information are significant and widespread among voters on both sides of the political spectrum.7 Many of the most attentive citizens tend to be highly biased “political fans.”8 They follow politics closely for much the same reasons as sports fans follow their favorite teams: not to get at the truth, but to enjoy the camaraderie of their fellow fans, the process of cheering on their preferred political “team,” and – in many cases – detesting opposing “teams” (opposing parties and their supporters). There is nothing necessarily wrong with being a political or sports fan. But fan behavior is often at odds with truth-seeking. People who follow politics primarily to enhance their fan experience cannot objectively evaluate political information. Rational ignorance and rational irrationality affect the decisions of altruistic voters and those who are narrowly self-interested. Even a citizen strongly motivated to help others has little incentive to devote more than a small amount of effort to acquiring political knowledge and reining their biases. Whether voters’ purposes are self-interested or not, the odds that seeking out new knowledge and curbing biases will pay off are extremely low. This low likelihood makes it rational for both egoists and altruists to severely limit the time and effort devoted to acquiring and analyzing political information and to make little effort to curb their biases.9 Rational ignorance does not necessarily require careful, calculated decision-making. In many cases, it involves merely the application of crude rules of thumb or an intuitive sense that there is little benefit to seeking out additional knowledge. Thus, the prediction of rational behavior in this sphere is not dependent on the assumption that voters are hyper-logical or capable of making complex calculations about odds. Indeed, such a detailed calculation may often actually be irrational, requiring more time and effort than can be justified, given the likely benefit.10 Decades of survey data indicate that voter knowledge levels are low and have undergone little or no increase, despite rising educational attainment and the development of the internet and other modern technology that makes information easier to access.11 The majority of the public often does not know basic information, such as which party controls Congress, what significant policies have been enacted in recent years, or which elected officials are responsible for which issues.12

290  Ilya Somin Just before the 2014 US federal election, in which the primary stake was control of Congress, only 38% of voters knew which party controlled the House of Representatives, and a similar percentage knew which controlled the Senate.13 Another 2014 survey found that only 36% of Americans can even name the three branches of the federal government: the executive, the legislative, and the judicial.14 Despite growing fiscal problems, most of the public also has little understanding of how the federal government spends its money – vastly underestimating the percentage of the budget that goes to major entitlement programs (among the largest categories of federal spending) while massively overestimating that allocated to foreign aid (which is only about 1% of the budget).15 Survey data indicate that there is similar ignorance in other democracies. It is not a problem unique to the United States. On the contrary, it exists in a wide range of democratic political systems in Europe and elsewhere.16 Some scholars have argued that voter ignorance is not as profound a problem as it seems because voters can use “information shortcuts” to make up for their knowledge deficits.17 Others contend that, even if individual voters make poor decisions, the effects of their errors cancel each other out through a “miracle of aggregation,” resulting in the electorate making well-informed decisions.18 I have criticized shortcut theories and “miracle of aggregation” arguments in some detail in previous work.19 Here, I assume, for present purposes, that these mechanisms do not come close to fully solving the problems of rational ignorance and irrationality. Thus, we must try other methods. It is, however, vital to emphasize that the enormous size, scope, and complexity of modern government both exacerbate the initial problem of political ignorance and weakens the effectiveness of any efforts to “solve” it.20 In most modern democracies, government spending accounts for one-third or more of GDP, and the state regulates almost every significant human activity, at least to some substantial degree.21 This reality ensures that even a substantial increase in voter knowledge might not be enough to enable the electorate to monitor more than a small fraction of government activity effectively. It also creates many opportunities for elites and interest groups to exploit the system.

15.3  Top-Down Approaches The most obvious potential solution to the problem of political ignorance is to find some way to concentrate power in the hands of the (at least relatively) more knowledgeable. I call such strategies “top-down” approaches to the problem. The most straightforward top-down approach is limiting the franchise to a more knowledgeable population segment or giving the latter

Top-Down and Bottom-Up Solutions to Political Ignorance 291 extra votes. Instead of democracy, we could have “epistocracy” – the “rule of the knowers” or at least move in an epistocratic direction.22 A more far-reaching extension of that approach would be to concentrate more authority in the hands of experts, at least relatively insulated from the democratic political process. Such officials could be technocrats, ­scientists, or others with (supposedly) superior insight into public policy issues. Finally, it may also be possible to give experts greater control over information flow – especially misinformation – to the general public. Under this approach, the voters would still have ultimate control over the selection of public officials and thereby still wield decisive influence over policy. Nevertheless, that influence would be guided and improved by curbing misinformation. I consider each of these three ideas, in turn. Despite their differences, they have some common limitations. The most significant is that all three create dangerous opportunities for the government to exploit the mechanism in question to promote its own partisan and ideological interests while suppressing opposition. Second, all three are likely to be less effective at coping with rational irrationality and bias in evaluating information than with simple ignorance. 15.3.1 Epistocracy23 The basic idea of epistocracy is far from new. As far back as the origins of democracy in ancient Greece, the franchise was limited to adult male citizens of the “hoplite” (military) class because it was thought they had superior knowledge and insight.24 In the nineteenth century, John Stuart Mill advocated giving more educated citizens extra votes for the same reasons.25 Modern democracies already pursue epistocratic strategies for limiting the franchise in some respects. In the United States, we exclude over 20% of our population from the franchise because we think they are ignorant and have poor judgment: children.26 Strikingly, very few of us feel much guilt for this massive exclusion of a large part of our population from political power. Even the idea of letting some children vote if they can prove they are more knowledgeable than the average adult is considered radical and dangerous.27 Similarly, the United States does not allow legal immigrants to become citizens and obtain the right to vote unless they can pass a civics test that most native-born Americans would likely fail if they had to take without studying.28 Many US states also exclude convicted felons and the mentally ill from the franchise out of concern over their competence and judgment (or lack thereof). More recently, political philosopher Jason Brennan has revised and extended the case for epistocracy in his influential 2016 book, Against

292  Ilya Somin Democracy.29 He proposes various strategies for limiting the franchise to more knowledgeable segments of the public while still ensuring that the electorate is large and representative.30 For example, we could limit voting rights to those who pass a political knowledge test, similar to that already imposed on immigrants who want to become naturalized citizens. While Brennan’s ideas may seem radical and off-putting, they can also be described as just modest extensions of the status quo. The exclusion of children, immigrants who cannot pass a citizenship test, felons, and the mentally ill already covers 25% or more of the population. If that is acceptable, why not eliminate the least knowledgeable 10% of the current electorate or the least knowledgeable 20%? Many of them might be no more competent voters than the groups we already exclude because of their supposed ignorance and poor judgment. If an adult native-born American is as bad a voter as the average child – or the average immigrant who cannot pass the citizenship test – it is hard to argue that they deserve to have the vote any more than members of these other groups do. Nonetheless, it is unlikely that epistocracy can overcome the problem of political ignorance or even be a significant tool in mitigating it. The idea has three serious shortcomings. First, it is difficult to believe that real-world governments can be trusted to objectively identify the more knowledgeable segments of the electorate and exclude the rest. For obvious reasons, they will be tempted to bias the system in favor of the supporters of their party or ideology.31 On top of that, there is a long history of using exclusions from the franchise to bar voters based on race, ethnicity, gender, and other similar invidious bases. Even if the government can purge itself of such long-standing prejudices or prevent them from influencing the rules for determining who qualifies as a sufficiently knowledgeable voter, the incentive for partisan and ideological bias will remain. That problem is a structural constraint inherent to government, not merely a consequence of biases specific to a particular time and place. We could perhaps overcome the issue by delegating the task of ­developing voter qualification criteria to a committee of non-partisan experts on political knowledge. But that merely kicks the process back to the problem of figuring out how to select those experts. A group with the power to determine qualifications for the electorate – and therefore the indirect ability to control the composition of future governments – is an obvious target for “capture” by parties, interest groups, and others. This latter problem also weakens the case for “enlightened preference” voting, a form of epistocracy under which everyone still gets to participate equally. The government then weights their votes based on calculations about what their preferences would have been if they had higher levels of political knowledge.32 As with more conventional approaches to epistocracy, the government would still have to have a test for determining what qualifies as political knowledge. In addition, it would also have to select

Top-Down and Bottom-Up Solutions to Political Ignorance 293 a formula for converting the “raw” preferences of low-­information voters to their “enlightened” version. Both steps in this process are rife with opportunities for capture and bias. Brennan is aware of these difficulties and has proposed ways of mitigating them, such as leaving the selection of qualifications to the public rather than a government agency.33 But I am skeptical that the public can carry out this function effectively or that any other currently known institutional arrangement can do it.34 A second limitation of epistocracy is that it would, at best, solve only one part of the political ignorance problem: low voter knowledge levels. If it works as intended, the system will ensure that the electorate is vastly more knowledgeable than what we presently have. That would be a genuine improvement over the status quo. Nevertheless, it would not address the problem of biased evaluation of information. An electorate that is well-informed but still in the throes of “rational irrationality” might make terrible decisions. Depending on how the epistocratic electorate is chosen, the problem of bias might even be worse than before because people who follow politics closely – and thereby have higher levels of factual political knowledge – are disproportionately likely to be biased “political fans.”35 Perhaps epistocratic selection can choose voters based not only on their superior knowledge but also on their relative lack of bias in evaluating what they know. But that would make epistocracy even more challenging to implement and create even more opportunities for partisan and other manipulation of the selection process. Finally, it is not clear that even the best possible epistocratic selection process will lead to an electorate capable of effectively monitoring modern government’s full range of functions. The knowledge burden required to do so is significant enough that even an electorate far more knowledgeable and far less biased than what we currently have would still not be sufficient. It is not my view that epistocracy should be wholly rejected. There are epistocratic elements in the status quo, and there is a plausible case for extending those elements at the margin. Regional or local governments should perhaps undertake experiments in more thoroughgoing epistocracy, especially in areas where partisan, racial, and ethnic conflicts are not too great, thereby reducing possible sources of bias. Nevertheless, barring some breakthrough in institutional design, it is unlikely that epistocracy can do more than marginally mitigate the problem of political ignorance. 15.3.2  The Rule of Experts If epistocracy is impractical or insufficient, why not go further in the same direction and transfer control over more policy areas to small groups

294  Ilya Somin of experts insulated from political pressure? Unlike epistocratic voters, there are well-established criteria for identifying experts in specific fields. For example, there are established and more or less widely recognized credentials for doctors, scientists, economists, defense policy specialists, and the like. As with epistocracy, status quo policy already incorporates the rule of experts in some respects. For example, most democracies delegate monetary policy to an independent central bank, such as the US Federal Reserve Board.36 During the coronavirus pandemic, many policy decisions were at least in part delegated to specialized public health agencies, such as the US Centers for Disease Control (CDC), in hopes that these experts could curb the spread of the virus by “following the science.” Experts, of course, also often wield discretionary authority in other areas of public policy, ranging from food and drug regulation to environmental protection. Some scholars and public policy analysts argue for greatly expanding the delegation of power to experts. Notable examples include Cass Sunstein and former US Supreme Court Justice Stephen Breyer.37 Why not expand the Federal Reserve model to other areas of policy? Despite its potential appeal, the rule of experts is a seriously flawed antidote to political ignorance. The most obvious danger is that experts will serve their interests or those of well-organized interest groups rather than the general public. Once insulated from the democratic process, experts will have little incentive to help the people from whose control they have been deliberately protected. There is, of course, a long history of small elite groups ruling in their interests. On the other hand, if expert bodies remain subject to external pressure, they can readily become politicized. During the Covid-19 pandemic, the CDC adopted measures such as a nationwide eviction moratorium and Title 42 “public health” expulsions of migrants that had little scientific basis but were advocated by powerful constituencies within the Trump and later the Biden administration.38 As a result, immense harm was inflicted on innocent people – most notably some 1.7 million vulnerable migrants cruelly denied opportunities to seek asylum – with little or no public health gain.39 Experts also suffer from knowledge and bias problems reminiscent of ordinary voters. As economist F. A. Hayek famously argued, it might be impossible for them to determine the actual preferences of the people they wish to serve because they cannot know how much any given individual values a particular activity.40 A health expert probably knows more than I do about the risks of drinking or smoking. Nevertheless, only I know how much enjoyment I derive from having a beer or smoking a cigarette. For that reason, the expert cannot readily tell whether trying to limit a given person’s alcohol intake will improve their welfare or not.41

Top-Down and Bottom-Up Solutions to Political Ignorance 295 This was a ubiquitous problem during the Covid-19 pandemic, when public health experts ordered sweeping “lockdowns” to contain the virus while trying to exempt “essential” activities, without considering that many people assigned widely divergent values to the activities in question. For much of the population, the lost opportunities might well have outweighed the marginal benefit of reducing potential exposure to the virus.42 The experts especially failed to consider the enormous costs of pandemic migration restrictions, including the stifling of future innovation that could improve health outcomes in the long run.43 In addition to limitations on their knowledge, experts also suffer from ideological and partisan bias in evaluating the information they know. Like ordinary voters, political elites are highly biased in evaluating political information; even in Denmark, a nation noted for having a relatively responsible and non-polarized elite, political leaders show severe motivated reasoning similar to that of ordinary “political fans.”44 Experts tasked with designing paternalistic policies also demonstrate a variety of biases similar to those of ordinary citizens.45 While they have a greater incentive to learn relevant information than ordinary voters do – because of the greater likelihood of decisiveness – they still have little incentive to carefully and objectively weigh costs and benefits instead of validating their preexisting biases and views. After all, most of the costs of their decisions are borne by the general public rather than by the experts themselves. These considerations do not prove that delegation of power to politically insulated experts is never justified. In some situations, the available alternatives may be even worse. For example, letting an insulated central bank set monetary policy may be a lesser evil than giving incumbent politicians control over the money supply, enabling them to debase the currency when doing so might help their reelection prospects. Delegation to experts may be helpful in cases where a clearly defined task requires specialized expertise and there is broad consensus about the associated costs and benefits that cut across ideological lines. In such situations, perverse incentives are reduced, and there is less possibility of ideological or partisan bias. Most major public policy issues, however, are not like that. Moreover, the more matters are delegated to experts, the greater the difficulties of monitoring the latter and avoiding biased and self-serving policies. 15.3.3  Controlling the Flow of Misinformation In recent years, some experts have argued that the ready availability of misinformation, facilitated by the internet and modern social media, has led many to believe “fake news,” conspiracy theories, and other dubious material on political issues.46 If so, the solution is to restrict the flow of misinformation, thereby preventing voters’ judgment from being influenced.

296  Ilya Somin Some advocates propose relatively modest measures, limited to curtailing extreme disinformation about election arrangements.47 Others would go much further. For example, French President Emmanuel Macron has proposed banning websites that spread “fake news” during elections.48 If voters lack the knowledge and discernment to sift through and reject misinformation on their own, perhaps we can improve their judgment by restricting access to dubious and dangerous information sources. Unlike epistocracy and the rule of experts, this strategy does not seek to displace the electorate from its role as the ultimate decision-maker in a democracy. But it would impose elite control over the information available to the voters. The main objections to this approach are too well-known to require extensive elaboration. The most obvious is that incumbent governments cannot be trusted to even-handedly target “misinformation” instead of selectively suppressing information sources that undercut the official line and support their opponents. This danger is especially significant in a deeply polarized political system like the United States and some European nations. Democrats would have good reason to distrust the objectivity of a Republican administration in determining what qualifies as “misinformation” or “fake news.” Republicans would, likewise, have good reason to distrust a Democratic one. Plans to suppress misinformation also suffer from the mistaken assumption that the problem is mainly a product of the new information environment created by the internet and social media, which facilitate “cheap speech” by unscrupulous manipulators, unconstrained by traditional media gatekeepers.49 Thus, it is argued that all we need to do to curtail misinformation seriously is a crackdown on its spread through these new pathways. In reality, it is far from clear that misinformation from new forms of media is significantly worse than that from the old or that political ignorance and irrationality are worse now than in the past.50 Political knowledge levels have been stably low for decades.51 While today’s political environment features plenty of irrational, widely believed misinformation, it is hard to show that it is worse than in the past. The ignorance and bias that contributed to the rise of fascism and communism in the early twentieth century were at least as bad as anything that exists today, and the distortions on which these ideologies were based were spread through what we would today consider “legacy” media: primarily newspapers and radio. Even today, new media are probably less significant sources of misinformation than traditional ones, such as broadcast media. More people follow news on broadcast media and conventional news websites than on social media.52 On the right, networks such as Fox News are far more significant purveyors of misinformation than anything on social media. Left-wing misinformation also has many alternative channels.

Top-Down and Bottom-Up Solutions to Political Ignorance 297 Ultimately, the main problem is not the supply of misinformation on specific media but the demand for it by political fans who crave validation for their preexisting biases rather than search for the truth.53 If that demand remains high, it will be met by purveyors of misinformation and exploited by unscrupulous politicians and activists. It is no surprise that right-wing misinformation – such as claims that Democrats stole the 2020 election from Donald Trump – appeals primarily to committed right-wingers, while left-wing political misinformation has the opposite valence.54 The root of the problem is the rational irrationality of voters, which makes many of them avid and gullible consumers of political claptrap, not the availability of misinformation. In theory, we can imagine a government that comprehensively suppresses the flow of misinformation across all media. Then, even the most gullible voters would not fall prey to it because they would never get the chance to see it. But such a quasi-totalitarian state would have every incentive to suppress all opposition and fill the websites and airwaves with its own misinformation. Near-comprehensive government control of electronic and broadcast media has yielded awful results in countries like Russia and China. We should not assume it would be much better in the West. Even if government regulation could significantly reduce the flow of “misinformation,” defined in some objective sense, it would still not solve the problem of political ignorance. Rationally ignorant voters would remain unaware of most areas of government policy and thus remain unable to monitor it effectively. Moreover, they would still be highly biased in evaluating the information they learn, even if it was not inherently misleading. Some of the biggest and most egregious deceptions might be banished from the political scene. But more prosaic forms of ignorance and bias would remain.

15.4  Bottom-Up Strategies In contrast with top-down proposals for alleviating political ignorance, bottom-up approaches seek to address the problem in ways that empower ordinary people rather than limiting their influence. Some straightforwardly seek to increase voters’ knowledge. Others seek to restructure decision-making institutions to incentivize ordinary citizens to make better-informed choices. Like top-down approaches, bottom-up strategies have a variety of shortcomings and constraints. But some are nonetheless promising. Empowering people to “vote with their feet” is likely to have the most significant potential. 15.4.1  Increasing Knowledge through Education The seemingly most straightforward bottom-up strategy for increasing voter knowledge is to rely on education. If voters are ignorant about

298  Ilya Somin politics and government, why not just teach them? The need to create competent voters is a long-standing traditional rationale for public education. In theory, it should be possible to require citizens to become knowledgeable voters as a condition of graduating from high school. Students will only get a diploma if they demonstrate their knowledge of the structure of the political system, key areas of government policy, basic economics, and any other information that may be essential to becoming a good voter.55 From Thomas Jefferson’s day to the present, reformers have often argued that we can alleviate public ignorance through education.56 Jefferson famously wrote that “the remedy” for voter ignorance is “to inform their discretion by education.”57 But using education to conquer voter ignorance is a much more difficult proposition than optimists assume. Strikingly, voter knowledge levels have mainly remained stagnant – and low – over the last fifty to sixty years, even as education levels have significantly risen.58 Today, the average American adult has several years more formal education than in the early 1960s. But that has done little to increase voter knowledge. Perhaps things would be different if only schools adopted better curricula. A school system focused on increasing political knowledge could surely make significant progress. But real-world politicians and school officials have little incentive to structure public education in this way. Having been elected by a largely ignorant electorate, they have little incentive to prioritize increasing that knowledge. Increases might even threaten their grip on power! Worse, the actual incentives of education officials are not merely to neglect political knowledge but to use public schools to indoctrinate students in the preferred ideology of the party in power or the majority of the community. Indoctrination was one of the main reasons governments established public education in the United States and Europe in the first place.59 And such tendencies persist today. In the United States, both conservative “red” states and liberal “blue” states seek to develop school curricula that promote their respective ideologies.60 Current battles over “critical race theory” in American schools are the latest iteration of a long history of ideological conflict over school curricula.61 With both left and right seeking to capture schools to promote their ideologies rather than create a balanced curriculum that increases citizen knowledge in some neutral sense, there is little chance that public schools will overcome the problem of political ignorance anytime soon. In theory, voters could carefully monitor school curricula and reward political leaders who adopt education policies that increase political knowledge while minimizing indoctrination. But if voters were that knowledgeable about education policy, it would mean there was not much of a problem of political ignorance to begin with. The very existence of widespread voter ignorance is an obstacle to its alleviation.62

Top-Down and Bottom-Up Solutions to Political Ignorance 299 Even if schools could be more effectively structured to increase political knowledge, it is not clear they could get students to learn enough to effectively monitor the many functions of modern government. That may require more instruction than any school system realistically has time for, at least if other educational goals are also pursued. Moreover, even if students, upon graduation, are fully knowledgeable about the significant policy issues of the day, their knowledge is likely to become dated as new problems emerge over time. Hopefully, an 18-year-old high school graduate will be a voter for many decades to come. During that time, new issues will undoubtedly arise. Unless voters are required to report for regular continuing political education – or reeducation – their knowledge will likely atrophy over the years.63 Some evidence suggests that political knowledge can be increased if governments adopt school choice policies, under which – instead of being assigned to a public school based on geography – parents are given vouchers that enable them to send their children to any public or private school of their choice.64 Such reforms are worth pursuing. Nevertheless, it is doubtful they will lead to more than modest increases in political knowledge.65 Even if education can significantly increase voter knowledge, there remains the problem of how to alleviate biased evaluation of political information. In theory, schools can pursue that objective as well. But it is likely to be even more difficult than increasing merely factual knowledge. Among other things, any such “de-biasing” education is itself susceptible to being biased due to capture by ideological factions and interest groups. 15.4.2  Paying Voters to Become Better Informed The core problem of political ignorance arises from voters’ incentives to be rationally ignorant. There is too little payoff to learning about political issues for the individual voter. If that is the problem, why not change voter incentives by paying them to learn? Economist Bryan Caplan, best known for his work on “rational irrationality,” has proposed the establishment of a “Voter Achievement Test,” under which any citizen who wants to can take a test of fundamental political and economic knowledge and get a monetary payment based on their performance.66 The higher the score, the higher the pay. This simple approach to addressing political ignorance deserves serious consideration. No one need be required to take the test. Moreover, unlike in the case of epistocracy, nobody need lose their right to vote if they refuse to take the test or perform poorly on it. The promise of payment could nonetheless incentivize large numbers of people to increase their political knowledge. If payments of $100 or even $1000 for high scores lead to creating a vastly more knowledgeable electorate, it would be a bargain!

300  Ilya Somin In addition, the test could be updated as new issues arise or additional policy-relevant information emerges. Unlike material learned in school years ago, the test content need never become outdated. Voters who have taken the test previously could be offered the opportunity to retake it every few years, earning more money. In this way, they can be incentivized to keep their knowledge of public policy up to date. The most obvious flaw in this strategy is the question of whether the government can be trusted to objectively determine what should be on the test and what the “right” answers are. It is easy to imagine how the test can be biased in favor of a particular political party or ideology.67 The dangers are similar to those that arise with epistocracy.68 But the risk is somewhat lessened because the test would not be used to exclude people from the franchise. Even if we cannot trust the state to develop an objective “Voter Achievement Test,” perhaps private philanthropists or educational organizations might do so. We might reduce the danger of bias by having an ideologically diverse group of experts design and revise the test. At least for now, I am skeptical that either the private or public sector can design a test-and-payment system that is likely to be both effective and widely accepted. Nevertheless, this bottom-up strategy for increasing voter knowledge deserves more significant consideration. At the very least, it is worth attempting as a small-scale experiment. Experimental evidence indicates that small payments for correct answers lead survey participants to give more accurate responses on tests of political knowledge – an increase of 11% if participants have no opportunity to study and 24% if they have a day to prepare.69 Larger payments combined with more study time might lead to far better results. Of all the strategies considered here, “paying for performance” is the one to which experts have given the slightest consideration.70 At the very least, we should devote more sustained analysis to it. That said, like the other options considered so far, it is more likely to be effective in combating pure ignorance than alleviating bias in evaluating information. In principle, a test can also be designed to improve performance on the latter front. But it would be a much more complicated task and even more susceptible to biases of its own. Similarly, we can augment the test to include an assessment of reasoning skills, aside from those needed to combat bias. For example, test-takers could be incentivized to learn economics, political theory, and other bodies of knowledge that could help them better assess the factual information they learn about specific policies and government structure. But selecting such background knowledge and determining what counts as a correct answer in this context would likely create even more opportunities for bias and manipulation than the selection of merely factual information.

Top-Down and Bottom-Up Solutions to Political Ignorance 301 15.4.3 Sortition Some political theorists and others have argued that we can use “sortition” to overcome the problem of political ignorance while simultaneously empowering ordinary voters in a way that does not cede vast power to elites.71 Sortition transfers decision-making authority over various issues to small groups of randomly selected citizens. It thereby overcomes the problem of rational ignorance by creating decision-making bodies with a smaller number of voters. Each vote matters more; thus, voters have more incentive to learn about the issues and assess opposing arguments without bias.. In addition, the participants could spend far more time evaluating issues than voters typically do in conventional elections. Because the participants are randomly selected, the resulting group can be representative of the whole population. Potentially, sortition might combine representative popular participation in government with a higher level of political knowledge than is likely under conventional democracy. Some sortition proposals would use randomly selected groups to decide on specific policy issues, such as education or environmental protection, or engage in oversight of government officials.72 More ambitious versions would replace the entire electorate with an “enfranchisement lottery,” under which only a small randomly selected group would get to vote in national elections. That group would then have the opportunity and incentive to discuss the issues in depth.73 Another option is to have a house of the legislature chosen by sortition, perhaps combined with one selected by traditional elections.74 Unfortunately, sortition is not nearly as good a solution to the problem of political ignorance as it might initially seem.75 Unless the participants study for a very long time, they are unlikely to become knowledgeable about more than a small fraction of the many issues the modern state addresses. We might alleviate this problem by having each body selected by sortition address only a narrow range of issues. But then there would be severe problems of coordination between them. Moreover, groups addressing a specific policy area might neglect significant trade-offs between that issue and others. For example, a group highly knowledgeable about education policy might still not know enough to assess the trade-off between devoting X amount of additional resources to public schools instead of devoting the same funds to law enforcement or environmental protection. Another way to make the participants better informed is to have them serve for long periods, perhaps even years at a time. But in that scenario, the participants would gradually become a kind of professional governing class and would no longer be just randomly selected, ordinary people. The problem of elite domination that sortition is intended to solve would return in a new form.

302  Ilya Somin Juries in the civil and criminal justice systems often have difficulty understanding the points at issue in cases with broad policy implications or complex scientific evidence.76 These problems are likely even more severe if we use jury-like sortition mechanisms to address a much more comprehensive range of policy issues. Among the difficulties likely to be encountered are the challenges of coordinating sortition-selected bodies that address different problems. Such coordination will become problematic if there are discrete juries or “mini-publics” addressing each issue. If we instead have one or a few bodies address many problems at once, that would dramatically increasing participants’ knowledge burden.77 Sortition systems are also vulnerable to manipulation in a variety of ways. The government could skew the selection procedure to ensure more of its supporters get selected. If, as in most proposals, the participants are expected to hear presentations about policy issues and engage in deliberation about them, there are many ways to bias the choice of presenters and the selection and framing of issues. A biased sortition process might well exacerbate the dangers of voter ignorance and irrationality rather than diminish them. Even without such biases, sortition systems will face difficult tradeoffs between representativeness and minimizing incentives for rational ignorance. If the group selected is small, rational ignorance is unlikely to be a problem, since each vote will have a high chance of decisiveness. But a small, randomly selected group can easily be unrepresentative. For example, a group of twenty people randomly selected from a population equally divided between Republicans and Democrats will have a 60% or larger majority for one party more than 50% of the time. The likelihood that some parts of the population will be significantly underrepresented increases if more than two groups have to be accounted for – for example, if there are more than two political parties or if we seek representativeness based on race, gender, ethnicity, religion, and other characteristics, in addition to political partisanship. Sortition may be a valuable tool for enabling popular participation in discrete decisions requiring evaluation of a narrow set of facts. The classic example is a jury trial to determine a specific defendant’s criminal or civil liability. By contrast, sortition is much more dubious when addressing complex, multifaceted issues.78 In those situations, the juror-deliberators are likely to face poor incentives, be too small a group to be representative, or have difficulty becoming informed unless they essentially become full-time policymakers – in which case they would no longer be randomly selected “ordinary” citizens at all. 15.4.4  Foot Voting Perhaps the most effective bottom-up strategy for empowering people to make better-informed political decisions is to increase their opportunities

Top-Down and Bottom-Up Solutions to Political Ignorance 303 to “vote with their feet.” People can vote with their feet in federal systems by choosing what regional or local government to live under, through international migration, and in the private sector. Both foot voting under federalism and foot voting through international migration are mechanisms by which individuals and families can exercise political choice – deciding what government policies they wish to live under. Many private-sector decisions have a similar function, particularly where private organizations provide services traditionally associated with local or regional governments, as in the case of private planned communities or private school choice.79 When it comes to the problem of ignorance, all three types of foot voting have enormous advantages over ballot-box voting. Whereas ballot-­ box voters have incentives to be rationally ignorant due to the very low probability of any one vote making a difference, foot voters have strong incentives to seek out relevant information because their decisions are highly likely to matter. If you are like most people, you probably devote much more time and effort to deciding what television or smartphone to buy than to choosing who to vote for in any election. That is not because the television is more important than who governs the country, but because the TV decision is highly likely to have a decisive effect. The TV you choose will probably end up in your house, and you will have the chance to use it. By contrast, if you turn on the TV and have the misfortune of seeing the president of the United States or some other powerful politician on screen, your chances of determining who holds that office and what policies they pursue are infinitesimally small. Basic economic theory and extensive empirical evidence indicate that foot voters seek out more and better information than ballot-box voters.80 They also have better incentives to evaluate the information they learn objectively.81 When a decision is highly likely to make a difference, people have stronger incentives to curb potential biases than if the odds of decisiveness are very low. These points even apply to foot voters in highly challenging situations, such as migrants fleeing authoritarian regimes that limit the flow of information or African-Americans deciding whether to leave the oppressive Jim Crow-era American South.82 Foot voting can effectively address both the problem of ignorance and bias in evaluating information. In that respect, it is more promising than any other strategy. Some argue that foot voting is not a truly “political” choice because it is often motivated by “economic” or personal considerations, such as seeking job opportunities. But the same is true of much ballot-box voting.83 The economic considerations that influence foot voting are often heavily influenced by government policy, such as housing and labor market regulation. In addition, many migrants move because of even more unambiguously “political” factors, such as the extent of personal liberty allowed

304  Ilya Somin by the government and the incidence of state-sponsored discrimination against their racial, ethnic, or religious group.84 International migrants tend to move to nations with greater freedom on various dimensions.85 Much can be done to expand opportunities for all three types of foot voting, most notably breaking down barriers to international migration and reducing restrictions on internal mobility within federal systems, such as exclusionary zoning, which blocks the construction of new housing in response to demand.86 Within the US and other federal systems, we can expand foot voting opportunities by decentralizing and limiting the powers of government, thereby leaving more room for local and regional variation in policy, and creating more opportunities for people to vote with their feet on a wide range of issues.87 It is likely impossible to subject every issue to foot voting. Some problems are so large-scale that they can only be addressed by a central government or, as in the case of global climate change, by international agreements and institutions. Nevertheless, there are ways to create far greater decentralization and foot voting opportunities than exist at present.88 Relative to other strategies for alleviating the problem of political ignorance, foot voting has the virtue of massively improving incentives for informed decision-making without empowering a small group of elites at the expense of the general public and without creating opportunities for incumbent government officials to skew the system in their favor. As we have seen, both flaws bedevil other, more conventional strategies for alleviating voter ignorance. There are a variety of objections to expanded foot voting, such as claims that it might lead to a “race to the bottom,” arguments that it will not benefit unpopular racial and ethnic minorities, and – most obvious of all – concerns that moving costs are too high for it to be effective, particularly for the poor. I have responded to each of these concerns in detail in previous work.89 Here, I will merely note that moving costs are balanced by the enormous gains many foot voters stand to achieve from going to places where better government policies and institutions can help them to become more productive and thereby earn vastly higher incomes than are possible in their original homes.90 Within the United States, reducing exclusionary zoning alone – aside from other barriers to mobility – could enable millions of people to move to areas where they could significantly increase their income – expanding US GDP by as much as 30% or more.91 Economists estimate that eliminating barriers to international migration could lead to a doubling of world GDP.92 And the biggest gains of this sort are likely to be realized by the poor and otherwise disadvantaged.93 This prospect can incentivize them to overcome moving cost barriers by borrowing against their future incomes, and other means – and also incentivize others to help finance these moves.

Top-Down and Bottom-Up Solutions to Political Ignorance 305 Far from promoting inequality, as some fear, expanded foot voting opportunities are a significant boon to the poor, disadvantaged, and oppressed.94 While fully exploiting the benefits of foot voting would require radical policy changes, especially regarding restrictions on international migration, even comparatively modest incremental reforms could enable significant progress. Partially breaking down barriers to internal mobility and partially decentralizing government policy would still expand the foot voting options available to millions of people. The same goes for incrementally reducing obstacles to international migration. If, for example, the United States were to increase its annual intake of immigrants by a mere 10% compared to pre-Covid-19 pandemic levels, that would provide an enormously valuable foot voting opportunity to some 100,000 people each year.95 In addition to practical concerns about the effectiveness of foot voting, there are also long-standing arguments that existing governments have a right to exclude would-be migrants based on analogies between the rights of governments and those of homeowners and private clubs.96 Alternatively, exclusion can be justified by claims that particular racial or ethnic groups are the rightful owners of a given territory and therefore have the right to exclude others.97 I have responded to such claims I detail in my book, Free to Move.98 There are also concerns that expanded foot voting might exacerbate various problems, such as increasing ethnic conflict, overburdening welfare states, increasing crime and terrorism, and spreading contagious diseases. These issues cannot be examined in detail here, though I have also addressed them in previous work.99 Here, I do not attempt to make anything approaching a comprehensive case for expanding foot voting. I merely highlight its potential value as a strategy for addressing the problem of political ignorance, and especially its advantages over other proposals for alleviating that danger.

15.5 Conclusion Voter ignorance is a severe flaw of democracy that is extremely difficult to overcome. A variety of strategies have been proposed to address it. To facilitate analysis, I have divided them into “top-down” and “bottom-up” approaches. We should not dismiss top-down processes out of hand. In some situations, they may be helpful or even unavoidable. But they have several common flaws. Bottom-up strategies have significant limitations of their own. Nevertheless, some are highly promising. The ones with the most significant potential are paying voters to increase their knowledge – an approach that deserves much more critical consideration than it has received – and expanding opportunities for foot voting.

306  Ilya Somin

Notes











1. This summary of the problem of political ignorance is partly adapted from Ilya Somin, Free to Move: Foot Voting, Migration, and Political Freedom (New York: Oxford University Press, rev. ed. 2022), ch. 1. 2. For different estimates of the likelihood of decisiveness, see Ilya Somin, Democracy and Political Ignorance: Why Smaller Government Is Smarter (Stanford: Stanford University Press, 2nd ed. 2016), 75–76. 3. For a more detailed discussion of the logic of rational ignorance, see Somin, Democracy and Political Ignorance, 75–91; cf. Ilya Somin, “Is Political Ignorance Rational?” in Routledge Handbook of Political Epistemology, eds. Michael Hannon and Jeroen de Ritter (London: Routledge, 2021). The idea of rational political ignorance was first developed by Anthony Downs, An Economic Theory of Democracy (New York: Harper & Row, 1957), ch. 13. 4. This section builds on my book Democracy and Political Ignorance: Why Smaller Government Is Smarter (Stanford: Stanford University Press, 2nd ed. 2016), which analyzes rational ignorance and its consequences in detail (see esp. chs. 1–4). I addressed a variety of criticisms of the arguments advanced in that book in Ilya Somin, “The Ongoing Debate Over Political Ignorance: Reply to My Critics,” Critical Review 27 (2015): 380–414. 5. See Somin, Democracy and Political Ignorance, ch. 3; Bryan Caplan, The Myth of the Rational Voter (Princeton, NJ: Princeton University Press, 2007), ch. 5; and Bryan Caplan, “Rational Ignorance vs. Rational Irrationality,” Kyklos 53 (2001): 3–21. 6. For a review of the evidence, see Somin, Democracy and Political Ignorance, 92–97. 7. For recent overviews of the evidence indicating widespread bias among voters on both left and right, see Brian Guay and Christopher D. Johnson, “Ideological Asymmetries and the Determinants of Politically Motivated Reasoning,” American Journal of Political Science (forthcoming); Peter H. Ditto et al., “At Least Bias Is Bipartisan: A Meta-Analytic Comparison of Partisan Bias in Liberals and Conservatives,” Perspectives on Psychological Science 14 (2019): 273–291; cf. Robin McKenna, “Asymmetrical Irrationality: Are Only Other People Stupid?” in Routledge Handbook of P ­ olitical Epistemology, eds. Michael Hannon and Jeroen de Ritter (London: Routledge, 2021); Keith E. Stanovich, “The Irrational Attempt to Impute Irrationality to One’s Political Opponents,” in Routledge Handbook of Political Epistemology, eds. Michael Hannon and Jeroen de Ritter (London: Routledge, 2021). 8. I introduced this analogy in Somin, Democracy and Political Ignorance, 93–94. 9. For a more detailed discussion, see ibid., 78. 10. For a more detailed discussion of why rational ignorance does not require careful calculation and is consistent with the use of crude heuristics, see Ilya Somin, “Rational Ignorance,” in Routledge International Handbook of Ignorance Studies, eds. Matthias Gross and Linsey J. McGoey (London: Routledge, 2015); and Brad R. Taylor, “The Psychological Foundations of Rational Ignorance: Biased Heuristics and Decision Costs,” University of Queensland, unpublished paper, Aug. 31, 2019, available at https://papers. ssrn.com/sol3/papers.cfm?abstract_id=3443280. 11. For recent overviews of the evidence, see, e.g., Somin, Democracy and Political Ignorance, ch. 1; Jason Brennan, Against Democracy (Princeton, NJ: Princeton University Press, 2016); Christopher Achen and Larry

Top-Down and Bottom-Up Solutions to Political Ignorance 307













Bartels, Democracy for Realists: Why Elections Do Not Produce Responsive Government (Princeton, NJ: Princeton University Press, 2016); Rick Shenkman, Just How Stupid Are We? Facing the Truth about the American Voter (New York: Basic Books, 2008). 12. For numerous examples, see Somin, Democracy and Political Ignorance, ch. 1. 13. Ibid., 1. 14. Ibid., 20. 15. Ibid., 18. 16. See, e.g., Ipsos-MORI, Perils of Perception: A Fourteen-Country Study (Ipsos-MORI, 2014) (detailing similar ignorance in many leading democracies); Ipsos-MORI, Perils of Perception 2018 (Ipsos-MORI, 2018) (same); Bobby Duffy, The Perils of Perception: Why We Are Wrong about Nearly Everything (London: Atlantic Books, 2018), chs. 4–9. 17. See, e.g., Morris Fiorina, Retrospective Voting in American Presidential Elections (New Haven, CT: Yale University Press, 1981); V. O. Key, The Responsible Electorate (Cambridge, MA: Harvard University Press, 1966), 60–61; Arthur Lupia and Matthew McCubbins, The Democratic Dilemma: Can Citizens Learn What They Need to Know? (New York: Cambridge University Press, 1998). 18. See, e.g., Donald Wittman, The Myth of Democratic Failure (Chicago: University of Chicago Press, 1995); Hélène Landemore, Democratic Reason: Politics, Collective Intelligence, and the Rule of the Many (Princeton, NJ: Princeton University Press, 2013); James Stimson, “A Macro Theory of Information Flow,” in Information and Democratic Processes, eds. John Ferejohn and James Kuklinski (Urbana: University of Illinois Press, 1990); Bernard Grofman and Julie Withers, “Information-Pooling Models of Electoral Politics,” in Information, Participation and Choice, ed. ­Bernard Grofman (Ann Arbor: University of Michigan Press, 1993); James Surowiecki, The Wisdom of Crowds: Why the Many Are Smarter Than the Few (New York: Doubleday, 2004), ch. 12. 19. See Somin, Democracy and Political Ignorance, ch. 4. 20. For a more detailed elaboration of this point, see ibid., 160–63. 21. Ibid. 22. The term has been popularized in recent political theory by Jason Brennan, Against Democracy (Princeton: Princeton University Press, 2016). See also Jason Brennan, “The Right to a Competent Electorate,” Philosophical Quarterly 61 (October 2011): 700–24; Jason Brennan, “In Defense of Epistocracy: Enlightened Preference Voting,” in Routledge Handbook of ­ Political Epistemology, eds. Michael Hannon and Jeroen de Ritter (London: Routledge, 2021); Jason Brennan and Christopher Freiman, “Why Paternalists Must Endorse Epistocracy,” Journal of Ethics and Social ­Philosophy 21 (2022): 329–53. 23. This critical evaluation of epistocracy draws on my earlier piece on the subject. See Ilya Somin, “The Promise and Peril of Epistocracy,” Inquiry 16 (2019): 27–34 (symposium on Brennan, Against Democracy). 24. I discuss this aspect of the limited franchise of Athenian democracy in Somin, “Democracy and Political Knowledge in Ancient Athens,” Ethics 119 (2009): 585–90. See also Josiah Ober, Knowledge and Democracy: Innovation and Learning in Classical Athens (Princeton: Princeton University Press, 2008). 25. John Stuart Mill, Considerations on Representative Government (Indianapolis: Bobbs-Merrill, 1958 [1861]), 140–42. 26. See US Census Bureau, “Quick Facts,” July 1, 2018, available at https:// www.census.gov/quickfacts/fact/table/US/PST045218 (estimating that 22.6% of the US population consists of persons under the age of 18).

308  Ilya Somin 27. I have tentatively proposed this myself. See, e.g., Ilya Somin, “Should We Let 16-Year-Olds Vote?” Volokh Conspiracy, Washington Post, Sept. 19, 2014, available at https://www.washingtonpost.com/news/volokhconspiracy/wp/2014/09/19/should-we-let-16-year-olds-vote/?utm_term=. ea9529c2c42a. 28. A 2018 study found that only 36% of Americans could pass the citizenship test required for legal immigrants to become citizens: Woodrow Wilson Foundation Survey, Oct. 2018, available at https://woodrow.org/news/ national-survey-finds-just-1-in-3-americans-would-pass-citizenship-test/. 29. Brennan, Against Democracy. 30. Ibid., ch. 8. 31. For a more detailed discussion of this problem, see Somin, “Promise and Peril,” 30–31. 32. Brennan, “In Defense of Epistocracy.” This can be done by comparing the views of the voters in question with those of people with similar background characteristics (e.g., income, race, gender, and so on) but with higher levels of political knowledge. 33. Brennan, Against Democracy, ch. 8. 34. For a more extended critique of Brennan’s proposals, see Somin’s “Promise and Peril,” 31–33. 35. For citations to relevant evidence, see Somin, Democracy and Political Ignorance, 94–100; see also Diana Mutz, Hearing the Other Side: Deliberative Versus Participatory Democracy (New York: Cambridge University Press, 2006), 29–41. 36. See Paul Tucker, Unelected Power: The Quest for Legitimacy in Central Banking and the Regulatory State (Princeton: Princeton University Press, 2018). 37. See, e.g., Stephen Breyer, Breaking the Vicious Circle: Toward Effective Risk Regulation (Cambridge: Harvard University Press, 1993), ch. 3; Cass R. Sunstein, Risk, and Reason: Safety, Law and the Environment (New York: Oxford University Press, 2002). 38. See Ilya Somin, “Nondelegation Limits on COVID Emergency Powers: Lessons from the Eviction Moratorium and Title 42 Cases,” NYU Journal of Law and Liberty 15 (2022): 658–98. 39. Ibid. 40. F. A. Hayek, “The Use of Knowledge in Society,” American Economic Review 4 (1945): 519–30. 41. This and other similar limitations of expert knowledge are outlined in Mario Rizzo and Glen Whitman, Escaping Paternalism: Rationality, Behavioral Economics, and Public Policy (Cambridge: Cambridge University Press, 2020), ch. 7. 42. This is especially likely in light of growing evidence that harsh lockdowns did little to reduce Covid-19 mortality. See, e.g., Jonas Herby, Lars Jonung, and Steve H. Hanke, “A Literature Review and Meta-Analysis of the Effects of Lockdowns on COVID-19 Mortality,” Studies in Applied Economics No. 200, Johns Hopkins Institute for Applied Economics, Global Health, and the Study of Business Enterprise, Jan. 2022, available at https://sites. krieger.jhu.edu/iae/files/2022/01/A-Literature-Review-and-Meta-Analysisof-the-Effects-of-Lockdowns-on-COVID-19-Mortality.pdf; Virat Agarwal et al., “The Impact of the COVID-19 Pandemic and Policy Responses on Excess Mortality,” National Bureau of Economic Research, June 2021, available at https://www.nber.org/papers/w28930. 43. See Somin, Free to Move, pp. 155–60. 44. See Martin Baekgaard et al., “The Role of Evidence in Politics: Motivated Reasoning and Persuasion among Politicians,” British Journal of Political Science 49 (2017): 1117–40.

Top-Down and Bottom-Up Solutions to Political Ignorance 309 45. For an overview, see Rizzo and Whitman, Escaping Paternalism, ch. 9, esp. 330–47. 46. See, e.g., Richard L. Hasen, Cheap Speech (New Haven: Yale University Press, 2021); Russell Muirhead and Nancy Rosenblum, Many People Are Saying: The New Conspiracism and the Assault on Democracy; James Ball, Post-Truth: How Bullshit Conquered the World (New York: Biteback, 2018); Cailyn O’Connor and James Owen Weatherall, The Misinformation Age: How False Beliefs Spread (New Haven: Yale University Press, 2018). 47. See, e.g., Hasen, Cheap Speech. 48. See Yasmeen Serhan, “Macron’s War on Fake News,” The Atlantic, Jan. 6, 2018. 49. For a recent statement of this view, see Hasen, Cheap Speech. 50. For an overview of reasons why that may not be the case, see Ilya Somin, “Are Public Ignorance and Misinformation Getting Worse?,” Reason, Feb. 15, 2022, available at https://reason.com/volokh/2022/02/15/are-publicignorance-and-misinformation-getting-worse/; Matthew Yglesias, “The ‘Misinformation Problem’ Seems like Misinformation,” Slow Boring, Feb. 15, 2022, available at https://www.slowboring.com/p/misinformation-myth?s=r. 51. See Somin, Democracy and Political Ignorance, ch. 1. 52. See Ilya Somin, “The Case against Imposing Common Carrier Restrictions on Social Media Sites,” Reason, July 8, 2021, available at https:// reason.com/volokh/2021/07/08/the-case-against-imposing-common-carrier-restrictions-on-social-media-sites/ (going over the data); Pew Research Center, “More Than Eight-in-Ten Americans Get Their News from Digital Devices,” Jan. 12, 2021, available at https://www.pewresearch.org/ fact-tank/2021/01/12/more-than-eight-in-ten-americans-get-news-fromdigital-devices/ (data showing that many more Americans follow television or media organization website news than do so on social media). 53. See discussion in Part I; see also Ilya Somin, “The Demand for Political Misinformation Is a Bigger Danger Than the Supply,” Reason, Apr. 15, 2022, available at https://reason.com/volokh/2022/04/15/the-demand-for-political-misinformation-is-a-bigger-danger-than-the-supply/. 54. Ibid. 55. Different theories of political participation imply additional knowledge prerequisites. See Somin, Democracy and Political Ignorance, ch. 2; Jamie Terence Kelly, Framing Democracy: A Behavioral Approach to Democratic Theory (Princeton: Princeton University Press, 2012), ch. 4. 56. See, e.g., Ben Berger, Attention Deficit Democracy: The Paradox of Civic Engagement (Princeton, NJ: Princeton University Press, 2011), 153–57. 57. Thomas Jefferson, “Letter to William C. Jarvis, Sept. 28, 1820,” in The Writings of Thomas Jefferson, Vol. 15, eds. Andrew Lipscomb and Albert Bergh (Washington, DC, 1904), 278. 58. Somin, Democracy and Political Ignorance, 198–200. 59. See, e.g., Eugen Weber, Peasants into Frenchmen (Stanford: Stanford University Press, 1976); E. G. West, Education and the State, 3rd ed. (Indianapolis: Liberty Press, 1994), 84–107; and John R. Lott Jr., “An Explanation for Public Provision of Schooling: The Importance of Indoctrination,” Journal of Law and Economics 33 (1990): 199–231. 60. For some recent examples, Ilya Somin, “Public Education as Public Indoctrination,” Reason, Jan. 12, 2020, available at https://reason.com/ volokh/2020/01/12/public-education-as-public-indoctrination/. 61. See, e.g., Laura Meckler and Hannah Natanson, “New Critical Race Theory Laws Have Teachers Scared, Confused and Self-Censoring,” Washington Post, Feb. 14, 2022, available at https://www.washingtonpost.com/ education/2022/02/14/critical-race-theory-teachers-fear-laws/.

310  Ilya Somin 62. I make this point, 202-0, in Somin, Democracy and Political Ignorance, 202–03. 63. Ibid., 203–04. 64. Ibid., 202–03. 65. Ibid (explaining the reasons why). 66. Bryan Caplan, “A Cheap, Inoffensive Way to Make Democracy Work Better,” Econlog, Oct. 9, 2013, available https://www.econlib.org/ archives/2013/10/a_cheap_inoffen.html. 67. I emphasize this limitation in evaluating the “pay for performance” approach in Somin, Democracy and Political Ignorance, 219–21. 68. See the discussion earlier in this chapter. 69. Arthur Lupia and Markus Prior, “Money, Time and Political Knowledge: Distinguishing Quick Recall and Political Learning Skills,” American Political Science Review 52 (2008), 169–83; for additional discussion, see Somin, Democracy and Political Ignorance, 219–20. 70. For rare exceptions, none of which are close to being comprehensive, see, e.g., Somin, Democracy and Political Ignorance, 219–21; Caplan, “A Cheap, Inoffensive Way to Make Democracy Work Better”; and Arthur Lupia, Uninformed: Why People Know So Little about Politics and What We Can Do about It (New York: Oxford University Press, 2016), 174–75. 71. See, e.g., Claudio López-Guerra, Democracy, and Disenfranchisement: The Morality of Electoral Exclusions (Oxford University Press, 2014); Claudio Lopez-Guerra, “The Enfranchisement Lottery,” Philosophy, Politics, and Economics (Oct. 2010); Helene Landemore, Open Democracy: Reinventing Democracy for the 21st Century (Princeton: Princeton University Press, 2020); Helene Landemore, “Deliberation, Cognitive Diversity, and Inclusiveness: An argument for the Random Selection of Representatives,” Synthese 190 (2013): 1209–31; Ethan Leib, Deliberative Democracy in America: A Proposal for a Popular Branch of Government (University Park: Penn State Press, 2004); Alexander Guerrero, “The Epistemic Case for Non-Electoral Forms of Democracy,” in Routledge Handbook of Political Epistemology, eds. Michael Hannon and Jeroen de Ritter (London: Routledge, 2021); Alexander Guerrero, The Lottocratic Alternative, (New York: Oxford University Press, 2021). 72. See, e.g., Landemore, Open Democracy (promoting a wide-ranging version of this approach); Samuel Bagg, “Sortition as Anti-Corruption: Popular Oversight against Elite Capture,” American Journal of Political Science (forthcoming). 73. See, e.g., Claudio López-Guerra,  Democracy and Disenfranchisement; Lopez-Guerra, “Enfranchisement Lottery.” 74. See, e.g., Arash Abizadeh, “Representation, Bicameralism, Political Equality, and Sortition: Reconstituting the Second Chamber as a Randomly Selected Assembly,” Perspectives on Politics 19 (2021): 791–806. 75. This section builds on criticisms of sortition in Somin, Democracy and Political Ignorance, 208–11. 76. To summarize some relevant evidence, see Ilya Somin, “Jury Ignorance and Political Ignorance,” William and Mary Law Review 55 (2014): 1167–93, 1179–87. 77. This problem is highlighted in Ilya Somin, “Deliberative Democracy and Political Ignorance,” Critical Review 22 (2010): 253–79, 271–72. I focused on proposals for local, small-scale deliberative democracy. But the same considerations apply with equal force to sortition. 78. For a more detailed discussion, see Somin’s “Jury Ignorance and Political Ignorance.”

Top-Down and Bottom-Up Solutions to Political Ignorance 311 79. See Somin, Free to Move, ch. 4. 80. For extensive reviews, see ibid., ch. 1, and Somin, Democracy and Political Ignorance, ch. 5. 81. See evidence discussed in works cited in the previous note. 82. Ibid. 83. For a more detailed discussion of this objection, see Somin, Free to Move, 39–42. 84. Ibid, 39–42, 69–70. 85. Ibid., 69–70. 86. For overviews of such barriers and what can be done to address them, see ibid., chs. 2–3. 87. Ibid., chs. 2 and 7. 88. Ibid. 89. Ibid., chs. 2 and 3. 90. For details, see ibid. 91. Ibid., 52–53. 92. Ibid., 71–74. 93. Ibid., 52–53 and studies cited therein. 94. For additional analysis of this point, see ibid., 53–57, 77–80. 95. Ibid., 202. 96. Both analogies are advanced in Christopher Heath Wellman, “Freedom of Movement and the Right to Enter and Exit,” in Migration in Political Theory: The Ethics of Movement and Membership, eds. Sarah Fine and Leah Ypi (New York: Oxford University Press, 2016), 83, 87; see also Christopher Heath Wellman, “Immigration and Freedom of Association,” Ethics 119 (2008): 109–41. 97. See, e.g., Michael Walzer, Spheres of Justice (New York: Basic Books, 1983), ch. 2; Michael Walzer, “Exclusion, Injustice, and the Democratic State,” Dissent 40 (1993): 55–64; David Miller, “Immigration: The Case for Limits,” in Contemporary Debates in Applied Ethics, eds. Andrew Cohen and Christopher Heath Wellman (Oxford: Blackwell, 2005); David Miller, Strangers in Our Midst: The Political Philosophy of Immigration (Cambridge, MA: Harvard University Press, 2016), 60–68. 98. Somin, Free to Move, ch. 5. 99. Ibid., ch. 6.

References Abizadeh, Arash. 2021. “Representation, Bicameralism, Political Equality, and Sortition: Reconstituting the Second Chamber as a Randomly Selected Assembly.” Perspective of Politics 19: 791–806. Achen, Christopher, and Bartels, Larry. 2016. Democracy for Realists: Why Elections Do Not Produce Responsive Government. Princeton: Princeton University Press. Agarwal, Virat, Cantor, Johnatan H., Sood, Neeraj, and Whaley, Christopher M. 2021. “The Impact of the COVID-19 Pandemic and Policy Responses on Excess Mortality.” NBER Working Papers 28930, National Bureau of Economic Research. Baekgaard, Martin, et al. 2017. “The Role of Evidence in Politics: Motivated Reasoning and Persuasion Among Politicians.” British Journal of Political Science 49: 1117–40. Bagg, Samuel. 2022. “Sortition as Anti-Corruption: Popular Oversight against Elite Capture.” American Journal of Political Science.

312  Ilya Somin Ball, James. 2018. Post-Truth: How Bullshit Conquered the World. New York: Biteback. Berger, Ben. 2011. Attention Deficit Democracy: The Paradox of Civic Engagement, 153–157. Princeton: Princeton University Press. Brennan, Jason. 2011. “The Right to a Competent Electorate.” Philosophical Quarterly 61: 700–24. ———2016. Against Democracy. Princeton: Princeton University Press. ———2021. “In Defense of Epistocracy: Enlightened Preference Voting.” In Routledge Handbook of Political Epistemology, eds. Michael Hannon and Jeroen de Ritter. London: Routledge. Brennan, Jason, and Freiman, Christopher. 2022. “Why Paternalist Must Endorse Epistocracy.” Journal of Ethics and Social Philosophy 21: 329–53. Breyer, Stephen. 1993. Breaking the Vicious Circle: Toward Effective Risk Regulation, ch. 3. Cambridge: Harvard University Press. Caplan, Byran. 2001. Rational Ignorance vs. Rational Irrationality. Kyklos 53: 3–12. ———2007. The Myth of the Rational Voter, ch. 5. Princeton: Princeton University Press. ———2013. “A Cheap, Inoffensive Way to Make Democracy Work Better.” Econlog, Oct. 9. Ditto, Peter H., et al. 2019. “At Least Bias Is Bipartisan: A Meta-Analytic Comparison of Partisan Bias in Liberal and Conservatives.” Perspective on Psychological Sciences 14: 273–91. Downs, Anthony. 1957. An Economic Theory of Democracy, ch. 13. New York: Harper & Row. Duffy, Bobby. 2018. The Perils of Perception: Why We Are Wrong About Nearly Everything, chs. 4–9. London: Atlantic Books. Fiorina, Morris. 1981. Retrospective Voting in American Presidential Elections. New Haven: Yale University Press. Grofman, Bernard, and Withers, Julia. 1993. “Information-Pooling Models of Electoral Politics.” In Information, Participation and Choice, ed. B. Grofman. Ann Arbor: University of Michigan Press. Guerrero, Alexander. 2021. The Lottocratic Alternative. New York: Oxford University Press. ———2021. “The Epistemic Case for Non-Electoral Forms of Democracy.” In Routledge Handbook of Political Epistemology, eds. Michael Hannon and Jeroen de Ritter. London: Routledge. Guay, Brian, and Christopher D. Johnson. 2021. “Ideological Asymmetries and the Determinant of Politically Motivated Reasoning.” American Journal of Political Science 66 (2): 285–301. Hasen, Richard L. 2021. Cheap Speech. New Haven: Yale University Press. Hayek, F.A. 1945. “The Use of Knowledge in Society.” American Economic Review 4: 519–30. Herby, Jonas, Jonung, Lars, and Steve H. Hanke. 2022. “A Literature Review and Meta-Analysis of the Effects of Lockdowns on COVID-19 Mortality.” Studies in Applied Economics 200. Ipsos-MORI. 2014. Perils of Perception: A Fourteen-Country Study. London: Ipsos-MORI. ———2018. Perils of Perception 2018. London: Ipsos-MORI.

Top-Down and Bottom-Up Solutions to Political Ignorance 313 Jefferson, Thomas. 1904. Letter to William C. Jarvis, Sept. 28, 1820. In The Writings of Thomas Jefferson, Vol. 15, eds. A. Lipscomb and A. Bergh, 278. Washington, DC: Nabu Press. Kelly, Jamie T. 2012. Framing Democracy: A Behavioral Approach to Democratic Theory, ch. 4. Princeton: Princeton University Press. Key, V.O. 1966. The Responsible Electorate, 60–61. Cambridge: Harvard University Press. Landemore, Hélène. 2013. Democratic Reason: Politics, Collective Intelligence, and the Rule of the Many. Princeton: Princeton University Press. ———2013. “Deliberation, Cognitive Diversity, and Inclusiveness: An Argument for the Random Selection of Representatives.” Synthese 190: 1209–31. ———2020. Open Democracy: Reinventing Democracy for the 21st Century. Princeton: Princeton University Press. Leib, Ethan. 2004. Deliberative Democracy in America: A Proposal for a Popular Branch of Government. University Park: Penn State Press. Lopez-Guerra, Claudio. 2010. “The Enfranchisement Lottery.” Philosophy, Politics, and Economics 10: 211–33. ———2014. Democracy and Disenfranchisement: The Moralist of Electoral Exclusions. New York: Oxford University Press. Lott, John R. Jr. 1990. “An Explanation for Public Provisions of Schooling: The Importance of Indoctrination.” Journal of Law and Economics 33: 199–231. Lupia, Arthur, and Prior, Markus. 2008. “Money, Time and Political Knowledge: Distinguished Quick Recall and Political Learning Skills.” American Political Science Review 52: 169–83. Lupia, Arthur. 2016. Uninformed: Why People Know So Little about Politics and What We Can Do about It. New York: Oxford University Press. Lupia, Arther, and McCubbins, Matthew. 1998. The Democratic Dilemma: Can Citizens Learn What They Need to Know? New York: Cambridge University Press. McKenna, Robin. 2021. “Asymmetrical Irrationality: Are Only Other People Stupid?” In Routledge Handbook of Political Epistemology, eds. Michael Hannon and Jeroen de Ritter. London: Routledge. Meckler, Laura, and Natanson, Hannah. 2022. “New Critical Race Theory Laws Have Teachers Scared, Confused and Self-Censoring.” Washington Post, Feb. 14. Mill, John S. 1861. Considerations of Representative Government. Indianapolis: Bobbs-Merrill. Miller, David. 2005. “Immigration: The Case for Limits.” In Contemporary Debates in Applied Ethics, eds. A. Cohen and C.H. Wellman. Oxford: Blackwell. ———2016. Strangers in Our Midst: The Political Philosophy of Immigration, 60–68. Cambridge: Harvard University Press. Muirhead, Russell, and Rosenblum, Nancy. 2018. Many People Are Saying: The New Conspiracism and the Assault on Democracy. Princeton: Princeton University Press. Mutz, Diana. 2006. Hearing the Other Side: Deliberative Versus Participatory Democracy. New York: Cambridge University Press. Ober, Josiah. 2008. Knowledge and Democracy: Innovation and Learning in Classical Athens. Princeton: Princeton University Press.

314  Ilya Somin O’Connor, Cailyn, and Weatherall, James O. 2018. The Misinformation Age: How False Beliefs Spread. New Haven: Yale University Press. Riccards, Patrick. 2018. “National Survey Finds Just 1 in 3 Americans Would Pass Citizenship Test.” The Woodrow Wilson National Fellowship Foundation, Oct. 3. Rizzo, Mario, and Whitman, Glen. 2020. Escaping Paternalism: Rationality, Behavioral Economics, and Public Policy. Cambridge: Cambridge University Press. Serhan, Yasmeen. 2018. “Macron’s War on Fake News.” The Atlantic, Jan. 6. Shearer, Elisa. 2021. “More Than Eight-In-Ten Americans Get News from Digital Devices.” Pew Research Center, Jan. 12. Shenkman, Rick. 2008. Just How Stupid Are We? Facing the Truth about the American Voter. New York: Basic Books. Somin, Ilya. 2009. “Democracy and Political Knowledge in Ancient Athens.” Ethics 119: 585–90. ———2010. “Deliberative Democracy and Political Ignorance.” Critical Review 22: 253–79. ———2014. “Jury Ignorance and Political Ignorance.” William and Mary Law Review 55: 1167–93. ———2014. “Should We Let 16-Year-Olds Vote?” Washington Post, Sept. 19. ———2015. “Rational Ignorance.” In Routledge International Handbook of Ignorance Studies, eds. M. Gross and L. J. McGoey. London: Routledge. ———2015. “The Ongoing Debate over Political Ignorance: Reply to My Critics.” Critical Review 27: 380–414. ———2016. Democracy and Political Ignorance: Why Smaller Government Is Smarter. 2nd ed. Stanford: Stanford University Press. ———2019. “The Promise and Peril of Epistocracy.” Inquiry 16: 27–34. ———2020. “Public Education as Public Indoctrination.” Reason, Jan. 12. ———2021. “Is Political Ignorance Rational.” In Routledge Handbook of Political Epistemology, eds. Michael Hannon and Jeroen de Ritter. London: Routledge. ———2021. “The Case against Imposing Common Carrier Restrictions on Social Media Sites.” Reason, July 8. ———2022. “Are Public Ignorance and Misinformation Getting Worse?” Reason, Feb. 15. ———2022. Free to Move: Foot Voting, Migration, and Political Freedom. Rev. ed. New York: Oxford University Press. ———2022. “The Demand for Political Misinformation Is a Bigger Danger Than the Supply.” Reason, Apr. 15. ———2022. “Nondelegation Limits on COVID Emergency Powers: Lessons from the Eviction Moratorium and Title 42 Cases.” NYU Journal of Law and Liberty 15: 658–89. Stanovich, Keith E. 2021. “The Irrational Attempt to Impure Irrationality to One’s Political Opponents.” In Routledge Handbook of Political Epistemology, eds Michael Hannon and Jeroen de Ritter. London: Routledge. Stimson, James. 1990. “A Macro Theory of Information Flow.” In Information and Democratic Processes, eds. J. Ferejohn and J. Kuklinski. Urbana: University of Illinois Press. Sunstein, Cass R. 2002. Risk and Reason: Safety, Law and the Environment. New York: Oxford University Press.

Top-Down and Bottom-Up Solutions to Political Ignorance 315 Surowiecki, James. 2004. The Wisdom of Crowds: Why the Many Are Smarter Than the Few, ch. 12. New York: Doubleday. Taylor, Brad R. 2019. The Psychological Foundation of Rational Ignorance: Biased Heuristics and Decision Costs. Unpublished paper, available at https:// papers.ssrn.com/sol3/papers.cfm?abstract_id=3443280 Tucker, Paul. 2018. Unelected Power: The Quest for Legitimacy in Central Banking and the Regulatory State. Princeton: Princeton University Press. US Census Bureau. 2018. Quick Facts. US Census Bureau, July 1. Walzer, Michael. 1983. Spheres of Justice, ch. 2. New York: Basic Books. ———1993. “Exclusion, Injustice, and the Democratic State.” Dissent 40: 55–64. Weber, Eugen. 1976. Peasants into Frenchmen. Stanford: Stanford University Press. Wellman, Christopher H. 2008. Immigration and Freedom of Association. Ethics 119: 109–41. ———2016. Freedom of Movement and the Right to Enter and Exit. In Migration in Political Theory: The Ethics of Movement and Membership, eds. S. Fine and L. Ypi, 83, 87. New York: Oxford University Press. West, E.G. 1994. Education and the State. 3rd ed. Indianapolis: Liberty Press. Wittman, Donald. 1995. The Myth of Democratic Failure. Chicago: University of Chicago Press. Yglesias, Matthew. 2022. “The ‘Misinformation Problem’ Seems like Misinformation.” Slow Boring, Feb. 15.

Index

Note: Page numbers followed by “n” refers notes. Aberdein, A. 61 academia 180–184; corporate world and 17; diversity in 17, 186 Achen, C. 28 acquisition: data 241; knowledge 3, 4, 208, 241; modes of corruption 157 activation modes of corruption 157 active listeners/listening 10, 221, 223 actively open-minded thinking (AOT) 171–172, 175, 179 adversarial collaboration 9, 183, 185–186 affective polarization 105–107, 110n53, 118–119, 129, 172, 284n3 Against Democracy (Brennan) 291–292 agential cynicism 162 Ahlstrom-Vij, K. 39, 59 Alfano, M. 138, 160 alliances 270, 273, 281, 282 alternative facts 34, 38, 47 Alternative for Germany (AfD) 63 American Dream 126, 166, 219–220 analytic epistemology 2–3, 75, 76–79, 84 anchoring biases 170, 171, 173 Anderson, E. 18, 136, 144 anger 118; moral disapproval and 119; partisan bias 98; political 118; and resentment 117–118, 123, 126, 128, 130n5; see also political resentment antagonistic narrative strategies 103 anthropological cynicism 162–163 AOT see actively open-minded thinking Appiah, K. A. 28 applied philosophy 53 Arendt, H. 52, 96, 108n26

argument 25, 202, 257; from bias 239– 241; consensus 41; from epistemic colonization 241–247; in philosophy of science 173; from polarization 222, 223, 224; populism and populist 61; Pyrrhonist 238; race to the bottom 304; rhetorical forms of 55, 64; skeptical 234–235, 236, 237 Aristotle 5, 17, 206, 231n27 Arneson, P. 164 Arnett, R. C. 164 Arnold, J. R. 109n41 assumptions: about communication 97; behavioral 25–30; of proof 18–25 attitudes 99, 216; associative 38; belief polarization 277; beliefs and 230n12; concurring agents 42; conservative political 37; convictions and 52; cynical 161–163; democratic 245; detrimental 36; epistemic 77, 163; false 77; negative 277; political 177; self-protective insistence 76; of Tea Party supporters 122; test hypotheses 59; toward scientific evidence 181; values and 159 authentic people 55, 62, 64 authoritarianism 178, 182, 186n3, 186n6 Bailey, O. 217, 220, 222, 230n12 Baldwin, J. 130n16 Bartels, L. M. 28, 36, 38, 257 Battaly, H. 3 Becker, J. 109n43 behavioral assumptions 25–30 Beitz, C. 261

Index 317 belief polarization 276–277, 278, 279–284 Berkeley, G. 235 Berlin, I. 231n26, 247n6 best-up-to-now strategy 137, 140–142, 149n9 better informed voters 299–300 biases: argument from 239–241; cognitive 2, 5, 30, 53, 56–59; confirmation 60, 64; empirically strange 171–173; heuristics and 52, 56–59; institutional 242–243; outlier 9; partisan 98, 173, 295; sortition 302; systemic 242; see also specific types Biden, J. 106 Biden administration 294 bigotry 118 Bildung 208 Black Sheep Effect 277 Blair premiership 158, 160 Blau, A. 266n19 Bloom, P. 223 bootstrapped corroboration 42–43 bottom-up strategies 287–288; foot voting 302–305; increasing knowledge through education 297–299; paying voters to become better informed 299–300; sortition 301–302 Bramson, A. 209n2 Brennan, J. 68, 157, 293; Against Democracy 291–292 Brexit 121, 135, 184, 224 Breyer, S. 294 Brostrom, S. 208 Brown, G. 130n4 burdened virtue 153 Cadwalladr, C. 99–100, 108n34 Cambridge Analytica 109n37 Cameron, D. 126, 127 campaign contributions 255–257, 259–263, 265n10 Caplan, B. 238–239, 245, 289, 299 Cassam, Q. 61, 67, 155; Vices of the Mind: From the Intellectual to the Political 3 Cato the Younger 234 ceilings as procedural elements 261 certainty 33, 36, 95, 104 Chaloupka, W.: Everybody Knows 165–166

character: affective 128; corruption and 154–157; epistemic 61, 75, 136–137, 140, 143, 149n4, 200, 202, 208; ethical and epistemic 45, 47; moral 118, 153, 157, 160; person of bad 219–220; of populism 62; social 79; traits 3, 4 Chavez, H. 51, 55 Christiano, T. 255, 264n3 civic duty 272 civic political informedness 39–40 civic political sophistication 36–39 civic respect 272–274, 279–281 civic virtue 154 classical cognitive biases 56–59 Clifford, W. K. 206 Clinton, H. 100, 119, 130n9 Code, L. 206 co-determination 208, 209n9 cognition: cultural identity-protective 116; need for 171–172, 175; social dimension of 76–79 cognitive biases 2, 5, 30, 53, 56–59 cognitive diversity 20 cognitive empathy 222 cognitive miserliness 56–67 Cohen, J. 52, 54, 62, 264n2, 265n9 collaboration see adversarial collaboration collective deliberation 200, 257, 262, 265n7, 265n14 collective governance 146 collective virtues 161 common knowledge effect 40–41 competence requirement 35–40 complex evaluative spaces (CES) 142, 143 Comprehensive Assessment of Rational Thinking (CART) 181, 186n5 Condorcet, N. 7, 32, 35–46 Condorcet’s Jury Theorem (CJT) 32– 48, 149n6; competence requirement 35–40; independence requirement 40–44; sincerity requirement 44–46 conspiracy beliefs 179, 183, 186n5 constitutive epistemic suboptimality 143–144 Converse, P. 36–38 Conway, L. G. 178 Cooper, A. J. 243 core epistemic functions 144–146, 148 correspondence theories 96, 108n23

318  Index corroboration 278; bootstrapped 42–43 corrupting/corruption 153; and character 154–156; conditions 156–160; and cynicism 163–167 corruptors 155–156; kinds of 158 Costello, T. H. 186n3 COVID/Covid-19 93, 104, 179, 183–184, 234, 294–295, 305 Cramer, K. 117 critical thinking 70–85; analytic epistemology 76–79; culture of ignorance 72–74; and epistemic authorities 79–82; epistemic circumstances 72–74; and intellectual autonomy 74–76; postmodern epistemology 76–79; social dimension of cognition 76–79; trusting experts as 82–83 criticism of deliberation 149n7 Croatian youth 45 culture of ignorance 72–74, 85n2 cultures of secrecy 158, 160 Cummings, D. 109n35 cumulative culture 138, 149n3 Currie, G. 218 The Cynical Society (Goldfarb) 152 cynicism 152; agential 162; anthropological 162–163; and corruption 163–167; curative property of 153; institutional 153, 154, 160–163; kinds of 162; politics of 161 D’Agostino, F. 142 Dahl, R. 93, 94, 107n14 decision-making 22–23; agent as heuristic 19; collective 22, 32, 35, 253; deliberative and 148; democratic 18, 46, 146; inclusive 17; political 48; procedures 251, 258, 266n16, 266n18; process 250, 251, 253, 262 deep cynicism 163; see also cynicism De La Torre, L. 62 deliberative democracy 41, 107n11 Delli Carpini, M. X. 34, 37, 39, 99 demand responsibility 99 democracy: defined 136; disinformation and 91–94; empathy and 222–225; epistemology of 1–6; nursery 209n10 Democracy Report 2022 90 democratic backsliding 51, 90, 106

democratic culture 153, 222, 225 democratic deliberation 9, 51, 52, 60–61, 65, 67, 257, 266n14 Democratic Faith (Deneen) 152 Democrats 28, 92, 102, 106, 180, 280, 296–297 Deneen, P.: Democratic Faith 152 Dennett, D. C. 175 de Ridder, J. 52; Handbook of Political Epistemology 5 Descartes, R. 76 destructive cynicism 164 Dewey, J. 74, 207 Diogenes of Sinope 152, 165 disagreements 103–106; skepticism from 234–237 discovery reliability 140, 141, 143 disinformation 90–107; defined 97–99; democracy and 91–94; disagreements 103–106; undemocratic 94–97; voter influence 99–103 Ditto, P. H. 172, 173, 177 diversity theorem 17–30; assumptions of proof 18–25; behavioral assumptions 25–30 Diversity Trumps Ability (DTA) 144 Diversity Trumps Ability Theorem 21, 30n12, 149n8 dogmatism 58–59, 67, 75, 96, 158, 164 doing good 159 doing well 159 doubling down 180–184 Du Bois, W. E. B 222–223, 243 Dunning-Kruger effect 240 early childhood education 207–208, 209n9 Eatwell, R. 130n15 Ebbw Vale 120, 121 echo chambers 33, 34–35, 41–44, 47, 85n2, 209n2, 283 education: early childhood 207–208, 209n9; increasing knowledge through 297–299 egalitarian approach 252, 260–262, 264n5 egalitarian democracy 107n11 egalitarian epistemic position 260–264 egalitarianism 77, 80, 251, 260, 264n6 electoral autocracy 94 electoral democracy 93–94, 97, 99, 106

Index 319 electoral ignorance 12, 36, 287–290, 297–299, 302, 304 elite polarization 275, 276 emotional empathy 10, 216–218, 220–221, 223, 229 emotionally charged perspectivetaking 218, 221–222, 230n14 empathy: cognitive 216, 221–222; democracy and 222–225; empathy 10, 216–218, 220–221, 223, 229; political 220; sensemaking and 216, 217–221 empirically strange bias 171–173 empty signifiers 62 English, A. 207 enlightened preference 292 enlightened understanding 93–94, 106 Environmental Protection Agency 120 epistemically corrupt 242 epistemic arrogance 203, 205 epistemic authorities 7, 8, 12, 73–77, 79–82, 83–84 epistemic autonomy 75–77, 82–84 epistemic bubbles 33–34, 41–44, 47, 85n2 epistemic capture 146–148 epistemic circumstances 72–74 epistemic colonization 242–243; argument from 241–247 epistemic community 195–208; civility approaches to public discourse 201; education and skill development 206–208; epistemic agency 199–201; epistemic exclusion 196–198; listening 199–201, 206–208; listening obligations 204–205; obligations to listen 204; polarization 196–198; stupidification 196–198; virtue of reciprocity 204–205 epistemic corruption 242–247 epistemic crisis 176–177, 195, 199, 202, 204–205, 208 epistemic democracy: and informal political sphere 251–253 epistemic estrangement 198 epistemic friction 164, 166 Epistemic Injustice: Power and the Ethics of Knowing (Fricker) 4 epistemic insouciance 58, 159 epistemic powers of democracy 136, 141, 143–146 epistemic reliability 136, 137–143 epistemology see specific types

epistocracy 291–293 epistocrats 135 establishment 55, 62, 117–119, 299 Estlund, D. 258, 260, 266n18 etiology 154 European Union (EU) 32, 120–122, 147 EU vs. Disinfo 102 Everybody Knows (Chaloupka) 165–166 expert acceptance scales 181 expert rule 5, 6, 12, 36, 85n3, 148 explanation and understanding 225 extremification 277, 278, 280 extremity shifts 278 Facebook 90, 100, 102, 109n37, 256, 263 fact-checkers 179, 183 fact-checking 183, 184 fact polarisation 104–105 factual beliefs 104 “faith in science” scales 180 Fakas, J. 93, 94; Post-Truth, Fake News and Democracy 92 fake news 295, 296 fallibilism 108n22 Fallis, D. 97, 98 false narrative 98, 108n31 Fanon, F. 242, 243 floors as procedural elements 261 Foot, P. 231n27 foot voting 288, 302–305 formalist approaches 252 framing biases 171 Freire, P. 207 Fricker, M. 141; Epistemic Injustice: Power and the Ethics of Knowing 4 Friedman, J. 148n1 The Function of Cynicism at the Present Time (Small) 153 funding political campaigns 254–256 Galston, W. 63 Gaus, G. 142 General Psychopathology (Jaspers) 225–226 Gigone, D. 41 Gilbert, M. 201 Gilens, M. 147, 255 glorious victories 57 Gluck, J. 56, 65 GMO foods 22 Goldfarb, J.: The Cynical Society 152

320  Index Goldman, A. I. 4 Goodin, R. E. 40 Goodwin, M. J. 130n15 Gordon, L. 242, 243 Gove, M. 61 government strategy 127 Great Paradox 215–217, 218, 226–227, 229 group polarization 284n7 Gunn, H. K. 209n2 Gunther, R. 109n42 Haider-Markel, D. P. 172 Haidt, J. 175 halo effect 63 Handbook of Political Epistemology (de Ridder and Hannon) 5 Hannon, M. 52, 224, 230n12; Handbook of Political Epistemology 5 Hastie, R. 41 Haus, D. E. 209n10 Hayek, F. 238–239, 244, 294 Hayekian individualism 245 Heldke, L. 197 Herbart, J. F. 207 heuristics 56–59 hindsight bias 171 Hochschild, A. R. 120–124, 126, 128, 216–217, 221, 230n16; Strangers in Their Own Land 215 homogenization 277–278, 280 Hong, L. 17–25, 30n6–7 Hong-Page theorem 18–19, 22–24, 29, 47, 140 hooks, bell 207 Howell, G. 264n4 How Propaganda Works (Stanley) 222 ideological belief system 37 ideological distance 275 ideological monoculture 185–186; scientific problem of 180–184 illiberal democracy 94, 120 illocutionary silencing 208n1 increasing disagreements 103 independence requirement: Condorcet’s Jury Theorem (CJT) 40–44; hypothetical affairs 34; obstacle to 35 independent dependence 84 individually epistemically vicious 136, 138, 140

individual understanding 216 Individual Virtue Approach (IVA) 135–148; epistemic capture and 146–148; epistemic powers of democracy 143–146; epistemic reliability 137–143; institutional epistemology 137–143; representative electoral democracy (RED) 143–146; theorists 149n2 informal political sphere and epistemic democracy 251–253 Information Disorder: The Essential Glossary (Wardle) 97 information technology 90 informed citizens 1, 272 inner censor 207 institutional bias 242–243 institutional cynicism 153, 154, 160–163, 165 institutional epistemology (IE) 135, 137–143 institutional ethoi 159, 160 intellectual autonomy 74–76 intellectual humility 75, 196, 203 intensification modes of corruption 157 Internet Research Agency 100 inter-partisan marriage 284n4 issue a mandate 99 Ivaldi, G. 130n2 Jaspers, K. 228–229; General Psychopathology 225–226 Jaworski, P. 157 Jones, P. E. 172 Joslyn, M. R. 172 Jury Theorem 144 Kahan, D. M. 28, 104, 105, 130n12, 180 Kahneman, D. 56, 57 Kant, I. 76 Kanter, D. L. 164 Kavanagh, J. 103, 104, 109n46 Keane, J. 95, 96, 108n23 Keeter, S. 34, 37, 39, 99 Kelly, T. 241 Kidd, I. J. 3, 203, 248n13 Klein, E. 274 Know-It-All Society (Lynch) 204 knowledge resistant 105 Krause, S. 230n14, 230n19 Kuklinsky, J. H. 37, 39

Index 321 Landemore, H. 18, 104 learning problems 139–143, 149n9 Leave campaign 109n35 leave supporters 117, 122, 130n11 Leave voters 121, 124, 125 left-leaning individuals 123 Le Pen, M. 55 Levy, N. 138 liberal democracy 93 liberal epistemic position 258 liberalism 237–239 Lin, D. Y. 176 listening 199–201; active 10, 221; competencies 10, 195; failures 196; obligations 204–205 Locke, J. 76 Lodge, M. 38 looking good 159 Lynch, M. P. 203; Know-It-All Society 204 Macron, E. 91, 92, 94, 107n10, 296 Maduro, N. 65 Maldonado-Torres, N. 242 manipulation 66, 98, 100, 158, 302 Marlin, R. 98–99 Mason, L. 115 McBrayer, J. P. 209n4 McGrath, S. 241 media bias 179 Medina, J. 154, 161, 164 memetics 174–175 Mercer, R. 109n37 Mercier, H. 58 meta-narratives 102 Meyerson, D. 230n23 Mignolo, W. 242 Mill, J. S. 36, 52, 141, 265n14, 291 miracle of aggregation 290 Mirvis, P. H. 164 miserliness: cognitive 56–67; social epistemic 67–68 miserly processing 58 misinformation: controlling the flow of 295–297; see also information misleading information 97, 98 models of democracy 144; see also democracy Montaigne, M. de. 238, 244 moral damage 154, 155, 156, 157, 163–166 morally inferior 161

motivated reasoning 28–29, 105, 110n52, 122, 130n12, 295 MS-13 27–28 Mueller, R. 100 Muslim: discrimination against 63; refugees/migrants 54, 57, 61, 63; ruler 57 myside bias 56, 59–60, 170–186; academia 180–184; adversarial collaboration 185–186; doubling down 180–184; empirically strange bias 171–173; epistemic crisis in universities 176–177; ideological monoculture 185–186; memetics 174–175; normative complications 173–174; properties 176–177; psychology’s self-correction problem 177–180; restoring epistemic legitimacy to social sciences 184–185; scientific problem of ideological monoculture 180– 184; theoretical alternative 174–175 Myside bias (Stanovich) 59 Nelson 44 neologism 94 New Diversity Theory 136, 142 New York Times 106, 178, 183 Nguyen, C. 34, 41 Nietzsche, F. 236–237 Nietzschean approach 237 nonaccidental feature of disinformation 97; see also disinformation non-Nordic 209n10 normative complications 173–174 normative diversity 141, 149n4 Northern League 116, 117 Nussbaum, M. 230n13 objectivism 108n27 objectivity of values 108n27 Ocasio-Cortez, A. 161 Orban, V. 51, 68, 94 Orwell, G. 234 ostracization 197 Ostrom, E. 138 our-side bias 59–60, 62, 68 outcome bias 171 outlier bias 173, 176 outlook 95, 115–119, 121, 123, 127, 129, 153, 161, 164, 220, 221 overconfidence bias 171 Oxford English Dictionary 91

322  Index Page, B. I. 147, 255 Page, S. 17–25, 30n6–7, 30n12 Palifka, B. 157 participatory democracy 107n11 Payne, B. K. 240 Pennycook, G. 105 personal understanding 216 Peyton, B. 37, 39 philosophy: analytic 5, 52; applied 53; critical thinking 74; engaged 68; political 48n1, 52; professors 71 Plato 5, 52, 76 political belief system 37 political campaigns: funding of 254–256 political disagreements 104 political enemies 270–284; democrat’s dilemma 271–274; managing belief polarization 281–284; polarization dynamic 270, 274–281 political epistemology 5, 52–53 political equality 93 political ignorance 287–305; problem of 288–290; top-down approaches 290–297 political inequalities 257–260 political influence 55, 92, 250–263, 264n5, 265n14, 266n17 politically motivated reasoning 28, 105 political polarization 7, 10, 44, 172, 224, 254, 275 political resentment 115–130; cruel hopes 125–128; good hopes 128–130; magic hopes 125–128; paradox of 120–122; polarization 116–120; queue jumping 123–125; wilful hopes 125–128 political skepticism 233–234, 247n6 politics 225–229; of cynicism 161; of hope 161 popular government 92, 93, 94 popular sovereignty 26, 29, 54, 62 populism 51–68, 115, 120, 129, 130n2, 131n18; cognitive miserliness 56–67; populist threat to democracy 54–56; social epistemic miserliness 67–68 postmodern epistemology 76–79 Post-Truth, Fake News and Democracy (Farkas and Schou) 92 preemptive evidencing 43

Problem-Solving a Tournament Thesis (PT) 137 procedural reform 261–263 Pronin, E. 176, 240 propagation modes of corruption 157; see also corrupting/corruption psychology’s self-correction problem 177–180 Public Discourse in America (Steinberg) 202 public opinion: influencing 253–254 racism 118, 127, 177–178, 243; symbolic 185, 186; systemic 147 radicalization 52, 61, 67, 68 Rand, D. G. 105 Rapacioli, P. 109n43 “rapefugees” 63 rational irrationality 289, 293 rationally ignorant/ce 63, 288–289, 290, 301–302 Rawls, J. 52, 251, 259, 263, 266n15, 266n21 reasonable criticism 282 reconquest of Spain 57 reform see procedural reform refugees, Muslim 54, 57, 61, 63 remain supporters 130n11 representative democracy 32, 48n2; see also democracy representative electoral democracy (RED) 136, 143–146, 147 Republicans 280 restoring epistemic legitimacy to social sciences 184–185 Rice, S. 206 Rich, F. 230n16 Rich, M. D. 103, 104, 109n46 Rini, R. 241 Rittel, H. 139 Rodin, J. 202 Rorty, R. 247, 247n1 Rose-Ackerman, S. 157 Rosenfeld, S. 108n18 Ross, L. 176 rule of experts 293–295 Russia, role in 2016 American election 100 Samaržija, S. P. 58 Sartre, J.-P. 163 Scapp, R. 207 Schaff, M. 215–216, 229 Schmidt, K. 240

Index 323 Schou, J. 93, 94; Post-Truth, Fake News and Democracy 92 Schwartz, S. H. 131n21 science skepticism 183 science trust 180 Scruton, R. 161 self-correction problem 177–180 sensemaking 215; empathy and 217–221 sexism 118, 127, 243 sexual harassment 243 Shaw, T. 236–237 Simas, E. N. 172 Simple Governance Thesis (SG) 137 simple reliability 138, 140, 141, 146 sincerity requirement 44–46 Singal, J. 186n6 skepticism: from disagreement 234– 237; political 233–234; science 183 Small, H. 166; The Function of Cynicism at the Present Time 153 social dimension 75; of cognition 76–79 social dominance 182–183 social-epistemic agency 200 social epistemic miserliness 67–68 social epistemology 4, 70 social identity theory 116, 124, 125 social structure reform 263–264 Somin, I. 68 Soros, G. 51 sortition 301–302 Sosa, E. 3 Sperber, D. 58 Spiekermann, K. 40 spillover mechanisms 253–257 Sputnik 103 stabilization modes of corruption 157; see also corrupting/corruption Stanley, J. 230n19; How Propaganda Works 222 Stanley, S. 152 Stanovich, K. 56, 58, 59 Steinberg, S. P.: Public Discourse in America 202 stereotyping 59–60, 68, 128 Strangers in Their Own Land (Hochschild) 215 strategies in learning governance 149n5 structural constraints 256–257 stupidification 195, 197 subsidiarity 147, 148 substantivist approaches 252

Sullivan, E. 160 Sunstein, C. R. 44 Swedish Democrats 102 Swedish Populist Party 102 Swire-Thompson 44 symbolic racism 185, 186; see also racism systemic bias 241–242 systemic racism 147; see also racism Taber, C. S. 38 Tajfel, H. 116 Talisse, R. B. 68, 205, 209n5; Sustaining Democracy: What We Owe to the Other Side 204 tame problems 137–138 Tanesini, A. 3, 203 Tappin, B. M. 105, 110n54 targeted political messaging 109n37 Tea Party 116, 120–125 Tesich, S. 107n5 theoretical alternative 174–175 threat of populism 54–56; see also populism Toplak, M. 179 Trump, D. 28, 32, 51, 54, 91, 92, 101, 107n37, 119, 135, 215–216, 218, 224, 287, 297 Trump administration 294 “trust in experts” measures 181 trusting experts 82–83 truth 85n1; and certainty 95; decay 103; as an objective property 96 truth-sensitivity 85n1 Tufekci, Z. 184 Turner, J. C. 116 Tversky, A. 57 undemocratic disinformation 94–97; see also disinformation understanding and explanation 225 unresponsive government 48n1 US Centers for Disease Control (CDC) 294 US Federal Reserve Board 294 values: beliefs and 38; customs and 120; epistemic and 3, 72, 76, 85n1; ethical and moral 103; liberal 129; progressive and conservative 46; political 6, 36, 48n2; voters 26 Varieties of Democracy (V-Dem) 107n1, 107n15 Vice, S. 161, 164, 165, 166

324  Index vice epistemology 3, 4, 53 Vices of the Mind: From the Intellectual to the Political (Cassam) 3 virtue epistemology 3, 4, 53, 70, 85n1 Virtues of the Mind (Zagzebski) 3 voter irrationality 289, 297, 302 Vuković, V. 45 Wagnsson, C. 103 Waks, L. J. 207 Wardle, C. 108n28; Information Disorder: The Essential Glossary 97 Washington Post 91 Webber, M. 139

Webster 44 West, R. 171 Whitcomb, D. 203 wicked problems 139, 142, 143 widespread polarization 275, 276 Wittgenstein 42, 243 Wolff, J. 52–53, 68 xenophobia 118, 127 Zagzebski, L. 80; Virtues of the Mind 3 Zaller, J. R. 38 Zemmour, E. 57 Zubčić, M.-L. 150n10