Handbook of Digital Politics: Second Edition (Elgar Handbooks in Political Science) [2 ed.] 1800377576, 9781800377578

Citation preview

HANDBOOK OF DIGITAL POLITICS

ELGAR HANDBOOKS IN POLITICAL SCIENCE Elgar Handbooks in Political Science provide an overview of recent research in all areas relating to the study of political science including comparative politics, international relations, political economy, political theory and research methods, ensuring a comprehensive and overarching guide to the field. The constituent volumes, edited by leading international scholars within the field, are high quality works of lasting significance, often interdisciplinary in approach. The Handbooks discuss both established and new research areas, expanding current debates within the field, as well as signposting how research may advance in the future. The series will form an essential reference point for all academics, researchers and students of political science. For a full list of Edward Elgar published titles, including the titles in this series, visit our website at www​.e​-elgar​.com

Handbook of Digital Politics SECOND EDITION

Edited by

Stephen Coleman Professor of Political Communication, University of Leeds, UK

Lone Sorensen Lecturer in Media and Communication, University of Leeds, UK

ELGAR HANDBOOKS IN POLITICAL SCIENCE

Cheltenham, UK • Northampton, MA, USA

© Stephen Coleman and Lone Sorensen 2023

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical or photocopying, recording, or otherwise without the prior permission of the publisher. Published by Edward Elgar Publishing Limited The Lypiatts 15 Lansdown Road Cheltenham Glos GL50 2JA UK Edward Elgar Publishing, Inc. William Pratt House 9 Dewey Court Northampton Massachusetts 01060 USA A catalogue record for this book is available from the British Library Library of Congress Control Number: 2023943155

This book is available electronically in the Political Science and Public Policy subject collection http://dx.doi.org/10.4337/9781800377585

ISBN 978 1 80037 757 8 (cased) ISBN 978 1 80037 758 5 (eBook)

EEP BoX

Contents

List of figuresviii List of tablesix List of contributorsx Introduction to the Handbook of Digital Politics (Second Edition)xiii Stephen Coleman and Lone Sorensen PART I

WAYS OF SEEING, LISTENING TO AND WRITING ABOUT DIGITAL POLITICS

1

A rhetoric of digital politics Stephen Coleman and Lone Sorensen

2

De-Westernizing digital politics: a Global South viewpoint Bruce Mutsvairo, Fabíola Ortiz dos Santos and Tenford Chitanana

3

Visual digital politics: imag(in)ing political activities and identities online Katy Parry

30

4

Revolution vs reaction: the role of social media in authoritarian regimes Anna Litvinenko

45

5

Transnational and global flows of political discussion online Yuan Zeng

PART II

2 16

59

CITIZENSHIP AND POLITICAL TALK

6

The Internet as a civic space Peter Dahlgren

76

7

Political filter bubbles and fragmented publics Cristian Vaccari and Augusto Valeriani

92

8

Computational approaches to online political expression: a framework for research Mengyu Li, Luhang Sun, Yiming Wang, Yibing Sun, Hyerin Kwon, Jiyoun Suk, JungHwan Yang and Dhavan V. Shah

9

Creating spaces for online deliberation Christopher Birchall and Stephen Coleman v

110

137

vi  Handbook of digital politics

10

New frontiers in two-screen politics Nick Anstead and Ben O’Loughlin

155

11

Gen Z’s civic engagement: news use, politics, and cultural engagement168 Ava Francesca Battocchio, Chris Wells, Emily Vraga, Kjerstin Thorson, Stephanie Edgerly and Leticia Bode

12

Gen Z’s civic engagement: civic skills, political expression, and identity Ava Francesca Battocchio, Leticia Bode, Chris Wells, Emily Vraga, Kjerstin Thorson and Stephanie Edgerly

181

PART III TECHNOLOGY AND PLATFORMS 13

Becoming eventful through data: the mediated construction of historic events in the age of data Heather Ford

14

Algorithms, power and digital politics Ulrike Klinger

15

Social media digital architectures: a platform-first approach to political communication and participation Michael Bossetta

16

Artificial intelligence in politics Leah Henrickson

242

17

Online content moderation during conflict Giovanni De Gregorio and Nicole Stremlau

259

196 210

226

PART IV CONTENTIOUS POLITICS, CIVIL AND NETWORKED SOCIETY 18

The Fifth Estate: a new source of democratic accountability William H. Dutton and Elizabeth Dubois

272

19

The logic of connective action: digital media and the personalization of contentious politics W. Lance Bennett and Alexandra Segerberg

20

Media ecologies, social movements and activism Emiliano Treré

313

21

E-petitioning and changing state–citizen engagement Scott Wright and Ariadne Vromen

327

287

Contents  vii

22

From Valencia filters to #BlackOutTuesday: collective action on Instagram Elena Sotelo-Prol

339

23

Post-Soviet digital democratization experiments: the promise and reality Yuri Misnikov

354

PART V

POLITICAL PARTIES, LEADERS AND GOVERNANCE

24

The digital performance of populism Thomas Wellings and Lone Sorensen

370

25

Political communication about data Brendan Lawson

388

26

Regulation of election communication Damian Tambini

401

PART VI JOURNALISM, APPARENT JOURNALISM AND MEDIA INSTITUTIONS 27

Social media as resources for journalistic struggle in politically restrictive settings Banafsheh Ranji

28

Fake news and digital politics Bente Kalsnes

432

29

Right-wing alternative news media and digital politics Kristoffer Holt

444

30

Research on the political implications of political entertainment Michael A. Xenos

457

418

Index472

Figures

7.1

Perceived exposure to political agreement and disagreement via the mass media, social media, mobile instant messaging apps, and offline conversations

102

Total volume of online posts for the Ford-Kavanaugh case on Facebook, Instagram, Reddit, Twitter, and YouTube from July 2018 to January 2019

118

Total volume of online posts for the Kim-Ahn case on Facebook, Twitter, YouTube, and two forums Nate Pann and DC Inside from March 2018 to September 2018

118

8.3

Total keyword frequency for the Ford-Kavanaugh case on Facebook, Instagram, Reddit, Twitter, and YouTube

120

8.4

Total keyword frequency for the Kim-Ahn case on Facebook, Twitter, YouTube, Nate Pann and DC Inside

120

19.1

Elements of collective and connective action networks

302

26.1

A tipping point? The rise of the Internet as a source of news

405

8.1

8.2

viii

Tables

8.1

Cross-correlation estimates for online posts related to the Ford-Kavanaugh case on Facebook, Instagram, Reddit, Twitter, and YouTube from July 2018 to January 2019

119

Cross-correlation estimates for online posts related to the Kim-Ahn case on Facebook, Twitter, forums, and YouTube from March 2018 to September 2018

119

8.3

High-frequency keywords for the Ford-Kavanaugh case on Facebook, Instagram, Reddit, Twitter, and YouTube

121

8.4

High-frequency keywords for the Kim-Ahn case on Facebook, Twitter, YouTube, and mainstream forums

123

8.5

Structural topic modeling of prominent topics in the Ford-Kavanaugh case

126

8.6

Structural topic modeling of prominent topics in the Kim-Ahn case

127

18.1

Examples of potentially effective strategies of the Fifth Estate

276

18.2

Networked institutions and networked individuals

279

8.2

ix

Contributors

Nick Anstead, Associate Professor, London School of Economics and Political Science Ava Francesca Battocchio, Doctoral Student, Michigan State University W. Lance Bennett, Ruddick C. Lawrence Professor Communication and Professor of Political Science, University of Washington Christopher Birchall, Lecturer in Digital Media, University of Leeds Leticia Bode, Associate Professor, Georgetown University Michael Bossetta, Associate Senior Lecturer in Media and Communication Studies, Lund University Tenford Chitanana, University of Technology Sydney Stephen Coleman, Professor of Political Communication, University of Leeds Peter Dahlgren, Professor Emeritus, Lund University Elizabeth Dubois, Associate Professor, University of Ottawa William H. Dutton, Emeritus Prof, University of Southern California and Martin Fellow, Oxford Martin School, Oxford University Stephanie Edgerly, Professor, Northwestern University Heather Ford, Associate Professor, University of Technology Sydney Giovanni De Gregorio, Research Associate, University of Oxford Leah Henrickson, Lecturer in Digital Media and Cultures, University of Queensland Kristoffer Holt, Professor in Media and Communication, Linnæus University Bente Kalsnes, Professor of Political Communication, Kristiania University College Ulrike Klinger, Professor of Digital Democracy, European New School of Digital Studies Hyerin Kwon, Doctoral Student, University of Wisconsin-Madison Brendan Lawson, University Teacher, Loughborough University Mengyu Li, Doctoral Student, University of Wisconsin-Madison x

Contributors  xi

Anna Litvinenko, Researcher, Freie Universität Berlin Yuri Misnikov, Associate Professor, Kaunas University of Technology Bruce Mutsvairo, Professor, Utrecht University Ben O’Loughlin, Professor of International Relations, Royal Holloway University of London Fabíola Ortiz dos Santos, University of Duisburg-Essen Katy Parry, Associate Professor in Media and Communication, University of Leeds Banafsheh Ranji, Postdoctoral Fellow, Norwegian University of Science and Technology Alexandra Segerberg, Associate Professor, Uppsala University Dhavan V. Shah, Maier-Bascom Professor, University of Wisconsin-Madison Lone Sorensen, Lecturer in Media and Communication, University of Leeds Elena Sotelo-Prol, Doctoral Student, University of Leeds Nicole Stremlau, Senior Research Fellow, University of Oxford, and Research Professor in the Humanities, University of Johannesburg Jiyoun Suk, Assistant Professor, University of Connecticut Luhang Sun, Doctoral Student, University of Wisconsin-Madison Yibing Sun, Doctoral Student, University of Wisconsin-Madison Damian Tambini, Distinguished Policy Fellow, London School of Economics and Political Science Kjerstin Thorson, Brandt Endowed Professor of Political Communication, Michigan State University Emiliano Treré, Reader in Data Agency and Media Ecologies, Cardiff University Cristian Vaccari, Chair in Future Governance, Public Policy and Technology, University of Edinburgh Augusto Valeriani, Associate Professor, University of Bologna Emily Vraga, Associate Professor, University of Minnesota Ariadne Vromen, Sir John Bunting Chair of Public Administration, Australian National University Yiming Wang, Doctoral Student, University of Wisconsin-Madison

xii  Handbook of digital politics

Thomas Wellings, Research Fellow, University of Leeds Chris Wells, Associate Professor in Emerging Media Studies, Boston University Scott Wright, Professor of Political Communication and Journalism, Bournemouth University Michael A. Xenos, Professor of Life Sciences Communication, University of Wisconsin-Madison JungHwan Yang, Assistant Professor, University of Illinois at Urbana-Champaign Yuan Zeng, Lecturer in Media and Communication, University of Leeds

Introduction to the Handbook of Digital Politics (Second Edition) Stephen Coleman and Lone Sorensen We are delighted that the positive reception of the first edition of the Handbook of Digital Politics has prompted the publication of this second edition. While ‘digital politics’ is no longer a new phenomenon, and its techniques, language and logic are now an integral part of contemporary political practice, there is still a widespread sense amongst political practitioners that the old ways of ‘doing politics’ are somehow under pressure from the relentless expansion of digital technologies, channels and styles. Whether these developments are characterized as enhancing, corrupting, broadening, narrowing or simply complicating is a matter of perspective. This book is neither a celebration of digital politics nor a cry of despair. It is instead an invitation to take stock by catching up with the latest knowledge and perspectives from some of the most impressive researchers in the field. We have intentionally selected contributors whom we knew would provide original, incisive and provocative thoughts. Their contributions are organized into six thematic sections. Part I explores ways of seeing, listening to and writing about digital politics. Its five chapters set out ways of thinking about what digital politics means. Stephen Coleman and Lone Sorensen begin by interrogating the rhetorical binaries of ‘cyber-optimism’ and ‘cyber-pessimism’ that have over-simplified our understandings of political change. Bruce Mutsvairo, Fabíola Ortiz dos Santos and Tenford Chitanana provide a vitally important corrective to the Anglo-American perspectives from which digital politics has tended to be described and evaluated. They remind us that there is a world beyond the West that is deeply affected by digital flows of information and communication. Katy Parry also broadens our horizons, taking us beyond the traditional territory of words into a visual landscape to which political communication is increasingly forced to adapt. Anna Litvinenko considers the various ways in which authoritarian regimes have come under pressure from digital publics, while often seeking to exploit and regulate digital technologies for their own ends. Yuan Zeng’s chapter focuses upon the transnational flows of political discussion that would have been hard to even imagine three decades ago. Each of these chapters attempts to answer framing questions about what it means to say that we are now living in a different kind of political world from the pre-digital era. The chapters in Part II focus upon citizenship and political talk. These chapters are connected by an interest in how civic practices have changed in the digital era and, particularly, how people are using their digital voices to engage with and disrupt the monological drone of political elites. Peter Dahlgren opens the section with a powerful argument for the Internet as a civic space. This is followed by a chapter by Cristian Vaccari and Augusto Valeriani which sets out valuable empirical findings about how citizens both encounter and avoid new ideas. Focusing upon experiential xiii

xiv  Handbook of digital politics

variables, they show how specific types of digital experience generate different participatory effects. A key determinant of digital experience is examined in a chapter by Mengyu Li, Luhang Sun, Yiming Wang, Yibing Sun, Hyerin Kwon, Jiyoun Suk, JungHwan Yang and Dhavan V. Shah. Focusing upon the #MeToo movement, the chapter takes a computational approach to understanding how certain voices come to be heard or ignored. In their chapter, Christopher Birchall and Stephen Coleman discuss innovations in platforms and spaces designed to encourage public deliberation and ask how design decisions affect democratic opportunities. Nick Anstead and Ben O’Loughlin’s chapter explores the new ways in which citizens consume televisual content through more than one screen, often asynchronously, and the ways in which some populist politicians have been exploiting this trend. The section concludes with two chapters by Ava Francesca Battocchio, Chris Wells, Emily Vraga, Kjerstin Thorson, Stephanie Edgerly and Leticia Bode exploring the ways in which Generation Z are accessing and engaging with political news and events, and exploring the new civic skills, expressive forms and identities that this digitally native generation are bringing to the political sphere. Part III focuses on technologies and platforms, pointing to the important ways in which they open up and constrain political affordances. Heather Ford’s opening chapter provides a vivid account of how the eventfulness and meaningfulness of the world are shaped by often unnoticed mediating processes. This is followed by Ulrike Klinger’s brilliant account of how algorithmic power operates upon digital political reality and the largely hidden asymmetries that materialize through it. Michael Bossetta’s insightful account of the politics of digital architecture offers an alternative or complementary approach to the affordances perspective on digital materiality. Using the example of political participation, he demonstrates the importance of paying attention to platform design features. Leah Henrickson’s penetrating chapter on the political implications of artificial intelligence gives an accessible introduction to both the practical uses of AI in politics and its potential application in research on politics. The section concludes with a chapter by Giovanni De Gregorio and Nicole Stremlau demonstrating how moderation practices and technologies intervene in the flow of public discussion and hate speech during the heat of war and conflict. In Part IV contributors turn to contentious politics, civil and networked society, showing how old political forms like media as ‘the fourth estate’ or ‘collective action’ are being digitally reconfigured. William H. Dutton and Elizabeth Dubois consider how the Internet has enabled new forms of social and political accountability which constitutes a ‘fifth estate’. W. Lance Bennett and Alexandra Segerberg discuss the transition from traditional collective action to what they call ‘connective action’. Emiliano Treré applies media ecology theory to the study of social movements and activism. He argues in his chapter that a holistic view of media use and practice enables us to disentangle the complexities of contentious politics in different ecologies across time and space. Looking at specific cases of digital activism, Scott Wright and Ariadne Vromen look at e-petitioning. Elena Sotelo-Prol explores civic activism on Instagram and discusses the ways in which the platform’s affordances

Introduction  xv

of visuality, visibility, shareability and performativity shape online collective action. Yuri Misnikov analyses digital developments and constraints in post-Soviet states. Turning to what used to be thought of as the institutional core of politics, the chapters in Part V examine digital changes in political parties, leaders and governance. Thomas Wellings and Lone Sorensen consider the ways in which populism has flourished amidst digital communication practices and landscapes. Brendan Lawson offers an account of how political communication has become increasingly data-driven and focused. He argues that we should pay attention to how politicians talk about data and the ways in which they use them rhetorically. Damian Tambini addresses the complex policy challenge of regulating digital information flows. In Part VI we look at journalism, apparent journalism and media institutions. It becomes apparent here that what used to be thought of as journalistic practices, institutions and standards have not easily made the transition to digital information spaces. Banafsheh Ranji considers how online journalism is faring in highly restrictive political spaces. She argues that we need to look beyond the restrictions of political systems as the single factor in journalists’ recourse to digital platforms and also consider the economic and organizational dynamics of news makers. Bente Kalsnes takes on the huge question of fake news and the politics of deception with a particular focus on its role in elections. Kristoffer Holt explores the murky area of right-wing ‘alternative media’ with a focus on their cultural dimension and with reference to the Canadian trucker movement in January 2022. And Michael A. Xenos looks at how many citizens now access news about the world through ‘infotainment’. Between them, our contributors vigorously shake the can of digital politics and bring its diverse elements to an accessible surface. We hope that you will enjoy and learn from them and the further readings and works cited at the end of each chapter.

PART I WAYS OF SEEING, LISTENING TO AND WRITING ABOUT DIGITAL POLITICS

1. A rhetoric of digital politics Stephen Coleman and Lone Sorensen

STORY TIME Unlike other chapters in this book, this one is less about the entanglements between digital technologies and democratic politics than how scholars have told the story of those entanglements. We are more interested in interrogating the rhetorical construction of the ‘digital politics’ narrative than pronouncing upon the empirical veracity of competing narratives. Our starting point is to say that all writing (and speaking) about social phenomena entails strategies of persuasiveness. To put it simply, the researcher has to try to demonstrate that they are speaking from a position of acute perspective; that their account possesses a credibility that is rooted in special forms of theoretical and methodological insight. As Paul Atkinson (2014, p. 16) points out: sociological texts in general are inescapably rhetorical. Whether they adopt an explicitly exhortatory tone, or purport merely to report social ‘fact’, they rely upon devices of persuasion to construct plausible accounts, striking contrasts, historical inevitabilities; to link data into convincing sequences of cause and effect; to embed theory into data and vice versa.

To state that both the best and worst writing about digital politics is rhetorically framed is not to suggest that writers are engaging in expressive ruses designed to trick readers into taking their point of view. The term rhetoric is commonly used these days to describe forms of communicative guile, evasion, spin, vacuity and mendacity. In contrast, we are using the term in its original Aristotelian sense of finding the most persuasive way of expressing a point of view. That is to say, every writer (and communicator in other forms) sets out to offer a plausible account of reality, using the most compelling arrangements of words and arguments to do so. Similarly, to say that writers about digital politics are telling stories is not to suggest that they are making things up, but that they are attempting to devise compelling accounts of the meaning of a social situation or sequence of events. With those clarifications in mind, we might say that this chapter is about how writers have rhetorically constructed stories about the relationship between digital technologies and political practices since the arrival of the Internet as a popular public resource at the end of the twentieth century. In 1996 the first author of this chapter, together with the economist Andrew Graham, organized the first ever conference in the United Kingdom to consider how (or, at that time, whether) the newly-arrived World Wide Web would impact upon politics. The event took place in the British Parliament and was attended by an impressive range of politicians, civic activists, technologists and academics. 2

A rhetoric of digital politics  3

Questions about how far this new communication network, to which fewer than 2 per cent of the UK population had home access at that time, would transform politics evoked wide-ranging speculations. As with most futurology, some expectations were fired up by the heat of technological determinism, leading to rash predictions that the Internet would be a panacea for the cumulative deficits of democratic politics; others, sticking with the incessant repeat cycle of ‘realism’, argued that politics would go on in the same old way despite the arrival of this latest Californian gee-whizzery; and others still sought to suggest that the building we were meeting in would become obsolete as soon as people became used to voting on every issue online, without the need for political representatives. Much of the early writing about digital politics dealt in such dichotomous appraisals: the Internet would either make possible the realization of the original and best principles of direct rule by the demos or it would be wholly subsumed by the iron law of oligarchy and the profit motive. Producing these competing versions entailed rhetorical work. Reading much of the popular and academic literature about digital politics from the turn of the century, one encounters a breathless rhetorical energy in which excitement in the face of fast movement often overcame nuanced thinking. Consider, for example, Nicholas Negroponte’s (1995, p. 231) seminal book, Being Digital, one of the first and most celebrated declarations that ‘the digital future is here’: The access, the mobility and the ability to effect change are what will make the future so different from the present. The information superhighway may be mostly hype today, but it is an understatement about tomorrow. It will exist beyond people’s wildest predictions.

The above words constitute the penultimate paragraph of a chapter entitled ‘An Age of Optimism’ and Negroponte introduces the paragraph by stating that ‘more than anything, my optimism comes from the empowering nature of being digital’. These references to optimism provide an explicit rhetorical steer. They invite the reader to open themselves to a disposition. They say, ‘I can see the great things that will come from this new situation and you will only be able to see them too if you share my unrestrained hope’. The author concedes that present mid-1990s’ talk about the Internet is ‘mostly hype’, but goes on to suggest that change is happening so fast that what is now hyperbole will soon be ‘understatement’. Hype serves as a measure not of our over-heated imaginations but our imaginative incapacity to predict the transformation surrounding us. All that we can be sure about is that that future ‘will exist beyond people’s wildest predictions’. It is not our intention here to disparage Negroponte’s feverishness. In fact, his book contains a number of perceptive insights. Our purpose is to illuminate a form of rhetorical construction that played an important part in shaping evaluation of the relationship between digital technologies and political power. A similar tone of breathless optimism pervades the following 1999 statement from the report of the European Information Society Forum:

4  Handbook of digital politics

The new information technologies may, for the first time in the history of industrial societies under liberal regimes, make it possible to recreate the perfect information arena, the agora of Ancient Greece, a meeting place where citizens could go to be fully informed and to participate directly, with no intermediary, in the government of the city, exercising all their political rights unconditionally and without restriction.

Three rhetorical tropes are at work here. Firstly, as in the passage from Negroponte, there are references to the profound historical significance of the developments outlined. The possibility that is being postulated is said to have emerged ‘for the first time in the history of industrial societies under liberal regimes’. We are informed that ‘new’ technologies are enabling the recreation of a democratic ideal that we had associated with the distant past. This sense of sweeping movement across history in the course of a single paragraph reflects the turbulence of the moment. We are in a whirlwind of tenses. Secondly, readers are urged to think of democracy, at least at the city scale, as exceeding the current mechanisms of representation. Everyone will soon be able to meet together in one (virtual) space. They will be able to become ‘fully informed’ and will be free to participate without any restriction. It is not clear whether the European Information Society Forum is advocating direct democracy or simply noting that it is about to become possible, but the rhetorical work has been done. The implication made is that unless such a new political order emerges, the Internet will have somehow failed in its potential. Thirdly, there is a gesture towards traditional utopian thought in the reference to recreating ‘the perfect information arena’. Contemporary social analysis tends to steer clear of concepts like perfection. What is meant by a ‘perfect information arena’ (one in which every point of view is accessible – and comprehensible to all – and open to deliberative contestation – and incorporated into policies, which themselves are known and understood by all?) is not said, but it does not need to be for the relationship between ‘new information technologies’ and perfection to be rhetorically planted as a seed. In his 1999 book, Vote.Com, Dick Morris, who had been one of the chief political advisers to US President Bill Clinton, goes even further in linking the Internet to a completely new democratic arrangement: Whether direct Internet democracy is good or bad is quite beside the point. It is inevitable. It is coming and we had better make our peace with it. We have to better educate ourselves so that we can make good decisions. Restricting the power of the people is no longer a viable option. The Internet made it obsolete. (Morris, 1999, p. 31)

By now the rhetorical elements should be apparent. The reference to inevitability; the need to educate ourselves and make peace with this imminent future; the obsolescence of the familiar present are all ways of orienting the reader towards the writer’s sense of certainty. Accorded an historical agency in its own right, the Internet emerges in this passage as an historical actor, regardless of our intentions. Much of what was written about digital politics at the turn of the century was driven by an impetus to answer a single question: Will the Internet be good or bad for democracy? (The same question had been asked about television thirty years

A rhetoric of digital politics  5

earlier.) In response to this normatively vague question scholars tended to gravitate towards one of two camps, referred to by Pippa Norris (2001) as ‘cyber-optimism’ and ‘cyber-pessimism’. What emerged from those camps were forms of rhetorical reductionism whereby complex cultural trajectories of a social innovation were eclipsed by the inducements of narrative. Cyber-optimists, as we have seen above, tended to employ a rhetoric of historical progress, inevitability and rupture. Their sceptical opponents were determined to show that social structures, political systems and human traits were more enduring than the optimists believed, but in doing so they tended to rely upon a rhetoric of ‘business as normal’, often failing to acknowledge that history is more than a binary between wholesale transformation and inert stasis. In their eagerness to counter the hyperbole of the optimists, proponents of the ‘normalization’ thesis too often failed to acknowledge the innovative affordances of digital technologies. The cyber-optimist-pessimist binary, which still persists within much of the digital politics literature, is founded upon the problematic assumption that the Internet somehow acts upon political behaviour. Drawing upon the language of media effects, this approach misses the reality that political technologies and their consequences are mediated by social practice. The political consequences of going online, be it to seek political information, exchange thoughts with friends or engage in collective protest, is determined by what Schraube (2009, p. 304) calls ‘the reciprocal interwovenness of materiality and sociability’. This relationship between political agency and digital technology is always shaped by social experience and practice. Is one a political citizen or an outsider? Is the behaviour legally permissible or illegal? Is action private or collective? Is political authority accountable or insensitive? Does a repertoire of online political activities already exist or is one engaging in innovatory practice? Are online platforms regulated or laws unto themselves? Is code explicit or hidden? Is data secure or precarious? How easy is it to build strong social networks with like-minded people? These and many other questions of practice override reductively binary questions about whether the Internet ‘changes everything’ or ‘changes nothing’. In contrast to the rhetorical binaries that we have been criticizing, we wish to make the case for a much less exciting mode of analytical expression: ambivalence. We speak of a situation or phenomenon as being ambivalent when it fits into more than one category of description. Unlike new-born babies (good) or poison (bad), ambivalent objects fall into several classes at the same time (Bauman, 1993, pp. 1–2). Ambivalence is the antithesis of what the philosopher Richard Rorty (1989, p. 74) refers to as a ‘final vocabulary’ capable of describing ‘a single permanent reality’. Ambivalence ‘eludes unequivocal allocation’ (Bauman, 1993, p. 9) by refusing definitional boundaries and binaries and acknowledging that objects can be more than one thing at a time. This is hardly a remarkable insight. We are simply stating that digital politics takes many forms that are more likely to be understood by being open to their contradictory elements, polysemic narratives and contextual variations than by succumbing to the illusory elegance of conclusive definition. But if we are to adopt such an analytical perspective this must entail breaking with rhetorical tra-

6  Handbook of digital politics

ditions that seek to evaluate digital politics through the encompassing dispositional lens of ‘optimism’ or ‘pessimism’. Ambivalence calls for a greater degree of analytical balance and nuance. In moving beyond analytical binaries and embracing theoretical ambivalence, new rhetorical options become available. We shall turn to these in the final section of this chapter, but in the next section we attempt to demonstrate at an empirical level how stories about digital politics are rhetorically shaped.

CREATING BINARIES Example 1: Coup d’état and Protest in Myanmar In the context of democratization, cyber optimism and pessimism took the form of the binary and unequivocal labels of ‘liberation technology’ (Diamond, 2010) and ‘net delusion’ (Morozov, 2012). In characterizing digital politics as ambivalent we do not mean to imply that it is too messy and indecipherable to explain. Rather, we can identify distinct crosscurrents and dynamics that shape digital politics in various complex ways. For instance, ambivalence suggests that we pay attention to the intents of different users, groups and networks of users on digital platforms. In the case of Myanmar, a struggle over the strategic goals of control and voice has been playing out in the digital political sphere. But what we have in mind is not a binary reading of this struggle, as it is presented in black and white in Box 1.1 and 1.2. Instead, ambivalence should encourage us to engage with the nuances of the relationship between technology and each side in the conflict as it pertains to power, capabilities and political economy.

BOX 1.1 LIBERATION TECHNOLOGY IN MYANMAR Liberation technology has lent Myanmar’s longstanding struggle for democracy the boost that it needed in the fight against its military regime. On 1 February 2021, a selfie fitness video performed on a busy roundabout in Myanmar’s capital, Naypyidaw, went viral. The video, produced by influencer and fitness instructor Khing Hnin Wai, was a demonstration of the social and economic progress that technology can bring to enterprising citizens in a country that emerged from military rule into a late dawn of technological progress only in 2011. But the video also captured and alerted the world to the country’s democratic regression in real-time. As Wai danced for her online audience, armoured vehicles rolled into parliament in the background of the shot. The military proceeded to once again seize power in Myanmar. While the regime initially attempted to shut down the Internet, they soon recognized that the old autocrat’s playbook of total communications control is no longer sustainable in the digital era. The country’s economic, as well as the regime’s

A rhetoric of digital politics  7

own, dependence on the Internet forced them to restore access. Some platforms, including Facebook and YouTube, then flexed their muscles and banned accounts associated with Myanmar’s military (Mozur, 2021). Lacking the capacity to create a bespoke online infrastructure like China’s, the new regime must accept that its citizens have access to diverse information and the tools to mobilize and deliberate. Indeed, the resistance movement Campaign for Civil Disobedience (CCD) was able to mobilize online – at first on Facebook and then by switching to Twitter and using free virtual private networks (VPNs) and censorship circumvention applications (Rao and Atmakuri, 2021). The switch to Twitter highlighted the importance of the architecture of individual platforms for democratization. The organization of content around hashtags on Twitter, for example, exposed Myanmar users to wider global perspectives on human rights and the Rohingya genocide four years earlier. Not only did digital media enable the mobilization of protest; it also gave protesters access to information and discussion on the art of democratic rights and responsibilities. As a result, the regime’s traditional propaganda efforts have failed to win the hearts and minds of Myanmar’s netizens who continue their fight for democracy.

BOX 1.2 MYANMAR’S DIGITAL DICTATORSHIP Witnessed in real-time by a powerless Twitter public, Myanmar’s military took control of the country in a coup on 1 February 2021. The military’s armoured vehicles rolling into parliament were unintentionally captured and live streamed by an oblivious fitness instructor recording a selfie video, which soon went viral for reasons other than intended. Yet the country’s recent digital connection to the rest of the world did it little good. The military immediately shut down the internet and, at the time of writing, continues to do so periodically in targeted ways in areas where it faces ongoing opposition or conducts military offensives against civilians. This impedes the work of journalists, human rights monitors and humanitarian organisations (United Nations, 2022). In those areas where the internet is again available, efficient protest remains hindered by the regime’s pressure on ISPs to block social media platforms central to protest mobilisation. Access to information for ordinary citizens is now limited to the wealthy as the regime has pressured telecommunication providers to make SIM card and data prices unaffordable. A new law is underway to curtail online speech and ban the use of virtual private networks (VPNs) (ibid.). Dark days lie ahead for freedom of expression in Myanmar. Meanwhile the new regime employs online digital technologies for its own purposes. In fact, their online preparations have been going on for some time. Since the election in November 2020, soldiers have incited violence on social media. Platforms have been slow and inefficient in their responses. For example, UN investigators found Facebook’s moderation and fact checking efforts far from suf-

8  Handbook of digital politics

ficient (Ratcliffe, 2021), and the regime was long able to continue to use TikTok unimpeded to provoke violence (Reuters, 2021) and confuse protesters by impersonating and misdirecting them. The role of platforms in the incitement of violence in Myanmar highlights two problems. First, the country’s uptake of the internet has not been accompanied by the proper development of critical digital literacy skills (Rao and Atmakuri, 2021). This has become critical in the face of the military’s disinformation campaign. Second, algorithms do not stand the test against coded inflammatory content, which can only be identified by people with local knowledge (Guest et al., 2021). Since social media platforms rely on business models that disincentivise measures against disinformation and content that incites violence, this is unlikely to change anytime soon. Our rhetorical position of ambivalence diverges from both the above accounts. Instead, the story of the coup in Myanmar resembles a tug of war. The military regime relied heavily on social media platforms for their disinformation campaigns and to run the country’s businesses and economy. Yet they hungered for control over protesters’ communication channels. Faced with international platform owners they could not coerce, and lacking the resources to create a bespoke online infrastructure like China’s, they resorted to temporary shutdowns, also of their own communication channels. Their experience was a learning curve; their methods of control only gradually turned to more targeted and sophisticated control of infrastructure, and citizens followed a similar learning curve in their attempts to circumvent new measures. Citizens and social movements use online platforms to enact resistance and solidarity. Their necessarily rapid development of critical digital literacy skills and their dependence on specific platforms for Internet access left them vulnerable to the regime’s online disinformation and to platform algorithms. Online affordances enabled regime members to act as impostors and intercept protest planning. Although Myanmar protesters displayed impressive agility in circumventing the regime’s attempts at closing their communication channels, these efforts by the state are becoming more targeted, restrictive and effective as we write this. Silicon Valley-based social media platforms like Facebook have moved into new and emerging markets by making them dependent on the platform for their Internet use. Their business model discourages the necessary human intervention in specific cultural and linguistic localities against disinformation and hate speech. Their algorithms serve the attention economy rather than democratic freedom. Yet platforms are keen to associate themselves with pro-democracy movements to retain an image of ‘liberation technology’, which is still alive in the public imagination. Facebook therefore did learn from past mistakes during the Rohingya genocide and was ready with local content moderators, even if UN investigators found their efforts inadequate because the military’s coded content got past both algorithms and untrained moderators (United Nations, 2022).

A rhetoric of digital politics  9

Our position of ambivalence reflects an observation made by Blumler and Coleman a decade ago: ‘the present-day political communication process is more complex than was its predecessor, more riddled with crosscurrents, and confronts many of its actors with more choice and greater uncertainty’ (2013, p. 177). Jostling for voice, position in and control over a complex media ecosystem has changed the ways in which authoritarian rulers, media and citizens act in situations such as the Myanmar coup. They are no longer unequivocal in their approach to communication power, nor are its social and political effects. Example 2: Constructing Political Authenticity Where our previous example of the Myanmar coup encouraged reflection on the ambivalent role of digital media in a single event, we now wish to consider its role in relation to a phenomenon. If power relations between different actors in the coup in Myanmar were ambivalent, might the notion of ambivalence also help us understand how digital performances by political actors engender perceptions of authenticity among their supporters? Again, we present a binary exposition of two cases of successful self-exposure and self-branding that both result in the construction of authentic political personas to their respective intended audiences (see Boxes 1.3 and 1.4). We then offer a more ambivalent reading of the phenomenon of digitally mediated authenticity that considers the simultaneous dynamics of vanity metrics, identity construction, deception and exposure that all characterize digital politics.

BOX 1.3 VOLODYMYR ZELENSKYY’S SELF-EXPOSURE: AUTHENTIC SERVANT OF THE PEOPLE Ukrainian President Volodymyr Zelenskyy was after Russia’s invasion of Ukraine in February 2022 voted the most powerful person in Europe by Politico’s readers (March 2022) (Posaner, 2022). Seventy-nine percent of US citizens deemed him a strong leader in April 2022 (YouGov and The Economist, 2022, p. 117), and in July 2022 he posed with his wife on the front cover of Vogue fashion magazine. Zelenskyy’s international image owes much to his authentic use of social media. Already in his election campaign in 2019 he showed himself to be different from Ukraine’s usual oligarch rulers, an ordinary guy, much like his teacher-turned-president character in the TV show Servant of the People in his former acting job. This political persona emerged naturally from his spontaneous self-exposure through low-cost selfie-style videos he shared on social media. His image could not be more different from the strategic game played in Ukraine’s corrupt establishment politics. The hallmark of his innovative campaign was authenticity (Sorensen, 2020). With Russia’s invasion of Ukraine, his authentic appeal extends to the international community. As of late July 2022, Zelenskyy’s Instagram had almost 17 million followers.

10  Handbook of digital politics

In nearly all his posts, he is dressed in an ordinary T-shirt in combat green, clearly identifying himself with the many Ukrainians who took up arms against the invading force. Here is a guy who doesn’t need to put on a suit to be an effective leader for, like most ordinary Ukrainians, he was born ready to face up to Russia. His authenticity remains consistent with his actions; at the outbreak of war, he remained right there in Kyiv with his men. The proof was a YouTube video shot in Kyiv in semi-darkness on his phone the day after the invasion (WFAA, 2022) when he also declined a US evacuation offer. Zelenskyy’s ability to show himself as a real person rather than a staged politician was similarly evident in social media posts at the start of the invasion in which he appeared with bags under his eyes but a clear sense of determination. No make-up or stage lighting were used to disguise his tiredness. As in his election campaign, he continues to shoot selfie-style videos. These lack the professional quality and paraphernalia such as teleprompters usually adopted by politicians and instead identify him with us ordinary folk who regularly use digital technology in the same way (Garber, 2022; Susarla, 2022). Thanks to social media, the West has mobilized behind the Ukrainian leader. His authenticity is not ephemeral; it has led to real international solidarity and on-the-ground results.

BOX 1.4 JAIR BOLSONARO’S SELF-BRANDING: AGGRESSIVE AUTHENTICITY Brazilian President Jair Bolsonaro, a supporter of the former dictatorship and formerly convicted army captain, owes at least part of his election victory in 2018 to his strategic use of social media, which focused on constructing his authentic image (Rocha et al., 2021, ch. 4). Bolsonaro’s use of social media has throughout his presidency strategically disseminated hate speech and disinformation. He has even established an ‘Office of Hate’ (Álvares, 2020). Combined with a populist strategy of identifying with the common man, social media are perfect vehicles for such content. Giving voice to feelings and opinions that the country’s progressive elite deem unsayable, and avoiding journalistic gatekeepers with liberal qualms, social media have given him a direct connection to his supporters’ dark hearts. The symbolic action of disseminating hate speech and disinformation has thus become a self-branding strategy for Bolsonaro. He presents himself as someone who does not hide his true feelings or opinions behind a self-censoring mask. Bolsonaro’s symbolic attempts to legislate against social media platforms’ moderation of hate speech and disinformation signal his obdurate intent to remain on this course (Caeiro, 2021). WhatsApp has been a convenient means of avoiding oversight of his illicit activities. The infrastructure of the platform has also enhanced his self-branding strategy. Large political WhatsApp discussion groups are common in Brazil and allow Bolsonaro’s supporters to play an active role in the construction of his au-

A rhetoric of digital politics  11

thenticity by sharing personal experiences that demonstrate the righteousness of his vitriolic outbursts. By demonstrating that he speaks for the common man, such supporters spread his messages and encourage others to vote for him (Owen, 2018). To protect this authentic image, Bolsonaro strives for purity rather than pluralism in his constructed image. His social media activity is his primary communication form with the public. Human Rights Watch (2021) express concern that the president is blocking critical followers on his social media account, including citizens, governmental institutions such as the Ministry of Justice, media and civil society organizations. Given Bolsonaro’s extensive use of social media for official purposes, these actions have implications for freedom of expression, including the right to seek, receive and impart information and to participate in the conduct of public affairs. They also demonstrate his concern with strategically manipulating his image of the authentic man of the people. Based on these accounts online platforms can be seen as means of constructing authentic political personas, for better or worse. As environments that cultivate norms of both self-branding (Khamis et al., 2017) and self-exposure (Halsema, 2021), social media platforms are ambiguous agents of authentic self-representation and identity construction. While they encourage users to expose who they really are, vanity metrics such as ‘Likes’ and retweets also push users to accentuate certain personality traits and modes of expression in self-aware performances. In other words, political actors’ self-branding and self-exposure are not either/or self-presentation strategies. They operate hand in hand in dynamic tension. Beyond this ambiguity, digital media are also sites of struggle over such constructed personas and realities. Scholars are tracking Bolsonaro’s disinformation campaigns (Dourado and Salgado, 2021) and fact-checkers are exposing his false claims (Palau, 2021). Yet direct refutations of disinformation have been found to encourage further propagation as well as political fatigue and cynicism in citizens (Deibert, 2019, p. 32). Also Zelenskyy’s performances of authenticity are contested, less for their dissimulation of his real self than for their authenticity of origin and risk of forgery. For example, in mid-March 2022 a deepfake video of the Ukrainian leader was constructed using artificial intelligence and circulated online (no longer available). In the video, Zelenskyy is moving his head and telling Ukrainian citizens and soldiers to surrender to Russia. Zelenskyy himself quickly debunked the video as fake in a Telegram selfie video (Digital Forensic Research Lab, 2022). Digital platforms may thus be used to curate a politician’s authentic image and to undermine this image when journalists or members of the public expose the staging of such curation. Yet malignant actors can also deploy archived material for manipulative purposes to create an alternative reality that is an apparent authentic political performance. The ambiguity inherent in social media’s relationship to authenticity suggests a more complex window on reality than black and white accounts present. We hope to have indicated in the above examples that while there is a superficial appeal to accounts that purge ambivalence, they are ultimately vulnerable to being

12  Handbook of digital politics

uncovered as crude rhetorical efforts to consolidate one impression by means of suppressing another. Beyond such classificatory certainty lies a more promising perspective.

CREATIVE AMBIVALENCE Our intention in making the case for ambivalence is not to embrace indeterminacy. The task of academic research is to describe, define and explain and it is a cop-out to avoid conclusions on the grounds that phenomena are slippery. The argument we are offering here is that rhetorical over-determination tends to blur the creative possibilities that are inherent in contingency. Ambivalence, unlike certainty, implies a creative dimension and an openness of mind. In academic enquiry, ambivalence is both a rhetoric and an epistemic position. It opens up the ground to contestation, deliberation, new opinion formation and the potential for change rather than a technologically determinist foreseeable future. It demands curiosity, empathy and deep listening to unfolding events and the subjectivities that shape them where entrenchment and equivocality closes minds. It is precisely these qualities of creativity and open-mindedness that have made digital technologies and spaces so exciting. While Facebook might be seen to imprison its users within a corporate stranglehold, what is politically intriguing is that some of its users have found ingenious ways of subverting the form for autonomous and collective ends. While the potential for connective action offered by digital networks (Bennett and Segerberg, 2013) cannot be denied, it is complicated by the proven capacity of elite political and economic structures to marginalize or co-opt this emancipatory promise. The scope of political opportunity lies in detail, itself commonly dependent upon the ambivalences of agency in context. When there exists what Merton (1976, p. 11) referred to as a ‘disjunction between culturally prescribed aspirations and socially structured avenues for realising these aspirations’ there is bound to be a creative tussle to determine whether a social phenomenon can be stretched towards its cultural capacity or will be stifled by structural constraints. Such contestation is at the core of politics. There is a paradoxical quality to the study of digital politics. Partly one is investigating the ways in which the digital stretches and constrains the political, but at the same time one is attending to the ways in which the digital is politically shaped and acted upon. Such a dialectic cannot avoid ambivalence, for there are moments in which the affordances of technology establish or loosen political manoeuvrability and others in which the exigencies of political agency disrupt seemingly intractable technical pathways. This incessant push and pull is best explained in terms that, rather than focusing upon the constantly fluctuating misalignments between cultural aspiration and structural constraint, are sensitive to the propulsive thrust that animates them. The question here has less to do with the direction of influence between the digital and the political than how these forms emerge as energies capable of constituting and regulating subjectivities. How do particular events, institutions and procedures come to be classified as political? How do devices, processes and networks come

A rhetoric of digital politics  13

to be categorized as technological? How are social phenomena such as citizenship, democracy, community or movements assembled through digital politics? These are rhetorical and performative questions. When we ask why data emanating from search engines come to be regarded as authoritative information we are dealing with a form of contextually-specific claim-making that relies upon techniques of persuasion. Tensions between Google as algorithmic manipulator and objective truth-teller have to be resolved though plausible narrative. Running alongside the much-celebrated connectivity engendered through digital networks are attempts to connect events, concepts and publics in ways that shape perceptions of reality. The effects of such efforts are bound to be ambivalent, depending not least upon the experiential differences between their recipients. Sweeping impulses towards digital optimism and pessimism lack sociological nuance, flattening experience into the breathless rhetorical tones of the technocratic utopian and the lugubrious realism of the cyber-sceptic. Both of these are wearisome deflections from the work of detecting creativity within nuance. In place of such binaries, a more percipient rhetoric of digital politics must come to terms with the fine-grained depth and distinctiveness of subjective experience. For it is the definition of experience that is a main prize of contemporary politics. We live in an era in which political communication depends increasingly upon the mobilization of affect – upon making people feel certain ways. Politicians have become experts in dispositional priming, making people worried about things they wouldn’t otherwise be worried about, ambitious for things they wouldn’t otherwise want, satiated by things that would in the ordinary course of events leave us feeling empty. Political rhetoric has come to rely upon opportunist appeals to emotional attention. Digital spaces are key strategic zones for such rhetorical jockeying. Political contestation on social media tends to be about the setting of atmospheres and the contestation of feelings. In their seductive efforts, political campaigners seek to define reality in ways that make people feel good about who they are; the communities to which they are attached; the values that they hold dear. More malignantly, digital politics seeks to other and undermine targets. Classification battles abound. In the absence of singular authorities, the Internet becomes a space for contesting legitimate labels, descriptions and evaluations. It can be ugly, but it is politics in the raw and if we want to understand it we need to pay at least as much attention to the rhetorical and technological strategies of political assertion as to the assertions themselves. When it first emerged we imagined that digital politics would be a new ground for the conduct of conflicts between old subjectivities. But it has turned out to be a space for the assemblage of reconfigured subjectivities. Who is the public? Who is us? Who is them? Who am I? Which I shall I be today? Digital politics shines a light upon the intrinsic ambivalence of social identity, power, connection and reality. The bigger question than ‘what shall we do with these new tools’ is ‘what will they make of us’ and then ‘how might we use them to make our better selves’.

14  Handbook of digital politics

FURTHER READING Coleman, S. 2012. Making the e-citizen: A sociotechnical approach to democracy. In S. Coleman and P. M. Shane (eds.), Connecting Democracy: Online Consultation and the Flow of Political Communication. Cambridge, MA: MIT Press, pp. 379–394. Finlayson, A. 2022. YouTube and political ideologies: Technology, populism and rhetorical form. Political Studies, 70(1), 62–80. Mansell, R. 2012. Imagining the Internet: Communication, Innovation, and Governance. Oxford: Oxford University Press. Yar, M. 2014. The Cultural Imaginary of the Internet: Virtual Utopias and Dystopias. Basingstoke: Palgrave Macmillan.

REFERENCES Álvares, D. 2020. Against vaccine, Bolsonaro son reactivates “Office of Hate.” The Brazilian Report. https://​brazilian​.report/​power/​2020/​10/​24/​against​-vaccine​-bolsonaro​-son​ -reactivates​-office​-of​-hate/​ (accessed 16 September 2022). Atkinson, P. 2014. The Ethnographic Imagination: Textual Constructions of Reality. Abingdon: Routledge. Bauman, Z. 1993. Modernity and Ambivalence. Cambridge: Polity Press. Bennett, W. L. and Segerberg, A. 2013. The Logic of Connective Action: Digital Media and the Personalization of Contentious Politics. Cambridge: Cambridge University Press. Blumler, J. G. and Coleman, S. 2013. Paradigms of civic communication. International Journal of Communication, 7(1), 173–187. Caeiro, C. 2021. Bolsonaro’s Social Media Plan Shows His Election Worry. Chatham House – International Affairs Think Tank. https://​www​.chathamhouse​.org/​2021/​09/​bolsonaros​ -social​-media​-plan​-shows​-his​-election​-worry (accessed 16 September 2022). Deibert, R. J. 2019. The road to digital unfreedom: Three painful truths about social media. Journal of Democracy, 30, 25–39. Diamond, L. 2010. Liberation technology. Journal of Democracy, 21, 69–83. Digital Forensic Research Lab. 2022. Russian War Report: Hacked News Program and Deepfake Video Spread False Zelenskyy Claims. Atlantic Council. https://​ www​ .atlanticcouncil​.org/​blogs/​new​-atlanticist/​russian​-war​-report​-hacked​-news​-program​-and​ -deepfake​-video​-spread​-false​-zelenskyy​-claims/​ (accessed 16 September 2022). Dourado, T. and Salgado, S. 2021. Disinformation in the Brazilian pre-election context: Probing the content, spread and implications of fake news about Lula da Silva. The Communication Review 24, 297–319. European Information Society Forum. 1999. Conference reports. IST 99 – HELSINKI, 22–24 November 1999, Information Society Technologies (IST) Conference, Kybernetes, Vol. 29, No. 3. https://​doi​.org/​10​.1108/​k​.2000​.06729cab​.004. Garber, M. 2022. The grim stagecraft of Zelensky’s selfie videos. The Atlantic. Halsema, A. 2021. Narrative self-exposure on social media: From Ricoeur to Arendt in the digital age. In W. Reijers, A. Romele, and M. Coeckelbergh (eds.), Interpreting Technology: Ricoeur on Questions Concerning Ethics and Philosophy of Technology. Lanham, MD: Rowman & Littlefield, pp. 99–116. Human Rights Watch. 2021. Brazil: Bolsonaro Blocks Critics on Social Media. Human Rights Watch. https://​www​.hrw​.org/​news/​2021/​08/​19/​brazil​-bolsonaro​-blocks​-critics​-social​ -media (accessed 16 September 2022). Khamis, S., Ang, L., and Welling, R. 2017. Self-branding, ‘micro-celebrity’ and the rise of social media influencers. Celebrity Studies, 8, 191–208.

A rhetoric of digital politics  15

Merton, R. K. 1976. Sociological Ambivalence and Other Essays. New York: Simon & Schuster. Morozov, E. 2012. The Net Delusion: How Not to Liberate the World. London: Penguin. Morris, D. 1999. Vote.Com: How Big-Money Lobbyists and the Media Are Losing Their Influence, and the Internet Is Giving Power Back to the People. Los Angeles: Renaissance Books. Mozur, P. 2021. YouTube bans Myanmar military channels as violence rises. The New York Times, 5 March. Negroponte, N. 1995. Being Digital. New York: Alfred A. Knopf. Norris, P. 2001. Digital Divide? Civic Engagement, Information Poverty and the Internet Worldwide. Cambridge: Cambridge University Press. Owen, L. H. 2018. What to know about WhatsApp in Brazil ahead of Sunday’s election. Nieman Lab. https://​www​.niemanlab​.org/​2018/​10/​what​-to​-know​-about​-whatsapp​-in​-brazil​ -ahead​-of​-sundays​-election/​ (accessed 16 September 2022). Palau, M. 2021. Inside Brazil’s dangerous battle over fake news. Americas Quarterly. https://​ americasquarterly​.org/​article/​inside​-brazils​-dangerous​-battle​-over​-fake​-news/​ (accessed 16 September 2022). Posaner, J. 2022. Ukraine’s Volodymyr Zelenskyy tops POLITICO 28 poll as most powerful person in Europe. POLITICO. https://​www​.politico​.eu/​article/​ukraine​-zelenskyy​-europe​ -most​-powerful​-person​-politico28​-2022​-poll/​ (accessed 15 September 2022). Rao, A. and Atmakuri, A. 2021. The Role of Social Media in Myanmar’s CDM: Strengths, Limitations and Perspectives from India. Working Papers, Long-term studies on trends and issues in South Asia. Institute of South Asian Studies, National University of Singapore. Rocha, C., Solano, E., and Medeiros, J. 2021. The Bolsonaro Paradox: The Public Sphere and Right-Wing Counterpublicity in Contemporary Brazil. Cham: Springer Nature. Rorty, R. 1989. Contingency, Irony, and Solidarity. Cambridge: Cambridge University Press. Schraube, E. 2009. Technology as materialized action and its ambivalences. Theory & Psychology, 19(2), 296–312. Sorensen, L. 2020. Symbolic politics meets digital media: Research on political meaning-making. In W. H. Dutton (ed.), A Research Agenda for Digital Politics. Cheltenham, UK and Northampton, MA, USA: Edward Elgar Publishing, pp. 133–145. Susarla, A. 2022. Why Zelenskyy’s ‘selfie videos’ are helping Ukraine win the PR war against Russia. The Conversation. http://​theconversation​.com/​why​-zelenskyys​-selfie​-videos​-are​ -helping​-ukraine​-win​-the​-pr​-war​-against​-russia​-178117 (accessed 15 September 2022). Ukrainian President Volodymyr Zelensky: “We are still here”. YouTube video, 25 February 2022. United Nations. 2022. Myanmar: UN Experts Condemn Military’s “Digital Dictatorship”. United Nations Human Rights Office of the High Commissioner. YouGov and The Economist (2022). The Economist/YouGov Poll, 26–29 March 2022 – 1500 US Adult Citizens (Economist Tables No. 28 March 2022). YouGov.

2. De-Westernizing digital politics: a Global South viewpoint Bruce Mutsvairo, Fabíola Ortiz dos Santos and Tenford Chitanana

INTRODUCTION Even though empirical studies on the influence of social media as pathways for social and political communication in non-Western societies are expanding (Gore, 2023; Barclay and Boobalakrishnan, 2022; Ndlela and Mano, 2020; Mutsvairo and Salgado 2021; Mitchelstein et al., 2020; Dwyer and Molony, 2019) there is a dearth of scholarship on digital politics particularly when viewed from a de-Westernized perspective, a central approach taken by this chapter. Indeed, it is no longer possible for political leaders and strategists to ignore the digital realms. This is the case even across the Global South, a diverse region mainly consisting of formerly colonized states in Latin America, Asia Pacific, the Arab World and Africa, which traditionally have been known epicenters associated with a myriad of challenges ranging from the digital divide to digital illiteracy, ultimately encumbering citizens’ participation in social and political deliberations online. Things are changing though, at least for now, particularly among the younger generations, who see digital communication platforms as the new game in town, using them for everything including shopping online and forcing their local political representatives to be more accessible and accountable to their constituencies. In this way, social media platforms have not only boosted political expediency among political players but also the mechanisms through which political support strategies such as fundraising are sourced and mediated between citizens and political players. Critically though, in many of these countries, traditional ways of conducting political business remain intact. In fact, it is very difficult to measure social media’s impact on politics because such platforms have largely turned out to be “fragmented, transient, polarizing and unreliable” as posited by Srinivasan et al. (2019) in an important special issue on publics in the African digital age, which was published by the Journal of Eastern African Studies. It also is no easy task to, for example, use online technologies to politically engage older people living in rural outposts, which are customarily excluded from Internet infrastructural development and in the case of several “developing world” countries, whose rural communities are excluded from key educational opportunities. More importantly, it is also not always easy to know if social media platforms will foster the political engagement that leads to democratic enhancement because digital authoritarianism (Yilmaz et al., 2022) remains deeply embedded within the non-Western polity. While research elsewhere 16

De-Westernizing digital politics: a Global South viewpoint  17

has shown the power of social media in improving political opinion expression (Oser and Boulianne, 2020; Coleman, 2017), many countries in the Global South have shut down the Internet whenever they have faced threats of popular uprisings. It has for example become normal for some African political establishments to discreetly switch off the Internet whenever they feel their power is threatened. Whether it is Cameroon, Ethiopia or Chad, social media platforms have become important tools for encouraging civic activism, forcing the African political elite to act by blocking Internet access, drawing criticism from Western governments, who see blocking the Internet as an affront to democracy. In fact, that is also where the problem begins. Any efforts to de-Westernize African digital politics should also start there. Why should democracy, a very difficult concept to define, be foisted upon these people? A de-Westernized viewpoint on African digital ecologies, for example, is an important step toward understanding why democracy is failing to take root in many nations across the Global South. This chapter is therefore relevant for three reasons. First, focusing mostly on the continent of Africa, it conceptually de-Westernizes digital politics, contributing fresh understanding on non-Western conceptualizations of the online sphere. Secondly, it critiques the digital networks from an anti-colonial perspective, articulating frequently overlooked perspectives on the inherent historical challenges inhibiting technological advancements across the Global South. Finally, it furthers knowledge on where future studies should focus if we are to embrace a de-Westernized notion of digital politics.

DE-WESTERNIZATION DEFINED The concept of de-Westernization “asks for a revision of the power relations in global academic knowledge production and dissemination” (Glück, 2018), attempting to deconstruct “dominant elitist Western axiology and epistemology” (Glück, 2018, p. 5), actualizing and promoting a wider pursuit of knowledge including enlisting ideas and ontologies from non-Western societies. To understand de-Westernization, one needs to embrace the role context plays in the establishment, extraction and distribution of knowledge. For example, while proponents of a possible “digital democracy” (Dahlberg, 2011; Iosifidis and Nicoli, 2020) have envisioned the possibility of digital technologies facilitating citizens’ voices in policy making, such ideas are not always in tandem with the realities of some societies in the Global South. The African situation, notwithstanding the dangers of lumping 54 nations together, shows the continent is a different terrain altogether. First, there are several countries that generally lack political legitimacy. They are almost always authoritarian in their dealings with citizens. They suppress information and put opponents in prison. Young people rarely get a chance to participate in important political decision-making. For example, faced with skyrocketing unemployment, youths in countries such as Mali and other nations across the Sahel, have been forced to join radical religious groups because they give them attention and a sense of identity (Vermeersch et al., 2020).

18  Handbook of digital politics

In such circumstances, it is difficult to envision a digital democracy because these youths probably have everything but democracy on their wish list. Neglected by their own governments, they seek a sense of belonging using social media primarily for non-political reasons that advance their self-seeking goals. Downtrodden and often hungry, democracy emerges as more of a pipe dream, a luxury. They may aspire for it, but they prefer a robust, pragmatic, trial-and-error approach, which ensures that their interests, perceived or real, are safe and secure. De-Westernizing digital politics also means questioning the veracity of widely shared Eurocentric perspectives suggesting the increasing reliance of citizens on social media enhances democracy in the Global South. While platforms are giving a voice to a wider pool of political players, enabling social engagement and shaping online and offline political narratives in Latin America (Mitchelstein et al., 2020), long before the digital era, African youths were disillusioned by their older political leaders (Bornman et al., 2021) providing clues to their less vibrant political engagement online. Indeed, they are disconnected with politics. A case in point is Zimbabwe, where for many political activists, the road toward democratic change has been rocky. Faced with democratic fatigue (Kelley, 2022), many of those seeking political change in the country end up using social media for non-political reasons, preferring to log in online to catch up with the latest gossip involving social media superstars who have taken advantage of the rising popularity of disruptive digital ecosystems to become instant online icons. One such celebrity is Sweden-based transgender socialite Tatelicious Karigambe-Sandberg, who clocks thousands of followers each time they go live on Facebook to discuss juicy gossip normally laced with profanities and expletives, which are mostly not tolerated in their deeply conservative real-world communities but accepted online. This serves to confirm Matthes’ (2022) suggestion that entertainment and relational reasons as opposed to political purposes are central to youth participation online. Their fatigue comes from decades of unsuccessfully seeking change with the ruling elite intolerant of all forms of dissent, online or otherwise. We realize there are still many Zimbabweans using social media for political purposes, but recent research has also shown urban youths seeking ways to detox from social media due to addiction and generational conflict through technological use (Mutsvairo et al., 2022). There are many reasons why universalizing Western standards and values as an expected norm for everyone is problematic. Next to its notable advantages including the promotion of awareness and new knowledge or helping politicians connect with a wider audience, platforms have also become key avenues for the dissemination of hate speech, misinformation and disinformation undermining the legitimacy of democracies (Posetti and Ireton, 2018). While the rapid spread of misinformation and disinformation on social networks is a global concern, false information poses a greater risk across the Global South because many of its countries, unlike the West in general, have weaker institutions, which struggle to make those responsible for disseminating and spreading false narratives online accountable for their actions. Not that legal jurisdictions in the West know how best to defeat disinformation but it is at the very least expected that stronger institutions ensure perpetrators of any crime face

De-Westernizing digital politics: a Global South viewpoint  19

justice. When corruption and impunity become the order of the day across a nation’s political and legal fraternities, leading to the promotion of unchecked violence, hate and harassment, many choose to move to countries where they feel safer. According to a recent study, just a paltry 32 percent of the 4,500 African youths aged 18–24 were positive about Africa, with many of them, battered by the pandemic, political instability and endless conflicts, considering moving abroad (Ahmed, 2022). At the same time, people of African descent living in major European and American cities are also moving back to Africa in search of a sense of belonging, with Ghana and Gambia popular destinations for the Back to Africa movement (Richards, 2020). The reactions of authoritarian regimes to digital activism or other forms of political engagement online also inform how citizens in different contexts adapt and utilize the Internet in their daily lives. In Myanmar, for instance, the military governments in their various iterations for over five decades have determined the country’s development trajectory, including suppressing the national telecommunications systems. The Internet only became widely available within the past decade with the introduction of mobile telephony technologies. While this connectivity coincided with the rapid growth of global telecommunication systems, particularly social media, and mobile telephony technologies, ostensibly opening up and linking Myanmar to the rest of the world, the military has remained largely in charge of the communication political economy. They have interests in telecommunication companies, and they have control over the regulatory system which undermines dissent and free speech. While people in Myanmar have incrementally used social media and other digital tools to advance diverse civil society causes, the heavy-handed military responses to digital dissent have also affected how people engage online. Thus, when trying to understand political communication in related contexts, the nuances of history and context come to the fore.

A CULTURE-CENTERED APPROACH Next to our calls to incorporate a de-Westernizing perspective on digital policy and new media research, a more culture-centered approach could be fostered in this decolonial turn.1 As a school of thought, decoloniality is associated with Latin American scholars.2 Decoloniality should not only be understood purely as the decolonization process – defined as the end of colonial occupation and administration – but as a wide rethinking of relations to “ongoing coloniality” that are still materialized through colonial structures that persist everywhere in people’s minds (Couldry and Mejias, 2019, p. 80). The decolonial consciousness seeks to decolonize the power, the “being” and the know-how (Maldonado-Torres, 2007). As suggested by Couldry and Mejias, decoloniality seeks to offer not only strategies for “surviving in a neo- or postcolonial context but also grants models for articulating an alternative worldview arising from the Global South” (2019, p. 80). It is a conception and apprehension of the world that challenges and refuses Eurocentric canons of modernity. With this in mind, we may

20  Handbook of digital politics

attempt to transpose this idea to digital politics and new media research and consider that although occasionally being internationally funded, it is possible to develop locally based narratives and solutions and embrace communities’ and nations’ own locus of enunciation with the perspective of creating new and decolonial accounts. In this sense, a perspective of fostering a more culture-centered approach to digital politics and media (including the use of Internet and social networks but not limited to that) which recognizes that platforms for communication and information exchange should not be used as “tools for attacking or taking advantage of people irrespective of age, sex, race, social status or sexual orientation” (Ephraim, 2013, p. 281) could be a good starting point. In some parts of the African continent, the use of mobile phones constituted the first opportunity ever to participate in telecommunications, because the few existing telephones often only worked on an irregular basis, especially in rural areas (Hahn and Kibora, 2008, p. 88). In a study that examined the specific modes of dealing with mobile phones in Burkina Faso, it was observed that the same technology could take on different meanings across changes and uses in cultural contexts and, thus, be used in radically different ways (Hahn and Kibora, 2008, p. 91). Another example of a culture-centered approach to the use of the new media is the case of web development within the Soninke ethnic group across Mali, southern Mauritania, eastern Senegal, Guinea, and The Gambia. The development of the Soninke web has not been an endeavor conducted by authorities (Galtier, 2011), it was rather an enterprise carried out by nationals, the diaspora, cultural activists and associations. Despite the challenges of graphic code and standardization of the Soninke language, their cultural websites have multiplied on the Web and offer great diversity of topics connecting Soninke social groups (Galtier, 2011).

DIGITAL TRANSFORMATION AND E-COLONIALISM Digital transformation is a “game-changer” for the African continent, and this has been acknowledged by G73 world leaders in an official document released during the 2019 Summit held in Biarritz, France. But despite “digital Africa” being one of the core topics discussed at this pre-Covid-19 summit, no African counterparts were invited to attend. Here, leaders of rich economies discussed among themselves how to ensure that the process of digital transformation on a continent that has historically been subjected to imperialism and colonization remains “open, free and secure” (G7, 2019). In the two-page Annex, the Summit stressed that digital technology could “drive innovation, economic growth and job creation in many key sectors of the economy” allowing for greater interconnection of African markets (G7, 2019, paragraph 2). Furthermore they were adamant that this common one-way global strategy for the continent would embrace an effort of enabling the necessary digital infrastructure so as to reduce the digital gap and inequality “including in isolated countries and regions that are excluded or underserved” (G7, 2019, paragraph 4). The need to develop digital literacy, foster digital entrepreneurship, and the sharing of

De-Westernizing digital politics: a Global South viewpoint  21

best practices including the creation of legislative and regulatory frameworks were, moreover, stressed. Additionally, leaders exchanged views on how to continue the fight against terrorist content online and recognized the “need to address the specific situation of fragile regions, such as the Sahel, Horn of Africa and Lake Chad regions” (G7, 2019, paragraph 7). This G7 Summit among other global fora is an example of how world leaders of big Western economies geographically placed in the Global North have enforced their beliefs and practices of what should be seen as universal needs and standards to be implemented and internationally accredited as good foreign policies and practices. These types of one-way cooperation reflect, in a way, the same old paradigm of imposition reproducing the colonial modus operandi, this time around centered on Web 2.0, which McPhail (2014) theorized as e-colonialism – the electronic colonialism theory: Electronic colonialism represents the dependent relationship of poorer regions on the post-industrial nations which is caused and established by the importation of communication hardware and foreign-produced software, along with engineers, technicians, and related information protocols. These establish a set of foreign norms, values, and expectations that, to varying degrees, alter domestic cultures, languages, habits, values, and the socialization process itself. (McPhail, 2014, paragraph 4)

Whereas the mercantile form of colonialism sought to control and exploit cheap manual labor, extract raw materials, and then ensure a market for finished industrialized products, the electronic form of colonialism aims at influencing and controlling the mind. It is aimed at manipulating attitudes, desires, beliefs, lifestyles, and consumer behavior (McPhail, 2014, paragraph 5). The major goal of e-colonialism is to control how mass media influence the mind. The digitally-based information revolution focuses more on the role and consequences of the intellect, consumer behavior, and the structural changes across aspects of life (McPhail, 2014). In line with McPhail, Atintande (2020) posits that digital communication in Africa is at a crossroads and has moved from the physical exploitation in the past to a position of virtual dominance once the digital space is controlled by a few powerful private Western technology giants. “Most African countries today no longer own or control their telecommunications infrastructures, as Africa’s telecommunications and digital infrastructures are being taken over by these big multinationals” (Atintande, 2020, p. 41; Mutsvairo and Moyo, 2022). This notion of e-colonialism portrays levels of dependency among nations and regions established by the importation of communication hardware and foreign-produced software, along with engineers, technicians, and related informational protocols and products. It relates to the “pervasive influence of large multimedia conglomerates which drive, control, produce, and spread the global flow of information” (Atintande, 2020, p. 62). This conceptual view of the world is evident across international fora as in the aforementioned G7 official document. The leaders of wealthy Western countries have been reluctant to work with African leaders so as to define the needs expressed by the African societies themselves concerning the

22  Handbook of digital politics

digital transformation. It was not acknowledged that there is a need for collaboration that could embrace the agency of African countries (typically seen as “recipients” of international aid or foreign policies) in an equal and more balanced way. These strategies and policies apparently do not consider the “beneficiaries” as core actors in the process of digital transformation and inclusion.

PHYSICAL CHALLENGES AND DIVIDES IN WEST AND CENTRAL AFRICA After the outbreak of the Covid-19 pandemic, the annual economic report released by the Organisation for Economic Co-operation and Development and the African Union (AUC and OECD, 2021) echoed similar sentiments proffered by the G7 remarking that the digital transformation could, now in the context of the pandemic, drive more innovative, inclusive and sustainable growth for the achievement of Agenda 2063 (AU, n.d.).4 The third edition of this annual economic report examined how digital transformation would support the creation of jobs and new opportunities for young people, having among its priorities: the universal access to the digital solutions best suited to local contexts; digital technology as a lever for productivity, especially for small and medium sized enterprises; the development of skills and expertise of the African workforce; and the coordination of multiple digital strategies at the continental, regional, national and local levels to better prioritize, implement, monitor and evaluate progress (AUC and OECD, 2021, p. 5). In a departure from the past, the acknowledgment for the need to search for solutions that are suited in different contexts is included in the last policy document. The actions to deepen regional and continental cooperation for digital transformation are also stressed in the report that specifically addressed Francophone African countries in regions such as Central5 and West Africa6) emphasizing the expansion of the digital coverage; the equipping of the workforce; the promotion of digital entrepreneurship and the support of entrepreneurs in using digital technologies, especially in agricultural sectors; strengthening governmental support to technology parks and start-up incubators; and the implementation of regulatory frameworks to develop fintech (AUC and OECD, 2021, p. 20). The focus on African countries is appropriate if not urgent because, for example, access to digital tools particularly in Central Africa is the lowest on the continent (AUC and OECD, 2021, p. 141). Although it still lags other parts of the continent, there has been a surge in Central Africa’s adoption of mobile phones. By 2018, 66 percent of the population had taken out a mobile phone subscription (ten percentage points lower than the African average), compared to 45 percent in 2010 (World Bank, 2020 World Development Indicators cited in AUC and OECD, 2021). Central Africa is struggling to increase the Internet penetration rate among the regional population in general, but also among public and private companies. Significant structural constraints have hindered any attempt for creating digital jobs – less than 48 in 100 people have access to electricity, while the mobile phone subscription rate

De-Westernizing digital politics: a Global South viewpoint  23

(66 percent in 2018) remains ten points below the African average (AUC and OECD, 2021, pp. 138–139). Weak purchasing power (with an average monthly per capita income of USD 195), combined with a lack of competition between operators explain why the cost of mobile phone communication remains a major obstacle to digital expansion in this part of the continent (with the average communication cost of more than 20 percent of the population’s monthly income; AUC and OECD, 2021, pp. 139–141). As an indication of the challenges in this part of Africa, only 5 percent of intermediary cities7 in the region are connected to fiber-optic broadband corresponding to the lowest rate in the continent. Nine out of every 100 people use a computer in Central Africa and one‑third of the region is covered by 4G. The high cost of subscriptions explains the low Internet penetration rate, which is 26 percent compared to an average of 35 percent for the continent (2021, pp. 138–139). Despite the fact that mobile money transactions increased nine-fold since 2010 – from USD 200 million to USD 1.8 billion in 2019 – the potential of digital entrepreneurship remains largely untapped since only nine start-ups raised over USD 100,000 during 2011–2020. The AU/OECD report declared that the structure of the region’s economies, especially those that export minerals, is not “conducive to digital development” (2021, p. 138). When we shift our attention to West African countries – five of them English-speaking but surrounded mainly by French-speaking countries – the access to communications infrastructure in the region has improved according to the same report. The proportion of the population with a cell phone in West Africa in 2018 was 40 percent (ITU cited in AUC and OECD, 2021, p. 222), and the fraction that use mobile phones regularly is 74 percent in 2018 (Gallup cited in AUC and OECD, 2021, p. 222). The 4G network coverage in the Western African region has expanded fast from 15 percent of the population in 2015 to 63 percent in 2020. The percentage of the population with Internet access in 2018 reached 24 percent among the poorest, and among rural inhabitants it also expanded to 26 percent (Gallup cited in AUC and OECD, 2021, pp. 221–222). E-commerce has experienced “robust growth” of 9 percent per year since 2010. And mobile banking has fostered financial inclusion, for example, of two thirds of adults in Senegal, who now have a mobile banking account, similarly in Burkina Faso, Ivory Coast, Togo, and Benin. Likewise in Central Africa, the West African labor market remains dominated by informal employment (AUC and OECD, 2021, p. 223). Although digital transformation has accelerated and been stimulated in the region by submarine cables,8 the gaps in infrastructure and skills expose “stark inequalities” (AUC and OECD, 2021, p. 224) and West Africa’s digital connectivity is still weak. The flagrant inequality is evidenced by the extremely high Internet connection costs that generally discourage the use of applications or technologies. A good example is the cost of 20 Gigabytes (GB) of mobile data rates in Ivory Coast, which is pegged at 30 Euros, while the same provider sells packages that include free phone calls and SMS with 100 GB of mobile data for less than 20 Euros throughout Europe (Kouamé, 2019). Despite the explicit reason for such high costs being the risk of submarine cable damage and

24  Handbook of digital politics

“digital isolation” (Cariolle and Goujon, 2019) the economic challenges should not penalize citizens. Digital media remain beyond many Africans’ reach, writes Conroy-Krutz (2020). There is a pronounced digital divide with younger, better-educated, wealthier, male, and urban-dwelling Africans much more likely to access the online world. As digital media access continues to rise across demographic groups and in most countries, the potential of shaping more knowledgeable and active populations appears to be exciting as the majority of Africans see digital media as having mostly positive effects on society. There is, nevertheless, distrust as new media is also seen as facilitating the spread of false information and hate speech (Conroy-Krutz and Koné, 2020). Indeed, rich countries have been criticized for fostering information disorder (Moyo, 2020). Facebook, for instance, announced in December 2020 that it removed almost 500 accounts and pages tied to French and Russian disinformation campaigns that largely focused on the Central African Republic elections – scheduled for December 27, 2020 – that also targeted users in 13 other African countries including Algeria, Cameroon, Libya and Sudan (Matiashe, 2020; Stubbs, 2020). The Russian disinformation campaign in African countries had been denounced a year earlier (Davey and Frenkel, 2019; Fidler, 2019) posing a test for Internet policy in Africa. The lack of digital literacy plays a role as well in the dissemination of false news in countries where access to the Internet is still precarious. It is also seen as a component that fuels the profusion of misconceptions and information disorder (Santos, 2021).

LOCALLY BASED DIGITAL SOLUTIONS The drive towards development across the continent has been nothing but “the extension of colonialism and the struggle to emancipate itself from same” (Musa, 2020, p. 76). In a book chapter on Africa and development policy in the digital era, the author later asks, “To what extent can the digital revolution offer a path to Africa’s development that does not require it to remain tethered, on a leash, to cultural imperialism and limiting conceptions of growth? (Musa, 2020, p. 76). The “modernist turn” in the continent implied “sacrificing the communal system” that had served as a “social glue that held society together” (Musa, 2020, p. 77). The Western cultural model considers the communal family system as disadvantageous to development and progress and its imposition disregards African values, argues Musa. While the digital revolution together with the infrastructure of digital communication have changed the African media and communication landscape, it is important to note that the media ecology perspective views changes in modes of communication as sub-structures of culture. In essence, once the forms of communication change, other institutions in society will follow (Musa, 2020, p. 86). When it comes to connecting individuals and groups in a two-way flow, digital communication allows for instant pluri-directional communication between senders and receivers. “Every person and group can have a say and have its voice heard. The needs of local communities can be brought to the forefront just as much as the

De-Westernizing digital politics: a Global South viewpoint  25

needs of the national groups” (Musa, 2020, p. 88). According to Musa, the “missing ingredient” is one that upholds both the community and individuals. “The digital revolution makes it possible to balance global and local needs, goals, and interests in development. In many ways, it will allow for [a] truly communal approach” (Musa, 2020, p. 88). It is important here to consider that despite the evidence of the uneven expansion of information and communications technologies (ICTs) on the continent, enabling many citizens easily to access social networks, several groups are left out of the digital participation since online activity is limited to those who can read and understand the colonial languages and afford the cost of connectivity (Mutsvairo and Ragnedda, 2019). A decade ago, there were doubts as to the possibility of Africa’s success in the global social media sphere because of the widespread poverty and the unequal distribution of access to ICT tools (Ephraim, 2013, p. 276). Now, however, the reality and spirit have changed.

CONCLUSION: CONSIDERATIONS FOR FUTURE RESEARCH This chapter has endeavored to present critical perspectives on the de-Westernization of digital politics by examining emerging narratives from the Global South. We argue that there is need to invest in research that helps us critique de-Westernization from a multidimensional perspective. De-Westernization can develop a tendency of being a catch-all phrase with no distinct meaning or boundaries. Our call is to encourage research-based understandings that deepen academic and societal knowledge on what de-Westernization entails. Without doubt, this presents a challenge as well as an opportunity for scholars to engage the visibly expanding theorization of this phenomenon. Part of the challenge in de-Westernizing digital politics is the fact that the broader global communication systems tend to be predominantly Western. The big Internet companies, social networking platforms and mass media outlets are driven by Western conglomerates with a worldview shaped by their environments and interests. The geopolitical interests of the Western powers are often reflected in how these mass communication platforms are structured and operate, and therein lies the problem for political communication scholars to deconstruct such Western centric notions. If their worldview is constricted by the Western media gaze of global realities, they are likely to be confined to Western geopolitical mappings of the world. Other nations located outside the Western polity are left out of critical policy and academic deliberations on digital technologies as we have argued in this chapter. There is therefore an opportunity for academia to challenge Western hegemony in the construction of the epistemology of digital politics as reflected by the dominance of Western journals and academic publishers as accepted knowledge dissemination platforms. It is important to challenge this dominant view which places Western epistemology at the core of theory that is then superimposed on Global South experiences. Resistance could come in many forms including the inclusion of non-Western

26  Handbook of digital politics

perspectives in important volumes such as this. Doing so makes such books relevant not only to Western universities but other knowledge-producing centers located in the Global South. It is also critical to encourage collaboration between non-Western and Western political communication scholars so that Global South experiences are not just used as case studies for clarifying, proving, or testing Western theory. Furthermore, the fragmentation of the broader Western hegemony as exemplified by the Covid-19 fallout internally and across geographies, Brexit, the Russia-Ukraine conflict and China’s growing financial muscle present opportunities to further probe the real meaning of “Western” and de-Westernization.

NOTES 1. For the concept of “coloniality of power” see Quijano (2000, 2007). 2. For more on decolonial thinking see Maldonado-Torres (2007, 2011, 2016, 2017), Mignolo and Walsh (2018), and Mignolo (2007, 2010, 2011). 3. This intergovernmental political forum convenes the seven (mainly Western) largest economies – Canada, France, Germany, Italy, Japan, the United Kingdom, the United States, with the European Union represented. 4. Adopted on January 31, 2015, the ‘Agenda 2063: The Africa We Want’ is a framework aiming to deliver concrete goals for inclusive and sustainable development (https://​au​.int/​ en/​agenda2063/​overview). 5. Central Africa has a population of around 150 million inhabitants and consists of mainly French-speaking countries but also Spanish and Portuguese – Burundi, Cameroon, Central African Republic, Chad, Republic of the Congo, Democratic Republic of the Congo, Equatorial Guinea, Gabon, and São Tomé and Principe. 6. Western Africa is composed of 16 countries: Benin, Burkina Faso, Cape Verde, The Gambia, Ghana, Guinea, Guinea-Bissau, Ivory Coast, Liberia, Mali, Mauritania, Niger, Nigeria, Senegal, Sierra Leone, and Togo. 7. Intermediary cities have a population of between 50,000 and one million people, and include small and mid-sized settlements or agglomerations that connect metropolitan and rural areas, as well as different groups of cities within urban systems. They play a major role in connecting rural and urban populations to basic facilities and services and relieving infrastructural pressures that can be endemic to megacities (Suri and Bonaglia, 2021). 8. In 2019, sub‑Saharan Africa was connected to the global telecommunications network via 18 active multilateral submarine cables, including eight on the west coast (Cariolle, 2021, p. 6). This expansion has led to a 5 percent increase in the Internet penetration rates in the region compared with the rest of the continent (AUC and OECD, 2021, p. 225).

FURTHER READING Chiumbu, S. (2016). Media, race and capital: A decolonial analysis of representation of miners’ strikes in South Africa. African Studies, 75(3), 417–435. Couldry, N. and Mejias, U. A. (2019). The Costs of Connection: How Data is Colonizing Human Life and Appropriating it for Capitalism. Stanford, CA: Stanford University Press. Langmia, K. and Lando, A. L. (eds.) (2020). Digital Communications at Crossroads in Africa: A Decolonial Approach. Cham: Springer International Publishing.

De-Westernizing digital politics: a Global South viewpoint  27

Mignolo, W. D. (2007). Introduction: Coloniality of power and de-colonial thinking. Cultural Studies, 21(2–3), 155–167. Mutsvairo, B. and Ragnedda, M. (eds.) (2019). Mapping the Digital Divide in Africa: A Mediated Analysis. Amsterdam: Amsterdam University Press. Ngomba, T. (2012). Circumnavigating de-Westernisation: Theoretical reflexivities in researching political communication in Africa. Communicatio, 38(2), 164–180.

REFERENCES Ahmed, K. (2022). Growing numbers of young Africans want to move abroad, survey suggests. The Guardian, 12 June. Atintande, M. (2020). Digital communication in Africa at crossroads: From physical exploitation in the past to virtual dominance now. In K. Langmia and A. L. Lando (eds.), Digital Communications at Crossroads in Africa: A Decolonial Approach (pp. 41–69). Cham: Springer International Publishing. AU (n.d.). Agenda 2063: The Africa We Want. African Union [Institutional website]. https://​ au​.int/​en/​agenda2063/​overview. AUC and OECD (2021). Africa’s Development Dynamics 2020: Digital Transformation for Quality Jobs. OECD. https://​doi​.org/​10​.1787/​0a5c9314​-en. Barclay, F. N. and Boobalakrishnan, N. (eds.) (2022). Social Media in India: Regulatory Needs, Issues and Challenges. London: Sage. Bornman, E., Harvey, J., Herman van Vuuren, J., Kekana, B., Matuludi, M. F., Mdakane, B. and Ramphele, L. (2021). Political engagement and opinions of youth in post-apartheid South Africa: A qualitative study. Politikon, 48(3), 372–390. Cariolle, J. (2021). International connectivity and the digital divide in Sub-Saharan Africa. Information Economics and Policy, 55, 100901. https://​doi​.org/​10​.1016/​j​.infoecopol​.2020​ .100901. Cariolle, J. and Goujon, M. (2019). Infrastructure et économie numérique en Afrique subsaharienne et dans l’UEMOA: état des lieux, acteurs, et nouvelles vulnérabilités. Note brève B186 [Policy Brief]. Fondation pour les études et recherches sur le développement international (Ferdi). https://​ferdi​.fr/​publications/​infrastructure​-et​-economie​-numerique​-en​ -afrique​-subsaharienne​-et​-dans​-l​-uemoa​-etat​-des​-lieux​-acteurs​-et​-nouvelles​-vulnerabilites. Coleman, S. (2017). Can the Internet Strengthen Democracy? Cambridge: Polity Press. Conroy-Krutz, J. (2020). Africans are concerned about ills of social media but oppose government restrictions. The Conversation, June 1. http://​theconversation​.com/​africans​-are​ -concerned​-about​-ills​-of​-social​-media​-but​-oppose​-government​-restrictions​-137653. Conroy-Krutz, J. and Koné, J. (2020). AD410: Promise and peril: In changing media landscape, Africans are concerned about social media but opposed to restricting access (Dispatch No. 410). Afrobarometer. http://​afrobarometer​.org/​publications/​ad410​-promise​ -and​-peril​-changing​-media​-landscape​-africans​-are​-concerned​-about​-social. Couldry, N. and Mejias, U. A. (2019). The Costs of Connection: How Data is Colonizing Human Life and Appropriating it for Capitalism. Stanford, CA: Stanford University Press. Dahlberg, L. (2011). Re-constructing digital democracy: An outline of four ‘positions’. New Media & Society, 13(6), 855–872. Davey, A. and Frenkel, S. (2019). Russia tests new disinformation tactics in Africa to expand influence. The New York Times, October 30. https://​www​.nytimes​.com/​2019/​10/​30/​ technology/​russia​-facebook​-disinformation​-africa​.html. Dwyer, M. and Molony, T. (eds.) (2019). Social Media and Politics in Africa: Democracy, Censorship and Security. London: Zed Books. Ephraim, P. E. (2013). African youths and the dangers of social networking: A culture-centered approach to using social media. Ethics and Information Technology, 15(4), 275–284.

28  Handbook of digital politics

Fidler, M. (2019). Disinformation Colonialism and African Internet Policy. Council on Foreign Relations, November 21. https://​www​.cfr​.org/​blog/​disinformation​-colonialism​-and​ -african​-internet​-policy. G7 (2019). Digital Transformation in Africa [Official document Annex 2]. G7 Summit. https://​www​.elysee​.fr/​admin/​upload/​default/​0001/​05/​54f72f2c31​6aab7cd837​73fde9e03d​ e3ab95b835​.pdf. Galtier, G. (2011). Internet, outil d’un nouveau discours identitaire soninké. Afrique contemporaine, 240(4), 149–151. Glück, A. (2018). De-Westernization and decolonization in media studies. In A. Glück (ed.), Oxford Research Encyclopedia of Communication. New York: Oxford University Press. Gore, C. D. (2023). The politics of the internet and social media in Africa: Three bases of knowledge for advancing research. Canadian Journal of African Studies, 57(1), 201–217. Hahn, H. P. and Kibora, L. (2008). The domestication of the mobile phone: Oral society and new ICT in Burkina Faso. The Journal of Modern African Studies, 46(1), 87–109. Iosifidis, P. and Nicoli, N. (2020). Digital Democracy, Social Media and Disinformation. New York: Routledge. Kelley, C. (2022). The Trump presidency: Democratic fatigue or fascism? Communication Research and Practice, 8(1), 4–18. Kouamé, Y. C. (2019). Analyse: La Fintech pour booster la finance inclusive en Afrique. EiC Corporation. AssoConnect. https://​www​.eic​-corporation​.org/​articles/​44183​-analyse​-la​ -fintech​-pour​-booster​-la​-finance​-inclusive​-en​-afrique. Maldonado-Torres, N. (2007). On the coloniality of being: Contributions to the development of a concept. Cultural Studies, 21(2–3), 240–270. Maldonado-Torres, N. (2011). Thinking through the decolonial turn: Post-continental interventions in theory, philosophy, and critique—an introduction. Transmodernity: Journal of Peripheral Cultural Production of the Luso-Hispanic World, 1(2), 1–15. Maldonado-Torres, N. (2016). Transdisciplinaridade e decolonialidade. Sociedade e Estado, 31(1), 75–97. Maldonado-Torres, N. (2017). On the coloniality of human rights. Revista Crítica de Ciências Sociais, 114, 117–136. Matiashe, F. S. (2020). Russian and French misinformation trolls have targeted African voters on Facebook. Quartz Africa, December 17. https://​qz​.com/​africa/​1947174/​facebook​-pulls​ -russian​-and​-french​-disinformation​-trolls​-in​-africa/​. Matthes, J. (2022). Social media and the political engagement of young adults: Between mobilization and distraction. Online Media and Global Communication, 1(1), 6–22. McPhail, T. L. (2014). eColonialism theory: How trends are changing the world. The World Financial Review, March 21. https://​worldfinancialreview​.com/​ecolonialism​-theory​-trends​ -changing​-world/​. Mignolo, W. D. (2007). Introduction: Coloniality of power and de-colonial thinking. Cultural Studies, 21(2–3), 155–167. Mignolo, W. D. (2010). Epistemic disobedience, independent thought and decolonial freedom. Theory, Culture & Society, 26(7–8), 159–181. Mignolo, W. D. (2011). The Darker Side of Western Modernity: Global Futures, Decolonial Options. Durham, NC: Duke University Press. Mignolo, W. and Walsh, C. E. (2018). On Decoloniality: Concepts, Analytics, Praxis. Durham, NC: Duke University Press. Mitchelstein, E., Matassi, M., and Boczkowski, P. J. (2020). Minimal effects, maximum panic: Social media and democracy in Latin America. Social Media + Society, 6(4), 1–11. Moyo, L. (2020). The Decolonial Turn in Media Studies in Africa and the Global South. Cham: Springer International. Musa, B. A. (2020). Africa at development policy and practice crossroads in the digital era: Navigating decolonization and glocalization. In K. Langmia and A. L. Lando (eds.), Digital

De-Westernizing digital politics: a Global South viewpoint  29

Communications at Crossroads in Africa: A Decolonial Approach (pp. 71–92). Cham: Springer International Publishing. Mutsvairo, B. and Moyo, L. (2022). Going beyond the digital divide debate: Critical reflections on the African media–economy matrix. In T. Flew, J. Holt, and J. Thomas (eds.), The SAGE Handbook of Digital Media Economy (pp. 325–341). London: Sage. Mutsvairo, B. and Ragnedda, M. (eds.) (2019). Mapping the Digital Divide in Africa: A Mediated Analysis. Amsterdam: Amsterdam University Press. Mutsvairo, B., Ragnedda, M., and Mabvundwi, K. (2022). ‘Our old pastor thinks the mobile phone is a source of evil’: Capturing contested and conflicting insights on digital wellbeing and digital detoxing in an age of rapid mobile connectivity. Media International Australia. OnlineFirst. Mutsvairo, B. and Salgado, S. (2021). Populism in Africa: Personalistic leaders and the illusion of representation. In H. Tumber and S. Waisbord (eds.), The Routledge Companion to Media Disinformation and Populism (pp. 101–120). London: Routledge. Ndlela, M. N. and Mano, W. (eds.) (2020). Social Media and Elections in Africa, Volume 1: Theoretical Perspectives and Election Campaigns. Cham: Palgrave Macmillan. Oser, J. and Boulianne, S. (2020). Reinforcement effects between digital media use and political participation: A meta-analysis of repeated-wave panel data. Public Opinion Quarterly, 84(S1), 355–365. Posetti, J. and Ireton, C. (2018). Journalism, Fake News and Disinformation: Handbook for Journalism Education and Training. Paris: UNESCO. Quijano, A. (2000). Coloniality of power and Eurocentrism in Latin America. International Sociology, 15(2), 215–232. Quijano, A. (2007). Coloniality and modernity/rationality. Cultural Studies, 21(2–3), 168–178. Richards, J. (2020). Going to Ghana: Black Americans explore identity living in Africa. https://​www​.nbcnews​.com/​news/​nbcblk/​going​-ghana​-black​-americans​-explore​-identity​ -living​-africa​-n1225646. Santos, F. O. dos (2021). Myths and misconceptions on Covid-19: ‘Congo Check’ and ‘Talato’ verification experiences. Frontiers in Communication, 6, 1–21. Srinivasan, S., Diepeveen, S., and Karekwaivanane, G. (2019). Rethinking publics in Africa in a digital age. Journal of Eastern African Studies, 13(1), 2–17. Stubbs, J. (2020). French and Russian trolls wrestle for influence in Africa, Facebook says. Reuters, December 15. https://​www​.reuters​.com/​article/​facebook​-africa​-disinformation​ -idINKBN28P261. Suri, S. N. and Bonaglia, F. (2021). Why local? Why now? Strengthening intermediary cities to achieve the SDGs. UN-Habitat, July 6. https://​unhabitat​.org/​why​-local​-why​-now​ -strengthening​-intermediary​-cities​-to​-achieve​-the​-sdgs. Vermeersch, E., Coleman, J., Demuynck, M., and Dal Santo, E. (2020). The Role of Social Media in Mali and its Relation to Violent Extremism: A Youth Perspective. ICCT: The International Centre for Counter-Terrorism Report. Yilmaz, I., Saleem, R. M. A., Pargoo, M., Shukri, S., Ismail, I., and Shakil, K. (2022). Religious Populism, Cyberspace and Digital Authoritarianism in Asia: India, Indonesia, Malaysia, Pakistan, and Turkey. European Center for Populism Studies (ECPS). https://​doi​ .org/​10​.55271/​rp0001.

3. Visual digital politics: imag(in)ing political activities and identities online Katy Parry

INTRODUCTION US Congresswoman Alexandra Ocasio-Cortez knows how to use interactive media platforms, whether streaming herself playing Among Us on her Twitch channel or answering questions on Instagram Live (D’Anastasio, 2020). In September 2021, Ocasio-Cortez uploaded an Instagram post in which she poses in her Met Gala dress, her back to the photographer as she looks over her shoulder at the camera; the white, otherwise-traditional gown displays ‘TAX THE RICH’ scrawled across the back in red capitals. Dress designer Aurora James is also pictured within the frame, holding the dress to ensure the key message is clear, and also looking directly at the camera. The first line of the caption reads: ‘The medium is the message’. This is almost too perfect a post for media and communication scholars. The phrase is of course media theorist Marshall McLuhan’s famous statement from his book Understanding Media: The Extensions of Man (1964), where he argues for close attention to each medium or new technology as ‘an extension of ourselves’, each affecting how we see the world around us. If we stay with McLuhan’s fairly loose understanding of ‘medium’ for now, and think about the languages or structures of the various ‘mediums’ at play here (which, we recall, act as extensions of ourselves), we can start with Ocasio-Cortez’s body, the vehicle used to wear the dress which is itself performing as a medium of communication. The dress carries the written slogan. This gown will later be worn to the Metropolitan Museum of Art (Met) Gala Ball and therefore extensively photographed and filmed by various international media at one of the renowned spectacles of the New York fashion calendar. The digital photograph, so simple to take on a smartphone, can be immediately uploaded to the next interrelated medium, the social media platform, Instagram, along with the typed written caption. The caption text emphasizes the identity of dressmaker Aurora James as a ‘Black woman immigrant designer’ who is working with Ocasio-Cortez to ‘kick open the doors at the Met’. As the image post is liked and commented upon by her millions of followers, it becomes viral, appearing across multiple global media channels. Why present this example as the opening vignette to this chapter? Because it speaks not only to the image management strategies of politicians in the digital era, whose tech-savviness and political authenticity are entwined with their use of visual imagery, but also to the blurring lines between official and unofficial visual repertoires. Ocasio-Cortez’s dress echoes the subversive ‘image politics’ (DeLuca, 1999) 30

Visual digital politics: imag(in)ing political activities and identities online  31

and embodied protests of activists and social movements: the white (almost bridal) gown clashing with red graffiti-style text. You could even argue it is a wearable rebuke to the jacket worn by Melania Trump during a trip to a migrant child detention centre in 2018 which had similar all-caps lettering: ‘I really don’t care, do u?’ (BBC, 2018). Ocasio-Cortez’s message is that she does care. Putting to one side the intersections of fashion and politics (Bartlett, 2019), Ocasio-Cortez is asserting a right to the personalized political spectacle that has long been criticized for debasing politics and the public sphere. She merges the entertainment of her political followers with the serious message that: ‘The time is now for childcare, healthcare, and climate action for all. Tax the Rich.’ Ocasio-Cortez knows that she will attract derision and cries of hypocrisy for her showy choices, sartorial and political, but this is a visually driven provocation aimed at those who routinely police her body and morals in media commentary. As an illustrative example, it both challenges and affirms some of the key characteristics of visual political communication in ways I go on to discuss in this chapter. Back in 2013, I was invited to write a chapter on visual politics online for the first edition of the Handbook of Digital Politics (Parry, 2015). In that chapter, I addressed the concerns about ‘visibility, vision and visuality in political communication and culture’. It felt pertinent then to write about the traditional suspicions and unease around the construction of political spectacles and other image-centred trends in mediated politics. Both socio-technological and scholarly developments over the past decade necessitate extensive revisions to this updated chapter. Social media platforms have become increasingly visually-led over this period, the rapidity and regularity of political memes in response to events has grown vastly, and the interdisciplinary interest in the role of visual images in global politics has likewise expanded. In this chapter I present a mapping of the field of visual digital politics, showing how both older and newer concerns about political aesthetics continue to be debated, how methods have evolved to better capture the ways in which new technologies shape political encounters, and how studies in political image-making are now abundant. Similar to the chapter for the first edition, I consider the visual imagery which is generated from both official and unofficial political realms, and indeed how those boundaries are becoming more difficult to draw. Journalists remain key players in the mediation of politics, but there is a diversification of image and knowledge producers in the contemporary mediascape, with increasing concerns over visual misinformation and unverifiable or uncredited images. Despite concerns over ‘deepfakes’, it is arguably the unguarded authentic image which continues to cause the most trouble for political leaders. The political functions of images mutate as the visual substance is re-mediated, recontextualized, and remixed for both serious and playful purposes. This chapter draws upon studies and cases from around the world to illustrate how the evidential power of the visual continues to offer a unique communicative force in politics, despite widespread knowledge of manipulation practices.

32  Handbook of digital politics

MAPPING THE FIELD OF VISUAL POLITICS: FROM THE PERIPHERAL TO ABUNDANCE There are broadly three avenues through which to chart the merging fields of the political, the visual, and communication. First, from the perspective of political communication and political studies, inquiries are often centred on how political encounters are shaped by the visual in various media forms. Doris Graber (1987) was an early innovator in pointing out how analysis of political television was incomplete without analysing the visual elements and their meaning-making capacity. Where democratic norms and opportunities for informed debate provide the normative underpinnings for investigations, the question becomes one of how the logic of the visual affects meaningful political action. Traditionally this has led to an ‘iconophobic’ strand of literature which is concerned with the ‘politics of ideals’ being replaced with the ‘politics of illusion’ (Barnhurst et al., 2004), where the televisual medium in particular is berated for selling news as a commodity to consumers rather than informing citizens. This tradition of concern around the distorting, seductive power of the visual is discussed in more detail in my earlier chapter (Parry, 2015). Despite persuasive insights from this earlier body of work responding primarily to television, it has been finessed and challenged by scholars this century, as the cultural, aesthetic and emotional turns across social sciences and humanities have led to more sustained interest in the intersections of popular culture and the political realm, enhancements of democratic life through more playful media genres, and how the affective dimensions of politics are part of a vibrant public sphere (Corner and Pels, 2003; Finnegan and Kang, 2004; Veneti et al., 2019). The ‘unproductive dichotomization’ of the ‘emotive visual’ against the ‘rational text’ (Parry, 2015), or the active citizen against the passive spectator, or indeed a narrow definition of political action which favours the official politics of institutions, can all work to oversimplify and negate the co-constitutive role of image and text, the symbolic and the imaginative, in interpreting our social worlds. It is also worth noting the advances from International Relations (IR), as opposed to studies more focused on national politics and political communication. Roland Bleiker has been a particularly influential thinker in opening up the field of global politics to the importance of aesthetic sources. Along with other (often Australia-based) scholars, Bleiker and collaborators have challenged disciplinary hostility to build a sub-field of ‘aesthetic politics’ that has burgeoned over the last few decades (Bleiker, 2012, 2015, 2021). In a recent piece, Bleiker (2021) reflects on the evolution of aesthetic approaches in IR and how developing innovative and creative modes of inquiry and analysis can struggle to gain legitimacy when disciplinary boundaries are conservatively policed. This academic ‘gatekeeping’ includes both cultural and structural barriers: through what is judged to be ‘proper’ rigorous research and valuable knowledge; in addition to funding, hiring and publishing decisions (Bleiker, 2021). This brief meta-narrative around the formation of the sub-field of ‘aesthetic politics’ is important to note for two reasons: it reminds us that whose knowledge and perspectives are valued is something we can actively shape in

Visual digital politics: imag(in)ing political activities and identities online  33

our own academic practices (including being alert to the Anglo-American predominance); and secondly, that truly interdisciplinary work that engages with alternative ways to better understand political realities can be rewarding but incredibly hard to achieve. As Bleiker (2021, p. 579) writes, aesthetic approaches to politics offer: a type of reflective understanding that emerges not from systematically applying the technical skills of analysis which prevail in the social sciences, but from cultivating a more open-ended level of sensibility about the political. This is why aesthetics is about far more than art: it is about the ability to step back, reflect and see political conflict and dilemmas in new ways.

The second field of study is one which takes for granted the value of aesthetic sources that Bleiker argues for above. For visual culture studies, artwork, film, and images are afforded representational complexity rather than associated with triviality, entertainment, and spectacle. The visual and symbolic are valued as worthy of close analytical attention on their own merits, and in their construction of the social and political world around us, not through a lens that is attuned to viewing the aesthetic as detrimental to political decision-making. Mieke Bal (2003, p. 19) sums it up thus: ‘Visual culture works towards a social theory of visuality, focusing on questions of what is made visible, who sees what, how seeing, knowing and power are interrelated’. To avoid reductively characterizing images as tools of manipulation, Mitchell (2002, p. 175) suggests treating visual images ‘as go-betweens in social transactions […] that structure our encounters with other human beings’, acknowledging that they are ‘the filters through which we recognize and of course misrecognize other people’. Finally, the third strand comes from the field of communication studies, albeit broadly defined and not necessarily straightforwardly distinguishable from the above two categories. It is worth noting that visual communication as a strand of communication studies has also emerged from interdisciplinary interests across anthropology, rhetoric, psychology, cultural studies and sociology aligning with approaches to photography, design and filmmaking (Pauwels and Mannay, 2020). Figures such as Walter Benjamin, Roland Barthes, Stuart Hall, Erwin Panofksy, and Susan Sontag, among others have provided some of the primary intellectual and philosophical foundations for those working with visual materials across the humanities and social sciences. In terms of mapping the field, then, in addition to the Barnhurst et al.’s (2004) mapping of ‘visual studies in communication’, mentioned above, which cites articles on political cartoons, advertising and the shaping of political perceptions, several reviews have more directly brought together the visual with political fields of research (Schill, 2012; Gerodimos, 2019). In his review of visual communication research within political communication, Dan Schill (2012, p. 119) wrote that: ‘The visual aspects of political communication remain one of the least studied and the least understood areas, and research focusing on visual symbols in political communication is severely lacking’. Schill offers an important overview of the function of visual symbols in politics, providing both an extensive literature review and a list of

34  Handbook of digital politics

functions of visual symbols for researchers to use and adapt. He also notes how this field of study is ‘often frustratingly complex and multidimensional’ with significant questions remaining about the normative implications of a visually dominant public sphere: ‘Are visual symbols better or worse for the public sphere than other forms of communication?’ (Schill, 2012, p. 134). He also highlights the neglect of audience research, where questions of how viewers process images are largely unanswered. Since Schill’s article there has indeed been a growth in research attempting to answer questions of how politicians’ visual strategies affect public interest and impressions, and the ‘severe lack’ or neglect of audience research is an argument that has become harder to make over the past decade (Lobinger and Brantner, 2015; Lindholm et al., 2021). Shifting back to the broader field of visual political communication, Bucy and Joo (2021, pp. 14–15) state: ‘As an area of study, visual politics is not just emerging – it is coming into its own’. We can briefly provide evidence for this in two ways: the emergence of edited collections and special issues in prestigious journals; and the significant growth in attention to visual politics across scholarly work more broadly. First, the interest paid to visual political communication in edited collections and special issues of journals has worked to enhance its position as worthy of serious study and to consolidate its theoretical and methodological underpinnings. Along with her co-editors, Anastasia Veneti is a driving force behind a number of these edited collections on visual political communication (Veneti et al., 2019; Lilleker and Veneti, 2023), including a collection focused on the Global South (Veneti and Rovisco, 2023), to complement a special issue on visual activism (Rovisco and Veneti, 2017). Other special issues on visual politics have appeared in the Cambridge Review of International Affairs (Crilley et al., 2020) and the International Journal of Press/Politics (Bucy and Joo, 2021), while the Journalism & Mass Communication Quarterly has also published a forum article on visual misinformation, social media and democracy (Dan et al., 2021), a growing area of concern across public communication scholarship. Other collections specifically on visual activism and protest include Aidan McGarry and co-editors’ The Aesthetics of Global Protest: Visual Culture and Communication (McGarry et al., 2020), to complement earlier special issues (for example, Doerr et al., 2013). Second, in term of sheer numbers, scholarly interest is clearly growing, and in a manner that transcends disciplinary boundaries. In my own very rough and ready search of the Web of Science website of academic articles, using the search term ‘visual’ AND ‘politic*’ appearing in the article abstract within the database, we can see a substantial increase in articles combining those two words (politic* is used to capture ‘politics’ and ‘political’). Including all topics or fields of research, most articles are categorized under ‘Communication’ (11.2 per cent), with ‘Humanities Multidisciplinary’ next (10.3 per cent). Possibly reflecting the traditional misgivings noted above, ‘Political Science’ is lower down the list, with 5.2 per cent (below ‘Art’, ‘History’ and ‘Cultural Studies’). Interestingly, no single discipline or subject area dominates at all in this list. Analysing the results via their publication year, we can see the number of articles soars from under 20 articles per year in 1997, to well

Visual digital politics: imag(in)ing political activities and identities online  35

over 500 per year by 2020. This could reflect the fact that more articles are being produced overall, due to the pressures to publish and growing numbers of journals. But if we use just one of the words, ‘politic*’, in the same kind of search, we see a tenfold increase in the number of articles over the same years (albeit in much larger numbers); whereas the more refined search for both ‘visual’ and ‘politic*’ appearing together in the abstract points to 25 times the number of articles in 2020 than in 1997. Admittedly this is a blunt measurement tool, but it reveals not only the growing abundance of articles which bring these two themes together, but the richness of the variety of approaches and disciplinary foci, whether in anthropological approaches to visual culture, cultural geography, visual news framing, visual narratives, digital-visual methods, or experimental studies examining effects on political knowledge. This list is taken from articles which all appeared on the first page of the search results, when sorted by ‘relevance’. Each sub-discipline outlined above is likely to hold a different emphasis in its identified research problems, or in its understanding of the visual in relation to the social world, but there are underlying persistent questions that recur across visual politics research: of how visuals construct meanings, and how they potentially persuade. How do visual qualities ‘work’ in the interests of their producers, and how might they ‘work’ on the viewer in different contexts?

WHERE ARE WE NOW? THREE PRIORITY AREAS FOR DIGITAL VISUAL POLITICS Where in earlier research the objects of study were television news, political cartoons, and newspaper photography, it is now the digital imagery on platforms such as news websites, image galleries, and social media that attract the lion’s share of attention for those working in the field of visual political communication. And social media imagery could refer to an array of visual content – screenshots, citizen witnessing, CCTV footage, news media videos, selfies, animations, GIFs – and so offers boundless possibilities for representing and supporting users’ politics, identities, values and morals (Frosh, 2019). Concentrating on digital visual politics, I note three priority areas: politician-focused visual imagery, from self-promotional to non-consensual; memes and participatory politics (from the mundane to the dangerously misleading or extreme); and finally, protest imagery. As noted at the start of the chapter, these distinctions are not always so clear-cut, but they are useful for organizing some of the key questions being posed in each area. The expansion of scholarly interest across these topics also requires a challenging degree of selectivity within this chapter. (Self-)Representation of Politicians As indicated in the opening section of the chapter, image management by politicians is no longer a matter of the occasional photo opportunity afforded by television or print media. The ‘fragility’ of ‘mediated visibility’ observed by John B. Thompson

36  Handbook of digital politics

(2005) has become an ever more delicate balance for politicians who hope to project their best qualities through social media accounts and websites, whilst avoiding the gaffes which can be edited and shared instantly. Online spaces provide eagerly monitored sites of contestation, of narratives and counter-narratives, publics and counter-publics: and if they started out as text-based platforms, they have become increasingly visually-led. Visual content has therefore become central to the digital communication strategies of politicians, attempting to create a coherent and authentic image across multiple platforms. Hashtags, emojis and tagging are part of these image-making practices, as Lalancette and Raynauld (2019) found in their study of Canadian Prime Minister Justin Trudeau’s use of Instagram. The personalization of politics often provides an analytical lens for studies of this nature. The content analysis of social media posts allows us to see the degree to which politicians merge their professional and personal lives in these intimate spaces: the values, qualities and popular cultural codes they highlight in their personal branding, and how other users and citizens respond in ‘likes’ and comments. We might expect politicians to promote themselves as trustworthy and honourable people. But the recent attention to populist communication styles suggests rule-breaking across aesthetics as well as democratic norms. A recent analysis of Brazilian President Jair Bolsonaro’s Instagram use shows how populists reject the more traditional leadership qualities as part of their transgressive appeal and their dismissal of the norms of representative politics. Mendonça and Caetano (2021, p. 213) find that Bolsonaro’s ‘eccentric rejection of basic social standards, over-the-top masculinity, and impromptu use of everyday objects as props work to construct an image that he is just an ordinary man, extraordinarily occupying the presidency’. Studies of this nature tend to use a mix of quantitative and qualitative content analysis of the social media posts to capture visual styles, themes and use of symbols (see also Uluçay and Melek, 2021). Gender-based differences in visual presentation have also attracted attention: for example, Bast et al.’s (2022) study focused on gender stereotypes on Instagram and also included an online experiment to gauge participants’ evaluation of traits such as ‘warmth’ and ‘competence’. Visual communication is at the heart of the political mediated persona in screen culture, whether by sophisticated or ironic design. But representations can also be non-consensual or aimed at exposure and scandal. The ‘Partygate’ scandal that occupied the UK government in winter 2021–2022 became energized through the release of videos and photographs that provided the evidential and newsworthy material to keep the story in the headlines. The scandal referred to gatherings that took place at 10 Downing Street and in other government buildings during the stringent Covid lockdown measures of 2020 and 2021, with the Metropolitan Police brought in to investigate in January 2022. Talk of 300 images being analysed by police and even the mere threat of more images being leaked unsettled those supportive of Prime Minister Boris Johnson, whose statements about his knowledge and participation in the parties revealed inconsistencies, to put it politely. Despite stating in the House of Commons before Christmas that he knew nothing of any of the parties and was ‘furious’, Johnson then defined them as ‘work events’ once it emerged that he had

Visual digital politics: imag(in)ing political activities and identities online  37

attended some of them. A photograph with a high vantage point looking down onto the Downing Street garden was released by the Guardian on 19 December 2021. It offered delicious details for those keen to dissect who was present alongside Johnson, the nature of relationships, the cheese and wine being consumed, whilst also causing anger for families who had been unable to see severely ill loved ones or attend funerals on 15 May 2020 when the photo was taken (Mason et al., 2021). As claims of more parties emerged, Johnson appeared to either not know the rules that he himself had set, or to have ignored them. Social media users responded with ridicule, arguably just as politically damaging as anger. Even airline company Ryanair’s official account tweeted an image-post inspired by the scandal, deploying lo-fi Internet aesthetics associated with 4Chan and Reddit. Johnson is depicted as ‘That Feel Guy’ standing awkwardly at a party (see Know Your Meme, n.d.). The fact that an airline company can attract ‘likes’ and retweets by ridiculing the prime minister and signalling their knowledge of a memetic in-joke via Twitter just goes to show the blurring boundaries between official and unofficial visual politics, where citizens, politicians, journalists and corporate digital communications workers participate in subversively humorous expression. It is to the participatory practice of memes that we next turn. Memes and Participatory Politics Meme cultures extend far beyond politics and the political, but the frequency and spread of politically themed memes means they are now difficult to avoid on many online social networks. Meme production by citizens, activists and political parties is especially intense around elections and scandals, but they are also undoubtedly part of everyday political conversations. As Jonathan Dean (2019) argues, political scientists would benefit from not only studying memes for whether they impact election outcomes, but for how they constitute political communities and contribute to the affective dynamics of political life. Visual images can serve to revive political participation through their associative, affective, creative and rhetorical appeals. For Limor Shifman, it is important to distinguish memetic texts from viral images: unlike an image which is simply shared, the memetic video ‘lures extensive creative user engagement in the form of parody, pastiche, mash-ups or other derivative work’ (2012, p. 190). Similar to the debates about soft news and political entertainment in the 1990s, it is often those citizens least likely to engage with official politics, especially young people, who are thought to find a sense of political belonging (and of their political adversaries) in contemporary memetic cultures. Of course, these activities are not always light-hearted in nature, and it is important to explore how symbols and flags can also become rallying features for violent and nationalistic movements. Another growing area of research concerns the intersections of populist, reactionary and far-right political formations, including the symbols and aesthetics employed to augment their tenets and mythologies in digital culture (Mortensen and Neumayer, 2021). Despite its creator Matt Furie’s objections, Pepe the Frog became an early icon of the alt-right, jumping from 4Chan to other platforms

38  Handbook of digital politics

and even embraced by Donald Trump during the 2016 US presidential election. Peters and Allan (2022) explore the ‘memetic weaponization’ of visual content and the critical role of journalists in explaining the contexts and public significance of memes such as Pepe, especially where hate-led agendas are normalized to create ‘us’ and ‘them’ binaries at the same time as claiming to be ‘just a bit of fun’. In the context of Middle East and North Africa (MENA), Moreno-Almeida and Gerbaudo (2021) examine how the Moroccan right are using Facebook meme pages to reshape local digital political landscapes, adapting far-right memes such as Pepe alongside symbols such as the Marinid flag to express pride in Moroccan identity and nostalgia for the Moorish Empire. As Mortensen and Neumayer (2021, pp. 2367–2368) point out, memes are currently ‘an inevitable and intrinsic part of visual communication in relation to political debate and conflict’, with humour and playfulness central to their uses, whether characterized as politicizing or depoliticizing, inclusionary or exclusionary. Visual Digital Activism There is a longer history of examining how mainstream media have covered protests, with the ‘protest paradigm’ an influential framework, and studies on the visual framing of protests showing how news images can work to marginalize dissent (Perlmutter and Wagner, 2004). But the continuing relevance of the news reporting paradigm has been questioned, as digital communication technologies enable an abundance of choice in media channels including alternative and activist media (Cottle and Lester, 2011). Sophisticated use of ‘image politics’ (DeLuca, 1999) includes the creation of humorous or compelling artwork, often merging the DIY aesthetic of homemade banners with digital branding freely available to download (see for example Extinction Rebellion’s website for materials). Crucially it is the interplay of the embodied demonstrations on the street with image-making practices across hybrid media forms that builds support, amplifies the message, and forms collective identities across borders. For those who are physically present on the ground, images of protests, vigils, or police brutality can be shared instantly during such events as they unfold. Digital pictures can also be easily edited together, have music or captions added (often in English), to produce cultural artefacts designed to attract attention beyond the immediacy of citizen or activist witnessing. This is thought to be especially transformative for diasporic communities or transnational protest movements whose supporters are able to express their solidarity via social media despite geographic distance. The concept of the carnivalesque has become particularly prominent in writing on protest repertoires, with the subversive humour and transgressive power-play of the carnival embraced through theatrical performances, colourful banners and costumes. The affective moods of both joy and rage are harnessed in street protests where the mischief-making of the festival-like gatherings becomes amplified through hybrid media forms. Janjira Sombatpoonsiri (2021) uses the example of Thailand’s Red Sunday group to demonstrate how the carnivalesque process works, and argues that anger-filled protests can be

Visual digital politics: imag(in)ing political activities and identities online  39

counter-productive, whereas more ‘friendly’ carnivalesque humour can be used to sustain public support: ‘Because of absurd and at times jocular features of the activities, fun displaced rage in this emotive space’ (Sombatpoonsiri, 2021, p. 10). Yet visuals which capture rage and anger are also central to garnering wider public support. The Black Lives Matter movement would not have gained such international prominence without mobile phone imagery and networked technologies. The fact that many of the cases of police brutality and murder became public knowledge in the first place due to mobile phone footage highlights how citizen-produced imagery is central to the cause. As others have noted, the very nature of police brutality as an issue is well-suited to an Internet-based campaign: ‘Unlike wealth or income inequality, police brutality is concrete, discrete in its manifestations, and above all, visual’ (Freelon et al., 2016, p. 82). In addition to those harrowing images which serve as evidence of injustice, the street protests and murals have also led to the wide circulation of photographs deemed ‘iconic’ due to their rhetorical power in contemporary discourses about race (Edrington and Gallagher, 2019; Aiello and Parry, 2020). The ‘new kid on the block’ in social media and politics at the time of writing is TikTok. Where research had largely focused on Instagram, Facebook, YouTube and Twitter, we now see a shift to TikTok activism. TikTok’s popularity soared during the first year of the Covid pandemic in 2020, appealing to age groups outside its core demographic of young people or ‘Gen Z’. When it comes to TikTok activism, interests often merge around young people’s own social media practices and causes. Hautea et al. (2021, p. 1) examine how young non-experts, grappling with ‘imperfect understandings and unpolished messaging techniques’, nevertheless spread the message that people care about climate change, noting the importance of the platform’s unique ‘affective affordances’, and the reshaping of publics through ‘affective contagion’ (Papacharissi, 2015). As with everyday politics outlined above, it is the memetic qualities and affective appeals, alongside the performance-centred genres of TikTok which intrigue scholars. Visually innovative TikTok genres are shaped by users’ practices and the functionalities of the platform which encourage intertextual borrowing and remixing with music, sound and images. The gestures, facial expressions, and posture of the human body on display are undoubtedly crucial elements for visual analysis, alongside the emojis, flags, text, and ‘stitching’ in dialogue with another video.

CONCLUSIONS As Bucy and Joo (2021, p. 15) write: ‘The scholarly interest in visual politics is palpable’. This statement has undeniably superseded earlier laments of its neglect. It also means that it is impossible to include all the innovative and consolidatory work in this chapter. One area I’ve not covered in the chapter is the continuing interest in the visual framing of issues and events (beyond protest), now focused on news media websites and their social media accounts, but often drawing upon Grabe and Bucy’s (2009)

40  Handbook of digital politics

earlier framework on ‘image bite politics’ in television news coverage of elections. Visual political news coverage remains a significant area of research, especially as global media outlets adapt to power shifts in the photojournalistic industry along with post-Covid international political relations (El Damanhoury and Garud-Paktar, 2022). The Covid-19 pandemic generated novel visual representations on an unprecedented global scale. News producers attempted to capture its ‘essence’ and ‘deep meaning’ through varied visual genres (Sonnevend, 2020), while political leaders had to stage their visual diplomacy through the ‘virtual summitry’ of online meetings (Danielson and Hedling, 2022). Continuing to underpin such studies is the belief that the necessarily selective mediated images are integral to shaping what is deemed important, appropriate and imaginable. I have also not dealt in detail with a group of studies which attempt to evaluate the impact of selected visual images or their effects on audience engagement. Such work tends to focus on measuring how people respond to certain images of political candidates, and often uses innovative technologies such as eye-tracking or computer vision techniques (Lindholm et al., 2021). Its neglect in this chapter possibly betrays my own humanities-led sensibility. Whilst experimental studies are part of a useful set of tools for distinguishing how people respond to selected modes of communication, my personal concern is that they tie researchers to a restrictive notion of the role of images in political life. The complexity and diversity of visual politics requires a range of theoretical and methodological knowledge to understand; as Bleiker (2015, p. 889) writes, ‘how images frame the conditions of possibility; how they influence what can and cannot be seen, thought and discussed; in short, how they delineate and shape the political’. In their special issue introduction, Bucy and Joo (2021, p. 9) argue that the complexities of contemporary visual politics necessitate collaboration across areas of expertise. For example, in response to the visual and symbolic forms of hate adopted by extremist groups, they suggest: a combined team of ethnographers to study the culture that produces and puts these signs into circulation, coders and computational scholars to identify and track them, network scientists to analyze the algorithms that accelerate and amplify their reach, and ethicists to describe the wider social implications of a media and political system that tolerates symbolic attacks on entire classes of people.

This speaks to the difficulty of the task ahead, and follows others’ calls for multi-disciplinarity: for a mixing of qualitative and quantitative methods, bringing together different logics and even incompatible ways of knowing as an ‘assemblage’ or ‘loose network of methodological connections’ (Bleiker, 2015, p. 883). We cannot investigate visual images out of context or without reference to the interplay with other modes of communication (text, music, sounds, etc.). I cited Mitchell in the earlier edition of this chapter and his argument still holds true: ‘the opening out of a general field of study does not abolish difference, but makes it available for investigation, as opposed to treating it as a barrier that must be policed and never crossed’

Visual digital politics: imag(in)ing political activities and identities online  41

(Mitchell, 2002, p. 173). New visual styles and strategies, new platforms and new genres all require careful attention, with different media formats analysed according to their own rhetorical functions or semiotic resources. Viral iconic images, memes, TikTok videos and GIFs operate across diverse contexts, encompassing different aesthetic strategies and cultural practices, serving different purposes and interests (Shifman, 2012; Miltner and Highfield, 2017). Combining recognition of specificity when it comes to multimodal formats, with curiosity and critique, informed and energized by multidisciplinary forms of seeing and knowing, will ensure the continued health of this burgeoning field of research. Mitchell’s famous claim that ‘there are no visual media’ is truer still in the smartphone age where visual objects and symbols are encountered as sensory experiences via devices which really have become ‘extensions’ of ourselves (McLuhan, 1964; Mitchell, 2005). Understanding how citizens relate to politics and ‘the political’ requires paying close attention to the structures and practices of image-making across a multiplicity of media forms.

FURTHER READING Bleiker, R. (2021). Seeing beyond disciplines: Aesthetic creativity in international theory. Australian Journal of International Affairs, 75(6), 573–590. Bucy, E. P. and Joo, J. (2021). Editors’ introduction: Visual politics, grand collaborative programs, and the opportunity to think big. International Journal of Press/Politics, 26(1), 5–21. Edrington, C. L. and Gallagher, V. J. (2019). Race and visibility: How and why visual images of black lives matter. Visual Communication Quarterly, 26(4), 195–207. Gerodimos, R. (2019). The interdisciplinary roots and digital branches of visual political communication research. In A. Veneti, D. Jackson and D. G. Lilleker (eds.), Visual Political Communication (pp. 53–73). Cham: Palgrave Macmillan. Hautea, S., Parks, P., Takahashi, B., and Zeng, J. (2021). Showing they care (or don’t): Affective publics and ambivalent climate activism on TikTok. Social Media + Society, 7(2). https://​doi​.org/​10​.1177/​20563051211012344 Mortensen, M. and Neumayer, C. (2021). The playful politics of memes. Information, Communication & Society, 24(16), 2367–2377.

REFERENCES Aiello, G. and Parry, K. (2020). Visual Communication: Understanding Images in Media Culture. London: Sage. Bal, M. (2003). Visual essentialism and the object of visual culture. Journal of Visual Culture, 2(1), 5–32. Barnhurst, K. G., Vari, M., and Rodríguez, Í. (2004). Mapping visual studies in communication. Journal of Communication, 54(4), 616–644. Bartlett, D. (ed.) (2019). Fashion and Politics. New Haven: Yale University Press. Bast, J., Oschatz, C., and Renner, A.-M. (2022). Successfully overcoming the “double bind”? A mixed-method analysis of the self-presentation of female right-wing populists on Instagram and the impact on voter attitudes. Political Communication, 39(3), 358–382.

42  Handbook of digital politics

BBC (2018). Melania Trump says ‘don’t care’ jacket was a message. BBC News, 14 October. https://​www​.bbc​.co​.uk/​news/​world​-us​-canada​-45853364. Bleiker, R. (2012). Aesthetics and World Politics. Basingstoke: Palgrave Macmillan. Bleiker, R. (2015). Pluralist methods for visual global politics. Millennium, 43(3), 872–890. Bleiker, R. (2021). Seeing beyond disciplines: Aesthetic creativity in international theory. Australian Journal of International Affairs, 75(6), 573–590. Bucy, E. P. and Joo, J. (2021). Editors’ introduction: Visual politics, grand collaborative programs, and the opportunity to think big. International Journal of Press/Politics, 26(1), 5–21. Corner, J. and Pels, D. (eds.) (2003). Media and the Restyling of Politics. London: Sage. Cottle, S. and Lester, L. (eds.) (2011). Transnational Protests and the Media. New York: Peter Lang. Crilley, R., Manor, I., and Bjola, C. (2020). Visual narratives of global politics in the digital age: An introduction. Cambridge Review of International Affairs, 33(5), 628–637. D’Anastasio, C. (2020). Alexandria Ocasio-Cortez storms Twitch. Wired, 28 October. https://​ www​.wired​.com/​story/​aoc​-among​-us​-twitch​-stream/​. Dan, V., Paris, B., Donovan, J., Hameleers, M., Roozenbeek, J., van der Linden, S., and von Sikorski, C. (2021). Visual mis- and disinformation, social media, and democracy. Journalism & Mass Communication Quarterly, 98(3), 641–664. Danielson, A. and Hedling, E. (2022). Visual diplomacy in virtual summitry: Status signalling during the coronavirus crisis. Review of International Studies, 48(2), 243–261. Dean, J. (2019). Sorted for memes and gifs: Visual media and everyday digital politics. Political Studies Review, 17(3), 255–266. DeLuca, K. (1999). Image Politics. New York: Guilford Press. Doerr, N., Mattoni, A., and Teune, S. (eds.) (2013). Advances in the Visual Analysis of Social Movements. Bingley: Emerald Group. Edrington, C. L. and Gallagher, V. J. (2019). Race and visibility: How and why visual images of black lives matter. Visual Communication Quarterly, 26(4), 195–207. El Damanhoury, K. and Garud-Paktar, N. (2022). Soft power journalism: A visual framing analysis of COVID-19 on Xinhua and VOA’s Instagram pages. Digital Journalism, 10(9), 1546–1568. Finnegan, C. A. and Kang, J. (2004). “Sighting” the public: Iconoclasm and public sphere theory. Quarterly Journal of Speech, 90(4), 377–402. Freelon, D., McIlwain, C. D., and Clark, M. D. (2016). Beyond the Hashtags: #Ferguson, #Blacklivesmatter, and the Online Struggle for Offline Justice. Centre for Media and Social Impact, 29 February. http://​cmsimpact​.org/​resource/​beyond​-hashtags​-ferguson​ -blacklivesmatter​-online​-struggle​-offline​-justice/​. Frosh, P. (2019). The Poetics of Digital Media. Cambridge: Polity Press. Gerodimos, R. (2019). The interdisciplinary roots and digital branches of visual political communication research. In A. Veneti, D. Jackson and D. G. Lilleker (eds.), Visual Political Communication (pp. 53–73). Cham: Palgrave Macmillan. Grabe, M. E. and Bucy, E. P. (2009). Image Bite Politics: News and the Visual Framing of Elections. New York: Oxford University Press. Graber, D. A. (1987). Kind words and harsh pictures: How television presents the candidates. In K. L. Schlozman (ed.), Elections in America (pp. 115–141). Winchester, MA: Allen & Unwin. Hautea, S., Parks, P., Takahashi, B., and Zeng, J. (2021). Showing they care (or don’t): Affective publics and ambivalent climate activism on TikTok. Social Media + Society, 7(2). https://​doi​.org/​10​.1177/​20563051211012344. Know Your Meme (n.d.). I Wish I Was At Home / They Don’t Know. Know Your Meme. https://​knowyourmeme​.com/​memes/​i​-wish​-i​-was​-at​-home​-they​-dont​-know.

Visual digital politics: imag(in)ing political activities and identities online  43

Lalancette, M. and Raynauld, V. (2019). The power of political image: Justin Trudeau, Instagram, and celebrity politics. American Behavioral Scientist, 63(7), 888–924. Lilleker, D. and Veneti, A. (eds.) (2023). Research Handbook on Visual Politics. Cheltenham, UK and Northampton, MA, USA: Edward Elgar Publishing. Lindholm, J., Carlson, T., and Högväg, J. (2021). See me, like me! Exploring viewers’ visual attention to and trait perceptions of party leaders on Instagram. International Journal of Press/Politics, 26(1), 167–187. Lobinger, K. and Brantner, C. (2015). Likable, funny or ridiculous? A Q-sort study on audience perceptions of visual portrayals of politicians. Visual Communication, 14(1), 15–40. Mason, R., Stewart, H., and Walker, P. (2021). Boris Johnson and staff pictured with wine in Downing Street garden in May 2020. The Guardian, 19 December. https://​ www​.theguardian​.com/​politics/​2021/​dec/​19/​boris​-johnson​-and​-staff​-pictured​-with​-wine​-in​ -downing​-street​-garden​-in​-may​-2020. McGarry, A., Erhart, I., Eslen-Ziya, H., Jenzen, O., and Korkut, U. (eds.) (2020). The Aesthetics of Global Protest: Visual Culture and Communication. Amsterdam: Amsterdam University Press. McLuhan, M. (1964). Understanding Media: The Extensions of Man. New York: Signet Books. Mendonça, R. F. and Caetano, R. D. (2021). Populism as parody: The visual self-presentation of Jair Bolsonaro on Instagram. International Journal of Press/Politics, 26(1), 210–235. Miltner, K. M. and Highfield, T. (2017). Never gonna GIF you up: Analyzing the cultural significance of the animated GIF. Social Media + Society. https://​doi​.org/​10​.1177/​ 2056305117725223. Mitchell, W. J. T. (2002). Showing seeing: A critique of visual culture. Journal of Visual Culture, 1(2), 165–181. Mitchell, W. J. T. (2005). There are no visual media. Journal of Visual Culture, 4(2), 257–266. Moreno-Almeida, C. and Gerbaudo, P. (2021). Memes and the Moroccan far-right. International Journal of Press/Politics, 26(4), 882–906. Mortensen, M. and Neumayer, C. (2021). The playful politics of memes. Information, Communication & Society, 24(16), 2367–2377. Papacharissi, Z. (2015). Affective Publics: Sentiment, Technology, and Politics. Oxford: Oxford University Press. Parry, K. (2015). Visibility and visualities: ‘Ways of seeing’ politics in the digital media environment. In S. Coleman and D. Freelon (eds.), Handbook of Digital Politics (pp. 417–432). Cheltenham, UK and Northampton, MA, USA: Edward Elgar Publishing. Pauwels, L. and Mannay, D. (eds.) (2020). The SAGE Handbook of Visual Research Methods, 2nd edition. London: Sage. Perlmutter, D. D. and Wagner, G. L. (2004). The anatomy of a photojournalistic icon: Marginalization of dissent in the selection and framing of ‘a death in Genoa’. Visual Communication, 3(1), 91–108. Peters, C. and Allan, S. (2022). Weaponizing memes: The journalistic mediation of visual politicization. Digital Journalism, 10(2), 217–229. Rovisco, M. and Veneti, A. (eds.) (2017). Special issue: Picturing protest – visuality, visibility and the public sphere. Visual Communication, 16(3) (various relevant articles included: http://​journals​.sagepub​.com/​toc/​vcja/​16/​3). Schill, D. (2012). The visual image and the political image: A review of visual communication research in the field of political communication. Review of Communication, 12(2), 118–142. Shifman, L. (2012). An anatomy of a YouTube meme. New Media & Society, 14(2), 187–203. Sombatpoonsiri, J. (2021). Carnivalesque humor, emotional paradoxes, and street protests in Thailand. Diogenes, 63(1–2). https://​doi​.org/​10​.1177/​0392192120970409. Sonnevend, J. (2020). A virus as an icon: The 2020 pandemic in images. American Journal of Cultural Sociology, 8, 451–461.

44  Handbook of digital politics

Thompson, J. B. (2005). The new visibility. Theory, Culture & Society, 22(6), 31–51. Uluçay, D. M. and Melek, G. (2021). Self-presentation strategies and the visual framing of political leaders on Instagram: Evidence from the eventful 2019 Istanbul mayoral elections. Visual Communication. https://​doi​.org/​10​.1177/​14703572211057595. Veneti, A., Jackson, D., and Lilleker, D. (eds.) (2019). Visual Political Communication. Cham: Palgrave Macmillan. Veneti, A. and Rovisco, M. (eds.) (2023). Visual Politics in the Global South. Cham: Palgrave Macmillan.

4. Revolution vs reaction: the role of social media in authoritarian regimes Anna Litvinenko

Over the past decade, authoritarian rulers around the globe have mastered mechanisms of control and co-optation of digital technologies. The liberation promise of social media celebrated by scholars after the Arab Spring (Diamond, 2012) was choked in many countries by elaborate Internet censorship and domination of pro-state discourse online. According to V-Dem data, the trend to autocratisation has increased over the past ten years, and today 68 per cent of the world population live in autocracies (Alizada et al., 2021). As of the beginning of the 2020s, the ‘repression’ arguments seem to have prevailed in the ‘liberation- vs repression-technology’ debate about the societal impact of ICTs (Deibert and Rohozinski, 2010; Rød and Weidmann, 2015; Keremoğlu and Weidmann, 2020). The case of authoritarian Russia is exemplary for this development. The protests ‘For Fair Elections’ in 2011–2012 demonstrated the mobilization power of social media (Bodrunova and Litvinenko, 2013; Reuter and Szakonyi, 2015). The state reacted by consequent suppression of Internet freedom, which culminated in the law on sovereign Internet in 2019 (Litvinenko, 2021). In 2022, during the full-scale invasion of Ukraine, the Russian state already had a wide range of possible tools of Internet control at hand in order to ensure military censorship of the Web. Moreover, the state propaganda in both traditional and online media seemed to have made the majority of the population immune to alternative sources of information, although they still had access to global social media platforms. How could a media system that over decades had had a substantial share of leadership-critical publics (Toepfl, 2020) be almost completely sterilized in such a short period of time, without major protests? How does digital connectivity work for disconnecting and disempowering people? Similar questions arise when we look at cases of other authoritarian regimes, like Egypt or Turkey. The protests on Tahir Square in 2011 and in Gezi Park in 2013 had a spirit of people demanding dignity and recognition similar to the protests ‘For Fair Elections’ in Russia. They employed similar strategies of political mobilization via social media (Tufekci, 2017). What came afterwards as a reaction of the above-mentioned states also had obvious parallels. Since 2013, we have observed, in both Egypt and Turkey, a restricting of Internet legislation, banning of independent media, and harsh penalties for political dissidents, which boosts self-censorship and fosters domination of pro-state narratives in public communication. At the same time, we observe cases divergent from this pattern, for instance in Tunisia, where after the Arab Spring at least partial structural transformation took place (Richter and Kozman, 2021). Hasty generalizations might create false expectations. Thus, when at the beginning of 2022, 45

46  Handbook of digital politics

popular protests were severely suppressed in Kazakhstan, many analysts awaited a repetition of the Belarusian scenario, where President Lukashenka brutally erased all kinds of dissent after the protests of 2020. Instead, at least in the immediate phase after the protests in Kazakhstan, we observed the introduction of a constitutional reform to limit the power of the president (RFERL, 2022), with some analysts even talking about a ‘perestroika’ spirit in the country (Turkstra, 2022). This proves yet again that trajectories of authoritarian regimes and of their social movements, even within one region, might diverge significantly (Geddes, 1999). The series of revolutions followed by reactionary waves in authoritarian states in the past decade have demonstrated that long-term observations are needed to better assess the roles and effects of social media in authoritarian settings. In this chapter, I present an overview of the studies on the topic, structured according to two major strands of research: (1) on co-optation of social media by states; and (2) on the use of social media by civil society actors. Subsequently, I define gaps and outline promising paths for future research in this area.

CONTROL AND CO-OPTATION: HOW AUTHORITARIAN RULERS USE SOCIAL MEDIA At the early stages of the development of social media, a laissez-faire attitude towards the Internet was widespread among autocrats (Tufekci, 2017). Many of them apparently underestimated the mobilization power of the Web and, as a result, tolerated a high level of Internet freedom. For instance, in Russia in 2011, at the beginning of the movement ‘For Fair Elections’, pro-state media would call liberal intellectuals ‘Facebook hamsters’. The name expressed the disparaging attitude to online protesters, who, in the view of authoritarian elites, were in fact harmless. However, at the end of 2011, online dissent did spill over to the streets and grew into the most powerful protest wave of Putin’s era. One of the famous rally posters was a satirical answer to the state’s mocking of online dissidents: ‘The hamster shrugged’. The state reacted to this revelation of the mobilization power of social media with a series of restrictive Internet laws as well as with sophisticated mechanisms for the co-optation of ICTs (Litvinenko and Toepfl, 2019). A similar attitude change could be observed in Turkey, where after years of rather half-hearted censorship measures, the 2013 Gezi Park protest marked a milestone in Erdoğan’s Internet policies. An illustration of this turn was the demonizing campaign against Twitter that culminated in its banning in 2014 (Tufekci, 2017). Bans and restrictions, that is, the first and the second generations of Internet control, were accompanied by co-optation measures, the so-called third generation of Internet control (Deibert, 2015), that helped foster regime stability. The shift to these more sophisticated control mechanisms was reflected in the growing body of research on authoritarian co-optation of social media (Pearce and Kendzior, 2012; Rød and Weidmann, 2015; Toepfl, 2020). Prior to the Arab Spring, such studies were rather rare and mostly focused on China, which was leading the

Revolution vs reaction: the role of social media in authoritarian regimes  47

way in state control over the Internet (Paltemaa and Vuori, 2009; MacKinnon, 2011). Most of the few existing studies on the effects of media use in authoritarian settings draw on data from China (Tang and Huhe, 2014; Hyun and Kim, 2015; Stockmann et al., 2020). They demonstrate a two-sided picture: on the one hand, there is some evidence that exposure to political news online might lead to a more negative evaluation of government performance (Tang and Huhe, 2014) and to the strengthening of civic culture under authoritarianism (Li et al., 2016); on the other hand, social media use overall positively correlates with regime support (Hyun and Kim, 2015; Li et al., 2016). The Chinese authoritarian regime has obviously mastered the art of what Schedler calls ‘institutional gardening’ (Schedler, 2009). This concept describes the process of a balanced use of ‘input institutions’ (Lagacé and Gandhi, 2015) in authoritarian settings, namely by calibrating them based on assessments of their benefits as well as risks for the regime. Social media can be considered an ‘input institution’ alongside authoritarian elections and traditional media, as they might serve as an important feedback mechanism that helps autocrats fine-tune their policies (Toepfl, 2020). However, the benefit of tolerating a certain amount of free speech online comes with a major risk of political dissent. Toepfl (2020) distinguishes three major types of publics in authoritarian regimes with regard to their approach to tolerating criticism in public communication: uncritical (North Korea, Turkmenistan), policy-critical (China, Iran), leadership-critical (Russia, Belarus, as of 2020). The post-pandemic ranking of press freedom shows a major decline in the freedom of speech in both Russia and Belarus, which can no longer be considered leadership-critical publics (RSF, 2022). Apparently, autocrats in these countries assessed the risks of tolerating critical political talk higher than its benefits. The overall shift to more restrictive Internet policies by authoritarian regimes during the pandemic and its effects for global authoritarian diffusion are yet to be explored in future research. Numerous studies have analysed how autocrats use social media for control and censorship as well as for the legitimization of their rule (Pearce and Kendzior, 2012; Lewis, 2016; Soest and Grauvogel, 2017; Paltemaa et al., 2020). Different types of covert and overt measures were discussed in connection with China (Paltemaa et al., 2020; Roberts, 2018), the post-Soviet space (Pearce and Kendzior, 2012; Toepfl, 2020), the MENA region (Sika, 2019; Uniacke, 2021), and other areas. King et al. (2013) analysed the content filtering of social media posts on the Chinese Internet and found that, contrary to expectations, not all state critical posts were censored, but rather those which contained references to collective action. The rare comparative studies reveal certain patterns in the employment of censorship measures across different authoritarian contexts. Hellmeier (2016) analysed Internet filtering cases across 34 autocratic regimes and came to the conclusion that certain regime types, for instance, monarchies or regimes experiencing a high level of social unrest, are more likely to filter Internet content. Toepfl and Litvinenko (2021) compared the levels of criticism in the comment sections of major networking platforms in three authoritarian post-Soviet states: Belarus, Azerbaijan, and Turkmenistan. They found that levels of tolerated criticism systematically differed depending on the openness of

48  Handbook of digital politics

the regime. Sinpeng (2020) focused on Southeast Asia, comparing four authoritarian regimes with different levels of Internet penetration, and found their Internet control strategies very similar. Differences were observed in the issues that usually caused censorship in a country. For instance, in Singapore, the main censored issues included criticism of authorities, social commentary, and satire, whereas in Indonesia, these were also corruption, LGBTQ-related topics, and blasphemy (Sinpeng, 2020, p. 36). What exactly the red lines are for different regimes and how autocrats ensure that these lines are not crossed by users in online communication are intriguing questions. They are connected to the bigger question of how exactly today’s autocrats shape discourse that legitimizes their rule. Although this field of research has been assessed by scholars as very promising in studies of today’s authoritarianism (Lewis, 2016; Crilley and Chatterje-Doody, 2021), the aspect of discourse construction in research of authoritarian (social) media has remained under-explored. Studies of authoritarian propaganda have shed light on some specific features of this discourse, such as depoliticization (Uniacke, 2021), spreading of fear, flooding with distracting content (Roberts, 2018), and producing ‘resignation, cynicism, and a sense of disempowerment’ (Tufekci, 2017, p. 228). In their study of the so-called 50 cent party, King et al. (2017) found that the strategy of the paid commentators was not to fight sceptics, but rather to distract the public from problematic issues and instead promote messages cheerleading the regime. In their study of memory narratives about commemorations of the October 1917 revolution in Russian online publics, Litvinenko and Zavadski (2020) came to the conclusion that the Russian state promoted different narratives on the same topic for various target groups, and the resulting patchwork of narratives with ‘memories on demand’ (Litvinenko and Zavadski, 2020, p. 1657) fostered societal fragmentation. Asmolov (2018) explored the practice of unfriending on social networks as an effect of disinformation campaigns of the Russian state, and suggested that disruption of horizontal ties – disconnection – was one of the main goals of propaganda, helping the state construct the image of an external enemy. The use of social media for disinformation campaigns by authoritarian states has grown into an important strand of research since the scandal over Russian interference in the US elections in 2016. The Oxford Internet Institute issues newsletters featuring research on computational propaganda, with special focus on China and Russia. Indeed, the biggest share of research on foreign influence is focused on Russian disinformation campaigns (Aro, 2016; Boyte, 2017; Freelon and Lokot, 2020). Several studies explored cross-platform use of computational propaganda (Howard et al., 2018; Lukito, 2020). Lukito (2020) suggested that Reddit was used by Russia’s Internet Research Agency, also known as a ‘troll fabric’, as a ‘testing ground’ for content that was subsequently spread via Twitter. Bolsover and Howard (2019) compared the use of automated propaganda on Twitter and Weibo, and found less automated content on Weibo compared to Twitter and overall little evidence of the use of automated accounts by the Chinese state, whereas a large amount of automated Twitter content was actually countering the Chinese narratives in simplified Mandarin. The fact that authoritarian states tend to massively use paid commenters

Revolution vs reaction: the role of social media in authoritarian regimes  49

alongside automated accounts makes it challenging to identify them (King et al., 2017). The role of platforms in authoritarian resilience has been addressed in many studies; however, platforms only rarely become the focus of studies in this research area. Several studies investigated the impact of search engines on shaping discourse on certain events under authoritarianism (Zavadski and Toepfl, 2019; Paltemaa et al., 2020). Kravets and Toepfl (2022) demonstrated reference and source bias of algorithms of the Russian-based search engine Yandex, compared to its Google counterpart. Paltemaa and colleagues (2020, p. 2064) called this type of bias ‘meta-information censorship’ and demonstrated that filtering on the search engine Baidu might also influence search results on the Chinese version of Google, spreading the effect of censorship far beyond the national borders of China. As the past decade has shown, authoritarian countries have become very fast and effective at co-opting digital technologies and learning from each other (Hall and Ambrosio, 2017). The mechanisms of diffusion of authoritarian practices and in particular the role of global online platforms therein are still to be better explored, as they not only contribute to authoritarian resilience worldwide, but also infuse illiberal movements in democracies (Michaelsen and Glasius, 2018).

‘BE WATER’: DIGITAL ACTIVISM ADAPTS TO INCREASINGLY RESTRICTIVE ENVIRONMENTS In 2019, as the major wave of anti-government protests erupted in Hong Kong, the Western press was filled with headlines containing the phrase ‘Be water’, the guiding principle of protesters borrowed from the quote of the actor and martial arts practitioner Bruce Lee. Journalists praised the creativity and digital savviness of activists, who managed to escape the police using innovative ways of organizing the crowd via different messengers (Dapiran, 2019). The moment of enthusiasm about the democratic wave did not last long. China cracked down on the protests, and the following two years of the Covid pandemic helped suppress dissidents even further (Maizland, 2022). The question of the fragility of leaderless protests, raised already after the Arab Spring (Della Porta, 2014; Tufekci, 2017), again proved to be crucial for social movement research. Can water seeping through a wall become a waterfall that breaks through a dam of networked censorship? If authoritarian regimes use social media to increase their resilience, can social movements become more sustainable and stronger by using digital technologies, too? A large body of research has analysed the benefits of social media for political mobilization in restrictive political settings (Diamond and Plattner, 2012; Della Porta, 2014). The strengths of connective action (Bennett and Segerberg, 2012), such as horizontal structure and anonymity, are specifically relevant for digital activists under authoritarianism. Breuer et al. (2015) explored the use of social media in the Tunisian revolution of 2011 through the lens of resource mobilization theory, and outlined the main roles that social media played for mobilizing civil society.

50  Handbook of digital politics

These included informing about the protest, facilitating collaboration, attracting new supporters, and ‘emotional mobilisation’ by telling the stories of regime brutalities (Breuer et al., 2015, p. 764). Lonkila et al. (2021) analysed forms of online activism in Russia and suggested distinguishing between technoactivism, based on manipulating technological systems, and communicative activism, which refers to human-to-human interactions (Lonkila et al., 2021, p. 139). Lokot (2018) explored the activist repertoires of Alexei Navalny’s network in Russia, and thematized the tension between the need for security and the importance of visibility as central to digital resistance practices. Many studies have proved the mobilization effects of social media in authoritarian settings (Diamond and Plattner, 2012; Della Porta, 2014; Breuer et al., 2015; Lee et al., 2015). These studies mostly focused on Facebook and Twitter as platforms of mobilization, with fewer examples for Telegram and other messengers (Urman et al., 2021; Wijermars and Lokot, 2022). Reuter and Szakonyi (2015) have found that in Russia, the use of Facebook and Twitter positively correlated with perceptions of electoral fraud, whereas the use of the Russia-based social networking sites VK and OK did not demonstrate the same effect. Studies have also looked at the specific roles that social media played in protests, as well as the limitations of their political efficacy (Breuer et al., 2015; Arafa and Armstrong, 2016). Thus, Agur and Frisch (2019) drawing on in-depth interviews with leading users of social media during the 2014 Umbrella protests in Hong Kong found that social media did enhance the mobilization and organization of protesters, but had only a limited effect on the persuasion of new protesters. Bodrunova and Litvinenko (2013) interviewed protesters of the 2011 ‘For Fair Elections’ movement in Russia about the perceived factors that influenced their decision to participate in the rallies. The protesters reported that the major trigger for them was the fact that they saw their friends, mostly on Facebook, openly voicing their dissent and expressing their willingness to attend the protests. After the initial enthusiasm with regard to the mobilization potential of social media, many scholars have pleaded for a more nuanced approach to studying media use by social movements as well as for caution with generalizations about social media effects (Alterman, 2011; Della Porta, 2014). Alterman (2011) called for revisiting the roles of Facebook and Twitter in the Arab uprisings, also mentioning the decisive role of satellite television in protest mobilization. Della Porta and Mattoni (2014) highlighted the importance of looking at the temporal phases of movements, arguing that during the 2011 Egyptian protests, for instance, Facebook was central only in the initial phase. Arafa and Armstrong (2016) explored the different roles that specific platforms such as YouTube, Facebook, and Twitter played during the Arab Spring uprisings. The 2005 Orange Revolution in Ukraine is often referred to as a herald for the rise of mobilization through ICTs, as digital media and mobile communication played a central role in mobilizing voters for successful protests in support of the oppositional candidate Viktor Yushchenko, albeit the Internet penetration in the country was not yet significant (Metzger and Tucker, 2017). Eight years later, the Euromaidan protests erupted in the country, against the autocrat Viktor Yanukovych who came

Revolution vs reaction: the role of social media in authoritarian regimes  51

to power after Yushchenko, and this new revolution already used a full range of social media tools for resource mobilization (Metzger and Tucker, 2017; Brantly, 2019). The Euromaidan was an example of a successful revolution that managed to overthrow the authoritarian ruler and usher in a new wave of democratization of the country. However, it was accompanied by anti-Euromaidan mobilization that resulted in the creation of pro-Russian separatist republics and the war in Donbas, which ultimately led to the full-scale Russian invasion in 2022. A study of media use during the Euromaidan showed the importance of pro-Russian television in the media diets of Euromaidan-sceptics (Onuch et al., 2021). However, with most research being focused on Euromaidan mobilization, the counter-movement remained practically out of sight of researchers. Generally, the counter-mobilization techniques used by autocrats during uprisings are mentioned in the studies but rarely become their focus, although they could shed light on the complex dynamics between repression and social mobilization, and on the reasons for certain outcomes of protests. Thus, Grimm and Harders (2018), in their study of the subjugation of the ‘anti-coup’ movement after the military coup against President Morsi in Egypt in 2013, demonstrated how exactly repressions had transformed protesters’ mobilization efforts. With authoritarian regimes co-opting social media and persecuting digital activism, new creative forms of protest appear (Lonkila et al., 2021). Bodrunova (2021) mentions digital gatherings on Yandex Maps as a form of online-only protests during the pandemic in Russia. Several studies have explored the role of humour, including memes and other entertaining content, on social networks as a form of political dissent (Miazhevich, 2015; Moreno-Almeida, 2021). Along with escaping into humour, political dissent also finds refuge in topics that activists themselves tend to perceive as non-political, such as ecology or gender issues (Oates, 2013). Pearce et al. (2014) suggest a two-dimensional framework for analysing civic engagement under authoritarianism, with a governmental/non-governmental dimension on the one side, and public/private on the other. With the example of Azerbaijan, they show that online communication is mostly associated with public civic engagement and in some cases, with private civic engagement. The effects of these subtler forms of activism on political mobilization have so far remained under-explored. There exist only a few studies on how audiences use and make sense of social media content under authoritarianism, due to the obvious challenges of this type of research (Stockmann et al., 2020). Koçer and Bozdag (2020) conducted a multi-method study of sharing news practices in Turkey using focus groups and semi-structured interviews. The study revealed self-censorship strategies of users, and demonstrated the perception of news sharing by some respondents as activism. In some regimes, for instance in Belarus, even simple following of oppositional Telegram channels can lead to jail sentences (Prince, 2021). The perspectives of ordinary users, who, in authoritarian settings, can be persecuted for liking or sharing a post, should prompt us to reconsider the usually rather dismissive attitude to clicktivism as a not really significant low-cost practice (Halupka, 2014). The growing pressure on dissidents in authoritarian regimes forces many of them to leave their countries and form oppositional communities abroad. Thus, after the

52  Handbook of digital politics

introduction of the so-called ‘Anti-fake news’ law, which de facto meant military censorship in Russia, shortly after Putin’s full-scale invasion of Ukraine, major oppositional media were forced to close and journalists had to flee the country. Several big media outlets split into dozens of small Telegram and YouTube channels. One of the major oppositional media outlets, the television channel ‘Dozhd’, which is often translated as ‘TV Rain’, started to call their journalists, who now operated separately, ‘rain drops’. However, despite the first impression of so many ‘rain drops’ – that is, individual oppositional channels – appearing on social media, they obviously could not make up for the loss of full-fledged editorial offices with nationwide outreach. Besides, the influence and trustworthiness of emigrant publics among those who stayed in the country are highly debated. As transborder activism via social media becomes increasingly important (Michaelsen, 2017), its practices and effects are still to be better explored in future research.

BLIND SPOTS AND PATHS FOR FUTURE RESEARCH Over the past decade, scholars have gathered an immense amount of evidence on how, and with what consequences, different actors in authoritarian regimes employ social media. Based on the overview of the growing strands of research in this area, I identify three promising paths for future research that could address the existing gaps in the study of authoritarian political communication. The first one I would broadly describe as looking at the bigger picture. This means that, on the one hand, we need to assess the long-term effects of social media, looking at the cycles of revolutions and counter-revolutions, and comparing the roles and effects of social media over time. On the other hand, it is important to ‘zoom out’ from focusing on social media only and to look at digital media in the larger context of authoritarian media systems, of which they are a part. The role of television and other ‘traditional’ media should not be neglected, as they continue to play an important role in shaping discourses. Online and offline realms should be looked at as intertwined and synergic. This is true especially for protest in restrictive political settings, as due to the massive digital surveillance activists are often forced to pursue traditional analogue activism creating a mixture of online and offline forms of protest. As authoritarian propaganda aims at depoliticizing public communication, it is essential to look for the political in seemingly unpolitical matters, paying attention, among other things, to entertaining content as part of the discourse. Speaking of discourse, unpacking the ways in which it is constructed using different media channels is a promising area of research that can contribute to a better understanding of political communication under authoritarianism. The second path is studying audiences under authoritarianism. How do people use, perceive and engage with various content? How do they make sense of propaganda messages on different media channels? How do they develop trust in certain sources of information? Do they consume media in exile and for what reasons? Are the efforts of fact-checkers effective? Autocrats usually deprive people of their agency,

Revolution vs reaction: the role of social media in authoritarian regimes  53

disempowering them. It is therefore even more important for researchers to return the agency to the people seeking to understand their perspective and repertoires of dealing with information, including propaganda. This would create a more nuanced understanding of public opinion mechanisms under authoritarianism and help avoid a generalized perception about a passive mass that blindly follows the autocrat. The third path would be focusing on the role of tech platforms as important players that provide communication infrastructure for both autocrats and activists. Although it has been thematized in many studies, no systematic research has been done in this direction, with the exception of a few studies (Youmans and York, 2012; Stockmann et al., 2020; Wijermars and Lokot, 2022). Stockmann et al. (2020) demonstrated that some affordances of platforms were more conducive to political talk than others. Comparative studies across different platforms are needed to better understand the role of discourse architectures of platforms (Freelon, 2015) in political talk. The role of Telegram, TikTok, and Instagram, which are increasingly being used both by authoritarian propaganda and social movements around the globe, are to be further explored. Innovative approaches are needed to study the role of private messengers, which – due to (perceived) anonymity – are highly popular for political talk in authoritarian regimes. This strand of research might result in practical outcomes for improving platform governance showing how Internet platforms could contribute to democratization rather than to authoritarian diffusion. In general, more comparative studies within as well as across different regions are needed to reveal patterns and contextual factors in the use and effects of social media. As this literature review has demonstrated, some world regions remain largely understudied, including Latin America (Lugo-Ocando et al., 2015) and sub-Saharan Africa (Mutsvairo and Rønning, 2020). Studies across different political contexts of both bottom-up political mobilization (Della Porta, 2014) and illiberal practices of Internet control (Michaelsen and Glasius, 2018) might be enlightening for analysing global diffusion of norms and practices. Expanding research on the above-mentioned areas will not only bring us to a better understanding of the role that social media play in authoritarian resilience and in the process of democratization, it could also help us in countering disinformation campaigns and contribute to the development of better Internet governance.

FURTHER READING Deibert, R. (2015). Authoritarianism goes global: Cyberspace under siege. Journal of Democracy, 26(3), 64–78. Della Porta, D. (2014). Mobilizing for Democracy: Comparing 1989 and 2011. Oxford: Oxford University Press. Roberts, M. E. (2018). Censored: Distraction and Diversion Inside China’s Great Firewall. Princeton, NJ: Princeton University Press. Stockmann, D., Luo, T., and Shen, M. (2020). Designing authoritarian deliberation: How social media platforms influence political talk in China. Democratization, 27(2), 243–264.

54  Handbook of digital politics

Toepfl, F. (2020). Comparing authoritarian publics: The benefits and risks of three types of publics for autocrats. Communication Theory, 30(2), 105–125. Tufekci, Z. (2017). Twitter and Tear Gas: The Power and Fragility of Networked Protest. New Haven: Yale University Press.

REFERENCES Agur, C. and Frisch, N. (2019). Digital disobedience and the limits of persuasion: Social media activism in Hong Kong’s 2014 Umbrella Movement. Social Media + Society, 5(1), 205630511982700. https://​doi​.org/​10​.1177/​2056305119827002. Alizada, N., Cole, R., Gastaldi, L., Grahn, S., Hellmeier, S., Kolvani, P., Lachapelle, J., Lührmann, A., Maerz, S. F., Pillai, S., and Lindberg, S. I. (2021). Autocratization Turns Viral: Democracy Report 2021. University of Gothenburg: V-Dem Institute. Alterman, J. B. (2011). The revolution will not be tweeted. The Washington Quarterly, 34(4), 103–116. Arafa, M. and Armstrong, C. (2016). Facebook to mobilize, Twitter to coordinate protests, and YouTube to tell the world’: New media, cyberactivism, and the Arab Spring. Journal of Global Initiatives: Policy, Pedagogy, Perspective, 10(1). https://​digitalcommons​.kennesaw​ .edu/​jgi/​vol10/​iss1/​6. Aro, J. (2016). The cyberspace war: Propaganda and trolling as warfare tools. European View, 15(1), 121–132. Asmolov, G. (2018). The disconnective power of disinformation campaigns. Journal of International Affairs, 71(1.5), 69–76. Bennett, W. L. and Segerberg, A. (2012). The logic of connective action. Information, Communication & Society, 15(5), 739–768. Bodrunova, S. (2021). Russia: A glass wall. In D. G. Lilleker, I. A. Coman, M. Gregor, and E. Novelli (eds.), Political Communication and COVID-19: Governance and Rhetoric in Times of Crisis (pp. 188–200). London and New York: Routledge. Bodrunova, S. and Litvinenko, A. (2013). New media and the political protest: The formation of a public counter-sphere in Russia of 2008–2012. In A. Makarychev and A. Mommen (eds.), Russia’s Changing Economic and Political Regimes (pp. 29–65). London: Routledge. Bolsover, G. and Howard, P. (2019). Chinese computational propaganda: Automation, algorithms and the manipulation of information about Chinese politics on Twitter and Weibo. Information, Communication & Society, 22(14), 2063–2080. Boyte, K. J. (2017). An analysis of the social-media technology, tactics, and narratives used to control perception in the propaganda war over Ukraine. Journal of Information Warfare, 16(1), 88–111. Brantly, A. F. (2019). From cyberspace to Independence Square: Understanding the impact of social media on physical protest mobilization during Ukraine’s Euromaidan revolution. Journal of Information Technology & Politics, 16(4), 360–378. Breuer, A., Landman, T., and Farquhar, D. (2015). Social media and protest mobilization: Evidence from the Tunisian revolution. Democratization, 22(4), 764–792. Crilley, R. and Chatterje-Doody, P. N. (2021). Government disinformation in war and conflict. In H. Tumber and S. Waisbord (eds.), The Routledge Companion to Media Disinformation and Populism (pp. 242–253). Abingdon: Routledge. Dapiran, A. (2019). ‘Be Water!’: Seven tactics that are winning Hong Kong’s democracy revolution. The New Statesman. https://​www​.newstatesman​.com/​politics/​2019/​08/​be​-water​ -seven​-tactics​-that​-are​-winning​-hong​-kongs​-democracy​-revolution​-2. Deibert, R. (2015). Authoritarianism goes global: Cyberspace under siege. Journal of Democracy, 26(3), 64–78.

Revolution vs reaction: the role of social media in authoritarian regimes  55

Deibert, R. and Rohozinski, R. (2010). Liberation vs. control: The future of cyberspace. Journal of Democracy, 21(4), 43–57. Della Porta, D. (2014). Mobilizing for Democracy: Comparing 1989 and 2011. Oxford: Oxford University Press. Della Porta, D. and Mattoni, A. (2014). Social networking sites in pro-democracy and anti-austerity protests: Some thoughts from a social movement perspective. In D. Trottier and C. Fuchs (eds.), Social Media, Politics and the State (pp. 39–63). New York: Routledge. Diamond, L. (2012). Liberation technology. In L. Diamond and M. F. Plattner (eds.), Liberation Technology: Social Media and the Struggle for Democracy (pp. 3–18). Baltimore, MD: Johns Hopkins University Press. Diamond, L. and Plattner, M. F. (eds.) (2012). Liberation Technology: Social Media and the Struggle for Democracy. Baltimore, MD: Johns Hopkins University Press. Freelon, D. (2015). Discourse architecture, ideology, and democratic norms in online political discussion. New Media & Society, 17(5), 772–791. Freelon, D. and Lokot, T. (2020). Russian disinformation campaigns on Twitter target political communities across the spectrum: Collaboration between opposed political groups might be the most effective way to counter it. Harvard Kennedy School Misinformation Review. https://​doi​.org/​10​.37016/​mr​-2020​-003. Geddes, B. (1999). What do we know about democratization after twenty years? Annual Review of Political Science, 2(1), 115–144. Grimm, J. and Harders, C. (2018). Unpacking the effects of repression: The evolution of Islamist repertoires of contention in Egypt after the fall of President Morsi. Social Movement Studies, 17(1), 1–18. Hall, S. G. F. and Ambrosio, T. (2017). Authoritarian learning: A conceptual overview. East European Politics, 33(2), 143–161. Halupka, M. (2014). Clicktivism: A systematic heuristic. Policy & Internet, 6(2), 115–132. Hellmeier, S. (2016). The dictator’s digital toolkit: Explaining variation in Internet filtering in authoritarian regimes. Politics & Policy, 44(6), 1158–1191. Howard, P., Ganesh, B., Liotsiou, D., Kelly, J., and François, C. (2018). The IRA, Social Media and Political Polarization in the United States, 2012–2018. Oxford University Institute: Project on Computational Propaganda. Hyun, K. D. and Kim, J. (2015). The role of new media in sustaining the status quo: Online political expression, nationalism, and system support in China. Information, Communication & Society, 18(7), 766–781. Keremoğlu, E. and Weidmann, N. B. (2020). How dictators control the Internet: A review essay. Comparative Political Studies, 53(10–11), 1690–1703. King, G., Pan, J., and Roberts, M. E. (2013). How censorship in China allows government criticism but silences collective expression. American Political Science Review, 107(2), 1–18. King, G., Pan, J., and Roberts, M. E. (2017). How the Chinese government fabricates social media posts for strategic distraction, not engaged argument. American Political Science Review, 111(3), 484–501. Koçer, S. and Bozdag, Ç. (2020). News-sharing repertoires on social media in the context of networked authoritarianism: The case of Turkey. International Journal of Communication, 14, 5292–5310. Kravets, D. and Toepfl, F. (2022). Gauging reference and source bias over time: How Russia’s partially state-controlled search engine Yandex mediated an anti-regime protest event. Information, Communication & Society, 25(15), 2207–2223. Lagacé, C. B. and Gandhi, J. (2015). Authoritarian institutions. In J. Gandhi and R. Ruiz-Rufino (eds.), The Routledge Handbook of Comparative Political Institutions (pp. 278–291). New York: Routledge. Lee, P. S. N., So, C. Y. K., and Leung, L. (2015). Social media and Umbrella Movement: Insurgent public sphere in formation. Chinese Journal of Communication, 8(4), 356–375.

56  Handbook of digital politics

Lewis, D. (2016). Blogging Zhanaozen: Hegemonic discourse and authoritarian resilience in Kazakhstan. Central Asian Survey, 35(3), 421–438. Li, X., Lee, F. L. F., and Li, Y. (2016). The dual impact of social media under networked authoritarianism: Social media use, civic attitudes, and system support. International Journal of Communication, 10(2016), 5143–5163. Litvinenko, A. (2021). Re-defining borders online: Russia’s strategic narrative on internet sovereignty. Media & Communication, 9(4), 5–15. Litvinenko, A. and Toepfl, F. (2019). The ‘gardening’ of an authoritarian public at large: How Russia’s ruling elites transformed the country’s media landscape after the 2011/12 protests ‘for fair elections’. Publizistik, 64(2), 225–240. Litvinenko, A. and Zavadski, A. (2020). Memories on demand: Narratives about 1917 in Russia’s online publics. Europe-Asia Studies, 72(10), 1657–1677. Lokot, T. (2018). Be safe or be seen? How Russian activists negotiate visibility and security in online resistance practices. Surveillance & Society, 16(3), 332–346. Lonkila, M., Shpakovskaya, L., and Torchinsky, P. (2021). Digital activism in Russia: The evolution and forms of online participation in an authoritarian state. In D. Gritsenko, M. Wijermars, and M. Kopotev (eds.), The Palgrave Handbook of Digital Russia Studies (pp. 137–153). Cham: Springer International Publishing. Lugo-Ocando, J., Hernández, A., and Marchesi, M. (2015). Social media and virality in the 2014 student protests in Venezuela: Rethinking engagement and dialogue in times of imitation. International Journal of Communication, 9, 3782–3802. Lukito, J. (2020). Coordinating a multi-platform disinformation campaign: Internet research agency activity on three U.S. social media platforms, 2015 to 2017. Political Communication, 37(2), 238–255. MacKinnon, R. (2011). China’s ‘networked authoritarianism’. Journal of Democracy, 22(2), 32–46. Maizland, L. (2022). Hong Kong’s Freedoms: What China Promised and How It’s Cracking Down. Council on Foreign Relations. https://​www​.cfr​.org/​backgrounder/​hong​-kong​ -freedoms​-democracy​-protests​-china​-crackdown. Metzger, M. M. and Tucker, J. A. (2017). Social media and EuroMaidan. Slavic Review, 76(1), 169–191. Miazhevich, G. (2015). Sites of subversion: Online political satire in two post-Soviet states. Media, Culture & Society, 37(3), 422–439. Michaelsen, M. (2017). Far away, so close: Transnational activism, digital surveillance and authoritarian control in Iran. Surveillance & Society, 15(3–4), 465–470. Michaelsen, M. and Glasius, M. (2018). Authoritarian practices in the digital age: Introduction. International Journal of Communication, 12, 3788–3794. Moreno-Almeida, C. (2021). Memes as snapshots of participation: The role of digital amateur activists in authoritarian regimes. New Media & Society, 23(6), 1545–1566. Mutsvairo, B. and Rønning, H. (2020). The Janus face of social media and democracy? Reflections on Africa. Media, Culture & Society, 42(3), 317–328. Oates, S. (2013). Revolution Stalled: The Political Limits of the Internet in the Post-Soviet Sphere. Oxford: Oxford University Press. Onuch, O., Mateo, E., and Waller, J. G. (2021). Mobilization, mass perceptions, and (dis)information: ‘New’ and ‘Old’ media consumption patterns and protest. Social Media + Society, 7(2), 205630512199965. https://​doi​.org/​10​.1177/​2056305121999656. Paltemaa, L. and Vuori, J. A. (2009). Regime transition and the Chinese politics of technology: From mass science to the controlled Internet. Asian Journal of Political Science, 17(1), 1–23. Paltemaa, L., Vuori, J. A., Mattlin, M., and Katajisto, J. (2020). Meta-information censorship and the creation of the Chinanet bubble. Information, Communication & Society, 23(14), 2064–2080.

Revolution vs reaction: the role of social media in authoritarian regimes  57

Pearce, K. E., Freelon, D., and Kendzior, S. (2014). The effect of the Internet on civic engagement under authoritarianism: The case of Azerbaijan. First Monday, 19(6). https://​doi​.org/​ 10​.5210/​fm​.v19i6​.5000. Pearce, K. E. and Kendzior, S. (2012). Networked authoritarianism and social media in Azerbaijan. Journal of Communication, 62(2), 283–298. Prince, T. (2021). Seven years in prison for following a telegram channel? Some Belarusians are unsubscribing, just in case. https://​www​.rferl​.org/​a/​belarus​-telegram​-extremist​-subscribers/​ 31511257​.html. Reuter, O. J. and Szakonyi, D. (2015). Online social media and political awareness in authoritarian regimes. British Journal of Political Science, 45(1), 29–51. RFERL (2022). Kazakh president pledges reforms in wake of deadly protests. https://​www​ .rferl​.org/​a/​kazakhstan​-president​-toqaev​-reforms​-protests/​31756553​.html. Richter, C. and Kozman, C. (2021). Introduction. In C. Richter and C. Kozman (eds.), Arab Media Systems (Vol. 3) (pp. xi–xliv). Open Book Publishers. https://​doi​.org/​10​.11647/​obp​ .0238. Roberts, M. E. (2018). Censored: Distraction and Diversion Inside China’s Great Firewall. Princeton, NJ: Princeton University Press. Rød, E. G. and Weidmann, N. B. (2015). Empowering activists or autocrats? The Internet in authoritarian regimes. Journal of Peace Research, 52(3), 338–351. RSF (2022). World Press Freedom Index: A New Era of Polarisation. Reporters Without Borders. https://​rsf​.org/​en/​rsf​%E2​%80​%99s​-2022​-world​-press​-freedom​-index​-new​-era​ -polarisation​?year​=​2022​&​data​_type​=​general. Schedler, A. (2009). The new institutionalism in the study of authoritarian regimes. Totalitarianism and Democracy, 6(2), 323–340. Sika, N. (2019). Repression, cooptation, and movement fragmentation in authoritarian regimes: Evidence from the youth movement in Egypt. Political Studies, 67(3), 676–692. Sinpeng, A. (2020). Digital media, political authoritarianism, and Internet controls in Southeast Asia. Media, Culture & Society, 42(1), 25–39. Soest, C. von and Grauvogel, J. (2017). Identity, procedures and performance: How authoritarian regimes legitimize their rule. Contemporary Politics, 23(3), 287–305. Stockmann, D., Luo, T., and Shen, M. (2020). Designing authoritarian deliberation: How social media platforms influence political talk in China. Democratization, 27(2), 243–264. Tang, M. and Huhe, N. (2014). Alternative framing: The effect of the Internet on political support in authoritarian China. International Political Science Review, 35(5), 559–576. Toepfl, F. (2020). Comparing authoritarian publics: The benefits and risks of three types of publics for autocrats. Communication Theory, 30(2), 105–125. Toepfl, F. and Litvinenko, A. (2021). Critically commenting publics as authoritarian input institutions: How citizens comment beneath their news in Azerbaijan, Russia, and Turkmenistan. Journalism Studies, 22(4), 475–495. Tufekci, Z. (2017). Twitter and Tear Gas: The Power and Fragility of Networked Protest. New Haven: Yale University Press. Turkstra, A. (2022). Kazakhstan undergoes changes comparable to ‘Perestroika’. https://​www​ .euractiv​.com/​section/​central​-asia/​opinion/​kazakhstan​-undergoes​-changes​-comparable​-to​ -perestroika/​. Uniacke, R. (2021). Authoritarianism in the information age: State branding, depoliticizing and ‘de-civilizing’ of online civil society in Saudi Arabia and the United Arab Emirates. British Journal of Middle Eastern Studies, 48(5), 979–999. Urman, A., Ho, J. C.-T., and Katz, S. (2021). Analyzing protest mobilization on Telegram: The case of 2019 Anti-Extradition Bill movement in Hong Kong. PloS One, 16(10), e0256675. https://​doi​.org/​10​.1371/​journal​.pone​.0256675.

58  Handbook of digital politics

Wijermars, M. and Lokot, T. (2022). Is Telegram a ‘harbinger of freedom’? The performance, practices, and perception of platforms as political actors in authoritarian states. Post-Soviet Affairs, 38(1–2), 125–145. Youmans, W. L. and York, J. C. (2012). Social media and the activist toolkit: User agreements, corporate interests, and the information infrastructure of modern social movements. Journal of Communication, 62(2), 315–329. Zavadski, A. and Toepfl, F. (2019). Querying the Internet as a mnemonic practice: How search engines mediate four types of past events in Russia. Media, Culture & Society, 41(1), 21–37.

5. Transnational and global flows of political discussion online Yuan Zeng

Political communication always been transcending national borders, especially so in the digital age. In the past years, media and the public across the world have actively participated in transnational or global issues such as climate change and women’s rights . Even on issues innately national, such as US presidential elections, the global public, galvanized by digital affordances of social media platforms and the retreat of traditional gatekeepers in international politics, convene on platforms such as Twitter, Clubhouse, contributing to the multi-modal and multi-dimensional flows of political discourse both globally and transnationally. Yet such global and transnational flows of political discussion are deeply ingrained with global inequality, between global north and south, between former colonisers and colonised. In the meantime, with the drums of nationalism beating ever louder around the world, authoritarian state actors especially in China and Russia are tightening the reins on the inflow of political discourse which proves to be much harder to tame than the domestic one. Amidst the technological, geopolitical, and ideological shifts that are redrawing and (dis)connecting the local and global spaces we live in, it might be a Herculean mission to try to map the global flows of political discussion online. What are the roles of nation-state, media, and citizens in curating and mobilizing online political discussion on a transnational and global scale? Who takes the lead in transnational and global political discussion? What are the new forms of the global flows of political discussion? Before addressing these questions, I present the readers with a vignette of the globally contested political narrative over Xinjiang, which might illustrate the intricate multimodality, multidimensionality, and transnationality of the topic. A remote far-western region of China, Xinjiang has in the past two years captured the attention of global media, political actors, activists, and ordinary citizens. Since 2018 China has imposed a harsh crackdown on the local Uyghur Muslim people and kept a tight lid on the narrative to fend off concerns raised by the international community. Some international media have reported the worrying situation, only with China vehemently rejecting such criticism, being emboldened by its decades-long practice and highly sophisticated state propaganda apparatus of information control at home. Meanwhile, across the Atlantic, an American teenager in New Jersey uploaded a makeup tutorial on TikTok, arguably the most popular short-video platform (owned by Chinese tech company ByteDance), where Gen Z share light-hearted and often funny lip-sync or dancing videos. Except that this makeup tutorial is not about makeup at all. Whilst curling her eyelashes in the video, the 17-year-old was 59

60  Handbook of digital politics

urging viewers to closely follow the situation in Xinjiang, where growing evidence suggests that up to one million Uyghur Muslims have been rounded up in internment camps. The video soon garnered millions of views on TikTok and millions more across other social media platforms including Twitter and Facebook. More young social media users followed suit in support of Uyghur Muslims uploading and sharing similar videos. Diasporic dissidents having fled Xinjiang also speak up both on digital platforms and to mainstream news outlets (Bonnenfant, 2022). These voices from citizens and activists across the world, form a global discourse divergent from China's state narrative, raising awareness of the issue across national borders, especially in multiple Western countries against the backdrop of worsening China-West relations. global businesses vowed to boycott Xinjiang cotton, allegedly produced by forced Uyghur labour. Faced with growing pressure from the international community which China can no longer brush aside, a pro-China narrative has received a gigantic boost on TikTok, Twitter, and Facebook, even though none of these platforms is accessible from inside China. (TikTok is not accessible in China, despite being owned by a Chinese company. Instead, the company operates its ‘sister app’ Douyin in China.) Among those promoting a pro-China narrative on social media are not only Chinese state media, but also diplomats, paid or unwitting social media influencers, or even worse – bots. Chinese embassies around the world are promoting the same content on both Western and Chinese social media platforms, in an apparent effort to target both foreign publics and overseas Chinese diaspora. Whereas inside China, overseas discussion on Xinjiang circles back into the country, but only to be meticulously cherry-picked and reframed by the state propaganda apparatus into a nationalistic narrative of a ‘West-led smear campaign against China’, usually invoking anti-imperialist and anti-colonialist rhetoric. Stirred by such rhetoric, the public anger has swarmed the digital space – not directed at the human rights abuse allegation, but at the foreign public and foreign media for ‘smearing China’. A propaganda campaign against British broadcaster BBC is particularly prominent, with state media and the public joining to accuse the BBC’s report on Xinjiang as ‘biased’. The public castigation was so harsh that BBC China correspondent John Sudworth was trolled and eventually had to relocate to Taiwan over safety concerns. This is just a snapshot of the transnationality and multimodality of political discussion online – a messy picture with multiple intertwined trajectories (transnational and national, inward and outward), a multiplicity of actors (states, citizens, journalists, businesses, activists), contesting narratives on and across multiple platforms, borderless yet still nation-state centred. In the space of one chapter, I do not plan to fully capture the messy picture of global political discussion online, or what Applebaum (2020) calls a ‘massive international wave of cacophony’. Instead, the following sections identify the strands of scholarly debate on this emerging area and pinpoint directions for future research.

Transnational and global flows of political discussion online  61

POLITICAL DISCUSSION ONLINE: TRANSNATIONAL AND MULTIMODAL Political discussion has always existed in both national and transnational spaces. But in recent decades, as we enter what Blumler (2016) calls ‘the fourth age of political communication’, globalization and digital technology have been further networking different political systems, actors, and publics together, reconfiguring the nature and forms of political discourse and debate, and ‘generating transcontinental or interregional flows and networks of activity, interaction and power’ (Held et al., 2000, p. 16). This poses new challenges as well as recasts old ones and requires urgent attention as the morass of digital politics, global activism, and geopolitical struggles is, albeit messy, (re)shaping our ways of living both individually and collectively. With the public becoming digital, transnational, and active, political information no longer flows as dictated by state leaders, international media conglomerates, or borders, morphing into what Castells (2013) calls a global network of networks of multimodal communication. Castells (2012), among others, provides a riveting account of the transnational networks of social movements, starting in Tunisia before soon sweeping around the world from 2010–13, with politicians resorting to social media of various national origins to reach diasporic constituents, and state actors of competing national interests contesting various versions of national stories to win public support, either foreign or global. The term ‘international media’ has lost its traditional meaning as now every institutional media outlet, regardless how local or international, potentially has a global reach thanks to the Internet; citizens online are enjoying easy access to discussions on global politics, not just as consumers of information, but more importantly as content producers. Whilst national political apathy is a widely acknowledged hindrance to political engagement, global engagement proves to be blossoming: public awareness on global challenges, as seen in climate change issues and social justice activism, is engaging public voices in a decentralized or network paradigm, pushing nation-states, together with international bodies, to make political commitments and change political agendas and public policies. Another new development in the current state of the flows of political discussion online is that the fabric of global network society is no longer meshed exclusively by rich Western democracies. Instead, it incorporates various political systems and media systems with distinct historical and cultural legacies, oftentimes of conflicting nature. Resourceful non-western countries such as China and Russia, whilst challenging the old west-led information order with their massive media apparatus and anti-colonial rhetoric, also actively employ both overt and covert tactics waging state-led misinformation campaigns on the open, inclusive, easily accessible digital platforms to bypass traditional gatekeepers and national regulators in transnational information flows. The British broadcasting regulator Ofcom has recently revoked the licences of China’s state broadcaster CGTN and Russia’s RT to broadcast in the UK, as a response to the two state propaganda bodies’ biased coverage on Hong Kong protests and Russia’s invasion of Ukraine, respectively. But any British viewer still can easily access their content online. On YouTube alone, for example, CGTN

62  Handbook of digital politics

has accumulated 2.9 million subscribers. RT had a much bigger overseas viewership until being taken down from YouTube and other big digital platforms, shortly after Russia’s invasion of Ukraine in 2022. Drawing on the ambient transnationality of late modernity whilst noting the decline of institutions (particularly national organizations as structuring principles of social and political action), some scholars suggest that we adopt ‘a cosmopolitan approach’ to the study of political communication (Moy et al., 2012; Rojecki, 2011). Yet given the colonial legacies, the notable rise in nationalism and inter-state contestation as shown in preceding paragraphs, compared to globalization or cosmopolitanism, transnationality better captures the relational nature of social, cultural processes that stream across spaces, and their ‘embeddedness in differently configured regimes of power’ (Ong, 1999, p. 4). I borrow this notion of transnationality from anthropology and sociology (Kearney, 1995; Ong, 1999) in discussing the interconnected and relational dynamics of global/transnational political discussion online and argue that this transnational practice and imagination not only is economic, social and cultural, but also has a substantial political dimension. This political dimension of transnationality underpins research in the fields of international relations, political science, and communication studies. Below I lay out four themes that have stood out in the global research agenda as prominent in both scholarly significance and practical implications.

NATION-STATE AS THE PROTAGONIST IN CURATING TRANSNATIONAL POLITICAL DISCUSSION Despite the multiplicity of actors as a defining feature of transnational political discussion online, state actors are still arguably the main protagonists in initiating, facilitating, or, in some cases, sabotaging transnational political discussion. Apart from the traditional nation-state model of political communication across borders – that of international institutions or transnational media – more recently academic interest, albeit still scarce, is increasingly shifting to two themes: state-led digital narrative contestation (figuring in digital diplomacy and digital information warfare) and diasporic digital politics. Digital Diplomacy and Information Warfare For international politics and international relations, cyberspace has become the new discursive battlefield for nation-states. To shape global public opinion and disrupt political discourse in foreign countries (in order to advance certain political or diplomatic interests), state-sponsored information operations, either defensive or offensive, are increasingly becoming part and parcel of global flows of political communication. These information operations, usually referred to as ‘information warfare’ in IR and security studies to underscore the nature of contestation, have drawn scholarly interest from political science, communication studies, and interna-

Transnational and global flows of political discussion online  63

tional relations. Scholarship in these fields highlights two particularly salient areas: digital diplomacy and digital disinformation. ‘Digital diplomacy’ refers to the practices of state actors engaging digital communication platforms in foreign policy and diplomacy, aiming to cultivate favourable foreign public opinion (Bjola and Holmes, 2015; Duncombe, 2019; Manor, 2019). In essence, it comprises the practices of ‘facilitating transnational discourse on key policy issues between state and non-state stakeholders’ (Duncombe, 2019, p. 104). Most studies to date centre on the United States, where former President Obama’s tactful use of Twitter and Facebook in engaging foreign publics is widely regarded as the inception of digital diplomacy (Holmes, 2015). The country’s erratic former President Donald Trump’s even more erratic use of Twitter has raised scholarly concerns over ‘technological populism’ and global engagement (Baldwin-Philippi, 2019; Cooper, 2019; Ott, 2017). State actors using open digital platforms, foremost, indeed opens up the opportunity for citizens and other non-traditional actors (other than diplomats) across national borders to join the open discussion that shapes foreign policy and international relations, but it also facilitates a versatile form of state propaganda weaponizing nationalism and global populism. In the Xinjiang case I presented at the beginning of this chapter, Chinese diplomats resort to Twitter to advance a state-sanctioned pro-China narrative and dismiss international accusations of China’s human rights violations as a ‘smear campaign’, feeding into nationalist discourse in China as well as the populist discourse in the West, as rebuttal and aggressiveness from foreign state actors would eventually appeal to those who feel politically ‘left alone’ at home (Fuchs, 2018). The most used platforms – social media such as Twitter – are found to be injecting too much adrenaline but too little rational deliberation into the political discussion, even among diplomats and the like. Duncombe (2019, p. 104) contends that social media platforms ‘facilitate technologically mediated forms of emotional contagion’, hence changing the dynamics of digital diplomacy. This affective dimension of propaganda, or ‘emotional diplomacy’ (Hall, 2015), albeit nothing new, is greatly galvanized by the openness and engagement of social media, as ‘the public nature of social media posts allows a global public to witness exchanges between political leaders and diplomats, and the speed of social media communication means that there is both much less time to absorb information and a need to respond equally quickly to ensure message dominance’ (Duncombe, 2019, p. 109). China’s so-called ‘wolf warrior diplomacy’ staging a combative tone to fight unfavourable narratives on Twitter best exemplifies such an emotion-driven tactic in digital diplomacy (Zeng, 2023). Concomitant with digital diplomacy are state-sponsored disinformation campaigns online, which disrupt the flow of accurate information at a deeply concerning speed and scale (Bennett and Livingston, 2018). Although the main target audiences of state-sponsored disinformation campaigns remain domestic, scholars identify a notable increase in disinformation campaigns targeting foreign publics, as seen in Russia targeting the US and Europe, and the disinformation operations between China and Taiwan, Israel and Palestine, Russia and Ukraine, etc. Acknowledging the urgency to better understand the growing use of disinformation in international

64  Handbook of digital politics

politics as bolstered by digital platforms, scholars have examined the scale, mechanisms, and effect of such transnational disinformation campaigns aimed at disrupting the democratic discussion, inflating/faking the popularity of certain opinions, and ultimately influencing the democratic elections in a foreign country (see, e.g. Benkler et al., 2018; Jamieson, 2018; Kragh and Åsberg, 2017; Lukito, 2020; Moore, 2018; Ziegler, 2018). Bradshaw and Howard (2018), for example, examined the global nature of what they call ‘computational propaganda’ (using automation, algorithms, and big data analytics) by state sponsored ‘cyber troops’ on social media platforms. Political science scholars approach the issue of disinformation campaigns, or ‘problematic information’ (Jack, 2017), from the angle of its detrimental effect on democratic deliberation. As Tenove (2020, p. 521) eloquently argues, ‘disinformation may undermine a deliberative system not only by increasing the quantity of false claims in circulation but also by decreasing people’s interest and opportunity to engage in public discussions on terms of reason giving, respect, and inclusivity’. State-sponsored disinformation at times is so intertwined with ‘digital diplomacy’ especially for rogue state actors that it is not uncommon to see diplomats propagating disinformation to foreign publics on social media. As I explained earlier in the Xinjiang case, Chinese diplomats are working together with state media and ‘cyber troops’ to flood the open cyberspace with disinformation, aiming to divert and dismantle the ‘human rights violation’ narrative. Whether ‘digital diplomacy’ or ‘digital disinformation’, the actual effect of state-sponsored information campaigns on swaying the political discourse in a foreign country is contested among scholars. Whilst many contend that such transnational information operations do leave a dent on the public opinion and the democratic procedure of the target country (Tenove, 2020), others hold that usually it is the domestic politics and traditional media of the target countries that matter (Benkler et al. 2018; Zeng, 2023). Whilst more empirical studies are needed, it seems that transnational (dis)information on social media is at least effective in pushing fringe opinions into the mainstream. As Schia and Gjesvik (2020, p. 415) note, the main challenges ‘have less to do with immediate attempts at swaying public opinion, and more to do with the splintering of a shared public discourse, leading to new forms of political organization with uncertain consequences’. Diasporic Digital Politics Another increasingly salient area is the rise of diasporic digital politics. Outside political communication, diaspora has largely been examined in terms of its social and cultural significance. But the ability of diaspora to influence both the sending and host countries’ political discussion, especially foreign policy, has only recently gained attention from political communication scholars. With the rise of global immigration flows, the ‘digital diaspora’ or ‘virtual diaspora’ (Laguerre, 2010) has conveniently bolstered the states (both of the sending country and the host country) to woo their diasporic constituents with transnational political messages, or to win

Transnational and global flows of political discussion online  65

support from diaspora in international conflicts (Ho and McConnell, 2019; Murti and Zaharna, 2014). Sun and Yu (2020) examined the role of WeChat in political communication and particularly citizen-making among Chinese diaspora in Australia. The Chinese migrant community in Australia, as the authors note, heavily rely on WeChat to get political information. The ubiquitous Chinese social media site WeChat, long criticized for colluding with the Chinese state’s heavy surveillance and censorship, has nonetheless become a key battlefield for political campaigning in countries with large Chinese diaspora communities. In the case of Australia, Candidates from major parties even held live sessions on WeChat with Chinese-speaking voters. Once on WeChat, information shared among, say, Australia-based Chinese diaspora, is converging with Chinese diaspora in other countries, or Chinese inside China, via personal links. Thus the ‘WeChat diaspora’, or ‘digital diaspora’ in general, is much more expansive than any other traditional form of diaspora, posing as a very enticing asset for the sending nation-state to conquer minds and hearts. This state-led communication effort to engage diaspora to advance state interests is also referred to as ‘migration diplomacy’ (Adamson and Tsourapas, 2019) or ‘diaspora diplomacy’ (Ho and McConnell, 2019). For authoritarian state actors, such digital communicative initiative oftentimes resembles diaspora governance with repressive strategies to coerce and co-opt diasporic dissidents across borders, featuring what scholars call ‘transnational authoritarianism’ (Baser and Ozturk, 2020; Tsourapas, 2021). It is well evidenced in Syria (Moss, 2018), Turkey (Baser and Ozturk, 2020), and Eritrea (Hirt and Saleh Mohammad, 2018). More often, such state-led messaging invokes patriotism rather than overt repression, appealing to a wide network of diaspora communities. In this sense, digital affordances facilitate authoritarian states to co-opt overseas diaspora. Diasporic digital politics to date is still an under-researched area. But with the rise of global immigration and new ICTs, how state actors engage with diaspora is becoming an increasingly salient issue in global politics. As Bernal (2006, p. 161) points out, ‘diasporas online may invent new forms of citizenship, community and political practices’. Chinese diasporas in the US, for example, are found to be relying on emotion-laden conservative narratives of American politics on WeChat, shaping the ‘new conservatism’ of Chinese immigrants as the emerging political constituency in the US (Zhang, 2018). In the case of Xinjiang, Uyghur diaspora network on TikTok and Twitter, using hashtags such as #MeTooUyghur to sustain a Uyghur identity and to counter Chinese state narrative (Ramzy, 2019); whereas Han Chinese diasporas are receiving the state-sanctioned narrative about Uyghurs on heavily censored WeChat, which inadvertently shapes their political identity and practice.

66  Handbook of digital politics

FADING BORDERS AND INSTITUTIONS: THE RISE OF (GLOBAL) DIGITAL CITIZENS AND EVERYDAY (GLOBAL) POLITICS Although as the preceding section argues, nation-states are still the main protagonists in curating transnational political discussion, non-state actors are increasingly seen on the front stage, actively engaging with global political issues, negotiating with, resisting, or reproducing the political narrative carefully curated by state actors or elites. A prominent feature of the transnationality of digital political discussion is the rise of digital citizens, whose self-actualizing style of civic participation outside institutional structures has been recognized by scholars as their curatorial agency (Bennett et al., 2011; Coleman, 2008; Penney, 2017). With late modern societies featuring the fading of institutions and organizational ties being replaced by large-scale, fluid social networks (Castells, 2004), an increasing number of studies focus on the non-traditional actors in transnational/global political communication – activists, ordinary citizens, or what scholars term as ‘digital citizenship’ (McCosker et al., 2016; Vromen, 2017). Citizens are engaging with politics across national borders on social media (Bossetta et al., 2017). In the aforementioned digital diasporic politics, for example, diaspora also actively network in resistance, as observed among Uyghur diaspora (Bonnenfant, 2022), Syrian diaspora (Tenove, 2019), and Iranian diaspora (Michaelsen, 2018). Social media does not only break down the geographic borders, thus facilitating the global flows of digital citizens’ political discussion, but also has given rise to the personalization of politics, or ‘everyday politics’ (Bennett, 2004; Boyte, 2004; Highfield, 2016). Never before have individuals been afforded such a degree of (convenient) agency in shaping political discourse online. Below I map out two main strands of scholarly discussion on the curatorial agency of citizens in global political discussion: global activism and everyday politics. Global Digital Activism Social movement studies have long identified the transnational dimension of activism (Bennett, 2005; Tarrow, 2005). Minor political players, for example, have been able to ‘reframe their disadvantageous relationships with the nation-states that encompass them by redefining their projects in the global space of environmentalism and human rights’ (Kearney, 1995, p. 560). This transnationality of activism is only bolstered by the digitalization of communication especially the Internet, as digitally mediated movements are decentralized, de-institutionalized, thus more inclusive and less costly for individuals to participate across borders. From the Arab revolutions a decade ago to today’s global climate change activism, social media has witnessed how citizens across national borders are networked and mobilized to raise global awareness, further personal engagement, and eventually push for social change at local/national/global levels.

Transnational and global flows of political discussion online  67

Social media is said to have bred a new logic of connective action, featuring individualized and digitally mediated informal networks (Bennett and Segerberg, 2012). Young activists on social media such as Twitter, Facebook, and more recently TikTok, are seen as ‘affective publics’ intervening in political discussion of global issues such as climate change and social justice (Hautea et al., 2021). This new form of social media activism, although largely symbolic, easily becomes ‘viral’ across platforms and national borders, thanks to the easy imitation and personal adaptation of these symbolic yet resistant messages (Bennett and Segerberg, 2012), and ‘the strength of weak ties’ breeds new forms of resistance, mobilizing people on an unprecedented scale to express political opinion (Cammaerts, 2015). Such connectivity facilitates translocal and transnational community building for social justice movements and activism (Sobré-Denton, 2016). As illustrated in the Xinjiang case, a 40-second-long video produced by an American teenager posted on TikTok overnight drew millions of viewers across platforms and national borders and grew into a global campaign joined by many other teenagers from around the world, because the message is easy to pass on, easy to view, and is for social justice. When global activism becomes a ‘political way of being’ (Castells, 2004, p. 154, quoted by Chadwick, 2006), unavoidably, not only progressive forces see social media as a breeding ground. Conservatives and populists also network on a transnational, if not global, scale. An emerging body of scholarship has shed light on how far-right activists network across Europe collaborating on an anti-immigrant narrative (Doerr, 2017), how alt-right activists use hashtag-based framing and political jamming to build a transnational community on Twitter (Xu, 2020), and how China’s nationalistic activists crusade Facebook to disrupt public discussion on Taiwan’s elections (Fang and Repnikova, 2018). Meanwhile, with the ease of dissemination and emotional charge on social media, a just social movement can be ‘hijacked’ to twist the political message, as is found in the case of transnational far-right actors hijacking the #MeToo movement (Knüpfer et al., 2022). Not only traditionally ‘minor’ political players, but also elites such as political journalists and pundits have also found a much wider and closer global audience base online. Readers can find abundant scholarship on how political journalism and political elites are increasingly relying on the active public online, which eventually reshape their positions in the field of political communication. Lifestyle and Consumerist Politics Political discussion online is now increasingly happening on non-political digital platforms, or what Wright (2012) conceptualizes as ‘third space’. This ‘third space’ is now further permeating into everyday life, as social media have made ‘everyday politics’ a convenient and personalized topic to engage with (Highfield, 2016) and everyday life becomes ‘politically contestable’ (Bennett, 2003; Chadwick, 2006). Apart from the previous example of TikTok, originally used for sharing funny short videos among young people, another prominent area is how global capitalism to a certain extent networks political discussion on a global scale. Indeed, when we take

68  Handbook of digital politics

into consideration the globalization of both communication and economic systems, tapping into global capitalism and late modern politics, it is hardly surprising to see the rise of a networked, individualized lifestyle politics (Bennett, 2004; Giddens, 1991). Citizen-consumers and multinational corporations are injecting political messages into production, marketing, and consumption, featuring what scholars call ‘political consumerism’ (Micheletti et al., 2004; Scammell, 2000), ‘branded lifestyle politics’ (Bennett, 2004), or the ‘new global political consumerism’ (Ward and de Vreese, 2011). Corporations brand political messages to woo global consumers; concurrently consumers also generate counter messages to boycott corporations whose political branding does not appeal to them. As Bennett (2004, p. 102) notes, ‘attaching political messages to corporate brands becomes a useful way to carry often radical ideas into diverse personal life spaces, as well as across national borders and cultural divides’. In the Xinjiang case, the active role of consumers both in China and in the West is evident in negotiating the political narrative on a global scale. Human rights groups outside China are calling for a boycott of global brands that use Xinjiang cotton, setting the agenda for traditional media, further pressuring businesses to take a stance via social media. Whereas in China, social media users are voicing discontent with these brands and rallying support for Xinjiang cotton. By calling for boycotting businesses that boycott Xinjiang cotton, Chinese consumers on social media are exerting a political identity which distinctively aligns with the state and making counter claims in the international community. Either for Western or Chinese citizen-consumer-activists, this ‘cross-border social media fashion crisis’ (as the New York Times calls it) in essence brands ‘Xinjiang cotton’ into a politically symbolic logo, wherein the complicated human rights and labour exploitation issue has now been simplified and reduced to a hardly traceable sourcing issue in the apparel industry. This echoes what Bennett (2004) sees as a strength of ‘branded political consumerism’, in that otherwise distant and complex political issues are branded in a much more communicative way for general publics ‘who may be more sensitive to their fashion statements than they are interested in the brute political logic of global economics’ (Bennett, 2004, pp. 106–107). Besides, as shown in the Xinjiang case, in authoritarian capitalist countries where citizens do not have the luxury of ‘conventional forms of political engagement’ such as voting but do enjoy a rich variety of fashion and shopping options, lifestyle or consumerist politics is the most accessible form of political expression, although in many cases a nationalistic expression rather than progressive one. Nonetheless, this further adds nuances and noises to the global political discussion online. Scholars studying lifestyle and consumerist politics on a global scale tend to use ‘global citizenship’ (Bennett, 2004) to refer to the broad coalitions of groups across borders using campaigns against corporations to press for social justice. However, in this Xinjiang case, we can clearly see that when authoritarian nation-states enter the picture, it gets more intricate than one monolith of ‘global citizenship’ pushing for corporate responsibility or social justice. Future studies should look at the nuances of and new tensions between these different layers of ‘global citizenship’,

Transnational and global flows of political discussion online  69

where, for example, citizen-consumers from authoritarian capitalist countries do see their political identity changing with interconnected global economies and use their options in lifestyle and consumerism to attach political messages to corporate brands across borders, but in quite opposite direction from the activism of Western citizen-consumers. In such cases, global businesses are left in limbo juggling the contentious political identities infused in consumerist politics. It is an interesting time to reflect on the new cultural and political implications of global capitalism.

INTERCONNECTED AND CONTENTIOUS: THE GREAT CONVERGENCE OR DECOUPLING? Thus far I have presented a messy picture of the global flows of political discussion unfolding online in one or multiple ‘hybrid media systems’ (Chadwick, 2013): on both international media platforms and national ones, from both democracies and authoritarian regimes. Further complicating the picture is the unprecedented multiplicity and multimodality of actors, not only genuine actors, either institutional or individual, but also bots masquerading as concerned citizens. What is emerging is not a global public sphere, as some scholars had hoped for (see, e.g. Volkmer, 2003). Instead of a transnational dialectic space, what we see can be best described as a contested social field with competing power relations, increasingly dependent on what Castells (2007) calls ‘socialised communication’ or ‘mass self-communication’ which connects individuals and individuals, individuals and institutions, and local and global sites. Messages, either conforming to norms or ‘alternative’, either progressive or reactionary, oftentimes emotion-laden, are ‘globally distributed and globally interactive’ (Castells, 2007, p. 248), contributing to both public spheres and ‘anti-public spheres’ (Cammaerts, 2007; Davis, 2021), each and every one of these spaces permeating across national borders in different ways in cyberspace. Although this chapter only focuses on the research areas relating to state actors and citizens, it is important to note the vital role of other non-state actors – most notably journalism. Institutional media in transnational political communication are still a key reality definer for digital citizens when it comes to international news and global politics (Zeng, 2022). Nevertheless, it is the changing curative agency of individual citizens that functions as the breeding ground for the changing dynamics of how media and state participate in political discussion. For example, the two forms of state actors’ evolving practice of curating political discussion online – i.e. digital (dis)information and diasporic politics – are, in essence, a response to the rise of active global digital citizens. Considering the connective and contentious nature of today’s messy global political communication online, are we striding into a ‘great convergence’, as Castells calls it, or a ‘great decoupling’ of the political cyberspace between nations and cultures? The state of global online political discussion showcases multiple distinctive features, as this chapter demonstrates: multidirectional, sometimes borderless, yet still largely nation-state centred. While we may live in a global network society,

70  Handbook of digital politics

undeniably resources are still in the hands of states and elites, who can either regulate or manipulate the transnational or global flows of political discussion in cyberspace. Indeed, cyberspace poses serious challenges to the traditional nation-state based regulatory model, but as the case of authoritarian China shows, by setting up digital barriers1 and co-opting citizens into state-sanctioned narrative, the influx of political information can be curbed and remoulded. But no decoupling is without loopholes in this digital age, especially when the global economy is so intertwined and digital citizenship is in the making. Again, think about TikTok, a Chinese-owned digital platform, accessible only outside China, yet enabling global citizens to advocate for Uyghur Muslims’ human rights in China. Even for the Chinese Internet, the sophisticated ‘Great Firewall’ inherently has cracks; alternative political discussion from transnational sources has always existed in the form of what could be called digital samizdat (Zeng, 2022). Volkmer (2003) uses the phrase ‘reciprocal political communication’ to refer to the information flow where critical political content which is censored in one country can be passed on via dispersed worldwide servers in order to bypass censorship. In this way, the networked and digital affordances of global digital citizens create an incubation for critical voices which operate a ‘digital exodus’, and sustain the networks and resources for critical voices (Zeng, 2022). The transnational flows of political discussion thus never really terminate even for highly censored domestic cyberspace. Increasingly, it is these multidirectional flows that eventually shape/reframe domestic political discussion at national and local levels. The global flows of political communication need to be further researched from the paradigms of transnationality and network. Digital ICTs have greatly facilitated the political engagement of global citizens, but also posed new challenges for achieving a globalized public sphere, as this chapter argues. The interconnected and relational power dynamics are structured in a global network and, as Castells (2007, p. 249) says, ‘played out in the realm of socialized communication, social movements … act on this global network structure and enter the battle over the minds by intervening in the global communication process’. It is the connectiveness and contention together that make the field important and fascinating to study.

NOTE 1.

The Internet in China, for example, is so disconnected from the rest of the world that it is nicknamed the ‘great firewall’.

FURTHER READING Bennett, W. L. and Segerberg, A. (2013). The Logic of Connective Action: Digital Media and the Personalization of Contentious Politics. Cambridge, UK: Cambridge University Press. Bradshaw, S. and Howard, P. N. (2019). Gilboa: 2019 Global Inventory of Organised Social Media Manipulation. Oxford Internet Institute.

Transnational and global flows of political discussion online  71

Castells, M. (2013). Communication Power. New York: Oxford University Press. Gilboa, E. (2016). Digital diplomacy. In C. M. Constantinou, P. Kerr, and P. Sharp (eds.), The SAGE Handbook of Diplomacy (pp. 540–551). London: Sage. Jones, A. and Kovacich, G. L. (2015). Global Information Warfare: The New Digital Battlefield. Boca Raton, FL: CRC Press. Juris, J. (2004). Networked social movements: The movement against corporate globalization. In M. Castells (ed.), The Network Society: A Cross-Cultural Perspective (pp. 341–362). Cheltenham, UK and Northampton, MA, USA: Edward Elgar Publishing. Moy, P., Bimber, B., Rojecki, A., Xenos, M. A., and Iyengar, S. (2012). Transnational connections: Shifting contours in political communication research. International Journal of Communication, 6, 247–254. Penney, J. (2017). The Citizen Marketer: Promoting Political Opinion in the Social Media Age. New York: Oxford University Press. Smith, M. P. and Guarnizo, L. E. (eds.) (1998). Transnationalism from Below. New Brunswick, NJ: Transaction Books. Tarrow, S. (2005). The New Transnational Activism. New York: Cambridge University Press. Trottier, D. and Fuchs, C. (eds.) (2015). Social Media, Politics and the State: Protests, Revolutions, Riots, Crime and Policing in the Age of Facebook, Twitter and YouTube. New York and London: Routledge. Volkmer, I. (2003). The global network society and the global public sphere. Development, 46(1), 9–16.

REFERENCES Adamson, F. B. and Tsourapas, G. (2019). Migration diplomacy in world politics. International Studies Perspectives, 20(2), 113–128. Applebaum, A. (2020). Twilight of Democracy: The Seductive Lure of Authoritarianism. New York: Doubleday. Baldwin-Philippi, J. (2019). The technological performance of populism. New Media & Society, 21(2), 376–397. Baser, B. and Ozturk, A. E. (2020). Positive and negative diaspora governance in context: From public diplomacy to transnational authoritarianism. Middle East Critique, 29(3), 319–334. Benkler, Y., Faris, R., and Roberts, H. (2018). Network Propaganda: Manipulation, Disinformation, and Radicalization in American Politics. New York: Oxford University Press. Bennett, W. L. (2003). Communicating global activism. Information, Communication & Society, 6(2), 143–168. Bennett, W. L. (2004). Branded political communication: Lifestyle politics, logo campaigns, and the rise of global citizenship. In M. Micheletti, A. Follesdal, and D. Stolle (eds.), Politics, Products, and Markets: Exploring Political Consumerism Past and Present (pp. 101–125). New Brunswick, NJ: Transaction Books. Bennett, W. L. (2005). Social movements beyond borders: Understanding two eras of transnational activism. In D. Della Porta and S. Tarrow (eds.), Transnational Protest and Global Activism (pp. 203–226). Lanham, MD: Rowman & Littlefield. Bennett, W. L. and Livingston, S. (2018). The disinformation order: Disruptive communication and the decline of democratic institutions. European Journal of Communication, 33(2), 122–139. Bennett, W. L. and Segerberg, A. (2012). The logic of connective action: Digital media and the personalization of contentious politics. Information, Communication & Society, 15(5), 739–768.

72  Handbook of digital politics

Bennett, W. L., Wells, C., and Freelon, D. (2011). Communicating civic engagement: Contrasting models of citizenship in the youth web sphere. Journal of Communication, 61(5), 835–856. Bernal, V. (2006). Diaspora, cyberspace and political imagination: The Eritrean diaspora online. Global Networks, 6(2), 161–179. Blumler, J. G. (2016). The fourth age of political communication. Politiques de communication, 6(1), 19–30. Bjola, C. and Holmes, M. (eds.) (2015). Digital Diplomacy: Theory and Practice. Abingdon: Routledge. Bonnenfant, I. K. (2022). Stateless Diasporas and China’s Uyghur Crisis in the 21st Century. In Oxford Research Encyclopedia of International Studies. Bossetta, M., Dutceac Segesten, A., and Trenz, H. J. (2017). Engaging with European politics through Twitter and Facebook: Participation beyond the national? In M. Barisione and A. Michailidou (eds.), Social Media and European Politics (pp. 53–76). London: Palgrave Macmillan. Boyte, H. C. (2004). Everyday Politics: Reconnecting Citizens and Public Life. Philadelphia: University of Pennsylvania Press. Bradshaw, S. and Howard, P. N. (2018). The global organization of social media disinformation campaigns. Journal of International Affairs, 71(1.5), 23–32. Cammaerts, B. (2007). Jamming the political: Beyond counter-hegemonic practices. Continuum: Journal of Media & Cultural Studies, 21(1), 71–90. Cammaerts, B. (2015). Social media and activism. In R. Mansell and P. Hwa (eds.), The International Encyclopedia of Digital Communication and Society (pp. 1027–1034). Oxford: Wiley-Blackwell. Castells, M. (2004). The Network Society. Oxford: Blackwell. Castells, M. (2007). Communication, power and counter-power in the network society. International Journal of Communication, 1(1), 238–266. Castells, M. (2012). Networks of Outrage and Hope: Social Movements in the Internet Age. Cambridge: Polity Press. Castells, M. (2013). Communication Power. New York: Oxford University Press. Chadwick, A. (2006). Internet Politics: States, Citizens, and New Communication Technologies. New York: Oxford University Press. Chadwick, A. (2013). The Hybrid Media System: Politics and Power. Oxford: Oxford University Press. Coleman, S. (2008). Doing IT for themselves: Management versus autonomy in youth e-citizenship. In W. L. Bennett (ed.), Civic Life Online: Learning How Digital Media Can Engage Youth (pp. 189–206). Cambridge, MA: MIT Press. Cooper, A. F. (2019). Adapting public diplomacy to the populist challenge. The Hague Journal of Diplomacy, 14(1–2), 36–50. Davis, M. (2021). The online anti-public sphere. European Journal of Cultural Studies, 24(1), 143–159. Doerr, N. (2017). Bridging language barriers, bonding against immigrants: A visual case study of transnational network publics created by far-right activists in Europe. Discourse & Society, 28(1), 3–23. Duncombe, C. (2019). Digital diplomacy: Emotion and identity in the public realm. In J. Melissen and J. Wang (eds.), Debating Public Diplomacy: Now and Next (pp. 102–116). Leiden: Brill. Fang, K. and Repnikova, M. (2018). Demystifying “Little Pink”: The creation and evolution of a gendered label for nationalistic activists in China. New Media & Society, 20(6), 2162–2185. Fuchs, C. (2018). Authoritarian capitalism, authoritarian movements and authoritarian communication. Media, Culture & Society, 40(5), 779–791.

Transnational and global flows of political discussion online  73

Giddens, A. (1991). Modernity and Self-Identity: Self and Society in the Late Modern Age. Cambridge: Polity Press. Hall, T. H. (2015). Emotional Diplomacy: Official Emotion on the International Stage. Ithaca, NY: Cornell University Press. Hautea, S., Parks, P., Takahashi, B., and Zeng, J. (2021). Showing they care (or don’t): Affective publics and ambivalent climate activism on TikTok. Social Media + Society, 7(2). https://​doi​.org/​10​.1177/​20563051211012344. Held, D., McGrew, A., Goldblatt, D., and Perraton, J. (2000). Global Transformations: Politics, Economics and Culture. Basingstoke: Palgrave Macmillan. Highfield, T. (2016). Social Media and Everyday Politics. Cambridge: Polity Press. Hirt, N. and Saleh Mohammad, A. (2018). By way of patriotism, coercion, or instrumentalization: How the Eritrean regime makes use of the diaspora to stabilize its rule. Globalizations, 15(2), 232–247. Ho, E. L. and McConnell, F. (2019). Conceptualizing ‘diaspora diplomacy’: Territory and populations betwixt the domestic and foreign. Progress in Human Geography, 43(2), 235–255. Holmes, M. (2015). Digital diplomacy and international change management. In C. Bjola and M. Holmes (eds.), Digital Diplomacy: Theory and Practice (pp. 27–46). Abingdon: Routledge. Jack, C. (2017). Lexicon of Lies: Terms for Problematic Information. Data & Society Research Institute. Jamieson, K. H. (2018). Cyberwar: How Russian Hackers and Trolls Helped Elect a President: What We Don’t, Can’t, and Do Know. New York: Oxford University Press. Kearney, M. (1995). The local and the global: The anthropology of globalization and transnationalism. Annual Review of Anthropology, 24(1), 547–565. Knüpfer, C., Hoffmann, M., and Voskresenskii, V. (2022). Hijacking MeToo: Transnational dynamics and networked frame contestation on the far right in the case of the ‘120 decibels’ campaign. Information, Communication & Society, 25(7), 1010–1028. Kragh, M. and Åsberg, S. (2017). Russia’s strategy for influence through public diplomacy and active measures: The Swedish case. Journal of Strategic Studies, 40(6), 773–816. Laguerre, M. S. (2010). Digital diaspora: Definition and models. In A. Alonso and P. Oiarzabal (eds.), Diasporas in the New Media Age: Identity, Politics, and Community (pp. 49–64). Reno: University of Nevada Press. Lukito, J. (2020). Coordinating a multi-platform disinformation campaign: Internet Research Agency activity on three US social media platforms, 2015 to 2017. Political Communication, 37(2), 238–255. Manor, I. (2019). The Digitalization of Public Diplomacy. Cham: Palgrave Macmillan. McCosker, A., Vivienne, S., and Johns, A. (eds.) (2016). Negotiating Digital Citizenship: Control, Contest and Culture. Lanham, MD: Rowman & Littlefield. Michaelsen, M. (2018). Exit and voice in a digital age: Iran’s exiled activists and the authoritarian state. Globalizations, 15(2), 248–264. Micheletti, M., Follesdal, A., and Stolle, D. (eds.) (2004). Politics, Products, and Markets: Exploring Political Consumerism Past and Present. New Brunswick, NJ: Transaction Books. Moore, M. (2018). Democracy Hacked: Political Turmoil and Information Warfare in the Digital Age. London: Oneworld. Moss, D. M. (2018). The ties that bind: Internet communication technologies, networked authoritarianism, and ‘voice’ in the Syrian diaspora. Globalizations, 15(2), 265–282. Moy, P., Bimber, B., Rojecki, A., Xenos, M. A., and Iyengar, S. (2012). Transnational connections: Shifting contours in political communication research. International Journal of Communication, 6, 247–254.

74  Handbook of digital politics

Murti, B. and Zaharna, R. S. (2014). India’s digital diaspora diplomacy: Operationalizing collaborative public diplomacy strategies for social media. Exchange: The Journal of Public Diplomacy, 5(1), 3–29. Ong, A. (1999). Flexible Citizenship: The Cultural Logics of Transnationality. Durham, NC: Duke University Press. Ott, B. L. (2017). The age of Twitter: Donald J. Trump and the politics of debasement. Critical Studies in Media Communication, 34(1), 59–68. Penney, J. (2017). The Citizen Marketer: Promoting Political Opinion in the Social Media Age. New York: Oxford University Press. Rojecki, A. (2011). Leaderless crowds, self-organizing publics, and virtual masses: The new media politics of dissent. In S. Cottle and L. Lester (eds.), Transnational Protests and the Media (pp. 87–97). New York: Peter Lang. Scammell, M. (2000). The internet and civic engagement: The age of the citizen-consumer. Political Communication, 17(4), 351–355. Schia, N. N. and Gjesvik, L. (2020). Hacking democracy: Managing influence campaigns and disinformation in the digital age. Journal of Cyber Policy, 5(3), 413–428. Sobré-Denton, M. (2016). Virtual intercultural bridgework: Social media, virtual cosmopolitanism, and activist community-building. New Media & Society, 18(8), 1715–1731. Sun, W. and Yu, H. (2020). WeChatting the Australian election: Mandarin-speaking migrants and the teaching of new citizenship practices. Social Media + Society, 6(1). https://​doi​.org/​ 10​.1177/​205630512090344. Tarrow, S. (2005). The New Transnational Activism. New York: Cambridge University Press. Tenove, C. (2019). Networking justice: Digitally-enabled engagement in transitional justice by the Syrian diaspora. Ethnic and Racial Studies, 42(11), 1950–1969. Tenove, C. (2020). Protecting democracy from disinformation: Normative threats and policy responses. The International Journal of Press/Politics, 25(3), 517–537. Tsourapas, G. (2021). Global autocracies: Strategies of transnational repression, legitimation, and co-optation in world politics. International Studies Review, 23(3), 616–644. Volkmer, I. (2003). The global network society and the global public sphere. Development, 46(1), 9–16. Vromen, A. (2017). Digital Citizenship and Political Engagement. London: Palgrave Macmillan. Ward, J. and de Vreese, C. (2011). Political consumerism, young citizens and the Internet. Media, Culture & Society, 33(3), 399–413. Wright, S. (2012). From “third place” to “third space”: Everyday political talk in non-political online spaces. Javnost: The Public, 19(3), 5–20. Xu, W. W. (2020). Mapping connective actions in the global alt-right and Antifa counterpublics. International Journal of Communication, 14(1), 1070–1091. Zeng, Y. (2023). Networked frame contestation from authoritarian to democracy: A case of China’s (failed) twiplomacy in contesting coronavirus narrative in the UK. European Journal of Communication. https://​doi​.org/​10​.1177/​0267323122115087. Zhang, C. (2018). WeChatting American Politics: Misinformation, Polarization, and Immigrant Chinese Media. Tow Center for Digital Journalism, Columbia University. Ziegler, C. E. (2018). International dimensions of electoral processes: Russia, the USA, and the 2016 elections. International Politics, 55(5), 557–574.

PART II CITIZENSHIP AND POLITICAL TALK

6. The Internet as a civic space Peter Dahlgren

Democracy is dependent upon the participation of its citizens, and such participation requires a variety of sites, places, and spaces. When the Internet emerged in the mid-1990s as a mass phenomenon, some observers dismissed it as insignificant for politics. However, it soon became apparent that this communication technology was to play an increasingly important role in the life of democracy. Yet today, as we shall see, there remains considerable contention as to just exactly what this role is, and whether or not the Internet ultimately is beneficial for democracy. (I signal here at the outset that for ease of exposition I use the term ‘Internet’ in a very broad way, to refer to both the hardware and software of this technical infrastructure, and to include such ancillary technologies as mobile telephony and the various platforms of social media.) As politics in society generally takes on a larger presence online, the prevailing structures of established power in society are increasingly mediated, solidified, negotiated and challenged to a great extent via the Internet. From the horizons of democracy, how should we view these developments? This chapter probes answers to that question.

THE SIGNIFICANCE OF THE INTERNET: CIVIC SPACES AND EVOLVING DEMOCRACY A Conceptual Continuum At an obvious level, the Internet, given its societal ubiquity, has become an understandably significant communication technology for civic space and for the functioning of democracy. However, to grasp this in a more analytic way, and to understand the issues that nonetheless arise in the process, let me begin by very briefly sketching some important conceptual background. First of all, it is important to remember that ‘democracy’ is both a complex and a contested term. There are not only a range of differing political systems in the world that claim to be democratic, but also, and more pertinent to my purposes here, there are different ideal models (see Held, 2006, for an overview). Without going through an entire inventory, I here simply note a decided polarity between two basic ways of looking at democracy, each with its own view of civic engagement – though it is probably more useful to think of the distinction as a continuum, rather than a simple either–or choice. On the one hand we have what is sometimes called elite democracy; its proponents take the view that the system works best via a rotation of various elite groups who come to power through elections, and where most citizens, aside from voting, do not engage themselves 76

The Internet as a civic space  77

much politically. Here civic participation is seen largely in terms of a formalized system based on elections. Alternatively there are various versions of republican models (see, for example, Dewey, 1923; Barber, 1984; Mouffe, 2005) that emphasize the ideal that citizens should engage themselves politically as much as possible, not just at election time. It is argued that such engagement is good not only for the vitality of democracy, but also for the individual citizen, since it offers potential for personal growth and development. In this perspective democratic involvement is understood as comprising not just an electoral system, but much larger societal domains. The adjective ‘democratic’ is something that should describe a society more generally, not just its voting mechanisms; democratic processes are seen as a part of an ongoing daily reality. Thus, while engagement in elections certainly requires civic spaces of various kinds, elite models put less emphasis on the need for such spaces beyond the context of electoral politics. Republican versions of democracy, on the other hand, underscore the significance of a broad and dynamic array of civic spaces. The distinction between elitist and republican models manifests itself also in the actual character of participation, that is, what actually goes on in civic spaces. Elite models highlight citizens’ needs for information, news, commentary and debate, in order to make (rational) voting decisions. Republican models concur but also demand a more participatory character of civic spaces, seeing them not just as sites where information can be obtained, but also as opportunities where citizens can interact, develop a sense of common interests, sharpen their opinions, and even engage in forms of decision-making. We see, in other words, a distinction in the ideal of the citizenship itself: reactive and restricted, versus proactive and robust. These are of course generalized and abstract conceptions, yet they inform, on a subtle and often unconscious way, the manner in which different kinds of power holders as well as citizens act. As a further context for the discussion at hand, in the past 25 years or so there has been growing international concern about democracy’s difficulties; indeed, the situation is often referred to as a crisis. This crisis is as complex as democracy itself, but for my purposes one basic feature is the decline in civic engagement in both the politics arena and the larger domain of what is often called civil society (which I will come to shortly). Not least, under the contemporary policy logic of neoliberalism, where representative democratic power is eroded and accumulates increasingly in unaccountable ways in the private corporate sector (Harvey, 2006), the grounds for trust and participation are eroded (Hay, 2007), as are societal norms central for democracy (Sandel, 2012). Parallel with these challenges, however, we have witnessed a growth in what can loosely be called ‘alternative politics’ that in various ways bypasses the electoral system (Rosanvallon, 2008, uses the term ‘counter-democracy’). Here political engagement lies outside of party structures, and both the issues that become politicized and the modes of engagement are evolving: the political becomes more closely linked to personal meaning, identity processes, and issues that often have to do with cultural matters (see, for example, Bennett and Segerberg, 2012). These transfor-

78  Handbook of digital politics

mations have served not least to focus attention specifically on the nature of civic engagement and its circumstances (a concern which is still very much with us; see Schachter and Yang, 2012). In these discussions, the media loom large, even if they are only part of the story. I can now go further and begin to make the notion of ‘civic spaces’ itself a bit more concrete by mobilizing two key terms that derive from several different trajectories in political theory. They can provide some helpful roadmaps, and they function well together, pointing to two kinds of civic spaces: the first is civil society, the second is the public sphere. Civil Society and the Public Sphere With ‘civil society’ (see Edwards, 2009, for an introductory overview) I refer to an eclectic tradition in democracy theory that accentuates citizens’ free association for common purpose outside the private sphere of the home, and independent of the market and the state. There are undeniably some unresolved issues with the concept, but the idea of civil society emphasizes that in a democracy people can exercise the freedom to interact in pursuit of their shared interests, in settings that are protected by the rights of expression and assembly. For example, dealing with friends, colleagues, communities, associations and social networks for non-commercial purposes are all a part of civil society. On the Internet, and especially in the context of social media, there is an almost infinite realm of shared engagement in meaningful and pleasurable activities around sports, hobbies, music (for example, amateur contributions on YouTube), fandom, wikis and so forth – though it is often not possible to keep market logics completely out. Thus, on one border, civil society has a porous demarcation between itself and what we can broadly call consumption, that is, commercial logics. Its other border is with politics: the political may arise in civil society settings, transforming them into what we call the public sphere, that is, the communicative space of politics. At what point the political actually emerges can be difficult to specify; most fundamentally, it materializes through talk: as people speak, topics may turn to – and become – political. At that point, conceptually, one could say that the discussion has entered the public sphere. Indeed civil society is important not just for the interaction and association it facilitates, but also precisely because it is in a sense a precondition for a functioning public sphere: without that free association, the public sphere could not survive (Cohen and Arato, 1992, underscore the links between the two domains in their classic treatment). Civil society comprises the sites where people can enact their roles as citizens, talk and work together; for this to happen there must exist a minimal foundation of trust and shared democratic values. Without such a sense of civic community and solidarity, civil society evaporates, undercutting democracy’s communicative dynamics (Alexander, 2006). If civil society atrophies, so does the vitality of democracy, as Putnam (2000) has famously argued. The concept of the public sphere, while having a somewhat mixed lineage, is more cohesive than the notion of civil society. The key text in English is Habermas (1989),

The Internet as a civic space  79

although since its first publication in the early 1960s there has been much debate on the theme, and Habermas himself has modified his views somewhat over the years. However, in simplified terms we can say that today the notion of the public sphere has become a key conceptual pillar in linking the media to democracy in a normative and critical way. As a normative ideal the public sphere is seen as the institutionalized communicative spaces that are accessible to all citizens and that help to promote the development of public opinion and political will formation. These public processes are to take place through the unhindered access to pertinent information, ideas and debate. In the modern world, much of the public sphere is comprised of media, especially in the form of journalism, yet face-to-face contexts remain essential, since this is where discussion and debate between citizens take place (and we can readily understand that the Internet has been offering mediated extensions to such civic deliberation). Habermas in his book proceeds to examine how various historical factors have served to constrict this ideal, not least the commercial logic of the media. Analysts have continued to use the concept as both a normative horizon and an empirical referent to be critically evaluated, especially with a strong emphasis on the affordances, limitations and actual modes of use of the Internet. Habermas and others make clear that the public sphere is far from unitary; empirically, it is comprised of many sprawling communicative spaces of considerable variety (see Habermas’s update from 2006). At the same time, these heterogeneous spheres are by no means equal in terms of access or political impact. Some are socially and politically more mainstream, and situated closer to decision-making power. Others are more geared toward the interests and needs of specific groups; emphasizing, for example, the need for collective group identity formation and/or the ambition to offer alternative political orientations, that is, subaltern, counter-public spheres (see Fraser, 1992). If one of the key normative elements of the public sphere is the ideal of universal access which permits citizens to participate in democracy, it is precisely on this point where much difficulty is encountered: ostensibly democratic societies have a variety of formal – but often informal – mechanisms that hinder democratic participation in civic spaces. With their emphasis on participation in broader societal contexts, republican versions of democracy push for a broader understanding of the political, one that readily extends beyond electoral politics and can potentially insert itself in just about any societal context where contention can arise. This ties in with another question that I will look at: if the public sphere is the communicative space of politics, what should civic engagement look like – how should political talk ideally proceed? As we shall see, the Internet becomes very salient in these discussions. To pull together the discussion thus far, we have the ideal of democracy, which can be understood as leaning towards more elite or more republican versions; the latter tendency underscores the importance of civic participation not only in elections but in the larger societal terrain as well. The role of citizens is today cogently actualized by the current crisis in democracy, where civic participation has become a central theme. Electoral politics is going through a difficult time; alternative politics,

80  Handbook of digital politics

although seemingly more robust, has a tenuous track record of success. Moreover, the character of participation, and of politics itself, is transforming, as social and cultural foundations of democracy become refigured; not least, there exist a variety of exclusionary mechanisms that obstruct universal participation in civic spaces. At some point the issue of the power relations that shape civic spaces becomes pertinent. Clearly the Internet figures prominently here, residing in a force-field of different premises and views about the political world generally and civic spaces in particular. The manner in which we might perceive the Internet’s normative and actual role is inevitably to some extent linked to how we view the contested and moving analytic target of democracy. I now turn to the key currents of research on the Internet as a civic space.

RESEARCH FINDINGS – AND CONTESTATIONS No Techno-Fix Since the mass-circulation printed press became an essential feature of democratic life in the nineteenth century, the media have been entwined with power structures, serving both to promote and encourage civic participation as well as to limit and deflect it. The specifics of course have varied greatly between different contexts and with changing circumstances. Almost every major revolution in media technology – radio, television, CB radios, desktop publishing, computers, Internet, Web 2.0 – has been accompanied by a rhetorical promotion of the respective technology’s democratic benefits. While such claims are not necessarily wrong, there is often a basic fallacy involved, namely technological determinism; that is, a view that analytically puts technology in the driver’s seat and discounts the modifying impact of socio-cultural settings. This, in short, is the vision of the quick techno-fix, which implicitly suggests that democracy’s ills at bottom have to do with an insufficiency of apparatus. This was certainly noticeable in the first few years of the Internet, when the technology was so new, so startling in its affordances. ‘Armchair theorists’ could uninhibitedly proclaim all sorts of wondrous developments for society, and democracy in particular, that would derive from the net – or, alternatively, predict the end of both democracy and civilization as we know it. Gradually, however, the research findings began coming in towards the end of the 1990s, and the discussions began to take on sharper contours. Beyond Business as Usual Most researchers have from the start explicitly or implicitly suggested that the Internet is a boon for civil society: it permits and indeed promotes horizontal communication in society. Individuals, groups and organizations can get in touch with each other, even on a global level, and exchange ideas, experiences and support.

The Internet as a civic space  81

While abuse of such communicative freedom can never be fully eliminated, various efforts (with varying degrees of success) have attempted to regulate or discourage such behaviour (for example, harassment, privacy violations, child pornography). The debates became more pointed when the issue had to do with the Internet’s impact on the public sphere. One major trajectory here was ‘business as usual’, that the Internet’s role in the public sphere, and democracy more broadly, was and would remain quite modest (for example, Margolis and Resnick, 2000). This view from the late 1990s acknowledged that the major political actors may engage in online campaigning, lobbying, policy advocacy, organizing, and so forth, but did not see the net as a significant space for civic activity: overall, the political landscape would remain basically the same. A few years later it was also noted that various experiments, usually on a rather small scale, to incorporate the Internet as ‘e-democracy’ or ‘e-participation’ into local governments had not been hugely successful (see, for example, Malina, 2003; Gibson et al., 2004; Chadwick, 2006). What should be emphasized is that this overall perspective was anchored in the formal political system, and coloured by the traditional role of the mass media in that system. Indeed, much of the evidence is based on American electoral politics (for example, Hill and Hughes, 1998). This view, however, began to change, as it became more and more apparent that citizens were using the Internet for political engagement in various discussion forums and so-called news groups, and that this could have consequences for how they vote. Certainly by the time of Barack Obama’s election victory in 2008, where it was clear that the strategic use of social media had played a major role, the Internet was firmly in place as a terrain of relevance for established political parties. In regard to alternative politics, one point emerged quite early: without the Internet, the sprawling landscape of activist groups, advocacy organizations, social movements and political networks would have a very difficult time of it. How effective their impact was, has been, and is, remains contested; sceptics point to the (probable) low numbers of people who are actually involved in these activities, and the (generally) low impact they have. There are exceptions, of course: the Occupy movement in the autumn of 2011 spread from New York City across the USA and went global. However, by the spring of 2012 there was not much left – it had dissipated. On the other hand, the crisis within the European Union (EU) has mobilized many people to alternative politics in recent years, especially in Southern Europe, and these manifestations, where the net has an important role to play, have thus far had more longevity. Online Civic Spaces and Practices In the evolution of the Internet itself, three areas or domains of convergence can be specified (Meikle and Young, 2012). There is the fundamental – and incessant – technological convergence of computers and digital media, where older media are constantly being reformatted and upgraded to be compatible with the ever-evolving

82  Handbook of digital politics

new possibilities. This leads to the second area of convergence: organizational ones, fusing convergences, the older institutions of the mass media and the newer online actors, with constantly new mergers, new trade-offs and bankruptcies taking place, with a very few giants emerging to dominate the web landscape. Finally, there are convergences of form and content: multimedia (where words, images and sounds can be integrated on the same device by virtue of the shared digital language); transmedia (where the same content is dispersed across a variety of platforms); and mash-ups (which involve sampling, remixing and reconstituting texts). Thus, we need to think of the Internet as a dynamic, highly protean milieu; which in some ways becomes problematic for democracy, since it requires a degree of institutional stability. For researchers it was becoming clear not only that the Internet had become a prime site of civic spaces, but also that citizens’ practices were becoming very diversified; the affordances that allowed for easily achieved user-generated content (UGC) were promoting more active modes of participation. Many citizens active in online civic spaces were moving from mere interactivity to full-fledged ‘produsers’, where UGC was becoming all the more relevant for politics, in both its electoral and, especially, its alternative variants. Moreover, the Internet has become inseparable from the daily life and social worlds of citizens; it is hyper-ubiquitous: it is everywhere, used by (almost) everyone in democratic societies, for a seemingly endless array of purposes. For many people it is no longer something they merely visit or occasionally check: we see especially the younger age cohorts spending significant amounts of time on the net, socializing, pursuing all manner of information, engaged in consumption, entertainment, and so on. Everyday life is increasingly embedded to a great extent on the Internet; it is where much of it takes place. These developments are predicated on the interplay between the transformation of the Internet and the uses to which it is put. Lievrouw (2011) underscores the continuing interplay between the affordances of communication technologies and the practices by which people utilize them for their own purposes, resulting in a sort of dialectic between technological innovation and creative adaption. Strict adherence to the formal criteria of deliberative democracy, while laudable and relevant for specialized contexts, seem far removed from the realities of today’s political communication on the net. Further, the mobile character of the net has important consequences for how we live: while the importance of place does not simply vanish, its relevance is in many circumstances diminished by mediated connectivity. We are more accessible than before, and we become more portable and flexible. A good deal of our social coordination and organization can be carried out from a distance. Surveillance can also be enhanced by mobile technologies, by authorities for a variety of purposes (crime-fighting, political suppression, routine monitoring), by peers and by parents (who often want their children to carry a mobile phone). These developments have significance for civic space: civil society can take on a more ambulatory character, obviously enough, but for the public sphere the changes become more profound. At bottom, the boundaries between public and private space have become negotiable (Meikle and Young, 2012). Public space can now be ‘refor-

The Internet as a civic space  83

matted’ in a variety of ways and for different purposes, including the interjection of the private (for example, a personal conversation). Such modulations begin to alter the basic coordinates of our social geography. While mobile devices are often used for personal purposes, crossing the thin line to public and political contexts is easily and often done. The public sphere becomes less demarcated from other domains, a development many of the republican persuasion support, since it allows politics to more easily seep into other, less traditional areas; and indeed, with social media, to go viral. Enthusiasts and Sceptics In the large and diverse literature from recent years are found enthusiasts such as Benkler (2006), Castells (2010) and Shirky (2008). More sceptical and critical voices, who argue that the democratic possibilities of the web have been seriously oversold, are found in Fuchs (2011), Hindman (2009) and Morozov (2011). The enthusiasts are no doubt easier to understand (and to like). They pick up on the horizontal, civic society character of the Internet, with its open quality and participatory affordances. They note how this in turn meshes with the ideas of social networks as the new organizing logic of society (Castells, 2010), and how the sharing and collective wisdom typified by wiki-logics can empower citizens and strengthen democracy (Shirky, 2008). Such authors point to social media in particular as spaces where interaction can readily shift from personal encounters to commerce, to civil society activity, and not least to public sphere communication. They highlight the almost infinite amount of information and views available online, and how this empowers citizens and broadens the spectrum of the public sphere. Sceptics, for their part, contend that using the Internet for political activities (at least defined in traditional terms) is certainly one of the less frequent usages; politics generally comes far behind consumption, entertainment, social connections, pornography, and so on. Today the opportunities for such kinds of involvement are overwhelmingly more numerous, more accessible and more enticing for most people, compared to civic or political pursuits. Moreover, it has been shown that access to the Internet in itself does not turn people towards political issues; in fact, younger cohorts, who are the most net savvy, are less likely to do politics on the net than older age cohorts. It is also argued that the very density of the symbolic environments on the online public sphere becomes a distraction, and they lead to massive competitions for attention. For political actors using the Internet, getting and holding an audience is a constant challenge (the case of political bloggers is often mentioned: most seem to fizzle out after a short time, while the big heavy ones, tied to major media organizations online, have more staying power). Hindman (2009) and Morozov (2011) are among the voices who are adamant that the benefits of the Internet for democracy have been much exaggerated; the latter author in particular makes a strong case for seeing the net as a tool for authoritarian control, as witnessed in places such as China and Belarus, and he asserts that similar patterns in web use by the authorities are also emerging in the Western democracies.

84  Handbook of digital politics

The revelations that the National Security Agency in the USA, and similar organizations in other countries, engage in massive surveillance on citizens in democracies suggests that we have entered the post-privacy era (Greenwald, 2014). Other critics point to structural issues about the Internet. They argue that the net, the regulation around it, the major operators that define how it functions, and the various platforms available on it – not least social media – are shaped by the commercial imperatives of political economy and the power relations that derive from them, to the detriment of the character of these civic spaces. On an even more fundamental level, other authors such as Carr (2010) argue that the architectural logic of the net and its impact on our modes of cognitive functioning have a deleterious impact on our capacity to think, read and remember. If many observers laud how the participatory ‘wisdom of the many’ (as manifested, for example, in Wikipedia and the blogosphere) is producing new and better forms of knowledge, others such as Keen (2008) warn of the dangers, asserting that it erodes our values, standards and creativity, as well as undermining cultural institutions. In a related vein, the argument is often made that the strongly affective character of the multimedia Internet, particularly social media, also contributes to the decline of rationality in the public sphere. While emotionality is of course essential for political engagement, many observers note that it often tips over in a manner that is counterproductive for sound democratic politics (for example, populist discourses). Moreover, affect can become an easy way to bypass what seems like infinite amounts of information yielding ambivalence from sources one may not fully trust (Andrejevic, 2013). Problematic Political Economy The Internet is not just a technological device, it is also a socially organized institution, enmeshed in power relations; these features of its political economy, as mentioned above, impact greatly on its character. For instance, the Internet is profoundly affected by Google, which greatly shapes how the net operates and what we can do with it (Cleland and Brodky, 2011; Fuchs, 2011; Vaidhyanathan, 2011). Moreover it has become the largest holder of information in world history, shaping not only how we search for information, but also what information is available, and how we organize, store and use it. In many ways it is an utterly astounding development, yet it has also grown into an enormous concentration of power that is largely unaccountable, hidden behind the cheery corporate motto ‘Don’t be evil’. We all strew daily personal electronic traces; these are gathered up, stored, sold and used for commercial purposes by Google (and other actors). This selling of personal information is done with our formal consent, but if we refuse we effectively cut ourselves off from the major utilities of the Internet. Increasingly very serious questions are being raised, and those struggling to defend the interests of the public in regard to privacy have begun, at least indirectly, to confront Google’s agenda to organize knowledge on a global scale.

The Internet as a civic space  85

All this is not to detract from Google’s truly impressive accomplishments; rather, the issue is that the position it has attained, and the activities it pursues (which are quite logical given its position), raise questions about information, democracy, accountability and power in regard to the Internet. Just to take one example: given the logic of personal profiling – the filtering of results to ‘fit your known locality, interests, obsessions, fetishes, and points of view’ (Vaidhyanathan, 2011, p. 183) – the answers that two people will receive based on the same search words may well differ significantly. This can wreak havoc with the whole idea of shared public knowledge (Pariser, 2011), which in the long run can potentially undermine the democratic culture of debate between differing points of view. Facebook, now with about 1 billion users, also compiles massive amounts of data on individuals, largely freely given. As with Google, the data gathered is for commercial purposes, but again, changing social contexts can generate new uses and meanings of personal information. With Facebook, the spillover from private to public is much easier, resulting in embarrassment, entanglements, defamation or even death. Data theft is also easier; digital storage systems are simply not fail-safe, as witnessed when hackers have even entered high-security military databases. Thus, to participate in Facebook and similar social media is to expose oneself to surveillance and to have one’s privacy put at risk. Moreover, such digital information is not erased; it is archived, and can be retrieved and inserted into new – and troubling – contexts of a person’s life. As noted above, social media sites such as YouTube, Facebook and Twitter have become incorporated into political communication. They have become important outlets and sources for journalism, and are increasingly a part of the public sphere of both electoral and alternative politics. Not least, they have become the sites for massive marketing efforts, as Dwyer (2010) underscores. In Facebook’s role as a site for political discussion, one can reflect on the familiar mechanism of ‘like’: one clicks to befriend people who are ‘like’ oneself, generating and cementing networks of like-mindedness. As time passes, people increasingly habituate themselves to encountering mostly people who think like they do, and getting their biases reinforced. The danger arises that citizens lose the capacity to discursively encounter different views; the art of argument erodes, and differences to one’s own views can become incomprehensible. What is ultimately required, as MacKinnon (2012) argues, is a global policy that can push regulation of the net such that it will be treated like a democratic, digital commons. We have a long way to go. A New Kind of Civic Space? Other implications of using screen-based social media for political life have been explored by a number of authors. Dean (2010) and Papacharissi (2010), for example, argue that it is not just a question of people choosing politics or consumption or popular culture, but that the Internet environment in its present form promotes a transformation of political practices and social relations whereby the political

86  Handbook of digital politics

becomes altered and embodied precisely in the practices and discourses of privatized consumption. Political practices become entangled with the drive for personalized visibility, self-promotion and self-revelation. When (especially) younger people do turn to politics, it seems that the patterns of digital social interaction increasingly carry over into the digital. Papacharissi (2010) argues that this is engendering a new form of civic space. I call it the ‘solo sphere’, and it can be seen as a historically new habitus for Internet-based political participation, a new social milieu for political agency. A networked, often mobile, yet oddly privatized sociality emerges, a personalized space from which the individual engages with the complex political outside world. Operating in this comfort zone often results in what is disparagingly called ‘slacktivism’ or ‘clicktivism’. It is easy to understand this stance as a safe retreat into an environment that many feel they have more control over. To the extent that this is true, however, it introduces a historically new – and problematic – set of ­circumstances for civic agency.

SYNTHESIZING AND SPLITTING THE DIFFERENCES The Internet has been contributing to the massive transformations of contemporary society at all levels for about two decades now, and it would be odd if it did not also alter the premises and infrastructure of political life. In making available vast amounts of information, fostering decentralization and diversity, facilitating interactivity and individual communication, while providing seemingly limitless communicative space for whoever wants it, at speeds that are instantaneous, it has redefined the practices and character of political engagement. Also, while politics remains a minor net usage, the vast universe of the Internet and its various (and ever evolving) technologies make it easier for the political to emerge in online communication, especially within the new kind of alternative politics that is on the rise. Contingencies as Dynamic Configurations There are thus grounds for optimistic views about the Internet’s significance for civic spaces, and there is a good deal of research which supports this view, some of which I have mentioned above. At the same time, as noted above, other voices are cautionary (and a very few are outright dismissive): once we leave the mythical realm of technological determinism and enter complex socio-cultural realities, the role of the Internet becomes more equivocal. Clearly it is not a question of coming to some simple resolution, a neat, all-purpose truth about the Internet as a civic space. The diverse approaches, assumptions and horizons in the extensive literature signal the complexity of the issues involved. A key theme that unites many of the diverse sceptical views is precisely their insistence on socio-cultural contingencies: that is, seeing the Internet (and all social phenomena) as products of circumstances that both engender and delimit them. There are only possibilities, nothing is necessary; any concrete phenomenon is shaped by

The Internet as a civic space  87

a series of other factors, in processes of dynamic configurations. We should keep in mind that the sceptics for the most part are not categorically rejecting any possible positive dimension in regard to the Internet as a civic space; rather, they are often reacting against the excessively enthusiastic and/or naïve view that has been circulated by some commentators, and not least by Internet industries themselves. Thus, the sceptical position challenges us to look critically at the contingencies of whatever social phenomenon we are addressing. A first step in such a direction is conceptual clarity in regard to the phenomenon and its dynamic configurations. In regard to the Internet, we should specify which aspects, services or platforms are relevant. Thus, for example, in regard to social media, different platforms can offer different forms of civic participation. For example, an activist group may need to: (1) internally discuss ideas and debate; (2) develop collective identities; (3) mobilize members; (4) strive to reach out to new members; (5) try to get mass media coverage; and (6) coordinate on-site during a demonstration. Facebook could well serve (1) and (2), Twitter may be very serviceable for (3) and (5), YouTube might be useful for (4), and mobile phone calls and SMS texts be especially useful for (6). There is nothing hard and fast here, yet one should be aware of how different platforms offer divergent affordances, and how this may shape the patterns of use in specific settings. Moreover, the various platforms can be and are used in convergent ways, with relays, feeds and sharing across the platforms (see, for example, Thorson et al., 2013). Among the contingencies to clarify are the zones of interface between on- and offline settings; to illuminate the contexts of use, the modes of usage, the social actors involved, their circumstantial settings, the overarching power relations, the links to their media and communicative spaces, and so forth. This kind of mapping of dynamic configurations, and the elucidation of the consequent contingencies at work, will provide a more rigorous and useful portrait of the specific civic spaces in question than sweeping generalizations. The actual technology itself is of course highly relevant, but it must be understood as being adapted for particular uses by certain actors; it does not operate as an independent, ahistorical force. An extended example of this kind of approach is found in Mattoni’s (2012) study of the media practices of activist workers in today’s crisis-ridden Italy; this movement in fact used such an analysis to devise its own media strategies. In my own work (Dahlgren, 2009) I have followed a version of this logic in looking at how the media may contribute to, or hinder, civic practices. My basic supposition is that for people to participate politically, to engage in civic spaces, they must be able to see engagement as both possible and meaningful. In other words, people need some kind of an empowering civic identity. Yet such identities cannot flourish in a vacuum; they need to be nourished by what I call ‘civic cultures’. Civic cultures are a way of answering, analytically and empirically, the question of what facilitates or hinders people from acting as political agents, from engaging in civic spaces. If we insert the Internet into this framework, we would want to highlight how various aspects interface with everyday life, how the particular citizens in question use it for political purposes, what the political means for them, the power relations in

88  Handbook of digital politics

which they find themselves, and so forth. Civic cultures serve as taken-for-granted resources that people can draw upon, while citizens in turn also contribute to the civic cultures development via their practices; that is, their political uses of the Internet. Further, civic cultures are comprised of a number of distinct dimensions that interact with each other. Participatory practices themselves constitute one key dimension of civic cultures; others include suitable knowledge about the political world and one’s place in it, democratic values to guide one’s actions, and appropriate levels of trust. A minimal level of ‘horizontal’ trust – that is, between citizens – is necessary for the emergence of the social bonds of cooperation between those who collectively engage in politics; there is an irreducible social dimension to doing politics. These dimensions could be elucidated in regard to specific affordances and usages of the net, in concrete situations. Moreover, civic cultures require communicative spaces where such agency can take place; the Internet as a civic space would be critically evaluated in relation to, for example, its political economy and technical architecture to clarify its democratic assets and drawbacks. Finally, forms of identity as political agents – my starting point above – are also a major dimension of civic cultures: people must be able to take on a civic self, to see themselves as actors who can make meaningful interventions in relevant political issues. Clarifying how these dimensions operate configurationally with each other (or not) in specific contexts would enhance our understanding of the Internet as a civic space. Proposals for Future Research A great deal of research on the Internet as a civic space has been done over the years, from varying perspectives. Yet there is still so much we need to know; indeed, from the broader perspective of mediatization and political participation, there is a need for developing a further research agenda, as suggested recently in Dahlgren and Alvares (2013). Based on that collective effort, I would propose that for the theme of the Internet as a civic space, researchers would do well to explore questions that continue some key trajectories in current research, such as the following: ● How does the use of the Internet contribute – in concrete situations – to the development of civic agency, knowledge, practices and identities? This would include a particular focus on alternative politics in the face of the continuing crises and the inadequacies of mainstream politics in dealing with them. ● How do these use strategies tend to promote or hinder actual political engagement, and shape its subjective perceptions and its concrete manifestations and expressions? ● How might existing engagement in popular culture, consumption and sociality be linked to the political, as civic spaces on the net (for example, social media) intersect all the more with societal domains beyond both civil society and the public sphere?

The Internet as a civic space  89

FURTHER READING Bennett, W. Lance and Alexandra Segerberg (2013). The Logic of Connective Action: Digital Media and the Personalization of Contentious Politics. Cambridge: Cambridge University Press. Castells, Manuel (2010). Communication Power. Oxford: Oxford University Press. Coleman, Stephen and Jay Blumler (2009). The Internet and Democratic Citizenship. New York: Cambridge University Press. Dahlgren, Peter (2013). The Political Web. Basingstoke: Palgrave Macmillan. Dewey, John (1923). The Public and its Problems. Chicago, IL: Swallow Press. Gerbaudo, Paolo (2012). Tweets and the Streets: Social Media and Contemporary Activism. London: Verso. Hindman, Mathew (2009). The Myth of Digital Democracy. Princeton, NJ: Oxford University Press. Loader, Brian and Dan Mercea (eds.) (2012). Social Media and Democracy. Abingdon: Routledge. McChesney, Robert W. (2013). Digital Disconnect. New York: New Press. Morozov, Evgeny (2011). The Net Delusion: How Not to Liberate the World. London: Allen Lane. Papacharissi, Zizi (2010). A Private Sphere: Democracy in a Digital Age. Cambridge: Polity Press. van Dijck, José (2013). The Culture of Connectivity: A Critical History of Social Media. Oxford: Oxford University Press.

REFERENCES Alexander, Jeffrey C. (2006). The Civil Sphere. New York: Oxford University Press. Andrejevic, Mark (2013). Infoglut: How Too Much Information Is Changing the Way We Think and Know. Abingdon: Routledge. Barber, Benjamin (1984). Strong Democracy: Participatory Politics for a New Age. Berkeley, CA: University of California Press. Benkler, Y. (2006). The Wealth of Networks: How Social Production Transforms Markets and Freedom. New Haven, CT: Yale University Press. Bennett, W. Lance and Alexandra Segerberg (2012). ‘The logic of connective action: Digital media and the personalization of contentious politics’. Information, Communication & Society 15(5), 739–768. Carr, Nicholas (2010). The Shallows: How the Internet is Changing the Way We Think, Read and Remember. London: Atlantic Books. Castells, Manuel (2010). Communication Power. Oxford: Oxford University Press. Chadwick, Andrew (2006). Internet Politics: States, Citizens, and New Communication Technologies. New York: Oxford University Press. Cleland, Scott and Ira Brodky (2011). Search and Destroy: Why You Can’t Trust Google. St Louis, MO: Telescope Books. Cohen, Jean and Andrew Arato (1992). Civil Society and Political Theory. Cambridge, MA: MIT Press. Dahlgren, Peter (2009). Media and Political Engagement. New York: Cambridge University Press. Dahlgren, Peter and Claudia Alvares (2013). ‘Political participation in an age of mediatisation: Toward a new research agenda’. Javnost/The Public 20(2), 47–66. Dean, Jodi (2010). Blog Theory. Cambridge: Polity Press.

90  Handbook of digital politics

Dewey, John (1923). The Public and its Problems. Chicago, IL: Swallow Press. Dwyer, Tim (2010). ‘Net worth: Popular social networks as colossal marketing machines’. In Gerald Sussman (ed.), Propaganda Society: Promotional Culture and Politics in Global Context. New York: Peter Lang, pp. 77–92. Edwards, Michael (2009). Civil Society, 2nd edition. Cambridge: Polity Press. Fraser, Nancy (1992). ‘Rethinking the public sphere: A contribution to the critique of actually existing democracy’. In Craig Calhoun (ed.), Habermas and the Public Sphere. Boston, MA: MIT Press, pp. 109–142. Fuchs, Christian (2011). Foundation of Critical Media and Information Studies. London: Routledge. Gibson, Rachel K., Andrea Römmele, and Stephen J. Ward (eds.) (2004). Electronic Democracy: Mobilisation, Organisation and Participation via New ICTs. London: Routledge. Greenwald, Glenn (2014). No Place to Hide: Edward Snowden, NSA, and the Surveillance State. New York: Macmillan. Habermas, Jürgen (1989). Structural Transformation of the Public Sphere. Cambridge: Polity Press. Habermas, Jürgen (2006). ‘Political communication in mediated society’. Communication Research 16(4), 411–426. Harvey, David (2006). A Brief History of Neoliberalism. Oxford: Oxford University Press. Hay, Colin (2007). Why We Hate Politics. Cambridge: Polity Press. Held, David (2006). Models of Democracy, 3rd edition. Cambridge: Polity Press. Hill, Kevin A. and John E. Hughes (1998). Cyberpolitics: Citizen Activism in the Age of the Internet. Lanham, MD: Rowman & Littlefield. Hindman, Mathew (2009). The Myth of Digital Democracy. Princeton, NJ: Oxford University Press. Keen, Andrew (2008). The Cult of the Amateur. New York: Doubleday. Lievrouw, Leah A. (2011). Alternative and Activist New Media. Cambridge: Polity Press. MacKinnon, Rebecca (2012). Consent of the Networked: The Worldwide Struggle for Internet Freedom. New York: Basic Books. Malina, Anna (2003). ‘e-Transforming democracy in the UK: Considerations of developments and suggestions for empirical research’. Communications: The European Journal of Communication Research 28(2), 135–155. Margolis, Michael and David Resnick (2000). Politics as Usual: The Cyberspace ‘Revolution’. London: Sage. Mattoni, Alice (2012). Media Practices and Protest Politics: How Precarious Workers Mobilise. Farnham: Ashgate Publishing. Meikle, Graham and Sherman Young (2012). Media Convergence: Networked Digital Media in Everyday Life. Basingstoke: Palgrave Macmillan. Morozov, Evgeny (2011). The Net Delusion: How Not to Liberate the World. London: Allen Lane. Mouffe, Chantal (2005). On the Political. London: Routledge. Papacharissi, Zizi (2010). A Private Sphere: Democracy in a Digital Age. Cambridge: Polity Press. Pariser, Eli (2011). The Filter Bubble: What the Internet is Hiding from You. London: Penguin. Putnam, Robert (2000). Bowling Alone: The Collapse and Revival of American Community. New York: Simon & Schuster. Rosanvallon, Pierre (2008). Counter-Democracy: Politics in an Age of Distrust. Cambridge: Cambridge University Press. Sandel, Michael (2012). What Money Can’t Buy: The Moral Limits of Markets. London: Allen Lane.

The Internet as a civic space  91

Schachter, Hindy Lauer and Kaifeng Yang (eds.) (2012). The State of Citizen Participation in America. Charlotte, NC: Information Age Publishing. Shirky, Clay (2008). Here Comes Everybody: The Power of Organizing Without Organizations. London: Allen Lane. Thorson, K., K. Driscol, B. Ekdale, S. Edgerly, L. G. Thompson, A. Schrock, L. Swartz, E. K. Vraga, and C. Wells (2013). ‘YouTube, Twitter, and the Occupy movement: Connecting content and circulation practices’. Information, Communication & Society 16(2), 1–31. Vaidhyanathan, Siva (2011). The Googlization of Everything: And Why We Should Worry. Berkeley, CA: University of California Press.

7. Political filter bubbles and fragmented publics Cristian Vaccari and Augusto Valeriani

INTRODUCTION: POLITICS IN HIGH CHOICE MEDIA ENVIRONMENTS Citizens’ exposure to sound political information and diverse opinions has been recurrently described as a vital component of a fully functioning democracy. According to Dahl (1992, p. 46), valid and updated knowledge about political issues and actors, as well as frequent participation in discussions with other citizens on issues of public relevance are among the key ingredients that constitute good citizenship. Such description, however, represents an ideal that only very rarely captures the reality of citizens’ engagement with political information and political discussion (Schudson, 1999). Yet, in the last two decades, the increasing gap between informed and uninformed citizens, and the gap between citizens exposed to diverse (also including adversarial) content and those exposed exclusively to homogeneous (likeminded) opinions, have been described as key challenges for contemporary democracies, particularly due to the shift towards increasingly high-choice media environments (Van Aelst et al., 2017). In 2007, Markus Prior coined the term “post-broadcast democracy” to describe a situation where, due to a multiplication of available media channels (Prior focused especially on the transition between broadcast and cable TV in the United States), media diets become highly diversified (Prior, 2007). In such context, Prior argued, the opportunities for incidental exposure to political content were highly reduced compared to the previous broadcast configuration, when choices were limited to few generalist channels, almost all of which featured at least some political news. This new media reality, Prior argued, hindered the so called “by-product learning” process of political information. For example, in the broadcast era many people got some political information from a TV newscast because they tuned in beforehand to be sure not to miss sports news at the end of the program, or a movie scheduled right after the news. Instead, with cable television, individuals had at their disposal all-movies, all-sports, and all-news channels. According to Prior, reduced opportunities for incidental exposure to political news would increase gaps between information seekers and information avoiders, in turn increasing inequalities in citizens’ political knowledge and involvement (Prior, 2007). Similarly, in 2009 Iyengar and Hahn highlighted that the multiplication of media channels in the US resulted in a fragmentation of news sources, with growing market opportunities for more partisan sources seeking to attract specific niches of the public 92

Political filter bubbles and fragmented publics  93

(Iyengar and Han, 2009). In this new environment, individuals highly engaged in politics had the opportunity to fine-tune their news consumption based on ideological proximity. From this standpoint, greater choice could result in a reduced exposure to diversity of opinions among the most politically involved. As citizens now had greater scope to engage with attitude-congruent political content, ideological polarization on issues of public relevance would increase. Conversely, opportunities to “hear the other side” would decrease, in turn reducing tolerance and making it more complicated for citizens to engage in deliberation and reasoned public debate (Mutz, 2006). The increasing relevance of the Internet and, more recently, of social media has arguably accelerated the two phenomena described by Prior (2007) and Iyengar and Han (2009). The combination of digital media affordances and individual preferences in seeking and processing different types of information may enhance selective exposure to political content among most, if not all, citizens. In the present chapter, we discuss the evidence for these claims based on a review of the diverse corpus of literature on the topic and on our own research.

SELECTIVE EXPOSURE TO POLITICAL CONTENT Differences in media experiences between information seekers and avoiders, as well as between citizens with different political orientations, are frequently discussed as the result of a phenomenon named selective exposure (Stroud, 2017). According to these theories, individuals tend to select (and avoid) media content based on several psychological mechanisms. Firstly, exposure to information conflicting with one’s viewpoints could result in cognitive overload, which tends to generate stress and fatigue (Festinger, 1957). Therefore, individuals try to avoid exposure to messages they disagree with, deliberately look for content confirming their beliefs (Nickerson, 1998), and process any information guided by the desire of being proven right in their convictions (Kunda, 1990). Moreover, for those who have little interest in, and knowledge of, political matters, engaging with political information and opinions could require additional cognitive work, especially if such content includes conflicting or nuanced views. This experience could generate confusion (Feldman and Price, 2008), which politically uninterested citizens might be willing to avoid in order to reduce stress and cognitive overload (Skovsgaard and Andersen, 2020). More generally, people organize their media choices based on goals and gratifications they are seeking from their media use, both in general and in specific situations (Katz et al., 1973; Krcmar and Strizhakova, 2009). As a result of these processes, people who are seeking support for their views may exclusively select attitude-congruent news sources and content, while those who mainly turn to the media for entertaining purposes may seek to avoid political news altogether (Toff and Kalogeropoulos, 2020). Finally, some citizens believe that there is no need to deliberately search for information and to organize their media choices accordingly since, thanks to exchanges with their social contacts and to the very functioning of

94  Handbook of digital politics

contemporary information environments, political news will “find” them anyway (Gil de Zúñiga et al., 2017; Toff and Nielsen, 2018). In sum, if individuals enjoy greater opportunities to select the content they get exposed to (or if selection becomes easier), then we should expect a widening in the information gap between citizens with different political orientations, as well as between those who are interested in news and those who are not. These two phenomena could both be seen as contributing to political polarization and fragmentation. In particular, increased selective exposure to attitude-congruent political content based on the political preferences of politically involved citizens could boost horizontal polarization between partisans of opposing sides. Conversely, increased selective exposure to political versus non-political content based on the media preferences of news-oriented versus entertainment-oriented citizens could enhance vertical polarization between information seekers and avoiders. The idea that digital media have increased the fragmentation and polarization of contemporary democratic publics is rooted in a vision of Internet and social media affordances as unavoidably enforcing and crystallizing these individual processes of selection, partly thanks to the action of filtering algorithms that are instructed by users’ previous (selective) actions, but end up reinforcing them by organizing and prioritizing available content in a way that replicates those patterns over time (Bucher, 2018). This representation is indeed based on some actual properties of digital spaces and of platforms’ algorithms. The Internet significantly reduced economic and organizational barriers for content creators to get their editorial work published and available for others. As a result, professional, alternative, and amateur outlets for political information have flourished online, most of them attracting very small niches of the public (Farrell and Drezner, 2008). From users’ perspective, this means that crafting a unique digital newsfeed perfectly tailored to individual political interests and preferences or, conversely, completely excluding politics, has become at least in theory possible, as predicted by Negroponte almost three decades ago (Negroponte, 1995). Secondly, the hyperlinks system that characterizes the Internet enables users to surf through an endless sea of interconnected content, and when it comes to politics this frequently entails navigating across nodes that espouse similar political positions (Adamic and Glance, 2005). The centrality that search engines have assumed in users’ experiences enables users to directly jump to pages automatically selected to meet their interests and expectations. Moreover, algorithms that rank search results take into consideration multiple dimensions, including information about the user who made the query (e.g. search history and behaviours). When applied to searches connected to political content, these mechanisms could end up mainly exposing users to information they already agree with (Robertson et al., 2018; but see Nechushtai and Lewis, 2019; Fletcher and Nielsen, 2018). The affordances of social media platforms have also been described as increasing the potential for selective exposure online due to the particular modalities of engagement they offer to their users (Bakshy et al., 2015). Social media algorithms are developed, among other things, to provide users with the kind of experiences that

Political filter bubbles and fragmented publics  95

are more likely to keep them glued to the platform and engaged with its content. The more users stay on the platform, connect with other users, and perform online actions that generate digital trace data in response to the messages they see, the more the algorithm gets information to curate their future experiences – and to profile them so their attention can be sold to advertisers with specific targeting requirements (Wu, 2017). In this way, users’ past choices inform algorithms’ responses, which in turn affects future users’ choices in a self-reinforcing cycle of selective exposure. Besides algorithmic filtering, many other social media affordances can be easily described as facilitating selective exposure. For example, social media enable users to craft and adjust their own networks of contacts, provide the socio-technical infrastructure for the emergence of ephemeral publics gathered around specific issues and sentiments through hashtag-type affordances (Papacharissi, 2015), and encourage users to – spontaneously or as part of coordinated efforts – endorse specific content, potentially leading to informational cascades where small-scale individual acts of propagation aggregate into large-scale impacts (Sunstein, 2017).

ACCIDENTAL (AND DELIBERATE) EXPOSURE TO POLITICAL NEWS AND DISAGREEMENT A more articulated and nuanced understanding of the socio-technical properties of digital environments suggests that we should not assume that all the dynamics of the Internet and social media propel contemporary democracies toward an unavoidable destiny of homophily and increased inequalities between news junkies and news avoiders (Vaccari and Valeriani, 2021, Chapter 1). Already in 2001, Tewksbury and colleagues observed that breaking news were widely disseminated online, especially as part of the content populating multi-service web portals and hubs which, at the time, most users visited as part of their Internet browsing routines. As a result, incidental exposure to unsearched political content was a common experience in this “post-broadcast” setting, as was the potential for learning about politics as a by-product of engaging with the news encountered on these websites (Tewksbury et al., 2001). Several aspects of the technical functioning and uses of social networking platforms suggest that coming across unsearched political information or disagreeing views could be a far from residual experience for many users. For one, content filtering algorithms are platform specific, complex, constantly redefined, and developed according to volatile corporate policies and goals. Organizing feeds according to previously expressed and inferred users’ preferences is just one of the multiple principles guiding their functioning. For example, we might not be interested in a topic or in a user (e.g. a politician or a party) but such user might be keen to reach us and might be willing to pay to pop up in our news feed via digital advertising. More broadly, algorithms are designed by humans based on their organizations’ goals and incentives, which in the current configuration of the digital economy arguably leads them to promoting “sameness” in pursuit of “stickiness” (Hindman, 2018). However,

96  Handbook of digital politics

different business models and goals might lead to the design of algorithms that promote other values. There is evidence that alterations of platform affordances can lead to meaningful changes in user behaviour, some of which may be democratically beneficial. For instance, when Twitter increased the maximum length of a post from 168 to 280 characters, discussions became more polite and constructive (Jaidka et al., 2019). Moreover, while some properties of platforms’ architecture facilitate selection and homophily, others are likely to promote serendipitous encounters with unsearched or unwanted content (Colleoni et al., 2014). For example, most social networking platforms feature sections highlighting topics that are “trending” at a given time. In this way, a user can learn about publicly relevant events, issues, or actors she was unaware of. Similarly, redistribution affordances such as sharing buttons are common to most digital platforms (Larsson, 2017) and constantly expose users to the “risk” of encountering content that others in their networks believe is worth sharing. While it is true that enabling users to craft their networks is a defining property of social media, it is also clear that people compose these networks based on multiple considerations, among which politics is unlikely to feature very prominently, especially among users who are not “political junkies” (Anspach, 2017). If we take the typical user of a mainstream platform such as Facebook, it is very likely that her network includes a mix of strong and weak ties resulting from multiple experiences and interests developed at different points of her professional and private life (Boyd and Ellison, 2007). Hence, it is far from impossible that such a relational environment will feature some degree of political diversity or different levels of interest in, and propensity to share, political news among its members. As a result, some political information could sneak into the Facebook feed even for someone who would usually avoid political news everywhere else. Similarly, someone who has a negative opinion of the current government may have among her Facebook friends a vocal supporter of said government – who became her friend not because of or despite politics, but because they met at the local basketball playground. Granted, once exposed to frequent messages expressing support for the government, our user could decide to “unfriend” her former playground buddy to stop having to deal with disagreeing political views. However, research suggests that unfriending and unfollowing based on political consideration are rare behaviours that are largely confined to a minority of politically involved users (Bode, 2016). More broadly, the appetite for opinion-reinforcing news does not necessarily entail deliberate avoidance of adversarial content (Garrett, 2009). In other words, selective avoidance is not necessarily the other side of the coin of selective exposure, and the multifaceted and multilayered structures of social media networks may promote encounters with diverse opinions that most users are not willing to silence even if they disagree. Finally, some people find enjoyment in engaging with adversarial political views and content online, for instance because they love political confrontation. These kinds of conflict-seeking users can leverage social media affordances to build the “contrarian clubs” they aspire to be part of (Vaccari et al., 2016).

Political filter bubbles and fragmented publics  97

“ECHO CHAMBERS” AND “FILTER BUBBLES”: DIFFERENCES AND CONNECTIONS Despite all the nuances discussed thus far, there is little doubt that the idea of selective exposure as the main law ruling digital realms currently characterizes much public understanding of, and debate around, social media platforms as spaces for public discussion (Bruns, 2019). This narrative has been sustained by the evocative power of two highly effective metaphors that have crossed the boundaries of academic debate to become common knowledge: the “echo chamber” and the “filter bubble”. These two concepts are both connected to the idea that the centrality of the Internet in contemporary communication ecosystems has increased the fragmentation of publics. However, although both terms are frequently employed as synonymous in academic reasoning (e.g. Flaxman et al., 2016), they capture different dimensions and dynamics (Bruns, 2019; Vaccari and Valeriani, 2021; Ross Arguedas et al., 2022). To define “echo chambers”, we borrow from Jamieson and Cappella (2008), who employed the term to describe the emergence of a conservative media ecosystem in the US at the start of the twenty-first century. The increasing supply of right-wing media had created a self-isolating ideological silo, a “safe haven” that reinforced “the views of these outlets’ likeminded audience members”, strengthening the ideological consistency of the audience while acting as a shield against exposure to counter-attitudinal views (Jamieson and Cappella, 2008, p. x). However, Jameson and Cappella mainly focused on analogue media, such as talk radio, newspapers, and cable television. The first author to apply the “echo chambers” metaphor to digital media was Cass Sunstein (2001, 2009, 2017). According to Sunstein, when individuals’ disposition towards homophily meets the choice affordances and choice-reinforcing algorithms of search engines and social media, the result is the emergence of small, tightly sealed chambers where people exclusively encounter discussion partners and news sources mirroring their political positions. In such a situation, while connection and identification with political in-groups become stronger, distance and separation from out-groups grow larger. The echo chamber metaphor thus mainly relates to what we have previously called horizontal polarization, i.e., the increasing gap between partisans on different sides. The “filter bubble” metaphor is the brainchild of activist and tech-entrepreneur Eli Pariser (2011). With this incisive image, Pariser specifically aimed at unveiling the role of algorithmic filtering in creating highly personalized experiences online, where users advertently or inadvertently instruct digital platforms with actions that signal their preferred types of content, and thus topics that fall out of their interest are progressively hidden from them. The filter bubble thus mainly addresses the vertical type of polarization between citizens who are interested in political news and those who are not. However, the idea of filter bubbles also highlights the potential for increased ideological (horizontal) polarization. As digital algorithms and affordances select and prioritize content based on users’ behaviours, they can also accommodate their political orientations, especially for those whose digital actions reveal clear ideolog-

98  Handbook of digital politics

ical preferences. Hence, Ross Arguedas and colleagues (2022) contend that the idea of “echo chamber” does not clearly differentiate between the individual and systemic factors that lead to the creation of like-minded bubbles. Conversely, they argue, “A filter bubble […] is an echo chamber primarily produced by ranking algorithms engaged in passive personalization without any active choice on our part” (Ross Arguedas et al., 2022, p. 11). In other words, filter bubbles entail lower levels of user agency and awareness, while echo chambers mainly result from conscious choices. Having discussed these key concepts and highlighted features of contemporary media ecosystems that both facilitate and hinder them, we will now briefly review some empirical research, including our own, that has assessed and explained their existence and diffusion.

HOW PREVALENT ARE ECHO CHAMBERS IN CONTEMPORARY DIGITAL MEDIA? The widespread concerns for the role that echo chambers and filter bubbles may play in limiting citizens’ information diets have spurred extensive research on this topic. Scholars have relied on various methods, each with different strengths and limitations, which we can only briefly summarize here. Surveys enable researchers to measure citizens’ experiences across a variety of online and offline channels and environments, thus offering a holistic representation of their media diets, but suffer from biases due to poor recall of past behaviour and social desirability. Social media data potentially capture individuals’ whole experience, including whether and how they engage with content by liking, commenting, or sharing, but they are platform-specific, pose huge ethical challenges, and, with the partial exception of Twitter and YouTube among the major platforms, are largely unavailable to social science researchers who do not work for, or collaborate with, the companies that own them (Tromble, 2021). Web tracking provides very precise and granular measures of website visits, but it is difficult and costly to recruit large and representative samples for these studies. Perhaps the best approach to address these complex issues is the combination of surveys with digital trace data, whether deriving from web tracking or social media, or both. This strategy can achieve both breadth (in terms of the attitudes and behaviours that can be measured and the environments where they occur) and depth (in terms of the precision and granularity of the actions that can be captured). Other methods, such as qualitative interviewing, media diaries, and digital ethnography, can also helpfully illuminate some relevant aspects of these problems, but they have been less frequently employed to conduct research on these topics (but see Magin et al., 2022). In this brief overview, we will discuss a few relevant contributions, without pretending to do justice to the panoply of studies that have emerged over the past decade. (For extensive reviews on these issues, see Ross Arguedas et al., 2022; Tucker et al., 2018.) Studies based on surveys have generally found little evidence that echo chambers are prevalent for most citizens. Dubois and Blank (2018, p. 740), for instance, sur-

Political filter bubbles and fragmented publics  99

veyed a representative sample of UK adults and concluded that “People regularly encounter things that they disagree with. People check multiple sources. People try to confirm information using search. Possibly most important, people discover things that change their political opinions.” Barnidge (2017) conducted a survey of a representative sample of US Internet users and found that social media users perceive they encounter political disagreement more often than non-users, and more often than in face-to-face conversations and anonymous exchanges online. Focusing on diversity of news sources, Fletcher and colleagues (2020) surveyed citizens in twelve Western democracies and assessed to what extent audiences are polarized, i.e., display clear aggregate-level divisions in the news sources used by right-wing and left-wing voters. They found that levels of polarization vary greatly across countries, with the US and Southern European democracies showing the starkest political divides in news consumption, while news audiences gravitate more towards “catch-all” centrist outlets (often public service media) in Germany, Northern European countries, and the UK. Importantly, online news audiences tend to be slightly more polarized than offline news audiences, although in some countries the pattern is reversed. Hence, the authors conclude that digital environments are not destined to become echo chambers where news consumption is primarily driven by citizens’ political preferences. Country-level systemic characteristics, more than technological developments, explain news audience polarization. The most comprehensive study to date based on social media data was conducted by Bakshy and colleagues (2015). Working with proprietary Facebook data, the authors studied patterns of exposure to news among 10 million US users who declared their ideological affiliation (liberal or conservative) on the platform. They then used machine learning to identify which, among the 7 million web links shared by these users, can be considered as hard news, and then classified the 226,000 links that belong to this category as liberal, neutral, or conservative, based on the ideological leanings of those users who had shared them. Based on these estimates, they then assessed to what extent the content of the news users see depends on their networks (which news their friends share), the Facebook algorithm (which stories are prioritized on their feed), and users’ choices (which links they click on). The results suggest that the key factor limiting ideological diversity on Facebook is that users tend to connect with others who share their political views, and who thus tend to post links to ideologically congruent news. Users were also more likely to click on articles confirming their views, further increasing exposure to congruent information. According to this study, whose authors were employed by Facebook at the time, the Facebook algorithm, at least as it functioned in the second half of 2014 when the data were collected, did not make a substantial contribution to homophily above and beyond network characteristics and user choices. Still, as discussed earlier and as we will further highlight below based on our own research, the US exhibits comparatively high levels of audience polarization online, so these results may not generalize to other liberal democracies. Even in this context, “on average more than 20 percent of an individual’s Facebook friends who report an ideological affiliation are from the opposing party” (Bakshy et al., 2015, p. 1131). After accounting for

100  Handbook of digital politics

networks, algorithms, and user choices, around 20 percent of the stories liberal users clicked on had a conservative slant; among conversative users, the percentage was close to 30 percent. Although there is no objective threshold for what constitutes an echo chamber or a pluralistic space of public debate, we suggest that these estimates do not provide strong evidence of the former. Scholars unable to access platforms’ proprietary user data have had to rely on public content, for instance on Facebook public pages, which are certainly relevant but only represent a small part of most users’ experience. An example of this approach is a study by Del Vicario and colleagues (2016), who collected posts and interactions on a sample of Facebook pages espousing conspiracy theories and science news. They found strong evidence of echo chambers, as users tend to only engage with content around a particular narrative and to mainly share it with others who are similar to them. However, the very specific focus of the pages from which the data were collected prevents generalizing these results to the experiences of ordinary Facebook users. Focusing on Twitter, Barberá (2015) developed a method to estimate users’ ideology and applied it to almost 200,000 accounts in Germany, Spain, and the United States. He then used the same approach to estimate the ideology of the accounts followed by these users over time and found that “over 75% of users in each country are embedded in networks that include 25% or more individuals with whom they disagree” (Barberá, 2015, p. 19). These two sets of estimates were then combined and, importantly, repeated for different periods of time to demonstrate that users’ ideology changes depending on changes in the ideology of the accounts they follow. A minority of users develops more ideologically congruent networks, by following more accounts they agree with, and as a result becomes more ideologically extreme. However, the majority of users tend to follow more users they disagree with, thus increasing the levels of diversity in their networks over time. After following more politically diverse sources, the estimated ideology of these users became more moderate. Studies based on web tracking have highlighted that most Internet users spend very little time engaging with political or news content, and that much of this exposure does not suggest a strong prevalence of echo chambers. For instance, Flaxman et al. (2016) used a browser add-on to collect data for 50,000 US-located users and showed that social media and search engines tend to expose users to both ideologically congruent and discordant news articles. Wojcieszak and colleagues (2021) combined panel surveys and web tracking data of US citizens and found that most users visited news websites very rarely, and most of these visits did not involve ideologically slanted news sites. Importantly, they also did not find any meaningful effects of visits to partisan websites on attitudinal or affective polarization. Guess (2021) analysed two surveys of representative samples of the US population combined with data on respondents’ Internet browsing and showed that, even in the highly polarized American context, most users consume minimal quantities of news but, when they do, they tend to gravitate around centrist sources with low levels of ideological slant. By contrast, a minority of highly partisan users makes up for the vast majority of the audience for partisan news. As we discuss below, these results are consistent with

Political filter bubbles and fragmented publics  101

those of our own research on the factors that predict the experience of echo chambers online.

ESTIMATING ECHO CHAMBERS AND FILTER BUBBLES ACROSS DIFFERENT COMMUNICATION ENVIRONMENTS As this brief review of the literature indicates, current research on echo chambers and filter bubbles is limited by the fact that, with few exceptions, it has mainly focused on individual social media platforms (mostly those for which data is readily available to researchers), in isolation from the broader context in which citizens can be exposed to different kinds of information in the hybrid media system (Chadwick, 2017), and in the context of single-country case studies that fail to capture the role of systemic factors. In our own work (Vaccari and Valeriani, 2021), we strived to overcome these limitations by asking respondents across six different Western democracies (France, Germany, Italy, Spain, the United Kingdom, and the United States) a series of standardized questions measuring the frequency with which they agree and disagree with the political information and opinions they encounter on various relevant communication environments. The surveys were conducted between 2015 and 2018 on online samples recruited to match the key demographic characteristics of the population with Internet access in each country (N=1,750 per country, N=2,500 in the US). We focused on four channels: the mass media (television and newspapers), social media (defined broadly as comprising the main public platforms such as Facebook, Twitter, and YouTube), mobile instant messaging apps (such as WhatsApp, Facebook Messenger, and Snapchat), and face-to-face conversations. For each of these channels, we asked respondents how often they encountered political content they agreed and disagreed with. (See Vaccari and Valeriani, 2021 for information on question wording and sampling.) With all the aforementioned limits of survey self-reports, these measures enable us to compare how different communication environments enhance diversity or homogeneity in citizens’ information diets, as well as capture the contribution of social media more broadly rather than based on any individual platforms. Here, we focus on the subset of respondents who use all types of platforms and who answered all the eight questions we asked to measure exposure to agreeing and disagreeing content across these environments (N=3,711). As Figure 7.1 shows, when compared with other sources of political information, social media do not look like the tightly sealed echo chambers they are often purported to be. Almost half the respondents (48 percent) report that they see information on social media that they disagree with equally as often as information they agree with; we label this experience “two-sided”. More than a quarter (28 percent) claim that they more often see content opposed to than congruent with their political views on these platforms, and we term this experience “one-sided oppositional”. (In another study mentioned earlier, we called these “contrarian clubs”; see Vaccari et al., 2016.) Only little more than one fifth (22 percent) answered that they more often agree than

102  Handbook of digital politics

Source:

Vaccari and Valeriani (2021).

Figure 7.1

Perceived exposure to political agreement and disagreement via the mass media, social media, mobile instant messaging apps, and offline conversations

they disagree with the political messages they see on social media, an experience we call “one-sided supportive”. Hence, according to these estimates, social media constitute echo chambers for a relatively small minority of the population, while serving as a channel that predominantly delivers politically diverse or even oppositional information to their users. Perhaps even more relevant, social media function less as echo chambers than both mobile instant messaging apps and offline conversations. In both these environments, while most respondents claim to encounter both agreeing and disagreeing content in equal measures, more users experience one-sided supportive contexts than oppositional ones, by a factor of two to one when it comes to offline conversations (37 percent one-sided supportive versus 19.5 percent one-sided oppositional). By comparison, only the mass media contribute to diversity of political information more than social media, with 38 percent of respondents experiencing one-sided oppositional and 18 percent one-sided supportive content on television and newspapers (Mutz, 2001). In sum, social media cannot be singled out as the sole culprit for the political fragmentation of contemporary public discourse. Social media more often than not expose users to diverse, often counter-attitudinal political content and they do so substantially more often than mobile instant messaging apps and face-to-face conversations (Vaccari and Valeriani, 2021, pp. 88–89). Although most social media users are not enveloped in political echo chambers, understanding the factors that explain why a substantial minority of around one-fifth predominantly encounters content they agree with can illuminate how these platforms distinctively contribute to political pluralism, fragmentation, and mobilization. To this end, we now discuss analyses of survey data collected in the six countries

Political filter bubbles and fragmented publics  103

listed above, plus Denmark, Greece, and Poland, focusing on respondents who use at least one major social media and answered the two questions we used to measure exposure to agreement and disagreement on these platforms (N=11,695). Regression models that predicted these outcomes based on socio-demographic characteristics, political attitudes, sources of political information, frequency of political talk online and offline, use of different social media platforms, and country of residence revealed several key explanatory factors. First, the older respondents were, the more likely they were to claim that they were exposed to different views on social media in roughly equal measure. The experience of echo chambers – and of contrarian clubs – is more common for younger social media users. Those with higher levels of education turned out to be less likely to be exposed to oppositional environments on social media, but not to predominantly supportive messages. Respondents who located themselves towards both the left and right poles of the ideological spectrum were significantly more likely to be part of echo chambers than they were to experience two-sided information flows on social media. Those who placed themselves at the centre-left and the centre-right were also more likely than centrists to encounter one-sided supportive environments than two-sided ones. Frequency of political talk on social media was one of the strongest predictors: the more respondents reported discussing politics on these platforms, the more likely they were to experience echo chambers and the less likely they were to be involved in contrarian clubs. Notably, higher levels of Facebook use did not predict the experience of one-sided supportive content, but they were associated with a lower likelihood to be part of one-sided oppositional environments than two-sided ones. By contrast, the more participants reported using Twitter and YouTube, the more likely they were to encounter one-sided supportive messages, although the magnitude of these relationships is small. Finally, and importantly, US respondents were significantly more likely to be part of echo chambers, and less likely to engage with contrarian clubs, than respondents in most of the eight other countries we studied. Although even among Americans one-sided supportive environments were less common than one-sided oppositional ones, the widespread concern for echo chambers might be more justified among US scholars (e.g. Sunstein, 2017) than among those studying other Western democracies where this phenomenon is less prevalent (Vaccari and Valeriani, 2021, pp. 90–98). These findings highlight that exposure to content one agrees with is, perhaps unsurprisingly, more common among social media users who are willing to take clear ideological sides and who frequently discuss politics online. Seeing their views vindicated is more important for these users than for the rest of the population, so they rationally employ the choice affordances of digital platforms to build supportive information environments that make them feel validated (Mutz, 2006). These political “power users” are clearly central for the flow of public information online, as they are more likely to be “opinion leaders” (Karlsen, 2015) and serve as hubs in their networks. However, their experience of political content on social media, which is often characterized by a prevalence of messages they agree with, should not be generalized to the whole population. For most social media users, politics is not a central part of

104  Handbook of digital politics

their online experience, and thus they are less likely to actively prune their feeds from political content they disagree with (Bode, 2016). Their limited engagement with political messages might also send weaker signals to the algorithms that curate their news feeds, thus reducing the likelihood that they will be automatically prevented from seeing disagreeable political information. As we will argue in the next section, lack of exposure to meaningful political content may be a more pressing concern than lack of diversity in such content.

CONCLUSIONS AND FUTURE DIRECTIONS Ever since the advent of the Internet as a mass medium, and continuing into the social media age, scholars of digital politics have been concerned with echo chambers, filter bubbles, and political fragmentation. This research agenda has generated relevant knowledge over the past twenty years, but it has also left open some important gaps. Research has mainly focused on how social media may contribute to what in this chapter we have defined horizontal polarization in news consumption, or the divide between the kinds of political information accessed by people located at different ends of the partisan or ideological spectrum. In the context of increasing concerns for ever growing political polarization in the United States and across other Western democracies, the choice affordance of the Internet and, later, social media were easy to see as part of this problem. The post-2016 reckoning compounded these worries, as echo chambers of ideologically extreme voters were seen as the natural breeding ground for the creation and spread of misinformation and disinformation online (Bennett and Livingston, 2018; Rhodes, 2022). And yet, an even bigger challenge for democracy might arguably be vertical polarization, between those who often get exposed to political content and those who rarely, if ever, encounter it. The choice affordances of social media might be more consequential in that they enable users to avoid (most) political information than because they allow power users to build echo chambers of agreeing content (Prior, 2007). There is a general consensus that the percentage of web traffic directed to news is very low, between 1.5 percent and 3 percent of the total time an average user spends online (Hindman, 2018, p. 134; Wojcieszak et al., 2021). Although some scholars – and we count ourselves among them – have written optimistically about the potential for social media to accidentally expose politically marginal users to relevant information about public affairs (Valeriani and Vaccari, 2016), the jury is still out on the extent and implications of this phenomenon (see e.g. Kümpel, 2020; Thorson, 2020). The fact that social media may be increasing the gap between information haves and have-nots may be more politically consequential, and worth investigating, than the fact that a robust and influential partisan minority may be using digital media to build the kinds of homophilic environments that in all likelihood resemble and augment their mass media news diets and face-to-face discussion networks. However, students of politics have also known for a long time that, for better or worse, most politically consequential phenomena in a democracy do not necessarily

Political filter bubbles and fragmented publics  105

involve majorities, or even large minorities, of the population (Sartori, 1987). From this standpoint, understanding the prevalence and effects of echo chambers on relatively small groups is just as important as assessing their role among the general public. For instance, scholars of terrorism pointed out how the secluded environments that characterize the experiences of some social media users may facilitate radicalization and recruitment to terrorist networks (O’Hara and Stevens, 2015). This potential risk became even more dramatically apparent when around two thousand Donald Trump supporters stormed the US Capitol on January 6, 2022, as Congress was certifying the results of the presidential election that saw Trump defeated by Joe Biden after a bitter campaign when the president and his supporters had incessantly and recklessly spread disinformation about the fairness and legality of the vote. Although research on the root causes of this direct attack on American democracy is still in its infancy (but see Finkel et al., 2020), it is highly likely that online homophilic networks, maintained on mainstream social media as well as alternative niche platforms, might have played a role in recruiting acolytes and organizing the insurrection (Munn, 2021). By the same token, homophilic networks on social media have arguably benefited social movements supporting democratic causes all around the world. Hashtag activism, witnessing and documenting injustice in real time, and sharing personal experiences of abuse via social media have helped committed minorities of activists reach out to potential supporters, recruit allies, and achieve visibility (Mendes et al., 2018; Richardson, 2020; Tufekci, 2017). The fact that these movements are often born out of, or at least nurtured by, online echo chambers of supportive voices is seldom discussed when assessing social media’s contribution to democracy. A final issue involves technological change. In the second half of the 2010s, online users and companies began to shift from (semi-)public social media, such as Facebook, YouTube, Instagram, and Twitter to (mainly) private messaging apps, such as WhatsApp, Facebook Messenger, and Snapchat. Communication in these environments is more strongly driven by users’ choices than by algorithms, as end-to-end encryption affords platforms limited control over the content seen by users (Rossini, 2023). As messaging apps are mainly used for maintaining relationships with strong and weak ties, they may enable users to express themselves politically in ways they would not contemplate in the more public contexts of social media (Valeriani and Vaccari, 2018). However, as the data in Figure 7.1 show, messaging apps may be much more hospitable to echo chambers than social media. One key factor that may explain this outcome is that users engaging with their social ties on these apps prefer to avoid conflict, for instance when others share information that is false, exaggerated, or highly partisan (Chadwick et al., 2022). Due to the lack of accessible digital trace data on users’ behaviours on these environments, scholars will need to creatively leverage social science research toolkits to shed light on how private messaging apps contribute to the diversity and fragmentation of contemporary media environments.

106  Handbook of digital politics

REFERENCES Adamic, L. A. and Glance, N. (2005). The political blogosphere and the 2004 US election: Divided they blog. In Proceedings of the 3rd International Workshop on Link Discovery (pp. 36–43). Anspach, N. M. (2017). The new personal influence: How our Facebook friends influence the news we read. Political Communication, 34(4), 590–606. Bakshy, E., Messing, S., and Adamic, L. A. (2015). Exposure to ideologically diverse news and opinion on Facebook. Science, 348(6239), 1130–1132. Barberá, P. (2015). How social media reduces mass political polarization: Evidence from Germany, Spain, and the US. Unpublished manuscript. Barnidge, M. (2017). Exposure to political disagreement in social media versus face-to-face and anonymous online settings. Political Communication, 34(2), 302–321. Bennett, W. L. and Livingston, S. (2018). The disinformation order: Disruptive communication and the decline of democratic institutions. European Journal of Communication, 33(2), 122–139. Bode, L. (2016). Pruning the news feed: Unfriending and unfollowing political content on social media. Research & Politics, 3(3), 2053168016661873. Boyd, D. M. and Ellison, N. B. (2007). Social network sites: Definition, history, and scholarship. Journal of Computer-Mediated Communication, 13(1), 210–230. Bruns, A. (2019). It’s not the technology, stupid: How the ‘echo chamber’ and ‘filter bubble’ metaphors have failed us. Paper presented at the IAMCR 2019 conference in Madrid, July 7–11. Bucher, T. (2018). If … Then: Algorithmic Power and Politics. New York: Oxford University Press. Chadwick, A. (2017). The Hybrid Media System: Politics and Power. New York: Oxford University Press. Chadwick, A., Vaccari, C., and Hall, N. A. (2022). What people talk about when they talk about misinformation on private social media: Affective contradiction, conflict avoidance, and trust as contexts for anti-misinformation interventions. Working Paper, Loughborough University. Colleoni, E., Rozza, A., and Arvidsson, A. (2014). Echo chamber or public sphere? Predicting political orientation and measuring political homophily in Twitter using big data. Journal of Communication, 64(2), 317–332. Dahl, R. A. (1992). The problem of civic competence. Journal of Democracy, 3(4), 45–59. Del Vicario, M., Bessi, A., Zollo, F., Petroni, F., Scala, A., Caldarelli, G., Stanley, H. E., and Quattrociocchi, W. (2016). The spreading of misinformation online. Proceedings of the National Academy of Sciences, 113(3), 554–559. Dubois, E. and Blank, G. (2018). The echo chamber is overstated: The moderating effect of political interest and diverse media. Information, Communication & Society, 21(5), 729–745. Farrell, H. and Drezner, D. W. (2008). The power and politics of blogs. Public Choice, 134(1), 15–30. Feldman, L. and Price, V. (2008). Confusion or enlightenment? How exposure to disagreement moderates the effects of political discussion and media use on candidate knowledge. Communication Research, 35(1), 61–87. Festinger, L. (1957). A theory of cognitive dissonance. Stanford University Press. Finkel, E. J., Bail, C. A., Cikara, M., Ditto, P. H., Iyengar, S., Klar, S., … Druckman, J. N. (2020). Political sectarianism in America. Science, 370(6516), 533–536. Flaxman, S., Goel, S., and Rao, J. M. (2016). Filter bubbles, echo chambers, and online news consumption. Public Opinion Quarterly, 80(S1), 298–320.

Political filter bubbles and fragmented publics  107

Fletcher, R., Cornia, A., and Nielsen, R. K. (2020). How polarized are online and offline news audiences? A comparative analysis of twelve countries. The International Journal of Press/ Politics, 25(2), 169–195. Fletcher, R. and Nielsen, R. K. (2018). Automated serendipity: The effect of using search engines on news repertoire balance and diversity. Digital Journalism, 6(8), 976–989. Garrett, R. K. (2009). Politically motivated reinforcement seeking: Reframing the selective exposure debate. Journal of Communication, 59(4), 676–699. Gil de Zúñiga, H., Weeks, B., and Ardèvol-Abreu, A. (2017). Effects of the news-finds-me perception in communication: Social media use implications for news seeking and learning about politics. Journal of Computer-Mediated Communication, 22(3), 105–123. Guess, A. M. (2021). (Almost) everything in moderation: New evidence on Americans’ online media diets. American Journal of Political Science, 65(4), 1007–1022. Hindman, M. (2018). The Internet Trap. Princeton, NJ: Princeton University Press. Iyengar, S. and Hahn, K. S. (2009). Red media, blue media: Evidence of ideological selectivity in media use. Journal of Communication, 59(1), 19–39. Jaidka, K., Zhou, A., and Lelkes, Y. (2019). Brevity is the soul of Twitter: The constraint affordance and political discussion. Journal of Communication, 69(4), 345–372. Jamieson, K. H. and Cappella, J. N. (2008). Echo Chamber: Rush Limbaugh and the Conservative Media Establishment. New York: Oxford University Press. Karlsen, R. (2015). Followers are opinion leaders: The role of people in the flow of political communication on and beyond social networking sites. European Journal of Communication, 30(3), 301–318. Katz, E., Blumler, J. G., and Gurevitch, M. (1973). Uses and gratifications research. The Public Opinion Quarterly, 37(4), 509–523. Krcmar, M. and Strizhakova, Y. (2009). Uses and gratifications as media choice. In T. Hartman (ed.), Media Choice: A Theoretical and Empirical Overview (pp. 67–83). New York: Routledge. Kümpel, A. S. (2020). The Matthew Effect in social media news use: Assessing inequalities in news exposure and news engagement on social network sites (SNS). Journalism, 21(8), 1083–1098. Kunda, Z. (1990). The case for motivated reasoning. Psychological Bulletin, 108(3), 480–498. Larsson, A. O. (2017). Going viral? Comparing parties on social media during the 2014 Swedish election. Convergence, 23(2), 117–131. Magin, M., Geiß, S., Stark, B., and Jürgens, P. (2022). Common core in danger? Personalized information and the fragmentation of the public agenda. The International Journal of Press/ Politics, 27(4), 887–909. Mendes, K., Ringrose, J., and Keller, J. (2018). #MeToo and the promise and pitfalls of challenging rape culture through digital feminist activism. European Journal of Women’s Studies, 25(2), 236–246. Munn, L. (2021). More than a mob: Parler as preparatory media for the US Capitol storming. First Monday. http://​dx​.doi​.org/​10​.5210/​fm​.v26i3​.11574. Mutz, D. C. (2001). Facilitating communication across lines of political difference: The role of mass media. American Political Science Review, 95(1), 97–114. Mutz, D. C. (2006). Hearing the Other Side: Deliberative versus Participatory Democracy. Cambridge: Cambridge University Press. Nechushtai, E. and Lewis, S. C. (2019). What kind of news gatekeepers do we want machines to be? Filter bubbles, fragmentation, and the normative dimensions of algorithmic recommendations. Computers in Human Behavior, 90, 298–307. Negroponte, N. (1995). Being Digital. New York: Vintage. Nickerson, R. S. (1998). Confirmation bias: A ubiquitous phenomenon in many guises. Review of General Psychology, 2(2), 175–220.

108  Handbook of digital politics

O’Hara, K. and Stevens, D. (2015). Echo chambers and online radicalism: Assessing the Internet’s complicity in violent extremism. Policy & Internet, 7(4), 401–422. Papacharissi, Z. (2015). Affective Publics: Sentiment, Technology, and Politics. New York: Oxford University Press. Pariser, E. (2011). The Filter Bubble: What the Internet is Hiding from You. London: Penguin. Prior, M. (2007). Post-Broadcast Democracy: How Media Choice Increases Inequality in Political Involvement and Polarizes Elections. Cambridge: Cambridge University Press. Rhodes, S. C. (2022). Filter bubbles, echo chambers, and fake news: How social media conditions individuals to be less critical of political misinformation. Political Communication, 39(1), 1–22. Richardson, A. V. (2020). Bearing Witness While Black: African Americans, Smartphones, and the New Protest #Journalism. New York: Oxford University Press. Robertson, R. E., Lazer, D., and Wilson, C. (2018). Auditing the personalization and composition of politically-related search engine results pages. Proceedings of the 2018 World Wide Web Conference (pp. 955–965). Ross Arguedas, A., Robertson, C., Fletcher, R., and Nielsen, R. (2022). Echo Chambers, Filter Bubbles, and Polarisation: A Literature Review. Report. Reuters Institute for the Study of Journalism. https://​reutersinstitute​.politics​.ox​.ac​.uk/​echo​-chambers​-filter​-bubbles​-and​ -polarisation​-literature​-review. Rossini, P. (2023). Farewell to big data? Studying misinformation in mobile messaging applications. Political Communication. https://​doi​.org/​10​.1080/​10584609​.2023​.2193563. Sartori, G. (1987). The Theory of Democracy Revisited. London: Chatham House Publishers. Schudson, M. (1999). The Good Citizen: A History of American Civic Life. Cambridge, MA: Harvard University Press. Skovsgaard, M. and Andersen, K. (2020). Conceptualizing news avoidance: Towards a shared understanding of different causes and potential solutions. Journalism Studies, 21(4), 459–476. Stroud, N. J. (2017). Selective exposure theories. In K. Kenski and K. H. Jamieson (eds.), The Oxford Handbook of Political Communication (pp. 531–548). New York: Oxford University Press. Sunstein, C. R. (2001). Republic.Com. Princeton, NJ: Princeton University Press. Sunstein, C. R. (2009). Republic.Com 2.0. Princeton, NJ: Princeton University Press. Sunstein, C. R. (2017). #Republic: Divided Democracy in the Age of Social Media. Princeton, NJ: Princeton University Press. Tewksbury, D., Weaver, A. J., and Maddex, B. D. (2001). Accidentally informed: Incidental news exposure on the world wide web. Journalism and Mass Communication Quarterly, 78(3), 533–554. Thorson, K. (2020). Attracting the news: Algorithms, platforms, and reframing incidental exposure. Journalism, 21(8), 1067–1082. Toff, B. and Kalogeropoulos, A. (2020). All the news that’s fit to ignore: How the information environment does and does not shape news avoidance. Public Opinion Quarterly, 84(S1), 366–390. Toff, B. and Nielsen, R. K. (2018). “I just Google it”: Folk theories of distributed discovery. Journal of Communication, 68(3), 636–657. Tromble, R. (2021). Where have all the data gone? A critical reflection on academic digital research in the post-API age. Social Media+ Society, 7(1). https://​doi​.org/​10​.1177/​ 205630512198892. Tucker, J., Guess, A., Barberá, P., Vaccari, C., Siegel, A., Sanovich, S., Stukal, D., and Nyhan, B. (2018). Social Media, Political Polarization, and Political Disinformation: A Review of the Scientific Literature. Hewlett Foundation. https://​papers​.ssrn​.com/​sol3/​papers​.cfm​ ?abstract​_id​=​3144139.

Political filter bubbles and fragmented publics  109

Tufekci, Z. (2017). Twitter and Tear Gas: The Power and Fragility of Networked Protest. New Haven, CT: Yale University Press. Vaccari, C. and Valeriani, A. (2021). Outside the Bubble: Social Media and Political Participation in Western Democracies. New York: Oxford University Press. Vaccari, C., Valeriani, A., Barberá, P., Jost, J. T., Nagler, J., and Tucker, J. A. (2016). Of echo chambers and contrarian clubs: Exposure to political disagreement among German and Italian users of Twitter. Social Media + Society, 2(3). https://​doi​.org/​10​.1177/​ 2056305116664221. Valeriani, A. and Vaccari, C. (2016). Accidental exposure to politics on social media as online participation equalizer in Germany, Italy, and the United Kingdom. New Media & Society, 18(9), 1857–1874. Valeriani, A. and Vaccari, C. (2018). Political talk on mobile instant messaging services: A comparative analysis of Germany, Italy, and the UK. Information, Communication & Society, 21(11), 1715–1731. Van Aelst, P., Strömbäck, J., Aalberg, T., Esser, F., De Vreese, C., Matthes, J., … Stanyer, J. (2017). Political communication in a high-choice media environment: A challenge for democracy? Annals of the International Communication Association, 41(1), 3–27. Wojcieszak, M., de Leeuw, S., Menchen-Trevino, E., Lee, S., Huang-Isherwood, K. M., and Weeks, B. (2021). No polarization from partisan news: Over-time evidence from trace data. The International Journal of Press/Politics. https://​doi​.org/​10​.1177/​19401612211047194. Wu, T. (2017). The Attention Merchants: The Epic Scramble to Get Inside Our Heads. New York: Vintage.

8. Computational approaches to online political expression: a framework for research Mengyu Li, Luhang Sun, Yiming Wang, Yibing Sun, Hyerin Kwon, Jiyoun Suk, JungHwan Yang and Dhavan V. Shah

Communication research’s emphasis on political talk can be traced back to the social interactionism of nineteenth-century French sociologist Gabriel Tarde. As Terry Clark (1969) and Elihu Katz (2006) remind us, Tarde argued for conversation’s place at the center of sociological inquiry, articulating a complex theory of “inter-mental activity” concerning how people influence one another. In so doing, he developed the concepts that later became known as the two-step flow of communication and opinion leadership, among other propositions of interpersonal influence (Lazarsfeld et al., 1944; Berelson et al., 1954; Katz and Lazarsfeld, 1955). Tarde’s insight about the late nineteenth century that “newspapers have transformed … the conversations of individuals, even those who do not read papers but who, talking to those who do, are forced to follow the groove of their borrowed thoughts. One pen suffices to set off a million tongues” (1969 [1898], p. 313), is just as true today of social media (Hanna et al., 2013). Tarde, often presented as the foil to Emile Durkheim’s efforts to distinguish the study of society from that of human psychology, was profoundly concerned with the interplay of mental and social forces, particularly as seen in the locus of conversational exchanges. Tarde was particularly focused on the relationship between mass communication and interpersonal talk for the formation of publics and their opinions. Katz (2006, p. 267) describes Tarde’s mediated process in the following manner: “To the press, he assigned the role of creating a public … The press, then, sets an agenda for the conversation of the cafes. Opinions are clarified and crystallized in these conversations, and then translated into actions in the world of politics”. The central tenets of this ordered model – press, conversation, opinion, and action – are supported by research on multi-step flow (Rogers and Shoemaker, 1971), opinion leadership (Shah and Scheufele, 2006), and communication mediation (Lee et al., 2013). Much of this research has employed cross-sectional and panel survey methods to examine the causes and consequences of political conversation, both face-to-face and online, and establish talk’s relationship with other factors involved in active citizenship (McLeod et al., 1999; Shah et al., 2005). Complementing this research is work on the actual content of conversations that relies on content analytic and ethnographic methods. For example, Papacharissi (2004) considered the quality of online political 110

Computational approaches to online political expression  111

talk by studying the level of civility in 287 discussion threads drawn randomly from 147 political newsgroups, concluding that discussions were heated but civil. Using an ethnographic approach, Walsh (2012) used participant observation of 37 recurring groups from 27 distinct communities across the state of Wisconsin to show how class- and place-based identity is linked with perceptions of relative deprivation. Sustained attention to conversation as a site of democratic understanding – and misunderstanding – testifies to its centrality in the study of political life (Schmitt-Beck and Lup, 2013; Rossini et al., 2021). Yet perhaps a more important aspect of Tarde’s work, given the growing convergence of information and conversation in social media (Shah et al., 2017), was his recognition “that opinions are really formed through the day-to-day exchange of comments and observations which goes on among people … by the very process of talking to one another, the vague dispositions which people have are crystallized, step by step, into specific attitudes, acts and votes” (Katz, 1992, p. 80). From this perspective, conversation is not simply a site of networked information exchange but also an opportunity for reasoning, reflection, and clarification of one’s own views (see Pingree, 2007; Shah, 2016). In digital media environments, as sources of news and sites of conversation have merged – amplified and reinforced within media ecologies containing increasingly polarized news outlets – attention to political talk continues to shape research on networked societies, including the reinforcement and distortion of views. The Laws of Imitation (Les lois de l’imitation) (1903 [1890]), Tarde’s most widely known work in English, speaks to communication and social influence within such settings. It also marks him as the “founding father of innovation diffusion research” (Kinnunen, 1996), charting processes of communicative invention, reproduction, and opposition. Methodologically, his calls for attention to observable interpersonal interactions, particularly within conversational processes, also presage the approaches central to computational social science, where each interaction “leaves digital traces that can be compiled into comprehensive pictures of both individual and group behavior, with the potential to transform our understanding of our lives, organizations, and societies” (Lazer et al., 2009, p. 721). The technological affordances of social media permit tracking of message creation and expression within a network, as well as reception and diffusion through systems (Namkoong et al., 2010; Han et al., 2011; Zhang et al., 2022). Tarde’s insights about “invention” and “imitation” provide ways to study online talk as it intersects with deliberative democracy, opinion, and citizenship (Schudson, 1978; Barber, 1984; Habermas, 1985; Katz, 1992, Kim et al., 1999; Price and Cappella, 2002; Mutz, 2006). Political expression and conversation in online environments lend themselves to the sort of large-scale, highly detailed interactional analysis that Tarde’s approach advocated and that Elihu Katz’s synthesis emphasized. As Bruno Latour (2010, p. 198) recognized, “it is indeed striking that at this very moment, the fast expanding fields of ‘data visualization’, ‘computational social science,’ or ‘biological networks’ are tracing, before our eyes, just the sort of data Tarde would have acclaimed.” “Big data” provides a way to understand everyday political talk online, its triggers,

112  Handbook of digital politics

content, and structures, especially in the more fragmented and polarized online information flow in the Web 2.0 era.

SHIFTING FORMS OF POLITICAL TALK With the rise of social media, information ecology became increasingly complex, coupled with change in mobile telephony (Ling and Campbell, 2017) and the broader political environment (Abramowitz and Saunders, 2008; Bafumi and Shapiro, 2009;). These changes have shifted the landscape of news outlets, how audiences consume news, and how political expression takes place. One significant change is that the boundary between interpersonal communication and mass communication has been further blurred (Friedland et al., 2006), and thus interpersonal communication has become unprecedentedly powerful in influencing the flow of information and processing. In the media environment where the sources of information and sites of conversation have begun to converge, ideological information amplifies and reinforces itself within increasingly polarized news media ecologies. Audiences can self-select into niche online platforms where they share similar views to reinforce and amplify their pre-existing political attitudes. For example, Jiang et al. (2021) found that when conservative and liberal Twitter users talk about Covid-19, conservative Twitter users are more likely to talk about the disadvantages of the vaccines; therefore, they talk “beyond” each other, looping in the expressive echo chamber without crosscutting conversation. This process has been specified in the form of a “citizen communication mediation model” (Shah et al., 2007, 2017), which finds that, consistent with Tarde’s framework, media influences are strong, but largely indirect, shaping opinion and action through effects upon face-to-face and online discussions about the news. Psychological processes such as selective exposure and motivated reasoning, facilitated by online communication mediation, push conversations along the ideological line. Online political talk not only channels media influence on efficacy and engagement but also sets the agenda for news coverage. The attention economy in the digital era has, to a large extent, changed how media institutions work. The public has greater power in driving the attention of media and politicians, potentially improving government accountability, and driving policy discussions (Jones and Baumgartner, 2005; Hanna et al., 2013). The public are no longer only the receivers of the political reality and agendas; they can now influence the political agenda and construct the political reality (Guo et al., 2012). For example, audience metrics tracked the online attention to the Black Lives Matter movement and drove both policy change and elites’ response (Freelon et al., 2018). Zhang et al. (2022) also found that discourses about sympathy and gun control on Twitter preceded news media rather than the other way around. On the other hand, people with different opinions can build their own agenda on various niche platforms where they can manipulate the factors that shape audience

Computational approaches to online political expression  113

attention to create an alternative reality. For example, people who participated in the storming of the US Capitol on January 6, 2021 used Parler and Gab to communicate with like-minded people who believed the election was rigged, eventually leading to the insurrection (Romero, 2021). The extremist group and fake news providers may utilize emotion’s role in driving attention to construct and spread their ideas, which can harm democracy (Wells et al., 2016). Zhang et al. (2018) also found that Trump’s followers intentionally amplified Trump’s Twitter account and posts and drove the attention paid to a specific person. It is thus essential to think about how society can ensure that the voices of the public and minorities are heard while holding oppositional actors accountable.

COMPUTATIONAL AND COMPARATIVE APPROACHES Facing these strong shifts, leading political communication scholars have debated whether and how the classic communication theories are still viable in the shifting media environment and what methodological innovations can be made to better understand contemporary communication phenomena (Bennett and Iyengar, 2008; Holbert et al., 2010). The complexity of online political expression calls for both theoretical reflection and methodological innovation. The affordances of social media provide people with tools to find, share and create content. And the process of generating content, whether images, videos, or posts, may demand deeper forms of reflective and compositional processing (Freelon, 2010; Ekström and Östman, 2013). Social media have also increased people’s political participation, especially in the form of hashtag activism (Bode et al., 2014; Suk et al., 2021). At the same time, this heightened political expression within cloistered social networks may create its own set of problems, such as declining social trust (Suk et al., 2021), contentious and oppositional activism (Zhang et al., 2022; Jiang et al., 2021), and source selectivity (Zhang et al., 2019). The changes have also encouraged a set of new explorations about how political engagement through social media may shift the elite-driven, top-down political process into more interpersonal talk and bottom-up participation. The trace of social media interactions allows scholars to further explore online discourses around major social issues, their process, content, and structures. The computational social science approach provides us with a way to better understand the public’s thoughts and reactions to everyday political talk as digital trace data often come with temporal and relational information. For example, Zhang et al. (2019) constructed event data around mass shootings to examine how Twitter users responded to those events. By applying natural language processing methods to time-stamped and networked Twitter data, they found that Twitter offers a space for public mourning and contestation, albeit the distinct communities of social media discourse reveal the limits of an online public sphere for deliberative exchanges. Within the context of social movement mobilization, computational methods can help to identify the temporal dynamics, prominent accounts, and networks of collective actions. Along these lines, Suk et al. (2021) used Twitter data to examine how the success of the

114  Handbook of digital politics

#MeToo movement is sustained by constructing a network of acknowledgment to build a sense of shared experience and identity. Work employing a range of methods – from ethnographic to computational – examines these phenomena across platforms, extending the work from Twitter, the most accessible research API, to Facebook, Instagram, YouTube, and Snapchat (Giglietto et al., 2012; Bossetta, 2018). It is important to note that social media platforms differ in affordances and audiences, with many studies focusing on the distinctions among social media platforms such as Facebook, Twitter, and WhatsApp as sites of political expression and disagreement (Kligler-Vilenchik et al., 2020), as well as the relationship of issue discourses across platforms (Lukito, 2020). Research on political talk may benefit from comparing diverse online discourses that can be found on different social media platforms and from understanding how different affordances and features may be related to what and how to talk about controversies online. Social media also operate in ways that often cross geographic boundaries, enabling social movements to traverse borders, such as environmental justice and #MeToo movements. The application of computational methods further enables comparative analysis across different countries with varying languages. Research on international and comparative communication processes could further benefit from computational methods to evaluate social media posts across countries, languages, and regional contexts. For instance, Maier et al. (2022) compared two methods for conducting cross-lingual analysis, machine translation and multilingual dictionaries, and applied them to topic modeling analysis. Similarly, Lopez et al. (2019) applied Linguistic Inquiry and Word Count (LIWC) to compare the #MeToo tweets in English and #BalanceTonPorc tweets in French. They found that although these two movements share some similarities, the discussions were largely shaped by local culture and social realities. This line of research has begun to explore the concept of “glocalization” (Kraidy, 1999), how global forms become localized as people construct meaning with them in their own ways. Research is also expanding to consider visual features of social media discourse, partly spurred by advances in computer vision and image detection techniques. For example, Bucy et al. (2020) compared visual, tonal, and verbal sub-indices and linked them to Twitter responses. Among the three cues, nonverbal cues in broadcast content appear to be the most effective in arousing social media reactions, further indicating their importance to research on political expression. Images, as an important feature, carry emotions that could further mobilize people.

#METOO AND GLOBAL ACTIVISM To illustrate this computational and comparative approach as a framework for research on political expression, we use public online data to understand the discourses surrounding two #MeToo cases that occurred in 2018 in the United States and South Korea, nearly a year after the 2017 eruption of #MeToo into the public consciousness: (1) Christine Blasey Ford’s sexual assault allegation against Brett

Computational approaches to online political expression  115

Kavanaugh, who was ultimately appointed as associate justice of the Supreme Court; and (2) the rape allegation against Ahn Hee-jung, a prominent South Korean governor, who was ultimately jailed for repeatedly assaulting his secretary Kim Ji-eun. The data used to analyze the two cases were collected from Synthesio (www​ .synthesio​.com), a commercial social listening platform that provides access to multilingual public data on various online domains, including both mainstream online platforms, such as Twitter, Facebook, and Instagram, as well as local forums in multiple languages, such as Spanish, Korean, Chinese, etc. To examine online discourses of the two #MeToo cases in the United States and South Korea, we filtered online posts containing both #MeToo-related keywords (e.g., sexual assault, sexual harassment, #MeToo and localized hashtags) and the names of accusers and the accused in English and Korean. The time frame for each data collection was six months from the date of the accusation, allowing for some time before this date. In the case of Ahn, the data spanned March 2018 to September 2018, and for Kavanaugh it spanned July 2018 to January 2019. The Korean-language #MeToo content surrounding Ahn Hee-jung/Kim Ji-eun in South Korea had a total data volume of 35,460 posts during the study period; the English-language #MeToo content concerning Brett Kavanaugh/Christine Blasey Ford in the United States had a total data volume of 1,014,461 posts. These posts spanned multiple platforms. Specifically, our dataset contains publicly available posts from prominent platforms and forums, comparing South Korea and the United States. For the Ahn Hee-jung case, we focus on publicly available posts on Facebook, Twitter, YouTube, and two most popular Korean forums:1 Nate Pann (pann.nate.com) and DC Inside (dcinside.com). For the Christine Blasey Ford case, we focus on publicly available posts on Facebook, Instagram, Reddit, Twitter and YouTube. To ensure the comparability of US’s Reddit and the Korean platforms, we combine data from the Nate Pann and DC Inside to a new category labeled as a forum in the following analysis. To analyze these data, we used a range of computational methods to understand the political expressions across online platforms, including temporal dynamics on different online platforms, analyses of keyword frequency, and structural topic modeling. After establishing the discourse patterns across platforms and contexts, we focus on the frequency of keywords to understand the linguistic features of online discourses on different platforms. Next, we use unsupervised natural language processing to build structural topic models of the themes and ideas clustering people’s discourse on different platforms. Before doing this, we provide contextual background information on the #MeToo allegations against Ahn Hee-jung and Brett Kavanaugh.

COMPARATIVE CASE STUDY APPROACH The cases of Ahn Hee-jung and Brett Kavanaugh were selected due to their similarities within the #MeToo movement and how they were treated in the distinctive socio-cultural environments of South Korea and the USA. Both high-profile #MeToo allegations brought forward by Kim Ji-eun and Christine Blasey Ford, respectively,

116  Handbook of digital politics

were directed at very prominent political figures with significant social implications. Despite different outcomes for the two accusations, with Ahn Hee-jung convicted and jailed and Kavanaugh exonerated and elevated, both accusations received considerable attention in the South Korean and US media systems. On the other hand, compared with the United States, Korea is a more conservative society characterized by patriarchal systems and traditional gender norms inherited from Confucianism. South Korea is also a collectivist society, placing enormous value on interdependence through deference to social norms, whereas the United States has a highly individualistic culture. Although later in time, we will begin with Christine Blasey Ford’s accusation and Brett Kavanaugh’s confirmation before turning to Ahn Hee-jung’s accusation and Kim Ji-eun’s eventual conviction. Christine Blasey Ford’s accusation: In July 2018, Christine Blasey Ford came forward with allegations that Brett Kavanaugh, the nominee to an open seat on the Supreme Court of the United States sexually assaulted her in the summer of 1982, when the two were high school students. Ford alleged that Kavanaugh and his friend Mark Judge pushed her into a bedroom at a party, where Kavanaugh pinned her down, attempted to remove her clothing, and held his hand over her mouth when she struggled to scream for help. She was able to escape from what she later referred to as a “rape attempt”. Kavanaugh immediately denied Ford’s accusation, with his nomination mired in controversy and increasing uncertainty. As a result, the Senate Judiciary Committee worked with Ford’s legal team, led by lawyer Debra Katz, to schedule a hearing about the allegation on September 27 in which Ford testified about her accusation. Social media reacted to Ford’s allegation and Kavanaugh’s hearing in real time. Some thanked and applauded Ford for her bravery, while others defended Kavanaugh and questioned the validity of Ford’s accusations. One of the most prominent defendants of Kavanaugh was ex-president Donald Trump. To cast doubt on Ford’s allegations, he tweeted “I have no doubt that, if the attack on Dr. Ford was as bad as she says, charges would have been immediately filed with local Law Enforcement Authorities by either her or her loving parents. I ask that she bring those filings forward so that we can learn date, time, and place!” Within hours of Trump’s tweet, #WhyIDidntReport became the top trending topic on Twitter. Thousands of survivors from all over the United States participated in this movement to tell their stories and disclose why they didn’t report their sexual violence experiences. Like the 2017 #MeToo movement, #WhyIDidntReport created a means of collective acknowledgment. Kim Ji-eun’s accusation: Kim Ji-eun, a female secretary of Ahn Hee-jung, governor of South Chungcheong Province, appeared on a news channel on March 5, 2018, to disclose Ahn’s pattern of indecent behavior and sexual assault, which included four rapes over eight months. Kim also mentioned that she knew there were other victims, and she hoped that her interview would encourage them to speak out. Ahn’s office initially denied the accusations and claimed that Ahn had consensual sex with Kim. This incident and its relevant topics rapidly became top trending items on Korean online space. With the allegations going viral on social media and receiving

Computational approaches to online political expression  117

attention from news media, Ahn retracted his office’s statement and apologized in a Facebook post, followed by his announcement of resignation and retirement from public life. However, Ahn was acquitted in the first trial on August 14, 2018, which spurred a second wave of online activism to condemn the court decision. The increasing volume of online discussion fostered new slogans, translated as “#Ahn Hee-jung_to_jail”, “#judiciary_guilty”, “police_unjust_investigation”, and “sexual_ assault_by_power”. A year and a half after Kim’s disclosure, the South Korean Supreme Court overturned his acquittal and found Ahn guilty of sexually assaulting his secretary using official powers. He was arrested and sentenced to three and a half years in jail.

#METOO OVER TIME AND ACROSS PLATFORMS We begin our analysis of online expression by examining the overtime distribution of posts concerning these #MeToo cases. Figure 8.1 (Ford-Kavanaugh) and Figure 8.2 (Kim-Ahn) present the total volume of online posts for the Kavanaugh case from July 2018 to January 2019 and Ahn case from March 2018 to September 2018 across multiple online platforms. Although Kavanaugh’s nomination was made in July 2018, a heated discussion concerning the #MeToo accusation started in mid-September and continued until mid-October in 2018, during the height of the revelations and confirmation hearings. For the discussion of the Kim-Ahn case, most of the discussions were from March to April 2018, while there was a second wave in mid-August 2018 on Twitter and two forums. This second wave of discussion coincided with the decision made by the Seoul Western District Court that acquitted Ahn Hee-jung on August 14, 2018. This decision triggered the second wave of online #MeToo discussion in South Korea. To identify how platforms’ trends match up with each other, we applied cross-correlation analysis. More specifically, we applied data imputation techniques2 for missing data across platforms and then conducted cross-correlation analysis by aligning each pair of series. Results indicate a high degree of temporal alignment between platforms in both Ford-Kavanaugh and Kim-Ahn cases. As shown in Tables 8.1 and 8.2, the value of the lag with the highest correlation coefficient represents the best fit between the two series. Discussion of accusations against Kavanaugh simultaneously spread across five platforms, except that the volume of Twitter posts occurred at a one-day lag later than on Facebook, Instagram, and YouTube. By contrast, public opinion on the Ahn case on YouTube preceded discussion on Twitter, Facebook, and Forum in a one-day period. The lag was somewhat longer for Twitter.

118  Handbook of digital politics

Figure 8.1

Total volume of online posts for the Ford-Kavanaugh case on Facebook, Instagram, Reddit, Twitter, and YouTube from July 2018 to January 2019

Figure 8.2

Total volume of online posts for the Kim-Ahn case on Facebook, Twitter, YouTube, and two forums Nate Pann and DC Inside from March 2018 to September 2018

Computational approaches to online political expression  119

Table 8.1

Cross-correlation estimates for online posts related to the Ford-Kavanaugh case on Facebook, Instagram, Reddit, Twitter, and YouTube from July 2018 to January 2019

Platforms

Facebook

Instagram

Reddit

Twitter

YouTube

Facebook

1.00

 

 

 

 

Instagram

0.91 (lag = 0)

1.00

 

 

 

Reddit

0.95 (lag = 0)

0.85 (lag = 0)

1.00

 

 

Twitter

0.91 (lag = 1)

0.79 (lag = 1)

0.90 (lag = 0)

1.00

 

YouTube

0.92 (lag = 0)

0.79 (lag = 0)

0.95 (lag = 0)

0.90 (lag = −1)

1.00

Table 8.2

Cross-correlation estimates for online posts related to the Kim-Ahn case on Facebook, Twitter, forums, and YouTube from March 2018 to September 2018

Platforms

Twitter

Facebook

Forum

YouTube

Twitter

1.00

 

 

 

Facebook

0.79 (lag = −4)

1.00

 

 

Forum

0.70 (lag = −5)

0.79 (lag = 0)

1.00

 

YouTube

0.74 (lag = −1)

0.80 (lag = −1)

0.69 (lag = −1)

1.00

#METOO KEYWORD FREQUENCY BY CONTEXT To examine the content of this overtime discourse, we analyzed the keywords that occurred frequently in the online discussions, considering both total keyword volume and relatively high-frequency keywords across different platforms for both #MeToo cases. In Figure 8.3 (Ford-Kavanaugh) and Figure 8.4 (Kim-Ahn) each bar represents the volume of the keyword on the vertical axes. The total keyword volume provides a broad understanding of online discussion of these two #MeToo cases across mainstream online platforms in the US and South Korea. Figure 8.3 shows the top 30 total high-frequency keywords of the Ford-Kavanaugh case. Keywords centered on two themes: sexual assault from the perspective of the #MeToo movement and political discussions around Trump nominating Kavanaugh to the Supreme Court. As one of the most politicized #MeToo cases in the US, Dr. Ford’s assault allegation gained considerable attention, spurred heated debates and drove contentiousness. Politicized words such as “Trump”, “Supreme”, “Court”, “Senate”, etc. were mentioned frequently. Meanwhile, terms such as “believe” “accus*”, “alleg*”, and “lie”, which convey uncertainty, may reflect the polarization and contentiousness of political expression in the US. In contrast, the Korean #MeToo case was decidedly less polarized as reflected in the keyword list in Figure 8.4. The top 30 high-frequency keywords for the Ahn Hee-jung case across Facebook, Twitter, YouTube and the two mainstream Korean forums, Nate Pann and DC Inside center on the sexual assault case itself, and the keywords include “sexual assault”, “victim”, “sexual violence”, “sexual harassment”, etc. in Korean, which are fundamental high-frequency keywords in the #MeToo

120  Handbook of digital politics

Figure 8.3

Total keyword frequency for the Ford-Kavanaugh case on Facebook, Instagram, Reddit, Twitter, and YouTube

Figure 8.4

Total keyword frequency for the Kim-Ahn case on Facebook, Twitter, YouTube, Nate Pann and DC Inside

movements, and similar to the Ford case. Although politics-related words such as “Chungnam Province Governor”, Ahn’s position before his resignation, and “Tak Hyun-min”, who was a secretary of former president Moon Jae-in, were mentioned

Computational approaches to online political expression  121

frequently in the Korean dataset, they were not deployed in a polarized manner. There is also less language questioning or doubting the veracity of the accusation. When turning to the high-frequency keywords by platform, we are able to see more nuanced linguistic features. Table 8.3 shows the top 30 high-frequency keywords for the Ford-Kavanaugh case on different platforms. Compared to other platforms, Twitter contained more contentious discussion directly related to accuser and accused and addressed the sexual assault aspects with the hashtags #MeToo, #believesurvivors and #stopkavanaugh. Twitter accounts of Republican politicians such as Donald Trump (@realdonaldtrump), who nominated Kavanaugh to the Supreme Court, and key Senators including Susan Collins (@senatorcollins), Chuck Grassley (@chuckgrassley), Jeff Flake (@jeffflake), and Lisa Murkowski (@lisamurkowski), along with @GOP, were all frequently mentioned around the discussion of the Ford-Kavanaugh case. Table 8.3

High-frequency keywords for the Ford-Kavanaugh case on Facebook, Instagram, Reddit, Twitter, and YouTube

Facebook

Instagram

Reddit

Twitter

YouTube

detent

#sexualassault

peopl

rt

subscrib

starr

#repost

evid

ford

cbs

center

#feminist

point

dr

cia

immigr

bio

think

sexual

democrat

clinton

#humanright

like

blasey

gorka

rauner

#femin

thing

christin

youtub

geo

#plannedparenthood

even

assault

guilti

parti

#humanist

also

#metoo

instagram

olson

#supremecourt

someon

#believesurvivor

news

role

#ppnycaction

seem

mock

trevor

arkansa

#dc

actual

#kavanaugh

channel

georg

#purjuri

feedback

#stopkavanaugh

parti

right-w

#activist

reason

survivor

facebook

strauss

#womenempower

mean

@realdonaldtrump

even

democrat

#activism

polit

ralli

like

state

@get_repost

possibl

mississippi

lang

border

#atheism

argument

misconduct

video

contract

#human

make

@senatorcollin

get

richard

#atheist

someth

#brettkavanaugh

newslett

cabal

#nov6

rape

breaking

year

bush

#elections2018

though

@chuckgrassley

innoc

hous

#protest

articl

trump

cbsn

compani

#genderjustic

probabl

brett

esterhazi

arbit

#womensright

pretti

@nbcnew

peopl

flow

#socialjustic

summari

#kavanaughhear

patreon

unname

#unitedst

comment

@jeffflak

feminist

year

#equal

parti

@msnbc

copyright

pretens

#washington

anyth

watch

googl

labour

#illustr

way

@lisamurkowski

fact

decent

#intersectionalfemin

fuck

@gop

pinterest

122  Handbook of digital politics

In contrast to online discourse on Twitter, the political talk on Facebook showed a general right-wing orientation, with discussion centered on a prior accusation of sexual assault against Bill Clinton, the former president, and his impeachment related to the cover-up of sexual misconduct, with both used to criticize Democrats and Ford. As the biggest mainstream forum, language on Reddit was more informal and conversational, which might be because forums like Reddit afford more spaces for engaging opinion sharing and discussion without word number limit. While on Instagram, a visual-based social media platform, hashtags dominated the textual content and discussion around the Ford-Kavanaugh case that leaned left. Both #MeToo-related hashtags and politicized hashtags were frequently used on Instagram supportive of women’s rights, feminist empowerment, and gender justice, with #MeToo embedded in a broader discussion about human rights and activism (e.g., #humanright, activist, #activism). Interestingly, as a key interlinking platform, YouTube contained high-frequency keywords of other platforms, such as Instagram, Facebook, and Pinterest, and the names of prominent hosts from the right and left (e.g., Sebastian Gorka, Trevor Noah), along with subscription and Patreon sponsorship references. Table 8.4 shares the top 30 high-frequency keywords for the Korean #MeToo case across the platforms of Facebook, Twitter, YouTube, and two mainstream forums (Nate Pann and DC Inside). We found different linguistic patterns compared with the Ford-Kavanaugh case. For the US case, all the discussion was centered on the #MeToo incident between Ford and Kavanaugh and the Supreme Court nomination of Kavanaugh. While in South Korea, people mentioned many other #MeToo perpetrators along with Kim-Ahn in online discussion, such as well-known Korean dancer Ha Yong-bu and director Lee Yoon-taek. Although many celebrities or politicians were also mentioned frequently for the Korean case, these people did not have direct relationships with the accused Ahn Hee-jung, but rather were referenced more generally. Similar to the Ford-Kavanaugh case on Twitter, the Kim-Ahn case also featured prominently in online discussion on Twitter, especially from a perspective of criminal investigation and legal procedure. Keywords such as “Antragsdelikt” (a crime requiring a victim complaint), “National Sexual Violence Counseling Council”, “judgment”, and “prison” were mentioned frequently on Twitter. In contrast, online discussion on Facebook was more politicized: keywords such as “mania”, “deodeumeo manjindang”,3 “left-wing”, “party” and “Moon administration” were mentioned frequently on Facebook. Since Ahn Hee-jung was a well-known and powerful Democratic politician, online discussion of a sexual assault accusation became polarized, particularly against the left. It is interesting to observe the similarity between online discussions of both US and South Korean #MeToo on Facebook. Different from the conversational discussion of the Ford-Kavanaugh case on Reddit, the Korean discussion of the Kim-Ahn case on the two biggest forums, Nate Pann and DC Inside, contained endorsements from various celebrities or professionals active in the #MeToo movement in South Korea, including Kim Soo-jeong, a female lawyer who fights for women’s rights and Roy Jo, a male YouTuber who

Facebook

안희정

매니아

성폭행

더듬어만져당

준비한

정당

성추행

구슬리다

우상화

역겹다

절친

상사

좌파

박재동

법정관리

생각함

클라스

페미니스트

김정은

고교동창

안전지사

전체보

Tak Hyun-mina

victim

accusation possible

expulsion

Antragsdelikt

request

like that

criminal

fiction

do not

resignation

Ahn Hee-jung

accusation

judgment

prison

lose

sexual assault

Kookmin bank

there is no

everyday

National Sexual Violence 만진당

문정부되고

 

Counseling Council

to be

perpetrator

Hong Joon-pyon

탁현민

피해자

고발가능

출당처리

친고죄

의뢰

저렇다

범죄자

허구

안하다

사퇴

안희정

고발

판단

감옥

지다

성폭행

국민은행

없다

일상으

전국성폭력

상담소협의회

되다

가해자

홍준표

제조

김수정

whole report

safety department

위하다

김동욱이

동성간

배틀그라운드

Manjindangm Moon administration

박근혜

남성

이재용

감염

뿌닛

베리칩

검은사막

카카오

김동욱

박찬주

페미니즘

심장

라이브

시청자

방송

김지호

high school alumni

Kim Jong-un

feminist

class

thinking

court management

Park Jae-dongk

left-wing

superior

best friend

disgust

idolization

cajole

sexual harassment

party

prepare

deodeumeo manjindang 문제

sexual assault

mania

Forum 조국

  Ahn Hee-jung

for

Kim Dong-wook

Same-sex

battleground

Park Geun-hye

male

Lee Jae-yongl

infection

bbunit

Berry chip

Black desert

Kakaoi

Kim Dong-wookh

Park Chan-juf

feminism

heart

live

viewers

broadcast

Kim Ji-hod

problem

produce

Kim Soo-jeongc

Jo Gukb

 

기사문의

계열사

질문

연구원

사람

검찰

귀가

충남도지사

흠결

트위

충남지사로부터

최영일

시사평론가

로이조

인터뷰

수행비서

유신쇼

손석희가

상식

제보

조사

베트남

김지은

유튜브

YouTube

report request

branch

questions

researcher

person

prosecution

homecoming

Governor

Chungnam Province

flaw

twit

Governor

from Chungnam

Choi Young-ilj

current criticism

Roy Jog

interview

secretary

Yushin Show

Son Seok-heee

common sense

report

inspection

Vietnam

Kim Ji-eun

Youtube

 

High-frequency keywords for the Kim-Ahn case on Facebook, Twitter, YouTube, and mainstream forums

Twitter

Table 8.4

Computational approaches to online political expression  123

직감

하용부

한만삼

형수보찢

블랙리스트

이윤택

sexual violence

curious

Kim Ji-eun

daily life

sweet

symbiosis

성폭력

궁금한

김지은

일상

달다

공생

 

Lee Yoon-taekr

blacklist

Hyungsu boggit (slur)

Hanman Samp

Ha Yong-buo

intuition

동성애

관하다

북두

대하다

프렌즈마블

추천

Forum

방문조사

treat 스마트폰 여성게스트

related homosexuality

손정혜

담담

Friends Marvel big dipper

남조선

YouTube

recommendation

 

Notes: a Tak Hyun-min, a secretary of former president Moon Jae-in. Tak is regarded as an anti-feminist politician in South Korea. b Jo Guk, a former Minister of Justice. c Kim Soo-jeong, a female lawyer who fights for women’s rights in South Korea. d Kim Ji-ho, a male lawyer related to the case of the accused Park Chan-ju in South Korea. e Son Seok-hee, an anchor and general director of JTBC (Joongang Tongyang Broadcasting Company). f Park Chan-ju, a South Korean four-star general. g Roy Jo is a male YouTuber and streamer in South Korea who supports #MeToo and feminist movements. h Kim Dong-wook, a subordinate of the general, Park Chan-ju in South Korea. i KakaoTalk, the No.1 instant messaging application in South Korea. j Choi Young-il, a news commentator who criticized #MeToo perpetrators in South Korea. k Park Jae-dong, a Korean artist who was accused of sexual harassment. l Lee Jae-yong, a vice chairman of Samsung company. m Manjindang, similar to “Deodeumeo manjindang”, a curse word to mock the Democrats in South Korea. n Hong Joon-pyo, a Republican politician and the current Mayor of Daegu in South Korea. o Ha Yong-bu, a famous Korean dancer, a perpetrator of #MeToo in South Korea. p Hanman Sam, a Korea priest and perpetrator of #MeToo in South Korea. q Son Jung-hye, a Korean lawyer who publicly talked about the #MeToo movement in South Korea. r Lee Yoon-taek, a Korean director, who was jailed for sex assaults.

Facebook

 

Twitter

female guest

Smartphone

Son Jung-hyeq

in-person inspection

calm

South Josun

 

124  Handbook of digital politics

Computational approaches to online political expression  125

supports #MeToo and feminist movements in South Korea. Other celebrities beyond the #MeToo movement were also mentioned frequently, likely reflecting the nature of forums. Although South Korea is in general more conservative in terms of gender and feminist issues, our results reveal that the online discussion of the Kim-Ahn #MeToo case was less contentious than the political talk of the Ford-Kavanaugh case in the US.

#METOO TOPIC PROMINENCE For a more comprehensive comparison between the two #MeToo cases, we applied Structural Topic Modeling (STM) to extract thematic clusters from the Ford-Kavanaugh and Kim-Ahn cases. Topics were derived from posts by probabilistic algorithms following the principle that co-occurrences of words in and across posts could reflect underlying topics. In other words, the more frequently words co-occur in documents (i.e., forum posts/tweets), the higher the probability that they constitute a topic. Moreover, given that STM allows for adding covariates in the estimation model, we use platforms of each post as the covariate to detect the variability of topics across platforms. We picked a 10 percent random sample for STM analysis of the content surrounding the Ford-Kavanaugh case and conducted STM on the full dataset of the Kim-Ahn case. As shown in Table 8.5, results of STM yielded seven topics from the US case, and as shown in Table 8.6, six topics from the Korean case. For the Ford-Kavanaugh case, “gender inequality” (19.30 percent) ranks top of the list of most prominent topics, including discussion on how gender power imbalance manifested in reality. For example, one representative post is “I want a world where people aren’t kept from doing something specifically because of their gender (being paid equally wages, being able to be pro-choice, having everyone’s emotions taken seriously regardless of their gender stigma)”. The second most prominent topic (19.10 percent) focuses on uncertainty around the accusation and calls for more investigation into the details of disclosure by the FBI. A related topic featured skeptical voices regarding Ford’s testimony (15.96 percent). The main reason for skeptics to discredit Ford was “it is all witness testimony, none of it is supported by material evidence of the event. Specifics of the location and time are unavailable”. The next most prominent topic of the Ford case is hashtag activism, characterized by a series of hashtag symbols (e.g., #stopkavanaugh #kavanaugh #protest, #kavanaugh, #IBelieveChristine) and targets of social media focus such as @ senatorchuckgrassley @senlisamurkowski @sensusancollins @jeffflake @joemanchinwv @senatorheitkamp. People are also divided in terms of their position on the nomination of Brett Kavanaugh to the Supreme Court (11.99 percent). Some people were strongly opposed to nominating Kavanaugh, while others supported the nomination. In addition, another topic concerns opinions surrounding President Donald Trump, who posted a tweet questioning why Christine Blasey Ford did not immediately report the alleged attack (10.46 percent). People also paid attention to

126  Handbook of digital politics

Table 8.5

Structural topic modeling of prominent topics in the Ford-Kavanaugh case

Topic

Proportion

T2 gender inequality

19.30%

Top terms Highest Prob: like, women, just, get, believe, people, think, assault, rape, men FREX: apologetic, feminine, pumpkin, frantic, incel, stat, entitle, duke, harder, angry

T4 investigating Ford

19.10%

accusation

Highest Prob: assault, sexual, ford, Dr, allege, investing, FBI, senate, Kavanaugh, Christine FREX: Debra, #rightsideofhistori, confine, Kavanaugh, #defendourmen, Ford, incomplete, decades-long, disclosure, Hogan

T6 skepticism of Christine 15.96%

Highest Prob: year, know, time, evident, happen, even, want, one, say, party

Blasey Ford’s credibility

FREX: date, redact, kosher, hrs, stair, Keyser, evidentiary, mischief, Smith, has

T3 hashtag activism

14.58%

Highest Prob: #metoo, ford, Dr, #believesurvivor, believe, Blasey, #kavanaugh, survivor, Christine, women FREX: #believesurvivor, #timesup, #ibelievechristineblaseyford, #resist, solidary, #cancelkavanaugh, #sexualassault, #theresist, #america, #republican

T5 opinions regarding Brett 11.99%

Highest Prob: Kavanaugh, accuse, Brett, court, supreme, sexual, lie, judge,

Kavanaugh’s nominee

allege, nominee FREX: bouf,* anti-women, scotus-perjuri, vacant, keyword, #dethronethek, FAQ, #ia04, Henderson, grenade

T7 opinions regarding

10.46%

Trump’s response

Highest Prob: trump, sexual, ford, Dr, Blasey, Christine, assault, president, mock, rally FREX: mock, rally, Mississippi, impersonal, cowardice, soulless, disdain, hideous, #releasethefbireport, slick

8.61%

T1

Highest Prob: democrat, politic, republican, now, work, clinton, vote, liberal,

#MeToo-related political

use, threat

events

FREX: click, stock, border, oklahoma, symbol, chart, trade, scientist, market, restaurant

Note:

* “Bouf” means to have sex, do the good thing on the bad leg.

other #MeToo-related political events and figures (8.61 percent). One example is to relate #MeToo to voting issues; “A vote for Beto is a vote for the continued normalization of sexual assault against women, e.g., Anthony Weiner, Terrance Patrick Bean, Russell Simmons, Eric Schneiderman, Tavis Smiley, Rep. Steve Lebsock, William Mendoza, Charles Wade (BLM), Cenk Uygur, Rep Bobby Scott, Andrea Ramsey … etc”.

Computational approaches to online political expression  127

Table 8.6

Structural topic modeling of prominent topics in the Kim-Ahn case

Topics

Proportion

Top terms (translated from Korean)

T3 Korean #MeToo online 27.77%

Highest Prob: Ahn Hee-jeong, sexual assault, governor, Hong Jun-pyo,

and offline activism

sexual violence, Kim Ji-eun, seoul, women, recommended, Moon Jae-in FREX: Gwanghwamun, homecoming, indicated, gun, plan, resolutely, full, Let’s get angry, in the first half of the year, bar

T5 opinions surrounding

19.13%

sexuality

Highest Prob: Ahn Hee-jung, sexual assault, same-sex, sexual harassment, #MeToo, rape, victim, sexual violence, AIDS, said FREX: same-sex, AIDS, heterosexual, pope, sexual contact, francisco, candidate, sexual contact, bisexual, Choi Young-ae

T4 legal procedures for

17.39%

accusation

Highest Prob: victim, Ahn Hee-jung, Kim Su-jeong, Kim Dong-wook, accused, criminal, resign, fictional, Tak Hyunmin, commissioned FREX: accuse, criminal, accusable, accuse sexual assault, expedited expeditiousness, disappearing, expelled from the party, accusation of criminal charges, Mr. Tak, judged

T2 conspiracy regarding

16.42%

Highest Prob: Ahn Hee-jung, sexual assault, Hong Jun-pyo, Samsung E,

business conglomerates and

victim, Lee Jae-yong, sexual harassment, Jang Chung-ki, Moon Jae-in,

election

sexual violence FREX: Jae-yong Lee, Choong-gi Jang, Kun-hee Lee, Samsung Family, Yoon-hyeong Lee, Message, Part-time Samsung, Samsung Blood Alliance, Occupation Release Department, Samsung Department

T1 gender politics in South 10.65%

Highest Prob: governor, Ahn Hee-jung, sexual assault, victim, #MeToo,

Korea

said, revealed, prosecutors, feminism, woman FREX: Hoju,* normal, thought, everyone, chief, sorry, capture, trace, slide, interview

T6 #MeToo-related political events

8.64%

Highest Prob: Ahn Hee-jung, sexual assault, sexual violence, #MeToo, governor, victim, sexual assault, Kim Ji-eun, case FREX: federal system, sexual scandal, passport, Mr. Min, fiction, Jungkook, second-party, refusal, extreme, sexual freedom

Note: * “Hoju” (Korean) means the “head of the family” or “head of the household”. In South Korea, it was formally introduced in 1953. The abolition of the hoju system in 2005 is regarded as a historic victory for the Korean women’s movement in which Korean women fight against their subordinate position in the family.  

Returning to the Korean case, the most prominent topic is online and offline activism around the Korean #MeToo campaign (27.77 percent), which focused on not only expressing opinions regarding the case but also organizing protest activities. For instance, hundreds of activists and citizens gathered for the International Women’s Day ceremony in Gwanghwamun Plaza, Seoul, South Korea. The second most prominent topic is the discourse underpinning issues of sexuality in society (19.13 percent), including heterosexual and homosexual relationships, moralization of sex and religion, sexual orientation and gender identity. Next, people also urged legal authorities to conduct criminal investigations into the sexual abuse cases in South Korea (17.39 percent).

128  Handbook of digital politics

One unique topic in this Korean case concerns conspiratorial beliefs (16.42 percent) that the #MeToo campaign is planned by Samsung, one of the most influential business conglomerates in South Korea, for the purpose of manipulating the election. Another unique Korean talking point is women’s activism and their political protests in the history of South Korea (10.65 percent). Consistent with a broader #MeToo focus on gender issues, people mention the abolition of the hoju (head of the household) system in 2005 as a historic victory for the Korean women’s movement. Despite some differences in topics across two cases, the Korean case shares similarities with the US case in discussion of other events associated with the #MeToo movement (8.64 percent). In general, political expression across the two #MeToo cases shared common themes of (1) opinions regarding two allegations of sexual misconduct, (2) expression concerning gender inequality and politics as well as (3) extended discussions on other political figures and issues associated with the #MeToo campaign. However, distinctive political and cultural backgrounds of the US and South Korea inevitably shaped online political discussion on #MeToo cases in unique ways. In the discussion of Ford-Kavanaugh case, the results of STM illustrate a political polarization of #MeToo movement in the US. In the Kim-Ahn case, they reveal broader themes on sexuality, morality, women’s activism and gender issues. This focus on gender politics in South Korea developed through democratic political progress, a male-dominated Asian society. Also notable is the prevalence of conspiracy theories regarding the Korean #MeToo movement’s connection to Samsung, one of the largest family-run conglomerates in South Korea. We further examined how these topics varied across platforms. To better reflect on the similarities and differences between the two cases, we selected four themes that appeared across multiple platforms for comparison. Two of them were gender inequality (USA) and gender politics (KR), which were categorized into a comparison group due to their similarities in semantic contents. Another two topics are hashtag activism (USA) and Korean #MeToo activism (KR), both of which concern growing forms of activist movements either online or offline. For the Ford-Kavanaugh case, discussion on gender and feminist agendas accounted for the highest proportion on YouTube, followed by forum-type platforms (e.g., Reddit), Facebook, and Twitter. However, opinion regarding gender politics in the Kim-Ahn case comprises a greater share in discussion on forums than on YouTube, Facebook, and Twitter. This suggested that YouTube serves as an important public sphere of discourse about women’s issues in the USA, while forums such as DC Inside and Pann Nate afforded more discourses of gender politics based on the accusation against Ahn. In the US case, hashtag activism played an important role especially on platforms such as Instagram and Twitter. Turning to the online and offline activism discussion in Korea, it earned a greater proportion on Facebook compared with other platforms. The patterns may suggest the intertwining influences of platforms’ affordances and socio-cultural factors on political discourse through the cases of #MeToo.

Computational approaches to online political expression  129

A FRAMEWORK FOR RESEARCH Rather than reflecting on the specifics of these cases, which were mainly deployed to illustrate some basic potential of computational approaches to understanding online political expression, we close by using them to offer some broader suggestions about online political research in response to a multi-platform environment with dynamic information flows. We also highlight the value of computational methods for tracing and analyzing political expression in unobtrusive ways. Through a comparison of US and South Korean cases, we presented how public discourse on social issues varied across platforms, and how these systems allow individuals to share political or social claims in a networked public sphere. The collective/individual interdependence makes it possible to spur activism that speaks on behalf of victims and calls for social changes across cultural settings. Of course, online expression and activism may also spur the polarized discourses and conspiracy beliefs that have become increasingly commonplace. In this chapter, we emphasize some basic time-series analysis and natural language processing techniques. Time-series analysis is a means of deeper temporal analysis of online political talk in two respects. First, through understanding the time-dependent dynamics of a single series, univariate time-series analysis allows researchers to capture how a variable of interest behaves in a specific span and overtime properties. The great advantage of univariate techniques, particularly the autoregressive integrated moving-average (ARIMA) model, is to detect if an autoregressive process exists in an individual series, which should be considered in many contexts (Shah et al., 2016). Built on these univariate models, multivariate time-series analysis serves to evaluate the relationships among these series. In addition to basic cross-correlation measures used in the chapter’s case study, multivariate methods including Prais– Winsten regression, Granger causality tests, and Vector Autoregression are useful to estimate leading, lagging and feedback relationships among multiple time series (Wells et al., 2019). In a multi-media environment, time-series analysis is particularly helpful to improve our understanding of how political discourse arises and develops across platforms. Natural language processing (NLP) has been applied widely by social scientists to understand the linguistic features of textual metadata for many years. At the most basic, tools like Linguistic Inquiry and Word Count (LIWC) (Tausczik and Pennebaker, 2010) provide metrics based on preexisting dictionaries representing concepts. Given these dictionaries have been developed in different languages, this tool permits comparative work on emotions, attentional focus, social relationships and other topics (Dudău and Sava, 2021). We identified high-frequency keywords across online platforms using the R package called Quanteda (https://​quanteda​.io/) and KoNLP (Korean Natural Language Processing package). As shown, generating word frequencies can provide insights about prominent linguistic features and points of emphasis. Meanwhile, we used STM, a semi-automated machine learning approach, to generate prevalent topics based on word co-occurrence. STM helps

130  Handbook of digital politics

produce the latent topical structure from large textual data and is useful for social scientists to categorize thematic and clustering features (see Ghosh et al., 2022). Scholars also use supervised machine learning to capture specific features in an accurate way. For example, researchers have used human-annotated content as a training set for machine learning input to track the volume and types of discourse following mass shootings, with a well-performing classifier applicable to large corpus of text (see Zhang et al., 2019). Supervised machine learning provides a more flexible and customizable way for scholars to attend to specific features within text data. Both unsupervised and supervised machine learning help social scientists to capture linguistic features, nuances, and topical structures of large text corpora. All the aforementioned linguistic analytical methods can be applied in multiple languages. Other techniques of analyzing textual data, such as language markers and part-of-speech tagging, have also been commonly applied by computational social science studies about political expression. Language markers, such as hashtags on social media platforms, are commonly used to measure online expressions. As discussed in Bode et al. (2015, pp. 149–150), hashtags afforded by social media platforms facilitate online expression and provide “discursive clusters around a shared interest”. It was argued that the hashtags are representative of the substantive content of the tweets, thus, studies on social media expression have used hashtags to study political networks and discourses streams (Zhang et al., 2019). By cross validating the results from language markers from hashtags and the supervised machine learning results, Zhang et al. (2019) demonstrated a high consistency between these two computational measures on Twitter expression about mass shootings. A part-of-speech approach is also a helpful technique to identify specific meanings within textual data, and it was widely used to detect verb tenses, pronouns, and identity languages in online political expression (Suk et al., 2021). These methods provide a range of powerful tools to analyze political expression and can be extended to include advanced syntactical coding and multilingual natural language processing (e.g., Lopez et al., 2019; Maier et al., 2022). The application of unsupervised and supervised machine learning techniques has great potential to contribute to our understanding of political talk by emphasizing the statistical co-occurrence of words or phrases on a large scale (Zhang et al., 2019; Kligler-Vilenchik et al., 2020; Suk et al., 2021). Such language processing approaches allow for the coding of a vast amount of data with high consistency and replicability. Through detection of the structure and content of online political discourse, these methods can provide diverse insights into an increasingly complicated media environment and globalized world. Although not featured in this chapter, the potential to extend this work on online political expression through more comprehensive multi-modal analysis holds tremendous promise. A large portion of current social media data comes in the form of image, audio, and video. Two of the most prominent platforms considered above, Instagram and YouTube, center on these formats, and yet so much current computational social science work, especially work on political expression, overlooks sounds

Computational approaches to online political expression  131

and images. Future research should use computational tools to analyze visual messages and their relationships with text and audio in a multi-modal fashion. Williams et al. (2020) provide a roadmap for social scientists to apply computational methods to study images. For instance, visual data could be used to detect protest activity, exhibited emotions, and degree of violence (Won et al., 2017). Such techniques could be further applied to political communication, especially to the study of political elites and their expressive behaviors (Sun et al., 2020). As a framework for research that can consider overtime and cross-platform differences, explore linguistic patterns, and synthesize themes and topics, multi-modal analysis that considers the full expressive potential of social media is needed as a next step for computational work.

NOTES 1. According to similarweb, Nate and DC Inside are the top two popular websites in the “Other Arts and Entertainment” category in South Korea (see https://​www​.similarweb​ .com/​website/​dcinside​.com/​#overview and https://​www​.similarweb​.com/​website/​nate​ .com/​#overview). Since Nate is a search engine, Korean counterpart of Google, we select Nate Pann (Nate’s forum) rather than Nate, along with DC Inside for representing publicly available posts on mainstream forums. We combine these two Korean forums and compare the Korean forum discourses with online discussions on Reddit for the US #MeToo case. 2. To prepare for cross-correlation analysis, we applied data imputation techniques via R package imputeTS for missing data across platforms. 3. Deodeumeo manjindang is a slur to deprecate the Democrats because the accused Ahn Hee-jung is from the Democratic party.

FURTHER READING Time-Series Analysis Box-Steffensmeier, J. M., Freeman, J. R., Hitt, M. P., and Pevehouse, J. C. (2014). Time Series Analysis for the Social Sciences. Cambridge: Cambridge University Press. Shah, D. V., Hanna, A., Bucy, E. P., Lassen, D. S., Van Thomme, J., Bialik, K., … Pevehouse, J. C. (2016). Dual screening during presidential debates: Political nonverbals and the volume and valence of online expression. American Behavioral Scientist, 60(14), 1816–1843. Wells, C., Shah, D. V., Pevehouse, J. C., Foley, J., Lukito, J., Pelled, A., and Yang, J. (2019). The temporal turn in communication research: Time series analyses using computational approaches. International Journal of Communication, 13, 4021–4043. Zhang, Y., Shah, D., Pevehouse, J., and Valenzuela, S. (2022). Reactive and asymmetric communication flows: Social media discourse and partisan news framing in the wake of mass shootings. The International Journal of Press/Politics. https://​doi​.org/​10​.1177/​ 19401612211072793.

132  Handbook of digital politics

Linguistic Analyses Dudău, D. P. and Sava, F. A. (2021). Performing multilingual analysis with linguistic inquiry and word Count 2015 (LIWC2015): An equivalence study of four languages. Frontiers in Psychology, 12, 2860. Ghosh, S., Su, M.-H., Abhishek, A., Suk, J., Tong, C., Kamath, K., Hills, O., Correa, T., Garlough, C., Borah, P., and Shah, D. (2022). Covering #MeToo across the news spectrum: Political accusation and public events as drivers of press attention. The International Journal of Press/Politics, 27(1), 158–185. Tausczik, Y. R. and Pennebaker, J. W. (2010). The psychological meaning of words: LIWC and computerized text analysis methods. Journal of Language and Social Psychology, 29(1), 24–54. Zhang, Y., Shah, D., Foley, J., Abhishek, A., Lukito, J., Suk, J., Kim, S. J., Sun, Z., Pevehouse, J., and Garlough, C. (2019). Whose lives matter? Mass shootings and social media discourses of sympathy and policy, 2012–2014. Journal of Computer-Mediated Communication, 24(4), 182–202.

Other Computational Methods Suk, J., Abhishek, A., Zhang, Y., Ahn, S. Y., Correa, T., Garlough, C., and Shah, D. V. (2021). #MeToo, networked acknowledgment, and connective action: How “empowerment through empathy” launched a social movement. Social Science Computer Review, 39(2), 276–294. Sun, Z., Sarma, P., Sethares, W., and Liang, Y. (2020). Learning relationships between text, audio, and video via deep canonical correlation for multimodal language analysis. Proceedings of the AAAI Conference on Artificial Intelligence (Vol. 34, No. 05, pp. 8992–8999). Williams, N. W., Casas, A., and Wilkerson, J. D. (2020). Images as Data for Social Science Research: An Introduction to Convolutional Neural Nets for Image Classification. Cambridge: Cambridge University Press.

REFERENCES Abramowitz, A. I. and Saunders, K. L. (2008). Is polarization a myth? The Journal of Politics, 70(2), 542–555. Bafumi, J. and Shapiro, R. Y. (2009). A new partisan voter. The Journal of Politics, 71(1), 1–24. Barber, B. (1984). Strong Democracy: Participatory Politics for a New Age. Berkeley, CA: University of California Press. Bennett, W. L. and Iyengar, S. (2008). A new era of minimal effects? The changing foundations of political communication. Journal of Communication, 58(4), 707–731. Berelson, B. R., Lazarsfeld, P. F., and McPhee, W. N. (1954). Voting: A Study of Opinion Formation in a Presidential Campaign. Chicago, IL: University of Chicago Press. Bode, L., Hanna, A., Yang, J., and Shah, D. V. (2015). Candidate networks, citizen clusters, and political expression: Strategic hashtag use in the 2010 midterms. The Annals of the American Academy of Political and Social Science, 659(1), 149–165. Bode, L., Vraga, E. K., Borah, P., and Shah, D. V. (2014). A new space for political behavior: Political social networking and its democratic consequences. Journal of Computer-Mediated Communication, 19(3), 414–429.

Computational approaches to online political expression  133

Bossetta, M. (2018). The digital architectures of social media: Comparing political campaigning on Facebook, Twitter, Instagram, and Snapchat in the 2016 US election. Journalism and Mass Communication Quarterly, 95(2), 471–496. Bucy, E. P., Foley, J. M., Lukito, J., Doroshenko, L., Shah, D. V., Pevehouse, J. C., and Wells, C. (2020). Performing populism: Trump’s transgressive debate style and the dynamics of Twitter response. New Media & Society, 22(4), 634–658. Clark, T. (ed.) (1969). Gabriel Tarde on Communication and Social Influence: Selected Papers. Chicago, IL: University of Chicago Press. Dudău, D. P. and Sava, F. A. (2021). Performing multilingual analysis with linguistic inquiry and word Count 2015 (LIWC2015): An equivalence study of four languages. Frontiers in Psychology, 12, 2860. Ekström, M. and Östman, J. (2013). Family talk, peer talk and young people’s civic orientation. European Journal of Communication, 28(3), 294–308. Freelon, D. G. (2010). Analyzing online political discussion using three models of democratic communication. New Media & Society, 12(7), 1172–1190. Freelon, D., McIlwain, C., and Clark, M. (2018). Quantifying the power and consequences of social media protest. New Media & Society, 20(3), 990–1011. Friedland, L. A., Hove, T., and Rojas, H. (2006). The networked public sphere. Javnost – The Public, 13(4), 5–26. Ghosh, S., Su, M.-H., Abhishek, A., Suk, J., Tong, C., Kamath, K., Hills, O., Correa, T., Garlough, C., Borah, P., and Shah, D. (2022). Covering #MeToo across the news spectrum: Political accusation and public events as drivers of press attention. The International Journal of Press/Politics, 27(1), 158–185. Giglietto, F., Rossi, L., and Bennato, D. (2012). The open laboratory: Limits and possibilities of using Facebook, Twitter, and YouTube as a research data source. Journal of Technology in Human Services, 30(3–4), 145–159. Guo, L., Vu, H. T., and McCombs, M. (2012). An expanded perspective on agenda-setting effects. Exploring the third level of agenda setting. Revista de comunicación, 11(1), 51–68. Habermas, J. (1985). The Theory of Communicative Action: Vol. 1. Reason and the Rationalization of Society, trans. T. McCarthy. Boston, MA: Beacon. Han, J. Y., Shah, D. V., Kim, E., Namkoong, K., Lee, S. Y., Moon, T. J., … Gustafson, D. H. (2011). Empathic exchanges in online cancer support groups: Distinguishing message expression and reception effects. Health Communication, 26(2), 185–197. Hanna, A., Wells, C., Maurer, P., Friedland, L., Shah, D., and Matthes, J. (2013). Partisan alignments and political polarization online: A computational approach to understanding the French and US presidential elections. Proceedings of the 2nd Workshop on Politics, Elections and Data (pp. 15–22). Holbert, R. L., Garrett, R. K., and Gleason, L. S. (2010). A new era of minimal effects? A response to Bennett and Iyengar. Journal of Communication, 60(1), 15–34. Jiang, X., Su, M. H., Hwang, J., Lian, R., Brauer, M., Kim, S., and Shah, D. (2021). Polarization over vaccination: Ideological differences in Twitter expression about COVID-19 vaccine favorability and specific hesitancy concerns. Social Media + Society, 7(3). Jones, B. D. and Baumgartner, F. R. (2005). The Politics of Attention: How Government Prioritizes Problems. Chicago: University of Chicago Press. Katz, E. (1992). On parenting a paradigm: Gabriel Tarde’s agenda for opinion and communication research. International Journal of Public Opinion Research, 4(1), 80–86. Katz, E. (2006). Rediscovering Gabriel Tarde. Political Communication, 23(3), 263–270. Katz, E. and Lazarsfeld, P. F. (1955). Personal Influence: The Part Played by People in the Flow of Mass Communications. Glencoe, IL: Free Press. Kim, J., Wyatt, R. O., and Katz, E. (1999). News, talk, opinion, participation: The part played by conversation in deliberative democracy. Political Communication, 16(4), 361–385.

134  Handbook of digital politics

Kinnunen, J. (1996). Gabriel Tarde as a founding father of innovation diffusion research. Acta Sociologica, 39(4), 431–442. Kligler-Vilenchik, N., Baden, C., and Yarchi, M. (2020). Interpretative polarization across platforms: How political disagreement develops over time on Facebook, Twitter, and WhatsApp. Social Media + Society, 6(3). Kraidy, M. M. (1999). The global, the local, and the hybrid: A native ethnography of glocalization. Critical Studies in Mass Communication, 16(4), 456–476. Latour, B. (2010). Tarde’s idea of quantification. In M. Candea (ed.), The Social After Gabriel Tarde: Debates and Assessments (pp. 187–202). London: Routledge. Lazarsfeld, P. F., Berelson, B., and Gaudet, H. (1944). The People’s Choice: How the Voter Makes Up His Mind in a Presidential Campaign. New York: Duell, Sloan & Pearce. Lazer, D., Pentland, A., Adamic, L., Aral, S., Barabasi, A. L., Brewer, D., … Van Alstyne, M. (2009). Social science. Computational social science. Science, 323(5915), 721–723. Lee, N. J., Shah, D. V., and McLeod, J. M. (2013). Processes of political socialization: A communication mediation approach to youth civic engagement. Communication Research, 40(5), 669–697. Ling, R. and Campbell, S. W. (2017). Mobile communication: Bringing us together and tearing us apart. In S. Campbell (ed.), Mobile Communication (pp. 11–26). London: Routledge. Lopez, I., Quillivic, R., Evans, H., and Arriaga, R. I. (2019). Denouncing sexual violence: A cross-language and cross-cultural analysis of #MeToo and #BalanceTonPorc. IFIP Conference on Human-Computer Interaction (pp. 733–743). Cham: Springer. Lukito, J. (2020). Coordinating a multi-platform disinformation campaign: Internet Research Agency activity on three US social media platforms, 2015 to 2017. Political Communication, 37(2), 238–255. Maier, D., Baden, C., Stoltenberg, D., De Vries-Kedem, M., and Waldherr, A. (2022). Machine translation vs. multilingual dictionaries assessing two strategies for the topic modeling of multilingual text collections. Communication Methods and Measures, 16(1), 19–38. McLeod, J. M., Scheufele, D. A., and Moy, P. (1999). Community, communication, and participation: The role of mass media and interpersonal discussion in local political participation. Political Communication, 16(3), 315–336. Mutz, D. C. (2006). Hearing the Other Side: Deliberative Versus Participatory Democracy. New York: Cambridge University Press. Namkoong, K., Shah, D. V., Han, J. Y., Kim, S. C., Yoo, W., Fan, D., … Gustafson, D. H. (2010). Expression and reception of treatment information in breast cancer support groups: How health self-efficacy moderates effects on emotional well-being. Patient Education and Counseling, 81, S41–S47. Papacharissi, Z. (2004). Democracy online: Civility, politeness, and the democratic potential of online political discussion groups. New Media & Society, 6(2), 259–283. Pingree, R. J. (2007). How messages affect their senders: A more general model of message effects and implications for deliberation. Communication Theory, 17(4), 439–461. Price, V. and Cappella, J. N. (2002). Online deliberation and its influence: The electronic dialogue project in campaign 2000. IT and Society, 1(1), 303–329. Rogers, E. M. and Shoemaker, F. F. (1971). Communication of Innovations: A Cross-Cultural Approach. New York: Free Press. Romero, L. (2021). Experts say echo chambers from apps like Parler and Gab contributed to attack on Capitol. ABC News. https://​abcnews​.go​.com/​US/​experts​-echo​-chambers​-apps​ -parler​-gab​-contributed​-attack/​story​?id​=​75141014. Rossini, P., Stromer-Galley, J., Baptista, E. A., and Veiga de Oliveira, V. (2021). Dysfunctional information sharing on WhatsApp and Facebook: The role of political talk, cross-cutting exposure and social corrections. New Media & Society, 23(8), 2430–2451. Schmitt-Beck, R. and Lup, O. (2013). Seeking the soul of democracy: A review of recent research into citizens’ political talk culture. Swiss Political Science Review, 19(4), 513–538.

Computational approaches to online political expression  135

Schudson, M. (1978). The ideal of conversation in the study of mass media. Communication Research, 5(3), 320–329. Shah, D. V. (2016). Conversation is the soul of democracy: Expression effects, communication mediation, and digital media. Communication and the Public, 1(1), 12–18. Shah, D. V., Cho, J., Eveland Jr, W. P., and Kwak, N. (2005). Information and expression in a digital age: Modeling Internet effects on civic participation. Communication Research, 32(5), 531–565. Shah, D. V., Cho, J., Nah, S., Gotlieb, M. R., Hwang, H., Lee, N. J., … McLeod, D. M. (2007). Campaign ads, online messaging, and participation: Extending the communication mediation model. Journal of Communication, 57(4), 676–703. Shah, D. V., Hanna, A., Bucy, E. P., Lassen, D. S., Van Thomme, J., Bialik, K., … Pevehouse, J. C. (2016). Dual screening during presidential debates: Political nonverbals and the volume and valence of online expression. American Behavioral Scientist, 60(14), 1816–1843. Shah, D. V., McLeod, D. M., Rojas, H., Cho, J., Wagner, M. W., and Friedland, L. A. (2017). Revising the communication mediation model for a new political communication ecology. Human Communication Research, 43(4), 491–504. Shah, D. V. and Scheufele, D. A. (2006). Explicating opinion leadership: Nonpolitical dispositions, information consumption, and civic participation. Political Communication, 23(1), 1–22. Suk, J., Abhishek, A., Zhang, Y., Ahn, S. Y., Correa, T., Garlough, C., and Shah, D. V. (2021). #MeToo, networked acknowledgment, and connective action: How “empowerment through empathy” launched a social movement. Social Science Computer Review, 39(2), 276–294. Sun, Z., Sarma, P., Sethares, W., and Liang, Y. (2020). Learning relationships between text, audio, and video via deep canonical correlation for multimodal language analysis. Proceedings of the AAAI Conference on Artificial Intelligence (Vol. 34, No. 05, pp. 8992–8999). Tarde, G. (1903 [1890]). The Laws of Imitation, trans. E. C. Parsons. New York: Henry Holt. Tarde, G. (1969 [1898]). Gabriel Tarde on Communication and Social Influence: Selected Papers, Vol. 334, ed. T. Clark. Chicago, IL: University of Chicago Press. Tausczik, Y. R. and Pennebaker, J. W. (2010). The psychological meaning of words: LIWC and computerized text analysis methods. Journal of Language and Social Psychology, 29(1), 24–54. Walsh, K. C. (2012). Putting inequality in its place: Rural consciousness and the power of perspective. American Political Science Review, 106(3), 517–532. Wells, C., Shah, D. V., Pevehouse, J. C., Foley, J., Lukito, J., Pelled, A., and Yang, J. (2019). The temporal turn in communication research: Time series analyses using computational approaches. International Journal of Communication, 13, 4021–4043. Wells, C., Shah, D. V., Pevehouse, J. C., Yang, J., Pelled, A., Boehm, F., … Schmidt, J. L. (2016). How Trump drove coverage to the nomination: Hybrid media campaigning. Political Communication, 33(4), 669–676. Williams, N. W., Casas, A., and Wilkerson, J. D. (2020). Images as Data for Social Science Research: An Introduction to Convolutional Neural Nets for Image Classification. Cambridge: Cambridge University Press. Won, D., Steinert-Threlkeld, Z. C., and Joo, J. (2017). Protest activity detection and perceived violence estimation from social media images. Proceedings of the 25th ACM International Conference on Multimedia (pp. 786–794). Zhang, Y., Shah, D., Foley, J., Abhishek, A., Lukito, J., Suk, J., … Garlough, C. (2019). Whose lives matter? Mass shootings and social media discourses of sympathy and policy, 2012–2014. Journal of Computer-Mediated Communication, 24(4), 182–202. Zhang, Y., Shah, D., Pevehouse, J., and Valenzuela, S. (2022). Reactive and asymmetric communication flows: Social media discourse and partisan news framing in the wake

136  Handbook of digital politics

of mass shootings. The International Journal of Press/Politics. https://​doi​.org/​10​.1177/​ 19401612211072793. Zhang, Y., Wells, C., Wang, S. and Rohe, K. (2018). Attention and amplification in the hybrid media system: The composition and activity of Donald Trump’s Twitter following during the 2016 presidential election. New Media & Society, 20(9), 3161–3182.

9. Creating spaces for online deliberation Christopher Birchall and Stephen Coleman

WHY ONLINE DELIBERATION? Contemporary political democracy is not short of political speech online, but the nature of the speech often presents two formidable challenges that have long been familiar in the public sphere. Firstly, there is the problem of under-informed, unconfident citizens who find it difficult to make up their minds on many of the important policy issues that face society. They rarely talk about politics because they think that nobody in authority will take any notice of them – and they are seldom listened to because they rarely talk about politics. We could compel such people to vote on issues, regardless of whether they feel able to form a competent judgement; we can offer them opportunities to follow parties and leaders who serve as containers of composite values and preferences (though in this way we may leave them vulnerable to the influence of the recently emerged forces of misinformation – for example, see Freelon and Wells, 2020; Jerit and Zhao, 2020; O’Connor and Weatherall, 2019); or we might leave them to disengage from politics, allowing those who feel confident that they are well-informed to make decisions for them. While such minimal terms of political engagement would be compatible with a highly parsimonious model of democracy, they would fall short of the norms of citizenship as formulated by participatory democrats. Secondly, there is the problem of dogmatic and inflexible citizens who have made up their minds on nearly all issues, often in accordance with an overarching ideological bias, and are open to neither new information nor ethical influence to change their rigidly-held values and preferences. Such people satisfy the normative democratic requirement of being willing to enter the political fray, but the quality of their engagement tends to be inconsistent with the democratic principle of intellectual openness and adaptability. In recent years as political discussion has become increasingly polarized (Pfetsch, 2018) and at the same time more noticeably tied to civic actions and behavioural choices – such as willingness to wear masks during a pandemic, or to accept vaccinations or movement restrictions – it has become even more important to advance rational public discussion to create informed publics. Neither citizens who can’t make up their minds nor citizens who have finally and forever made up their minds are ideal inhabitants of a healthy democracy. Arguments and practical proposals for democratic deliberation respond to both of these challenges. Including the least confident or vocal members of society in something approaching a public conversation, while encouraging the permanently certain to encounter a wider range of perspectives and information, can only be good for democratic politics. Public deliberation fills a conspicuous vacuum in the public 137

138  Handbook of digital politics

sphere in which self-referential political and media elites have often seemed to crowd out the voices of the citizenry. The principles of deliberation are well known: all propositions should be on the table for inclusive and uncensored discussion; arguments for and against must be open to public scrutiny; those who deliberate must be regarded as equals (at least, in the context of the deliberative moment) and must listen with attention and respect to all arguments, evidence and experiential narratives; and, ideally at least, deliberative judgements should be based on the force of the strongest argument rather than narrow interests, blind commitments or appeals to external authority (Dryzek, 2000; Gastil, 2000; Habermas, 1994; Steiner, 2012). There are several other conditions that theorists might want to add to the list of deliberative requirements, with some setting the bar so high that it sometimes seems as if deliberation could only ever work in small-scale, experimental environments. Other scholars argue that even if full-blown deliberative democracy is too ambitious an objective, the creation of a more deliberative democracy (Coleman and Blumler, 2009) would at least be preferable to the current situation in which the diverse testimonies of civic experience are drowned out by the relentless outpouring of sensational media headlines. The case for democratic deliberation, in contrast to the mainly aggregative forms of decision-making associated with voting and mass parties, has gained momentum in recent decades, partly in response to the two challenges discussed above and partly because democratic legitimacy in a more culturally egalitarian era is ever more dependent upon the strength of communicative relationships between government and governed. Governments, parliaments, local authorities and parties, as suppliers of proposed solutions to social problems, can no longer depend upon popular deference, but are under increasing pressure to acknowledge the experience and expertise that lies beyond them, often within local neighbourhoods or communities of practice. Such inputs cannot be collected through the ballot box, which is a crude mechanism for capturing the rationale and multidimensionality of the public will (Coleman, 2013). These public institutions also need to compete within a public sphere that also now includes powerful actors such as influencers, campaign groups and other opinion leaders, and thus must engage with the public debate in order to exert any control (Miller and Vaccari, 2020). By enabling lots of different kinds of people to have the space and confidence to form, rehearse and articulate their views, and encouraging people to develop hitherto incomplete or inconsistent arguments, deliberation at its best helps people to acknowledge the political reality that it is sometimes politically preferable to engage in effective compromise than to remain isolated and impotent. By inviting citizens to account for their views rather than simply counting their bundled preferences, deliberation may be able to challenge some of the polarization of public discussion, and democratic outcomes might be more likely to reflect the values and experiences of citizens, stand a chance of being implemented with public support and be regarded as fair. However, establishing spaces, processes and cultural habits that are likely to result in meaningful, inclusive and consequential deliberation has proved to be a difficult challenge. Most citizens know where to go to vote when elections come around

Creating spaces for online deliberation  139

– and many have at least a clue about where to go to complain when elected representatives let them down in between elections. But where do citizens go to deliberate about the issues, policies and global forces that affect them? Deliberation has tended to invoke images of market squares, coffee houses and modern community centres, buzzing with civic dialogue; but how might these romantically quaint metaphors of deliberative space be reinvented as twenty-first century arenas of democratic talk? For some democratic theorists, the emergence of the Internet offered a potential solution to this problem. From the outset of the World Wide Web as a public network in the mid-1990s, theorists in search of contemporary space for deliberation and online enthusiasts in search of a democratizing role for the Internet gravitated towards visions of e-democracy: the potential of online space as an environment for a new kind of more inclusive and deliberative political practice. Millions of conversations and interactions of various kinds are going on all the time within online spaces that are now a routine domain of everyday interaction for a vast proportion – though not all – of the global population. But deliberative spaces do not form themselves. While they sometimes develop unexpectedly – see Graham et al. (2016), for example – they are nonetheless the consequence of intentionality and design in the development of participatory spaces. This chapter focuses on the problematics of designing space for online deliberation. Our aim is to consider what has been learned from research about the ways in which tools, protocols, structures and interfaces affect the quality of democratic deliberation. We then turn to the implications of these factors for future research regarding the promotion and evaluation of online deliberation.

PRINCIPLES OF DELIBERATIVE QUALITY There is a theoretical distinction to be made between political deliberation, which seeks to encapsulate the benefits of focused, purposeful and honest talk, and everyday talk about politics, which is often fragmented, purposeless, uninformed and unequal. Whereas the latter ‘is not always self-conscious, reflective or considered’ (Mansbridge, 1999, p. 211), the quality of deliberative practice lies in its commitment to a process of shared reflection that eschews mere competitive self-interest and embedded injustice. In reality, the theoretical distinction between deliberation and everyday talk is less obvious; there can be greater or lesser degrees of the former within the latter. Several commentators have observed that what passes for political debate online tends to be far from deliberative; that most online political exchanges seem to be partisan, prejudiced and uncivil; and that this raises significant doubts about the potential relationship between the Internet and more deliberative democracy (Hill and Hughes, 1999; Jerit and Zhao, 2020; Morozov, 2012; Pfetsch, 2018; Wilhelm, 2000). A weakness of these studies is that they have tended to be based upon limited cases, such as fora in which members of the same party gather together to reinforce their collective values or random exchanges between friends on social media sites. To dismiss arguments for online deliberation on the grounds that most online politi-

140  Handbook of digital politics

cal talk is shallow, angry or uninformed is to miss the point of trying to design spaces that attempt to reduce the anti-deliberative influences of conversational homophily and group herding. The case for online deliberation rests on the assumption that it is a means of enhancing the quality of public debate and that such enhancement is unlikely to happen without well-planned design. However, there is a temptation for scholars to ‘discover’ online deliberation by adopting the circular perspective that deliberation only occurs when people talk to one another in ways anticipated and facilitated by deliberative theorists. As Coleman and Moss (2012, p. 5) have argued, Most researchers … continue to speak and write as if deliberation and the capacities it presupposes are naturally occurring and universal rather than constructed and contingent. Holding on to an essentialist conception of liberal citizenship, they fail to consider the extent to which the deliberative citizen is ‘formed and normed’, in Ivison’s (1997: 41) evocative phrase, and to which they contribute to the construction of the object of their own research.

Rather than thinking of deliberation as an objective or formulaic practice in which one kind of technical platform can serve the needs of all citizens and all of the vast range of subjects they might want to discuss, it makes sense to acknowledge that different social groups behave differently in varying online spaces. Several important studies have identified determinants of online deliberative behaviour that preclude essentialism and recognize that there is no single way to realize the quality of deliberative outcomes (Dahlgren, 2005; Freelon, 2010; Pickard, 2008; Wright et al., 2020). A first key factor determining deliberative outcomes, on or offline, is that most people prefer to talk to other people when they feel secure and comfortable rather than intimidated or under pressure. This accounts for the well-established finding that in both offline and online contexts people discuss politics with likeminded people and feel more comfortable in environments where their points of view and modes of expression are unlikely to be fundamentally challenged (McPherson et al., 2001; Nahon and Hemsley, 2014). The attraction of homophilic political communication presents a challenge to democracy, as the most likely effect of exchanging ideas with people who share one’s views is to make such beliefs seem obviously right and to distance and marginalize alternative perspectives (Sunstein, 2002). A key mark of deliberative quality is the extent to which people find themselves in situations where they are compelled to justify their values and preferences; where, indeed, they might come to question or even change their original positions. Self-questioning and preference-shifting are strong empirical effects of high-quality deliberation. Of course, questioning one’s opinions can be uncomfortable and all too often deliberative quality is realized at the expense of decreased participation in politics (Mutz, 2006). A well-designed deliberative online environment would allow people to feel safe in disclosing their views to strangers, while exposing them to perspectives that they would not usually encounter. As with the design of any public space, the aim should be to expose participants to the worldliness of politics without crushing personal dispositions. In the case of online deliberation, this entails an effective

Creating spaces for online deliberation  141

balance between the normative requirements of rational-critical interaction and the social practices and customs that people adopt as part of their personal performance of citizenship. In this regard Freelon’s (2010) framework for exploring the ways in which distinctive ‘democratic styles’ lead people to deliberate in different ways provides a useful way of thinking about the pluralistic design of deliberative space. As he puts it, ‘Rather than simply analyzing online forums in terms of the extent to which they adhere to a singular set of deliberative standards, scholars [should] bring to bear on their data an understanding that different kinds of public spheres exist’. Freelon argues that people come to public discussion with various ideas about what it means to perform as citizens. Liberal-individualists, he argues, are mainly interested in self-expression and self-actualization, while communitarians are mainly interested in strengthening collective ties and classic deliberators are motivated by a search for the best argument. According to Freelon, both liberal-individualist and communitarian modes of discussion can incorporate elements of deliberation, but this calls for careful design to make it happen. That is to say, even in the absence of citizens who meet the normative requirements of fully-fledged deliberators, the design of discursive environments can encourage degrees of deliberative outcomes. Taking this insight into account, designers of spaces for online deliberative talk might aim to create interfaces and protocols that allow discussants to pursue their own ‘democratic styles’, while being gently encouraged to interact with others committed to different styles. The important point here is that designers should acknowledge the nuances of cultural practice and expressive habit that frame deliberative interaction rather than expecting such habits and practices to bend to the rigours of deliberative theory. A second factor likely to affect deliberative quality is the subject matter being discussed. While citizens may be more willing to participate in deliberation than is often thought (Neblo et al., 2010), some political topics are likely to arouse passions more than others (Coe et al., 2014). Karlsson (2010) analysed 28 online discussion forums, each sharing the same platform design, but in which contributors discussed different topics related to EU policy. Significant variation was observed in levels of deliberative participation per visitor between the respective forums, suggesting that different discussion topics may make people more or less likely to participate in online deliberation. Just as citizens are often more likely to deliberate when in a comfortable environment, one might assume that they are more likely to deliberate about topics that make them feel safe, informed and relatively invulnerable to hostile feedback. Interestingly, Karlsson’s study found that forums with the highest proportions of deliberative content were the ones that generated the most user engagement. Indeed, his conclusion that ‘deliberation is more likely to be successful if the issue of deliberation is surrounded by a high level of engagement and conflicted opinions rather than being an issue that renders participants indifferent or is surrounded by a high level of consensus regarding the topics under investigation’ is very promising from a democratic perspective. It suggests that contributors are more likely to put in the effort required for deliberation (as opposed to ranting) when they are exposed to a subject that they find not only engaging, but intellectually challenging. Perhaps, then, an important requirement of a deliberative system is that it make topics

142  Handbook of digital politics

attractive and challenging to participants, particularly when they are outside of the target participants’ usual areas of interest or comfort. Indeed, it might be that taking people beyond their ideological comfort zones is more likely to trigger deliberative activity than pandering to an imagined popular desire to avoid agonistic contestation. However, there are other reasons that participants may seek to keep conversations civil and rational, such as maintaining the accord that makes participants comfortable in the community in which the conversation is happening. Social ties have been shown to be of importance in maintaining productive deliberation where a conversational space is able to allow participants to go beyond ideological comfort zones, stray into potential conflict, but use freedoms in the design of the space to conciliate and rebuild relationships through off-topic interactions, before re-joining the debate (Birchall, 2018). Of course, it is not only the willingness of contributors to participate that matters, but also their ability to do so effectively. Designing spaces for online deliberation that compensate for structural inequalities offline (such as class, gender or ethnic inequality) can sometimes result in greater equality of voice between discussion participants. Monnoyer-Smith (2012, p. 203) describes how online spaces can be designed in a fashion ‘that welcomes women, the less informed, and the socioculturally deprived’, thereby restructuring, but not eliminating, some of the unjust power structures that might be expected to prevail in the offline world. This potential is less often realized than not, however, with social inequalities, gendered behaviours and learning preferences often being overlooked in online deliberative space design (see Shortall et al., 2021, for an interesting study of this). In short, design could be used better to help people to discuss a diverse range of sometimes complex or sensitive subjects as well as in broadening the range of voices taking part. A third factor likely to influence deliberative quality is the relationship (actual and perceived) between spaces in which people are invited to deliberate and institutions of power that are likely to be making decisions related to what is being discussed. A deliberative space discussing a proposed national policy might have clear links to the government, parliament or political party that has proposed it. If such institutions are involved in the discussion as sponsors, participants or respondents, this could have either positive or negative impacts upon deliberative outcomes. If participants’ trust in the institution is high – if they believe that it is really listening to what they have to say, are minded to take their views and experiences into account before making a policy decision and are genuinely willing to offer honest feedback – this may well enhance the quality of deliberation. After all, people are more likely to engage in the hard work of deliberating if they believe that their efforts will have real-world consequences. Alternatively, if a governmental, legislative or corporate institution is deemed untrustworthy and people believe that a deliberative exercise is merely tokenistic or, worse still, an exercise in surveillance or data-gathering, this would surely diminish deliberative quality. In such circumstances, participants might decide to use the occasion to merely reaffirm their original positions or voice their scepticism towards the process. Some forms of online public deliberation are intentionally autonomous, refusing to be connected to any dominant political interest,

Creating spaces for online deliberation  143

especially government. These tend to entail lateral exchange of views between citizens, either for mainly epistemic ends or as a prelude to civic mobilization. Studies of discussions in online ‘third spaces’ – defined as ‘online discussion spaces with a primarily non-political focus, but where political talk emerges within conversations’ (Graham et al., 2015, p. 651) – have found that such venues enable people to rehearse their own identities and encounter (often inadvertently) other perspectives and values (Graham et al., 2016; Graham and Wright, 2014; Wright, 2012). These spaces may provide a crucial foundation for democratic deliberation. Indeed, where participants seek change, there is evidence to suggest that peer-to-peer policy deliberation is often not regarded by participants as ‘mere talk’, but as a means of shaping policy by influencing public opinion, which in turn will put pressure upon elite decision-makers (Coleman et al., 2011, show how online protesters against the Iraq war had much more confidence in their capacity to influence fellow citizens than government per se; Graham et al., 2015, 2016 highlight the value of political talk online in fomenting political action). In this sense, effective deliberation in third spaces may be a valuable entry point to the informal political sphere. Within such informal contexts people learn to develop the quality of their arguments and gain the confidence to take more institutionally related collective action when necessary. Taking these three factors into consideration can help deliberative practitioners to design spaces and interfaces that reflect the structural features of normatively effective deliberation. While some features of online deliberative quality call for the replication of offline practices that have proven to be effective, other features are distinctive to the online context. Offline deliberative theory and practice may not be directly applicable to online environments and so designers should ‘strive to take advantage of the unique design flexibility of the online discussion environment’ (Pingree, 2009, p. 309). De Cindio (2012) urges designers of online deliberative spaces to consider three key factors: the social grouping who are expected to deliberate (which she calls the gemeinschaft dimension); the social contract between developers, administrators and contributors (the gesellschaft dimension); and the technologies to be used in consolidating these relationships. Much research literature on online deliberation has focused upon the first of these considerations: who deliberates and how their preferences change or stay the same. Below we focus upon the other two considerations – developing the appropriate technological functionalities to facilitate deliberation and devising rules and moderation structures that are most likely to generate productive deliberative outcomes.

DELIBERATIVE DESIGN: SOME TECHNICAL CONSIDERATIONS A multitude of niches exist online in which conversation occurs with a greater or lesser degree of deliberative quality. Some attract user groups whose views are partisan; others attract participants whose views are more reflexive, reciprocal and cross-cutting. Some harbour highly deliberative political discussions almost by

144  Handbook of digital politics

accident (Graham, 2012), while others generate deliberative content despite the design of the space. Occasionally, elements of deliberation emerge amongst the character-limited conversations on Twitter (Thimm et al., 2014; Upadhyay, 2014), while other sites devote considerable resources to the design of tools to facilitate public debate, but fail utterly (the UK government’s ‘Spending Challenge’ from 2010 is a case in point). Some deliberative success stories result from participants feeling safe and at home within a community. Other online sites, such as many of those established for official policy consultations, aim to attract politically disengaged citizens to specially designed spaces, outside of the familiar environments in which they might usually express themselves. We consider below five technical factors that have been identified by online deliberation researchers as being significant for effective design. Engendering Substantive Debate Creating the right environment for online deliberation to take place entails something of a balancing act. On the one hand, motivating people to participate in political talk with strangers often involves appealing to their passions; on the other hand, ensuring that debate is constructive often entails suppressing those same passions and encouraging some degree of dispassionate rationality. Scholars have given considerable thought to ways of engendering such a balance (Barton, 2005; Coleman and Moss, 2012; Friess and Eilders, 2015; Schlosberg et al., 2007). Here we discuss the extent to which designs for online deliberation have addressed the need to balance participant commitment and the informational foundations of thoughtful interaction. Information provision, and mechanisms to ensure that participants utilize information, have long been a common feature of deliberative online spaces. The real-time discussion tool Unchat, created by Noveck for small-group deliberation, featured ‘speed bumps’, designed to force users to encounter relevant information prior to participating in debate. Transcripts were provided to help latecomers to ‘catch up’ with previous discussion. Similarly, the Deme interface (Davies et al., 2009) attempted to foster informed debate by providing access to relevant background information as well as features to enhance participant collaboration, including document-centred discussion and the sharing of files and links. The Deliberative Community Networks (OpenDCN) project (De Cindio, 2012) built on these and other previous projects by including an ‘informed discussion’ tool that allows participants to upload their own background information in a wide array of formats. Participants used built-in templates to supply their own datasets or links to external datasets. In this way, they were able to offer their own interpretations of evidence, thereby transcending the rather artificial distinction between background information and deliberative practice. Implicit here is the principle of generating reciprocal interactions amongst participants, removing barriers between agenda-setting initiators of deliberative exercises and deliberative publics.

Creating spaces for online deliberation  145

Real-Time or Asynchronicity Some advocates of online deliberation claim that carefully designed interfaces for synchronous conversation can replicate the vivacity of face-to-face interactions. For example, Noveck’s Unchat gathered dispersed people together online, as if they were in a single place at the same time. Today, messaging apps and mobile notifications bring near-synchronous communication into the mainstream (Colom, 2021), but full-quorum participation at the same time is rarely guaranteed without careful organization. Synchronous conversations are difficult to schedule for large numbers of participants, so may need to be constrained by rules limiting group size and contribution frequency (Cavalier et al., 2009; Tucey, 2010). Such an approach sacrifices inclusive spontaneity for the sake of deliberative quality. Other scholars argue that asynchronous deliberation makes it more convenient for people to participate on their own terms and leads to more reflective outcomes because users have more time to think before committing themselves to a position. Asynchronous conversation typical of online spaces can help participants to join in when they can, helping to improve reciprocity (de Brasi and Gutierrez, 2020). Until recently, the majority of deliberative tools and models were asynchronous, but these give rise to their own particular challenges. Entering into a large-scale asynchronous discussion that has already started presents users with a need to process, understand and organize the content that has emerged before they arrived. In the case of a large-scale discussion comprising thousands of threads and messages, this can prove to be a time-consuming challenge. As Pingree (2009, p. 310) puts it, ‘The Problem of Scale manifests as a difficulty in keeping up with all messages being sent’ while the ‘Problem of Memory and Mental Organization’ arises from the limitations of human memory in assimilating argumentative material. Designers have sought to alleviate these problems by designing interface features that diminish the disadvantages faced by latecomers to a discussion. For example, OpenDCN seeks to optimize interactivity between participants by organizing content in such a way that specific individuals and arguments can be easily located within the overall discussion. Nested posts and replies help participants to visualize arguments, identify authors and find appropriate locations for their own contributions. Social rating features, such as ‘likes’ and ‘recommends’ organize the content further and open the door to efforts to automate insight and knowledge from deliberations to report to policy makers. Such features help participants to place themselves within debates (Spiliotopoulou and Charalabidis, 2015), but can at times run counter to the principle of deliberation which expects everyone to be open to all arguments. By allowing users to rate the most popular comments, they are failing to reflect the quality of reasoning behind particular contributions, thereby shifting debate to the surface level of existing preferences (Buckingham Shum et al., 2014).

146  Handbook of digital politics

Visualizing the Arguments The challenge of levelling the point of entry to deliberation, so that all participants are exposed not only to background information and each other’s positions, but the core questions motivating the debate, is particularly necessary in the case of policy-related public deliberation, where it is of paramount importance that all contributors acknowledge a common agenda (Coleman and Blumler, 2009). Macintosh (2008) argued that more complex discussion platforms are necessary to facilitate ‘access to and analysis of factual information’, ‘preference formation’, and ‘community building’ – systems that generate and present community knowledge as well as just information. Many deliberative theorists, turned to argument visualization (AV) systems to provide not just spaces for people to pursue arguments, but a way of making visible the flow of argumentation through graphical representations depicting the collision and convergence of arguments (King, 2018; Klein, 2015). AV’s roots are in electronic collaborative theory which dates back over fifty years to the creation of systems designed to support legal and political decision-making (Conklin and Begeman, 1987; Kunz and Rittel, 1970). Expanding upon the Issue-Based Information System (IBIS) of Kunz and Rittel, AV formally structures conversations, the flows and components of which are used to create ‘maps’ of the arguments and evidence. This allows users to locate places within the debate where they feel that they can add value. Examples of AV include Pingree’s Decision Structured Deliberation system (DSD), the Deliberatorium from MIT (Klein, 2011) and later projects (see Buckingham Shum et al., 2014, for more) which have advanced AV by utilizing Web 2.0 features, such as user profiles, ratings and filtering. The Deliberatorium provides participants with a personal homepage, which includes watchlists to help them to keep up with conversations that might be of particular interest to them. Such systems are yet to have a widespread impact on the norms of online participation, though examples such as DebateGraph have been used in a number of governmental and third-sector-initiated deliberative consultations and may well in the future become useful facilitators of deliberative consultation (Iandoli et al., 2012, King, 2018). Moderating the Discussion Designing for online deliberation is not simply a matter of coming up with ever more sophisticated technical tools. Some qualities of deliberation depend upon more basic communicative interventions, such as moderation and facilitation. Wright and Street (2007) found that the social contract between contributors and administrators is a vital dimension to the success of deliberative spaces (see also Coleman and Gøtze, 2001; De Cindio, 2012; Noveck, 2003, 2010; Wright, 2006, 2009). The ways in which rules and protocols of a discussion space are maintained, contributors encouraged to interact and discussion outcomes are encouraged can make the difference between friendly, sharing interaction and a breakdown in trust and civility. There is now considerable research evidence to suggest that open and uncontrolled discussion between large groups of people who do not know one another often results in reduced

Creating spaces for online deliberation  147

deliberative quality, measured in terms of rational content and contributor interaction (Sobieraj and Berry, 2011) and facilitation of discussion can help to focus discussion and maintain civility (Epstein and Leshed, 2020). Moderation practices can be particularly sensitive in the case of governmental platforms where the management and structuring of discussion can be seen as a form of censorship. Wright (2009) showed how discussion moderation can be vital in turning random position-stating into more focused and productive discourse. He describes two models of moderation: content moderation, in which humans (and also possibly automated programs) pre-moderate content against pre-defined criteria, and interactive moderation, in which the moderator acts as a facilitator, giving feedback, supplying resources and directing the conversation in productive ways. The latter can be seen in the Deliberatorium (Klein, 2011), in which the moderators have a ‘part education and part quality control’ role and can communicate with contributors to help them to produce acceptable posts. Studies of journalists’ involvement in online discussions generated by their stories shows that the presence of an ‘official’ or qualified voice in such debates often results in a more civil conversation (Lewis et al., 2014; Meyer and Carey, 2014) and increased reciprocity (Wright et al., 2020). An example of content moderation can be seen in the AV-based E-Liberate system which was built around the use of Robert’s Rules of Order, a set of directives that designated an orderly process for equitable decision making in face-to-face meetings (Schuler, 2009). However, this feature has not always been popular with users, who felt that their free expression was being constrained by overly-formal rules. In response, the designers incorporated an ‘auto pilot’ feature into the system, allowing users to express themselves without constraint, but only when they considered that moderation was impeding their conversation. Designers of the Unchat system (Noveck, 2003) included a more flexible moderation tool in which moderators were elected from amongst the discussion participants, who have the right to depose them if they disagree with their decisions. Moderation practices are widespread now to counter incivility and abuse on digital platforms and can consist of automated and/or professional moderation or self-moderation drawn from the participant community. While this requirement will persist, careful planning of moderation strategy is needed to ensure that opinion diversity is not too heavily limited (Perrault and Zhang, 2019). Participant Authentication Whether or not discussion participants are required to provide authentication before entering a deliberative space is a further pressing question for deliberative design. Authentication methods vary in strength, from strong forms, such as postal confirmation of offline addresses used by banks and government departments to weaker forms where email addresses or pseudonyms are all that is required to identify a participant (Marx, 1999). The case for requiring user authentication is that strong identities are more likely to contribute to trusting relationships between participants. In an experimental situation, Rhee and Kim (2009) found that when contributors to a discussion were required to reveal social identity cues this resulted in them being more atten-

148  Handbook of digital politics

tive to messages and more likely to elaborate their arguments at a higher cognitive level than in a control group of anonymous discussants. However, authentication introduces barriers to participation (particularly for members of marginalized communities) and there is surely a case for distinguishing between weak authentication required for comment-posters on a political blog and strong authentication required for contributors to a consequential exercise in policy deliberation. Stronger authentication may be particularly important in institutional deliberative initiatives, as De Cindio and Peraboni (2011, p. 104) observe: ‘in order to create a trustworthy social environment that encourages government officers and representatives to undertake online dialogue with citizens, [a] weak form of identification is not adequate: the online identity should, as much as possible, reflect the offline identity’. In less formal situations, deliberation is often aided by weaker forms of authentication, such as stable pseudonyms that allow the maintenance of social ties (Birchall, 2018) or anonymity that can reduce barriers to entry and allow participants to feel freer in their expressions (de Brasi and Gutierrez, 2020). However, in areas such as e-rulemaking, identity matters, as fake accounts and bots can severely diminish the quality of participation (Rinfret et al., 2021).

SOME THOUGHTS ABOUT FUTURE RESEARCH We began by referring to two types of citizen: the unconfidently undecided and the over-confidently dogmatic. Most deliberative practice has been geared towards helping the latter to be more flexible in formulating their preferences. By encouraging holders of hard preferences to justify their positions explicitly and publicly and exposing them to counter-arguments, often stemming from radically different experience, some online deliberative exercises have proved to be a force for greater democratic understanding. A key research question here relates to the durability of such preference shifts. Do people adopt more open-minded outlooks during and shortly after exposure to other perspectives, but then return to ideological intransigence once the deliberative air has cleared? If so, might there be ways of sustaining such democratic outcomes beyond one-off mini-deliberations? Much thought has been devoted to designing spaces for time-limited deliberative events, but what about the possibility of establishing ongoing online deliberative institutions within which citizens might acquire enduring habits of democratic communication? Several researchers have attempted to move beyond the notion of deliberation, both on and offline, as a discrete event. They argue that deliberative norms can best be realized in a scaled-up fashion: as macro rather than micro-deliberation. Parkinson and Mansbridge’s (2012) innovative notion of a ‘deliberative system’ in which there is division of labour and functions between individuals and institutions, each playing distinctive roles in the generation of deliberative outcomes, could have important implications for online deliberative design. If, instead of online spaces having to provide for all the complex norms of deliberation, they were to be seen as one element within a democratic media ecosystem, it would be possible to focus upon

Creating spaces for online deliberation  149

those aspects of public discussion that are best supported by digital technologies, leaving other elements to be provided elsewhere, such as television or newspaper content or local, face-to-face meetings. The practical, political and technical conditions and implications of the institutional interaction that could sustain a deliberative system have yet to be explored in any depth. The role of digital technologies within a macro-social order committed to democratic deliberation gives rise to much more complex problems than the relatively simple communicative challenge of creating isolated silos of high-level deliberation. The three principles of online deliberative quality considered above could be valuable in thinking through deliberation at a systemic level. The principle of encouraging cross-cutting debate, in which citizens encounter strangers and unsought for perspectives, is a key precondition for normatively successful deliberation, but runs counter to the institutional structure of contemporary politics, whereby activists cluster together in partisan formations – a structure exacerbated by the automated targeting of content in digital platforms that can lead to balkanization of participant communities (Feezell, 2018; Sîrbu et al., 2019). There has been little research conducted on ways of enabling mass political parties to deliberate, either internally, with the public or with one another. Indeed, much online deliberative experimentation has proceeded as if parties were irrelevant and preference formation and expression could be reconfigured at the micro-level. Freelon’s acknowledgement of divergent democratic styles is helpful here in opening up space for a more pluralistic sociology of discursive motivation. There is space for more imaginative research on ways of supporting and empowering the first (possibly larger) group of disengaged citizens mentioned at the beginning of this chapter: those who are the least confident, informed and vocal. Such research might involve the development of hybrid spaces of deliberation, in which mass-media audiences are encouraged to go online and participate in debates triggered by television stories and images. Graham’s work on the ways in which audiences of popular cultural content often use their viewing experience as a basis for broader social deliberation is highly promising in this regard. Might it be that the least politically confident or engaged people in contemporary society are unlikely to be attracted to the kinds of innovative web-based spaces in which most deliberative innovation has occurred? The current popularity of social media platforms may well offer a more appropriate space for introducing elements of democratic deliberation. Most of the design innovations highlighted in this chapter have tended to work (when they do work) as niche products, operating within realms of specific consultative environments, rather than reaching out to the general public. Many researchers have analysed the communicative dynamics of Facebook, Twitter, Weibo and YouTube, and studies of deliberation in such spaces have emerged that show both success and failure, in relation to access and civility, rational discussion and reciprocity, extreme views and misinformation (Feezell, 2018; Freelon, 2015; Jennings et al., 2021; Sîrbu et al., 2019; Thimm et al., 2014; Upadhyay, 2014). These massive social networks pose formidable challenges for the scoping of deliberative projects; in a world of global access to online media how does one generate a community of use that is

150  Handbook of digital politics

open enough to be representative but controlled enough to connect a local or expert community to a local or expert discussion? Here lies a major research challenge to the field. In an age of seemingly endless choice in information source and participatory space and as online debate becomes increasingly fragmented and linked to ideological and emotional identity (Bouko and Garcia, 2020) leading to the ‘enclave deliberation’ described by Sunstein (2017), how can spaces be designed to encourage people to step outside of their comfort zones, listen to opposing opinion about difficult topics, and do so in spaces where efficacy might ensue? Many different niches exist on the web in which conversation occurs with a greater or lesser degree of deliberative quality. Such digital niches are formed through complex combinations of social and technical dimensions that lead to varied conditions for effective deliberation. The challenge for designers of deliberative spaces is to translate the successful characteristics of these deliberative niches into more broadly inclusive spaces, shaped by interface design techniques and regulatory protocols that combine sensitivity to democratic normativity and an acknowledgement of cultural practice.

FURTHER READING Bächtiger, A. and Parkinson, J. (2019). Mapping and Measuring Deliberation: Towards a New Deliberative Quality. New York: Oxford University Press. Culloty, E. and Suiter, J. (2021). Disinformation and Manipulation in Digital Media: Information Pathologies. New York: Routledge. Fishkin, J. S. (2018). Democracy When the People Are Thinking: Revitalizing Our Politics Through Public Deliberation. New York: Oxford University Press. Margolis, M. and Moreno-Riaño, G. (2016). The Prospect of Internet Democracy. New York: Routledge. Persily, N. and Tucker, J. A. (eds.) (2020). Social Media and Democracy: The State of the Field and Prospects for Reform. Cambridge: Cambridge University Press.

REFERENCES Barton, M. D. (2005). The future of rational-critical debate in online public spheres. Computers and Composition, 22(2), 177–190. Birchall, C. (2018). Trying not to fall out: The importance of non-political social ties in online political conversation. Information, Communication & Society, 23(7), 963–979. Bouko, C. and Garcia, D. (2020). Patterns of emotional tweets: The case of Brexit after the referendum results. In G. Bouvier and J. Rosenbaum (eds.), Twitter, the Public Sphere, and the Chaos of Online Deliberation (pp. 175–203). Cham: Palgrave Macmillan. Buckingham Shum, S., De Liddo, A., and Klein, M. (2014). DCLA Meet CIDA: Collective intelligence deliberation analytics. The 4th International Conference on Learning Analytics and Knowledge, Indianapolis, IN, USA. Cavalier, R., Kim, M. and Zeiss, Z. S. (2009). Deliberative democracy, online discussion, and project PICOLA. In T. Davies and S. P. Gangadharan (eds.), Online Deliberation: Design, Research and Practice. Stanford, CA: CSLI Publications.

Creating spaces for online deliberation  151

Coe, K., Kenski, K., and Rains, S. A. (2014). Online and uncivil? Patterns and determinants of incivility in newspaper website comments. Journal of Communication, 64(4), 658–679. Coleman, S. (2013). How Voters Feel. New York: Cambridge University Press. Coleman, S. and Blumler, J. G. (2009). The Internet and Democratic Citizenship: Theory, Practice and Policy. Cambridge: Cambridge University Press. Coleman, S. and Gøtze, J. (2001). Bowling Together: Online Public Engagement in Policy Deliberation. London: Hansard Society. Coleman, S., Morrison, D. E., and Yates, S. (2011). The mediation of political disconnection. In K. Brants and K. Voltmer (eds.), Political Communication in Postmodern Democracy (pp. 215–230). Basingstoke: Palgrave Macmillan. Coleman, S. and Moss, G. (2012). Under construction: The field of online deliberation research. Journal of Information Technology & Politics, 9(1), 1–15. Colom, A. (2021). Using WhatsApp for focus group discussions: Ecological validity, inclusion and deliberation. Qualitative Research. https://​doi​.org/​10​.1177/​1468794120986074. Conklin, J. and Begeman, M. L. (1987). gIBIS: A hypertext tool for team design deliberation. Proceedings of the ACM Conference on Hypertext. ACM. Dahlgren, P. (2005). The Internet, public spheres, and political communication: Dispersion and deliberation. Political Communication, 22(2), 147–162. Davies, T., O’Connor, B., Cochran, A., Effrat, J., Parker, A. Newman, B., and Tam, A. (2009). An online environment for democratic deliberation: Motivations, principles, and design. In T. Davies and S. P. Gangadharan (eds.), Online Deliberation: Design, Research and Practice. Stanford, CA: CSLI Publications. de Brasi, L. and Gutierrez, C. (2020). Anonymity and asynchronicity as key design dimensions for the reciprocity of online democratic deliberation. International Journal of Applied Philosophy, 34(2), 183–200. De Cindio, F. (2012). Guidelines for designing deliberative digital habitats: Learning from e-participation for open data initiatives. The Journal of Community Informatics, 8(2). De Cindio, F. and Peraboni, C. (2011). Building digital participation hives: Toward a local public sphere. In M. Foth, L. Forlano, C. Satchell, and M. Gibbs (eds.), From Social Butterfly to Engaged Citizen: Urban Informatics, Social Media, Ubiquitous Computing, and Mobile Technology to Support Citizen Engagement (pp. 93–113). Cambridge, MA: MIT Press. Dryzek, J. S. (2000). Deliberative Democracy and Beyond: Liberals, Critics, Contestations. New York: Oxford University Press. Epstein, D. and Leshed, G. (2020). The magic sauce: Practices of facilitation in online policy deliberation. Journal of Deliberative Democracy, 12(1), 6–10. Feezell, J. T. (2018). Agenda setting through social media: The importance of incidental news exposure and social filtering in the digital era. Political Research Quarterly, 71(2), 482–494. Freelon, D. (2010). Analyzing online political discussion using three models of democratic communication. New Media & Society, 12(7). Freelon, D. (2015). Discourse architecture, ideology, and democratic norms in online political discussion. New Media & Society, 17(5), 772–791. Freelon, D. and Wells, C. (2020). Disinformation as political communication. Political Communication, 37(2), 145–156. Friess, D. and Eilders, C. (2015). A systematic review of online deliberation research. Policy and Internet, 7(3), 319–339. Gastil, J. (2000). By Popular Demand: Revitalizing Representative Democracy Through Deliberative Elections. Berkeley: University of California Press. Graham, T. (2012). Beyond “political” communicative spaces: Talking politics on the Wife Swap Discussion Forum. Journal of Information Technology & Politics, 9(1) 31–45.

152  Handbook of digital politics

Graham, T., Jackson, D., and Wright, S. (2015). From everyday conversation to political action: Talking austerity in online ‘third spaces’. European Journal of Communication, 30(6), 648–665. Graham, T., Jackson, D., and Wright, S. (2016). ‘We need to get together and make ourselves heard’: Everyday online spaces as incubators of political action. Information, Communication & Society, 19(10), 1373–1389. Graham, T. and Wright, S. (2014). Discursive equality and everyday talk online: The impact of “superparticipants.” Journal of Computer-Mediated Communication, 19(3), 625–642. Habermas, J. (1994). Three normative models of democracy. Constellations, 1(1), 1–10. Hill, K. A., and Hughes, J. E. (1999). Cyberpolitics: Citizen Activism in the Age of the Internet. Lanham, MD: Rowman & Littlefield. Iandoli, L., Introne, J., and Klein, M. (2012). Rating in large-scale argumentation systems. In H. Masum and M. Tovey (eds.), The Reputation Society: How Online Opinions Are Reshaping the Offline World (pp. 163–172). Cambridge, MA: MIT Press. Ivison, D. (1997). The Self at Liberty: Political Argument and the Arts of Government. Ithaca, NY: Cornell University Press. Jennings, F. J., Suzuki, V. P., and Hubbard, A. (2021). Social media and democracy: Fostering political deliberation and participation. Western Journal of Communication, 85(2), 147–167. Jerit, J. and Zhao, Y. (2020). Political misinformation. Annual Review of Political Science, 23, 77–94. Karlsson, M. (2010). What does it take to make online deliberation happen? A comparative analysis of 28 online discussion forums. In F. De Cindio, A. Macintosh, and C. Peraboni (eds.), From e-Participation to Online Deliberation: Proceedings of the Fourth International Conference on Online Deliberation, OD2010. Leeds. King, M. (2018). A Review of Online Argument Representation Platforms: A Working Paper for the Scholio Project. December. https://​www​.scholio​.net. Klein, M. (2011). Enabling large-scale deliberation using attention-mediation metrics. Computer Supported Cooperative Work, 21, 449–473. Klein, M. (2015). A critical review of crowd-scale online deliberation technologies. SSRN Electronic Journal. http://​dx​.doi​.org/​10​.2139/​ssrn​.2652888. Kunz, W. and Rittel, H. (1970). Issues as Elements of Information Systems. Working Paper #131, Institut für Grundlagen der Planung I.A., University of Stuttgart. Lewis, S. C., Holton, A. E., and Coddington, M. (2014). Reciprocal journalism: A concept of mutual exchange between journalists and audiences. Journalism Practice, 8(2), 229–241. Macintosh, A. (2008). The emergence of digital governance. Significance, 5(4), 176–178. Mansbridge, J. (1999). Everyday talk in the deliberative system. In S. Macedo (ed.), Deliberative Politics: Essays on Democracy and Disagreement (pp. 211–240). Oxford: Oxford University Press. Marx, G. T. (1999). What’s in a name? Some reflections on the sociology of anonymity. Information Society, 15, 99–112. McPherson, M., Smith-Lovin, L., and Cook, J. M. (2001). Birds of a feather: Homophily in social networks. Annual Review of Sociology, 27, 415–444. Meyer, H. K. and Carey, M. C. (2014). In moderation: Examining how journalists’ attitudes toward online comments affect the creation of community. Journalism Practice, 8(2), 213–228. Miller, M. L. and Vaccari, C. (2020). Digital threats to democracy: Comparative lessons and possible remedies. International Journal of Press/Politics, 25(3), 333–356. Monnoyer-Smith, L. (2012). The technological dimension of deliberation: A comparison between online and offline participation. In S. Coleman and P. J. Shane (eds.), Connecting Democracy: Online Consultation and the Flow of Political Communication (pp. 191–208). Cambridge, MA: MIT Press.

Creating spaces for online deliberation  153

Morozov, E. (2012). The Net Delusion: The Dark Side of Internet Freedom. New York: Public Affairs. Mutz, D. C. (2006). Hearing the Other Side: Deliberative versus Participatory Democracy. New York: Cambridge University Press. Nahon, K. and Hemsley, J. (2014). Homophily in the guise of cross-linking: Political blogs and content. American Behavioral Scientist, 58(10), 1294–1313. Neblo, M. A., Esterling, K. M., Kennedy, R. P., Lazer, D. M. J., and Sokhey, A. E. (2010). Who wants to deliberate and why? American Political Science Review, 104(3), 566–583. Noveck, B. S. (2003). Designing deliberative democracy in cyberspace: The role of the cyber-lawyer. Boston University Journal of Science & Technology Law, 9, 1–90. Noveck, B. S. (2010). Wiki Government: How Technology Can Make Government Better, Democracy Stronger, and Citizens More Powerful. Washington, DC: Brookings Institution Press. O’Connor, C. and Weatherall, J. O. (2019). The Misinformation Age. New Haven: Yale University Press. Parkinson, J. and Mansbridge, J. (eds.) (2012). Deliberative Systems: Deliberative Democracy at the Large Scale. Cambridge: Cambridge University Press. Perrault, S. T. and Zhang, W. (2019). Effects of moderation and opinion heterogeneity on attitude towards the online deliberation experience. Conference on Human Factors in Computing Systems – Proceedings. Pfetsch, B. (2018). Dissonant and disconnected public spheres as challenge for political communication research. Journal of the European Institute for Communication and Culture, 25(1–2), 59–65. Pickard, V. W. (2008). Cooptation and cooperation: Institutional exemplars of democratic internet technology. New Media & Society, 10(4), 625–645. Pingree, R. J. (2009). Decision structure: A new approach to three problems in deliberation. In T. Davies and S. P. Gangadharan (eds.), Online Deliberation: Design, Research and Practice (pp. 309–316). Stanford, CA: CSLI Publications. Rhee, J. W. and Kim, E. M. (2009). Deliberation on the net: Lessons from a field experiment. In T. Davies and S. P. Gangadharan (eds.), Online Deliberation: Design, Research, and Practice (pp. 223–232). San Francisco: CSLI Publications. Rinfret, S., Duffy, R., Cook, J., and St. Onge, S. (2021). Bots, fake comments, and E-rulemaking: The impact on federal regulations. International Journal of Public Administration. https://​ doi​.org/​10​.1080/​01900692​.2021​.1931314. Schlosberg, D., Zavestoski, S., and Shulman, S. W. (2007). Democracy and e-rulemaking: Web-based technologies, and the potential for deliberation. eRulemaking Research Group, Paper 1, University of Massachusetts – Amherst. Schuler, D. (2009). Online civic deliberation with E-Liberate. In T. Davies and S. P. Gangadharan (eds.), Online Deliberation: Design, Research and Practice. Stanford, CA: CSLI Publications. Shortall, R., Itten, A., Delft, T. U., Van Der Meer, M., Murukannaiah, P. K., and Jonker, C. M. (2021). Reason against the machine: Future directions for mass online deliberation. ArXiv Preprint. Sîrbu, A., Pedreschi, D., Giannotti, F., and Kertész, J. (2019). Algorithmic bias amplifies opinion fragmentation and polarization: A bounded confidence model. PLoS ONE, 14(3). Sobieraj, S. and Berry, J. M. (2011). From incivility to outrage: Political discourse in blogs, talk radio, and cable news. Political Communication, 28(1), 19–41. Spiliotopoulou, L. and Charalabidis, Y. (2015). Web 2.0 in governance: A framework for utilizing social media and opinion mining methods and tools in policy deliberation. In Social Media and Networking: Concepts, Methodologies, Tools, and Applications (vol. 1, pp. 281–303). Pennsylvania: IGI Global.

154  Handbook of digital politics

Steiner, J. (2012). The Foundations of Deliberative Democracy: Empirical Research and Normative Implications. Cambridge: Cambridge University Press. Sunstein, C. (2002). The law of group polarization. Journal of Political Philosophy, 10(2), 175–195. Sunstein, C. (2017). #Republic: Divided Democracy in the Age of Social Media. Princeton, NJ: Princeton University Press. Thimm, C., Dang-Anh, M., and Einspänner, J. (2014). Mediatized politics: Structures and strategies of discursive participation and online deliberation on Twitter. In A. Hepp and F. Krotz (eds.), Mediatized Worlds: Culture and Society in a Media Age (pp. 253–270). Basingstoke: Palgrave Macmillan. Tucey, C. B. (2010). Online vs. face-to-face deliberation on the global warming and stem cell issues. Western Political Science Association 2010 Annual Meeting Paper. Upadhyay, M. (2014). Political deliberation on Twitter: Is Twitter emerging as an opinion leader? International Conference on People, Politics and Media (ICPPM), 25–26 April, Jagran Lakecity University, India. Wilhelm, A. G. (2000). Democracy in the Digital Age: Challenges to Political Life in Cyberspace. New York: Psychology Press. Wright, S. (2006). Government-run online discussion fora: Moderation, censorship and the shadow of control. The British Journal of Politics & International Relations, 8(4), 550–568. Wright, S. (2009). The role of the moderator: Problems and possibilities for government-run online discussion forums. In T. Davies and S. P. Gangadharan (eds.), Online Deliberation: Design, Research and Practice. Stanford, CA: CSLI Publications. Wright, S. (2012). From “third place” to “third space”: Everyday political talk in non-political online spaces. Javnost, 19, 5–20. Wright, S., Jackson, D., and Graham, T. (2020). When journalists go “below the line”: Comment spaces at The Guardian (2006–2017). Journalism Studies, 21(1), 107–126. Wright, S. and Street, J. (2007). Democracy, deliberation and design: The case of online discussion forums. New Media & Society, 9(5), 849–869. Wright, S., Trott, V., and Jones, C. (2020). ‘The pussy ain’t worth it, bro’: Assessing the discourse and structure of MGTOW. Information, Communication & Society, 23(6), 908–925.

10. New frontiers in two-screen politics Nick Anstead and Ben O’Loughlin

INTRODUCTION As is inevitable when researching rapidly evolving topics, revisiting a piece of work about political communication even just a few years after it was published can be instructive, surprising and, on occasions, chastening. Certainly, re-reading the chapter that we wrote for the first edition of this book (Anstead and O’Loughlin, 2015b) was a salutary experience. The chapter was built on a body of work we were then in the process of constructing, which examined the different ways in which people were using new media to consume, comment on and interact with traditional live television broadcasting, and the ways in which the data this created was being analysed and then recycled back into the public sphere (see Ampofo et al., 2011; Anstead and O’Loughlin, 2011; Anstead and O’Loughlin, 2015a). In the chapter in the first edition, we focused our attentions on the development of two-screen viewing. We defined these practices as ‘simultaneous television watching while reading and creating online content on a smart phone, laptop or tablet’ (Anstead and O’Loughlin, 2015b, p. 306). In many ways, the ideas in the original chapter continue to be prescient. This is evident in the growing size of the datasets being created on social media sites in response to broadcast events. To offer some context: the very first paper we authored on two-screen viewing was an analysis of how Twitter was used during an episode of the BBC political panel programme Question Time broadcast in 2009 (Anstead and O’Loughlin, 2011). This broadcast featured Nick Griffin, leader of the far-right British National Party, as a panellist. As a result, it was highly controversial and generated significant comment from politicians and journalists, as well as public interest. Using a crude tool to access the Twitter API, we were able to gather approximately 40,000 tweets from the period of the broadcast (a significant proportion of the 53,000 tweets that contemporary commentary suggested appeared on Twitter during the programme). At the time, this seemed like a massive corpus. In retrospect and with the continuing expansion of two-screening practices, these numbers seem almost laughably small. The first ever televised election debate in the UK in 2010 saw 200,000 related tweets being published while the programme was on air (Linguamatics, 2010). By the 2015 general election, a ‘seven-way’ party leaders debate saw 1.5 million debate-related tweets appearing during the broadcast (BBC Online, 2015). Numbers from the UK are dwarfed by the quantity of content produced by second screening during election debates in the United States. The first televised election debate in the 2012 presidential election saw 10.3 million debate related tweets published. In more recent US elections, Twitter has ceased to publish official figures, 155

156  Handbook of digital politics

although other sources can give an indication as to the scale of second screening. For example, Nielsen social media monitoring suggests that the 2016 debate saw 17.1 million interactions on Twitter, and 65.9 million debate related interactions on Facebook (Jarvey, 2016).1 More broadly, we predicted that the development of two-screen viewing would have two important ramifications for those researching political communication. First, social media has the potential to provide researchers with vast amounts of data for their endeavours, and even create new forms of ‘live research’ (Elmer, 2013). Second, and as significant, other political actors – including governments, politicians, broadcasters, advertisers, and corporations – would immerse themselves in these new streams of data and the networks created by two-screen viewing. They would do this to harness them for their own purposes. This development has the potential to create feedback loops where elite actors could use social media to manage and steer public conversations. What the intervening years have made abundantly clear is the tensions between these two predictions. On the one hand, we had a vision of a raw stream of data allowing researchers to tap into instant audience reactions to broadcast events. High profile political programmes, such as televised election debates, and the conversations they generate on social media can provide an insight into public opinion and citizens attitudes to democratic institutions (Anstead and O’Loughlin, 2015a). This idea is embodied in research that focuses on citizens and examines how two-screen viewing relates to their political participation, engagement, and learning. On the other hand, two-screening creates a networked environment with a range of participants, including institutional actors such as political parties, important political office holders and traditional media figures. Inevitably, actors in this network have unequal levels of influence. This leads to a second strand of research, which is more interested in questions of power and how it is distributed in online conversations. There are certainly tensions between these two approaches. However, as we shall argue below, it can be a fruitful tension. This chapter proceeds as follows. First, we examine the changing technological and political landscape since our original chapter, and how this has shaped both the practices and perceptions of two-screen viewing. Second, we consider some of the key literature in the field and its findings. Finally, we look at future directions for research and the challenges faced by scholars studying two-screen practices.

TWO-SCREEN VIEWING IN THE CONTEXT OF CHANGING TECHNOLOGY AND EVOLVING POLITICS In the years since our original chapter, the social media environment has continued to evolve rapidly. One obvious reason for this is that social network sites have continued to increase their user base. In 2014, Facebook had 1.3 billion users globally. By 2021, this figure had more than doubled to 2.9 billion users. Over the same period, Twitter had grown from 235 million users to 322 million users (Statistica, 2021).

New frontiers in two-screen politics  157

Additionally, a host of new and distinctive social media companies and services – including WhatsApp and TikTok – have grown to become major platforms for content creation and sharing. Each of these networks has idiosyncratic user-bases and attributes. As a result, each platform poses distinctive challenges for researchers. The televisual element of two-screen viewing has also changed dramatically in recent years, becoming decreasingly recognizable from the broadcast technology of the mass media era. The traditional broadcast form of television has been augmented by catch-up and streaming services, which have been created by older legacy television companies (such as the BBC’s iPlayer service), by major rights holders and producers (such as Disney), by technology behemoths moving into television production (for example Apple and Amazon), and new entrants dedicated to content production and distribution (notably Netflix). The move to streaming has ramifications for the temporal dimensions of television consumption. With increased choice and the ability to watch ‘on demand’ viewers are less likely to watch programmes simultaneously. It is not a coincidence that much of the work on second screening has focused on ‘event television’ – election debates, sports, and even the Eurovision Song Contest – which more closely conform to the model of a traditional, simultaneously consumed broadcast event. However, even these examples of ‘event television’ are increasingly likely to be consumed asynchronously or even experienced on social media, where small sections of the broadcast are clipped, shared, and then go viral. This may occur for the purpose of entertainment or to make a political point. As we discuss below, non-simultaneous viewing and new ways of consuming content challenges some of the core assumptions our earlier work made about the relationship between broadcast and social media. An additional technology-related factor, which was wholly unpredictable at the time we wrote the original version of the chapter, is the Covid-19 pandemic. This event had a huge impact on how technology was used and greatly increased reliance on mediated interaction between individuals who could not be physically co-present with each other. Two-screen experiences of television consumption have become more normalized, and taken on new forms, whether it is friends and family interacting on Zoom or using chat applications such as WhatsApp while they watch their favourite programmes together. The pandemic even saw the emergence of dedicated applications, like TeleParty, which allowed viewers in multiple locations to share the streaming experience while interacting with each other. If the technological landscape has dramatically changed, the political landscape has also evolved beyond recognition. The ‘End of History’ consensus of the 1990s and early 2000s, marked by a belief in the supremacy and stability of liberal democracies has been swept aside. The worries from this earlier era – specifically about relatively low participation in democratic institutions by recent historic standards – have been swept aside by much greater concerns about the underlying viability of liberal democracy in times of crisis. Recent years have seen at least three international crises (the financial crisis, the Covid pandemic, and the Russian–Ukrainian war), any of which would have arguably been era defining in their own right. This ‘age of crisis’ led to broader concern about what has been termed democratic backsliding

158  Handbook of digital politics

(Waldner and Lust, 2018), where the institutions of liberal democracy and the rule of law are subverted and co-opted by populist would-be authoritarians. These processes are perhaps most evident in central Europe, notably Hungry, where Victor Orbán’s Fidesz regime has systematically set about attacking and replacing the independent judiciary, electoral regulators, the media, and higher education. However, backsliding has also been evident in older, established democracies. The Brexit process in the United Kingdom saw the deployment of populist rhetoric attacking politicians and the judiciary and an illegal prorogation of Parliament. Donald Trump’s refusal to accept his defeat in the 2020 presidential election directly led to the storming of the US Capitol. Even in the world’s premier democracy, the most basic of democratic tests – the smooth transition of power from one administration to another – looks to be in jeopardy. What are the practical ramifications of these changes for scholars researching two-screen viewing? The most significant change relates to the questions that have tended to dominate our thinking. The broad sense that new communication technologies can be beneficial to democratic institutions, which hit its height after Barack Obama’s first presidential election victory in 2008 and the Arab Spring of 2011 has markedly receded, being replaced by fears that social media platforms provide an environment where misinformation can circulate, radicalization can occur, and misleading political advertising can be targeted at voters with little or no regulatory oversight. We have gone from a period where the intersection between technology and politics was largely a site of optimism, to one where that same intersection is now understood through the prism of a very clear pessimism. Two-screen viewing offers a particularly interesting case study of changing attitudes to the technological-political interface because, at least in its original conception, it posited a relatively gentle idea of change, where established media and practices in the form of broadcast television were augmented by newer practices in the form of social media commentary. While that commentary could well be critical of the content of television broadcasts, there was still a strong sense that the two components of two-screening were ultimately complimentary to each other. It is questionable whether this position is sustainable given the tone of contemporary politics.

RECENT ACADEMIC RESEARCH INTO TWO-SCREEN VIEWING Understanding the Relationship between Broadcast Events and Online Discussion In its simplest form, the relationship between broadcast events and the online conversations which they generate on social media might seem self-evident, but that connection still needs to be documented. As a result, several early studies on two-screen practices set out to show empirically that social media users responded directly to content that was being broadcast on the television screen. These studies demonstrated

New frontiers in two-screen politics  159

a form of stimulus/response relationship (Anstead and O’Loughlin, 2011; Giglietto and Selva, 2014). This model of interaction is supported by studies that compare the topic saliency in debate transcripts and social media discussion of debate broadcasts, and find a relationship between the level of attention particular issues receive across the two (Vergeer and Franses, 2016). However, it is also evident that the two-screen relationship is more complex than a simple stimulus/response model would suggest, as the content of online conversations also has the potential to frame dominant understandings of the broadcast event itself. In their study of the 2012 US election debates, for example, Freelon and Karpf (2015) note that many of the breakout moments of the debate that received post broadcast coverage on mass media were first framed on social media commentary. Only then did they cross over into more traditional media. Similarly, in their study of South Korean Mayoral TV debates Heo et al (2016) found that the discourse which emerged from the debate was shaped by a combination of traditional broadcast commentary and a parallel online discussion. Two-screen viewing practices have also challenged more traditional social dimensions of media consumption. D’Heer and Courtois (2016) argue that two-screen viewing undermines interactions between household members, even when they are watching broadcast media in the same space, as it leads to verbal avoidance and less frequent conversations. A counter-position to this though is offered by the idea of ‘many-screening’ where individuals in a household all consume broadcast media, while also interacting with their personal networks through other devices (Anstead et al., 2014). However, they use the content that they are consuming on the online networks to start real-world conversations with others in their physical space. As such, many-screening theorized a more dynamic and multi-directional interaction between broadcast content, online social networks, and face-to-face conversations. Relationship between Two-Screening Practices and Democracy A broader question addressed in the field focuses on how two-screening relates to democratic practices. In some ways, this body of work can be seen as a response to earlier arguments, both scholarly and polemical, that have downplayed the democratic significance of online engagement. This earlier work has argued that online political engagement can only ever amount to ‘clicktivism’ and, as such, is little more than ersatz politics, and is certainly inferior to real-world engagement (Gladwell, 2010; Morozov, 2011). In contrast though, research into second-screening practices has found that online discussions in response to broadcast events are positively related to various democratic ‘goods’. Vaccari et al. (2015) note that there is a correlation between citizens who are most engaged with two-screening (especially those who undertake more complex two-screening practices such as writing comments or engaging in conversation) and higher levels of offline political engagement. Other studies have also argued that second screening is correlated with higher levels of political learning (Gil de Zúñiga et al., 2015) and increased levels of social capital (Huber et al., 2019).

160  Handbook of digital politics

A different question relates to the quality of the debate generated by two-screening. Again, literature on two-screening offers a somewhat revisionist response to dominant contemporary paradigms, which have tended to focus on the weaknesses of online debate, characterizing these interactions as polarized and uncivil (Chen, 2017). In contrast, research has suggested that two-screening can promote deliberative forms of debate (Camaj, 2021). Another study found that citizens who participated in two-screening are more likely to change their opinions on political issues, indicative of a level of open mindedness in approaching a political discussion (Barnidge et al., 2017). Two-screening practices may also be shaped by democratic culture and institutions. Evidence for this is found in the form that two-screen online networks and the content that flows through them takes in different national contexts. One study comparing the United States and France, for example, found that French two-screen conversations were more likely to be driven by elite opinion than those in the United States. The authors hypothesize that these different practices are the product of the two countries’ distinctive political cultures (Wells et al., 2016b). How Do Two-Screen Practices Relate to Power Relations? Any discussion of democratic practices necessarily needs to extend into questions of power. In the context of two-screening, a key question relates to how power is distributed across the networks that facilitate online conversations. At the outset it is worth noting that two-screeners, particularly for political events, are a minority of a minority of a minority in the population (i.e. they are those who are consuming a political broadcast event, who have an account on a particular social network, and who use that account to post about politics). It is this fact that leads to accusations of non-representative social media bubbles which are best avoided by politicians and journalists who want to understand the general public’s reaction to events. Even within the content networks generated by second screening, power is not distributed evenly. Instead, it seems to be the case that the bulk of the content is produced by relatively few participants (Anstead and O’Loughlin, 2011; Bentivegna and Marchetti, 2015). Two-screening networks may also reflect structural inequality within and between societies. Men are more likely to two-screen than women for example (Houston et al., 2013), while those who wield network power and act as conversational gatekeepers may come from a relatively narrow strata of society (Heo et al., 2016). This problem may be even more pronounced in economically less developed countries with weaker digital infrastructures. In these circumstances, we may see two-screen networks emerging which reflect these digital inequalities (Barnidge et al., 2019). However, two-screening also has the potential to challenge structural inequalities within democracies, by providing a space for groups of people who are excluded or under-represented in traditional media coverage or political participation to voice their opinions. Some evidence, for example, suggests that young people are proportionately more likely to second screen, as opposed to older people (McKinney

New frontiers in two-screen politics  161

et al., 2014). This is the converse of many other forms of political participation, most notably voting. Other research has argued that the most important effects of two-screen viewing are found among low interest voters (Vaccari and Valeriani, 2018). In terms of political discourse, Ceron and Splendore (2018) argue that two-screening provides a space for ideas that are normally excluded from political debate to be discussed (the main example they focus on are anti-political ideas. It is of course open to question whether the increased expression of anti-political ideas is beneficial for politics more generally).

THE CHALLENGES OF RESEARCHING TWO-SCREEN VIEWING As we detail above, much of the original literature on two-screening was built on a specific set of assumptions about how the practice would occur. Implicit in this was the idea that, while the media landscape was changing, the televisual medium remained prime in the sense that it was the stimulus and the social media posts were the response. Participants’ shared experience would be created by the broadcast event, and they would respond to what was being shown on a television screen by commenting and engaging on social media. It is now apparent that the simplicity of the vision we offered – citizens watching political television while reading and commenting in real time on social media platforms – has not aged that well, as the flows of information between television and social media (and back again) are clearly more multifaceted. Non-Simultaneous Two-Screening The way audiences consume television has changed dramatically in recent years, with a significant shift to non-simultaneous viewing through streaming and catch-up services. Unsurprisingly, research conducted on this suggests that these developments fundamentally change the nature of the two-screening experience. Pittman and Tefertiller (2015) undertook a natural experiment, drawing Twitter data both from what they term appointment television (which is broadcast at a set time on a traditional network) and streamed television, which is made and distributed by online only companies like Netflix, so is exclusively available via streaming services. The study found that the two-screening dataset created by the two types of programmes had different temporal characteristics for the two-screening dataset. For appointment television, most comments were published during the period of broadcast, and then the numbers fell away quickly. For streamed television, the content is produced steadily over a much longer period. Interestingly though, the same article also notes that there are not substantive differences in terms of the level of use of the social functions of Twitter (for example retweeting or directing a tweet to another user by including their Twitter name) across different types of viewing.2

162  Handbook of digital politics

This research is useful, but it still leaves a lot of questions unaddressed. This is particularly true of what could be termed ‘event television’. This is distinct from Pittman and Tefertiller’s category of appointment television because it tends to involve live and unpredictable events. The obvious examples of event television occur in politics and sports. The value of watching live and in real time is clearly far greater. However, even for these sorts of events it seems likely that distinct audiences may be emerging, some watching the event in real-time and some watching on catch-up (even if the delay is only a few minutes long). Yet we know very little about how this increased level of synchronicity effects the two-screen experience. An additional audience category that requires further research are people who experience televisual events, but only through the products of other people’s two-screening. In this situation, social media users may not ever see the original broadcast source material, but instead see the comments or clips posted by others online. Certainly we know that it is increasingly common for some citizens (particularly young people) to experience politics almost exclusively through social media (Ford et al., 2021). However, it is important recognize that this does not remove them from the processes of second screening, as reactions to and content from broadcast television posted by other users will play a significant role in shaping the content they are exposed to. Studying Diverse Online Environments Most of the research on two-screen viewing appearing in recent years has focused on the interaction between traditional broadcast media and one social platform – namely, Twitter. This is because Twitter has a relatively open API (Application Programming Interface), meaning that there are very low barriers to accessing significant quantities of data from the service. There is a risk that researchers are falling into the trap of being led by data that is readily available to them, and missing practices in other spaces. This is particularly problematic if these other spaces are used by groups of citizens who are substantively different to Twitter users. As cited earlier in this chapter, the data relating to the 2016 US presidential election debates suggests there were nearly four times the quantity of debate related content published on Facebook than on Twitter (Jarvey, 2016). For researchers though, extracting data from Facebook remains a much more difficult proposition, not least because the platform has increasingly closed possible avenues for data gathering. This has been done on the grounds that the platform is safeguarding users’ privacy, an ironic consequence of the Cambridge Analytica scandal. While these events made it clear that Facebook urgently needed greater scrutiny from academia and civil society, it was also the product of illegal data harvesting. As a result, Facebook responded by closing their (already limited) APIs that allowed researchers to engage in data-gathering. The result is that transparency and oversight have been hampered still further. The relative availability of Facebook and Twitter data (and the biases it can generate towards the latter) matters, as the two platforms have distinctive userbases. To

New frontiers in two-screen politics  163

take the UK as an example, Facebook is used by 83 per cent of people over the age of 16. In contrast, Twitter has a penetration rate of 37 per cent. Even more striking is the age distribution of users on these platforms. Twitter is used by 47 per cent of 16- to 24-year-olds, in comparison with the 69 per cent in this age group who use Facebook. Among those over the age of 65, only 21 per cent use Twitter, in comparison with 91 per cent who use Facebook (Ofcom, 2021, Figure 1.30). The challenges of understanding two-screening become even more acute when we think about other platforms. WhatsApp, for example, is a private encrypted chat application. In many parts of the world, it has become central both to politics generally and other more specific practices, such as the spread of misinformation or the organization of political violence (Banaji et al., 2019). However, since it is a secure private chat network, primary data from the platform will never be available to researchers. Instead, studying WhatsApp requires the development of a distinctive methodological toolkit, which might include in-depth interviews or participant observation, for example. Similarly, different methods would be required for analysing two-screening practices that rely on video technologies, such as Zoom or TeleParty. As a result, researchers looking at two-screening on private or non-text platforms may need to move away from the types of research associated with big data, and instead start to employ more qualitative approaches. This would be a fundamental shift from the vast majority of the work that has been published in the field in recent years. Populism and Populist Leaders Despite the many additional ambiguities and complexities related to the idea of second screening we outline above, it is ironic that Donald Trump’s use of Twitter frequently met the requirements of a very traditional definition of the phenomenon, with his Twitter account frequently publishing comments that were clearly a direct, real-time response to content that was being broadcast on Fox News (Gertz, 2018a). While some earlier work on two-screening has identified the possibility that politicians could participate in and reshape online conversations (Wallsten, 2014), it would have taken an impressive leap of the imagination to consider the possibility of the most powerful democratically elected figure in the world doing this on a regular basis. Trump’s tweets created a feedback loop, wherein they would play a significant role in shaping the media agenda (Wells et al., 2016a). Sometimes the effects could be dramatic: on one occasion, Trump appeared to threaten nuclear strikes against North Korea in response to a segment on the morning show Fox and Friends (Gertz, 2018b). Indeed, a dystopian fictionalized account of a United States/North Korean nuclear war written by the security analyst Jeffrey Lewis takes these events to their logical conclusion, when an aggressive set of tweets published by Trump in response to a Fox News segment is interpreted as a prelude to war in Pyongyang, triggering a North Korean pre-emptive strike on the United States (Lewis, 2018). More broadly, the wider success of populist politics in recent times has challenged many of the core assumptions in the second-screening literature, which had tended

164  Handbook of digital politics

towards the hopeful. The changes being wrought by technology were assumed to be gradualist, co-existing with traditional broadcast media. Scholars researching the field were keen to find democratic ‘goods’ that emerged from the practice, such as increasing levels of engagement, political learning or social capital. They attempted to prove that second screening could lead to high quality political debate. These predispositions were certainly evident in our own work on the 2009 Nick Griffin episode of BBC Question Time, where one of our key findings involved viewers employing social networks to attack and undermine a neo-fascist political leader appearing on television, an observation we rather relished (Anstead and O’Loughlin, 2011). However, in a contemporary setting, similar sorts of practices can start to look more sinister. Populist politics is constructed based on creating an in-group (representing some version of the pure people) and a deviant out-group (Mudde and Kaltwasser, 2017). This out-group might be made up of the political class, the judiciary or an ethnic or religious minority. It might also include media professionals. In this context, two-screening practices can start to look more like a mob, constantly attacking their ideological opponents. This is particularly true if populist leaders and their supporters are willing to make such attacks systematically and strategically, building networks of activists and bots to circulate false information (Howard, 2020). As a result, research on second screening needs to start to ask some additional, difficult questions. Specifically, do second-screening practices have the potential to be anti-democratic or illiberal, in the sense that they can – when deployed as a form of large scale and aggressive attack on individuals, broadcast networks or institutions – actually undermine open and free debate. Addressing questions of this kind requires a shift away from the audience/data focus that has thus far dominated the field, and instead a move into understanding how both elite actors and institutions respond to second screening, an area rarely covered in existing research.

CONCLUSION The challenge of researching second screening and political populism points to an idea which is important for any research in the area: power within a second-screening network (and how it constitutes and is constituted by that network) is a feature not a bug. Put differently, we should not set up understanding the audiences’ reactions to televisual events as being distinct from understanding the power relations within that network. This point is perhaps best illustrated with an example. Second-screening research has tended to bemoan the influence of bots, even seeking technical mechanisms to filter the content that they generate out of datasets. The underlying assumption behind this choice is that they distort what the ‘real’ conversation would look like if it only involved human actors. However, bots and their deployment are products of economic and institutional power, so their use can also be seen as reflective of the political context in which the network has developed. As such, there is great value in understanding it.

New frontiers in two-screen politics  165

The concept of two-screening started off with a very simple idea: people sitting, watching television, posting comments on social networks in response to what they were seeing. It is now clear this concept needs to be stretched in a variety of ways, recognizing the changing temporal aspects of television consumption, the role that social media plays in framing a wider understanding of televised events, and the impact that second screening has not just on audiences, but also on politicians and journalists, as well as political and media institutions. Only by understanding these various dynamics and interrelationships can we really understand two-screen viewing.

NOTES 1. No official reason has been given for Twitter not providing this data. It may simply be that the scale of the numbers is now so vast it has become too hard to offer a precise figure. Alternatively, it might be because the relationship between Twitter and electoral politics has become increasingly fraught, especially in the Trump era. Despite a thorough search and reaching out to contacts at Twitter, the authors were unable to find comparable figures for the 2020 televised US election debates. 2. There is one major flaw with this study (which in fairness, it should be said the authors acknowledge). The data is gathered in each case by collecting tweets that use a particular hashtag, i.e. #DowntonAbbey. On a practical level, this means any tweets not using the hashtag or using a variation of the hashtag will not be collected. An important consequence of this is that the authors are not able to compare the prominence of hashtags across the two types of programmes. A hashtag is after all a device for searching and organizing content. This capability will arguably become more central to the two-screen experience when viewers are removed from a shared temporal context.

REFERENCES Ampofo, L., Anstead, N., and O’Loughlin, B. (2011). Trust, confidence, and credibility: Citizen responses on Twitter to opinion polls during the 2010 UK general election. Information, Communication & Society, 14(6), 850–871. Anstead, E., Benford, S., and Houghton, R. J. (2014). Many-screen viewing: Evaluating an Olympics companion application. Proceedings of the ACM. Anstead, N. and O’Loughlin, B. (2011). The emerging viewertariat and BBC Question Time: Television debate and real-time commenting online. International Journal of Press/ Politics, 16(4), 440–462. Anstead, N. and O’Loughlin, B. (2015a). Social media analysis and public opinion: The 2010 UK general election. Journal of Computer-Mediated Communication, 20(2), 204–220. Anstead, N. and O’Loughlin, B. (2015b). Two-screen politics: Evidence, theory and challenges. In S. Coleman and D. Freelon (eds.), Handbook of Digital Politics (pp. 306–323). Cheltenham, UK and Northampton, MA, USA: Edward Elgar Publishing. Banaji, S., Bhat, R., Agarwal, A., Passanha, N., and Sadhana Pravin, M. (2019). WhatsApp vigilantes: An exploration of citizen reception and circulation of WhatsApp misinformation linked to mob violence in India. London: Department of Media and Communications, London School of Economics and Political Science. http://​eprints​.lse​.ac​.uk/​104316.

166  Handbook of digital politics

Barnidge, M., Diehl, T., and Rojas, H. (2019). Second screening for news and digital divides. Social Science Computer Review, 37(1), 55–72. Barnidge, M., Gil de Zúñiga, H., and Diehl, T. (2017). Second screening and political persuasion on social media. Journal of Broadcasting and Electronic Media, 61(2), 309–331. BBC Online (2015). How the internet reacted to the leaders’ debate. London: BBC Online. https://​www​.bbc​.co​.uk/​news/​election​-2015​-32174120. Bentivegna, S. and Marchetti, R. (2015). Live tweeting a political debate: The case of the ‘Italia bene comune’. European Journal of Communication, 30(6), 631–647. Camaj, L. (2021). Real time political deliberation on social media: Can televised debates lead to rational and civil discussions on broadcasters’ Facebook pages. Information, Communication & Society, 24(13), 1907–1924. Ceron, A. and Splendore, S. (2018). From contents to comments: Social TV and perceived pluralism in political talk shows. New Media & Society, 20(2), 659–675. Chen, G. M. (2017). Online Incivility and Public Debate: Nasty Talk. Cham: Palgrave Macmillan. D’Heer, E. and Courtois, C. (2016). The changing dynamics of television consumption in the multimedia living room. Convergence, 22(1), 3–17. Elmer, G. (2013). Live research: Twittering an election debate. New Media & Society, 15(1), 18–30. Ford, R., Bale, T., Jennings, W., and Surridge, P. (2021). Fragmented and polarised: Broadcasting and social media. In R. Ford, T. Bale, W. Jennings, and P. Surridge (eds.), The British General Election of 2019 (pp. 305–345). New York: Springer. Freelon, D. and Karpf, D. (2015). Of big birds and bayonets: Hybrid Twitter interactivity in the 2012 Presidential debates. Information, Communication & Society, 18(4), 390–406. Gertz, M. (2018a). I’ve studied the Trump-Fox feedback loop for months. It’s crazier than you think. Politico. https://​www​.politico​.com/​magazine/​story/​2018/​01/​05/​trump​-media​ -feedback​-loop​-216248/​. Gertz, M. (2018b). The president just threatened a nuclear strike while live-tweeting a Fox News segment. https://​twitter​.com/​MattGertz/​status/​948359183526694913. Giglietto, F. and Selva, D. (2014). Second screen and participation: A content analysis on a full season dataset of tweets. Journal of Communication, 64(2), 260–277. Gil de Zúñiga, H., Garcia-Perdomo, V., and McGregor, S. C. (2015). What is second screening? Exploring motivations of second screen use and its effect on online political participation. Journal of Communication, 65(5), 793–815. Gladwell, M. (2010). Small change. The New Yorker, 27 September. Heo, Y.-C., Park, J.-Y., Kim, J.-Y., and Park, H.-W. (2016). The emerging viewertariat in South Korea: The Seoul mayoral TV debate on Twitter, Facebook, and blogs. Telematics and Informatics, 33(2), 570–583. Houston, J. B., McKinney, M. S., Hawthorne, J., and Spialek, M. L. (2013). Frequency of Tweeting during presidential debates: Effect on debate attitudes and knowledge. Communication Studies, 64(5), 548–560. Howard, P. (2020). Lie Machines: How to Save Democracy from Troll Armies, Deceitful Robots, Junk News Operations, and Political Operatives. New Haven: Yale University Press. Huber, B., De Zúñiga, H. G., Diehl, T., and Liu, J. (2019). Effects of second screening: Building social media social capital through dual screen use. Human Communication Research, 45(3), 334–365. Jarvey, N. (2016). First presidential debate breaks Twitter Record. Los Angeles: The Hollywood Reporter. https://​www​.hollywoodreporter​.com/​news/​politics​-news/​first​ -presidential​-debate​-breaks​-twitter​-932779/​. Lewis, J. (2018). The 2020 Commission Report on the North Korean Nuclear Attacks Against the United States. New York: Random House.

New frontiers in two-screen politics  167

Linguamatics (2010). Linguamatics reveals instant reactions on Twitter to final televised election debate. London: Linguamatics. https://​www​.linguamatics​.com/​blog/​linguamatics​ -reveals​-instant​-reactions​-twitter​-final​-televised​-election​-debate. McKinney, M. S., Houston, J. B., and Hawthorne, J. (2014). Social watching a 2012 Republican presidential primary debate. American Behavioral Scientist, 58(4), 556–573. Morozov, E. (2011). The Net Delusion: How Not to Liberate the World. London: Penguin. Mudde, C. and Kaltwasser, C. R. (2017). Populism: A Very Short Introduction. Oxford: Oxford University Press. Ofcom (2021). Online Nation. London: Ofcom. https://​www​.ofcom​.org​.uk/​_​_data/​assets/​pdf​ _file/​0013/​220414/​online​-nation​-2021​-report​.pdf. Pittman, M. and Tefertiller, A. C. (2015). With or without you: Connected viewing and co-viewing Twitter activity for traditional appointment and asynchronous broadcast television models. First Monday. https://​doi​.org/​10​.5210/​fm​.v20i7​.5935. Statistica (2021). Social Media & User-Generated Content. https://​www​.statista​.com/​markets/​ 424/​topic/​540/​social​-media​-user​-generated​-content/​#overview. Vaccari, C., Chadwick, A., and O’Loughlin, B. (2015). Dual screening the political: Media events, social media, and citizen engagement. Journal of Communication, 65(6), 1041–1061. Vaccari, C. and Valeriani, A. (2018). Dual screening, public service broadcasting, and political participation in eight Western democracies. International Journal of Press/Politics, 23(3), 367–388. Vergeer, M. and Franses, P. H. (2016). Live audience responses to live televised election debates: Time series analysis of issue salience and party salience on audience behavior. Information, Communication & Society, 19(10), 1390–1410. Waldner, D. and Lust, E. (2018). Unwelcome change: Coming to terms with democratic backsliding. Annual Review of Political Science, 21, 93–113. Wallsten, K. (2014). Microblogging and the news: Political elites and the ultimate retweet. In IGI Global, Political Campaigning in the Information Age (pp. 128–147). Pennsylvania: IGI Global. Wells, C., Shah, D. V., Pevehouse, J. C., Yang, J., Pelled, A., Boehm, F., … Schmidt, J. L. (2016a). How Trump drove coverage to the nomination: Hybrid media campaigning. Political Communication, 33(4), 669–676. Wells, C., Van Thomme, J., Maurer, P., Hanna, A., Pevehouse, J., Shah, D. V., and Bucy, E. (2016b). Coproduction or cooptation? Real-time spin and social media response during the 2012 French and US presidential debates. French Politics, 14(2), 206–233.

11. Gen Z’s civic engagement: news use, politics, and cultural engagement Ava Francesca Battocchio, Chris Wells, Emily Vraga, Kjerstin Thorson, Stephanie Edgerly and Leticia Bode

INTRODUCTION For as long as there has been a study of digital politics, young citizens have occupied a special place in it. Why? Two significant reasons stand out. First, young people early on were recognized as “digital natives”, a term meant to capture something special about the relationship between youth and digital media (Prensky, 2001): a supposed electronic sixth sense to explain aptitudes for videocassette recorder (VCR) programming in the 1980s, web surfing in the 1990s, and social media use in the twenty-first century. For scholars of political engagement, one early assumption was that previously disengaged youth might be reached with a preferred medium and so brought back to civic life. Originating as it did with perceptions of older people as less inclined to use technology, the concept of digital native has limitations in describing the practices and abilities of a large and diverse group. Yet the fact that the generation entering adulthood today has come of age wholly immersed in an environment of often rapidly shifting digital media landscapes is one worth paying attention to – although not simply because of their facility with technology. A second reason for interest in the political uses to which young people have put digital media is high-profile content creation examples. Youth remain at the forefront of experimenting with digital communication in political life. Millennials were the first generation of digital natives to break down divisions between digital communication and politics that pushed the limits of how we understood digital technology and social media usage around the 2010s. Where millennials had prominent examples of pop culture engagement with politics (e.g., “Crush on Obama”), Gen Z had a revolving door of musicians and memes in the 2020 election (Postema, 2020). Young people have been vital in protest politics, such as the DREAMer protests against US anti-immigration policies, Global Black Lives Matter actions for racial justice, and Hong Kong’s anti-extradition protests. These activities are not unique to Gen Z: political protest has often emerged among students and other young activists. However, between 2016 and 2020, protest participation increased 5 percent (CIRCLE, 2021). A further reason for the importance of following young citizens as their political involvements via digital media take shape is the uniqueness of the global youth generation. In the US and much of the developed world, Gen Z are markedly tolerant and liberal; but many are overburdened with student loan debt, inflation, increasing 168

Gen Z’s civic engagement: news use, politics, and cultural engagement  169

living costs, and economic precarity (Parker and Igielnik, 2020; Race and Timmins, 2022). Moreover, inequality is growing between those in higher and lower social classes, features of late modern societies generally but often felt most acutely at the youth level (Dawson, 2021). The emerging generation faces significant challenges, and using digital media to engage with public life will surely be at the heart of how they approach those challenges. The young people who were the subjects of the first articles about the fascinating nexus of young citizens, digital media and politics are now approaching middle age. Even as Internet time churns ahead rapidly enough that “generations” of distinct digital experience turn over every couple of years, we take this chapter as an opportunity to look back on two decades of research on the possibilities that young people will increase their engagement in politics through or because of digital media. In this chapter, we train our lens on four areas of special interest when it comes to young citizens and digital politics: ● the changing bases of the civic identities of citizens in industrialized democracies; ● how younger generations are consuming – or not consuming – news; ● the practices and patterns emerging as formal political campaigns attempt to reach young people through digital media; and ● how content creation and interaction in digital media enables, for some young citizens, a form of cultural engagement that pushes the boundaries of the political. To complement this discussion, we will expand upon the role of socio-economic status and social identity, online practices, socialization and civic education in Chapter 12. Before proceeding, a note on what is meant by “young” citizens: the literature on youth civic engagement is inconsistent in its definitions of the precise boundaries of youth. However, for the purposes of both chapters, we are comfortable adopting the term “Gen Z” to refer to members born from 1997 and onward (Dimock, 2019).

SITUATING THE DIGITAL CITIZEN IN LATE MODERN SOCIETY Millennials are often considered novel in their merging of technology and political engagement (see previous edition), in part due to their status as the earliest “digital natives”. However, our increasing immersion in a technological society means that the norm, rather than the exception, is for up-and-coming youth to have no residual memory of a predominantly analogue world. Gen Z has continued the tradition of digital innovation that millennials began, but with more frequent and far-reaching digital media use (Auxier and Anderson, 2021). Several important social-structural changes should inform our understanding of young people’s civic identities, relationships to politics, and use of digital media to those ends. First, the period preceding and encompassing the childhoods of con-

170  Handbook of digital politics

temporary young people was one of marked economic change. Led by processes of economic globalization, this period saw the completion of the interpenetration of national economies, the rising power of transnational corporations, and the decline of unions; and correspondingly, a faltering working-class way of life for many in the rust belts of the United States and other developed nations. Concurrent with economic globalization was a shift toward network structures of organization, in associational contexts from global capital and international finance to civil society and local communities (Castells, 1996). Paired with new economic stresses, these changes meant a constellation of pressures that undermined the group-based associational life of high modern society (Bennett, 1998). The resulting decline in participation in place-based, face-to-face community organizations (Putnam, 2000) means that many Gen Zers have had less exposure to the traditional interpersonal community interactions that were formative for older citizens. In their place, argue some scholars, is a networked individualism in which young people, in particular, are comfortable creating interest-based communities via online social networks (Rainie and Wellman, 2012). Parallels can be seen in the media structures younger citizens have grown up alongside. With cord-cutting most prevalent amongst young adults (Kim et al., 2021), television and radio are increasingly defined by subscription-based services like Netflix, Hulu, and Spotify (Tefertiller, 2018), creating an environment and expectation of on-demand media. Thus, whereas their parents and grandparents inhabited a world in which television watchers were simultaneously seeing the evening newscast, Gen Z is tethered to an ever-present panoply of media choices, to suit whatever particular interest (or ennui) grips them at the moment. All of this demonstrates that what is special about young people is much more than the ability to create TikTok videos. Compared to other generational cohorts, Gen Z is the most ethnically and racially diverse and on track to be the best-educated (Parker and Igielnik, 2020). Gen Z has grown up in a world in which potent labour unions, single-employer careers, long-term economic security (Parker and Igielnik, 2020), limited-channel media systems, and exclusively space-based communities were historical artifacts, not lived experiences. However, as of March 2020, those artifacts drifted even farther into the past. Young adults have suffered the combined misfortune of a poorly timed and prolonged pandemic, coinciding with key transitional education and career milestones. Though previously on track to avoid economic and employment issues experienced by millennials and Gen X, Gen Z has instead felt the greatest strain to education and career aspirations, interpersonal relationships, economic stability, and mental and physical health (American Psychological Association, 2021). While social media provided a social lifeline during lockdown, it also put young adults front and centre to an overwhelming amount of Covid-19 news while navigating public health-related misinformation (Islam et al., 2020). For some young adults, this combination of social isolation and Covid-related information overload triggered unhealthy coping mechanisms, like increased alcohol consumption (Mohr et al., 2021). Despite these struggles and setbacks, young adults are hopeful about

Gen Z’s civic engagement: news use, politics, and cultural engagement  171

post-pandemic recovery (American Psychological Association, 2021). Gen Z is resilient, perhaps because they have to be. With this unparalleled and precarious environment in mind, we interrogate how Gen Z makes sense of their civic life. How do these changes play out in young people’s actual experiences and activities as they engage the political world with digital media? How can we understand the participation of citizens experimenting with tools, who also had formative experiences – of community, of media, of civic life – quite unlike their elders? This attention to the larger contexts in which young citizens leverage digital media for political purposes yields many insights described below.

YOUTH AND NEWS If we wish to understand the implications of economic, social, and media changes for young citizens’ civic activity, news use is an excellent first place to look. The habits of news media consumption among today’s youth are dramatically different from that of previous generations at their age. While previous generations depended on traditional media, digital media are increasingly the primary means of accessing news. For example, in 1963, 79 percent of US high school seniors reported reading about politics in the newspaper more than three days a week, and 70 percent watched political news on television more than three days a week (data from Jennings et al., 2005, authors’ analysis). A 2021 Pew Research study found that only 20 percent of US 18to 29-year-olds get their news from print publications “at least sometimes”, while 45 percent watched news on television “at least sometimes” (Matsa and Naseer, 2021). News consumption as a habit of a particular time and place is being replaced by the digital media-enabled possibilities for news on demand. In 2021, 90 percent of American 18- to 29-year-olds reported using digital devices to get news “at least sometimes” (Matsa and Naseer, 2021). While all age groups are becoming increasingly dependent on digital news sources, especially in light of newspaper closures, young people rely more on social media for news than older cohorts. For example, 42 percent of American 18- to 29-year-olds report getting news from social media sites, compared to only 15 percent of 50- to 64-year-olds (Shearer, 2021). Though young adults know the importance of keeping up to date on news and politics, available sources and their impact on citizenship, daily practices and awareness about current events often do not reflect consistent news use (Swart and Broersma, 2022). However, this may be compounded by the issue that social media-delivered news is not a sufficient substitute for the breadth of general political news that traditional sources afford (Shehata and Strömbäck, 2021). Furthermore, there is little evidence that the use of the Internet and mobile technologies for news is becoming a regular habit of replacing the news routines of earlier generations (Edgerly et al., 2018a, 2018b, 2018c), with news being of low importance in the daily lives of young adults (Kümpel, 2020). Instead, youth seek out news content less and less and often rely on a “news finds me” attitude in which they encounter news incidentally while they are

172  Handbook of digital politics

online for other purposes (Boczkowski et al., 2018; de Zúñiga et al., 2017; Swart, 2021). The continued growth and introduction of new social media platforms mean that young adults construct diverse and personalized media repertoires that affect how they encounter news and politics (Edgerly et al., 2018b; Swart et al., 2017). Cross-cutting social media repertoires expose young adults to a range of potential news sources such as mainstream news outlets, alternative news sources, politicians and political activists, celebrities and influencers, and ordinary people (Cotter and Thorson, 2022; Walker and Matsa, 2021). Moving beyond Facebook, some young adults are beginning to incorporate regular news use into their everyday leisure routines on platforms such as Snapchat, TikTok, Instagram, Reddit, and Twitter (Walker and Matsa, 2021). News app push notifications are also a growing digital news source (Stroud et al., 2020). Though new avenues for accessing news is a positive development, the increasing role that platforms play in the political process, both in terms of how platform logics interact with users’ political interests (Thorson et al., 2021) and the increasing prominence of platforms in civic life (Thorson et al., 2020) is a cause for concern. Various observers have pointed out that algorithmic curatorial practices increasingly drive consumption of news media content. Personal preferences have heightened influence in shaping future content visibility, enabling those with a great deal of interest in keeping up with the world to gain greater knowledge, while those who do not enjoy the news may find it increasingly easier to avoid such content altogether (Aharoni et al., 2021; DeVito, 2017; Kümpel, 2020). Furthermore, the personal preferences of others on social media platforms can also shape news and political content visibility (Bode, 2016; Karnowski et al., 2017; Kümpel, 2019). While user preferences can shape the content that individuals are incidentally exposed to, often the result is very little public affairs content on young adults’ feeds (Wells and Thorson, 2017). Young adults are increasingly aware of both their role and the role of algorithms in shaping what they see (and what they don’t) – some attempting to trick the algorithms. This awareness leads to varying levels of trust and sentiment towards algorithmic curation (Swart, 2021). Research has shown that exposure to news content on sites like Facebook can lead to political learning (Bode, 2016), at least under some circumstances (Edgerly et al., 2018b), as well as electoral participation (Moeller et al., 2018). However, the extent to which such engagement occurs among younger citizens is not yet settled. Many individuals feel social pressures to avoid exchanging opinionated political content, particularly on Facebook, largely arising from the complexities of networked audiences (Thorson, 2014). As political tensions grow, social media as a platform for news has become increasingly associated with uncivil and draining political discussion and practices such as unfriending, blocking, and muting (Merten, 2021). Not surprisingly, polls regularly show younger citizens to be at the bottom of the heap regarding the standard measures of political knowledge and news quiz questions (Wattenberg, 2020). Curiously, in the face of most classic accounts of citizen engagement in politics, which posit learning about politics through news as

Gen Z’s civic engagement: news use, politics, and cultural engagement  173

a necessary precursor to action, some research is unearthing evidence of decoupling of knowledge and action. For example, Dimitrova and colleagues (2014) found regarding democratic functionality that social media was relatively low in its ability to generate informational opportunities for voters but generated higher levels of political involvement, connection and mobilization in comparison. The inverse was true of online news sites. More research is needed to consider whether the role of knowledge in civic engagement is indeed changing as the nature of the media system, and young citizens’ inclinations evolve.

FORMAL POLITICS This brings us to the question of how digital technologies impact young people’s participation rates. In the United States and other Western countries, the most elemental act of formal political participation has long been the vote. Historically, young adult turnout has been quite low compared to other age cohorts. For example, during the 2020 US presidential election, 18- to 29-year-olds comprised the smallest percentage of voters at only 16.5 percent of the population (Fabina and Scherer, 2022). Moreover, this trend has endured since 1988 (Fabina and Scherer, 2022), suggesting that it is a characteristic of a life process stage more than any particular generation. Not surprisingly, the low – and, through 2016, declining – level of youth participation in formal politics has long been a source of great concern, a state of affairs that had many observers prepared to hope for a turnaround prompted by new communication technologies. The integration of digital media strategies has become increasingly part and parcel of campaign efforts (Stromer-Galley, 2019). However, regulating political advertising and electoral politics is often a tug of war between corporate policies and democratic processes, with political consulting mirroring many paid client relationship practices in non-political digital advertising scenarios (Barrett, 2022; Kreiss and Mcgregor, 2019). While platforms are cause for political concern, they are also a source of political creativity, blending popular culture and political media, precipitating the growth of political fan cultures. While politicians may benefit from campaign exposure and subsequent opportunities for political talk, amongst youth target audiences campaigns run the risk of user-generated content taking on a life of their own, as was the case in the US with Bernie Sanders’ 2016 digital campaign. Sanders’ campaign mirrored that of Obama in merging traditional, grassroots outreach and online strategies. However, the Connect with Bernie app was both an information and mobilization platform, which collated information from the campaign’s official social media accounts (Facebook, Twitter, YouTube, Instagram, and Tumblr) with the ability to share to users’ social media profiles and compensate for other candidates’ extensive mainstream media platforms. Lastly, political campaign strategies extended into virtual worlds in 2020, employing pop culture fandoms to identify audiences who may be excluded from more tradi-

174  Handbook of digital politics

tional social media campaigning efforts, particularly gamers. For example, the Biden campaign, taking advantage of the video game Animal Crossing: New Horizons’ pandemic popularity, paid for virtual lawn signs, in addition to rolling out a “Build Back Better” map in the online game Fortnite (Roose, 2020). Politicians also used Twitch’s game-streaming platform to connect with young adults. For example, Bernie Sanders and Joe Biden used the platform to extend more traditional campaign tactics, like fireside chats and live streaming speeches. In contrast, Alexandria Ocasio-Cortez (D-New York) promoted early voting, answered viewer questions, and discussed issues with other politicians, as did guest Ilhan Omar (D-Minnesota), during gameplay (D’Anastasio, 2020). Technology integration into all facets of everyday life is now normal, expected, and uneventful. So the bigger question is, if the novelty of platforms as part of political campaigns has worn off, what might be contributing to the uptick in youth political participation? One possible answer to the recent uptick in civic participation lies in the ability of technology to help facilitate political cross-talk amongst youth. Leading up to the 2020 US election, 51 percent of 18- to 29-year-olds reported trying to persuade other youth to vote, and 70 percent noted that they had discussed political issues and the elections with their friends (CIRCLE, 2021). For some young adults, social media were spaces to attempt to counteract candidates’ potentially impactful rhetoric, such as responding to Trump’s anti-immigration rhetoric. Weighing in on social media in response to politicians as a means of social correction is seen as a duty (Penney, 2019). However, rates of youth political expression on platforms may be constrained by impression management and anonymity (Lane, 2020a). While it seems that digital media mobilizing youth may be an oversimplification, the reality is that young people are finding new ways to talk about news and politics that impact their lives, whether through public expression or direct messaging. In addition, platform affordances provide young people with different means of engagement with some spaces more conducive to including softer voices in the conversation. See Chapter 12 for further discussion.

CULTURAL ENGAGEMENT Digital media enable an assortment of activities that exist at the boundary of what is commonly accepted as “the political”. Foremost in testing those boundaries has been work on what has been described as content production, user-generated content, engagement with participatory culture, cultural engagement, or interactivity. New digital platforms have put the tools of participatory culture (Jenkins, 2006) in the hands of more youth (Poell et al., 2019; van Dijck et al., 2018). Similarly, creative multimedia tools available on mobile devices and social networking have changed our understanding of youth political engagement and the expressive citizen (Lane, 2020b; Literat and Kligler-Vilenchik, 2019). Platforms provide youth with various engagement options, from creating a public TikTok video to privately sharing

Gen Z’s civic engagement: news use, politics, and cultural engagement  175

a political post in closed WhatsApp groups. That said, it is not “better” technology that drives participation, but rather cultural practices that inspire individuals to use technology for acts of participation (Jenkins, 2006). Thus, a key civic benefit of participatory cultures is the potential for youth to develop skill sets that extend beyond personal expression to include the social skills needed to interact with a larger community. Fandoms are participatory cultures in which members hone communication skills and coordinate with others to accomplish goals. Fandom as a mechanism for youth political participation and mass mobilization has exploded since earlier fandom mashups of popular culture and social issues (e.g., Harry Potter Alliance). One recent example is K-pop, idol culture-oriented Korean popular music where social media savvy young adults hone their creativity and skills to support their favourite group. In the past, K-pop fandoms have connected with select philanthropic initiatives and social justice commentary (see Diaz Pino, 2021). One example of K-pop’s mobilization power responded to Donald Trump’s free registration for a 2020 campaign rally with limited attendance due to social distancing. Young adults created content to promote fake event registration to generate low event turnout and block interested Trump supporters from attending, expanding beyond K-pop circles, particularly on TikTok (Bandy and Diakopoulos, 2020). K-pop fans also mobilized around human rights violations, state-sanctioned violence and police brutality. During the racial justice protests following the murder of George Floyd, the Dallas Police Department promoted the launch of a crime tip app on Twitter, encouraging the public to submit videos of illegal activities. Subsequently, the Dallas PD was bombarded with user-generated content of K-pop idols in the Twitter comments, obscuring any legitimate content, and crashing the app. The BTS Army (the fandom of K-pop Superstars BTS) also staged a strategic takeover of #WhiteLivesMatter and #AllLivesMatter, and other allegedly white supremacist hashtags (Kanozia and Ganghariya, 2021). Youth online engagement has also been demonstrated through civic gaming. Research in this area focuses on digital games’ potential to foster the learning and practice of civic skills. This is an up-and-coming line of investigation given that a 2021 survey in the US found 76 percent of individuals under the age of 18 reported playing some type of video game (Entertainment Software Association, 2021). Jenkins (2006) argues that gaming cultures can provide individuals with the core experiences of play, simulation, and performance, which are precursors to participation. Games require young people to make decisions, communicate their ideas effectively, and in many cases, work with multiple players to achieve an end goal. Coordinated political actions are not a pandemic-related advent in the gaming world, as they often provide safety from state-sanctioned and reprisal violence. However, during the pandemic, popular video games such as Animal Crossing: New Horizons (ACNH), Grand Theft Auto, World of Warcraft, and The Sims, increasingly resembled traditional acts of political participation, becoming the sites of virtual memorialization acts and protests in response to police brutality and the murders of George Floyd and Breonna Taylor in addition to pro-Hong Kong democracy activism. ACNH

176  Handbook of digital politics

protests were often paired with the streaming platform Twitch to increase reach, participation, and fundraising efforts for Black Lives Matter (Ismangil and Lee, 2021; Schofield, 2020). Pokémon GO players in Hong Kong planned and promoted meetup events used as a cover for anti-extradition and anti-authoritarian protests. Players also adopted Pokémon imagery to communicate safety guidelines and messaging strategies for protesters ahead of these events (Ismangil and Lee, 2021). Examples such as these, well beyond the conventional territory of civic engagement, point to the ongoing need for conceptual and empirical work to understand the involvement of young people through digital channels.

CONCLUSION We hesitate to speculate about the future of Internet time. However, perhaps the maturation of both the first digital generation and our first generation of research on digital media allows us to imagine ourselves near the “end of the beginning” of youth digital politics. That is, though we are far from coherent answers to many of our questions about the implications of digital media on the political world (in a time of change as rapid as ours, giving any such answers would be foolhardy), we are accumulating bodies of evidence on several critical issues. What is the relationship between digital politics and youth civic engagement? Our reading of the evidence leads us to paraphrase Kranzberg’s first law of technology (Kranzberg, 1986): digital media are sometimes good for youth engagement, sometimes bad, and almost always complicated.

FURTHER READING Aharoni, T., Kligler-Vilenchik, N., and Tenenboim-Weinblatt, K. (2021). “Be less of a slave to the news”: A texto-material perspective on news avoidance among young adults. Journalism Studies, 22(1), 42–59. Andersen, K., Ohme, J., Bjarnøe, C., Bordacconi, M. J., Albæk, E., and De Vreese, C. H. (2021). Generational Gaps in Political Media Use and Civic Engagement: From Baby Boomers to Generation Z. New York: Routledge. Cotter, K. and Thorson, K. (2022). Judging value in a time of information cacophony: Young adults, social media, and the messiness of do-it-yourself expertise. International Journal of Press/Politics, 27(3), 629–647. Kanozia, R. and Ganghariya, G. (2021). More than K-pop fans: BTS fandom and activism amid COVID-19 outbreak. Media Asia, 48(4), 338–345. Merten, L. (2021). Block, hide or follow: Personal news curation practices on social media. Digital Journalism, 9(8), 1018–1039. Swart, J. (2021). Tactics of news literacy: How young people access, evaluate, and engage with news on social media. New Media & Society, 25(3). https://​doi​.org/​10​.1177/​ 14614448211011447.

Gen Z’s civic engagement: news use, politics, and cultural engagement  177

REFERENCES Aharoni, T., Kligler-Vilenchik, N., and Tenenboim-Weinblatt, K. (2021). “Be less of a slave to the news”: A texto-material perspective on news avoidance among young adults. Journalism Studies, 22(1), 42–59. American Psychological Association (2021). Stress in America™ 2021: Stress and decision-making during the pandemic. https://​www​.apa​.org/​news/​press/​releases/​stress/​ 2021/​decision​-making​-october​-2021​.pdf. Auxier, B. and Anderson, M. (2021). Social Media Use in 2021. Pew Research Center. https://​ www​.pewresearch​.org/​internet/​2021/​04/​07/​social​-media​-use​-in​-2021/​. Bandy, J. and Diakopoulos, N. (2020). #TulsaFlop: A case study of algorithmically-influenced collective action on TikTok (arXiv:2012.07716). arXiv. https://​ doi​ .org/​ 10​ .48550/​ arXiv​ .2012​.07716. Barrett, B. (2022). Commercial companies in party networks: Digital advertising firms in US elections from 2006–2016. Political Communication, 39(2), 147–165. Bennett, W. L. (1998). The uncivic culture: Communication, identity, and the rise of lifestyle politics. PS: Political Science and Politics, 31(4), 741–761. Boczkowski, P. J., Mitchelstein, E., and Matassi, M. (2018). “News comes across when I’m in a moment of leisure”: Understanding the practices of incidental news consumption on social media. New Media & Society, 20(10), 3523–3539. Bode, L. (2016). Political news in the news feed: Learning politics from social media. Mass Communication and Society, 19(1), 24–48. Castells, M. (1996). The Rise of the Network Society. Malden, MA: Blackwell Publishers. CIRCLE (2021). 2020 Election Center. CIRCLE, Tufts University. https://​circle​.tufts​.edu/​ 2020​-election​-center. Cotter, K. and Thorson, K. (2022). Judging value in a time of information cacophony: Young adults, social media, and the messiness of do-it-yourself expertise. International Journal of Press/Politics, 27(3), 629–647. D’Anastasio, C. (2020). Alexandria Ocasio-Cortez storms Twitch. Wired, October 20. https://​ www​.wired​.com/​story/​aoc​-among​-us​-twitch​-stream/​. Dawson, B. (2021). The State of America’s Children® 2021. The Children’s Defense Fund. https://​www​.childrensdefense​.org/​wp​-content/​uploads/​2021/​04/​The​-State​-of​-Americas​ -Children​-2021​.pdf. de Zúñiga, H. G., Weeks, B. E., and Ardèvol-Abreu, A. (2017). Effects of the news-finds-me perception in communication: Social media use implications for news seeking and learning about politics. Journal of Computer-Mediated Communication, 22(3), 105–123. DeVito, M. A. (2017). From editors to algorithms. Digital Journalism, 5(6), 753–773. Diaz Pino, C. (2021). “K-pop is rupturing Chilean society”: Fighting with globalized objects in localized conflicts. Communication, Culture & Critique, 14(4), 551–567. Dimitrova, D. V., Shehata, A., Strömbäck, J., and Nord, L. W. (2014). The effects of digital media on political knowledge and participation in election campaigns: Evidence from panel data. Communication Research, 41(1), 95–118. Dimock, M. (2019). Defining Generations: Where Millennials End and Generation Z Begins. Pew Research Center. https://​www​.pewresearch​.org/​fact​-tank/​2019/​01/​17/​where​ -millennials​-end​-and​-generation​-z​-begins/​. Edgerly, S., Thorson, K., Thorson, E., Vraga, E. K., and Bode, L. (2018a). Do parents still model news consumption? Socializing news use among adolescents in a multi-device world. New Media & Society, 20(4), 1263–1281. Edgerly, S., Thorson, K., and Wells, C. (2018b). Young citizens, social media, and the dynamics of political learning in the US presidential primary election. American Behavioral Scientist, 62(8), 1042–1060.

178  Handbook of digital politics

Edgerly, S., Vraga, E. K., Bode, L., Thorson, K., and Thorson, E. (2018c). New media, new relationship to participation? A closer look at youth news repertoires and political participation. Journalism & Mass Communication Quarterly, 95(1), 192–212. Entertainment Software Association (2021). 2021 Essential Facts About the Video Game Industry. Entertainment Software Association. https://​www​.theesa​.com/​resource/​2021​ -essential​-facts​-about​-the​-video​-game​-industry/​. Fabina, J. and Scherer, Z. (2022). Voting and Registration in the Election of November 2020. Current Population Survey Reports, P20-585, US Census Bureau, Washington, DC. Islam, A. K. M. N., Laato, S., Talukder, S., and Sutinen, E. (2020). Misinformation sharing and social media fatigue during COVID-19: An affordance and cognitive load perspective. Technological Forecasting and Social Change, 159, 120201. https://​doi​.org/​10​.1016/​j​ .techfore​.2020​.120201. Ismangil, M. and Lee, M. (2021). Protests in Hong Kong during the Covid-19 pandemic. Crime, Media, Culture, 17(1), 17–20. Jenkins, H. (2006). Convergence Culture: Where Old and New Media Collide. New York: New York University Press. Jennings, M. K., Markus, G. B., Niemi, R. G., and Stoker, L. (2005). Youth–parent socialization panel study, 1965–1997: Four waves combined. Ann Arbor, MI: Inter-University Consortium for Political and Social Research. https://​doi​.org/​10​.3886/​ICPSR04037​.v1. Kanozia, R. and Ganghariya, G. (2021). More than K-pop fans: BTS fandom and activism amid COVID-19 outbreak. Media Asia, 48(4), 338–345. Karnowski, V., Kümpel, A. S., Leonhard, L., and Leiner, D. J. (2017). From incidental news exposure to news engagement. How perceptions of the news post and news usage patterns influence engagement with news articles encountered on Facebook. Computers in Human Behavior, 76, 42–50. Kim, H., Chan-Olmsted, S. M., Hwang, K.-H., and Chang, B.-H. (2021). Examining the use, perception, and motivation of cord-cutting: A consumer segment approach. Journalism & Mass Communication Quarterly, 98(1), 126–147. Kranzberg, M. (1986). Technology and history: “Kranzberg’s Laws”. Technology and Culture, 27(3), 544–560. Kreiss, D. and Mcgregor, S. C. (2019). The “arbiters of what our voters see”: Facebook and Google’s struggle with policy, process, and enforcement around political advertising. Political Communication, 36(4), 499–522. Kümpel, A. S. (2019). The issue takes it all? Incidental news exposure and news engagement on Facebook. Digital Journalism, 7(2), 165–186. Kümpel, A. S. (2020). Nebenbei, mobil und ohne Ziel? Eine Mehrmethodenstudie zu Nachrichtennutzung und -verständnis von jungen Erwachsenen. Medien & Kommunikationswissenschaft, 68(1–2), 11–31. Lane, D. S. (2020a). In search of the expressive citizen: Citizenship norms and youth political expression on social media. Public Opinion Quarterly, 84(S1), 257–283. Lane, D. S. (2020b). Social media design for youth political expression: Testing the roles of identifiability and geo-boundedness. New Media & Society, 22(8), 1394–1413. Literat, I. and Kligler-Vilenchik, N. (2019). Youth collective political expression on social media: The role of affordances and memetic dimensions for voicing political views. New Media & Society, 21(9), 1988–2009. Matsa, K. E. and Naseer, S. (2021). News Platform Fact Sheet. Pew Research Center. https://​ www​.pewresearch​.org/​journalism/​fact​-sheet/​news​-platform​-fact​-sheet/​. Merten, L. (2021). Block, hide or follow: Personal news curation practices on social media. Digital Journalism, 9(8), 1018–1039. Moeller, J., Kühne, R., and De Vreese, C. (2018). Mobilizing youth in the 21st century: How digital media use fosters civic duty, information efficacy, and political participation. Journal of Broadcasting & Electronic Media, 62(3), 445–460.

Gen Z’s civic engagement: news use, politics, and cultural engagement  179

Mohr, C. D., Umemoto, S. K., Rounds, T. W., Bouleh, P., and Arpin, S. N. (2021). Drinking to cope in the COVID-19 era: An investigation among college students. Journal of Studies on Alcohol and Drugs, 82(2), 178–187. Parker, K. and Igielnik, R. (2020). What We Know About Gen Z So Far. Pew Research Center. https://​www​.pewresearch​.org/​social​-trends/​2020/​05/​14/​on​-the​-cusp​-of​-adulthood​ -and​-facing​-an​-uncertain​-future​-what​-we​-know​-about​-gen​-z​-so​-far​-2/​. Penney, J. (2019). It’s my duty to be like ‘this is wrong’: Youth political social media practices in the Trump era. Journal of Computer-Mediated Communication, 24(6), 319–334. Poell, T., Nieborg, D., and van Dijck, J. (2019). Platformisation. Internet Policy Review, 8(4), 1–13. Postema, S. (2020). How pop culture encourages political participation. EUPHORIA, September 17. https://​www​.euphoriazine​.com/​blog/​2020/​09/​culture​-how​-pop​-culture​ -encourages​-political​-participation/​. Prensky, M. (2001). Digital natives, digital immigrants. On the Horizon, 9(5), 1–6. Putnam, R. (2000). Bowling Alone: The Collapse and Revival of American Community. New York: Simon & Schuster. Race, M. and Timmins, B. (2022). Gen Z on battling soaring inflation for the first time. BBC News, January 19. https://​www​.bbc​.com/​news/​business​-60024716. Rainie, L. and Wellman, B. (2012). Networked: The New Social Operating System. Cambridge, MA: MIT Press. Roose, K. (2020). How Joe Biden’s digital team tamed the MAGA Internet. The New York Times, December 6. https://​www​.nytimes​.com/​2020/​12/​06/​technology/​joe​-biden​-internet​ -election​.html. Schofield, D. (2020). Black Lives Matter meets Animal Crossing: How protesters take their activism into video games. The Guardian, August 7. https://​www​.theguardian​.com/​games/​ 2020/​aug/​07/​black​-lives​-matter​-meets​-animal​-crossing​-how​-protesters​-take​-their​-activism​ -into​-video​-games. Shearer, E. (2021). 86% of Americans Get News Online from Smartphone, Computer or Tablet. Pew Research Center. https://​www​.pewresearch​.org/​fact​-tank/​2021/​01/​12/​more​ -than​-eight​-in​-ten​-americans​-get​-news​-from​-digital​-devices/​. Shehata, A. and Strömbäck, J. (2021). Learning political news from social media: Network media logic and current affairs news learning in a high-choice media environment. Communication Research, 48(1), 125–147. Stromer-Galley, J. (2019). Presidential Campaigning in the Internet Age. New York: Oxford University Press. Stroud, N. J., Peacock, C., and Curry, A. L. (2020). The effects of mobile push notifications on news consumption and learning. Digital Journalism, 8(1), 32–48. Swart, J. (2021). Tactics of news literacy: How young people access, evaluate, and engage with news on social media. New Media & Society, 25(3). https://​doi​.org/​10​.1177/​ 14614448211011447. Swart, J. and Broersma, M. (2022). The trust gap: Young people’s tactics for assessing the reliability of political news. International Journal of Press/Politics, 27(2), 396–416. Swart, J., Peters, C., and Broersma, M. (2017). Navigating cross-media news use: Media repertoires and the value of news in everyday life. Journalism Studies, 18(11), 1343–1362. Tefertiller, A. (2018). Media substitution in cable cord-cutting: The adoption of web-streaming television. Journal of Broadcasting & Electronic Media, 62(3), 390–407. Thorson, K. (2014). Facing an uncertain reception: Young citizens and political interaction on Facebook. Information, Communication & Society, 17(2), 203–216. Thorson, K., Cotter, K., Medeiros, M., and Pak, C. (2021). Algorithmic inference, political interest, and exposure to news and politics on Facebook. Information, Communication & Society, 24(2), 183–200.

180  Handbook of digital politics

Thorson, K., Medeiros, M., Cotter, K., Chen, Y., Rodgers, C., Bae, A., and Baykaldi, S. (2020). Platform civics: Facebook in the local information infrastructure. Digital Journalism, 8(10), 1231–1257. van Dijck, J., Poell, T., and de Waal, M. (2018). The Platform Society: Public Values in a Connective World. New York: Oxford University Press. Walker, M. and Matsa, K. E. (2021). News Consumption Across Social Media in 2021. Pew Research Center. https://​www​.pewresearch​.org/​journalism/​2021/​09/​20/​news​-consumption​ -across​-social​-media​-in​-2021/​. Wattenberg, M. P. (2020). Is Voting for Young People? (5th edition). New York: Routledge. Wells, C. and Thorson, K. (2017). Combining big data and survey techniques to model effects of political content flows in Facebook. Social Science Computer Review, 35(1), 33–52.

12. Gen Z’s civic engagement: civic skills, political expression, and identity Ava Francesca Battocchio, Leticia Bode, Chris Wells, Emily Vraga, Kjerstin Thorson and Stephanie Edgerly

INTRODUCTION Gen Zers around the world have grown up and come of age in a period rife with the implications of climate change, heightened right-wing extremism, threats to democracy, and rising inflation. Gen Z has also been characterized by an enhanced awareness of mental health care and body positivity, LGBTQIA+ and the spectrum of gender identity, as well as racial justice, diversity, equity and inclusion. The previous chapter on youth political engagement focused on digital media’s role in shaping engagement modes. In this chapter, we explore shifts in the basis for civic identity for many young people. Drawing on sociologists such as Giddens (1991) and Beck (1999), some scholars contend that because of their different experiences, new generations of citizens are embracing “lifestyle” politics (Bennett, 1998) or “rights-bearing” citizenship (Schudson, 1998). These perspectives share a view of citizenship that entails a decreased experience of duty and obligation, decreased identification with and trust in parties and official leaders, and decreased inclination to participate in organized, bounded protests. In place of these old norms are rising demands for expression, individuality, personalization, and flexibility in the acting out of civic identity, which may take the form of acts that can be practiced on a daily, lifestyle basis, such as becoming a vegetarian or making (at least occasional) conscious consumer choices, or non-political “community” participation, such as volunteering (Zukin et al., 2006). From this perspective, both changes in civic participation and digital media uses are seen as products of young people’s situatedness in a changing civic order and the particular technologies available (and developing) at that time (Wells, 2013). In this chapter, we focus on three areas of particular interest when it comes to young citizens and digital politics: ● the role of socio-economic status and affinity groups; ● skills, online practices, and the definition of political engagement; ● what these changes imply for the study of political socialization and the practice of civic education. A note before we continue – in this chapter, we intend to draw attention to how skills, identity, and socialization uniquely shape youth participation and practices. 181

182  Handbook of digital politics

Global youth are not monolithic, so why should we assume their civic engagement and online practices would be? Our discussion is by no means exhaustive, and while we include some international examples, our lens is primarily trained on American youth.

SOCIO-ECONOMIC STATUS AND AFFINITY GROUPS Scholars of youth engagement have generally been moving away from searches for the direct effects of digital media, instead turning toward more nuanced perspectives in which technology is seen as one of many factors influencing an individual or group’s likelihood of participating. The emerging consensus is, “it depends”. But the factors on which it depends are coming into better focus. One factor on which the impact of digital media on youth engagement depends is socio-economic status (SES). The study of how SES impacts youth engagement has deep implications, foremost among them the question of whether digital media significantly change the makeup of who becomes engaged. SES has long been a primary predictor of political participation. However, on a positive note, digital media seems to level the playing field in terms of political expression in ways that traditional offline political participation does not (Lane et al., 2023). Systemic disenfranchisement and marginalization has received increased attention in recent years through digital media presentations of violence against Black and Brown individuals at the hands of the police, mounting Islamophobia and anti-Asian violence, growing anti-immigration sentiment, and threats to reproductive rights. Unsurprisingly, counter-narratives and movements have also experienced heightened discourse. It seems that every social concern has a related hashtag or two. While discursive opportunities are perhaps some of the more visible examples of youth civic engagement associated with race and ethnicity, there are several ways that digital media influences behaviours both on and offline. Black youth are relatively more engaged in community and political action rather than activism, with political efficacy and social responsibility being key for civic engagement (Hope, 2016). Furthermore, engagement with critical reflection on social inequality often shapes participation in voting and socio-political action amongst Latinx and Black youth (Bañales et al., 2020). Overall, Black and Latinx youth seem to engage in more digital acts of political expression than their white peers (Lane et al., 2023). Often motivated by social responsibility, combating injustice, and generating social change, undocumented and other immigrant-origin youth employ non-formal mechanisms of civic engagement, many of which are heavily dependent on digital media. For example, some youth utilize technology to network, participate in and lead community organizations or provide varying translation work of civic information (Suárez-Orozco et al., 2015). Overall, social media plays the part of amplification within a larger transmedia strategy that scaffolds youth-led, local-level civil disobedience (Zimmerman, 2016).

Gen Z’s civic engagement: civic skills, political expression, and identity  183

It is difficult to determine the true extent of voting participation amongst Indigenous youth in the US or Canada, often compounded by access and trust issues (Canadian Heritage, 2021; CIRCLE, 2021). What is more apparent is that Indigenous youth, particularly First Nations, Métis, and Inuit youth in Canada, are more likely to engage in non-electoral political activities than electoral activities (Canadian Heritage, 2021). One area that sets Indigenous content creators and social movements apart is the degree to which culture, beliefs, and language shape political discourse and practices (Raynauld et al., 2018). This is particularly evident on “Native TikTok” (Cole, 2021) and other platforms through content on topics such as climate activism and water governance, cultural genocide and the Indian Residential School System, land ownership restoration, language revitalization following racialized policies (Meighan, 2021), and Missing and Murdered Indigenous Women. In recent years, social media have often been venues to raise issues that disproportionately impact women, such as reproductive rights and sexual assault and harassment, and intimate partner violence, mainly through the use of hashtag-driven campaigns (e.g., #MeToo, #TimesUp). For example, the recent social media campaign #MeTooK12 (Lu, 2018) demonstrates that youth are shifting focus from issues that centre adult voices to those of youth. Despite the visible uptick in digital political mobilization, there is a representative gender gap in online political participation. One potential cause is incivility, particularly in that female and transgender individuals are often targeted in online harassment and tend to tolerate rather than respond directly (Haslop et al., 2021). In turn, visible uncivil behaviour towards women in political discourse may influence the willingness of other women to engage politically online (Koc-Michalska et al., 2021). However, there is an ongoing debate about whether perceptions of incivility or political socialization drive how frequently women engage in visible political discourse (cf. Bode, 2017; Van Duyn et al., 2021). Discrimination against sexual minorities can generate political participation. American queer college students are twice as likely as their non-queer peers to engage in more radical political non-electoral activity such as protests, rallies, and marches (Swank and Fahs, 2017). While some queer youth are more comfortable engaging in radical politics both on and offline, others search for safe spaces for socialization. Tumblr, a microblogging and social networking platform, is a popular, digital enclave amongst LGBTQIA2S+ youth, providing shelter from some of the more discriminatory corners of the Internet, enabling the exploration of gender, sexuality, and progressive politics (Cavalcante, 2019; Lucero, 2017). Fostering community is particularly important in that for transgender individuals, there is a reciprocal relationship between community connectedness and civic engagement, albeit the offline connection is stronger than online (Billard, 2022). For youth with disabilities, digital devices are a potent tool for daily life (Baumgartner et al., 2023), political inclusion, and political engagement (Trevisan, 2020). Before the pandemic’s influx of Zoom and other video-driven protests, virtual protests and additional online collective mobilization were long-established and integral to the disability rights movement (Trevisan, 2018). Youth with disabilities

184  Handbook of digital politics

are also tasked with identifying Assistive Technology and platform affordances that make civic information accessible. For instance, members of the American Sign Language (ASL) community reported using YouTube for dedicated ASL current affairs programming (Trevisan, 2020). Likewise, Facebook, Messenger, and WhatsApp groups are relatively accessible for socialization and information for students with visual impairments, though there was little political discussion (Della Líbera and Jurberg, 2020). In recent years, there has been a growing awareness of the spectrum of disability, both visible and non-visible, driven partly by digital media. Many of these individuals utilize hashtags (Sarkar et al., 2021) and personal storytelling that surface concerns and challenge stereotypes associated with their visible and invisible disabilities, including chronic illness and mental health (Lawson, 2021). For instance, Instagram and TikTok have given rise to active figures in the disability rights and inclusion movement. While this content can generate less-positive outcomes, like online harassment, creators can foster advocacy and community-building (Rauchberg, 2022). However, on platforms like TikTok, content that normalizes the lived experience of not only disabled but queer and trans creators is often disproportionately impacted by practices such as shadow banning and other forms of algorithmic suppression that limit the reach of their messaging (Köver and Reuter, 2019; Rauchberg, 2022). Many of these areas have seen little work that centres on youth. We hope that this section serves as a call to action for scholars to devote more resources and energy to investigating the intersection between digital media, politics, and various affinity groups and the impact on youth political participation.

SKILL AND ONLINE PRACTICES “Digital media” can refer to various platforms, devices, and uses, and some uses of digital media are more tightly linked to the emergence of political behaviour than others (Valenzuela et al., 2012). Young people tend to primarily use social media for non-political and entertainment content (Binder et al., 2021). However, digital media, mainly social media, may increase incidental exposure to political news and political knowledge, with potential implications for how youth engage civically (Boulianne and Theocharis, 2020), or avoid such engagement (Milhailidis, 2020). Furthermore, such exposure can be a critical way to get information about social and political issues that might not be covered in the curriculum into the hands of marginalized young adults while circumnavigating adults whom they may view as untrustworthy (Kaskazi and Kitzie, 2021). Regardless of whether it is incidental or intentional, digital news exposure requires that the user has technical skills for using the Internet and the literacy to understand how online platforms and information are structured. While there is debate about different types of literacy – for example, media, news, information, or digital literacies – the term literacy incorporates both knowledge of the environment and the skills to navigate it (Vraga et al., 2021b). For online environments, relevant literacies

Gen Z’s civic engagement: civic skills, political expression, and identity  185

have been theoretically (Vraga et al., 2021b) and empirically (Vraga and Tully, 2021) related to civic engagement and quality information recognition and consumption. However, the link between literacy and information consumption and civic engagement is not clear cut; there are other examples wherein greater literacy and/or educational efforts to boost literacy have failed (Vraga et al., 2021a). Indeed, literacy can sometimes create cynicism about the media environment and disengagement rather than healthy scepticism. Additionally, it is vital to consider demography’s role in shaping skill acquisition and subsequent political behavioural outcomes. Youth media research tends to focus on those from WEIRD settings (Western, Educated, Industrialized, Rich, and Democratic), often overlooking marginalized youth (Jordan and Prendella, 2019). Scholars argue that curriculum differences amongst these populations highlight the roles that different skill development plays in predicting youth civic engagement (Moon and Bai, 2020). For instance, among high-school-aged Korean students, media literacy skills are often the driving force behind civic engagement rather than technical ability to navigate technology. Understanding which types of knowledge and skills for which individuals can encourage civic engagement remains an important avenue for future research (Moon and Bai, 2020). Finally, much literacy work is limited to educational settings. But the emphasis on education, while important, leaves out the majority of the population who are not in school or whose school days are long in the past. New research has tested the success of short literacy interventions that can reach these overlooked groups, but has only shown mixed success (van der Meer and Hameleers, 2022). In order to develop a better understanding of the implications of skill acquisition for youth civic engagement, researchers should pay more attention to non-WEIRD and non-educational settings. It is also crucial to note that interest in politics continues to play a significant role in whether a young person’s digital media use leads them to news consumption (Boulianne and Shehata, 2021), or to become politically engaged (Levy and Akiva, 2019). Ultimately, younger citizens’ lower interest in politics will always remain a barrier to equal engagement, insurmountable by any degree of digital media innovation (Bode et al., 2017). Earlier research highlighted the connection between non-political, interest-driven online practices and political engagement, giving optimism to the idea that increased digital engagement would translate into civic engagement. However, scholars are increasingly of the position that non-political online practices do not necessarily convert to political mobilization (Matthes, 2022; though see Lee et al., 2020). There is also ongoing conversation on the role of platform design in shaping youth political expression. Perceived platform affordances can create an environment or “civic laboratory” where young adults feel that they can safely explore political expression (Lane et al., 2019). Platforms that allow for anonymity result in a higher likelihood of expressing political opinions because of decreased concerns about political self-presentation (Lane, 2020) and social risk (Lane et al., 2019). While platform affordances can shield youth from real and imagined communities, they

186  Handbook of digital politics

can also provide access to a shared community of those with similar interests and beliefs, cultivating an environment ripe for collective political expression (Literat and Kligler-Vilenchik, 2019). As such, both individual and collective political expression can take on many digital forms (Literat and Klingler-Vilenchik, 2019; Penney, 2019), such as resharing a news article, creating a unique post or using a hashtag. Furthermore, young adults also tend to engage in online civic engagement and content creation practices that centre on self-identity-related “restorying”, community building, and collective action (Wilf and Wray-Lake, 2021). In addition to creating their own content, young adults seem to be increasingly aware of the importance of uplifting and amplifying historically marginalized voices in digital spaces without further commentary. This is particularly visible when looking at the frequency in which BLM content is retweeted versus quote-tweeted by young adults on Twitter, compared to other age groups (Shugars et al., 2021). Other types of political expression that draw on social and political issues might be more creative, visual expressions. One example is when a digital cultural artifact located on one platform can generate political expression both within and across several social media platforms. This was the case with a virtual representation of Trump’s border wall in the video game Fortnite (Literat and Kligler-Vilenchik, 2021). Although memes demonstrate a questionable amount of political knowledge (McLoughlin and Southern, 2021), they are still perceived by some as an act of political participation in their creation (McLoughlin and Southern, 2021). While pop culture artifacts as catalysts for political expression are perhaps a novel, though limited, approach to generating increased political knowledge and participation, there is also a dark side. The humorous nature of memes often helps to cultivate a political aesthetic and obscure their potential as entry points to the socialization of alt-right, white supremacist, and other extremist ideologies amongst youth (DeCook, 2018, 2020). These varied forms of expression and participation bring us to perhaps one of the most active and contentious areas of debate concerning young citizens and engagement – drawing the lines around what constitutes engagement and the legitimacy of activities with varying participation modalities. Clearly, a discussion of only news consumption and formal political participation (see Chapter 11) is no longer adequate to describe youth uses of digital media to participate in public life, and discussions over the levels of youth engagement depend increasingly on one’s definition of engagement. No longer is it sufficient to examine youth rates of voting, contacting public officials, contributing money, and following conventional news: young people now inhabit a political communication sphere in which their options are much more varied and include a host of opportunities to learn, share, and express ideas on topics. Virtual participation in traditional in-person activities has become more normalized during the pandemic. However, common terms to describe online-only engagement (e.g., sharing or liking political content, online petitions, boycotting or purchasing products that support a cause), such as “Slacktivism”, “clicktivism”, and “armchair activism”, are still frequently pejorative.

Gen Z’s civic engagement: civic skills, political expression, and identity  187

Moreover, the debate often extends to whether or not online participation translates into offline participation or vice versa (Boulianne and Theorcharis, 2020). In a UK-based study, Leyva (2017) found that frequent use of and exposure to politics on social media had a positive but weak relationship with offline engagement, both formal and activist. In contrast, there was a strong association between use and exposure with online “slacktivism”. Overall though, there seems to be a growing consensus that many young adults are engaging in “hybrid activism” as there is a strong relationship between youth engagement in online and offline civic activities (Boulianne and Theocharis, 2020). Lastly, it is worth noting that online-centered practices are essential in their own right, regardless of their connection to offline behaviours (Lane and Cin, 2018). Virtual participation allows engagement that might not otherwise exist offline, especially for those fearing reprisal from authoritarian governments (Otiono, 2021) and youth dealing with accessibility issues. As we continue as a society to grapple with the ongoing pandemic as a “mass disabling event, in addition to state-sanctioned violence and police brutality” (Pomeroy, 2021), we must recognize that for many young people, especially those from historically marginalized populations, political engagement is often finely balanced with personal safety. Online engagement offers options for engagement that may not easily translate to offline activity, yet add to the growing importance and strength of youth support of social movements, filling a gap where there may not be a participatory contribution otherwise.

POLITICAL SOCIALIZATION AND CIVIC EDUCATION One area working especially hard to come to terms with these changes is that of political socialization and civic education. Socialization scholars have moved away from the transmission model of socialization, which focused almost entirely on the role of parents in fostering political attitudes (Niemi and Jennings, 1991). Instead, recent work recognizes the diversity of influences shaping young adults’ involvement in the political process and developing their political identity (Thorson et al., 2018). Similarly, emergent research highlights the wide diversity of ways parents and children can co-orient themselves towards the political process, moving beyond studies of transmission versus trickle-up socialization. Intra-family dynamics are more complex than simple transmission, and there are even cases in which youth seem to “socialize” their parents (Shehata and Amnå, 2019). Although research has long suggested that mass media contribute to socialization by focusing attention on the political process and intersecting with parental, classroom, and peer discussions on the topic, the more complex media and online environment require scholars to differentiate between forms of online engagement. Within this field, particular attention has been paid to youth engagement on social networking sites, which often promote civic and political engagement (Heiss et al., 2020), and lead youth to diverge from their parents in political activism. This important online engagement may have several roots: the influence of peer discussion on polit-

188  Handbook of digital politics

ical and citizenship identities (Wegemer, 2022), the impact of supportive network ties on political participation (Maher and Earl, 2019; Terriquez et al., 2020), and the development of norms and “citizenship vocabularies” regarding appropriate political behaviour and action (Bergan et al., 2022) and the ability to engage in self-expression and identity formation (Lane et al., 2019; Literat and Kligler-Vilenchik, 2021). This more complex media environment has also contributed to a renewed interest in civic education’s role in socialization. While earlier research found little impact of school on political socialization, more recent evidence suggests that strong civic curricula that encourage children to actively debate and discuss issues in class can play a vital role in socialization processes, particularly in conjunction with media exposure (Pontes et al., 2019). Similarly, researchers are increasingly emphasizing that schools can play an important role in helping youth develop the skills they need to interpret online information correctly (Bowyer and Kahne, 2020), as well as encouraging youth to both seek out and value exposure to diverse perspectives (Tully and Vraga, 2018), long recognized as important to promoting tolerance in the back-and-forth conflict inherent in the democratic process (Siegel-Stechler, 2021). Despite social media’s constant presence, in-school civic education continues to strongly influence youth political engagement (Ohme et al., 2020). Perhaps in response to youth digital dependence, gamification of civic participation (Hassan and Hamari, 2020) and interactive school media programming (Geers et al., 2020) are increasingly making their way into civic education. Together, scholars have begun to develop a fairly robust measure of the forces expected to contribute to youth socialization into civic and political life. But despite the recognition that studying socialization is important because the orientations developed during youth tend to endure throughout the life cycle and shape engagement with politics (Thorson, 2012), scholars differ on a fundamental question: when are these orientations actually established? Are some ideological preferences genetically inherited, while others are socialized over time (Wajzer and Dragan, 2021)? Do children develop stable partisan attitudes aged 5 to 8 years, before they enter grade school (van Deth et al., 2011), or is adolescence the key time to observe changes in partisan identity (Rekker et al., 2017)? Does political interest stabilise by early adulthood (Russo and Stattin, 2017)? Does socialization occur gradually, in a linear fashion, as youth recognize their place in the political process, or does it occur in fits and starts during campaigns, building on the agenda-setting potential for the mass media to make political discussion salient (Kiousis et al., 2005)? Or perhaps the bigger question for the future is: for a generation that is inundated with so many pressing social issues, what role do events such as Black Lives Matter protests, school shootings, the pandemic, amongst many others play as socializing events for young people?

Gen Z’s civic engagement: civic skills, political expression, and identity  189

CONCLUSION One of the most exciting things about research on young people’s uses of digital media for political purposes is how often they, and the technologies they use, defy our expectations: it is more complicated than that. What we found time and again in our survey of the literature is that, first, almost all the “logical” assumptions about youth uses of digital media depend on a host of factors; and second, fully understanding this field will always be impossible without a rich appreciation for the social-political contexts young people inhabit. However, this is why we do research: to get to grips with these complexities and render them into something that occasionally resembles understanding. We have made the case that the field has made a good start in understanding how young citizens are situated to engage in digital politics. Nevertheless, there is much more research – and surprises, surely – ahead.

FURTHER READING Andersen, K., Ohme, J., Bjarnøe, C., Bordacconi, M. J., Albæk, E., and De Vreese, C. H. (2021). Generational Gaps in Political Media Use and Civic Engagement: From Baby Boomers to Generation Z. New York: Routledge. Lane, D. S., Das, V., and Hiaeshutter-Rice, D. (2019). Civic laboratories: Youth political expression in anonymous, ephemeral, geo-bounded social media. Information, Communication & Society, 22(14), 2171–2186. Literat, I. and Kligler-Vilenchik, N. (2021). How popular culture prompts youth collective political expression and cross-cutting political talk on social media: A cross-platform analysis. Social Media + Society, 7(2), 20563051211008820. https://​doi​.org/​10/​gn2r48. Matthes, J. (2022). Social media and the political engagement of young adults: Between mobilization and distraction. Online Media and Global Communication. https://​doi​.org/​10​ .1515/​omgc​-2022​-0006. Rauchberg, J. S. (2022). #Shadowbanned: Queer, trans, and disabled creator responses to algorithmic oppression on TikTok. In P. Pain (ed.), LGBTQ Digital Cultures: A Global Perspective. New York: Routledge. Raynauld, V., Richez, E., and Boudreau Morris, K. (2018). Canada is# IdleNoMore: Exploring dynamics of Indigenous political and civic protest in the Twitterverse. Information, Communication & Society, 21(4), 626–642.

REFERENCES Bañales, J., Mathews, C., Hayat, N., Anyiwo, N., and Diemer, M. A. (2020). Latinx and Black young adults’ pathways to civic/political engagement. Cultural Diversity and Ethnic Minority Psychology, 26(2), 176–188. Baumgartner, A., Rohrbach, T., and Schönhagen, P. (2023). ‘If the phone were broken, I’d be screwed’: Media use of people with disabilities in the digital era. Disability and Society, 38(1), 73–97. Beck, U. (1999). World Risk Society. Cambridge: Polity Press. Bennett, W. L. (1998). The uncivic culture: Communication, identity, and the rise of lifestyle politics. PS: Political Science and Politics, 31(4), 741–761.

190  Handbook of digital politics

Bergan, D. E., Carnahan, D., Lajevardi, N., Medeiros, M., Reckhow, S., and Thorson, K. (2022). Promoting the youth vote: The role of informational cues and social pressure. Political Behavior, 44, 2027–2047. Billard, T. J. (2022). Together we rise: The role of communication and community connectedness in transgender citizens’ civic engagement in the United States. Mass Communication and Society, 25(3), 335–360. Binder, A., Heiss, R., Matthes, J., and Sander, D. (2021). Dealigned but mobilized? Insights from a citizen science study on youth political engagement, Journal of Youth Studies, 24(2), 232–249. Bode, L. (2017). Closing the gap: Gender parity in political engagement on social media, Information, Communication & Society, 20(4), 587–603. Bode, L., Vraga, E. K., and Troller-Renfree, S. (2017). Skipping politics: Measuring avoidance of political content in social media. Research and Politics, 4(2), 2053168017702990. https://​doi​.org/​10​.1177/​2053168017702990. Boulianne, S. and Shehata, A. (2021). Age differences in online news consumption and online political expression in the United States, United Kingdom, and France. International Journal of Press/Politics, 27(3), 19401612211060271. Boulianne, S. and Theocharis, Y. (2020). Young people, digital media, and engagement: A meta-analysis of research. Social Science Computer Review, 38(2), 111–127. Bowyer, B. and Kahne, J. (2020). The digital dimensions of civic education: Assessing the effects of learning opportunities. Journal of Applied Developmental Psychology, 69, 101162. http://​dx​.doi​.org/​10​.1016/​j​.appdev​.2020​.101162. Canadian Heritage (2021). Canada’s First State of Youth Report: For Youth, with Youth, by Youth. Government of Canada. https://​epe​.lac​-bac​.gc​.ca/​100/​201/​301/​weekly​ _acquisitions​_list​-ef/​2021/​21​-35/​publications​.gc​.ca/​collections/​collection​_2021/​pch/​CH4​ -193​-2021​-eng​.pdf. Cavalcante, A. (2019). Tumbling into queer utopias and vortexes: Experiences of LGBTQ social media users on Tumblr. Journal of Homosexuality, 66(12), 1715–1735. CIRCLE (2021). 2020 Election Center. CIRCLE, Tufts University. https://​circle​.tufts​.edu/​ 2020​-election​-center. https://​circle​.tufts​.edu/​latest​-research/​2020​-youth​-voter​-turnout​ -raceethnicity​-and​-gender. Cole, K. (2021). How Indigenous creators are nurturing a space on TikTok to educate and entertain. PBS NewsHour, October 1. https://​www​.pbs​.org/​newshour/​arts/​how​-indigenous​ -creators​-are​-nurturing​-a​-space​-on​-tiktok​-to​-educate​-and​-entertain. DeCook, J. R. (2018). Memes and symbolic violence: #Proudboys and the use of memes for propaganda and the construction of collective identity. Learning, Media and Technology, 43(4), 485–504. DeCook, J. R. (2020). Trust me, I’m trolling: Irony and the alt-right’s political aesthetic. M/C Journal, 23(3). http://​dx​.doi​.org/​10​.5204/​mcj​.1655. Della Líbera, B. and Jurberg, C. (2020). Communities of practice on WhatsApp: A tool for promoting citizenship among students with visual impairments. British Journal of Visual Impairment, 38(1), 58–78. Geers, S., Boukes, M., and Moeller, J. (2020). Bridging the gap? The impact of a media literacy educational intervention on news media literacy, political knowledge, political efficacy among lower-educated youth. Journal of Media Literacy Education, 12(2), 41–53. Giddens, A. (1991). Modernity and Self-Identity: Self and Society in the Late Modern Age. Stanford, CA: Stanford University Press. Haslop, C., O’Rourke, F., and Southern, R. (2021). #NoSnowflakes: The toleration of harassment and an emergent gender-related digital divide, in a UK student online culture. Convergence, 27(5), 1418–1438.

Gen Z’s civic engagement: civic skills, political expression, and identity  191

Hassan, L. and Hamari, J. (2020). Gameful civic engagement: A review of the literature on gamification of e-participation. Government Information Quarterly, 37(3), 101461. https://​ doi​.org/​10​.1016/​j​.giq​.2020​.101461. Heiss, R., Knoll, J., and Matthes, J. (2020). Pathways to political (dis-)engagement: Motivations behind social media use and the role of incidental and intentional exposure modes in adolescents’ political engagement. Communications, 45(s1), 671–693. Hope, E. C. (2016). Preparing to participate: The role of youth social responsibility and political efficacy on civic engagement for black early adolescents. Child Indicators Research, 9(3), 609–630. Jordan, A. and Prendella, K. (2019). The invisible children of media research. Journal of Children and Media, 13(2), 235–240. Kaskazi, A. and Kitzie, V. (2021). Engagement at the margins: Investigating how marginalized teens use digital media for political participation. New Media & Society, 25(1), 72–94. Kiousis, S., McDevitt, M. and Wu, X. (2005). The genesis of civic awareness: Agenda setting in political socialization. Journal of Communication, 55, 756–774. Koc-Michalska, K., Schiffrin, A., Lopez, A., Boulianne, S., and Bimber, B. (2021). From online political posting to mansplaining: The gender gap and social media in political discussion. Social Science Computer Review, 39(2), 197–210. Köver, C. and Reuter, M. (2019). TikTok curbed reach for people with disabilities. https://​ netzpolitik​.org/​2019/​discrimination​-tiktok​-curbed​-reach​-for​-people​-with​-disabilities/​. Lane, D. S. (2020). Social media design for youth political expression: Testing the roles of identifiability and geo-boundedness. New Media & Society, 22(8), 1394–1413. Lane, D. S. and Cin, S. D. (2018). Sharing beyond slacktivism: The effect of socially observable prosocial media sharing on subsequent offline helping behaviour. Information, Communication & Society, 21(11), 1523–1540. Lane, D. S., Das, V., and Hiaeshutter-Rice, D. (2019). Civic laboratories: Youth political expression in anonymous, ephemeral, geo-bounded social media. Information, Communication & Society, 22(14), 2171–2186. Lane, D. S., Thorson, K., and Xu, Y. (2023). Uninterested and unequal? Examining SES-based gaps in youth political behavior on social media. Information, Communication & Society, 26(4), 663–681. Lawson, M. (2021). Disabled creators on TikTok are going beyond “inspiration porn”. Allure, January 9. https://​www​.allure​.com/​story/​disabled​-creators​-on​-tiktok. Lee, S. S., Lane, D. S., and Kwak, N. (2020). When social media get political: How perceptions of open-mindedness influence political expression on Facebook. Social Media + Society, 6(2). https://​doi​.org/​10​.1177/​2056305120919382. Levy, B. L. M. and Akiva, T. (2019). Motivating political participation among youth: An analysis of factors related to adolescents’ political engagement. Political Psychology, 40(5), 1039–1055. Leyva, R. (2017). Exploring UK millennials’ social media consumption patterns and participation in elections, activism, and “slacktivism”. Social Science Computer Review, 35(4), 462–479. Literat, I. and Kligler-Vilenchik, N. (2019). Youth collective political expression on social media: The role of affordances and memetic dimensions for voicing political views. New Media & Society, 21(9), 1988–2009. Literat, I. and Kligler-Vilenchik, N. (2021). How popular culture prompts youth collective political expression and cross-cutting political talk on social media: A cross-platform analysis. Social Media + Society, 7(2), 20563051211008820. https://​doi​.org/​10/​gn2r48. Lu, W. (2018). What #MeToo means to teenagers. The New York Times, April 19. https://​ www​.nytimes​.com/​2018/​04/​19/​well/​family/​metoo​-me​-too​-teenagers​-teens​-adolescents​ -high​-school​.html.

192  Handbook of digital politics

Lucero, L. (2017). Safe spaces in online places: Social media and LGBTQ youth. Multicultural Education Review, 9(2), 117–128. Maher, T. V. and Earl, J. (2019). Barrier or booster? Digital media, social networks, and youth micromobilization. Sociological Perspectives, 62(6), 865–883. Matthes, J. (2022). Social media and the political engagement of young adults: Between mobilization and distraction. Online Media and Global Communication. https://​doi​.org/​10​ .1515/​omgc​-2022​-0006. McLoughlin, L. and Southern, R. (2021). By any memes necessary? Small political acts, incidental exposure and memes during the 2017 UK general election. The British Journal of Politics and International Relations, 23(1), 60–84. Meighan, P. J. (2021). Decolonizing the digital landscape: The role of technology in Indigenous language revitalization. AlterNative: An International Journal of Indigenous Peoples, 17(3), 397–405. Mihailidis, P. (2020). The civic potential of memes and hashtags in the lives of young people. Discourse: Studies in the Cultural Politics of Education, 41(5), 762–781. Moon, S. J. and Bai, S. Y. (2020). Components of digital literacy as predictors of youth civic engagement and the role of social media news attention: The case of Korea. Journal of Children and Media, 14(4), 458–474. Niemi, R. G. and Jennings, M. K. (1991). Issues and inheritance in the formation of party identification. American Political Science Review, 35, 970–988. Ohme, J., Marquart, F., and Kristensen, L. M. (2020). School lessons, social media and political events in a get-out-the-vote campaign: Successful drivers of political engagement among youth? Journal of Youth Studies, 23(7), 886–908. Otiono, N. (2021). Dream delayed or dream betrayed: Politics, youth agency and the mobile revolution in Africa. Canadian Journal of African Studies / Revue Canadienne des Études Africaines, 55(1), 121–140. Penney, J. (2019). It’s my duty to be like ‘this is wrong’: Youth political social media practices in the Trump era. Journal of Computer-Mediated Communication, 24(6), 319–334. Pomeroy, C. (2021). A tsunami of disability is coming as a result of “long COVID”. Scientific American, July 6. https://​www​.scientificamerican​.com/​article/​a​-tsunami​-of​-disability​-is​ -coming​-as​-a​-result​-of​-lsquo​-long​-covid​-rsquo/​. Pontes, A. I., Henn, M., and Griffiths, M. D. (2019). Youth political (dis)engagement and the need for citizenship education: Encouraging young people’s civic and political participation through the curriculum. Education, Citizenship and Social Justice, 14(1), 3–21. Rauchberg, J. S. (2022). #Shadowbanned: Queer, trans, and disabled creator responses to algorithmic oppression on TikTok. In P. Pain (ed.), LGBTQ Digital Cultures: A Global Perspective. New York: Routledge. Raynauld, V., Richez, E., and Boudreau Morris, K. (2018). Canada is# IdleNoMore: Exploring dynamics of Indigenous political and civic protest in the Twitterverse. Information, Communication & Society, 21(4), 626–642. Rekker, R., Keijsers, L., Branje, S., and Meeus, W. (2017). The dynamics of political identity and issue attitudes in adolescence and early adulthood. Electoral Studies, 46, 101–111. Russo, S. and Stattin, H. (2017). Stability and change in youths’ political interest. Social Indicators Research, 132(2), 643–658. Sarkar, T., Forber-Pratt, A.J., Hanebutt, R., and Cohen, M. (2021). “Good morning, Twitter! What are you doing today to support the voice of people with #disability?”: Disability and digital organizing. Journal of Community Practice, 29(3), 299–318. Schudson, M. (1998). The Good Citizen: A History of American Civic Life. New York: Martin Kessler Books. Shehata, A. and Amnå, E. (2019). The development of political interest among adolescents: A communication mediation approach using five waves of panel data. Communication Research, 46(8), 1055–1077.

Gen Z’s civic engagement: civic skills, political expression, and identity  193

Shugars, S., Gitomer, A., McCabe, S., Gallagher, R. J., Joseph, K., Grinberg, N., Doroshenko, L., Foucault Welles, B., and Lazer, D. (2021). Pandemics, protests, and publics: Demographic activity and engagement on Twitter in 2020. Journal of Quantitative Description: Digital Media, 1. https://​doi​.org/​10​.51685/​jqd​.2021​.002. Siegel-Stechler, K. (2021). Teaching for citizenship: Instructional practices and open classroom climate. Theory and Research in Social Education, 49(4), 570–601. Suárez-Orozco, C., Hernández, M. G., and Casanova, S. (2015). “It’s sort of my calling”: The civic engagement and social responsibility of Latino immigrant-origin young adults. Research in Human Development, 12(1–2), 84–99. Swank, E. and Fahs, B. (2017). College students, sexualities identities, and participation in political marches. Sexuality Research and Social Policy, 14(2), 122–132. Terriquez, V., Villegas, R., Villalobos, R., and Xu, J. (2020). The political socialization of Latinx youth in a conservative political context. Journal of Applied Developmental Psychology, 70, 101188. Thorson, K. (2012). What does it mean to be a good citizen? Citizenship vocabularies as resources for action. Annals of the American Academy of Political and Social Science, 644, 70–85. Thorson, K., Xu, Y., and Edgerly, S. (2018). Political inequalities start at home: Parents, children, and the socialization of civic infrastructure online. Political Communication, 35(2), 178–195. Trevisan, F. (2018). Connective action mechanisms in a time of political turmoil: Virtual disability rights protest at Donald Trump’s inauguration. Australian Journal of Political Science, 53(1), 103–115. Trevisan, F. (2020). “Do you want to be a well-informed citizen, or do you want to be sane?” Social media, disability, mental health, and political marginality. Social Media + Society, 6(1). https://​doi​.org/​10​.1177/​2056305120913909. Tully, M. and Vraga, E. K. (2018). Who experiences growth in news media literacy and why does it matter? Examining education, individual differences, and democratic outcomes. Journalism and Mass Communication Educator, 73(2), 167–181. Valenzuela, S., Arriagada, A., and Scherman, A. (2012). The social media basis of youth protest behavior: The case of Chile. Journal of Communication, 62(2), 299–314. van der Meer, T. G. and Hameleers, M. (2022). I knew it, the world is falling apart! Combatting a confirmatory negativity bias in audiences’ news selection through news media literacy interventions. Digital Journalism, 10(3), 473–492. van Deth, J. W., Abendschon, S., and Vollmar, M. (2011). Children and politics: An empirical reassessment of early political socialization. Political Psychology, 32(1), 147–174. Van Duyn, E., Peacock, C., and Stroud, N. J. (2021). The gender gap in online news comment sections. Social Science Computer Review, 39(2), 181–196. Vraga, E. K. and Tully, M. (2021). News literacy, social media behaviors, and skepticism toward information on social media. Information, Communication & Society, 24(2), 150–166. Vraga, E., Tully, M., and Bode, L. (2021a). Assessing the relative merits of news literacy and corrections in responding to misinformation on Twitter. New Media & Society, 24(10), 2354–2371. Vraga, E. K., Tully, M., Maksl, A., Craft, S., and Ashley, S. (2021b). Theorizing news literacy behaviors. Communication Theory, 31(1), 1–21. Wajzer, M. and Dragan, W. Ł. (2021). It is not only the environment that matters: A short introduction to research on the heritability of political attitudes. Political Studies Review, 21(1). https://​doi​.org/​10​.1177/​14789299211053780. Wegemer, C. M. (2022). Service, activism, and friendships in high school: A longitudinal social network analysis of peer influence and critical beliefs. Journal of Youth and Adolescence, 51, 1–15.

194  Handbook of digital politics

Wells, C. (2013). Two eras of civic information and the evolving relationship between civil society organizations and young citizens. New Media & Society, 16(4), 615–636. Wilf, S. and Wray-Lake, L. (2021). “That’s how revolutions happen”: Psychopolitical resistance in youth’s online civic engagement. Journal of Adolescent Research. https://​doi​.org/​ 10​.1177/​07435584211062121. Zimmerman, A. (2016). Transmedia testimonio: Examining undocumented youth’s political activism in the digital age. International Journal of Communication, 10, 1886–1906. Zukin, C., Keeter, S., Andolina, M., Jenkins, K., and Delli Carpini, M. X. (2006). A New Engagement: Political Participation, Civic Life, and the Changing American Citizen. New York: Oxford University Press.

PART III TECHNOLOGY AND PLATFORMS

13. Becoming eventful through data: the mediated construction of historic events in the age of data Heather Ford

When asked what sort of thing was most likely to blow governments off course, British Prime Minister Harold Macmillan famously replied, ‘Events, dear boy, events’. Newsweek journalist George F. Will wrote about Macmillan in 2002: ‘He would know. The 1956 Suez crisis had catapulted Macmillan into the role of prime minister, and it was the Profumo scandal1 that would usher him out’. But were the events themselves in charge? Were they, as Will wrote, ‘in the saddle, riding mankind’? Media theorists would disagree. Events on their own don’t affect politics. The mediation of significant political events influences the ways in which they are experienced, interpreted and remembered. Not all events qualify. Global iconic events (both ceremonial and unexpected) involve major ruptures: upending social and political identities, ideological commitments, political orientations and power. Such events are political not only because they involve members of the political class (e.g. snap elections or historic press conferences) but because their outcomes affect the distribution of power and authority. They can include violent protests, online activism, natural disasters and human-made catastrophes that become political because they have an impact on the distribution of attention and resources. Iconic global events are interesting sites of study for media and communication researchers because they enable us to understand the role of media and mediation in determining who are the winners and losers in battles over minds and resources. Mediation is central to the ways in which disruptive political events are experienced and the outcomes that result from them. How such events are framed and the narratives that drive their interpretation influence people’s shared experiences and memories. Mediation influences who are recognized as the victims and who the oppressors. These framings ultimately influence how events play out over time. In this chapter I discuss major contributions to media events theory to date and outline key areas of scholarly disagreement. I go on to articulate the need to move beyond the analysis of how a medium represents historic or significant events to an understanding of how the media (a) constructs an event as historic or significant and (b) in the context of a web increasingly dominated by the logics and materialities of data. Focusing on the need for the meaning of an event’s significance to be rapidly shared across communities, platforms, media and borders in order for it to gain historic status, I suggest methods from science and technology studies (STS) for 196

Becoming eventful through data: the mediated construction of historic events  197

achieving this understanding in the context of a hybrid media system (Chadwick, 2013). I conclude with a series of research questions that are ripe for the study of media events today. In my work on online knowledge construction in the context of the Egyptian revolution, I highlight the important role of platforms and media that host and share data about historic events as key signifiers (Ford, 2022). The web is increasingly governed by the logics and materialities of data. Data now plays a key role both in constructing certain events rather than others as significant and in classifying and labelling them in ways that might enable their travel. Online venues in which data is created, curated, shared and translated by both humans and algorithms serve as important sites where the struggle over the meaning and importance of an event intensifies.

MEDIA EVENTS: CONSENSUS AND DEBATE Interest in the media events phenomenon can be traced back to the earliest days of media research (e.g. see Cantril, 1940) but the phenomenon was brought to public consciousness with Daniel Dayan and Elihu Katz’s pathbreaking 1992 book, Media Events: The Live Broadcasting of History. Exploring the role of television in the live broadcasting of historic events like the Olympic Games, the wedding of Prince Charles and Lady Diana, and the funeral of John F. Kennedy, the book demonstrated the transforming power of the live television broadcast. Dayan and Katz argued that media technology (in this case broadcasting) had the potential to transform ‘not only a “message”, not only the nature of response, but an entire structure of social relations’ (Dayan and Katz, 1992, p. 217). Such media events were powerful, argued Dayan and Katz, because they served as unifying rituals. Broadcast on television to transfixed viewers around the world, historic media events enabled the collective experience that served to bind societies together. Media events were ‘centring’ performances, co-produced by broadcasters and organizers – the latter, according to Dayan and Katz, ‘typically’ included ‘public bodies with whom the media cooperate, such as governments, parliaments … political parties … international bodies … and the like’ (Dayan and Katz, 1992, p. 6). Focusing on events that were ‘preplanned, announced and advertised in advance’, Dayan and Katz defined media events as those presented ‘with reverence and ceremony’ (Dayan and Katz, 1992, p. 7, emphasis in original). Dayan and Katz’s Media Events was revolutionary because it articulated the ways in which the media were an important feature of social life. It identified the mediation of historic, ceremonial events, in particular, as a useful demonstration of the media’s importance. According to Paddy Scannell, ‘[t]he originality and genius of Dayan and Katz lie in having identified media events as a topic for serious academic study’ (2017, p. 153). And for Stuart M. Hoover, Dayan and Katz provided ‘a paradigm-shifting definition of an entirely new ground on which to understand the role of the media and of mediation’ (2010, p. 284). Understanding how mediation had

198  Handbook of digital politics

become so integral to societies and what were its effects became a popular project for media studies researchers on the back of Media Events. Today, media events have become one of the most discussed and referenced topics in media studies. There has been a 25th anniversary special edition of Media, Culture & Society in celebration of Dayan and Katz’s 1992 text and an edited book collection about the concept in the context of global media cultures edited by Couldry, Hepp and Krotz in 2009. Despite the longevity and popularity of Dayan and Katz’s encapsulation of ‘media events’, it is not without its critics. Andreas Hepp and Nick Couldry set out two key criticisms of Dayan and Katz’s work in order to rethink and extend what they nonetheless believe was ‘pathbreaking’ (Hepp and Couldry, 2010, p. 1). First, Hepp and Couldry (2010, p. 8) critique Dayan and Katz’s definition of media events as overly narrow. The ‘reverent and priestly’ style that Dayan and Katz defined as characteristic of the ways in which media events were presented was not a given, they argued. Dayan and Katz’s typification of media events as contests, conquests and coronations was also overly limiting – especially given the increasing prevalence of ‘disaster marathons’ (Liebes, 1998, p. 72) where the public domain is overtaken by oppositional rather than hegemonic forces and popular media events like the TV series Big Brother involving celebratory culture (Hepp and Couldry, 2010, p. 8). Out of this critique, Hepp and Couldry generate an expanded definition of media events as ‘situated, thickened, centering performances of mediated communication that are focused on a specific thematic core, cross different media products and reach a wide and diverse multiplicity of audiences and participants’ (2010, p. 12). Second, they disagree with Dayan and Katz’s claim that the effect of media events was to reaffirm shared values and restore order in order to hold society together. There are neither a common set of shared values nor evidence of a society that is stable – especially considering the fragmented, globalized societies in which we currently live. It is important not to assume ‘an integrative role of ritual media events’, Hepp and Couldry argued (2010, p. 5). In many cases, audiences actually bypass or reinterpret the intended centring that the organizers of media events prioritize. The construction of a ‘mediated centre’, therefore, remains ‘an uncertain and contested process, however totalizing the claims that such a construction involved’ (Hepp and Couldry, 2010, p. 13). The effects of media events, in other words, cannot be known in advance but depend on how media events are used to ‘establish certain discursive positions and to maintain those actors’ power’ (Hepp and Couldry, 2010, p. 12). Rather, argue Hepp and Couldry, we need to investigate historic events as ‘media rituals’ (Couldry, 2005) in a different sense. The outcomes of media events cannot be taken as a given or assumed in advance. In many cases, audiences circumvent or reinterpret the intended centring of media events. Media events are rituals that construct the myth of the mediated centre by articulating ‘the power-related, hegemonic imagination of the media as the centre of present societies, as the expression of the important incidents within that society’ (Hepp and Couldry, 2010, p. 5). The media, in other words, are so central that they don’t only represent important events but also actually construct events as important.

Becoming eventful through data: the mediated construction of historic events  199

How, then, do we understand the role of the massive variety of actors that work together to construct and represent historic, globally important events? Dayan and Katz presented media events such as the wedding of Charles and Diana as orchestrated by the crown in collaboration with broadcasters. They write that ‘these events are not organized by the broadcasters even if they are planned with television “in mind”. The media are asked, or ask, to join’ (Dayan and Katz, 1992, p. 6). Since then, there has been a recognition of the growing variety of actors involved in co-producing media events. Hepp and Couldry point out that in the current ‘globalized media cultures’ media events are ‘produced not only by the mass media (television, radio) but also by the Internet and other digital media’ (2010, p. 11). The ‘organizers’ of media events have also expanded. Rather than representatives of the state or the crown, event organizers include terrorists who use the media to hijack attention to highlight their cause. Unexpected events such as pandemics and natural disasters also introduce a range of institutional and civil society actors in their construction and representation. Scholars have recognized that the Internet has invited a number of new actors into the process of representing events. There is, however, debate about who the key actors are (and their relative influence) in the co-construction of media events. Previous accounts usually put a single actor in charge. Early accounts of the 2011 Egyptian revolution in the context of the Arab Spring were attributed to Twitter and Facebook with some calling this a ‘Facebook revolution’ (Smith, 2011). And yet there is always a contingency to the outcomes of events that require multiple actors to work together to propel particular narratives towards consensus.

THE MEDIATION OF EVENTS IN THE CONTEXT OF SOCIAL MEDIA AND DATA Theories relating to media events try to unpick the role of the media in the context of significant events in history that are mediated for global audiences. This is nowhere nearly as important as in our current age of mediatization (Hepp, 2019) in which everyday life and the events that punctuate it are steeped in media of multiple kinds. Global pandemics, natural disasters, protests, terrorist attacks, incidents and accidents are now represented by myriad actors on social and digital media platforms, often beginning minutes or hours after a bomb goes off, a shot is fired, a drone attacks, an earthquake hits, a pandemic is announced. Technologies of real-time connection have facilitated this immediacy. The first reports of the air raid that killed Osama Bin Laden in 2011 were made by an IT consultant living in Abbottabad, Pakistan when he unknowingly tweeted details of the US-led operation as it happened (BBC, 2011). When the Tōhoku earthquake hit Japan in 2011, it took just minutes after the six-minute earthquake stopped for Wikipedia editors to start an encyclopaedic article documenting the event and its aftermath (Keegan et al., 2013). Active audiences relay their experiences of events on social

200  Handbook of digital politics

media as immediate or remote witnesses. These messages are, in turn, re-presented in the traditional media as primary sources or a reflection of public opinion. Digital media has introduced new, influential actors in the catalysing of media events. The #MeToo movement, for example, began with sexual assault survivor and activist Tarana Burke using the phrase ‘Me Too’ in 2006 on Myspace (Ohlheiser, 2017). The SOPA (Stop Online Piracy Act) and PIPA (PROTECT IP act) blackout protests were constituted by a coordinated refusal of service by US digital media platforms on 18 January 2012 against the two proposed laws relating to intellectual property rights online. In these cases, online actors have been able to catalyse historic events before their coverage by traditional media. Whereas Dayan and Katz’s focus in the 1990s was on anticipated events (weddings, funerals, sporting events), today, unanticipated events seem to be the new normal. The defining characteristic of an unplanned event is that there is an information vacuum and a corresponding need from publics for information in order to ease the sense of dis-ease. Unexpected events are followed by a public search for meaning. Why did it happen? What is happening as a result? How can we ensure that it doesn’t happen again? As the events scholar Robin Wagner-Pacifici writes, ‘The public rides the shockwave of the flow of an event, twisting and turning to find a centre that will hold, or an authoritative voice that will define the situation, or a frame or a label that will settle the rupture into a set of predictable forms’ (Wagner-Pacifici, 2017, p. 12). Digital and social media have clearly changed the ways in which we experience and document historic events. Digital media has the power of ‘not only setting the agenda of transnational news outlets, but of creating transnational political events’ (Volkmer and Deffner, 2010, p. 218). Such events are transnational because they expand beyond national boundaries and reach the attention of those whose identities and cultures are implicated by them. But is has been difficult to tease out the role of social media as opposed to traditional event stakeholders, including traditional media, activists, governments, institutions and all the other actors that co-produce the historic event. The 2011 Egyptian revolution was the first major event to foreground the role of platforms in enabling social and political dissent and where the role of platforms has been heavily debated since. Some nominated social media as the lead actor in driving these events forward. There are now a host of accounts of how Facebook and Twitter proved instrumental for activists as an organizing tool for the daily protests that culminated in Hosni Mubarak’s resignation. On Twitter, Jared Cohen cited an Egyptian activist summing up the roles of social media in the revolution: ‘Facebook used to set the date, twitter used to share logistics, YouTube to show the world, all to connect people’. In the book Tweets and the Streets (2012), Paolo Gerbaudo argues that the role of social media in the wave of social movements that took place in the early part of this century is much more expansive than merely facilitating the practical operations (setting dates, sharing logistics and showing the world) that Cohen’s statement suggests. Social media is not limited to facilitating practical operations common to activism before the Internet, nor does social media result in ‘unrestrained participa-

Becoming eventful through data: the mediated construction of historic events  201

tion’ by all. Instead, platforms like Twitter, Facebook, and YouTube enable the symbolic construction of public space in which influential online activists can become ‘choreographers’ of collective action, argues Gerbaudo. Not all individuals have equal agency to effect change. Influential activists have used social media for ‘setting the scene, and constructing an emotional space within which collective action can unfold’ (Gerbaudo, 2012, p. 5). There are now several articulate accounts of the ways in which digital and social media have been used in political protest (see, for example, Bennett and Segerberg, 2012; Tufekci, 2017). And yet, their limitations lie in starting with a focus on (particular) social media platforms without being open to how the event itself is constructed and mediated by a variety of actors not usually recognized in these debates. Media events theory and methods are useful for understanding how multiple stakeholders catalyse and mediate the historic events that serve to reorganize social order. But the focus on particular types of media (Twitter, Facebook, broadcasting, etc.) has foregrounded certain media and ignored others that play a significant role in the ways that events are now constructed and represented.

OPERATIONALIZING ‘MEDIATED EVENTS’ Scholars like David Moats (2019) have proposed an alternative, ‘mediated events’ framework inspired by STS as a way of disclosing media technologies and practices at work in the mediation of an event. Moats believes that the disruptions from routine that are emblematic of mediated events enable privileged glimpses into how socio-technical assemblages work when they break down (Moats, 2019, p. 1166). They can help media scholars to break down dichotomies like producer/audience, social/technical, content/material that have been counterproductive for the field in the past. Instead of following a particular medium (e.g. television or social media), Moats encourages asking what media matter in a particular case. Drawing from Actor Network Theory (ANT), Moats describes a method that entails not deciding in advance what types of actors are consequential for a controversy’s settlement but rather to study the event and understand who and what actors are enrolled in its evolution. ‘Thus, non-human entities, technologies and nature are not interesting for their own sake but only to the extent that they matter for the stabilization of knowledge and society, and are worth foregrounding to the extent that their role has been underappreciated in the past’ (Moats, 2019, p. 1166). Following ANT, the goal of studying mediated events is to merely ‘describe the development of the controversy using participants’ own (multiple, provisional) articulations of it’ (Moats, 2019, p. 1166). Moats calls ‘mediated events’ a ‘radically empiricist’ interpretation of events ‘to emphasize that we are interested in how the event is mediated and translated while remaining agnostic about which sorts of actors are doing the mediating’ (Moats, 2019, p. 1172). Studying the event in this way enables us to answer important ques-

202  Handbook of digital politics

tions relating to the role of particular media in the evolution of the event and about how power is distributed within hybrid media systems. Such questions include: How does the reliance on particular sources favour certain versions of events? Who are the primary definers, and how do they get the upper hand in making their framings stick? What other actors are critical to constructing the event and/or issue? Moats’ approach is useful for scholars interested in teasing out the role of all actors in the hybrid media system that become actively engaged in the mediation of events. Governments, civil society organizations, international governmental institutions and crowds play an increasing role in the mediation of significant events, along with traditional media. It is best, then, not to decide up front what media are important for the study of media events but rather to follow the event to see which media matter in its construction. Such an approach can bring to light actors we usually do not recognize as important. The problem is that Moats doesn’t specify what to focus on in order to do justice to the myriad actors involved in representing iconic global events over time. Following the event as it is mediated across all media channels (including social media) quickly becomes overwhelming. Which representations does one focus on? Which actors across the millions that come to represent the event as they comment and report on it from multiple points around the world? I expand on Moats’ approach of following events as they are mediated by focusing on how key facts about the event stabilize over time. Boundaries between news media and historical knowledge produced about catalytic events were determined by norms of temporality and expertise. The first draft of history was the newspaper story, but news was always treated as a primary source and therefore subservient to the expert work of professional historians; a work of opinion versus knowledge. Today, we see the merging of news and history both in terms of temporality and expertise. Encyclopaedias like Wikipedia are no longer limiting themselves to the representation of events after they have occurred. Events are being documented as they take place (Keegan et al., 2013; Ford, 2022). And it is no longer only professional historians who produce knowledge about catalytic events that is deemed authoritative. Today, a host of new actors have become enrolled in developing knowledge that is elevated to the status of consensus truth rather than mere opinion or primary reportage.

FOLLOWING DATAFIED FACTS AS THEY TRAVEL Moats suggests following catalytic events that disrupt media routines in order to understand ‘how socio-technical assemblages like media work when they break down’ (Moats, 2019, p. 1166). I suggest three principles when analysing the mediation of catalytic events. First is to follow key facts about the event. Second is to focus on facts as they travel. Third is to pay close attention to locations (and practices) in which facts are translated across sites. In each case, we ask what media and actors

Becoming eventful through data: the mediated construction of historic events  203

matter to the stabilization of knowledge about the event in question. I expand on each below. I adopt Mary S. Morgan’s definition of facts as ‘autonomous, short, specific and reliable pieces of knowledge’ (Morgan, 2010, p. 7). Morgan writes that these characteristics define the statements that come in a ‘bewildering variety of forms’ (Morgan, 2010, p. 6) in a huge variety of fields. In the context of catalytic events, I define key facts as those that can answer a series of classificatory questions: What type of event is this? Who are its victims and perpetrators? What is the scale of their loss and success? These questions underly the facts that are created, debated, translated and made to travel between different sites. Facts that are born as data (or datafied when metadata is added to them) can more easily be transported across sites and are therefore particularly useful focal points. Following facts as they travel is key to understanding the role of mediation in the construction of meaning. Facts need to travel beyond the communities of practice in which they were originally created in order to reach consensus. Egyptians in Tahrir Square celebrating after the resignation of Hosni Mubarak on 11 February 2011 needed to have the knowledge that this event was a revolution translated into a fact that would travel beyond Egypt in order for it to become taken for granted as common knowledge. In order for facts about the event to attain consensus status, they need to be handed off by a variety of actors across different media. Mary Morgan identifies a fact’s companions, allies, labels and packaging as key actors that help facts to travel. Companions and allies are particular individuals, groups, or institutions that help them in their travels. As Morgan writes, ‘We depend upon systems, conventions, authorities, and all sorts of good companions to get facts to travel well – in various senses – and danger may lurk when these are subverted or fail to work’ (Morgan, 2010, p. 7). In the context of my Wikipedia work, I determine a fact’s allies as its authors (the Wikipedia editors who add, remove, and change facts as they work in small groups and large crowds) as well as non-human algorithms (or bots) that automatically select a fact and extract or copy it across databases where it is represented in different contexts. A fact’s travelling companions include the sources from which the fact originated. Given that facts’ travelling companions are often discarded along the way in the context of data systems, this step requires qualitatively analysing conversations among editors about what sources they value and quantitatively assessing which sources survive innumerable editing attempts. I also consider conventions that enable data from Wikipedia to travel to powerful knowledge platforms like the search engine (notably Google) and as an answer to a question in a question-answer machine such as Apple’s Siri or Amazon’s Alexa. In order to reach the status of fact, claims need to be structured according to the rules of linked data. Data constitutes a fact’s labels and packaging that explain its meaning to machines and provides instructions on how it should be translated to other sites across the web. Datafied facts will travel farther through the infrastructure of the Internet than facts that are merely expressed as HTML code. Data is rocket fuel for

204  Handbook of digital politics

facts because it enables machines to recognize and extract facts that can later be represented as answers to user queries. One of the most popular travel routes for datafied facts organized this way is between Wikipedia and search engines like Google, and increasingly by question-answer machines in virtual assistants such as Amazon’s Alexa, Apple’s Siri and Google’s Assistant. We ask what is happening in Ukraine, and our Google Assistant provides us with details. Current events are categorized as a ‘war’. Deaths are calculated (‘At least 46k’ when I searched on 17 May 2022). Property damage is estimated at ‘Approximately USD 600B’. The source is ‘Reuters’ updated ‘3 hours ago’. The facts that are highlighted in the right hand side of the page in Google search and those that are spoken by smart speakers are recognized by many as the consensus truth, delivered from on high by automated machines untainted by the mess and politics of human institutions. In the same way that Moats encourages asking what media (and actors) matter in a particular case, I traced how datafied facts about the 2011 Egyptian revolution crystallized and were made to travel across digital infrastructures in the days, weeks, months and years following the first protests on 25 January 2011 (Ford, 2022). Instead of all media in general, I chose to focus on a particular travel route that has not yet been considered in relation to the mediation of events: the travel of semantic data from the moments in which it datafied facts are created, to the ways in which they travel to big data pools such as Google’s knowledge graph. Semantic data is data organized according to principles that enable computers to interpret them without human intervention. They are usually organized according to two objects and their relationship. For example, if one wanted to represent the country Egypt existing within the Middle East, the data organization might look like this: EGYPT MIDDLE EAST. The objects (Egypt and Middle East) are interpreted with regard to their relationship (existing within). Google’s knowledge graph is perhaps one of the most important semantic databases in the world because it powers new features of search engines and responses to user queries from its smart speakers and digital assistants. Paying attention to semantic data and its travels enabled me to foreground actors (human and non-human actors) that were essential for the stabilization of knowledge about the revolution that had been underappreciated in the past. In this case, I followed how the semantic connection between the events in Egypt in January and February 2011 became indelibly linked to the concept of revolution and what it took to achieve this task. The final principle involves paying attention to the locations (and practices) in which facts are translated across sites. I followed facts as they were constructed, curated and edited, paying particular attention to moments and locations where facts are datafied for travel to other sites. In the path between Wikipedia and Google this includes locations such as Wikipedia’s infobox, Wikipedia article titles (and their alternatives), category tags, interwiki links, Wikidata items, Google knowledge panels, Google’s featured snippets and the ‘People also ask’ feature. It required following debates and arguments, in what may appear to be tedious technical discus-

Becoming eventful through data: the mediated construction of historic events  205

sions but are actually passionate debates about who will control the narrative of an historic event. During this following, I collected the activities and perspectives of actors who attached themselves to event data as it moved through Wikipedia (attended by sources on the web), to Wikidata (a new sister project to Wikipedia that houses data) and then to Google. I started this work with a single dominant representation – that of Google’s representation of the event and its answer to the question: ‘What happened in Egypt in 2011?’ I then moved to understand how the datafied facts highlighted by Google were originally created, starting with the first version of the English Wikipedia article from which Google is extracting its answers and featured snippets. These actors do not necessarily need to be human: they simply need to be consequential for ‘the stabilization of knowledge and society’ (Moats, 2019, p. 1168). This following of facts about events through Internet infrastructures revealed actors that are not usually recognized as important but that emerge as critical to what becomes accepted as common knowledge about what happened and why. I focus, in particular, on how consensus is arrived at in relation to the categorization of an event when there are multiple alternative options vying for favour. The result is a fine-grained analysis of how data mediates the production of knowledge about events and provides insights into the role of data and AI technologies in mediating our understanding of the world and of each other. Doing this work, I recognized the importance of data (produced by key actors and at key locations) that played a significant role in the ways in which events are constructed and archived. Doing so reveals a series of new actors that are coming to play an important role not only in the representation of events and their dominant meanings, but also in the construction of certain events as significant and exceptional in the context of the many others that are not accorded this status, despite the wishes of their stakeholders. I learned how sites like Wikipedia and Google are performing their representation of events in a totalizing way that characterizes the centring performativity of media events discussed by Hepp and Couldry.

CONCLUSION In 1992, it seemed impossible for historic events to have any public meaning without their transmission by television broadcasting. Today, it is impossible for historic events to have any public meaning without their transmission as data. In contrast to news reports, historic books, documentaries and other fact-based media, event data is a powerful representative force because it can travel most fruitfully to reach massive audiences and its prize location as the single answer to user queries on the web. Today, terrorist attacks, natural disasters, protests and pandemics – incidents and accidents – have been added to the weddings, funerals and sporting events that Dayan and Katz wrote about in 1992. The events that seem to arise on a daily basis to upend our societies are unexpected rather than expected. Rather than unifying societies through a single narrative arc, media events are characterized by a fracturing of

206  Handbook of digital politics

voices that compete to determine the meaning of what happened. In addition to television broadcasters, the live documentation of historic events as they happen is being conducted by myriad actors across digital platforms. The authority to determine the significance and meaning of the event, as well as its geographical scope, has been fundamentally altered. Despite these changes, there are few accounts that adequately capture the kinds of actors involved in both catalysing and representing events in the context of data and the semantic web. This is largely because, as Moats states, researchers tend to follow particular types of actors (e.g. Facebook, Twitter, ‘social media’, newspapers, etc.) rather than following the event itself and analysing how actors emerge as important mediators in the context of the event. The role of data that structures sources of multiple kinds, I argue, is yet to be fully explored and remains a key gap in terms of media events research. I identify three key research themes following from this. The first set of questions relates to actors and their political struggles over the meaning of events. Who (or what) is involved in constructing, structuring and sharing data about political events? How do they resolve their differences? Are they making decisions driven by particular ideologies, e.g. driven by governments or commerce, etc.? Following the event according to its datafied facts surfaces a range of new actors involved in attempting to shape the meaning of political events. Their goal is to influence event data, some of the most powerful event representations. The resultant struggles reflect new tactics of power and knowledge. Facts’ allies and companions have much to gain in either enabling facts to travel or stopping them dead in their tracks. Skirmishes over the meaning of events are ramped up by the forces that would have them represented as simplified (and often singular) data about events. The web has catalysed a host of data representations by journalists, citizens, civil society and state actors, via social media and digital platforms. The battle over event data has intensified. The second set of related questions involves the role of data specifically. What is the role of data in representing and catalysing media events? How are particular meanings of events encoded into data? How does data frame the narrative about what happened? The restructuring of the event according to data doesn’t only influence who might become involved in shaping the event. It also influences the meaning of the event as data is used to re-represent it. The ways in which datafied representations differ from purely digital ones reflect a particular kind of logic and understanding that influences the events they represent directly. Finding answers to these questions requires understanding how actors producing and using data structures explain their value. It requires understanding how data sharing and processing in particular contexts works to surface certain datafied facts rather than others. Data is constructed out of both social and technical or material relations. Finally, we need to understand human agency in the context of data-driven events. Does the wide access to data production means enable greater ability by publics to

Becoming eventful through data: the mediated construction of historic events  207

affect and effect political change? What is the role of automation and AI in determining how events are reflected and remembered? To many, the brute force of the algorithm in translating the meaning of events means that human history making has been ceded to automated technologies. ‘Where history was once written by its victors, and later by its nerds, it’s now being shaped by its algorithms’, wrote Washington Post journalist Caitlyn Dewey in an article about Google’s knowledge panels and their rapidly disappearing Wikipedia data sources (Dewey, 2019). But the event is produced by a number of (often opposing) forces, not by a single technology or platform. In my work on Wikipedia’s representation of events, I’ve learned that forces both internal and external to Wikipedia help shape data representations of events. Events in Egypt in January 2011 drove crowds to the Wikipedia article on waves of media attention. They influenced the market of sources available to Wikipedia editors. According to events scholar Robin Wagner-Pacifici, ‘Great things are at stake [in the representation of an event], including the making, dissolution and remaking of social and political identities and the redistribution of power and resources’ (Wagner-Pacifici, 2017, p. 11). Tracing digital facts as they travel through the semantic web is a useful mechanism for understanding how truth is constructed in the age of data. It is clear that there is still much to learn in the relations between the contingent, fluctuating knowledge that accompanies an unexpected media event as it evolves over time and the definitive, inflexible datafied fact on the other. Further investigations will yield important understandings on who and what are wresting control of these important phenomena and who is set to benefit from the events that serve to either rock the status quo from its foundations or to significantly reinforce the power of those who already hold sway.

NOTE 1. The Profumo scandal was a major controversy in British politics in which Secretary of State for War John Profumo lied about an affair with the 19-year-old model Christine Keeler in 1961.

FURTHER READING Dayan, D. and Katz, E. (1992). Media Events: The Live Broadcasting of History. Cambridge, MA: Harvard University Press. Ford, H. (2022). Writing the Revolution: Wikipedia and the Survival of Facts in the Digital Age. Cambridge, MA: MIT Press. Hepp, A. and Couldry, N. (2010). Introduction: Media events in globalized media cultures. In N. Couldry, A. Hepp, and F. Krotz (eds.), Media Events in a Global Age (pp. 1–20). New York: Routledge.

208  Handbook of digital politics

Moats, D. (2019). From media technologies to mediated events: A different settlement between media studies and science and technology studies. Information, Communication & Society, 22(8), 1165–1180. Morgan, M. (2010). Traveling facts. In P. Howlett and M. Morgan (eds.), How Well Do Facts Travel? The Dissemination of Reliable Knowledge (pp. 3–42). Cambridge: Cambridge University Press. Ohlheiser, A. (2017). The woman behind ‘Me Too’ knew the power of the phrase when she created it – 10 years ago. The Washington Post, 19 October. https://​web​.archive​.org/​web/​ 20171019201825/​https:/​www​.washingtonpost​.com/​news/​the​-intersect/​wp/​2017/​10/​19/​the​ -woman​-behind​-me​-too​-knew​-the​-power​-of​-the​-phrase​-when​-she​-created​-it​-10​-years​-ago/​. Wagner-Pacifici, R. (2010). Theorizing the restlessness of events. The American Journal of Sociology, 115(5), 1351–1386. Wagner-Pacifici, R. (2017). What Is an Event? Chicago: University of Chicago Press.

REFERENCES BBC (2011). Bin Laden raid was revealed on Twitter. https://​www​.bbc​.com/​news/​technology​ -13257940. Bennett, W. L. and Segerberg, A. (2012). The logic of connective action. Information, Communication & Society, 15(5), 739–768. Cantril, H. (1940). America faces the war: A study in public opinion. Public Opinion Quarterly, 4(3), 387–407. Chadwick, A. (2013). The Hybrid Media System: Politics and Power. New York: Oxford University Press. Couldry, N. (2005). Media Rituals: A Critical Approach. New York: Routledge. Couldry, A. Hepp, and F. Krotz (eds.) (2009). Media Events in a Global Age (pp. 1–20). New York: Routledge. Dayan, D. and Katz, E. (1992). Media Events: The Live Broadcasting of History. Cambridge, MA: Harvard University Press. Dewey, C. (2019). You probably haven’t even noticed Google’s sketchy quest to control the world’s knowledge. The Washington Post, 26 April. Ford, H. (2022). Writing the Revolution: Wikipedia and the Survival of Facts in the Digital Age. Cambridge, MA: MIT Press. Gerbaudo, P. (2012). Tweets and the Streets: Social Media and Contemporary Activism. London: Pluto Press. Hepp, A. (2019). Deep Mediatization. New York: Routledge. Hepp, A. and Couldry, N. (2010). Introduction: Media events in globalized media cultures. In N. Couldry, A. Hepp, and F. Krotz (eds.), Media Events in a Global Age (pp. 1–20). New York: Routledge. Hoover, S. M. (2010). Conclusion: The media events debate – moving to the next stage. In N. Couldry, A. Hepp, and F. Krotz (eds.), Media Events in a Global Age (pp. 283–299). New York: Routledge. Keegan, B., Gergle, D., and Contractor, N. (2013). Hot off the Wiki: Structures and dynamics of Wikipedia’s coverage of breaking news events. American Behavioral Scientist, 57(5), 595–622. Liebes, T. (1998). Television’s disaster marathons: A danger for democratic processes? In T. Liebes and J. Curran (eds.), Media, Ritual and Identity (pp. 71–84). London: Routledge. Moats, D. (2019). From media technologies to mediated events: A different settlement between media studies and science and technology studies. Information, Communication & Society, 22(8), 1165–1180.

Becoming eventful through data: the mediated construction of historic events  209

Morgan, M. (2010). Traveling facts. In P. Howlett and M. Morgan (eds.), How Well Do Facts Travel? The Dissemination of Reliable Knowledge (pp. 3–42). Cambridge: Cambridge University Press. Ohlheiser, A. (2017). The woman behind ‘Me Too’ knew the power of the phrase when she created it – 10 years ago. The Washington Post, 19 October. https://​web​.archive​.org/​web/​ 20171019201825/​https:/​www​.washingtonpost​.com/​news/​the​-intersect/​wp/​2017/​10/​19/​the​ -woman​-behind​-me​-too​-knew​-the​-power​-of​-the​-phrase​-when​-she​-created​-it​-10​-years​-ago/​. Scannell, P. (2017). Media events: An afterword. Media, Culture & Society, 40(1), 153–157. Smith, C. (2011). Egypt’s Facebook revolution: Wael Ghonim thanks the social network. Huffington Post, 2 November. https://​www​.huffpost​.com/​entry/​egypt​-facebook​-revolution​ -wael​-ghonim​_n​_822078. Tufekci, Z. (2017). Twitter and Tear Gas: The Power and Fragility of Networked Protest. New Haven: Yale University Press. Volkmer, I. and Deffner, F. (2010). Eventspheres as discursive forms: (Re-)negotiating the “mediated center” in new network cultures. In N. Couldry, A. Hepp, and F. Krotz (eds.), Media Events in a Global Age (pp. 217–236). New York: Routledge. Wagner-Pacifici, R. (2017). What Is an Event? Chicago: University of Chicago Press. Will, G. F. (2002,). ‘Events, dear boy, events’. Newsweek, 27 January. https://​www​.newsweek​ .com/​events​-dear​-boy​-events​-143481.

14. Algorithms, power and digital politics Ulrike Klinger

INTRODUCTION We live in a world of algorithms – computational operations making automated decisions on our behalf and in the interest of their creators, in almost all aspects of our life worlds. In the age of information abundance, algorithms curate news feeds, search results and social media timelines, they recommend groups and channels for us to join, determine which ads we see, how likely we are to buy certain products, estimate how much we are willing to pay for products and services, select music, movies and games we might like and even candidates on dating apps we might find attractive. Algorithms make decisions about the payment options we have when shopping online, our credit limits, calculate the probability we commit crimes and prioritize who gets a Covid-19 vaccine at what time. Algorithms are everywhere. At the same time, algorithms are anything but neutral or objective (Gillespie, 2014). They might appear as simple mathematical operations, splitting complex problems into small steps that can be operationalized and solved, but algorithms can and do quickly turn into “weapons of math destruction” (O’Neil, 2016). Algorithms are not more objective and less biased than the people who program them and the data which they process. Many examples and critical studies have shown that algorithms make racist decisions, discriminate against the poor, and reinforce and amplify various forms of discrimination (Eubanks, 2018; Sandvig et al., 2016). Safiya Umoja Noble (2018) has, along many other scholars, shown how automated decision-making systems create and deepen social inequalities, how sexism and racism are encoded into what she calls algorithmic oppression. In a more harmless application, algorithms used in beauty pageants detected beauty only in white bodies.1 In a more life-changing situation, algorithms were used to assign school exam grades in the United Kingdom, when exams could not take place during the pandemic in 2020. Because the algorithm put strong weight on the historical performance of the schools, it discriminated against high performing students in poorer schools, while underperforming students in expensive, private schools benefited from the automated grades.2 Algorithms can even make bias worse, because citizens may not even know when algorithms discriminate against them; also they mask human action and deflect accountability, and it is difficult to hold vastly opaque systems accountable for their decisions. Algorithms are also not better at decision-making than humans and make many mistakes. For instance, when during the Covid pandemic Facebook, Twitter and Google had to send human content moderators home and left the content moderation of the platform entirely to algorithms, the results were disastrous.3 Algorithmic 210

Algorithms, power and digital politics  211

decisions often cannot be appealed, and algorithmic accountability is very hard to attain. Moreover, despite their omnipresence and impact, only few citizens understand how algorithms work. A 2019 representative survey in 28 EU member states found that 48 percent of citizens do not know what an algorithm is, 46 percent have rather positive evaluations and expect positive outcomes from algorithmic decision-making, but 74 percent also want more rigorous control of using algorithms (Grzymek and Puntschuh, 2019). According to the same survey, in France, 21 percent of respondents find algorithms scary, and in the United Kingdom, 25 percent had never heard about algorithms. These results connect well to the rather paradoxical findings of the Reuters Digital News Report (2021) regarding use of and trust in social media in Europe. While between 31 percent (Germany) and 69 percent (Greece) say they use social media as a news source, only between 6 percent (United Kingdom) and 37 percent (Poland) say they trust social media as a news source.4 These numbers reflect the massive power asymmetry that materializes in algorithms – the power asymmetry between those who create, shape and maintain algorithms and those who only use them, sometimes knowingly, mostly unknowingly, or are subjected to algorithmic decision-making. Definitions There is no universally agreed definition of what an “algorithm” is, and often the term is used to deflect responsibility for human decisions on technology – “the algorithm did it” (Lum and Chowdhury, 2021). Etymologically the word algorithm comes from the Greek word for number “arithmos” and the Arabic word for calculation “al-jabr” (from which algebra stems, see Striphas, 2015). Most definitions centre on algorithms being a set of rules (Stone 1971), a multiple step process to produce specific outputs (Kitchin, 2017) and a way of automated decision-making (e.g. in the US Algorithmic Accountability Act of 2019). While most research focuses on the output of algorithms and discussions about the social consequences of their output, it is perhaps even more important to understand the input side. Algorithms are technologies of automation, they operate in the contexts of large and complex datasets and are trained on data. The results of their operations are contingent on the data – the types of data, the quality of data. The myth of “objective data” has been repeatedly debunked (e.g. Hong, 2020) – data are no neutral collections of facts and figures, but constructs, results of operationalizations, approximations. Data are uncertain, messy and biased, imperfect, so that, as Hong (2020) argues, datafication creates “data-driven fabrication” and ultimately, “technologies of speculation”, rather than objective truth. Algorithms, too, are human-made. They are industrial products, and the contexts of their production and maintenance are programmed into them – they are instrumental, they are programmed for certain goals, against the background of business models, the worldviews and experiences of their creators. Algorithms are never perfect, never finished, they are work in progress, with unintended consequences, with mistakes

212  Handbook of digital politics

being made, with collateral effects. There is bias in algorithms, in the data they were trained on, and in the data they run on. Thus we can understand algorithms as material as well as social processes – their calculations may be based on direct articulations from the programmer, on data they draw their calculations from, or data generated from calculations in previous steps. Algorithms as processes include (a) input, the data and instructions as code, which (b) lead to the calculations which operate in a big-data context, that then (c) result in decisions, some kind of output. As political communication and digital politics is taking place in digital environments – campaigning on platforms, micro-targeted ads, mobilizing social protest through connective action, crowdfunding political movements or building apps for political parties – algorithms are more than just “neutral” infrastructure to facilitate these political processes. Here I will first discuss how algorithms shape and impact key political processes and second reflect on how algorithms themselves have and exert power in these processes.

ALGORITHMS AND DIGITAL POLITICS Not only social media, but also search engines, apps, recommendations systems, streaming services, online shopping and dating portals run on algorithms that filter, select, curate or prioritize information. Algorithms do not “stay” in online, purely digital environments. In hybrid media systems, the digital world is the real world. Journalism has in many way “automated the news”, blended how humans and algorithms can work together, be it through data mining, automated text production, or the dissemination of information with newsbots (Diakopoulos, 2019). Mass mobilizations on the streets may have started and gained traction in Telegram channels or on Instagram. Disinformation in “alternative” media and radicalization in online forums may lead to people making political choices against their own interests, or, in the case of anti-vaxxers, even dying from preventable diseases. Algorithmic bias and outright discrimination can lead to dire consequences for entire social groups, such as longer prison terms, declined credit applications, or non-invitations to job interviews. Regarding digital politics, we will focus here on four areas in which algorithms are particularly relevant and have been intensively studied: (1) information, (2) participation and opinion formation, (3) personalization, and (4) polarization and radicalization. Information Social media and other online platforms have become relevant news sources for a large part of the populations in democratic societies, thus having an impact on how information is obtained and distributed and the level of political knowledge. In European countries, between 31 percent (Germany) and 69 percent (Greece) of Internet users get their news through social networks, in most cases on Facebook or WhatsApp – with similar numbers in other parts of the world (Reuters Digital News

Algorithms, power and digital politics  213

Report, 2021). While only a very small share of them relies entirely on social media for news, it is clear that TV, radio or newspapers are no longer the main conveyors of news. Instead, citizens increasingly leave the curation of their political information to automated processes instead of professional journalists, with algorithms choosing relevant news for them based on data and their previous behavior. While researchers at first assumed that this algorithmic personalization might have negative effects on the level of information and political knowledge – because algorithms are very good at showing content people want to see, not what they should see – recent empirical research rather identified other problems. Getting the news on social media is indeed associated with decreasing political knowledge and political interest, an attitude of passively waiting for news instead of actively informing oneself (the “news finds me” perception, Gil de Zúñiga and Diehl, 2019). A study found that one reason for this is that there is simply not enough “hard” news available on Facebook, and that users who are combining Facebook with traditional news sources experience feelings of information overload, which is keeping them from acquiring political knowledge (van Erkel and van Aelst, 2021). This confirms Bode’s (2016) argument that social media are low-choice media environments, with the potential that even politically uninterested citizens learn through incidental news exposure – a potential that is unfortunately often not realized. Indeed, replacing traditional news sources with social media leads to less political knowledge, and social media cannot compensate for not reading newspapers, watching news or actively reading online news sites (Shehata and Strömbäck, 2021). At the same time, disinformation travels faster, farther and deeper on social media (Vosoughi et al., 2018), as there is no epistemic editing on social media, a completely unedited public sphere (Bimber and Gil de Zúñiga, 2020). Algorithms are optimized to keep users on the platform, to hold their attention, to distribute content based on users’ presumed preferences and content that has previously triggered user interactions. Platforms, their affordances and algorithms are not optimized for political information or debate, and accordingly they perform pretty poorly in this regard. Participation and Opinion Formation Research has shown that people do not participate equally in online debates – only a small fraction actually writes or comments, while most people only read, share or lurk – a pattern that correlates with gender, age and education (Jensen and Schwartz, 2021; Koc-Michalska et al., 2021). The question of who speaks, who has voice, who is seen and heard in algorithmically curated environments is particularly relevant when it comes to digital politics. As the spiral of silence concept has argued for decades, citizens derive information about the world from their information environments: if they perceive themselves consistent with a majority opinion climate, they will express their opinion. If not, they will not change their opinion, but remain silent about it. This dynamic can lead to loud minorities and distortion of reality perceptions. Studies have shown that an extremely small number of highly active users impact online discourse and game the platform’s algorithms. Because most people do

214  Handbook of digital politics

not contribute much content and do not engage much, it does not take much to be an outlier, or a superspreader. Papakyriakopoulos et al. (2019) found that users leaving three or more comments were already “hyperactive” users, outliers in the general participation pattern. Klinger et al. (2022) found that just over 1 percent of Twitter accounts generated over 22 percent of all interactions in an anti-immigration campaign in Germany, and Martini (2020) showed that the German #MeToo discourse was “hash-jacked” by a tiny minority of anti-feminist, far-right accounts. The impact these few superspreaders can have on political discourse in algorithmically curated environments is impressive: they can create the false impression of grass-roots mobilizations, with engagement-based algorithms attributing relevance and further pushing their already outsized visibility. This of course distorts the basic idea of the public sphere as a marketplace of ideas. Algorithms also play an increasing role in the participation in social movements and the mobilization of protest. With the advent of connective action and the mobilizing effects of peripheral networks, participation is no longer brokered by organizations and their collective action, but through networked structures in algorithmically curated environments. Large movements across the entire political spectrum, from Fridays for Future and Black Lives Matter to protest against Covid vaccines or the MAGA movement, algorithms connect these activists and facilitate the distribution of their messages. Personalization An immediate effect, and perhaps the most tangible consequence of algorithms in digital politics is the personalization of information and participation. Accordingly, soon after social media platforms became key players in political communication, activists, journalists and scholars raised concerns about their unintended consequences, most prominently about filter bubbles and echo chambers (Pariser, 2011; Guo et al., 2020). The worry about homophily and like-mindedness, was that algorithms would serve users an endless stream of more of the same contents, information they want to see, but not what they need to see. Similarly, scholars worried that users would self-select to engage only with other users who agree with their opinions, leading to social fragmentation. The ultimate dystopia, then, was a society that could not agree on a shared reality anymore – a world in which journalists no longer conveyed a constructed version of reality to a mass society, but a world in which everyone could piece together their own reality, detached from each other. Empirical studies, however, have found that the evidence for the filter bubble and echo chamber hypotheses is rather thin, or that “if there are echo chambers, the walls are pretty porous” (Zuiderveen Borgesius et al., 2016, p. 6). In their comprehensive literature review on the topic, Ross Arguedas et al., (2022) concluded that “echo chambers are much less widespread than is commonly assumed” and found no support for the filter bubble hypothesis. On the contrary, many studies found that algorithmic curation on social media or search engines provided even slightly more diverse information, and only a small fraction of highly-partisan Internet users self-select to remain in

Algorithms, power and digital politics  215

like-minded echo chambers. Axel Bruns (2019c; see also Bruns, 2019b, 2021) put it even more pointedly: [E]cho chambers and filter bubbles principally constitute an unfounded moral panic that presents a convenient technological scapegoat (search and social platforms and their affordances and algorithms) for a much more critical problem: growing social and political polarisation. But this is a problem that has fundamentally social and societal causes, and therefore cannot be solved by technological means alone.

To sum up: yes, algorithms personalize political information and communication, citizens do not encounter the same news about the world. But algorithmic curation does not lock them into impermeable information bubbles, because of incidental news exposure and because most citizens do not get their news entirely on social media, but still watch television and talk to their spouses and acquaintances. Platforms might make it easier to avoid news altogether and to disconnect from people with other opinions, but only a small fraction of Internet users behave in this way. Polarization and Radicalization Another problem scholars have discussed with regard to algorithmic environments is their impact on polarization and radicalization. And indeed, several internal, leaked studies from Facebook have made the point that machine-learning algorithms based on engagement, seeking to increase user engagement, can have harmful consequences in terms of polarization and radicalization. The company had known for years that its own recommendation systems were recruiting users for extremist groups on the platforms. As one internal report stated for the case of Germany in 2016, “64% of all extremist group joins are due to our recommendation tools.”5 It is important to keep in mind here that we only know about these things from leaked internal documents, as it is impossible to gain such insights from the data available to researchers. In experimental settings, researchers have shown that indeed algorithmic recommendation can lead to opinion reinforcement and polarization, especially when running on user behavior data (Cho et al., 2020), but other studies could not find evidence for algorithmically generated content to drive polarization. For instance, Feezell et al. (2021) tried to empirically test Zeynep Tufekci’s (2018) famous observation that YouTube radicalized its users by suggesting escalating content in autoplay videos, independent of the topic: “Videos about vegetarianism led to videos about veganism. Videos about jogging led to videos about running ultramarathons. It seems as if you are never ‘hard core’ enough for YouTube’s recommendation algorithm” (Tufekci, 2018). However, Feezell et al. could not find evidence for this in their study, and suggest that social media platforms may play a role in facilitating related phenomena like affective polarization, but are perhaps not the source of polarizing partisan beliefs. Another problem with studies of polarization is that the overwhelming majority of them comes from the US with its two-party system. Research based on data from other parts of the world has previously challenged the

216  Handbook of digital politics

notion that partisan media reinforces polarization, but can indeed lead to moderation (Conroy-Krutz and Moehler, 2015), and there may be a similar effect in algorithmically curated environments. While it is certainly true that radical groups use social media platforms to further their cause and polarized issues are feverishly debated on these platforms, the research we have so far does not seem conclusive on the question as to whether and, if so, how, algorithms can be blamed for polarization and radicalization.

ALGORITHMS AND POWER Social science has so far intensively studied the impact of algorithms on democratic processes and political communication – as far as it is possible to study this with the little data that is available for independent research (Bruns, 2019a). However, all this research is overly focused on the output side, understanding algorithms as an external, a given force. Clearly, algorithms have an impact on digital politics, but also on all other aspects of our life worlds. But do algorithms have agency, and do they have power? And what can we learn from shifting the focus to the input side of algorithmic decision-making? At first sight, the question of whether we can regard algorithms as actors may seem very simple. Algorithms clearly “do” something, they make decisions, they make a difference when they identify patterns, they deliver results. They act, they change the status quo. That agency is not necessarily limited to human action has long been accepted in social science, and the proponents of actor-network theory (ANT) have famously brought forward the argument that actors can have “human, nonhuman, unhuman, inhuman characteristics” (Latour, 1996, p. 7). However, this perception focuses strongly on the output side, on the results algorithms deliver. The agency of algorithms is indeed quite different from human agency (Klinger and Svensson, 2018). Humans and organizations act through algorithms, and their actions are shaped by technology and the data they process. On the input side, the actions of algorithms are based on how programmers, engineers, tech companies or hacker collectives design them – through the business models, worldviews, economic or political interests of their creators and maintainers, the contexts of their creation are inherent in their design. Algorithms cannot move beyond themselves, they cannot change their purpose or even self-learn; non-supervised algorithmic systems can only act, adjust, improvise or change within a certain range. They know no serendipity, unless serendipity is programmed into them, e.g. by adding elements of randomness into recommendation systems.6 Algorithms have purpose, they are programmed to perform specific steps of defined actions. This means that algorithms act on behalf of someone else, and that those who impact the agency through their input (data, code) are not identical with those who are impacted by the output. There is nothing democratic about algorithms and their agency. Algorithms are mostly untransparent and there is more that we don’t know than we do know. Algorithms are embedded in complex algorithmic systems that are in most

Algorithms, power and digital politics  217

cases proprietary – companies own algorithms and keep their design secret. Platforms prohibit any forms of reverse engineering when providing data for researchers (e.g. Crowdtangle) and threaten scholars with legal claims. As a case in point, in 2021 Facebook aggressively moved against and finally closed down a research project “Ad Observatory” at New York University that had analysed the platform’s targeting of political ads. Already in 2013, Diakopoulos had cautioned that “what we generally lack as a public is clarity about how algorithms exercise their power over us” (p. 2). Since then, things have not improved and there are many known unknowns. For instance, no one outside the platform companies can know who sees which political ads, how much disinformation about Covid-19 and vaccines circulates on platforms and who is driving these, or how political mobilization dynamics move across platforms, e.g. from YouTube into Telegram channels to WhatsApp groups to Facebook or Instagram posts. Scholars have found that negativity increases user engagement on Facebook, that posts making people angry yield more reactions, shares and comments (e.g. Klinger et al., 2022). But there is no way to find out what role the platform’s algorithms play in this. As in many other cases, the few things we do know originate from internal studies that were leaked by whistleblowers. Facebook, for instance, diversified the old “Like” button with new buttons signifying emotional reactions (wow-astonishment, haha-humor and irony, sad, angry, love) in 2016. But Facebook did not treat those reactions equally. An “angry” reaction was assigned five times the weight of an ordinary “like”. This means that posts with angry reactions were deemed five times more valuable and that consequently, the platform’s algorithms would push more content like that: “Facebook CEO Mark Zuckerberg even encouraged users to use the angry face emoji to react to posts they didn’t like, without the users knowing it would push more content they didn’t like.”7 We only know this because of the internal documents known as the Facebook Papers leaked by Frances Haugen in 2021. The problem with opacity is that it reifies the power asymmetries between the companies that create algorithms and own the data on which they operate, and everyone else, who is affected by the operations of the algorithmic systems. Not only is knowledge power, but the ability to uphold secrecy and opacity around how and why decisions are made, to avoid scrutiny while owning and processing the data to scrutinize others, is the resource of extreme wealth and the power of tech companies. As a result, we find ourselves in what Frank Pasquale (2015) has termed a “black-box society”: We do not live in a peaceable kingdom of private walled gardens; the contemporary world more closely resembles a one-way mirror. Important corporate actors have unprecedented knowledge of the minutiae of our daily lives, while we know little to nothing about how they use this knowledge to influence important decisions that we – and they – make. (Pasquale, 2015, p. 9)

So, algorithms have agency, but a different agency from humans, and their opacity is an expression of power asymmetries. But do algorithms in themselves have and exert power?

218  Handbook of digital politics

In classic political theory, power refers to the opportunity to impose one’s will on others, even against their resistance (Max Weber). The ability to do so can rest on various resources – brute force, accepted hierarchy, money, convincing arguments or charisma. However, when it comes to media or platforms, power often rather refers to strong impact, the ability to influence or even manipulate public opinion through gate-keeping, agenda-setting or curating information flows by filtering, selecting, and prioritizing. Media and platforms can amplify or understate information and opinions; the messages they convey are often taken for granted, but are the results of professional and/or technological curation processes. Communication power, as Manuel Castells defines it, thus means that “the media are not the holders of power, but they constitute by and large the space where power is decided” (2007, p. 242). In 2014, Moíses Naím put forward the argument that power is not what it used to be, that it is easier to get and to lose power, and that power delivers less for those who have it than in earlier decades and centuries. Interestingly, he also provided a very useful definition of power that goes beyond the classical reading and helps to understand the power of algorithms and platforms: Power is the ability to direct or prevent the current or future actions of other groups and individuals. Or, put differently, power is what we exercise over others that leads them to behave in ways they would not otherwise have behaved. (Naím, 2014, p. 16)

Building on Ian MacMillan, Naím differentiates four dimensions of exercising power: through persuasion (through argument or advertisement), through coercion (by law or force), through obligation (such as in religious or traditional duties) and through rewards. This differentiation is helpful because it includes influence through persuasion, but clearly locates it as a subset of power instead of confusing it with power itself. Persuasion “changes the perception of a situation, not the situation itself” (Naím, 2014, p. 27).

RESEARCHING ALGORITHMS, POWER AND DIGITAL POLITICS In most cases, algorithms run in the background with most users giving little thought to them – unless something goes wrong. When algorithms discriminate against social groups, when they make racist decisions, or simply when they fail, e.g. in content moderation, the blame game begins. Companies blame “the algorithm”, users blame “the companies”, observers and experts blame a “lack of regulation”. What future research needs to do is not so much untangle how algorithmic decision-making impacts digital politics, but to address and critically discuss the power structures behind the interfaces of digital communication. Interestingly, social scientists have so far not explored many alternative ways of how to study algorithms beyond their immediate effects (Lee and Björklund Larsen, 2019). Algorithms are not just “technologies” or “mathematics”, and in contrast to the well-researched world of

Algorithms, power and digital politics  219

information production in the mass media era, namely the economic, cultural and professional contexts of journalism, we do not know much about the contexts of the industrial production of code, such as algorithms. Here I suggest differentiating at least three dimensions, power of algorithms, power through algorithms, and power over algorithms. Power of Algorithms As we have discussed already, algorithms have agency and their opacity is an expression of power asymmetries. But how exactly do they have power? This is not always so obvious, as in cases when algorithms decide if someone gets a credit card, or not. Algorithms can have the ability to direct or prevent the current or future actions of groups and individuals – by persuasion, by coercion and through rewards. Algorithms can persuade in more or less subtle ways, because they make decisions on behalf of a user based on the user’s presumed preferences, as manifested in data. Algorithms persuade users to click on content, to stay on a platform, to buy products, to look at ads, to share messages, to join groups, to play videos, or to share more behavioural data. They coerce them into specific behavior by limiting or extending the options users have – e.g. by allocating eligibility for Covid vaccination appointments, or by shadow-banning political groups. And algorithms can nudge user behavior through rewards, e.g. Uber incentivizing its drivers to work longer shifts through reward badges.8 Algorithms do not only make decisions for users, e.g. on personalized search results or news feeds, but steer their behavior and predict their future behavior, as Zuboff (2019) put it, they trade human futures. Algorithms make users do things they would not have done without them, to behave in a way they would not have otherwise. As a very simple example, platform algorithms are optimized to sell users’ attention to advertisers, to keep them on the platform as long as possible and to engage them cognitively. As a result, users stay on platforms longer, give away more of their most limited resource (time) than they otherwise would have; they spend their time with more negative emotions than they otherwise would have (because anger keeps them engaged, algorithms prioritize postings evoking negative emotions). Power through Algorithms There is, however, more to the power of algorithms than just the output side, the decisions they make and the results they produce. To understand this better, one must change the perspective and scrutinize the input side of algorithms: the business models, work routines, worldviews and ideologies and the data (and their biases) that are all at work behind the user interfaces (Klinger and Svensson, 2021). Behind the interfaces and affordances of digital technologies is an entire industry, manufacturing our digital world and the preconditions for digital politics – through algorithms. It is astonishing how little we know, and how little research there is, about the makers and maintainers of these systems, as Jakob Svensson wonders in his book Wizards of the

220  Handbook of digital politics

Web (2021), in which he explores the humans behind the machines, the mathemagicians, the hippies, the entrepreneurs. The teams of designers and programmers have power through algorithms, because through their decisions, their code, they are able to direct or prevent the current or future actions of other groups and individuals. Their job is to implement psychological knowledge to make users behave in a beneficial way to the platform-owning company, e.g. interact more, watch more and longer videos, upload more photos and information about their lives. Again, this works through persuasion, coercion and rewards. For instance, many apps only function properly when users share their geolocation data, or in the case of TikTok, users share biometric personal data like faceprints, voiceprints, or keystroke patterns and rhythms.9 The proprietary structure of most digital environments also means that those who create and maintain these environments have the power to unilaterally (and often without appeal) “moderate” content, de-platform users and to be in control over microtargeted advertisements. Social movements rely on connective action – when organization and participation are based on interpersonal networks facilitated though technology, then those who control the technology can direct and/or prevent further actions of these networked movements. Political parties, just as commercial advertisers, can direct tailored messages to specific, predefined granular target groups – but without any control over the process. They pay platforms like Google or Facebook for micro-targeting their messages in election campaigns, but they have no opportunity to scrutinize the process and to validate who sees what, there is no independent arbiter between advertisers and platforms, and it may well be the case that micro-targeting is not working as well as platforms promise.10 But also political parties themselves can have power through algorithms, when they engage in data-driven campaigning to inform and mobilize voters, or in a more subversive way, to demobilize and disinform (Römmele and Gibson, 2020), or when they create apps for voters and collect big datasets about voters and supporters. Power over Algorithms Taina Bucher (2018) has convincingly argued that understanding algorithms as “black boxes” is a misleading metaphor. Indeed, while it is popular for researchers, social actors or regulatory institutions to demand more transparency, to open up the black box, it will not help much in better understanding or governing algorithms. The main reason for this is the emergent, relational and performative “nature” of algorithms. It is true that algorithms are often proprietary; they are industrial products intentionally designed by commercial organizations. But other than public discourse about them suggests, there is not one “Google algorithm” or “Facebook algorithm” or “Netflix algorithm”. Platforms like Google, Facebook or Netflix run on billions of lines of code, complex algorithmic systems, with multitudes of different versions of themselves at the same time. Many different algorithms interact with each other, while they are constantly changing in machine-learning processes. Platforms change

Algorithms, power and digital politics  221

all the time, are running and experimenting with different versions, the hands and minds of (human) engineers, designers and maintainers are permanently reaching into them, tweaking, adapting, replacing parts and versions of code. Code may be law, as Lessig famously put it. But it is a law that is never finished, or ready for inspection, a law that is constantly being rewritten over and again. Those who have power over algorithms thus may themselves not know all the time what exactly their algorithms look like and how exactly they operate. And they do not have to. Opening the black box for inspection, let’s say for a public auditing or a code review with the common good of society in mind, would not challenge their power or change algorithms. Power over algorithms, in essence and with regards to Naím’s definition, means the ability to steer their behavior, their performance – for instance making a platform’s algorithms behave in ways these would not otherwise have behaved. One example here would be Facebook’s decision to tweak its news algorithms in the crucial period shortly before and after the 2020 US elections – to make it “nicer”, to prioritize news from authoritative sources over hyper-partisan sources, to combat disinformation about the election. Soon after the election, the platform returned to the original version, optimized for economic success with advertisements: “This was a temporary change we made to help limit the spread of inaccurate claims about the election,” said Joe Osborne, a Facebook spokesman. […] Other measures Facebook has developed to combat political misinformation and hate speech have been scaled back or vetoed by executives in the past, either because they hurt Facebook’s usage numbers or because executives feared they would disproportionately harm right-wing publishers, several Facebook employees told The Times last month.11

In a similar way, Twitter has stopped “enforcing its civic integrity policy”, i.e. all its actions against the spread of lies about the 2020 election in March 2021.12 Power over algorithms means being in a position to make strategic decisions about the performance and the objectives of algorithmic systems and to have them implemented. It is not the individual coders, and perhaps not even the teams who design and maintain algorithms – it is perhaps not even people in tech leadership positions, such as architects. Power over algorithms is indeed limited to a very small group of people within the leadership of the organizations that own these algorithms. They have the final decisions about strategic objectives, can green-light changes – or not, and control their algorithms’ priorities. For example, Facebook repeatedly adjusts its News Feed algorithm, which has an impact on how much news users see in their timelines and which news sources are prioritized. In 2018, the company decided to prioritize content from family and friends over other content, including news – which had a measurable effect on news media organizations and their reach (Bailo et al., 2021). These decisions are not in the hands of any team or employee, but are controlled by the top leadership.

222  Handbook of digital politics

CONCLUSION: IT’S NOT ABOUT TECHNOLOGY, BUT ABOUT ACCOUNTABILITY In this chapter I have argued that algorithms are not just technology and not just math, but the outcome of social processes. They are industrial products, manufactured by humans with specific intentions for defined tasks. Because the digital world is based on algorithms – not single operations, but massive algorithmic systems that are constantly emerging, changing, intersecting with various versions of themselves – they matter for political communication and digital politics. Algorithms are not neutral or objective, but messy, biased and imperfect. To understand their impact, we should not focus only on the output side, on the decisions they make and the results they produce, but on their makers, designers, and their owners. Joseph Weizenbaum, computer science pioneer and critical thinker, already warned in the 1970s: “The fact that our society is increasingly relying on computer systems that have long been beyond the comprehension of those who work with them and are becoming more and more indispensable to them is a very serious development” (Weizenbaum 1980/1976, p. 311, translation by author). And still, this hits the crucial point: We cannot hold algorithms accountable for their actions, their failures, their consequences. But as a democratic society, we can demand accountability from those who create them, operate them, own and make money with them. Thus the core tenet is not so much that algorithms shape and impact political communication (yes, they do, it is actually quite banal), but that technology companies are increasingly shaping political communication and digital politics (Kreiss and McGregor, 2018, 2019). Algorithms do not just shape and impact digital politics, but they have power – and there is a massive power asymmetry between those who use digital technologies (citizens, activists), those who are in a position to shape or at least choose how to implement digital technologies (programmers, but also political parties when making decisions about data-driven campaigns or party apps) and particularly those who own and have power over digital technologies and platforms. But there is no reason to despair over the opacity and complexity of algorithms, and their power. The good news is that everything human-made can be re-made, or as Diakopoulos (2019, p. 240) put it: Algorithms lack animus; they are inert. What animates algorithms are the people who design, develop, operate, and manage them. It is people who define, measure, and sample data to feed and train algorithms. […] We, the people, have agency in – and responsibility for – how these systems ultimately operate and influence the media. The future of algorithmic media must be human-centered.

The call for algorithmic accountability does not end with demanding it from social media platforms, but to think about how we, as democratic societies, can design hybrid media systems, the interaction of automated and human agency, in a way that minimizes potentially harmful collateral effects on public debates, political

Algorithms, power and digital politics  223

knowledge and participation. To achieve this, it will be necessary that society and its actors have a seat at the table, to be included in the production of algorithmic systems that impact public debate and political processes. Making algorithms transparent, opening the black box, will hardly be enough.

FURTHER READING Arguedas, A. R., Robertson, C. T., Fletcher, R., and Nielsen, R. K. (2022). Echo chambers, filter bubbles, and polarisation: A literature review. https://​reutersinstitute​.politics​.ox​.ac​.uk/​ echo​-chambers​-filter​-bubbles​-and​-polarisation​-literature​-review. Bruns, A. (2019). Are Filter Bubbles Real? New York: John Wiley & Sons. Bucher, T. (2018). If … Then: Algorithmic Power and Politics. New York: Oxford University Press. Diakopoulos, N. (2019). Automating the News. Cambridge, MA: Harvard University Press. Gillespie, T. (2018). Custodians of the Internet. New Haven: Yale University Press. Svensson, J. (2021). Wizards of the Web: An Outsider’s Journey into Tech Culture Programming and Mathematics. Gothenburg: Nordicom.

REFERENCES Arguedas, A. R., Robertson, C. T., Fletcher, R., and Nielsen, R. K. (2022). Echo chambers, filter bubbles, and polarisation: A literature review. https://​reutersinstitute​.politics​.ox​.ac​.uk/​ echo​-chambers​-filter​-bubbles​-and​-polarisation​-literature​-review. Bailo, F., Meese, J., and Hurcombe, E. (2021). The institutional impacts of algorithmic distribution: Facebook and the Australian news media. Social Media + Society, 7(2), 20563051211024963. Bimber, B., & Gil de Zúñiga, H. (2020). The unedited public sphere. New Media & Society, 22(4), 700–715. Bode, L. (2016). Political news in the news feed: Learning politics from social media. Mass Communication and Society, 19(1), 24–48. Bruns, A. (2019a). After the ‘APIcalypse’: Social media platforms and their fight against critical scholarly research. Information, Communication & Society, 22(11), 1544–1566. Bruns, A. (2019b). Are Filter Bubbles Real? New York: John Wiley & Sons. Bruns, A. (2019c). It’s not the technology, stupid: How the ‘echo chamber’ and ‘filter bubble’ metaphors have failed us. Paper presented at the IAMCR 2019 conference in Madrid, July 7–11. Bruns, A. (2021). Echo chambers? Filter bubbles? The misleading metaphors that obscure the real problem. In M. Pérez-Escolar and J. M. Noguera-Vivo (eds.), Hate Speech and Polarization in Participatory Society (pp. 33–48). New York: Routledge. Bucher, T. (2018). If … Then: Algorithmic Power and Politics. New York: Oxford University Press. Castells, M. (2007). Communication, power and counter-power in the network society. International Journal of Communication, 1(1), 238–266. Cho, J., Ahmed, S., Hilbert, M., Liu, B., and Luu, J. (2020). Do search algorithms endanger democracy? An experimental investigation of algorithm effects on political polarization. Journal of Broadcasting & Electronic Media, 64(2), 150–172. Conroy-Krutz, J., & Moehler, D. (2015). Moderation from bias: A field experiment on partisan media in a new democracy. The Journal of Politics, 77(1), 575–587.

224  Handbook of digital politics

Diakopoulos, N. (2013). Algorithmic accountability: On the investigation of black boxes. Columbia Journalism Review. https://​www​.cjr​.org/​tow​_center​_reports/​algorithmic​ _accountability​_on​_the​_investigation​_of​_black​_boxes​.php​#:​~:​text​=​What​%20we​ %20generally​%20lack​%20as​,of​%20any​%20particular​%20algorithmic​%20power. Diakopoulos, N. (2019). Automating the News. Cambridge, MA: Harvard University Press. Eubanks, V. (2018). Automating Inequality: How High-Tech Tools Profile, Police, and Punish the Poor. New York: St. Martin’s Press. Feezell, J. T., Wagner, J. K., and Conroy, M. (2021). Exploring the effects of algorithm-driven news sources on political behavior and polarization. Computers in Human Behavior, 116, 106626. Gil de Zúñiga, H. and Diehl, T. (2019). News finds me perception and democracy: Effects on political knowledge, political interest, and voting. New Media & Society, 21(6), 1253–1271. Gillespie, T. (2014). The relevance of algorithms. In T. Gillespie, P. J. Boczkowski, and K. A. Foot (eds.), Media Technologies: Essays on Communication, Materiality, and Society (pp. 167–194). Cambridge, MA: MIT Press. Grzymek, V. and Puntschuh, M. (2019). What Europe knows and thinks about algorithms. Discussion Paper Ethics of Algorithms #10. Bertelsmann Foundation, Gütersloh. https://​ www​.Bertelsmann​-Stiftung​.De/​Fileadmin/​Files/​Bst/​Publikationen/​Grauepublikationen/​ Whateur​opeknowsan​dthinkabou​talgorithm​.Pdf. Guo, L., Rohde, J. A., and Wu, H. D. (2020). Who is responsible for Twitter’s echo chamber problem? Evidence from 2016 US election networks. Information, Communication & Society, 23(2), 234–251. Hong, S. H. (2020). Technologies of Speculation. New York: New York University Press. Jensen, J. L. and Schwartz, S. A. (2021). The return of the “lurker”: A longitudinal study of citizens’ use of social media in Danish elections 2011, 2015, and 2019. Social Media + Society, 7(4), 20563051211063463. Kitchin, R. (2017). Thinking critically about and researching algorithms. Information, Communication & Society, 20(1), 14–29. Klinger, U., & Svensson, J. (2018). The end of media logics? On algorithms and agency. New Media & Society, 20(12), 4653–4670. Klinger, U., & Svensson, J. (2021). The power of code: Women and the making of the digital world. Information, Communication & Society, 24(14), 2075–2090. Klinger, U., Lance Bennett, W., Knüpfer, C. B., Martini, F., & Zhang, X. (2022). From the fringes into mainstream politics: intermediary networks and movement-party coordination of a global anti-immigration campaign in Germany. Information, Communication & Society, 1–18. Koc-Michalska, K., Schiffrin, A., Lopez, A., Boulianne, S., and Bimber, B. (2021). From online political posting to mansplaining: The gender gap and social media in political discussion. Social Science Computer Review, 39(2), 197–210. Kreiss, D. and McGregor, S. C. (2018). Technology firms shape political communication: The work of Microsoft, Facebook, Twitter, and Google with campaigns during the 2016 US presidential cycle. Political Communication, 35(2), 155–177. Kreiss, D. and McGregor, S. C. (2019). The “arbiters of what our voters see”: Facebook and Google’s struggle with policy, process, and enforcement around political advertising. Political Communication, 36(4), 499–522. Latour, B. (1996). On actor-network theory: A few clarifications. Soziale Welt, 47, 369–381. http://​www​.jstor​.org/​stable/​40878163. Lee, F. and Björklund Larsen, L. (2019). How should we theorize algorithms? Five ideal types in analyzing algorithmic normativities. Big Data & Society, 6(2), 2053951719867349. Lum, K. and Chowdhury, R. (2021). What is an “algorithm”? It depends whom you ask. MIT Technology Review. https://​www​.technologyreview​.com/​2021/​02/​26/​1020007/​what​-is​-an​ -algorithm/​.

Algorithms, power and digital politics  225

Martini, F. (2020). Wer ist# MeToo? Eine netzwerkanalytische Untersuchung (anti-) feministischen Protests auf Twitter. M&K Medien & Kommunikationswissenschaft, 68(3), 255–272. Naím, M. (2014). The End of Power: From Boardrooms to Battlefields and Churches to States, Why Being in Charge Isn’t What It Used to Be. New York: Basic Books. Noble, S. U. (2018). Algorithms of Oppression. New York: New York University Press. O’Neil, C. (2016). Weapons of Math Destruction: How Big Data Increases Inequality and Threatens Democracy. New York: Crown. Papakyriakopoulos, O., Shahrezaye, M., Serrano, J. C. M., and Hegelich, S. (2019). Distorting political communication: The effect of hyperactive users in online social networks. IEEE INFOCOM 2019 – IEEE Conference on Computer Communications Workshops (INFOCOM WKSHPS) (pp. 157–164). IEEE. Pariser, E. (2011). The Filter Bubble: How the New Personalized Web Is Changing What We Read and How We Think. New York: Penguin. Pasquale, F. (2015). The Black Box Society: The Secret Algorithms That Control Money and Information. Cambridge, MA: Harvard University Press. Römmele, A. and Gibson, R. (2020). Scientific and subversive: The two faces of the fourth era of political campaigning. New Media & Society, 22(4), 595–610. Sandvig, C., Hamilton, K., Karahalios, K., and Langbort, C. (2016). Automation, algorithms, and politics when the algorithm itself is a racist: Diagnosing ethical harm in the basic components of software. International Journal of Communication, 10, 4972–4990. Shehata, A. and Strömbäck, J. (2021). Learning political news from social media: Network media logic and current affairs news learning in a high-choice media environment. Communication Research, 48(1), 125–147. Stone, H. S. (1971). Introduction to Computer Organization and Data Structures. New York: McGraw-Hil. Striphas, T. (2015). Algorithmic culture. European Journal of Cultural Studies, 18(4–5), 395–412. Svensson, J. (2021). Wizards of the Web: An Outsider’s Journey into Tech Culture Programming and Mathematics. Gothenburg: Nordicom. Tufekci, Z. (2018). YouTube: The great radicalizer. The New York Times, March 10. van Erkel, P. F. and Van Aelst, P. (2021). Why don’t we learn from social media? Studying effects of and mechanisms behind social media news use on general surveillance political knowledge. Political Communication, 38(4), 407–425. Vosoughi, S., Roy, D., and Aral, S. (2018). The spread of true and false news online. Science, 359(6380), 1146–1151. Weizenbaum, J. (1980/1976). Die Macht der Computer und die Ohnmacht der Vernunft. Frankfurt am Main: Suhrkamp. Zuboff, S. (2019). The Age of Surveillance Capitalism: The Fight for a Human Future at the New Frontier of Power: Barack Obama’s Books of 2019. Profile Books. Zuiderveen Borgesius, F., Trilling, D., Möller, J., Bodó, B., De Vreese, C. H., and Helberger, N. (2016). Should we worry about filter bubbles? Internet Policy Review – Journal on Internet Regulation, 5(1).

15. Social media digital architectures: a platform-first approach to political communication and participation Michael Bossetta

Have you ever noticed that when you open a social media app, the post at the bottom of your screen is cut off? Go ahead, try it. You’ll find that most of the time, the bottom post in your feed is only partially visible. The software designers and engineers who build social media platforms – the architects – are trying to tell you: “Scroll down!” Doing so exposes you to advertisements, which helps pay the architects’ salaries. Some might argue that scrolling – or “scrollability” – is an affordance of social media (e.g., McGrenere and Ho, 2000). After all, scrolling is a possibility for action that is activated through a relationship between user and technology. Others would argue that scrolling is not an affordance but rather, an established cultural convention (Norman, 1999). Alternatively, we could sidestep the largely terminological debate about affordances and try to observe how specific elements of platform design – such as not showing a full post at the bottom of the landing page – shapes users’ practices of social media engagement. Although apolitical, I open with the example of scrolling to highlight two approaches to conceptualizing social media’s impact on politics. The first is the affordances approach, which theorizes how the relation between users and technology affects political processes. Here, abstract concepts such as visibility, anonymity, and persistence are argued to be valuable lenses for studying digital politics, even though these concepts lack coherent definitions and are often inconsistently applied (Evans et al., 2017). The second approach is the architectural approach, which theorizes how specific components of platform design shape political processes. Here, concrete excavations of platform structures try to explain their effects on users’ political agency, both online and offline. In this chapter, I advocate for an architectural approach to studying digital politics. Relative to affordances, I argue that a focus on platform architectures is more conducive to concept building, clearer to operationalize, and provides the added benefit of archiving platform changes across research designs. I mount this argument through a focus on digital architectures: the collective suite of technical protocols that enable, constrain, and shape user behavior in a virtual space. Before outlining digital architectures in more detail, I first discuss previous architectural approaches to studying virtual publics and political discourses online. Then, I build on this prior work to introduce digital architectures and offer a conceptual roadmap on how to map political functions to platforms’ architectural characteristics. Before concluding 226

Social media digital architectures  227

the chapter with future research avenues, I provide an illustration of how digital architectures influence various facets of citizens’ political participation.

THE DISCOURSE ARCHITECTURE APPROACH TO PUBLICS AND CONVERSATIONS Prior to social media, scholars focused on how the structure of an online space – such as a forum, website, or blog – shaped the formation of publics and political discourses. The concept of “discourse architecture”, originating from the literature on Information Systems, described the “technology base and features that help structure discourses” (Jones and Rafaeli, 2000, p. 218). Jones and Rafaeli (2000), who examined how virtual publics could be fostered for e-commerce through computer-mediated spaces, considered aspects of discourse architectures to be: the types of media supported (e.g., sound, images, and video), whether communication is synchronous or asynchronous, and the length of time a post remains visible online. Jones and Rafaeli’s (2000) core argument was that virtual publics are not infinitely scalable. Rather, software developers needed to actively create virtual publics through appropriate audience segmentation strategies – achieved through the design of the discourse architecture – in order to attract, retain, and manage the growth of simultaneous user communities. In hindsight, their conceptual argument was highly prescient, given the effectiveness of Facebook groups and Twitter hashtags in creating numerous and concurrent virtual publics a decade later. As more users gained access to the Internet, Sack (2005) helped carry the concept of discourse architectures into the social sciences. At the time, a critical mass of users began posting and commenting on early versions of online forums, such as Usenet, which was a prototype of today’s Reddit. Noting the “very large-scale” of these conversations, Sack (2005, p. 243) urged web developers to establish practices for designing discourse architectures, defined as “environments to support conversation, discussion, and exchange between people”. To help evaluate the large-scale exchanges within these architectures, Sack (2005) argued that the development of analytical methods could be inspired by, and help contribute to, the social sciences and humanities. Although he was not the first to interpret Usenet data through a democratic lens, Sack (2005) showed how components of deliberative theory could guide the operationalization and interpretation of user activity at scale. For example, visualizing the interactions between users (i.e., social network analysis) could reveal insights about the conversation’s equality of participation. Or, an analysis of words used in the conversation (i.e., computational text mining) could measure the conversation’s diversity of viewpoints represented. Political communication scholars have since applied the discourse architecture approach to study how digital structures influence the quality and norms of online political discussions. Wright and Street (2007, p. 863) compared online forums to argue that levels of moderation, as well as features such as “a threaded system of replies”, affect an online discussion’s content in ways that indicate the quality of

228  Handbook of digital politics

deliberative discourse. In comparing political conversations on Twitter and online news sites, Freelon (2015, p. 776) argued that variations in discourse architectures, understood as distinct “packages of technological characteristics”, can nudge users toward participating in conversations that enact various democratic norms: deliberative, communitarian, or liberal individualist. Freelon’s (2015) analysis suggests that Twitter’s hashtag feature promotes bonding between users and therefore supports communitarian ideals, whereas a lack of character limits in online news sites supports more long-form opinion justification in line with deliberative ideals. These studies of discourse architecture highlight the force that digital structures exert upon user agency in online environments (e.g., virtual public formation or the discursive expression of democratic ideals). Each acknowledges that the design elements of an online space have a direct or mediating effect on user behavior within it. Following this logic, scholars can deconstruct the elements of an online space that may cause, explain, or restrict certain political behaviors. In the digital architectures approach I present next, I keep this core theoretical focus on digital structures but extend its implications to forms of political agency beyond discourse. I argue that by conceptually mapping which aspects of platform design matter for users’ political agency within a platform, we can also build knowledge about how platform design shapes political agency outside of platforms as well.

DEFINING DIGITAL ARCHITECTURES Digital architectures can be defined as the collective suite of technical protocols that enable, constrain, and shape user behavior in a virtual space. The first term in this definition to clarify is “technical protocols”, which has a very specific meaning in network infrastructure studies (such as Hypertext Transfer Protocol [HTTP]). Here, however, technical protocols refer broadly to “computational rule systems”. All aspects of a social media platform – its aesthetic design, what it does, and what it doesn’t do – are constructed through a series of rule systems. These rule systems are written in code and form the core building blocks of a platform’s digital architecture. For example, there are rule systems that encode how individual buttons look, such as Facebook Reactions or Instagram Likes, and rule systems for what happens when we use them (or don’t). There are rule systems for what information is presented to us, and rule systems guiding what data should be collected, organized, and transferred based on how we respond to that information. There are rule systems that govern how far our posts travel and whom they reach. The point is that each platform has a dizzying compilation of rule systems that work together to dictate how the platform operates, and the interoperability of these rule systems to generate seamless functionality under a common user interface is characteristic of a “software suite”. Thus, technical protocols are the constellation of features and functions that give a platform its form while also governing its operation. On social media, the technical protocols that govern a platform’s operation make possible certain actions, place limitations on others, and ultimately shape user behav-

Social media digital architectures  229

ior on the platform. In line with the discourse architecture approach, these protocols are the rule systems that decide, for example: the types of media technically supported by a platform, the limitations placed on those media (such as text character limits or video length limits), and the extent to which algorithms filter content. However, in a departure from the discourse architectures approach, the conceptual focus of digital architectures goes beyond how digital structures influence the content of what users say. Of course, technical protocols will necessarily influence the content of discourse (in terms of what is allowed to be broadcast and visible on the platform), but these protocols may also influence users’ behavioural activity – both on and off the platform. For example, an architecture that maximizes privacy through anonymity and encryption (such as Telegram or Signal) will influence the content of discourse on the platform. However, such an architecture may also affect real-world behavior outside the platform, such as purchasing illicit goods, enacting political demonstrations, or carrying out terrorist attacks. Whether we speak of discourse architectures or digital architectures, both differ from the concept of affordances by placing a theoretical focus on how platform structures shape user behavior (i.e., they are “platform-first” approaches). Affordances, meanwhile, places its emphasis on the “relation” between users and technologies (Evans et al., 2017). Such a relational focus is problematic for social media research, since the relationship between user and technology is (a) different for each individual and (b) constantly changing as platform structures evolve. Social media platforms are transient environments (Barrett and Kreiss, 2019), where software developers release hundreds of updates on a daily basis through a software practice known as continuous deployment (Savor et al., 2016). Thus, the dynamic and unstable nature of platform structures, coupled with the individualized and often unpredictable ways in which users interact with them, leads to extreme difficulty in theorizing or empirically testing relationships between users and platforms in a way that accounts for individual variance. In essence, the concept of affordances falsely assumes a stable relationship between user and technology, when in fact the underlying technology is evolving on a daily basis. However, by first mapping out platform structures and observing how users interact with them at a single point in time, scholars can build research designs that effectively accumulate knowledge regarding how platforms shape digital politics.

DIGITAL ARCHITECTURES AND POLITICAL AGENCY ONLINE AND OFFLINE Social media platforms are not single composite structures. Rather, they often include various intra-platform spaces that serve various functions for users. For example, a user may land on a centralized broadcast feed when first logging into a platform (such as Facebook’s News Feed or Twitter’s Timeline), but they then can navigate to different intra-platform spaces, such as: moderated community groups, direct messages, or spaces that allow users to change their privacy settings.

230  Handbook of digital politics

Digital architectures structure how such intra-platform spaces – or “digital rooms” – are designed and how users navigate across them. Some platforms host spaces where users meet to discuss topical themes, such as pages and groups on Facebook or Subreddits on Reddit. The same might be said for YouTube videos, which open up a commenting space underneath the video. Other platforms like WhatsApp and Telegram host less publicly visible groups, whereas Instagram, Snapchat, and TikTok generally lack a space for like-minded users to congregate and interact publicly. While Twitter also lacks such defined group spaces, users can generate quasi-spaces through hashtags, which index conversations in a way that opens a thematic, conversational space within the platform. Such intra-platform spaces matter for what users do and say on a platform, how likely users are to encounter like-minded or opposing viewpoints, and how the platform may function for the user politically. While there are various online behaviors that these architectures influence (such as what content users are exposed to, what users they interact with, and the discourse of what they say), digital architectures can also influence political agency outside of a platform. Keeping with the example of intra-platform spaces, the types of spaces offered by a platform can directly influence political behaviors, such as offline protest mobilization. The Aganaktismenoi movement, which was a series of anti-austerity protests in Greece, began with a protest that was organized overnight through a Facebook page (Treré et al., 2017). While the Facebook page directly called for a protest that would attract over 20,000 citizens, post hoc analysis of the Twitter conversation reveals that only 6 percent of tweets from citizens included a call to action that would signal mobilization (Theocharis et al., 2015). Instead, the majority of tweets were used to spread information about the protests. One explanation for this difference between Facebook and Twitter is differences in their intra-platform spaces. Facebook offers features such as pages, events, and groups, which lend themselves to coordinating offline mobilization. Twitter, meanwhile, lacks these spaces with organizing potential, and therefore serves a different function for activists. As one activist from Treré et al.’s (2017, p. 414) study noted: Twitter is the king of protest reporting, because it’s faster, direct … But then when you want to comment on what has happened Twitter is problematic because it doesn’t give you enough space to post and you need longer posts. You may need to write an article and publish it through a blog or a website, and maybe after that you link it through Twitter. Or even if you don’t have time or don’t have enough materials to write an article, you write a post on Facebook that can be something in between.

Here, the activist notes how Twitter’s lack of algorithmic filtering (i.e., being faster and more direct) as well as character limits (i.e., Facebook allowing longer posts) influence how they are used during protests. In a similar vein, Gerbaudo (2012, p. 17) finds that across protests in Egypt, Spain, and the US, “Facebook is used as a recruitment platform”, while “Twitter is mainly employed as a means of internal coordination within the activist community”. The similar use of these platforms across such varied contexts suggests that certain architectural components of platforms contribute to how they are deployed for protest activity.

Social media digital architectures  231

So far, we’ve seen how the structure of intra-platform spaces, algorithms, and limitations on character limits can influence how users leverage platforms to achieve a political function, such as protests. These platform features (and the limits placed on them) have direct implications for users’ online platform use, but they can also affect users’ offline political agency. Without Facebook pages, the Aganaktismenoi movement might not have happened (or at least, it would have manifest in a different way). Thus, digital structures matter not only because they shape online discourses; a focus on digital architectures can illuminate how platform design can structure political processes outside of a platform.

DIGITAL ARCHITECTURES AFFECT ALL USERS The political processes that digital architectures affect are not limited to citizens and their protest mobilization. Digital architectures shape the activity of all users online, and they are therefore relevant to consider for any political actor. In political campaigning, for example, the bottom-up emergence of Facebook pages and groups has impacted official campaign messaging strategies and internal organizational hierarchies (Penney, 2017). Similarly, Facebook’s targeting and advertising features have influenced political campaign organization and operation (through increasing staff and resources around digital outreach) and likely, a concrete impact on governance through helping candidates get elected who otherwise might not. Focusing on political parties and leaders, Haßler et al. (2023) provide an excellent example of how digital architectures can be used to structure an analysis of the visual political communication on Instagram. For the mainstream media, digital architectures influence how news articles on a platform are presented, which limits journalists’ ability to signal the importance of a story through traditional cues such as its position in a paper or the size of text accompanying the story. Instead, the presentation of news is shaped by a platform’s architecture, which generally sets a predefined format for both text and pictures. Crucially, algorithms influence the reach of news content, which devolves control over news dissemination from the media to the platform. Since digital architectures shape the actions of journalists and the news that users attract online, these architectures likely exhibit demonstrable effects on modern journalistic reporting practices as well as citizens’ levels of political knowledge. For example, research finds that some users have become so reliant on platform algorithms for news that they disengage from news-seeking altogether (Gil de Zúñiga et al., 2017). Other, more niche actors will leverage certain aspects of digital architectures to fit their political needs. For example, Mitts (2022) shows how offline counter-extremism events correlate with increased mentions of Telegram among terrorist sympathizers, who leverage the platform’s encryption to dodge law enforcement surveillance. Furthermore, malicious actors engaging in coordinated inauthentic behavior such as harassment or the spread of disinformation need to work within the rule systems of platforms (and often try to “game” them through the mass creation of fake accounts

232  Handbook of digital politics

or deployment of bots). Thus, both normatively positive and negative political behaviors are mediated by a platform’s digital architecture. As these examples from various political actor-types highlight, an architectural approach to social media entails identifying the specific aspects of a platform’s form to understand how they function politically for users. In the next section, I outline this conceptual process of “mapping”, before applying it to the concrete example of citizens’ political participation.

MAPPING DIGITAL ARCHITECTURES FOR POLITICS Social media platforms are designed products. From the platform side, features are first conceptualized in a planning stage and then built through code, tested on users, and refined based on user feedback. Therefore, platform companies design features to enact certain functions, and how those functions are achieved is built through a series of encoded rule systems (i.e., the digital architecture). Given the scale of social media platforms, these features need to be designed in ways that are easily interpretable and accessible for a wide range of different users (who will vary, for example, in the hardware they use to access the platform, their technological proficiency, and their intentions for using a platform). In short, this means that complex rule systems on the back-end (platform side) deliver rather simplistic functionalities on the front-end (user side). When we log into a social media app, what we see is a highly optimized product that has been designed to be as simplistic and user-friendly as possible for a massively diverse array of people. I argue that even though researchers cannot see the developer code on the platform back-end, we can conceptually map how certain front-end features shape users’ political communication practices and behavior. Given that the front-end of platforms are readily observable and designed to be simplistic, an architectural approach can feasibly excavate which specific components of a platform matter for politics by approaching social media platforms as designed products. In other words, researchers can observe a social media’s form and conceptualize its political function for users. While we can only speculate about what a certain feature accomplishes for a platform company, we can build theories and research designs around how and why these features enable, constrain, or shape particular political processes. How can we wade through the various design features of social media platforms to decide which ones matter for politics? In product design, the first step is to conceptually separate the functions of a product – what it does – from the product’s arrangement of physical (or digital) characteristics (Ulrich, 1995). For social media, we can identify several basic functions that are shared across platforms: connecting users, allowing them to create and engage with content, ranking this content to be displayed to users, and generating data for marketers. These functions serve specific purposes for platforms (such as attracting users and monetizing their activity) and are not necessarily designed with political implications in mind. However, each of these functional categories can matter for politics, depending on how specific users

Social media digital architectures  233

leverage a platform to achieve their political goals. Thus, a good first step in excavating digital architectures is to conceptualize the basic functions that social media platforms share as a genre. Once the core functions of a platform are identified, the next step is mapping, “where the functions to be implemented are assigned (mapped) to the components of the architecture” (Ferrari and Sangiovanni-Vincentelli, 1999, p. 5). It is here where we begin to see that similar functions take different architectural forms across platforms. For example, the basic function of connecting users differs on Facebook and Twitter. Both platforms facilitate connections between users but do so through different rule systems, which create the “Friend” versus “Follow” rule systems respectively. Empirical research finds that these different user networks (i.e., those built on strong-tie [Friend] versus weak-tie [Follow] connections), promote different pathways for citizens to engage in political activities, such as protest participation (Valenzuela et al., 2018). Even though both Facebook and Twitter can encourage protest activity, they do so through different dynamics that result from architectural differences in how users connect on a platform. This example should illustrate that we do not need to know why Facebook and Twitter implemented the Friend versus Follow mechanism, and we do not need to know how exactly these rule systems work in terms of software development or engineering. Rather, we can observe that both platforms exhibit a shared function (connecting users – or “connectivity”), and map that function onto specific rule systems of the platform (the Friend and Follow mechanisms). Other platforms, like Snapchat, have a similar Follow mechanism to Twitter, but user privacy settings are strict by default, lending to a more a Friend-like structure that users must toggle off to trigger a Follow mechanism. On Instagram, privacy settings are open by default resulting in a Follow structure, but many users toggle their privacy settings to support a Friend structure. Furthermore, we might contrast those systems to a platform like YouTube, where users can interact through comments but only establish enduring connections to content creators through channel subscriptions. So far, this demonstration of mapping has been largely apolitical, which is why I consider digital architectures a “platform-first” approach. Once functions are mapped onto platforms’ digital architectures, scholars can begin to conceptualize and test which elements interact with their political process of interest. For example, one could build theory-driven hypotheses about how different network structures influence protest participation (Valenzuela et al., 2018) or how certain digital architectures might help explain a political process through exploratory analysis of a case study (such as the role of a social media platform in a particular election or social movement). Irrespective of whether one deploys digital architectures as a cause or explanation for political phenomena, it is crucial to take a platform-first approach and map out the architectural components that are important to consider for any given social media analysis. The reason is that platforms change constantly, and scholars have largely missed the opportunity to archive these changes in the scholarly literature. Through mapping digital architectures and both conceptualizing and testing their impact on

234  Handbook of digital politics

politics, scholars can build a better archive of platform changes while accumulating knowledge about how specific design elements shape political behavior. This is why I advocate for an architectural approach over a focus on affordances. In being highly relational between user and technology, there will never be a singular theory of affordances (Evans et al., 2017). Instead, scholars continue to develop new types of affordances, resulting in a “proliferation of adjectives” that confounds, rather than clarifies, a unified understanding of the concept (Marsteintredet and Malamud, 2020). Bucher and Helmond (2018), for example, identify eight such adjectives: perceived affordances, technology affordances, social affordances, communicative affordances, high-level affordances, low-level affordances, imagined affordances, and vernacular affordances. The multiplicity of affordances means the concept is building out horizontally into more and more categories (which often overlap), rather than vertically toward conceptual clarity. Instead, I propose to start at the platform-level and map out digital architectures, thereby building up the ladder of abstraction vertically (Sartori, 1970). Not only will this provide conceptual clarity regarding the components of platforms that matter for political processes, it will also facilitate the refinement of methodologies to test architectural components while archiving platform changes in the process.

DIGITAL ARCHITECTURES AND POLITICAL PARTICIPATION In previous research, I have mapped four aspects of a platform’s digital architecture – network structure, functionality, algorithmic filtering, and datafication – and shown how they shape practices of American political campaigning (Bossetta, 2018). To highlight how a platform-first approach to mapping digital architectures provides conceptual clarity and flexibility in political research designs, I take these same categories and argue that they also influence citizens’ political participation in various ways. Conceptually, this exercise illustrates an important point alluded to previously: politically relevant components of a digital architecture should be applicable across user-types and be extendable to new platforms. Thus, a key task of digital architectures research is to identify the shared technical protocols that define social media as a genre, and then drill down into how slight or major differences across platforms affect how they function politically for different user-types (e.g., politicians, journalists, citizens, and even platform companies themselves). Defining Political Participation First though, we need to define political participation. Traditionally in political science, political participation has been considered the physical “acts” (Verba and Nie, 1972, p. 2) that citizens take to influence politics. A focus on physical acts makes little sense when studying political participation through social media, where citizens’ physical actions are limited to mouse clicks, keystrokes, and finger taps.

Social media digital architectures  235

Instead, I conceptualize political participation as a process comprising three phases: latent participation, manifest participation, and mobilization (Dutceac Segesten and Bossetta, 2017, p. 1627). The first phase is latent participation, defined as citizens’ “readiness and willingness to take political action” (Ekman and Amnå, 2012, p. 296). Forms of latent participation include reading political news, discussing politics with others, and reflecting upon one’s political viewpoints. These latent forms of participation condition and prime manifest participation, which refers to citizens’ actions aimed at affecting “politics and political outcomes in society, or the decisions that affect public affairs” (Ekman and Amnå, 2012, p. 289). These manifest forms of participation can be electoral, such as voting or canvassing, as well as extra-parliamentary, such as protesting or taking everyday actions that attempt to curb climate change. The third phase, mobilization, refers to the transition from latent to manifest participation. A citizen can either “be mobilized” by an external event (such as voting in an election to fulfil a civic duty or protesting in response to a legal change) or citizens can actively “be mobilizers” by calling others to action. Importantly, this political participation process is not necessarily linear. Citizens may move across the three phases in various contexts and in response to particular circumstances, or they may only enact one phase (such as being engaged in latent participation but never becoming mobilized to enact manifest participation). Conceptualizing political participation as a process accommodates the various ways that citizens can participate in politics online as well as offline. Moreover, it also reflects how manifest forms of participation are linked to prior attributes such as political knowledge, which is accumulated through latent participatory forms like consuming political news and discussing politics with others. Importantly, while this understanding of political participation is conceptually broad, it also keeps the door open for new forms of online political participation that may arise in the future. For example, few would have foreseen Snapchat’s “Run for Office” feature, where users can say “I want to run for office” and Snapchat will provide them links to the necessary resources and partner organizations for starting a political campaign. This platform-mediated form of latent participation may very well translate into some users running for office, highlighting how platforms can enable the mobilization link between latent and manifest participation. Network Structure Network structure refers to the technical protocols that govern how connections between users are initiated, established, and maintained. Differences in the protocols underpinning a platform’s network structure affect at least three aspects of user connections. The first is searchability, which refers to how users can identify new accounts and subscribe to their content. The second is connectivity, referring here to how connections between accounts are initiated and established (discussed above in the Friend versus Follow mechanisms). The third aspect of network structure is

236  Handbook of digital politics

privacy, which allows users to influence who can identify them through searches (searchability) as well as how connections become established (connectivity). In the context of political participation, network structures are crucial since they influence the sources of political content or mobilization (e.g., a close friend, influencer, journalist, politician, or NGO), the configuration of potential discussion partners, and levels of involvement, attachment, or trust in a virtual community. Whereas some platforms like Facebook, Instagram, and Twitter suggest connections based on recommender systems (and allow exchanges between non-connected users in public commenting spaces), other platforms like Signal, Telegram, and WhatsApp require specific information about a potential connection (such as a phone number). In requiring such specific information, the resulting network structure on these encrypted messaging services will likely support a highly dense, private, and less searchable network. Just as political campaigns use these platforms to coordinate the canvassing activities of ardent volunteers, activists and terrorists can leverage the architectures of encrypted messaging apps to plan and carry out extra-parliamentary forms of activism, like protests or politically-motivated violence. Functionality Functionality refers to the technical protocols allowing users to access, create, and redistribute content through interaction with the platform interface. Here, hardware is a key component to consider: is the platform accessible through a computer, smartphone, or wearable accessory like a smartwatch, eyewear, or a virtual reality headset? Second, the design of the Graphical User Interface (GUI) is crucial for how users learn about and navigate through the various features and spaces offered by the platform. Functionality’s third category is the broadcast feed, which aggregates, ranks, and displays content based on computational rule systems (i.e., algorithms). A fourth component is supported media, comprising the multimedia formats that the platform supports technically (e.g., text, images, video, GIFs), the size and length constraints placed on acceptable media (text character limits or video lengths), and the rules governing hyperlinking. The fifth element of functionality is cross-platform integration, which refers to whether content generated on one platform is technically supported by another. The functionality of a platform will impact whether and how citizens can leverage it for latent participation, manifest participation, and mobilization. For example, a broadcast feed that aggregates politically relevant content (such as Facebook’s News Feed or Twitter’s Timeline) is conducive for latent participation in the form of reading news. However, limitations on Twitter’s character limit impact its utility for long-form discussions about politics (Freelon, 2015), whereas platforms with less restrictive limits – like Reddit – may be conducive to thoughtful political exchanges. In terms of mobilization and spreading calls to action, platforms that lack a feature to share into the broadcast feed (such as private messaging apps) may be less impactful in organizing large-scale manifest action, such as protests. However, a controversial video first shared through these platforms may then be seeded onto more public plat-

Social media digital architectures  237

forms and spark mobilization, insofar as the video would be supported technically (i.e., cross-platform integration). As users toggle between platforms on a daily basis, cross-platform integration is important to consider in terms of how the same content can also travel across platforms. The best example here is TikTok, which is not a particularly political platform but does host users broadcasting videos tied to social issues. One can share videos from TikTok to Instagram directly, and the vertical video format popularized by TikTok is highly interoperable with other platforms. A TikTok video can be shared onto platforms like Twitter, Facebook, and Instagram, with the latter two even designing “Reels” to compete with TikTok (while also hosting TikTok content on their own platforms as Reels). Thus, functionality is an important architectural component for both single and cross-platform research designs. If two or more platforms share similar functionalities, it is possible that the same political content will be shared across them. Scholars can therefore seek to investigate how the same forms of participation are issued across platforms – such as posting a mobilizing call to action – and work to explain why (from an architectural perspective) certain forms of participation are impactful on some platforms but not others. Furthermore, scholars should be aware that some platforms allow third-party programs to “plug-into” the functionalities of platforms’ existing digital architecture. For example, the company uCampaign developed a gamified app for conservative campaigns (including the Trump 2016 and Brexit Vote Leave campaigns). These apps encourage users to attend rallies in person and share posts from the campaigns on mainstream social media to earn “Action Points” (Bossetta, 2022). While encouraging a gamified version of political participation, these apps serve as coordination hubs that crowdsource user engagement in ways that can game platform algorithms, thereby increasing the organic reach of a campaign’s message. Thus, when thinking about a how a platform’s functionality affects participation, one should consider the role of individual features, a platform’s overall “suite” of collective features, and how these features may interact with other platform technologies in the social media or digital campaigning ecosystem. Algorithmic Filtering Algorithmic filtering structures how content is ranked and displayed through computational rule systems. For political participation, algorithmic filtering influences both what type of content users are exposed to (on the reception-side) as well as how far a post is disseminated across users’ broadcast feeds (on the production-side). Although the exact operation of algorithms is often “black-box” and difficult to discern, the rule systems guiding algorithms’ filtering can be based on factors such as relevance or chronology. An algorithm is guided by predicted relevance when content is presented to a user based on their previous behavior on the platform. Examples include Facebook’s News Feed algorithm and YouTube’s WatchTime algorithm. Other platforms, like Twitter and now Instagram, prioritize filtered chronology by presenting posts in

238  Handbook of digital politics

reverse chronological order from the time of user log-in. Meanwhile, direct message platforms like Snapchat and private messaging apps have low levels of algorithmic filtering, and content is displayed to users based on unfiltered chronology; i.e., in the chronological order that messages are received. While algorithmic filtering is mostly important for platforms with a broadcast feed, it is important in the context of participation because algorithms can be manipulated to increase the exposure of particular posts. For example, the coordination by activists (through, for example, the third-party apps previously mentioned), the mass creation of fake accounts, or the deployments of bots and botnets can send a post or hashtag trending to increase its visibility. Moreover, most platforms with a broadcast feed allow business accounts set up by organizations to override algorithmic filtering by paying to promote content through boosting or advertising. By overriding algorithms to place content in front of users who may not see it otherwise, organizations can push messages or aim to grow their online audiences for future mobilization initiatives. Datafication Datafication dictates how users’ activity is quantified and measured. Although datafication primarily serves platforms’ internal development by gauging user activity for future optimization, or for political campaigns to segment and target voters, datafication is also rendered to users through the interactions data they receive on their posts. Nearly all platforms provide feedback to users based on how their posts are performing, in terms of quantifiable metrics such as “Likes” or qualitative indicators such as “Comments”. In terms of political participation, these datafied responses to users’ political posts may condition how users leverage a platform for political discussion or community-building. Bail (2021), for example, notes how politically-motivated Twitter users signal partisanship to increase their status on the platform, especially when they lack substantial offline relationships, steady employment, or mental well-being. Thus, receiving engagement from like-minded partisans on social media may encourage a spiral of increasingly extreme partisan expressions, which influences the dynamics of online political talk (a form of latent participation). Alternatively, as one digital campaigner told me during an episode of the Social Media and Politics podcast (Bossetta, 2019), online volunteers for a political campaign can form “Welcome Teams” that actively scout out and collectively engage with posts from users who express positive sentiments or ask questions about a candidate. The intent is to generate a dopamine rush for that user (through receiving a surge of interactions) and create positive associations with the campaign. These examples show how the datafication of post reactions may influence the content of users’ posts, how they interact with others, and potentially their recruitment into the online canvassing initiatives for political campaigns or organizations.

Social media digital architectures  239

FUTURE AVENUES FOR DIGITAL ARCHITECTURES RESEARCH Although these examples are far from an exhaustive list, hopefully this chapter has inspired you to think about how digital architectures shape social media users’ political agency online and offline. Borrowing theories from product and systems design, I have offered a conceptual roadmap to map political functions to platform architectures as an alternative heuristic to affordances. In my view, an architectural approach offers more promise for building conceptual clarity and operational precision through political research designs. However, some readers may wish to view the two approaches as complementary, with digital architectures giving rise to affordances. That is, any discussion of social media affordances necessitates thinking about the elements of platform architecture that make the affordance possible (e.g., “the affordance of visibility through hashtags” or “the affordance of anonymity through account creation protocols”). So, one could start by mapping a digital architecture and then theorizing about the affordances it offers. Yet, I maintain that the concept of affordances is vague, lacks analytical utility, and should be abandoned (Oliver, 2005). Scholars interested in ground truth should avoid such imprecise constructs. Either way, a critical task for research is to deconstruct, isolate, and test the components of digital architectures that matter for politics. Experiments are perfectly suited for this endeavour, where particular architectural features can be varied across treatment and control groups. Mock-up platforms, while losing ecological validity in relation to real platforms, can still isolate and test the effects of specific platform features on political behavior outcomes. Studies of digital trace data from platforms are also valuable for exploring digital architectures (Haßler et al., 2023). Especially for cross-platform studies, architectural mapping can be performed to generate confirmatory hypotheses that compare user behavior across platforms. For single platform studies, the formulation of hypotheses testing the effects of digital architectures will be more difficult (due to the lack of a comparative benchmark); however, digital architectures may help explain how users leveraged a particular platform during the context of a specific election or social movement. Additionally, qualitative interviews and focus groups can be a particularly powerful method to understand how and why social media users leverage certain platforms for political participation (as demonstrated earlier in this chapter by the quotes from the Greek anti-austerity activist). Similarly, ethnographic approaches such as participant observations could provide first-hand insight into how digital architectures shape users’ political agency in practice. In closing, one potential critique of an architectural approach is that since platforms evolve at a rapid pace, their digital architectures may be unstable objects of analysis. However, social media platforms share a set of core functions that are fundamental to the genre, slow to change, and can be mapped onto architectures at specific points in time. Archiving these changes, while mapping new functions and architectures as they emerge, will build a solid conceptual foundation and appropriately flexible

240  Handbook of digital politics

heuristic to assess the impact of digital structures on political agency – now and in the future.

FURTHER READING Bossetta, M. (2018). The digital architectures of social media: Comparing political campaigning on Facebook, Twitter, Instagram, and Snapchat in the 2016 US election. Journalism & Mass Communication Quarterly, 95(2), 471–496. Freelon, D. (2015). Discourse architecture, ideology, and democratic norms in online political discussion. New Media & Society, 17(5), 772–791. Jones, Q. and Rafaeli, S. (2000). Time to split, virtually: ‘Discourse architecture’ and ‘community building’ create vibrant virtual publics. Electronic Markets, 10(4), 214–223. Haßler, J., Kümpel, A. S., and Keller, J. (2023). Instagram and political campaigning in the 2017 German federal election: A quantitative content analysis of German top politicians’ and parliamentary parties’ posts. Information, Communication & Society, 26(3), 530–550. Oliver, M. (2005). The problem with affordance. E-Learning and Digital Media, 2(4), 402–413. Wright, S. and Street, J. (2007). Democracy, deliberation and design: The case of online discussion forums. New Media & Society, 9(5), 849–869.

REFERENCES Bail, C. (2021). Breaking the Social Media Prism: How to Make Our Platforms Less Polarizing. Princeton: Princeton University Press. Barrett, B. and Kreiss, D. (2019). Platform transience: Changes in Facebook’s policies, procedures, and affordances in global electoral politics. Internet Policy Review, 8(4), 1–22. Bossetta, M. (2018). The digital architectures of social media: Comparing political campaigning on Facebook, Twitter, Instagram, and Snapchat in the 2016 US election. Journalism & Mass Communication Quarterly, 95(2), 471–496. Bossetta, M. (Host) (2019). Online engagement and digital campaigning for Pete Buttigieg, with Stefan Smith (No. 104) [Audio podcast episode, April 12]. Social Media and Politics. https://​so​cialmediaa​ndpolitics​.org/​online​-engagement​-and​-digital​-campaigning​-for​-pete​ -buttigieg​-with​-stefan​-smith/​. Bossetta, M. (2022). Gamification in politics. In A. Ceron (ed.), Elgar Encyclopedia of Technology & Politics (pp. 304–307). Cheltenham, UK and Northampton, MA, USA: Edward Elgar Publishing. Bucher, T. and Helmond, A. (2018). The affordances of social media platforms. In J. Burgess, T. Poell, and A. Marwick (eds.), The SAGE Handbook of Social Media (pp. 233–255). London: Sage. Dutceac Segesten, A. and Bossetta, M. (2017). A typology of political participation online: How citizens used Twitter to mobilize during the 2015 British general elections. Information, Communication & Society, 20(11), 1625–1643. Ekman, J. and Amnå, E. (2012). Political participation and civic engagement: Towards a new typology. Human Affairs, 22(3), 283–300. Evans, S. K., Pearce, K. E., Vitak, J., and Treem, J. W. (2017). Explicating affordances: A conceptual framework for understanding affordances in communication research. Journal of Computer-Mediated Communication, 22(1), 35–52.

Social media digital architectures  241

Ferrari, A. and Sangiovanni-Vincentelli, A. (1999). System design: Traditional concepts and new paradigms. Proceedings 1999 IEEE International Conference on Computer Design: VLSI in Computers and Processors (pp. 2–12). IEEE. Freelon, D. (2015). Discourse architecture, ideology, and democratic norms in online political discussion. New Media & Society, 17(5), 772–791. Gerbaudo, P. (2012). Tweets and the Streets. London: Pluto Press. Gil de Zúñiga, H., Weeks, B., and Ardèvol-Abreu, A. (2017). Effects of the news-finds-me perception in communication: Social media use implications for news seeking and learning about politics. Journal of Computer-Mediated Communication, 22(3), 105–123. Haßler, J., Kümpel, A. S., and Keller, J. (2023). Instagram and political campaigning in the 2017 German federal election: A quantitative content analysis of German top politicians’ and parliamentary parties’ posts. Information, Communication & Society, 26(3), 530–550. Jones, Q. and Rafaeli, S. (2000). Time to split, virtually: ‘Discourse architecture’ and ‘community building’ create vibrant virtual publics. Electronic Markets, 10(4), 214–223. Marsteintredet, L. and Malamud, A. (2020). Coup with adjectives: Conceptual stretching or innovation in comparative research? Political Studies, 68(4), 1014–1035. McGrenere, J. and Ho, W. (2000). Affordances: Clarifying and evolving a concept. In S. Fels and P. Poulin (eds.), Proceedings Graphics Interface 2000 Conference (pp. 179–186). Canadian Human-Computer Communications Society. Mitts, T. (2022). Countering violent extremism and radical rhetoric. International Organization, 76(1), 251–272. Norman, D. (1999). Affordance, conventions, and design. Interactions, 6(3), 38–43. Oliver, M. (2005). The problem with affordance. E-Learning and Digital Media, 2(4), 402–413. Penney, J. (2017). Social media and citizen participation in “official” and “unofficial” electoral promotion: A structural analysis of the 2016 Bernie Sanders digital campaign. Journal of Communication, 67(3), 402–423. Sack, W. (2005). Discourse architecture and very large-scale conversation. In R. Latham and S. Sassen (eds.), Digital Formations: IT and New Architectures in the Global Realm (pp. 242–282). Princeton: Princeton University Press. Sartori, G. (1970). Concept misformation in comparative politics. American Political Science Review, 64(4), 1033–1053. Savor, T., Douglas, M., Gentili, M., Williams, L., Beck, K., and Stumm, M. (2016). Continuous deployment at Facebook and OANDA. In 2016 IEEE/ACM 38th International Conference on Software Engineering Companion (ICSE-C) (pp. 21–30). IEEE. Theocharis, Y., Lowe, W., Van Deth, J. W., and García-Albacete, G. (2015). Using Twitter to mobilize protest action: Online mobilization patterns and action repertoires in the Occupy Wall Street, Indignados, and Aganaktismenoi movements. Information, Communication & Society, 18(2), 202–220. Treré, E., Jeppesen, S., and Mattoni, A. (2017). Comparing digital protest media imaginaries: Anti-austerity movements in Greece, Italy & Spain. TripleC: Communication, Capitalism & Critique, 15(2), 404–422. Ulrich, K. (1995). The role of product architecture in the manufacturing firm. Research Policy, 24(3), 419–440. Valenzuela, S., Correa, T., and Gil de Zúñiga, H. (2018). Ties, likes, and tweets: Using strong and weak ties to explain differences in protest participation across Facebook and Twitter use. Political Communication, 35(1), 117–134. Verba, S. and Nie, N. H. (1972). Participation in America. New York: Harper & Row. Wright, S. and Street, J. (2007). Democracy, deliberation and design: The case of online discussion forums. New Media & Society, 9(5), 849–869.

16. Artificial intelligence in politics Leah Henrickson

INTRODUCTION Artificial intelligence (AI) is a hot topic. In just the last decade, we’ve seen films about lonely men falling in love with AI voice assistants, robot overlords taking over, and basketball-loving avatars. We’ve heard advertisements for AI-powered software to help write university essays, garner business intelligence, and maintain health and fitness. News articles about the latest AI technologies abound. Whether representations of AI are positive or negative, it’s clear that AI is pervasive and powerful – or at least seems to be. But what is AI, exactly? What do developments in AI mean for digital politics? How might we be able to harness AI for new insights into socio-political happenings, and for influencing political discourse? This chapter begins with an introduction to AI, before reviewing some of the many ways that AI is currently being used in politics and how AI may be used in politics research. I conclude with a discussion that highlights AI’s potential to influence, illuminate, and impair public consciousness. Although AI may help us complete tasks, make sense of data, and share information faster and more effectively than ever before, we must strive towards AI that’s recognized as the fundamentally human technology that it is. Politics is, after all, about people, and our use of AI must support rather than contradict this human emphasis.

A BRIEF INTRODUCTION TO AI AI is a fast-moving and ever-changing area of technological experimentation and theoretical rumination. Developers are trying to come up with the next ‘big thing’. Scholars are reflecting on the implications of previous ‘big things’ for their respective fields. Artists are responding to these efforts by imagining and reimagining presents and futures. Journalists are trying to bring all of these perspectives together in ways comprehensible to the general public. Most of these contributions assume a basic level of understanding of what AI is, though. Below, I provide an explanation that aims to fill that ‘basic understanding’ gap. The term ‘artificial intelligence’ is often attributed to American computer scientist John McCarthy, coined in a proposal for a 1956 event at Dartmouth College. ‘We propose that a 2 month, 10 man study of artificial intelligence be carried out during the summer of 1956 at Dartmouth College in Hanover, New Hampshire.’ McCarthy et al.’s proposal (1955, p. 2) begins: 242

Artificial intelligence in politics  243

The study is to proceed on the basis of the conjecture that every aspect of learning or any other feature of intelligence can in principle be so precisely described that a machine can be made to simulate it. An attempt will be made to find how to make machines use language, form abstractions and concepts, solve kinds of problems now reserved for humans, and improve themselves.

Before the Dartmouth event, work in what would now be called AI was well underway, with some scholars even attributing ancient scholarship and storytelling to the field’s lineage (Cave et al., 2020). AI research continues to be driven by efforts to simulate human intelligence, whether through language usage, conceptual abstraction, or problem solving. AI is, put simply, any effort to mimic or match (usually human) intelligent behaviour. AI can be loosely categorized into two types: narrow and general. Narrow systems are trained in specific domains, only capable of completing a limited number of tasks addressing predefined problems. General systems can respond to a broader array of tasks, capable of navigating multiple domains simultaneously and identifying appropriate solutions for problems that may not have been anticipated. Most of our current AI is narrow. A system might be able to play chess or help us select the most lucrative investments, but those same systems would not cross into one another. A chess-playing AI system won’t be much use if you were to ask it what stocks to buy. Substantial research continues to strive towards general AI, however, with speculation about ‘super AI’ or ‘superintelligence’: AI that outperforms humans in decision-making tasks (Bostrom, 2014). Representations of AI are often embodied. For example, news articles about AI may be accompanied by photos of robots in various forms. These robots, some scholars have observed, are usually white in colour, and are often made to look ethnically White when given more ‘human’ appearances (Cave and Dihal, 2020). However, the fields of robotics and AI are distinct. AI is usually unembodied, largely contained to motherboards and processing units. Indeed, at its core, any AI system is really just a set of statistical probabilities. The system determines these probabilities, and its corresponding responses, by following sets of instructions called algorithms. Although the term ‘algorithm’ is widely used to refer to virtually any kind of computational functionality, it more specifically refers to a coded instruction or set of instructions that a system uses to complete a task. It is a step-by-step guide to solving a problem using a defined dataset. For example, ‘the TikTok algorithm’ that offers personalized video recommendations to TikTok users applies a set of evaluative criteria to the vast dataset of videos uploaded to the platform; it identifies content and viewership patterns to solve the ‘problem’ of personalized recommendations that keep users active. AI system output always depends upon patterns identified across data. In instances of machine learning (ML), those patterns can be identified through reinforcement learning, supervised learning, or unsupervised learning. Reinforcement ML rewards desirable outcomes and punishes undesirable outcomes; these systems learn from trial and error. Supervised ML requires labelled data to start; a system then tries to

244  Handbook of digital politics

find patterns across these data, adjusting itself as needed when faced with larger and more diverse datasets. Unsupervised ML doesn’t require labelled data to start; systems create the labels themselves. Even though unsupervised ML doesn’t require human labour to find patterns, it still requires human moderation and curation of results. To be sure, all of these kinds of ML necessitate human involvement in different capacities. This human involvement means that AI systems are never neutral. Rather, they are always the result of overt and covert human labour, representing human intention that so very often perpetuates or creates power imbalances (Crawford, 2021). Humans write the algorithms, comprise and provide the datasets, maintain the systems, and curate the output. As long as we are human, we will never be able to create a totally unbiased AI system. This is both a bad and good thing. On the one hand, we can encode harmful biases, intentionally or unintentionally, into systems, contributing to the marginalization of certain groups. However, we can also encode efforts to overcome the marginalization of these groups. It is therefore vital that we take time to consider who is developing systems, why, how, and for whom. These are questions related to the power and politics of AI.

APPLICATIONS OF AI As I write this chapter, we are in the midst of a crisis in Ukraine. While Ukraine– Russian relations have long been tense, the world is currently witnessing an atrocious series of attacks on Ukrainian soil. Although Russian media refer to these attacks as a ‘military operation’, they more accurately constitute a war. It is still too early to conclusively determine how AI is being used in this war – to influence popular discourse, to inform military decision making, and so forth – but we are seeing its use. Given its international prominence, the Ukraine crisis will serve as my primary source of illustration for the ‘AI in Politics’ section. My use of this crisis throughout this chapter serves two purposes: to provide timely examples of AI in real-world, high-stakes contexts, and to contribute to a snapshot of what will undoubtedly be regarded as a historically momentous time. The use of such examples is especially appropriate for a discussion of AI given that much of the field’s development has been driven and funded by national governments and their militaristic organizations. Note that the global context I present here is informed by my being embedded in an Anglocentric context at the time of writing. This section is divided into two sub-sections: AI in Politics and AI in Politics Research. As I adopt a survey approach, citations are used liberally to signpost readers to complementary resources. AI in Politics AI is now evident in virtually every part of our social infrastructure, including politics. Some scholars have published surveys of AI in public governance, but these

Artificial intelligence in politics  245

surveys are few and vary in their critical depth (for example, Araya and King, 2022; Coeckelbergh, 2022; Gupta and Pal, 2021; Mehr, 2017; Zuiderwijk et al., 2021). In this chapter, I focus on three applications of AI in politics: logistical assistance, forecasting and decision making, and information production and dissemination. These are loose categories, and my overview is not comprehensive. Further, the examples I draw from are all narrow AI, developed for particular applications in pre-specified domains. While SAM – ‘the world’s first Virtual Politician’, whose views evolve with voter input – emerged in the New Zealand political scene in 2017, the concept of a virtual politician has achieved neither the popularity nor the traction to warrant more than a fleeting mention (Sam, n.d.). Yet the potential influence of virtual politicians like SAM raises questions about AI being granted legal personhood, which are hotly debated (Bryson et al., 2017). In most cases, the AI systems at the centre of the personhood debates are examples of general AI. Logistical Assistance AI systems are especially well suited for completing routine tasks with predictable data. These tasks might include scheduling appointments, answering simple queries, and filling out forms. Chatbots – virtual assistants interacted with through text or speech – have become popular options for providing such logistical assistance, although some have lambasted their use in political settings (Susskind, 2018). Of course, politics is not the only field to attempt increased operational efficiency supported by AI (Tariq, 2021). Logistical systems may also prove useful for fields like law, which draw from massive corpora of data that require substantial amounts of time and memory to navigate (Dervanovi, 2018). Logistical assistance may also extend to the area of machine translation (MT): automated translation from one natural human language to another. MT has a long history, much of which has been characterized by manual efforts to encode words and grammar in ways that allow computers to automatically identify like terms and syntactic structures across different languages (Mitchell, 2022; Slocum, 1985). Although MT still depends on human moderation, modern MT integrates ‘deep learning’ ML structures like neural networks and generative adversarial networks (GANs) to automate parts of the encoding process (Dabre et al., 2021). Language is political, though, and the implications of incorrect translations could be serious. In 2016, Google rushed to fix some automatically embedded errors in Google Translate’s Ukrainian to Russian service: ‘Russian Federation’ became ‘Mordor’ (the hellish Lord of the Rings setting); ‘Russians’ became ‘occupiers’; and the surname of Russia’s Foreign Minister Sergey Lavrov became ‘sad little horse’. A statement from Google explained that its Translate service finds patterns across hundreds of millions of documents; in this case, the incorrect translations appeared to reflect the language used by some Ukrainians following the 2014 Crimea conflict (BBC, 2016). It is still unclear precisely when these translations began to appear in Translate results, or whether these translations were intentionally inputted to denigrate Russians in more roundabout ways. However, politically charged mistrans-

246  Handbook of digital politics

lation between Ukrainian and Russian is not unprecedented, even without the use of MT tools (Kolomiyets, 2020). In this instance, the Google Translate service had mechanized human ‘error’ by capturing and encoding human textual experiences. In addition to its linguistic applications, AI is frequently used for visual analysis in what is called ‘computer vision’. Computer vision can be used to classify, sort, and monitor aspects of the visual world. For example, it is computer vision that constitutes much of the software embedded in self-driving cars. If I am letting my Tesla take control, I want to feel confident that it knows to stop when it ‘sees’ someone crossing the street. Computer vision also includes facial recognition software, which is applied in military contexts, as is the case with its current uses to identify Russian soldiers (Simonite, 2022b). Facial recognition may too be used to help identify the dead (Dave and Dastin, 2022). However, we know that the datasets used to train facial recognition software are often dominated by light-skinned people with masculine features, resulting in inaccurate identifications that may have dire consequences (Buolamwini and Gebru, 2018). Recall the countless times these systems have miscategorized darker-skinned people as ‘gorillas’ and West Asian people as ‘blinking’ (Waelen, 2022). So, while AI can offer useful logistical assistance to make our lives easier, this assistance can also be used to inflict harm, intentionally or not. Such potential affirms a truth long accepted by mystery novel fans: the butler is capable of so much more than simply following orders. Forecasting and Decision Making To be sure, AI systems do do more than simply follow orders. Some systems are also being programmed to forecast future events by sifting through historical and current data. For example, the use of predictive analytics – which have depended on ‘big data’ (massive and complex datasets subject to AI analysis) – in multiple US Presidential election cycles has been observed, but academic documentation of this use is sparse (Laterza, 2021; Nickerson and Rogers, 2014). Indeed, we are not yet sure of the effectiveness of algorithms for influencing voter behaviour because studies of algorithmic influence yield mixed results (Agudo and Matute, 2021; Rusch et al., 2013). If these forecasting systems do work, they may help us better prepare for upcoming events and more effectively mitigate potential risks through automated decision making. We are already seeing efforts to apply AI this way in response to social issues like climate change, population planning, and resource distribution. But AI is not just being used for predictions – it is also being used to aid decision making by making its own recommendations. For example, ML models can help discover new ways of combatting toxins in the human body. When inverted, though, these same models can suggest the development of new chemical warfare agents (Urbina et al., 2022). The same training set and technology may be used for either purpose and, as the researchers using these models have emphasized, human monitoring of the technology is vital to ensure that the models don’t spiral into harm.

Artificial intelligence in politics  247

At its most extreme, AI may be combined with deadly machinery to become what we call ‘autonomous weapons’ or ‘killer robots’. There are, of course, many social, logistic, and moral issues with this kind of weaponry. To start, who is responsible for their behaviour? The jury is still out (Taylor, 2021). A large part of why we are not able to definitively assign responsibility for these systems is that, bluntly, we don’t always know precisely how they are reaching their conclusions. This is an obvious problem. After all, if my killer robot is committing murder on my behalf, I would like to know why. Thankfully, the area of explainable artificial intelligence (XAI) continues to be an active space for research about what we are supposed to do with AI-based decisions, how those decisions were reached in the first place, and how we can achieve more transparent and trustworthy AI systems. AI may be used as a tool for forecasting or as a kind of partner for decision making. An AI system’s insights, however, must always be checked by humans who are able to assess the suitability of predictions and recommendations for their contexts of application. Information Production and Dissemination In addition to helping us with logistics, forecasting, and decision making, AI systems are also producing unique content that keeps us informed and entertained. AI journalists already abound; these are systems either capable of generating meaningful news articles themselves or used to support the work of human journalists through fact-finding, fact-checking, suggesting story ideas, or co-writing stories. Through automatically generated news articles, we are able to read up-to-the-minute news in language that is more understandable and compelling than, say, a table filled with raw numerical data. These AI journalists apply what is called natural language generation (NLG), which means that they are producing pieces in everyday human languages like English. For those who would rather listen to a story than read it, AI speech generation tools are also available. NLG tools, though, may facilitate plagiarism and production of ‘fake news’ (Fyfe, 2022). OpenAI’s GPT (generative pre-trained transformer) systems have proven especially good at generating texts including various markers of authenticity – including quotations (often from sources that the system has created itself), footnotes, and meticulous detail – but may nonetheless be largely, even entirely, fictional. The term ‘fake news’ is at best nebulous (Egelhofer et al., 2020), but these texts can get about as fake as the term might allow. Nevertheless, differentiating between human-written and computer-generated texts can be remarkably difficult. Doing so becomes even harder as content is produced and shared at increasingly rapid rates, with social media platforms making it ever easier to post and repost. Facebook, for example, reported anti-Ukraine ‘coordinated inauthentic behaviour’ stemming from a small network run by people in Russia and Ukraine. This network supposedly ran websites framed as independent news sources that included fake personae with humanlike profile pictures – ‘deepfakes’, fake likenesses produced through deep learning – probably created using GANs (Gleicher, 2022). Deepfakes

248  Handbook of digital politics

are not just limited to static imagery. They have also been used to create videos wherein a real person’s face or body is augmented so that it looks like they are saying or doing something that they actually didn’t. Actor Tom Cruise has been an especially well-known subject – or victim, depending on one’s perspective – of experimental deepfakes for entertainment. Yet deepfakes have also made their way into politics. For example, some have claimed that the deepfakes consciously and legally used in the latest South Korean Presidential election contributed to Yoon Suk Yeol’s win, helping him ‘win the hearts of the people’ by providing ‘wittier and more likeable’ answers to questions than the candidate himself (Fabiani, 2022). Unofficial deepfakes have also emerged. Ukraine President Volodymyr Zelenskyy has been the subject of a deepfake video in which he appears to instruct Ukrainian soldiers to lay down their weapons; another older deepfake video shows Russian President Vladimir Putin declaring Russia’s victory over Ukraine (Simonite, 2022a). Thankfully, these latter videos have been so poorly made that they are easily identifiable as inauthentic. However, some deepfake videos are extremely convincing, and it can be difficult to discern what is and isn’t doctored. Deepfakes contribute to the cyberwarfare we are seeing alongside the physical atrocities of the current Ukraine crisis. AI isn’t just enhancing military strategy through forecasting and decision making; it’s also being used to produce and share disinformation, as well as flood quantitative text trackers (e.g. Twitter’s ‘Trending’ function) with keywords to bring certain conversations and sentiments to the cultural fore. The latter practice is sometimes called ‘content polluting’, and it depends upon targeted efforts to exploit AI’s tendency towards pattern recognition. Russian Twitterbots (automated Twitter accounts) have gained particular notoriety, especially following their apparent prominence in Donald Trump’s 2016 Presidential campaign (Sanovich, 2018). Yet Ukraine has borne the brunt of much of Russia’s disinformation effort, with some scholars even claiming that ‘the conflict between Ukraine and Russia has been characterized by a fierce standoff in the realm of information’ (Zhdanova and Orlova, 2018, p. 45.) These scholars highlight examples of bot usage in Ukraine, which is notably anti-Ukrainian and pro-Russian despite Ukrainian information management and counterintelligence efforts. These bots use AI to sway public discourse by both generating content and flooding ‘trending’ sections with keywords. AI can help us stay informed of newsworthy events through its ability to automate information production and dissemination. However, this ability may also be applied to the creation of ‘news’ that seems legitimate but is actually misleading, libellous, or strategically persuasive. To safeguard the sanctity of democracy, we must ensure that citizens are able to critically reflect on the news they consume. It isn’t always possible to determine if news is ‘real’ or ‘fake’, but it is possible to inform people of this technological capability and encourage them to consider not just what the news is saying, but also where that news has come from and why it is being shared.

Artificial intelligence in politics  249

AI IN POLITICS RESEARCH Above, I’ve outlined some of the many ways AI is used in politics, highlighting its logistic, predictive, and informative potentials. We can also exploit these potentials in research practice. Using AI, we can look at our data in new ways by processing datasets far too large or multifaceted for humans to fully comprehend alone. Further, AI can help us share research findings in alternative ways, reaching audiences that have historically been underserved in dissemination efforts. For consistency, I use the same categories here as above: logistical assistance, forecasting and decision making, and information production and dissemination. Recall that these are loose categories. I must also stress that the applications of AI for research listed here usually depend on interdisciplinary collaborations between those with subject knowledge, data scientists, computer scientists, design professionals, and other relevant specialists, or some combination thereof. Most people cannot do AI-facilitated research on their own, and those who can probably shouldn’t. This work requires disciplinary understanding, technical know-how, and mathematical prowess; these requirements all entail different skillsets. To ensure appropriate system use and accurate data interpretation, collaborative work isn’t just desirable – it’s almost always necessary, even if only informally, to mitigate the risks associated with system misuse and output misinterpretation. Logistical Assistance AI is already making our lives easier by taking on those tasks we don’t want to do. As I write this, a robot vacuum cleaner whizzes by my feet; my word processor proposes alternative wording; this same word processor evaluates my writing for readability. While researching for this chapter, I learned more about subjects of interest by searching on Google. Google’s search function uses AI algorithms that don’t just identify relevant sources, but also rank those sources according to their anticipated ability to respond to queries. Google documents constant improvements to its search function on its company blog (‘Blog: The Keyword’, n.d.). But this AI isn’t without bias. Positively, Google has declared its efforts to ensure that certain results (for example, those touting extremism) are ranked lower in listings, if they are included at all. AI also helps rapidly remove problematic content, although such automatic analysis frequently flags permissible content. Less positively, search engine optimisation (SEO) marketing professionals specialize in exploiting search engine algorithms to ensure that their content is ranked higher in results lists than it might be otherwise. Google’s search results are also influenced by a user’s geographical location, search history, general activity, and other factors that may actually perpetuate inequalities, or even create new ones (Noble, 2018). We must therefore be critical not just of the information we find using AI, but also of the AI we use to get that information. AI may also benefit researchers through MT of sources in other languages. Although translations are not always ideal or even correct, access to alternative perspectives on global issues as they are expressed in different linguistic forms allows us

250  Handbook of digital politics

to consider these issues from fresh vantage points. After all, each language group has unique ways of structuring and communicating ideas. If one isn’t up to the challenge of achieving fluency in multiple other languages, MT is another option, although not one without issue (Bowker and Buitrago Ciro, 2019). Forecasting and Decision Making Sometimes, we aren’t sure what to make of our data. This is especially so when we’re faced with large and complex datasets. Through AI and ML, we may get some computational assistance with our work; ML systems can quickly signpost us to potentially meaningful patterns within a dataset using reinforced, supervised, or unsupervised approaches. In the UK, millions of pounds are spent annually on supporting the implementation of AI in the National Health Service (NHS), using the technology to – in ideal cases – aid diagnosis, support recovery and wellbeing, and improve productivity (Joshi and Morley, 2019). By identifying patterns across NHS data, these systems can contribute to a clearer picture of the current state of health and healthcare across the UK, helping us figure out what works and what doesn’t. Yet this data collection and usage hasn’t been without controversy; such sensitive data necessitates its own ethical care, and the expertise and intuition of medical professionals remains vital. AI systems can use the patterns it finds to then forecast future happenings. For example, numerous attempts to predict election results through analysis of social media data have been made, albeit with mixed results (dos Santos Brito et al., 2021). These attempts include the use of AI-driven volume, topic, and sentiment analysis tools that identify how frequently a topic appears in a dataset, what is said about that topic, and what feelings (sentiments) are reflected, respectively. While there is undoubtedly value to this kind of information, the claim that social media data can accurately predict election results assumes that there’s a representative sample of the voting population not just on the platform under investigation, but also actively engaging in political discourse on the platform. It also assumes that such engagement clearly and honestly reflects personal political beliefs. However, as anyone who has used any social media platform can confirm, posted content may be multimodal, satirical, partial, or altogether untrue. Not everyone is using social media platforms, and those who are using these platforms engage in different ways, for different purposes. Nevertheless, AI’s use in social media research shows that this technology isn’t just useful for making sense of quantitative data, but also qualitative data. AI has, for example, been used to detect disinformation through natural language processing (NLP) (Ozbay and Alatas, 2020). NLP applies computational techniques to textual datasets, analysing large amounts of human-language data that would be too much for any one researcher to read. NLP can also be used to detect linguistic trends in political speeches, potentially contributing to deeper understandings of the impacts of these speeches’ stylistic and semantic qualities (Savoy, 2020). Similar methods for analysing authorial style – a practice called stylometry – have been applied to

Artificial intelligence in politics  251

assessments of political leanings of media outlets (Hu and Rayz, 2021), analysis of trends in politicians’ tweets (García-Díaz et al., 2022), and identification of those oh-so-pesky Russian content polluters (Nasrin et al., 2019). These are just some of the ways NLP can help us better understand political climates locally, nationally, and globally. NLP still depends on quantifying the qualitative by giving words statistical ‘scores’ to conduct analysis, but results may alert us to trends and possibilities that we may never have otherwise considered. It is, after all, still up to us as researchers to decide if AI output is accurate and appropriate. Information Production and Dissemination Just as AI is being used to generate news, it may also be used to generate readable summaries of research findings in various languages, as well as in varying levels of detail for audiences with different literacy needs. Additionally, AI systems are serving as useful writing tools for those writing in non-native languages, looking to improve their authorial styles, and wishing to ensure that all of the necessary information is included (Razack et al., 2021). These tools can even help us rethink our own perspectives by summarizing and simplifying our long-form prose; I have recently benefited from ‘tldr papers’, which uses NLP and NLG to produce ‘science abstracts a second grader can understand’. AI systems can also produce informative and entertaining visualizations. For example, there are currently experimental efforts to generate images, both lifelike and more abstract, using AI. These images may be used in the future to illustrate news and research articles, supplementing explanatory text to enhance readers’ understanding of the material. However, the datasets used to train these image generators have been informed by human perspectives and expectations, often through manual tagging. This is why, as Kate Crawford and Trevor Paglen observe, a woman in a bikini is labelled a ‘slut’ and a man drinking beer is labelled an ‘alcoholic’. In their article about image-based ML training sets, Crawford and Paglen (2019) argue that training sets are unavoidably political: Datasets aren’t simply raw materials to feed algorithms, but are political interventions. As such, much of the discussion around ‘bias’ in AI systems misses the mark: there is no ‘neutral,’ ‘natural,’ or ‘apolitical’ vantage point that training data can be built upon. There is no easy technical ‘fix’ by shifting demographics, deleting offensive terms, or seeking equal representation by skin tone. The whole endeavor of collecting images, categorizing them, and labeling them is itself a form of politics, filled with questions about who gets to decide what images mean and what kinds of social and political work those representations perform.

Research using AI for information production and dissemination still requires substantial human oversight. Nonetheless, AI systems offer novel and exciting opportunities for research dissemination, cultural impact, and audience engagement. AI can transform storytelling associated with research, although this transformation has yet to gain widespread traction in academia.

252  Handbook of digital politics

DISCUSSION Throughout this chapter, I’ve highlighted both positive and negative applications of AI. AI can support fresh ways of thinking about stale problems, contributing to timely and novel solutions. However, it can also support the creation of new problems. Such problems may stem from cultural hype surrounding AI: hype that sets unrealistic expectations of system functionality and autonomy. As is evident in all of the above examples, AI depends upon human involvement not just for development, but also for system maintenance and output curation. AI systems are only as useful or harmful as humans allow them to be. Computer-generated output is never ‘objective fact’, nor is it ever an evidential end. Just as we must train individuals to develop and manage computational systems, we must too train them to critically assess computational output in its many forms. In its National AI Strategy, the UK Government presents a decade-long plan ‘to make Britain a global AI superpower’. This plan includes new visa routes for international researchers, targeted training opportunities for students and those changing careers, and funding opportunities for private commercial organizations. However, the strategy makes little reference to the societal and ethical implications of AI, with its emphasis being on quantitative measures of evaluation (HM Government, 2021). Similarly, UK Research and Innovation – the government body that directs research funding – has focused primarily on quantitative measures of success (UKRI, 2021). The implied assumption that only quantitative measures are suitable for assessing AI impact is, as was made clear above, unacceptable. How are we to apply only quantitative evaluation to an investigation into the efficacy and readability of computer-generated texts? Can calculation alone capture the human experiences of watching deepfake videos? What numbers need we collect to represent the ethical dilemmas of computer vision? We need humanities and social sciences researchers to make sense of what drives and emerges from AI. In the UK, the political choice to prioritize quantitative over qualitative data is indicative of current policymakers’ values and priorities. By reducing people and institutions to machine-readable categories and scales of measurement, a government may garner ‘objective’ evidence of success that ultimately obscures individual circumstance. Put another way, by depending on AI to analyse citizens without that AI being held to account by those trained to embrace grey areas, governments dehumanize their own constituencies and further marginalize social outliers. We silence those voices that don’t fit into our categories, whether those categories are identified by humans or machines. On the flip side, AI has potential to aid democracy. With AI systems taking on some of the logistical work associated with finding and making sense of vast amounts of information, policymakers and laypeople alike may be more informed about issues of interest, as well as better equipped to respond to those issues. Moving away from those tasks often considered tedious – data collection, translation, and so forth – and towards deeper consideration of the patterns identified by AI systems may support consideration of current affairs from a more thorough range of perspectives. Used

Artificial intelligence in politics  253

in this way, AI systems may serve as critical friends, contributing to interpersonal understanding by enhancing exposure to different points of view. But note how frequently I have used ‘may’ and ‘might’ throughout this chapter. These systems can help or they can harm. Often, the actual uses and implications of systems aren’t anticipated by those systems’ developers. I would like to believe that few developers wake up and think let’s see how I can ruin someone’s day through automation. Yet use of these systems may have unintended consequences, and those consequences that are intended are so very often political: users have their own perspectives, motivations, and desired outcomes. Whether AI is used in politics or in politics research, periodic evaluation of system functionality and suitability for purpose – ideally by those representing a diverse demographic pool – is imperative. We must also allow our minds to wander into worst-case scenarios. Who might be paying the price for the ‘advancement’ of AI?

DIRECTIONS FOR FUTURE RESEARCH Where do we go from here, then? There are gaps in our understanding of every one of the AI applications that I have listed above, and these gaps serve as opportunities for future research. For AI in digital politics, the issues I believe to be most pressing include amplifying a more diverse set of voices in AI and AI-adjacent conversations, managing expectations of system capacity, and – above all – never forgetting that AI is inextricably bound to its human creators and users. It is significant that the participants of the 1956 Dartmouth event from which ‘AI’ as a term emerged were all men, all of whom were embedded in Western, English-speaking contexts. The field’s pale-skinned phallicism continues, despite concentrated efforts to diversify AI development, training sets, and output (Cave and Dihal, 2021; Stathoulopoulos and Mateos-Garcia, 2019). There is even a term that coalesces this bias with a call for countering exaggeration of data-driven projects: Big Dick Data. ‘Big Dick Data is a formal, academic term […] to denote big data projects that are characterized by masculinist totalizing fantasies of world domination as enacted through data capture and analysis’, Catherine D’Ignazio and Lauren Klein (2020, p. 151) explain in their book Data Feminism. ‘Big Dick Data projects ignore context, fetishize size, and inflate their technical and scientific capabilities.’ This is the unfortunate reality of many current AI-driven projects. Projects proposals titillate funders with promises to make sense of big data and provide solutions to complex problems that humans have not been able to solve themselves. But such ‘powered by AI’ claims do not adequately encapsulate the limitations or risks associated with AI. Rather, they contribute to hype surrounding AI by overstating computational capacity. Moreover, while AI could be regarded as only in its teenage years, somewhat established but still in the uncomfortable throes of puberty, there is already an urgent need to decolonize the field. Eurocentric approaches to problem-solving are not the only ones worth coding, and are not suitable in every circumstance (Schwartz Reisman Institute, 2022). Independent watchdogs (Algorithmic Justice League;

254  Handbook of digital politics

Black in AI; Center for Humane Technology) play a key role in holding AI developers and companies to account, advocating for more representative approaches to system creation and use. These watchdogs’ advocacy for diversity and regulation is complemented by similar efforts by public (Daly et al., 2019; ‘European AI Alliance’; National Artificial Intelligence Initiative; OECD AI Policy Observatory), commercial (‘AI Policy Principles’, 2017; ‘Artificial Intelligence at Google’; ‘Responsible AI’), and religious (RenAIssance Foundation) organizations, as well as collaborations between these organizations (Partnership on AI). Many of these efforts, though, are directed by vague language about ambiguous concepts like ‘transparency’ and ‘fairness’. Such language can accommodate rapid technological change, but also lends itself to loopholes exploited by those with unsavoury intentions. Further, critics continue to identify unsuitable functionality of AI that disadvantages those with marginalized identities (Raji and Buolamwini, 2019), and incorrectly profiles individuals through automated classification (O’Neil, 2016). The Dutch government, for instance, is currently facing the aftermath of the ‘toeslagenaffaire’ (loosely translating to ‘benefits scandal’). The toeslagenaffaire saw incorrect ‘risk’ profiling drive some citizen victims to poverty, homelessness, and suicide, and led to children being unnecessarily removed from their homes (Amnesty International, 2021). In addition to social issues, we must also consider the environmental and financial costs of training ML systems on large datasets. These costs are prohibitive, and may contribute to the establishment of technocratic social infrastructures governed by a privileged few (Strubell et al., 2019). Historically, these few have been men, although efforts to highlight the past and present contributions of women to AI and computing more generally are underway (Evans, 2018; Hicks, 2017). Big data and the AI algorithms that help us make sense of those data can – and do – have life or death consequences. Future research must prioritize the improvement of everyone’s quality of life, rather than tend towards more exclusive techno-chauvinism: the belief that technological solutions are always best (Broussard, 2018). AI is just statistics, but people are not wholly quantifiable.

FURTHER READING Broussard, M. (2018). Artificial Unintelligence: How Computers Misunderstand the World. Cambridge, MA: MIT Press. [A critical consideration of hype surrounding AI, and an argument that technological solutions are not always the most appropriate.] Cave, S. Dihal, K., and Dillon, S. (2020). AI Narratives: A History of Imaginative Thinking about Intelligent Machines. Oxford: Oxford University Press. [An edited collection exploring representations of AI throughout history.] Chubb, J., Cowling, P., and Reed, D. (2021). ‘Speeding up to keep up: Exploring the use of AI in the research process’, AI & SOCIETY. https://​doi​.org/​10​.1007/​s00146​-021​-01259​-0. [A survey of attitudes towards AI in research.] Coeckelbergh, M. (2022). The Political Philosophy of AI: An Introduction. Cambridge: Polity Press. [An analysis of AI applications from a political philosophy angle in particular.]

Artificial intelligence in politics  255

OECD AI Policy Observatory (n.d.). https://​oecd​.ai. [A hub of news, data, and analyses related to AI, highlighting national and global perspectives.] Winterson, J. (2021). 12 Bytes: How We Got Here. Where We Might Go Next. London: Jonathan Cape. [A broad overview of some of the many social and ethical issues posed by AI.] Woolley, S. C. and Howard, P. N. (eds.) (2016). ‘Automation, algorithms, and politics’, Special Section, International Journal of Communication 10, 4882–5055. [A collection of articles exploring computational elements of modern politics (e.g. big data, personalization systems, etc.).]

REFERENCES Agudo, U. and Matute, H. (2021). ‘The influence of algorithms on political and dating decisions’, PLOS ONE. https://​doi​.org/​10​.1371/​journal​.pone​.0249454. ‘AI Policy Principles’, Information Technology Industry Council. 2017. https://​www​.itic​.org/​ resources/​AI​-Policy​-Principles​-FullReport2​.pdf. Algorithmic Justice League. https://​www​.ajl​.org. Amnesty International (2021). Xenophobic Machines: Discrimination Through Unregulated Use of Algorithms in the Dutch Childcare Benefits Scandal. London: Amnesty International. Araya, D. and King, M. (2022). The Impact of Artificial Intelligence on Military Defence and Security. Waterloo, ON: Centre for International Governance Innovation. ‘Artificial Intelligence at Google: Our Principles’, Google AI. https://​ai​.google/​principles. BBC (2016). ‘Google translated Russia to “Mordor” in “automated” error’, BBC News. https://​ www​.bbc​.co​.uk/​news/​technology​-35251478. Black in AI. https://​blackinai​.github​.io. ‘Blog: The Keyword’, Google. https://​blog​.google​.com. Bostrom, N. (2014). Superintelligence: Paths, Dangers, Strategies. Oxford: Oxford University Press. Bowker, L. and Buitrago Ciro, J. (2019). Machine Translation and Global Research: Towards Improved Machine Translation Literacy in the Scholarly Community. Bingley: Emerald. Broussard, M. (2018). Artificial Unintelligence: How Computers Misunderstand the World. Cambridge, MA: MIT Press. Bryson, J. J., Diamantis, M. E., and Grant, T. D. (2017). ‘Of, for, and by the people: The legal lacuna of synthetic persons’, Artificial Intelligence and Law 25, 273–291. Buolamwini, J. and Gebru, T. (2018). ‘Gender shades: Intersectional accuracy disparities in commercial gender classification’, Proceedings of the 1st Conference on Fairness, Accountability and Transparency 81 (pp. 77–91). New York: PMLR. Cave, S. and Dihal, K. (2020). ‘The whiteness of AI’, Philosophy & Technology 33, 685–703. Cave, S. and Dihal, K. (2021). ‘Race and AI: The diversity dilemma’, Philosophy & Technology 34, 1775–1779. Cave, S., Dihal, K., and Dillon, S. (2020). AI Narratives: A History of Imaginative Thinking About Intelligent Machines. Oxford: Oxford University Press. Center for Humane Technology. https://​www​.humanetech​.com. Coeckelbergh, M. (2022). The Political Philosophy of AI: An Introduction. Cambridge: Polity Press. Crawford, K. (2021). Atlas of AI: Power, Politics, and the Planetary Costs of Artificial Intelligence. New Haven: Yale University Press. Crawford, K. and Paglen, T. (2019). Excavating AI: The Politics of Training Sets for Machine Learning. https://​excavating​.ai. Dabre, R., Chu, C., and Kunchukuttan, A. (2021). ‘A survey of multilingual neural machine translation’, ACM Computing Surveys 53(5). https://​doi​.org/​10​.1145/​3406095.

256  Handbook of digital politics

Daly, A., Hagendorff, T., Li, H., Mann, M., Marda, V., Wagner, B., Wang, W. W., and Witteborn, S. (2019). Artificial Intelligence Governance and Ethics: Global Perspectives [Research Paper No. 2019-15]. Hong Kong: Chinese University of Hong Kong. Dave, P. and Dastin, J. (2022). ‘Ukraine starts using AI to identify the dead, uncover assailants amid Russian invasion’, Global News. https://​globalnews​.ca/​news/​8680053/​ukraine​-ai​ -technology​-russia. Dervanovi, D. (2018). ‘I, inhuman lawyer: Developing artificial intelligence in the legal profession’, in M. Corrales, M. Fenwick, and N. Forgó (eds.), Robotics, AI and the Future of Law (pp. 209–234). Singapore: Springer. D’Ignazio C. and Klein, L. F. (2020). Data Feminism. Cambridge, MA: MIT Press. dos Santos Brito, L., Silva Filho, R. L. C., and Leitão Adeodato, P. J. (2021). ‘A systematic review of predicting elections based on social media data: Research challenges and future directions’, IEEE Transactions on Computational Social Systems 8(4), 819–843. Egelhofer, J. L., Aaldering, L., Eber, J.-M., Galyga, S., and Lecheler, S. (2020). ‘From novelty to normalization? How journalists use the term “fake news” in their reporting’, Journalism Studies 21(10), 1323–1343. ‘European AI Alliance’, European Commission. https://​futurium​.ec​.europa​.eu/​en/​european​-ai​ -alliance. Evans, C. L. (2018). Broad Band: The Untold Story of the Women Who Made the Internet. New York: Portfolio. Fabiani, A. (2022). ‘In South Korea, AI deepfakes of politicians are being used to win the hearts of the people’, Screenshot. https://​screenshot​-media​.com/​technology/​ai/​south​-korea​ -deepfakes​-politicians. Fyfe, P. (2022). ‘How to cheat on your final paper: Assigning AI for student writing’, AI & Society. https://​doi​.org/​10​.1007/​s00146​-022​-01397​-z. García-Díaz, J. A., Colomo-Palacios, R., and Valencia-García, R. (2022). ‘Psychographic traits identification based on political ideology: An author analysis study on Spanish politicians’ tweets posted in 2020’, Future Generation Computer Systems 130, 59–74. Gleicher, N. (2022). ‘Updates on our security work in Ukraine’, Facebook. https://​about​.fb​ .com/​news/​2022/​02/​security​-updates​-ukraine. Gupta, R. and Pal, S. K. (2021). Introduction to Algorithmic Government. Cham: Palgrave Macmillan. Hicks, M. (2017). Programmed Inequality: How Britain Discarded Women Technologists and Lost Its Edge in Computing. Cambridge, MA: MIT Press. HM Government (2021). National AI Strategy. London: Office for Artificial Intelligence. Hu, Y. and Rayz, J. (2021). ‘A study of media polarization with stylometry methods’, The International FLAIRS Conference Proceedings, 34. https://​doi​.org/​10​.32473/​flairs​.v34i1​ .128477. Joshi, I. and Morley, J. (eds.) (2019). Artificial Intelligence: How to Get It Right. Putting Policy into Practice for Safe Data-Driven Innovation in Health and Care. London: NHSX. Kolomiyets, L. (2020). ‘Manipulative mistranslations in official documents and media discourses on contemporary Ukraine’, Harvard Ukrainian Studies 37(3–4), 367–405. Laterza, V. (2021). ‘Could Cambridge Analytica have delivered Donald Trump’s 2016 presidential victory? An anthropologist’s look at big data and political campaigning’, Public Anthropologist 3, 119–147. McCarthy, J., Minsky, M. L., Rochester, N., and Shannon, C. E. (1955). A Proposal for the Dartmouth Summer Research Project on Artificial Intelligence. http://​jmc​.stanford​.edu/​ articles/​dartmouth/​dartmouth​.pdf. Mehr, H. (2017). Artificial Intelligence for Citizen Services and Government. Cambridge, MA: Harvard Kennedy School. Mitchell, C. (2022). ‘How Canada accidentally helped crack computer translation’, The Walrus. https://​thewalrus​.ca/​how​-canada​-accidentally​-helped​-crack​-computer​-translation.

Artificial intelligence in politics  257

Nasrin, N., Choo, K.-K. R., Ko, M., and Rios, A. (2019). ‘How many users are enough? Exploring semi-supervision and stylometric features to uncover a Russian troll farm’, Proceedings of the 2nd Workshop on NLP for Internet Freedom: Censorship, Disinformation, and Propaganda (pp. 20–30). Hong Kong: ACL. National Artificial Intelligence Initiative. https://​www​.ai​.gov. Nickerson, D. W. and Rogers, T. (2014). ‘Political campaigns and big data’, Journal of Economic Perspectives 28(2), 51–74. Noble, S. U. (2018). Algorithms of Oppression: How Search Engines Reinforce Racism. New York: New York University Press. O’Neil, C. (2016). Weapons of Math Destruction. New York: Crown. OECD AI Policy Observatory. https://​oecd​.ai. Ozbay, F.A. and Alatas, B. (2020). ‘Fake news detection within online social media using supervised artificial intelligence algorithms’, Physica A 540. https://​doi​.org/​10​.1016/​j​ .physa​.2019​.123174. Partnership on AI. https://​www​.partnershiponai​.org. Raji, I. D. and Buolamwini, J. (2019). ‘Actional auditing: Investigating the impact of publicly naming biased performance results of commercial AI products’, in V. Conitzer, G. Hadfield, and S. Vallor (eds.), Proceedings of the 2019 AAAI/ACM Conference on AI, Ethics, and Society (pp. 429–435). New York: ACM. Razack, H. I. A., Mathew, S. T., Saad, F. F. A., and Alqahtani, S. A. (2021). ‘Artificial intelligence-assisted tools for redefining the communication landscape of the scholarly world’, Science Editing 8(2), 134–144. RenAIssance Foundation. https://​www​.romecall​.org. ‘Responsible AI’, Microsoft. https://​www​.microsoft​.com/​en​-us/​ai/​responsible​-ai. Rusch, T., Lee, I., Hornik, K., Jank, W., and Zeileis, A. (2013). ‘Influencing elections with statistics: Targeting voters with logistic regression trees’, The Annals of Applied Statistics 7(3), 1612–1639. Sam – The virtual politician of the future. http://​www​.politiciansam​.nz. Sanovich, S. (2018). ‘Russia: The origins of digital misinformation’, in S. C. Woolley and P. N. Howard (eds.), Computational Propaganda: Political Parties, Politicians, and Political Manipulation on Social Media. Oxford: Oxford University Press Scholarship Online. https://​doi​.org/​10​.1093/​oso/​9780190931407​.001​.0001. Savoy, J. (2020). ‘Applications to political speeches’, in J. Savoy, Machine Learning Methods for Stylometry: Authorship Attribution and Author Profiling (pp. 229–249). Cham: Springer. Schwartz Reisman Institute (2022). ‘Syed Ishtiaque Ahmed | Whose intelligence? Whose ethics? Ethical pluralism and decolonizing AI’, YouTube. https://​www​.youtube​.com/​watch​ ?v​=​ReSbgRSJ4WY. Simonite, T. (2022a). ‘A Zelensky deepfake was quickly defeated: The next one might not be’, WIRED. https://​www​.wired​.com/​story/​zelensky​-deepfake​-facebook​-twitter​-playbook. Simonite, T. (2022b). ‘Online sleuths are using face recognition to ID Russian soldiers’, WIRED. https://​www​.wired​.com/​story/​facial​-recognition​-identify​-russian​-soldiers. Slocum, J. (1985). ‘A survey of machine translation: Its history, current status, and future prospects’, Computational Linguistics 11(1), 1–17. Stathoulopoulos, K. and Mateos-Garcia, J. (2019). ‘Gender diversity in AI research’, SSRN. http://​dx​.doi​.org/​10​.2139/​ssrn​.3428240. Strubell, E., Ganesh, A., and McCallum, A. (2019). ‘Energy and policy considerations for deep learning in NLP’, Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics. Florence: ACL. Susskind, J. (2018). ‘Chatbots are a danger to democracy’, The New York Times. https://​www​ .nytimes​.com/​2018/​12/​04/​opinion/​chatbots​-ai​-democracy​-free​-speech​.html. Tariq, M. U. (2021). ‘Achieving operational excellence through artificial intelligence: Driving forces and barriers’, Frontiers in Psychology. https://​doi​.org/​10​.3389/​fpsyg​.2021​.686624.

258  Handbook of digital politics

Taylor, I. (2021). ‘Who is responsible for killer robots? Autonomous weapons, group agency, and the military-industrial complex’, Journal of Applied Philosophy 38(2), 320–334. UKRI (2021). Transforming Our World with AI: UKRI’s Role in Embracing the Opportunity. Swindon: UK Research and Innovation. Urbina, F., Lentzos, F., Invernizzi, C., and Ekins, S. (2022). ‘Dual use of artificial-intelligence-powered drug discovery’, Nature Machine Intelligence. https://​doi​ .org/​10​.1038/​s42256​-022​-00465​-9. Waelen, R. A. (2022). ‘The struggle for recognition in the age of facial recognition technology’, AI and Ethics. https://​doi​.org/​10​.1007/​s43681​-022​-00146​-8. Zhdanova, M. and Orlova, D. (2018). ‘Ukraine: External threats and internal challenges’, in S. C. Woolley and P. N. Howard (eds.), Computational Propaganda: Political Parties, Politicians, and Political Manipulation on Social Media. Oxford: Oxford University Press Scholarship Online. https://​doi​.org/​10​.1093/​oso/​9780190931407​.001​.0001. Zuiderwijk, A., Chen Y.-C., and Salem, F. (2021). ‘Implications of the use of artificial intelligence in public governance: A systematic literature review and a research agenda’, Government Information Quarterly 38(3). https://​doi​.org/​10​.1016/​j​.giq​.2021​.101577.

17. Online content moderation during conflict Giovanni De Gregorio and Nicole Stremlau

INTRODUCTION War increasingly has a digital dimension. Recent conflicts around the world have underlined the role of social media in providing channels to fight propaganda and denounce atrocities but also facilitating the sharing of disinformation or hate speech that contribute to inflaming conflicts. The conflict in Ukraine is another example of how social media goes to war (Scott and Kern, 2022). Meta’s decision to change hate speech policies to enlarge the dissemination of content in times of war is one example that underlines the central role of social media in mediating conflicts in the digital age (Vengattil and Culliford, 2022). This is far from the first time such problems have arisen. In 2018, a UN fact-finding mission found social media – Facebook in particular – to have had a ‘determining role’ in suspected genocide in Myanmar (Miles, 2018). In 2021, whistleblower Frances Haugen brought renewed attention to concerns about online speech and offline violence when she argued that Facebook was ‘literally fanning ethnic violence’ in Ethiopia and continued to do so in Myanmar (Akinwoto, 2021). This situation follows a longer legacy of the role of mass media in violent conflict, from the use of newspapers in Nazi propaganda campaigns (Herzstein, 1978) to the more recent conflicts involving radio in Rwanda and satellite television in Somalia (Allen and Stremlau, 2005; Stremlau, 2018). Radio and television have long demonstrated an ability to weaponize hate against minorities or targeted groups (Larson and Whitton, 1963). Media outlets have contributed to guiding and organizing entire military forces and promoting propaganda with a view to attracting new proselytists. While traditional media outlets still have a role in times of war, social media have amplified the reach of content and engagement on a global scale. As also underlined by Gagliardone (2019), social media can add complexity to the spread of hate speech, or what may be better termed extreme speech, during conflict. The automated content moderation, user activism and social media incentives in governing content constitute only some of the forces shaping the dynamics of conflicts in the digital age. In the Central African Republic, online hate speech has been attributed to provoking mass atrocities between Christians and Muslims (Schlein, 2018). In Sri Lanka, rumours on social media are widely regarded as provoking a number of religious attacks, including the 2019 Easter Sunday church and hotel bombings (Fisher, 2019). The growing prominence of social media in times of war prompts urgent questions about how to mitigate this situation. This does not only concern questions around how to address the spread of online hate and disinformation but also how to increase transparency and set procedural safeguards and oversight in the process of content 259

260  Handbook of digital politics

moderation. The cooperation with social media companies is then key. The conflict in Ukraine has underlined how social media representatives have been involved in long discussions with governments and international organizations to define a strategy to tackle Russian propaganda. Without collaborating with social media, the only way is to restrict access to sites or to shut down the Internet as has already occurred in many countries around the world (De Gregorio and Stremlau, 2020). Within this framework, this chapter explores the role of social media in times of conflicts. Issues of censorship, online hate speech, and disinformation during conflict are a huge concern given their potential to exacerbate violence and erode election integrity. Yet how and when social media companies respond has long been an overlooked and under-addressed problem, particularly in countries beyond their priority as profitable markets. This chapter underlines the role of social media in moderating conflicts through artificial intelligence systems and the inequalities in this process. In the first part, we focus on the role of social media in influencing conflicts, focusing particularly on online hate and disinformation. In the second part, we explore the inequalities in the field of social media and conflicts. In the third part, we underline the trends to address the governance of content moderation in times of conflicts. Particularly, we look at systems of oversight, information intervention and crisis protocols.

MODERATING CONFLICTS In recent years, the spread of harmful content such as online hate and disinformation has increasingly shaped conflicts, thus underlining how online speech produces effects beyond digital boundaries. This situation does not only concern Western democracies as the events on Capitol Hill illustrated. Looking at Africa and Asia, in the Central African Republic, the spread of online hate speech and disinformation on social media has contributed to mass atrocities between Christians and Muslims (OHCHR, 2019). Many other African countries have shut down the Internet to deal with the protests and violence resulting from the spread of online hate and disinformation on social media. In Sri Lanka, rumours on social media led to several religious attacks, including the 2019 Easter Sunday church and hotel bombings (Purnell, 2019), while the use of Facebook in inciting violence against Myanmar’s minority Muslim population has elevated concerns about the role of social media in perpetrating genocides (Potkin and McPherson, 2020). Even in situations of conflict, social media discretionarily determine how to moderate content, thus becoming the governors of online speech (Klonick, 2017). Although social media actors are rarely based in the country where conflicts and war escalate, they exercise broad discretion in determining the rules according to which information circulates online and, therefore, how content is shared between communities (Dimitrakopoulou et al., 2014). The dissemination of content in social media spaces is characterized by multiple layers of governance based on private norms and technological standards. Images of battlefields and propaganda do not

Online content moderation during conflict  261

just neutrally flow in social media spaces but are organized according to the logic of content moderation. Online platforms determine how to show and organize online content according to predictive analysis based on the processing of users’ data (Bloch-Wehba, 2019). The information uploaded by users is processed by automated systems that define (or at least suggest to human moderators) not only which content to retain or remove but also its organization. Online platforms that process content rely on a mix of human moderators and artificial intelligence systems that define which content must be removed according to non-transparent standards and without explanation, providing very few opportunities for remedies (Roberts, 2019). Considering the limit of artificial intelligence to understand contexts and the distance of human moderators from where the conflicts take place, the question is about whether the process of content moderation can effectively mitigate the spread of content inflaming conflicts. Besides, the massive implementation of artificial intelligence technologies in content moderation leads to privately determined computing standards for protection that contribute to determine rights and freedoms in the digital age. Legal norms and social media policies are interpreted by a mere algorithmic calculation that is opaque and not entirely accountable even for social media which are primarily driven by profit maximization. The global pandemic has amplified this problem while underlining how problematic artificial intelligence technologies can be when it comes to effectiveness; during the pandemic there have been significant cases when the involvement of human moderators was restricted and an over-reliance on the automated system led to the spread of disinformation and blocking of accounts that were actually countering disinformation (Statt, 2020). This case underlines the role social media plays not only in organizing content through processes of moderation, but also in creating digital spaces where the spread of hate speech and disinformation is not neutral. This opaque framework of governance defines the rules based on which information is disseminated. And it is not surprising that some content flows more widely and with greater speed, especially in cases involving strong messages of hate or dissent. Algorithmic content moderation contributes to driving people to online hate and disinformation (Nicas, 2018), which can also lead to discrimination (Noble, 2018). In areas characterized by tensions and conflicts, this can inflame and escalate violence and conflicts. This ‘content moderation paradox’ explains why, on the one hand, social media commit to protecting free speech, while, on the other hand, they moderate content regulating their communities for business purposes. This risk is also connected to the role of social media in providing spaces that transform users into active sources of communication. This can in turn shape conflicts (Dahlgren, 2005). As observed by Zeitoff (2017), communications in conflicts have typically been defined in two ways: ‘elite-level communication’ focused on tactical and logistical aims; and ‘mass-based appeals’ aimed at coordinating or inhibiting public behaviour through control of the narrative and manipulating mass channels of communication (Howard, 2010). Social media provide users opportunities to challenge elite-dominated discourse, especially in authoritarian regimes, which tend

262  Handbook of digital politics

to exercise public control of traditional media outlets. Information spread on social media can be immediately shared with other communities of users, potentially going viral. The possibility to use these channels to contest central authorities and spread disinformation has encouraged governments to censor online speech or even use social media as an instrument of surveillance.

SOCIAL MEDIA, INEQUALITIES AND CONFLICTS The process of content moderation is deeply influenced by the business nature of social media. Notwithstanding company statements that attempt to emphasize the potential of social media to enhance rights and freedoms transnationally, companies need to moderate content to protect their business interests by avoiding users’ rejection because of the dissemination of content like terrorism and hate (Gillespie, 2018). At the same time, the business model of large social media is based on advertising revenues coming from the organization of content. These actors do moderate content for profit. It should not surprise then if some areas of the world have not always been considered ‘priorities’ (Wong, 2021). Online hate speech and disinformation have long incited violence, and sometimes mass atrocities. However, when this has happened in the Global South, from Ethiopia to Myanmar, much of the world has looked away. The imbalance of economic power between the Global South and social media companies – their valuations can be many times the GDP of poorer countries – means the tech titans pay little heed to the concerns of hate speech or disinformation campaigns in such nations, not least because they are marginal markets. This is part of a broader picture of inequality characterizing the process of content moderation on a global scale. Since social media relies on automated decision-making to scale due to the vast amount of content (Gillespie, 2020), they have fewer incentives to invest resources to address the challenges of language diversity in areas which cannot ensure a certain economic return in terms of advertising revenues. A video showing the torture of Samuel Doe, former Liberian President, has been available on YouTube for years. Although this content was expressly contrary to the social media’s community guidelines, content moderation has not worked in this case, even though the content was published in English. If we compare this case to the fast blocking of content concerning the New Zealand shooting at Christchurch, the Doe video would suggest that content moderation is not geographically unique. This is also a matter of scale. The number of dialects and vernacular languages in some regions in Africa or Asia does not encourage the development of a unique system for detecting hate speech or disinformation thus making the process of content moderation less effective. While language processing tools are more developed to detect hate in certain languages, they still fail to detect content in the vast majority of idioms. Content moderation can indeed perform tasks of language detection with more accuracy just for online content published in some languages or areas, thus, leading to a technical inequality. For instance, in Africa, there are not many

Online content moderation during conflict  263

language processing tools and resources to detect hate speech. Although, in Kenya, the detection of hate speech is easier due to the use of English (Ombui et al., 2019), the language diversity in Africa is one of the primary challenges to promote common standards for moderating content and detecting online hate speech. The Western focal point is also the reason why social media tend to neglect the risk of injecting biases when programming content moderation technologies. For instance, search engines have already shown that their results can be discriminatory for marginalized categories (Noble, 2018). This form of speech colonization is also the result of lack of incentives in moderating content in areas which are far from technical, such as social and political influence on social media. The Myanmar genocide has underlined the inability of Facebook to detect and limit the spread of hate speech (Stecklow, 2018). The spread of hate speech on Facebook supported ethnic cleansing in Myanmar, but this went mostly unchecked due to the lack of moderation tools and human moderators fluent in Burmese. While Facebook significantly expanded its team of Burmese speakers to create a dataset of hate and violent expressions, the international pressure to act also led to overreactions including the banning of some armed groups (Sablosky, 2021). Online community standards do not exist in many non-European languages, and social media companies have been slow to adapt their platforms accordingly. For example, Facebook’s own research indicates that its algorithms incorrectly delete Arabic content more than three out of four times (Scott, 2021). And given that Amharic, Swahili, and Somali are much less prevalent and low resource (less text and material for AI to be trained on), accurate automated content removal for these languages happens even less often. Many governments have neither the capacity nor the tools to effectively address these kinds of issues. Given these challenges, the first reaction by many governments has been to criminalize the spread of online hate and disinformation by users and social media. The spread of online hate and disinformation on social media is increasingly considered by some governments to be a justification (or legitimate aim) to censor speech and shut down the Internet. This is often perceived as the only immediately effective remedy to deal with the escalation of violence in the context of company-led discretion in responding to and moderating content. Even though there is very limited evidence about the effects of these practices in tackling the misinformation and hate they purport to address, shutdowns have been implemented to curtail online speech, and particularly content that is seen to be provoking violence or promoting dissent. Nonetheless, the indirect role of social media in Internet shutdowns is often glazed over. States do not always have the power to regulate social media such as in the case of the European Union. The spread of Internet shutdowns is also connected to the lack of control and skills on how to govern the circulation of online content that is primarily driven by social media. Therefore, in some cases, the only way states can intervene to deal with protests or the spread of hate and violence online in the absence of concerted cooperation from social media companies is by shutting down the entire network or specific websites.

264  Handbook of digital politics

Besides, some critics correctly observe that social media can and has been utilized by authoritarian regimes to monitor individuals (Morozov, 2012). Thus, while social media can indeed be used to promote democratic values, it can equally be used by hate groups seeking to recruit or organize members, or by governments for surveillance or propaganda purposes. The war in Ukraine makes it difficult to ignore how social media is being weaponized in conflict. And this conflict has been showing the inequalities of the efforts of social media companies to address the spread of online hate and disinformation. Twitter responded to Ukraine by taking those seen to be inflaming the conflict offline – though its efforts were deemed insufficient by the Ukrainian government and its allies. However, Meta, Facebook’s parent company, then made a controversial change to its hate speech policy, allowing posts that would normally violate its rules, including calls for violence, such as ‘death to Russian invaders’ (Vengattil and Culliford, 2022). Poland, Lithuania, Estonia, and Latvia, meanwhile, sent an impassioned plea to executives at Google, Twitter, and Facebook demanding they censor and take down accounts justifying the war or praising war crimes, many of which came from Russia. They called for government accounts to be suspended, including state-controlled media in Russia and Belarus. The reactions to the Ukrainian conflict underline the extent to which the governance of online content is driven by multiple influences. This situation raises questions on how it is possible to increase transparency and oversight in content moderation and ensure collaboration between government and social media to minimize the reliance of general censoring measures.

CONTROL, OVERSIGHT AND INTERVENTION The challenges raised by social media have triggered government reactions, particularly in poorer and less geopolitically influential countries to address the spread of online hate and disinformation. They have accused online platforms of disseminating hate and disinformation online, criminalized the spread of hate and disinformation (Olewe, 2018), have used platforms for surveillance, worked to push alternative narratives (sometimes flooding platforms with disinformation), and have attempted to censor content (Shahbaz, 2018). The spread of hate on social media has also been one of the primary reasons why governments have increasingly justified the use of Internet shutdowns. Whereas only a few years ago such forms of censorship would have been seen as a grave violation of freedom of expression, increasingly they are understood as one of the few mechanisms available for addressing online speech and offline harms in a moment of crisis such as conflicts and war. Addressing the challenges of social media in times of war has also triggered discussion on how to fix algorithmic biases and lack of transparency. There are also different proposals to deal with content moderation. It would be possible to focus on algorithmic diversity (Helberger et al., 2018) or rely on human moderators to check content to reduce the biases and fallacies of algorithms. However, human moderation raises several challenges (Roberts, 2019) while also not a viable solution for global

Online content moderation during conflict  265

online platforms such as Facebook due to the high amount of content to moderate (Lessin, 2016). Within this framework, the European Union is proposing a new model for mitigating social media discretion in content moderation. The proposal for the Digital Services Act is a clear example of a legal instrument which aims to regulate the process of content moderation without focusing on content. This approach increasingly tends to promote a governance approach where social media are considered as regulated centres of collaboration, social infrastructures or digital utilities (Guggenberger, 2021; Rahmnan, 2018). Particularly, among the procedural safeguards, the Digital Services Act requires social media to enforce their Terms of Services consistently and protect the right to freedom of expression. Likewise, very large online platforms are called to identify, analyse and assess, once a year and before launching new services, the probability and severity of any significant systemic risks stemming from the design, algorithmic systems, intrinsic characteristics, functioning and use made of their services in the Union. In the case of extraordinary circumstances, such as public security, the Digital Services Act recognizes the power of the Commission to initiate the drawing up of crisis protocols to coordinate a rapid, collective and cross-border response in the online environment. Extraordinary circumstances will include serious threats to public interests such as war. In this case, social media would be used as instruments to limit the spread of harmful content such as disinformation and to spread reliable content. In any case, the Digital Services Act clarifies that these crisis protocols should be activated only for a limited period of time and the measures adopted should also be limited to what is strictly necessary to address the extraordinary circumstance. Another proposal is to introduce a mechanism for the oversight of content moderation through social media councils (Docquir, 2019; Tworek, 2019). The recent attempts of Facebook to establish independent oversight would be a first step towards institutionalizing mechanisms of transparency and redress in the field of content moderation (Douek, 2019). The Facebook Oversight Board could play a critical role to foster oversight in the absence of the state or when states do not have the power to introduce transparency and accountability in content moderation. Nonetheless, this self-regulatory body could introduce another layer of governance and the institutionalization of procedural rules resulting from the bureaucracy of content moderation. Indeed, the Facebook Oversight Board is a paradigmatic example of a new trend showing platforms institutionalizing procedures to foster transparency and accountability in the process of content moderation. There is also a question about the role of the international community in case of violent conflicts triggered by the spread of harmful online content. In some cases of violence and mass atrocities, international actors, including the United Nations (UN), have relied on ‘information interventions’ (Metzl, 1997), an expression which surfaced in the 1990s in response to the conflict in Rwanda and the Balkans (Price and Thompson, 2002). During cases of escalating violence, an information intervention might consist of media and conflict monitoring of the target state; peace broadcasting, which seeks to provide an outlet for non-violent voices; or media shutdowns,

266  Handbook of digital politics

which involves censoring the media whose messages are provoking conflict, such as through the bombing of radio towers. In situations where conflict is winding down – or at least where this is the hope – intervention usually focuses on measures promoting what the international community or funders believe to be a democratic and sustainable media environment. Thus, in conflict situations, information intervention consists of both short- and long-term strategies aimed at stabilizing the media environment within a specific country (Larson and Whitton, 1963). While information interventions have been applied to traditional media outlets, it is critical to underline how such a response could be relevant for social media, particularly when they have a leading role in disseminating content directly associated with mass atrocities. This would not be the first time international actors, such as the UN, have intervened to halt incendiary media. And such actions can make a difference. For example, during the Rwandan genocide in 1994, international forces overlooked the role radio had in mobilizing violence. As a result of the lessons learnt, NATO forces became more interventionist in their response to propaganda in Bosnia and Herzegovina, targeting media outlets and seizing radio transmitters that were found to be mobilizing violence. At the time, there were calls for formalizing these efforts to address the spread of inflammatory speech through the establishment of an information intervention unit (Metzl, 1997), among other solutions, and there were attempts to craft the legal and policy tools to enable media interventions within a human rights framework in times of conflict (Price and Thompson, 2002). However, these initiatives were not consolidated, and this kind of action has never effectively been translated to the world of social media. There are, inevitably, risks. Formalizing information interventions might legitimize moves towards certain modes of censorship; the threat of intervention could also disincentivize social media actors from operating in conflict-affected countries, resulting in collateral censorship, removing social media spaces for those who need it for information and communication, especially in times of war. Besides, any such information intervention runs the risk of social media actors choosing not to operate in conflict-affected countries. This would result in collateral censorship (Balkin, 1999; Wu, 2013) that could involve not just the deletion of content, but the wholesale removal of specific social media spaces. Unlike traditional media outlets, which operate within a specific region and play an important role in providing information to those in that area, the presence of a social media platform is purely down to business opportunities. Therefore, social media would potentially be incentivized to cease operating in regions where information interventions might be enacted, provoking financial and reputational losses. The international recognition of social media’s involvement in escalating violent conflict could lead to social media companies declining to provide services to countries afflicted by such conflicts. Effectively, this could mean the creation of a ‘social media vacuum’ in some areas of the world. The consequences of such a situation could be serious, especially in countries where social media is the most popular way people experience the Internet, thus providing a valuable source for accessing and engaging with international information.

Online content moderation during conflict  267

As greater self-moderation by large corporations remains a distant prospect, an Information Intervention Council (De Gregorio and Stremlau, 2021), operating within a human rights framework and ideally grounded in the UN or supranational organizations such as the African Union, is one of the few ways to effectively address online hate and disinformation in times of war and conflict. Such a council would oversee, advise, support, and guide interventions to increase transparency and accountability in the process. The body would be responsible for conducting research: for example, independent investigations into the role of social media platforms in spreading hate during conflicts (recently recommended by Meta’s Oversight Board but yet to be actioned by the company), or verifications of the roles played by other media in potential target states. It could also establish guidelines for information interventions in violent conflicts, and recommend which actions might be taken by particular actors. It would act as a key pillar of support for the development of those guidelines to nudge the private sector – social media companies in particular – to comply with specific standards in times of conflict. Given the aim of the body would not be to solve disputes or interpret international law, it should not be structured like an international tribunal, but rather take the form of a dynamic council, hosting members committed to addressing specific situations. In addition to members representing any international organizations involved, temporary members should include representatives of social media companies operating in conflict zones, individuals with an understanding of international norms relating to the international and UN-backed norm on the Responsibility to Protect, as well as members of civil society organizations.

CONCLUSIONS The governance of online speech is increasingly being shaped by a mix of public and private policies in an ad hoc and (often) arbitrary manner. This system also applies in times of war and conflicts. The process of content moderation is primarily driven by business purposes rather than human rights, or more generally public interest. Social media incentives are not always enough to lead to more accountability and efforts in addressing the spread of online hate and disinformation, and this was particularly clear in the case of inequalities in the efforts of social media in addressing conflicts in Africa or Asia and now the war in Ukraine. The current approach towards social media and conflict is dominated by an opaque governance of online content, and particularly in times of war, would benefit from greater transparency and oversight. First, how social media deal with conflicts situations and the measures they implement to avoid radicalization and escalation is not always clear. Most of the decisions are taken not only by opaque algorithmic systems or human moderators but also by the unaccountable cooperation between government and social media. Enlarging a procedural approach to content moderation will also lead to more proportionate reactions rather than general censorship. The offline harms associated with hate speech are a central justification as to why governments have proposed to criminalize

268  Handbook of digital politics

online hate and disinformation, and have, at times, turned to blunter mechanisms, such as internet shutdowns, to regulate content online. Procedural safeguards can help to provide multiple solutions to address the spread of online hate and disinformation while also mitigating general censoring measures. Second, we believe that it is critical to engage more independent and external actors. Information intervention councils can play a primary role in assessing the risks relating to the spread of online hate and disinformation in some regions and anticipate the escalation of conflicts through proportional measures. Escalating concerns between online content and offline harms call for urgent action, particularly by independent bodies such as the United Nations. The doctrine of information intervention offers one starting point to think about the potential role and responsibilities of international actors to intervene and address the most severe, or egregious cases, where online speech contributes to mass atrocities and escalating conflicts.

FURTHER READING De Gregorio, G. and Stremlau, N. (2020). Internet shutdowns and the limits of law. International Journal of Communication 14, 4224–4243. De Gregorio, G. and Stremlau, N. (2021). Information interventions and social media. Internet Policy Review 10(2). https://​policyreview​.info/​pdf/​policyreview​-2021​-2​-1567​.pdf. Gillespie T. (2018). Custodians of the Internet: Platforms, Content Moderation, and the Hidden Decisions That Shape Social Media. New Haven: Yale University Press. Zeitzoff, T. (2017). How social media is changing conflict. Journal of Conflict Resolution 61(20).

REFERENCES Akinwoto, E. (2021). Facebook’s role in Myanmar and Ethiopia under new scrutiny. The Guardian, 7 October. https://​www​.theguardian​.com/​technology/​2021/​oct/​07/​facebooks​ -role​-in​-myanmar​-and​-ethiopia​-under​-new​-scrutiny. Allen, T. and Stremlau, N. (2005). Media Policy, Peace and State Reconstruction. Crisis States Research Centre Discussion Papers, Crisis States Research Centre, London School of Economics and Political Science. http://​eprints​.lse​.ac​.uk/​28347/​. Balkin, J. M. (1999). Free speech and hostile environments. Columbia Law Review 99(8), 2295–2320. Bloch-Wehba, H. (2019). Global platform governance: Private power in the shadow of the state, SMU Law Review 72, 27. Dahlgren, P. (2005). The internet, public spheres, and political communication: Dispersion and deliberation. Political Communication 22(2), 147. De Gregorio, G. and Stremlau, N. (2020). Internet shutdowns and the limits of law. International Journal of Communication 14, 4224–4243. De Gregorio, G. and Stremlau, N. (2021). Information interventions and social media. Internet Policy Review 10(2). https://​policyreview​.info/​pdf/​policyreview​-2021​-2​-1567​.pdf. Dimitrakopoulou, D., Tzogopoulos, G., and Nikolakopoulou, A. (2014). The Role of Social Media in Violent Conflict. INFOCORE Working Paper 2014/05.

Online content moderation during conflict  269

Docquir, P. R. (2019). The Social Media Council: Bringing Human Rights Standards to Content Moderation on Social Media. CIGI, 28 October. https://​ www​ .cigionline​ .org/​ articles/​social​-media​-council​-bringing​-human​-rights​-standards​-content​-moderation​-social​ -media/​. Douek, E. (2019). Facebook’s Oversight Board: Move fast with stable infrastructure and humility. North Carolina Journal of Law & Technology 21(1). Fisher, M. (2019). Sri Lanka blocks social media, fearing more violence. The New York Times, 21 April. https://​www​.nytimes​.com/​2019/​04/​21/​world/​asia/​sri​-lanka​-social​-media​.html. Gagliardone, I. (2019). Defining online hate and its “public lives”: What is the place for “extreme speech”? International Journal of Communication 13, 3069–3087. Gillespie T. (2018). Custodians of the Internet: Platforms, Content Moderation, and the Hidden Decisions That Shape Social Media. New Haven: Yale University Press. Gillespie, T. (2020). Content moderation, AI, and the question of scale. Big Data & Society 7(2). Guggenberger, N. (2021). Essential platforms. Stanford Technology Law Review 24, 237. Helberger, N., Karppinen, K., and D’Acunto L. (2018). Exposure diversity as a design principle for recommender systems. Information, Communication & Society, 21(2), 191–207. Herzstein, R. E. (1978). The War That Hitler Won: The Most Infamous Propaganda Campaign in History. New York: Putnam Publishing Group. Howard, P. N. (2010). The Digital Origins of Dictatorship and Democracy: Information Technology and Political Islam. Oxford: Oxford University Press. Klonick, K. (2017). The new governors: The people, rules, and processes governing online speech. Harvard Law Review 131, 1598. Larson, A. and Whitton, B. (1963). Propaganda towards Disarmament in the War of Words. New York: Oceana Publications. Lessin, J. (2016). Facebook shouldn’t fact-check. The New York Times, 29 November. Metzl, J. M. (1997). Information intervention: When switching channels isn’t enough. Foreign Affairs (November), 15–20. Miles, T. (2018). UN investigators cite Facebook role in Myanmar crisis. Reuters, 12 March. https://​www​.reuters​.com/​article/​us​-myanmar​-rohingya​-facebook​-idUSKCN1GO2PN. Morozov, E. (2012). The Net Delusion: The Dark Side of Internet Freedom. New York: Public Affairs. Nicas, J. (2018). How YouTube drives people to the internet’s darkest corners. Wall Street Journal, 7 February. https://​www​.wsj​.com/​articles/​how​-youtube​-drives​-viewers​-to​-the​ -internets​-darkest​-corners​-1518020478. Noble, S. U. (2018). Algorithms of Oppression: How Search Engines Reinforce Racism. New York: New York University Press. Office of the High Commissioner for Human Rights (2019). Preventing Incitement to Hatred and Violence in the Central African Republic. https://​www​.ohchr​.org/​EN/​NewsEvents/​ Pages/​PeacekeepersDay2019​.aspx. Olewe, D. (2018). Kenya, Uganda and Tanzania in “anti-fake news campaign”. BBC News, 16 May. Ombui, E., Karani, M., and Muchemi, L. (2019). Annotation framework for hate speech identification in tweets: Case study of tweets during Kenyan Elections. In P. Cunningham and M. Cunningham (eds.), IST-Africa 2019 Conference Proceedings. IIMC International Information Management Corporation. Potkin, F. and McPherson, P. (2020). Spreading like wildfire: Facebook fights hate speech before Myanmar poll. Reuters, 5 November. https://​www​.reuters​.com/​article/​myanmar​ -election​-facebook​-idUSL4N2HQ3QU. Price, M. E. and Thompson, M. (eds.) (2002). Forging Peace: Intervention, Human Rights, and the Management of Media Space. Bloomington: Indiana University Press.

270  Handbook of digital politics

Purnell, N. (2019). Sri Lankan Islamist called for violence on Facebook before Easter attacks. Wall Street Journal, 30 April. https://​www​.wsj​.com/​articles/​sri​-lankan​-islamist​-called​-for​ -violence​-on​-facebook​-before​-easter​-attacks​-11556650954. Rahman, K. S. (2018). Regulating informational infrastructure: Internet platforms as the new public utilities. Georgetown Law and Technology Review 2(2). Roberts, S. T. (2019). Behind the Screen. New Haven: Yale University Press. Sablosky, J. (2021). Dangerous organizations: Facebook’s content moderation decisions and ethnic visibility in Myanmar. Media, Culture & Society 43(6). Schlein, L. (2018). Hate speech on social media inflaming divisions in CAR. VOA. https://​ www​.voanews​.com/​africa/​hate​-speech​-social​-media​-inflaming​-divisions​-car. Scott, M. (2021). Facebook did little to moderate posts in the world’s most violent countries. Politico, 25 October. https://​www​.politico​.eu/​article/​facebook​-content​-moderation​-posts​ -wars​-afghanistan​-middle​-east​-arabic/​. Scott, M. and Kern, R. (2022). Social media goes to war. Politico, 2 March. https://​www​ .politico​.eu/​article/​social​-media​-goes​-to​-war/​. Shahbaz, A. (2018). The rise of digital authoritarianism: Freedom on the net 2018. Freedom House, October. https://​freedomhouse​.org/​report/​freedom​-net/​2018/​rise​-digital​ -authoritarianism. Statt, N. (2020). How Facebook is using AI to combat COVID misinformation and detect ‘hateful memes’. The Verge, 12 May. https://​www​.theverge​.com/​2020/​5/​12/​21254960/​ facebook-ai-moderation-covid-19-coronavirus-hateful-memes-hate-speech. Stecklow, S. (2018). Why Facebook is losing the war on hate speech in Myanmar. Reuters. https://​www​.reuters​.com/​investigates/​special​-report/​myanmar​-facebook​-hate/​. Stremlau, N. (2018). Media, Conflict, and the State in Africa. Cambridge: Cambridge University Press. Tworek, H. (2019). Social Media Councils. Centre for International Governance Innovation. Vengattil, M. and Culliford, E. (2022). Facebook allows war posts urging violence against Russian invaders. Reuters, 11 March. https://​www​.reuters​.com/​world/​europe/​exclusive​ -facebook​-instagram​-temporarily​-allow​-calls​-violence​-against​-russians​-2022​-03​-10/​. Wong, J. C. (2021). Revealed: The Facebook loophole that lets world leaders deceive and harass their citizens. The Guardian, 12 April. https://​www​.theguardian​.com/​technology/​ 2021/​apr/​12/​facebook​-loophole​-state​-backed​-manipulation. Wu, F. T. (2013). Collateral censorship and the limits of intermediary immunity. Notre Dame Law Review 87(1), 293. Zeitzoff, T. (2017). How social media is changing conflict. Journal of Conflict Resolution 61(20).

PART IV CONTENTIOUS POLITICS, CIVIL AND NETWORKED SOCIETY

18. The Fifth Estate: a new source of democratic accountability William H. Dutton and Elizabeth Dubois1

INTRODUCTION Digital technologies offer new options for the practice of politics captured by the idea of a Fifth Estate. The concept of a Fifth Estate is based on empirical research that shows how ordinary people have been enabled by the Internet to hold institutions like the government and press more accountable. Though potentially empowering, this focus on the Fifth Estate role of networked individuals has rarely been the approach of scholars in recent studies of digital politics or in the practice of introducing digital tools to politics. That said, some scholars have identified the growing use of alternative media in politics (Lievrouw, 2011), how the Internet can enable the many to monitor the few (Bauman, 1999) or hold government accountable (Johnson et al., 2004), and how new ways are developing outside of government for networked individuals to register their opinions on policy issues (Boczkowski and Papacharissi, 2018). There have also been calls for more direct online democracy such as online petitions and voting. However, opposition to more direct forms is dominant and most often tied to three general concerns. One is a fear of direct forms of participation undermining the dialogue and debate that occurs with more representative forms of democracy. The second is a concern over equity and the disenfranchisement of citizens who are not online, even though it might be less than ten percent of the public in Canada, the United Kingdom, or United States. A third and more recent concern is over the potential manipulation of direct forms of participation by bad actors, such as in the use of Internet bots to create false levels of support or opposition and other disinformation campaigns. Reservations over direct forms of digital governance have shifted efforts to design technical innovations that will reinforce and enhance the institutions of representative democracy, such as consultation between citizens and their elected officials (Coleman, 2004). Open government initiatives have been another target of developments that could support greater transparency of democratic institutions, such as enabling governments to publish more information online in forms that the public can more easily access, such as through innovations in open linked data. As enduring as this debate over innovations in direct versus representative democracy may be, it has ignored the rise of a new organizational form which we have called the ‘Fifth Estate’ (Dutton, 2007, 2009, 2023; Dubois and Dutton, 2013). This new form poses the most realistic potential for enhancing democratic governance in 272

The Fifth Estate: a new source of democratic accountability  273

the digital age. The following sections explain the concept of a Fifth Estate, providing an illustration of how the Internet does indeed create the potential for enhancing democratic accountability, but in ways that the advocates of direct and representative democracy have largely ignored. It does not replace representative government but promises to hold representatives more accountable. It does not require universal access to the Internet, only a critical mass of citizens online. However, this new form of democratic accountability is not an inevitable outcome of technical change and raises new risks. Are members of the Fifth Estate representative of the public? Are there developments in technology or policy that could undermine the Fifth Estate? Therefore, after describing the Fifth Estate and the empirical research that supports the emergence of this new form of accountability, we briefly outline some of the most serious risks posed by contemporary developments.

THE CONCEPT OF A FIFTH ESTATE The concept of a Fifth Estate refers to the ways in which the Internet is enabling a collectivity of networked individuals to support social accountability across many sectors, including business and industry, government, politics, and the media. Members of the Fifth Estate are Internet-enabled networked individuals who enhance their informational and communicative power by accessing people and other resources online. Independent of established institutions, members of the Fifth Estate can hold institutions such as government, the press and businesses accountable through their networked interactions online. Just as the press created the potential for a Fourth Estate in the eighteenth century, the Internet is enabling a Fifth Estate in the twenty-first. The notion draws from the concept of ‘estates of the realm’, originally related to divisions in pre-industrial societies such as Ireland with respect to distinctions between the clergy, nobility and the commons. Much licence has been taken with their characterizations over the centuries. For example, American social scientists linked the three estates with Montesquieu’s tripartite conception to be identified with the legislative, executive and judicial separation of powers. However, the Irish-born philosopher Edmund Burke, echoing the historian Thomas Macaulay, stood in parliament and pointed to the press as a Fourth Estate, arguing (according to Carlyle, 1905): ‘there were Three Estates in Parliament; but, in the Reporters’ Gallery yonder, there sat a Fourth Estate more important far than they all. It is not a figure of speech, or witty saying; it is a literal fact – very momentous to us in these times.’ The rise in the twentieth century of press, radio, television and other mass media consolidated this reality as a central feature of pluralist democratic processes in countries where the press gained an independence from government and commercial control. More recently, the growing use of the Internet, social media, mobile ICTs and other digital online capabilities is creating another new ‘literal fact’: networked individuals reach out across traditional institutional and physical boundaries into what Castells (2001) has called a ‘space of flows’, rather than a ‘space of places’,

274  Handbook of digital politics

reflecting contemporary perspectives on hybrid media systems (Chadwick, 2013) and governance processes that are ‘hybrid and multijurisdictional with plural stakeholders who come together in networks’ (Bevir, 2011, p. 2). Within this hybrid space of flows, people increasingly go to find information and services. These could be located anywhere in the world and relate to local issues (e.g. taxes, political representatives, schools) and regional, national or international activities. This signals that increasingly people are likely to go first to a search engine or to a site recommended through social networking sites, rather than directly to an organization’s site or to a physical place, such as a government office, library, newspaper, university or other institution (Dutton, 2009). Our 2017 survey of seven high-income nations found that the majority of individuals are more likely to go to online search for information about politics than to any other source. Only television is as prominent as online search for access to political information (Dutton et al., 2019). When access to information is no longer tightly controlled by institutions the door is open for others to exert their own influence and control in new ways. In other words, there are potential shifts in communicative power which the Internet is enabling citizens to advance. Members of the Fifth Estate are Internet-enabled networked individuals who, independent of any single institution, are able to better access resources and people online in ways that enhance their informational and communicative power relative to other actors. This power allows them to render other institutions, like government, the press, and businesses, more accountable. Challenging Traditional Approaches The concept of a Fifth Estate challenges major perspectives on the political role of the Internet in society. In the early decades of the Internet’s diffusion, many scholars viewed it as an unimportant political resource – a technical novelty or passing fad. It was regarded as peripheral to political campaigns and elections. In contrast, others have adopted one or another form of determinism, such as seeing computer-mediated communication as inherently a technology of freedom (de Sola Pool, 1983), or quite the opposite, as inherently centralizing (Hindman, 2009). Others have found computer-based systems to most often be used to reinforce existing power structures in organizations and society – what has been called ‘reinforcement politics’ (Danziger et al., 1982). For example, Evgeny Morozov (2011) echoes this perspective, arguing that authoritarian states have used the Internet to reinforce their control of citizens and experts who are not in positions of authority. Most of these perspectives view the Internet as a significant political resource that is changing patterns of governance across multiple sectors, even if reinforcing them. Technologically deterministic perspectives continue to be prominent, such as in notions of filter bubbles, but research has increasingly supported the social shaping of the role of the Internet by a multiplicity of economic, cultural, political and other social factors influencing the behaviour of multiple actors. While the idea of a Fifth Estate recognizes this social shaping of the Internet and its role in society, it is tied to observations of how networked individuals can be empowered by virtue of their

The Fifth Estate: a new source of democratic accountability  275

strategic use of these digital media to search, originate, network, collaborate and even leak information in ways that can hold others more accountable (Table 18.1; Dutton, 2023). However, the role of a Fifth Estate is not inevitable or an inherent feature of the technology, but a description of a pattern of use observed over time. It can be undermined by actors from other estates as well as bad actors, such as malicious users. The concept of a Fifth Estate suggests that the choices and uses of the Internet by multiple actors will lead to differing political outcomes depending on the situation. Through a growing range of digital and Internet-enabled technologies, including search, social networking, emailing, texting, messaging, video blogging, tweeting, and more, individuals are reshaping not only how they connect with information, people and services, but also reshaping what they know, who they communicate with, and what services they access (Dutton, 1999). If strategic, individuals can move across the boundaries of existing institutions, sourcing their own information and networks in ways that open new opportunities for calling to account politicians, journalists, experts and other loci of power and influence. Networked individuals are often embedded in and related to various institutions, but they can act more independently, similar to when journalists blog about stories they cannot cover or publish in their newspaper. A complex but dramatic example is the former actor turned president of Ukraine, Volodymyr Zelenskyy. Zelenskyy is embedded at the top of Ukraine’s government, but he was able to exploit his communication skills and digital devices, like his mobile phone, to reach out to politicians and publics across the world. A UK journalist likened him to ‘Churchill with an iPhone’. As president and as a networked individual, he strategically used Internet and digital media to enhance his communicative power as president but also as a networked individual in dramatically effective ways. A similar case could be made about Donald Trump’s use of Twitter when he was president of the United States. Aides protesting about his communication being out of their control – for better or worse – is illustrative of a more independent networked individual of the Fifth Estate. He even tweeted about his own staff and other members of government, such as former presidents, thereby ‘violating virtually every rule of presidential messaging, decorum, and press management’ (Douglas, 2018, p. 133).

COMMUNICATIVE STRATEGIES WITH REAL-WORLD POWER SHIFTS Members of the Fifth Estate can hold others within sectors ranging from government to media to business more accountable by exerting communicative power. These members of the Fifth Estate can source information in new ways, access different people and groups, and connect with others to advance goals, independent of institutions. Table 18.1 provides an overview of the five main ways in which members of the Fifth Estate can exert communicative power. While often uninfluential, on occasion

276  Handbook of digital politics

Table 18.1

Examples of potentially effective strategies of the Fifth Estate

Strategies

Description

Example(s)

Searching

Individual finds information through search

Voter locates information about an elected

or their social network

official or policy issue

Individual creates information

Creating original post of information, images,

Originating

observations, such as an eyewitness video of an event Networking

Individuals join self-selected network(s)

Networks of friends potentially mobilized around a neighbourhood cause; social networks of patients; or climate activists

Collaborating

Using platforms to aggregate information,

Environmental sensing; churnalism, identifying

observations, experiences

poor journalistic practices; 38 Degrees; bribery websites

Leaking

Individuals leak information in ways that can Whistleblowers reach sites that distribute be publicly accessible online

information, such as blogs, WikiLeaks or news media

Source:

Adapted from Dutton (2023).

they are dramatically successful. A growing variety of cases demonstrate the potential for these strategies to enable new forms of accountability. Searching and originating information are perhaps the most obvious Fifth Estate actions that can be strategic but in some cases accidentally powerful. The Internet enables individuals to produce original content and design groups by reconfiguring access – changing the ways people seek out information and services, and the outcome of these activities (Dutton, 1999). The ability to search and access desired information from a variety of sources paired with the ability to produce original information highlight the fact that members of the Fifth Estate can act more independently of existing institutions. A local level example comes from a primary school. In 2012, with encouragement from her father, a 9-year-old girl in Scotland created a blog for her school writing project, called ‘NeverSeconds’. She took a photo of her school lunches on her mobile phone and posted them on her blog with commentary. NeverSeconds became popular, leading officials to request that she did not bring her camera to school. This fuelled greater interest in her site, generating millions of page views, funding for charities, and a book with her father about her project (Payne and Payne, 2012). As a young girl, Martha Payne was able to produce original information, distribute it globally, and foster a debate that led to change in the practices of her school and other schools across the UK. Greta Thunberg presents another example. In 2018, when she was 15 years old, she began a strike, spending her school days sitting outside the Swedish Parliament with a sign that read ‘school strike for climate’. Her photo was posted on Instagram and other social media, soon creating massive news coverage. Her impact worldwide has been phenomenal.

The Fifth Estate: a new source of democratic accountability  277

These examples illustrate that the Fifth Estate can be relatively independent of an institution and that it is not the equivalent of a social or political movement. One networked individual can play a critical role in holding an institution more accountable. Even social movements, which often have strong hierarchical structures, must cope with the potential of networked individuals to source their own information and counter the narrative of a movement. They also show how the Fifth Estate is not a homogeneous group. Multiple Fifth Estate actors may act independently but interact in varying ways. Greta Thunberg’s origination of her strike action led many other networked individuals to create content from similar actions around climate change and network with like-minded individuals. Following from this point, the examples also illustrate that networking is another effective form of Fifth Estate action. Networked individuals seek like-minded people online through such means as social networking sites, instant messaging, blogs, and even email. With the Internet they can bypass such traditional barriers, such as space and time, and access people in a ‘space of flows’ that spans the ‘space of places’ anchored in common geographical and institutional boundaries (Castells, 1996, p. 411). By reconfiguring how people source information and reshape networks of individuals, the Internet can enhance their communicative power vis-à-vis institutions and set the stage for increased accountability. For example, a patient within a health service can use the Internet to source their own information or join networks of other individuals with similar health or medical issues, such as the network of hundreds of families linked to the UK Children with Diabetes Advocacy Group. Likewise, physicians can rely on online institutional resources but also explore new sources of information online and network with other physicians who are outside their own institutional setting, as can be found on Sermo, one of the largest social networking platforms for physicians. The Fifth Estate is not equivalent to the mass media, but the strategies of bloggers and other Fifth Estate activities often entail gaining media coverage as one means to enhance their communicative power. Reflecting what Chadwick calls hybridity (2013), there can be a clear complementarity between the traditional media and the Fifth Estate. While the Internet’s broad social roles in government, politics and other sectors may parallel those of traditional media at times, such as in simply getting information to a broader public, the Fifth Estate differs from traditional media in how it helps networked individuals drive opportunities for greater social accountability across institutions, even occasionally setting the mass media agenda. However, bad actors can undermine the press and the Fifth Estate, creating a need for Internet users, viewers, and the general public to be sceptical of any one source of information and use the Internet and related media to fact check and consult alternative sources. Likewise, collaboration as a strategy of the Fifth Estate provides another means for networked individuals to exert communicative power (Table 18.1). This collaborative strategy is illustrated most vividly by the ability of individuals to contribute to collective intelligence and collective or distributed observation. For example, collective intelligence such as through smartphone apps enables the creation of inde-

278  Handbook of digital politics

pendent sources of information about pollution levels in cities. Individuals can pool their observations, such as used in the creation of bribery websites to monitor and map incidences of corruption. Such forms of distributed collaboration are not directly dependent on any one institutional source or any single estate. Another and more problematic Fifth Estate strategy is focused on leaking. An insider leaking information from an institution, such as Daniel Elsberg’s copying and leaking the Pentagon Papers to the press in 1971, is not new. However, the ability of a leak to be posted online and instantaneously available worldwide has been one reason why leaks have become a more common feature of contemporary politics. The Fifth Estate might be described in terms of networks of accountability rather than a formally organized institution. Internet use enables the creation of alternative sources of information and collaboration that are not directly dependent on any one institutional source or any single estate. Internet-enabled individuals, even those whose primary networking activities are social, can often create networks that span across standard geographical, organizational, and institutional networks, to link with others online. Networked individuals are enabled to build and exercise their ‘communicative power’ by using ICTs to reconfigure networks in ways that can lead to real-world power-shifts. For instance, the relationships between media producers, gatekeepers and consumers are changed profoundly when previously passive audiences generate and distribute their own content and when search engines point to alternative sources reflecting different views on a topic. Access to online resources that incorporate and go beyond more traditionally accessible institutions is supported by the near-ubiquity of the Internet and related digital technologies like the smartphone. Individuals can then network with information and people to change their relationships with more institutionalized centres of authority in the other estates, thereby holding them more socially accountable through the interplay between ever-changing networks of networks. Many cases could illustrate this role of the Fifth Estate that leads to real-word power shifts. They would include: the video of a 17-year-old black man, Trayvon Martin, being shot in Sanford, Florida, in 2012; the role of an ordinary pedestrian out shopping in 2020 when she was able to video and post her eyewitness account on Facebook of the arrest of George Floyd in Minneapolis, Minnesota, whose death resulted from an arresting officer’s knee on Floyd’s neck for 9 minutes – one of the most prominent events sparking the BlackLivesMatter (BLM) movement; and the rise of the #MeToo movement as a consequence of individual women exposing online the sexual abuse and sexual harassment they had experienced.

NETWORKED INSTITUTIONS AND INDIVIDUALS Individuals are not the only ones making use of the Internet and related technologies. Indeed, institutions are also becoming increasingly ‘networked’, using digital technology to advance their goals. Table 18.2 outlines examples of the comparative

The Fifth Estate: a new source of democratic accountability  279

Table 18.2

Networked institutions and networked individuals

Arena

Networked individuals of the Fifth Estate

Networked institutions of the other estates

Governance and democracy Social networking; blogging about government Digital government; digital democracy; or policy; individual podcasts on politics

inviting online comments on proposed legislation or regulation

Press and media

Bloggers, Vlogs; online news aggregators;

Online journalism; radio and TV news

Wikipedia contributors; citizen journalists

podcasts, enabling news readers to comment online

Business and commerce

Peer-to-peer file sharing (e.g. music

Online business-to-business; business-to-

downloads); collaborative network

consumer (e.g. e-shopping, e-banking);

organizations; self-publishing; rating products virtual organizations and services Work and the organization

Self-selected work collaborations; systems for Flatter networked structures; enabling co-creation and distribution (e.g. open source

networking to create flexible work

software); working from home or on the move; location and times; corporate blogs leaking online

Source: Adapted from Dutton (2009, p. 7).

activities of networked individuals and networked institutions across a variety of sectors, often overlapping and interrelated in complementary ways. There are many common but also distinctive ways in which digital technologies are used across different social and political systems. The following section looks into the ways networked individuals of the Fifth Estate and networked institutions interact within governance, the press, business, and work to illustrate a pattern that extends to many other sectors not illustrated here, such as education. Government and Democracy on the Line Digital government initiatives, such as enabling citizens and businesses to obtain public services via the Internet, such as in paying taxes, are becoming the norm in many nations. For example, the Covid-19 pandemic has generated a step-jump in the provision of digital government services. Governmental initiatives have been paralleled by efforts to use the Internet to support democratic institutions and processes. Some critics suggest digital democracy could erode traditional institutions of representative deliberative democracy by offering over-simplistic and effortless ‘point and click’ participation, what has been variously called slacktivism or clicktivism. These criticisms are based on an institution-centric view of the Internet, focused on voting or parliamentary consultations, for example, rather than considering the role of the individual outside of existing institutional processes, such as in sourcing their own information. As Greta Thunberg said, ‘blah, blah, blah’, she wanted action not words, but her words mattered. The Fifth Estate is not necessarily reliant on institutions and therefore presents new opportunities and threats. Networked individuals can challenge institutional author-

280  Handbook of digital politics

ity and provide a novel means for holding politicians and mainstream institutions to account through ever-changing networks of individuals, who form and re-form continuously depending on the issue generating the particular network (e.g. to form ad hoc ‘flash mob’ meetings at short notice through social networks and mobile communication). For example, the use of texting after the March 2004 Madrid train bombings to organize anti-government rallies challenging the government’s claims and contributing to unseating that administration by quickly providing people with important information and instruction which enabled mobilization. Pro-democracy protests across the Middle East and North Africa in early 2011 or Ukraine in 2014 further illustrate the potential for networked individuals to challenge governmental authorities and other institutions. In 2022, the Russian invasion of Ukraine was accompanied by a war of words online, including individuals sourcing evidence of war crimes by using open-source intelligence. Governments, such as in the US and UK, are making information available online in ‘user friendly’ forms as a key element of open government initiatives promoting greater transparency and accountability. These initiatives illustrate how other estates can support the role of the Fifth Estate. Examples of governments using technology to limit the power of the Fifth Estate are also extensive, such as by filtering and censoring the press and the Internet. During the Russo-Ukraine War, Russia blocked access to independent press as well as Internet and social media sites in an effort to use state broadcasting to support the government’s narrative. However, a minority of the Russian public were able to gain access to some alternative sources via virtual private networks and some social media, such as Telegram albeit they were also subject to influence campaigns financed by the Kremlin. Attempts to control access to digital information often fail, but such threats to the Fifth Estate remains real. The Press As citizen journalists, bloggers, information seekers and producers, networked individuals are increasingly able to contest claims made by traditional media sources, provide new details and perspectives on specific news items, and thus, hold the Fourth Estate more accountable. Bloggers have been said to be ‘watching the watchdogs’ – the press (Cooper, 2006). On the flip side, traditional media outlets are increasingly incorporating products of the Fifth Estate into their own reporting. Many news programmes include reviews of social media responses to an issue. Other programmes create and promote specific Twitter hashtags to generate and follow conversation related to their programme. Yet, the relationship between the Fourth and Fifth Estate is not always complementary. The Internet and its users are criticized for eroding the quality of the public’s information environment and undermining the integrative role of traditional Fourth Estate media in society. This includes claims the Internet is marginalizing expertise or high-quality journalistic coverage by proliferating misinformation, trivial non-information and propaganda created by amateurs (Keen, 2007; Nichols, 2017). However, the Fifth Estate is not a substitute for expertise or journalism and in

The Fifth Estate: a new source of democratic accountability  281

many cases the experts and citizen journalists have the expertise or the experience, such as being eyewitnesses, to make positive contributions. Others have claimed that social media are locking people into virtual ‘echo chambers’ where personal prejudices are reinforced as Internet users choose to access only a narrow spectrum from the vast array of content that agrees with their prejudices (Sunstein, 2017). Counter to such expectations, studies found that Internet users consult multiple (four or more) sources of information, including the Internet and mass media, making it unlikely that any one technology traps them into any echo chamber. To the contrary, it provides many more alternative sources of information, which are often used (Dutton et al., 2019; Dubois and Blank, 2018). The traditional mass media embodies equivalent weaknesses (e.g. a focus on sensational negative news stories, poor quality reporting and celebrity trivia). For example, the networks of communication among journalists can create what has been called ‘pack journalism’ that can be exacerbated online (Crouse, 1972). There is an unjustified assumption that the Internet substitutes for, rather than complements, traditional media. Many Internet users read online newspapers or news services, although not always the same newspaper as they read offline (Dutton et al., 2019). Thus, the Internet is indeed a source of news that in part complements, or even helps sustain, the Fourth Estate (Dutton et al., 2019). Business Organizations and Work The Fifth Estate has transformative potential at all levels in businesses and other private sector organizations. Geographically distributed individuals cooperating together to form collaborative network organizations (CNOs) to co-create or co-produce information products and services are one example. The online encyclopaedia Wikipedia and open-source software products such as the Firefox Web browser are examples of this phenomenon, becoming widely used and trusted despite initial doubts about the merits of their methods of creative co-production. There are concerns CNOs may blur the boundaries and operations of the firm or undermine the firm’s productivity. Instead, evidence suggests individuals generally choose to collaborate in such networks primarily to enhance their own productivity, performance or esteem (Surowiecki, 2004). As consumers become increasingly empowered to hold businesses accountable, such as through Internet orchestrated boycotts, or better-informed consumer groups, the role of the Fifth Estate in business and industry will increase. For example, networked individuals are increasingly challenging the information practices of major Internet companies, such as Facebook and Google, and leading them to alter their approaches to protecting the privacy of their users and many other information policies driven by what can easily be identified as Fifth Estate accountability.

282  Handbook of digital politics

IMPLICATIONS OF A FIFTH ESTATE This chapter provides illustrations and evidence of the emergence of a Fifth Estate and systematic research is needed to address issues over its political dynamics and societal implications. The Internet can be used to empower networked individuals because it enables them to harness communicative power to hold other institutions to account. However, there are threats as well, such as from malevolent actors. There are also well-intentioned actors that can develop policy and technical initiatives that might inadvertently undermine the Fifth Estate. For example, the enhanced communicative power of networked individuals has led to efforts to censor and otherwise control the Fifth Estate, including calls for disconnecting the Internet, blocking sites, such as Twitter, and arresting bloggers. This mirrors familiar forms of governmental control of traditional media, such as bans, closures and the arrest of journalists of the Fourth Estate. The Internet’s opening of doors to an array of user-generated content equally invites techniques deployed by governments and others to block, monitor, filter and otherwise constrain Internet traffic or create an Internet ‘kill-switch’ (Deibert et al., 2008). Government efforts to control Internet content include the ‘Great Firewall of China’, and the Russian government’s efforts to restrict the country’s Internet and social media services during the Russo-Ukraine War launched in 2022. Moreover, even liberal democratic regimes, such as the UK, EU, and USA are devising approaches to protect the safety of children and the public in ways that will involve greater censorship of Internet content. Networked individuals continue to challenge attempts to control Internet access and circumvent censorship. A once prominent university research project, www​ .herdict​.org, accepted and published reports from Internet users of inaccessible websites around the world, but as research projects end, it was stopped in 2019. Reporters Sans Frontiers and other civil society organizations continue to support worldwide efforts to sustain and reinforce a free and open Internet. Moreover, networked individuals acting as a Fifth Estate use the Internet and other digital media, such as social media, blogs, Twitter and messaging, to criticize media and governmental actions that restrict an open and global Internet and provide alternative information and support to each other, and to call the government and other estates to account. The chapter has argued that the Fifth Estate does not require universal access. For that reason, despite digital divides, the Internet has achieved a critical mass in many nations enabling networked individuals to become a significant force for accountability. This enables the Fifth Estate to play an important political role even in nations such as Indonesia, which has relatively low proportions, but large numbers of Internet users. Nevertheless, because Internet use has become an increasingly central aspect of everyday life and work in networked societies, disparities in access to the Internet are increasingly of substantive social, economic and political significance. Similarly, when considering the implications of a Fifth Estate on governance systems broadly, we see that not all citizens need to consciously strategize the use of the Internet to harness communicative power for the Fifth Estate to be a meaningful

The Fifth Estate: a new source of democratic accountability  283

political force. Individuals who simply make use of the Internet and related tools for everyday life and work, such as for social networking, are creating the social and technical infrastructure than can contribute to social accountability. An ordinary person going about their everyday life can record an eyewitness account of a protest, police action, or speech that has the potential to shape policy and practice. It is these members of the Fifth Estate who are challenging governance structures, at times indirectly, when holding others within the governance system to account. Members of the Fifth Estate are inserting themselves into political conversations which were once reserved for institutional bodies and political elites by using the Internet to source, originate, network, collaborate or leak critical information that can reshape the outcome of political debate and public policy.

DIRECTIONS FOR FUTURE RESEARCH The role of the Internet in ‘reconfiguring access’ is why information politics has become so central to the digital age (Dutton, 1999). How the Internet potentially reconfigures access is shaped by patterns of digital choices by networked individuals, which can impact control over access to information, communication, services and technologies of the digital age. As argued in this chapter, this reconfiguration of access has enabled networked individuals to enhance their informational and communicative power, giving rise to a Fifth Estate. However, its role could be enhanced or undermined by changes in technology, policy and regulation. The impact of networked individuals seeking to shape access to and from the outside world, in local and global contexts, has supported the rise of the Fifth Estate, but also the role of the Internet in all the other estates. In this sense, the Fifth Estate is the unintended outcome of an ecology of choices made by multiple actors, rather than a specific organizational form people seek to join. This is demonstrated, for instance, in the strategies of government agencies, politicians, lobbying groups, news media, bloggers, business owners and others trying to gain access to citizens over the Internet, countered by networked individuals seeking to source their own information and networks. What individuals know is one outcome of this ecology of choices and strategies that creates a Fifth Estate role. This chapter has argued that distinctive features of the Fifth Estate make it worthy of being considered a new estate of at least equal importance to the Fourth, in addition to being the first estate not to be essentially institution-centric. There are three main directions for research on the Fifth Estate. First, an increased understanding of the social dynamics of how members of the Fifth Estate interact and collaborate is important, such as might be gained through case studies of its role in particular political events. While many researchers focus on the ways in which social movements and campaigns make use of the Internet, few examine the ways in which individuals use the Internet to hold institutions from a variety of sectors accountable. The focus on the individual in their personal network is an important perspective

284  Handbook of digital politics

and which potentially provides insight into political systems, but which is especially difficult to do while protecting the privacy and trust of individuals. Second, research into how governance systems and specific institutions do and do not respond to and integrate the Fifth Estate is needed. While research looking at how various institutions use the Internet and how individuals use the Internet exist, there is a lack of understanding about how each might integrate the other into their strategic planning, daily activities, and/or organizational structure. Organizations increasingly are anticipating potential insider leaks but less often focused on how networked individuals can source information in other ways that affect their accountability. Finally, systematic research on the rise and fall of the Fifth Estate in varying contexts is paramount. The potential implications of economic, social, cultural, and political factors are not well known in the context of Fifth Estate research and must be developed further. Policy changes underway in many nations, such as the UK’s online safety bill and European Union regulatory initiatives, could dramatically undermine the vitality of the Fifth Estate and need to be tracked over time. As the Fifth Estate better understands its own existence and power, it is more likely to speak the truth to power and protect its rights and responsibilities as a Fifth Estate of governance.

NOTE 1. This chapter is a revision of a chapter by Dutton and Dubois entitled ‘The Fifth Estate: a rising force of pluralistic accountability’, pp. 51–66 in the 2015 edition of The Handbook of Digital Politics, ed. Stephen Coleman and Deen Freelon, Cheltenham, UK and Northampton, MA, USA: Edward Elgar Publishing.

FURTHER READING Dubois, E. and Dutton, W. H. (2013). ‘The fifth estate in internet governance: Collective accountability of a Canadian policy initiative’, Revue française d’Etudes Américaines RFEA, 134. Dutton, W. H. (2005). ‘The internet and social transformation: Reconfiguring access’, in W. H. Dutton, B. Kahin, R. O’Callaghan, and A. W. Wyckoff (eds.), Transforming Enterprise (pp. 375–397). Cambridge, MA: MIT Press. Dutton, W. H. (2009). ‘The fifth estate emerging through the network of networks’, Prometheus, 27(1), 1–15. Dutton, W. H. (2015). ‘The internet’s gift to democratic governance: The fifth estate’, in S. Coleman, G. Moss, and K. Parry (eds.), Can the Media Serve Democracy? Essays in Honour of Jay G. Blumler (pp. 164–173). Basingstoke: Palgrave Macmillan. Dutton, W. H. (2023). The Fifth Estate: The Power Shift of the Digital Age. New York: Oxford University Press.

The Fifth Estate: a new source of democratic accountability  285

REFERENCES Bauman, Z. (1999). In Search of Politics. Cambridge: Polity Press. Bevir, M. (2011). ‘Governance as theory, practice, and dilemma’, in M. Bevir (ed.), The SAGE Handbook of Governance (pp.1–16). London: Sage. Boczkowski, P. J. and Papacharissi, Z. (eds.) (2018). Trump and the Media. Cambridge, MA: MIT Press. Carlyle, T. (1905). On Heroes: Hero Worship and the Heroic in History. Repr. of the Sterling Edition of Carlyle’s Complete Works. Teddington, Middlesex: The Echo Library. Castells, M. (1996). The Rise of the Network Society. Oxford: Blackwell. Castells, M. (2001). The Internet Galaxy. Oxford: Oxford University Press. Chadwick, A. (2013). The Hybrid Media System: Politics and Power. Oxford and New York: Oxford University Press. Coleman, S. (2004). Connecting parliament to the public via the internet. Information, Communication & Society, 7(1), 1–22. Cooper, S. D. (2006). Watching the Watchdog: Bloggers as the Fifth Estate. Sokane, WA: Marquette Books. Crouse, T. (1972). The Boys on the Bus. New York: Random House. Danziger, J. N., Dutton, W. H., Kling, R., and Kraemer, K. L. (1982). Computers and Politics. New York: Columbia University Press. de Sola Pool, I. (1983). Technologies of Freedom. Cambridge, MA: Harvard University Press. Deibert, R., Palfrey, J., Rohozinski, R., and Zittrain, J. (eds.) (2008). Access Controlled. Cambridge, MA: MIT Press. Douglas, S. J. (2018). ‘Breaking the rules of political communication: Trump’s successes and miscalculations’, in P. J. Boczkowski and Z. Papacharissi (eds.), Trump and the Media (pp. 133–141). Cambridge, MA: MIT Press. Dubois, E. and Blank, G. (2018). ‘The echo chamber is overstated’, Information, Communication & Society, 21(5), 729–745. Dubois, E. and Dutton, W. H. (2013). ‘The fifth estate in internet governance: Collective accountability of a Canadian policy initiative’, Revue française d’Etudes Américaines RFEA, 134. Dutton, W. H. (1999). Society on the Line: Information Politics in the Digital Age. Oxford and New York: Oxford University Press. Dutton, W. H. (2007). ‘Through the network (of networks) – the fifth estate’. Inaugural lecture, Examination Schools, University of Oxford, 15 October. http://​webcast​.oii​.ox​.ac​.uk/​?view​ =​Webcast​&​ID​=​20071015​_208. Dutton, W. H. (2009). ‘The fifth estate emerging through the network of networks’, Prometheus, 27(1), 1–15. Dutton, W. H. (2023). The Fifth Estate: The Power Shift of the Digital Age. New York: Oxford University Press. Dutton, W. H. and Dubois, E. (2013). ‘The fifth estate of the digital world’, in G. Youngs (ed.), Digital World: Connectivity, Creativity and Rights (pp. 131–143). London: Routledge. Dutton, W. H., Reisdorf, B. C., Blank, G., Dubois, E., and Fernandez, L. (2019). ‘The internet and access to information about politics: Searching through filter bubbles, echo chambers, and disinformation’, in M. Graham and W. H. Dutton (eds.), Society and the Internet: How Networks of Information and Communication are Changing our Lives, 2nd edition (pp. 228–247). Oxford: Oxford University Press. Hindman, M. (2009). The Myth of Digital Democracy. Princeton and Oxford: Princeton University Press. Johnson, D. R., Crawford, S. P., and Palfrey, J. G. (2004). ‘The accountable net: Peer production of internet governance’, Virginia Journal of Law and Technology, 9. http://​ssrn​.com/​ abstract​=​529022.

286  Handbook of digital politics

Keen, A. (2007). The Cult of the Amateur: How Today’s Internet is Killing Our Culture. New York: Doubleday. Lievrouw, L. A. (2011). Alternative and Activist New Media. Cambridge: Polity Press. Morozov, E. (2011). The Net Delusion: How Not to Liberate the World. London: Penguin Books. Nichols, T. (2017). The Death of Expertise: The Campaign Against Established Knowledge and Why It Matters. New York: Oxford University Press. Payne, M. and Payne, D. (2012). NeverSeconds. Glasgow: Cargo Publishing. Sunstein, C. R. (2017). #Republic: Divided Democracy in the Age of Social Media. Princeton and Oxford: Princeton University Press. Surowiecki, J. (2004). The Wisdom of Crowds. New York: Doubleday.

19. The logic of connective action: digital media and the personalization of contentious politics W. Lance Bennett and Alexandra Segerberg1

From the Arab Spring and los indignados in Spain, to Occupy Wall Street (and beyond), large-scale, sustained protests are using digital media in ways that go beyond sending and receiving messages. Some of these action formations contain relatively small roles for formal brick-and-mortar organizations. Others involve well-established advocacy organizations, in hybrid relations with other organizations, using technologies that enable personalized public engagement. Both stand in contrast to the more familiar organizationally managed and brokered action conventionally associated with social movement and issue advocacy. This chapter examines the organizational dynamics that emerge when communication becomes a prominent part of organizational structure. It argues that understanding such variations in large-scale action networks requires distinguishing between at least two logics that may be in play: the familiar logic of collective action associated with high levels of organizational resources and the formation of collective identities, and the less familiar logic of connective action based on personalized content sharing across media networks. In the former, introducing digital media does not change the core dynamics of the action. In the case of the latter, it does. Building on these distinctions, the chapter presents three ideal types of large-scale action networks that are becoming prominent in the contentious politics of the contemporary era. With the world economy in crisis, the heads of the 20 leading economies held a series of meetings beginning in the fall of 2008 to coordinate financial rescue policies. Wherever the G20 leaders met, whether in Washington, London, St Andrews, Pittsburgh, Toronto, or Seoul, they were greeted by protests. In London, anti-capitalist, environmental direct activist, and non-governmental organization (NGO)-sponsored actions were coordinated across different days. The largest of these demonstrations was sponsored by a number of prominent NGOs including Oxfam, Friends of the Earth, Save the Children and World Vision. This loose coalition launched a Put People First (PPF) campaign promoting public mobilization against social and environmental harms of “business-as-usual” solutions to the financial crisis. The website for the campaign carried the simple statement: Even before the banking collapse, the world suffered poverty, inequality and the threat of climate chaos. The world has followed a financial model that has created an economy fuelled by ever-increasing debt, both financial and environmental. Our future depends on 287

288  Handbook of digital politics

creating an economy based on fair distribution of wealth, decent jobs for all and a low carbon future. (Put People First, 2009)

The centrepiece of this PPF campaign was a march of some 35,000 people through the streets of London a few days ahead of the G20 meeting, to give voice and show commitment to the campaign’s simple theme. The London PPF protest drew together a large and diverse protest with the emphasis on personal expression, but it still displayed what Tilly (2004, 2006) termed “WUNC”: worthiness, embodied by the endorsements by some 160 prominent civil society organizations and recognition of their demands by various prominent officials; unity, reflected in the orderliness of the event; numbers of participants, that made PPF the largest of a series of London G20 protests and the largest demonstration during the string of G20 meetings in different world locations; and commitment, reflected in the presence of delegations from some 20 nations who joined local citizens in spending much of the day listening to speakers in Hyde Park or attending religious services sponsored by church-based development organizations.2 The large volume of generally positive press coverage reflected all of these characteristics, and responses from heads of state to the demonstrators accentuated the worthiness of the event (Bennett and Segerberg, 2011).3 The protests continued as the G20 in 2010 issued a policy statement making it clear that debt reduction and austerity would be the centrepieces of a political program that could send shocks through economies from the United States and the UK, to Greece, Italy, and Spain, while pushing more decisive action on climate change onto the back burner. Public anger swept cities from Madison to Madrid, as citizens protested that their governments, no matter what their political stripe, offered no alternatives to the economic dictates of a so-called neoliberal economic regime that seemed to operate from corporate and financial power centres beyond popular accountability and, some argued, even beyond the control of states. Some of these protests seemed to operate with surprisingly light involvement from conventional organizations. For example, in Spain los indignados (the indignant ones) mobilized in 2011 under the name of 15M for the date (May 15) of the mass mobilization that involved protests in some 60 cities. One of the most remarkable aspects of this sustained protest organization was its success at keeping political parties, unions, and other powerful political organizations out: indeed, they were targeted as part of the political problem. There were, of course, civil society organizations supporting 15M, but they generally stayed in the background to honour the personalized identity of the movement: the faces and voices of millions of ordinary people displaced by financial and political crises. The most visible organization consisted of the richly layered digital and interpersonal communication networks centering around the media hub of Democracia real YA!4 This network included links to more than 80 local Spanish city nodes, and a number of international solidarity networks. On the one hand, Democracia real YA! seemed to be a website, and on the other, it was a densely populated and effective organization. It makes sense to think

The logic of connective action  289

of the core organization of the indignados as both of these and more, revealing the hybrid nature of digitally mediated organization (Chadwick, 2013). Given its seemingly informal organization, the 15M mobilization surprised many observers by sustaining and even building strength over time, using a mix of online media and offline activities that included face-to-face organizing, encampments in city centres, and marches across the country. Throughout, the participants communicated a collective identity of being leaderless, signalling that labour unions, parties, and more radical movement groups should stay at the margins. A survey of 15M protesters by a team of Spanish researchers showed that the relationships between individuals and organizations differed in at least three ways from participants in an array of other more conventional movement protests, including a general strike, a regional protest, and a pro-life demonstration: (1) where strong majorities of participants in other protests recognized the involvement of key organizations with brick-and-mortar addresses, only 38 percent of indignados did so; (2) only 13 percent of the organizations cited by 15M participants offered any membership or affiliation possibilities, in contrast to large majorities who listed membership organizations as being important in the other demonstrations; and (3) the mean age range of organizations (such as parties and unions) listed in the comparison protests ranged from 10 to over 40 years, while the organizations cited in association with 15M were, on average, less than three years old (Anduiza et al., 2014). Despite, or perhaps because of, these interesting organizational differences, the ongoing series of 15M protests attracted participation from somewhere between 6 and 8 million people, a remarkable number in a nation of 40 million (RTVE, 2011). Similar to PPF, the indignados achieved impressive levels of communication with outside publics both directly via images and messages spread virally across social networks, and indirectly when anonymous Twitter streams and YouTube videos were taken up as mainstream press sources. Their actions became daily news fare in Spain and abroad, with the protesters receiving generally positive coverage of their personal messages in local and national news; again defying familiar observations about the difficulty of gaining positive news coverage for collective actions that spill outside the bounds of institutions and take to the streets (Gitlin, 1980).5 In addition to communicating concerns about jobs and the economy, the clear message was that people felt the democratic system had broken to the point that all parties and leaders were under the influence of banks and international financial powers. Despite avoiding association with familiar civil society organizations, lacking leaders, and displaying little conventional organization, los indignados, similar to PPF, achieved high levels of WUNC. Two broad organizational patterns characterize these increasingly common digitally enabled action networks. Some cases, such as PPF, are coordinated behind the scenes by networks of established issue advocacy organizations that step back from branding the actions in terms of particular organizations, memberships, or conventional collective action frames. Instead, they cast a broader public engagement net using interactive digital media and easy-to-personalize action themes, often deploying batteries of social technologies to help citizens spread the word over their

290  Handbook of digital politics

personal networks. The second pattern, typified by the indignados and the Occupy protests in the United States, entails technology platforms and applications taking the role of established political organizations. In this network mode, political demands and grievances are often shared in very personalized accounts that travel over social networking platforms, e-mail lists, and online coordinating platforms. For example, the easily personalized action frame, “We are the 99%”, that emerged from the US Occupy protests in 2011 quickly travelled the world via personal stories and images shared on social networks such as Tumblr, Twitter, and Facebook. Compared to many conventional social movement protests, with identifiable membership organizations leading the way under common banners and collective identity frames, these more personalized, digitally mediated collective action formations have frequently been larger; have scaled up more quickly; and have been flexible in tracking moving political targets and bridging different issues. Whether we look at PPF, Arab Spring, the indignados, or Occupy, we note surprising success in communicating simple political messages directly to outside publics using common digital technologies such as Facebook or Twitter. Those media feeds are often picked up as news sources by conventional journalism organizations.6 In addition, these digitally mediated action networks often seem to be accorded higher levels of WUNC than their more conventional social movement counterparts. This observation is based on comparisons of more conventional anti-capitalist collective actions organized by ­movement groups, in contrast with both the organizationally enabled PPF protests and the crowd-enabled 15M mobilizations in Spain and the Occupy Wall Street protests, which quickly spread to thousands of other places. The differences between both types of digitally mediated action and more conventional organization-centred and brokered collective actions led us to see interesting differences in underlying organizational logics and in the role of communication as an organizing principle. The rise of digitally networked action (DNA) has been met with some understandable skepticism about what really is so very new about it, mixed with concerns about what it means for the political capacities of organized dissent. We are interested in understanding how these more personalized varieties of collective action work: how they are organized, what sustains them, and when they are politically effective. We submit that convincingly addressing such questions requires recognizing the differing logics of action that underpin distinct kinds of collective action networks. This chapter thus develops a conceptual framework of such logics, on the basis of which further questions about DNA may then be tackled. We propose that more fully understanding contemporary large-scale networks of contentious action involves distinguishing between at least two logics of action that may be in play: the familiar logic of collective action, and the less familiar logic of connective action. Doing so in turn allows us to discern three ideal action types, of which one is characterized by the familiar logic of collective action, and two other types involve more personalized action formations that differ in terms of whether formal organizations are more or less central in enabling a connective communication logic. A first step in understanding DNA, the DNA at the core of connective

The logic of connective action  291

action, lies in defining personalized communication and its role along with digital media in the organization of what we call c­ onnective action.

PERSONAL ACTION FRAMES AND SOCIAL MEDIA NETWORKS Structural fragmentation and individualization in many contemporary societies constitute an important backdrop to the present discussion. Various breakdowns in group memberships and institutional loyalties have trended in the more economically developed industrial democracies, resulting from pressures of economic globalization spanning a period from roughly the 1970s through to the end of the last century (Bennett, 1998; Putnam, 2000). These sweeping changes have produced a shift in social and political orientations among younger generations in the nations that we now term the post-industrial democracies (Inglehart, 1997). These individualized orientations result in engagement with politics as an expression of personal hopes, lifestyles, and grievances. When enabled by various kinds of communication technologies, the resulting DNAs in post-industrial democracies bear some remarkable similarities to action formations in decidedly undemocratic regimes such as those swept by the Arab Spring. In both contexts, large numbers of similarly disaffected individuals seized upon opportunities to organize collectively through access to various technologies (Howard and Hussain, 2011). Those connectivities fed in and out of the often intense face-to-face interactions going on in squares, encampments, mosques, and general assembly meetings. In personalized action formations, the nominal issues may resemble older movement or party concerns in terms of topics (environment, rights, women’s equality, and trade fairness) but the ideas and mechanisms for organizing action become more personalized than in cases where action is organized on the basis of social group identity, membership, or ideology. These multifaceted processes of individualization are articulated differently in different societies, but include the propensity to develop flexible political identifications based on personal lifestyles (Giddens, 1991; Inglehart, 1997; Bennett, 1998; Bauman, 2000; Beck and Beck-Gernsheim, 2002), with implications in collective action (McDonald, 2002; Micheletti, 2003; Della Porta, 2005) and organizational participation (Putnam, 2000; Bimber et al., 2012). People may still join actions in large numbers, but the identity reference is more derived through inclusive and diverse large-scale personal expression rather than through common group or ideological identification. This shift from group-based to individualized societies is accompanied by the emergence of flexible social “weak tie” networks (Granovetter, 1973) that enable identity expression and the navigation of complex and changing social and political landscapes. Networks have always been part of society, to help people navigate life within groups or between groups, but the late modern society involves networks that become more central organizational forms that transcend groups and constitute core organizations in their own right (Castells, 2000). These networks are estab-

292  Handbook of digital politics

lished and scaled through various sorts of digital technologies that are by no means value-neutral in enabling quite different kinds of communities to form and diverse actions to be organized, from auctions on eBay to protests in different cultural and social settings. Thus, the two elements of “personalized communication” that we identify as particularly important in large-scale connective action formations are: 1. Political content in the form of easily personalized ideas such as PPF in the London 2009 protests, or “We are the 99%” in the later Occupy protests. These frames require little in the way of persuasion, reason, or reframing to bridge differences in how others may feel about a common problem. These personal action frames are inclusive of different personal reasons for contesting a situation that needs to be changed. 2. Various personal communication technologies that enable sharing these themes. Whether through texts, tweets, social network sharing, or posting YouTube mashups, the communication process itself often involves further personalization through the spreading of digital connections among friends or trusted others. Some more sophisticated custom coordinating platforms can resemble organizations that exist more online than off. As we followed various world protests, we noticed a dazzling array of personal action frames that spread through social media. Both the acts of sharing these personal calls to action and the social technologies through which they spread help to explain both how events are communicated to external audiences and how the action itself is organized. Indeed, in the limiting case, the communication network becomes the organizational form of the political action (Earl and Kimport, 2011). We explore the range of differently organized forms of contention using personalized communication up to the point at which they enter the part of the range conventionally understood as social movements. This is the boundary zone within which what we refer to as connective action gives way to collective action. The case of PPF occupies an interesting part of this range of contentious action because there were many conventional organizations involved in the mobilization, from churches to social justice NGOs. Yet, visitors to the sophisticated, stand-alone, PPF coordinating platform (which served as an interesting kind of organization in itself) were not asked to pledge allegiance to specific political demands on the organizational agendas of the protest sponsors. Instead, visitors to the organizing site were met with an impressive array of social technologies, enabling them to communicate in their own terms with each other and with various political targets. The centrepiece of the PPF site was a prominent text box under an image of a megaphone that invited the visitor to “Send Your Own Message to the G20”. Many of the messages to the G20 echoed the easy-to-personalize action frame of PPF, and they also revealed a broad range of personal thoughts about the crisis and possible solutions. PPF as a personal action frame was easy to shape and share with friends near and far. It became a powerful example of what students of viral communication refer to as a meme: a symbolic packet that travels easily across large and diverse populations

The logic of connective action  293

because it is easy to imitate, adapt personally, and share broadly with others. Memes are network-building and bridging units of social information transmission similar to genes in the biological sphere (Dawkins, 1989). They travel through personal appropriation, and then by imitation and personalized expression via social sharing in ways that help others to appropriate, imitate, and share in turn (Shifman, 2013). The simple PPF protest meme travelled interpersonally, echoing through newspapers, blogs, Facebook friend networks, Twitter streams, Flickr pages, and other sites on the Internet, leaving traces for years after the events.7 Indeed, part of the meme travelled to Toronto more than a year later where the leading civil society groups gave the name “People First” to their demonstrations. And many people in the large crowds in Seoul in the last G20 meeting of the series could be seen holding up red and white “PPF” signs in both English and Korean (Weller, 2010). Something similar happened in the case of the indignados, where protesters raised banners and chanted “Shhh … the Greeks are sleeping”, with reference to the crushing debt crisis and severe austerity measures facing that country. This idea swiftly travelled to Greece where Facebook networks agreed to set alarm clocks at the same time to wake up and demonstrate. Banners in Athens proclaimed: “We’ve awakened! What time is it? Time for them to leave!” and “Shhh … the Italians are sleeping” and “Shhh … the French are sleeping”. These efforts to send personalized protest themes across national and cultural boundaries met with varying success, making for an important cautionary point: we want to stress that not all personal action frames travel equally well or equally far. The fact that these messages travelled more easily in Spain and Greece than in France or Italy is an interesting example pointing to the need to study failures as well as successes. Just being easy to personalize (for example, I am personally indignant about x, y, and z, and so I join with los indignados) does not ensure successful diffusion. Both political opportunities and conditions for social adoption may differ from situation to situation. For example, the limits in the Italian case may reflect an already established popular anti-government network centred on comedian-activist Beppe Grillo. The French case may involve the ironic efforts of established groups on the left to lead incipient solidarity protests with the indignados, and becoming too heavy-handed in suggesting messages and action programs. Personal action frames do not spread automatically. People must show each other how they can appropriate, shape, and share themes. In this interactive process of personalization and sharing, communication networks may become scaled up and stabilized through the digital technologies people use to share ideas and relationships with others. These technologies and their use patterns often remain in place as organizational mechanisms. In the PPF and the indignados protests, the communication processes themselves represented important forms of organization. In contrast to personal action frames, other calls to action more clearly require joining with established groups or ideologies. These more conventionally understood collective action frames are more likely to stop at the edges of communities, and may require resources beyond communication technologies to bridge the gaps or align different collective frames (Snow and Benford, 1988; Benford and Snow, 2000).

294  Handbook of digital politics

For example, another set of protests in London at the start of the financial crisis was organized by a coalition of more radical groups under the name G20 Meltdown. Instead of mobilizing the expression of large-scale personal concerns, they demanded ending the so-called neoliberal economic policies of the G20, and some even called for the end of capitalism itself. Such demands typically come packaged with more demanding calls to join in particular repertoires of collective action. Whether those repertoires are violent or non-violent, they typically require adoption of shared ideas and behaviors. These anarcho-socialist demonstrations drew on familiar anti-capitalist slogans and calls to “storm the banks” or “eat the rich” while staging dramatic marches behind the four horsemen of the economic apocalypse riding from the gates of old London to the Bank of England. These more radical London events drew smaller turnouts (some 5,000 for the Bank of England march and 2,000 for a climate encampment), higher levels of violence, and generally negative press coverage (Bennett and Segerberg, 2011). While scoring high on commitment in terms of the personal costs of civil disobedience, and displaying unity around anti-capitalist collective action frames, these demonstrations lacked the attributions of public worthiness (for example, recognition from public officials, getting their messages into the news) and the numbers that gave PPF its higher levels of WUNC. Collective action frames that place greater demands on individuals to share common identifications or political claims can also be regarded as memes, in the sense that slogans such as “eat the rich” have rich histories of social transmission. This particular iconic phrase may possibly date to Rousseau’s quip: “When the people shall have nothing more to eat, they will eat the rich”. The crazy course of that meme’s passage down through the ages includes its appearance on T-shirts in the 1960s and in rock songs of that title by Aerosmith and Motorhead, just to scratch the surface of its history of travel through time and space, reflecting the sequence of appropriation, personal expression, and sharing. One distinction between personal action and collective action memes seems to be that the latter require somewhat more elaborate packaging and ritualized action to reintroduce them into new contexts. For example, the organizers of the “storm the banks” events staged an elaborate theatrical ritual with carnivalesque opportunities for creative expression as costumed demonstrators marched behind the Four Horsemen of the financial apocalypse.8 At the same time, the G20 Meltdown discourse was rather closed, requiring adopters to make common cause with others. The Meltdown coalition had an online presence, but they did not offer easy means for participants to express themselves in their own voices (Bennett and Segerberg, 2011). This suggests that more demanding and exclusive collective action frames can also travel as memes, but more often they hit barriers at the intersections of social networks defined by established political organizations, ideologies, interests, class, gender, race, or ethnicity. These barriers often require resources beyond social technologies to overcome. While the idea of memes may help to focus differences in transmission mechanisms involved in more personal versus collective framing of action, we will use the terms “personal action frames” and “collective action frames” as our general concepts. This conceptual pairing locates our work alongside analytical categories used by social

The logic of connective action  295

movement scholars (Snow and Benford, 1988; Benford and Snow, 2000). As should be obvious, the differences we are sketching between personal and collective action frames are not about being online versus offline. All contentious action networks are in important ways embodied and enacted by people on the ground (Juris, 2008; Routledge and Cumbers, 2009). Moreover, most formal political organizations have discovered that the growing sophistication and ubiquity of social media can reduce the resource costs of public outreach and coordination, but these uses of media do not change the action dynamics by altering the fundamental principles of organizing collectivities. By contrast, digital media networking can change the organizational game, given the right interplay of technology, personal action frames, and, when organizations get in the game, their willingness to relax collective identification requirements in favour of personalized social networking among followers. The logic of collective action that typifies the modern social order of hierarchical institutions and membership groups stresses the organizational dilemma of getting individuals to overcome resistance to joining actions where personal participation costs may outweigh marginal gains, particularly when people can ride on the efforts of others for free, and reap the benefits if those others win the day. In short, conventional collective action typically requires people to make more difficult choices and adopt more self-changing social identities than DNA based on personal action frames organized around social technologies. The spread of collective identifications typically requires more education, pressure, or socialization, which in turn makes higher demands on formal organization and resources such as money to pay rent for organization offices, to generate publicity, and to hire professional staff organizers (McAdam et al., 1996).9 Digital media may help to reduce some costs in these processes, but they do not fundamentally change the action dynamics. As noted above, the emerging alternative model that we call the logic of connective action applies increasingly to life in late modern societies in which formal organizations are losing their grip on individuals, and group ties are being replaced by large-scale, fluid social networks (Castells, 2000).10 The organizational processes of social media play an important role in how these networks operate, and their logic does not require strong organizational control or the symbolic construction of a united “we”. The logic of connective action, we suggest, entails a dynamic of its own and thus deserves analysis on its own analytical terms.

TWO LOGICS: COLLECTIVE AND CONNECTIVE ACTION Social movements and contentious politics extend over many different kinds of phenomena and action (Melucci, 1996; McAdam et al., 2001; Tarrow, 2011). The talk about new forms of collective action may reflect ecologies of action that are increasingly complex (Chesters and Welsh, 2006). Multiple organizational forms operating within such ecologies may be hard to categorize, not least because they may morph over time or context, displaying hybridity of various kinds (Chadwick, 2013). In addition, protest and organizational work is occurring both online and off,

296  Handbook of digital politics

using technologies of different capabilities, sometimes making the online/offline distinction relevant, but more often not (Earl and Kimport, 2011; Bimber et al., 2012). Some observers mark a turning point in patterns of contemporary contentious politics, which mix different styles of organization and communication, along with the intersection of different issues with the iconic union of “teamsters and turtles” in the Battle of Seattle in 1999, during which burly union members marched alongside environmental activists wearing turtle costumes in battling a rising neoliberal trade regime that was seen as a threat to democratic control of both national economies and the world environment. Studies of such events show that there are still plenty of old-fashioned meetings, and issue brokering and coalition building, going on (Polletta, 2002). At the same time, however, there is increasing coordination of action by organizations and individuals using digital media to create networks, structure activities, and communicate their views directly to the world. This means that there is also an important degree of technology-enabled networking (Livingston and Asmolov, 2010) that makes highly personalized, socially mediated communication processes fundamental structuring elements in the organization of many forms of connective action. How do we sort out what organizational processes contribute what qualities to collective and connective action networks? How do we identify the borders between fundamentally different types of action formations: that is, what are the differences between collective and connective action, and where are the hybrid overlaps? We propose a starting point for sorting out some of the complexity and overlap in the forms of action by distinguishing between two logics of action. The two logics are associated with distinct dynamics, and thus draw attention to different dimensions for analysis. It is important to separate them analytically as one is less familiar than the other, and this in turn constitutes an important stumbling block for the study of much contemporary political action that we term connective action.11 The more familiar action logic is the logic of collective action, which emphasizes the problems of getting individuals to contribute to the collective endeavour that typically involves seeking some sort of public good (for example, democratic reforms) that may be better attained through forging a common cause. The classical formulation of this problem was articulated by Olson (1965), but the implications of his general logic have reached far beyond the original formulation. Olson’s intriguing observation was that people in fact cannot be expected to act together just because they share a common problem or goal. He held that in large groups in which individual contributions are less noticeable, rational individuals will free-ride on the efforts of others: it is more cost-efficient not to contribute if you can enjoy the good without contributing. Moreover, if not enough people join in creating the good, your efforts are wasted anyway. Either way, it is individually rational not to contribute, even if all agree that all would be better off if everyone did. This thinking fixes attention on the problematic dynamics attending the rational action of atomistic individuals, and at the same time makes resource-rich organizations a central concern. Both the solutions Olson discerned – coercion and selective incentives – implied organizations with substantial capacity to monitor, administer, and distribute such measures.

The logic of connective action  297

In this view, formal organizations with resources are essential to harnessing and coordinating individuals in common action. The early application of this logic to contentious collective action was most straightforwardly exemplified in resource mobilization theory (RMT), in which social movement scholars explicitly adopted Olson’s framing of the collective action problem and its organization-centred solution. Part of a broader wave rejecting the idea of social movements as irrational behavior erupting out of social dysfunction, early RMT scholars accepted the problem of rational free-riders as a fundamental challenge and regarded organizations and their ability to mobilize resources as critical elements of social movement success. Classic formulations came from McCarthy and Zald (1973, 1977) who theorized the rise of external support and resources available to social movement organizations (SMOs), and focused attention on the professionalization of movement organizations and leaders in enabling more resource-intensive mobilization efforts. The contemporary social movement field has moved well beyond the rational choice orientation of such earlier work. Indeed, important traditions developed independently of, or by rejecting, all or parts of the resource mobilization perspective and by proposing that we pay more attention to the role of identity, culture, emotion, social networks, political process, and opportunity structures (Melucci, 1996; McAdam et al., 2001; Della Porta and Diani, 2006). We do not suggest that these later approaches cling to rational choice principles. We do, however, suggest that echoes of the modernist logic of collective action can still be found to play a background role even in work that is in other ways far removed from the rational choice orientation of Olson’s original argument. This comes out in assumptions about the importance of particular forms of organizational coordination and identity in the attention given to organizations, resources, leaders, coalitions, brokering differences, cultural or epistemic communities, the importance of formulating collective action frames, and bridging of differences among those frames. Connective action networks may vary in terms of stability, scale, and coherence, but they are organized by different principles. Connective action networks are typically far more individualized and technologically organized sets of processes, that result in action without the requirement of collective identity framing or the levels of organizational resources required to respond effectively to opportunities. One of the most widely adopted approaches that moved social movement research away from the rational choice roots toward a more expansive collective action logic is the analysis of collective action frames, which centres on the processes of negotiating common interpretations of collective identity linked to the contentious issues at hand (Snow et al., 1986; Snow and Benford, 1988; Hunt et al., 1994; Benford and Snow, 2000). Such framing work may help to mobilize individuals and ultimately lower resource costs by retaining their emotional commitment to action. At the same time, the formulation of ideologically demanding, socially exclusive, or high-conflict collective frames also invites fractures, leading to an analytical focus on how organizations manage or fail to bridge these differences. Resolving these frame conflicts may require the mobilization of resources to bridge differences between groups that have different goals and ways of understanding their issues. Thus, while

298  Handbook of digital politics

the evolution of different strands of social movement theory has moved away from economic collective action models, many still tend to emphasize the importance of organizations that have strong ties to members and followers, and the resulting ways in which collective identities are forged and fractured among coalitions of those organizations and their networks. Sustainable and effective collective action from the perspective of the broader logic of collective action typically requires varying levels of organizational resource mobilization deployed in organizing, leadership, developing common action frames, and brokerage to bridge organizational differences. The opening or closing of political opportunities affects this resource calculus (Tarrow, 2011), but overall, large-scale action networks that reflect this collective action logic tend to be characterized in terms of numbers of distinct groups networking to bring members and affiliated participants into the action and to keep them there. On the individual level, collective action logic emphasizes the role of social network relationships and connections as informal preconditions for more centralized mobilization (for example, in forming and spreading action frames, and forging common identifications and relations of solidarity and trust). At the organizational level, the strategic work of brokering and bridging coalitions between organizations with different standpoints and constituencies becomes the central activity for analysis (see also Diani, forthcoming). Since the dynamics of action in networks characterized by this logic tends not to change significantly with digital media, it primarily invites analysis of how such tools help actors do what they were already doing (see also Bimber et al., 2009; Earl and Kimport, 2011). Movements and action networks characterized by these variations on the logic of collective action are clearly visible in contemporary society. They have been joined by many other mobilizations that may superficially seem like movements, but on closer inspection lack many of the traditional defining characteristics. Efforts to push these kinds of organization into recognizable social movement categories diminish our capacity to understand one of the most interesting developments of our times: how fragmented, individualized populations, that are hard to reach and even harder to induce to share personally transforming collective identities, somehow find ways to mobilize protest networks from Wall Street to Madrid to Cairo. Indeed, when people are individualized in their social orientations, and thus structurally or psychologically unavailable to modernist forms of political movement organization, resource mobilization becomes increasingly costly and has diminishing returns. Organizing such populations to overcome free-riding and helping them to shape identities in common is not necessarily the most successful or effective logic for organizing collective action. When people who seek more personalized paths to concerted action are familiar with practices of social networking in everyday life, and when they have access to technologies from mobile phones to computers, they are already familiar with a different logic of organization: the logic of connective action. The logic of connective action foregrounds a different set of dynamics from the ones just outlined. At the core of this logic is the recognition of digital media as organizing agents. Several collective action scholars have explored how digital com-

The logic of connective action  299

munication technology alters the parameters of Olson’s original theory of collective action. Lupia and Sin (2003) show how Olson’s core assumption about weak individual commitment in large groups (free-riding) may play out differently under conditions of radically reduced communication costs. Bimber et al. (2005) in turn argue that public goods themselves may take on new theoretical definition as erstwhile free-riders find it easier to become participants in political networks that diminish the boundaries between public and private; boundaries that are blurred in part by the simultaneous public–private boundary crossing of ubiquitous social media. Important for our purposes here is the underlying economic logic of digitally mediated social networks, as explained most fully by Benkler (2006). He proposes that participation becomes self-motivating as personally expressive content is shared with, and recognized by, others who in turn repeat these networked sharing activities. When these interpersonal networks are enabled by technology platforms of various designs that coordinate and scale the networks, the resulting actions can resemble collective action, yet without the same role played by formal organizations or transforming social identifications. In place of content that is distributed and relationships that are brokered by hierarchical organizations, social networking involves co-production and co-distribution, revealing a different economic and psychological logic: co-production and sharing based on personalized expression. This does not mean that all online communication works this way. Looking at most online newspapers, blogs, or political campaign sites makes it clear that the logic of the organization-centred brick-and-mortar world is often reproduced online, with little change in organizational logic beyond possible efficiency gains (Bimber and Davis, 2003; Foot and Schneider, 2006). Yet, many socially mediated networks do operate with an alternative logic that also helps to explain why people labour collectively for free to create such things as open source software, Wikipedia, WikiLeaks, and the free and open source software that powers many protest networks (Calderaro, 2011). In this connective logic, taking public action or contributing to a common good becomes an act of personal expression and recognition or self-validation achieved by sharing ideas and actions in trusted relationships. Sometimes the people in these exchanges may be on the other side of the world, but they do not require a club, a party, or a shared ideological frame to make the connection. In place of the initial collective action problem of getting the individual to contribute, the starting point of connective action is the self-motivated (though not necessarily self-centred) sharing of already internalized or personalized ideas, plans, images, and resources with networks of others. This “sharing” may take place in networking sites such as Facebook, or via more public media such as Twitter and YouTube through, for example, comments and re-tweets.12 Action networks characterized by this logic may scale up rapidly through the combination of easily spreadable personal action frames and digital technology enabling such communication. This invites analytical attention to the network as an organizational structure in itself. Technology-enabled networks of personalized communication involve more than just exchanging information or messages. The flexible, recombinant nature of DNA makes these web spheres and their offline extensions more than just communication

300  Handbook of digital politics

systems. Such networks are flexible organizations in themselves, often enabling coordinated adjustments and rapid action aimed at often shifting political targets, even crossing geographic and temporal boundaries in the process. As Diani (forthcoming) argues, networks are not just precursors or building blocks of collective action: they are in themselves organizational structures that can transcend the elemental units of organizations and individuals.13 As noted earlier, communication technologies do not change the action dynamics in large-scale networks characterized by the logic of collective action. In the n­ etworks characterized by connective action, they do. The organizational structure of people and social technology emerges more clearly if we draw on the actor-network theory of Latour (2005) in recognizing digital networking mechanisms (for example, various social media and devices that run them) as potential network agents alongside human actors (that is, individuals and organizations). Such digital mechanisms may include organizational connectors (for example, web links), event coordination (for example, protest calendars), information sharing (for example, YouTube and Facebook), and multifunction networking platforms in which other networks become embedded (for example, links in Twitter and Facebook posts), along with various capacities of the devices that run them. These technologies not only create online meeting places and coordinate offline activities, but they also help to calibrate relationships by establishing levels of transparency, privacy, security, and interpersonal trust. It is also important that these digital traces may remain behind on the web to provide memory records or action repertoires that might be passed on via different mechanisms associated with more conventional collective action such as rituals or formal documentation. The simple point here is that collective and connective logics are distinct logics of action (in terms of both identity and choice processes), and thus both deserve analysis on their own terms. Just as traditional collective action efforts can fail to result in sustained or effective movements, there is nothing preordained about the results of digitally mediated networking processes. More often than not, they fail badly. The transmission of personal expression across networks may or may not become scaled up, stable, or capable of various kinds of targeted action depending on the kinds of social technology designed and appropriated by participants, and the kinds of opportunities that may motivate anger or compassion across large numbers of individuals. Thus, the Occupy Wall Street protests that spread in a month from New York to more than 80 countries and 900 cities around the world might not have succeeded without the inspiring models of the Arab Spring or the indignados in Spain, or the worsening economic conditions that provoked anger among increasing numbers of displaced individuals. Yet, when the Occupy networks spread under the easy-to-personalize action frame of “We are the 99%”, there were few identifiable established political organizations at the centre of them. There was even a conscious effort to avoid designating leaders and official spokespeople. The most obvious organizational forms were the layers of social technologies and websites that carried news reported by participants and displayed tools for personalized networking. One of the sites was “15.10.11 united for #global change”.14 Instead of the usual “Who are we?” section of the website, #globalchange asked: “Who are you?”.

The logic of connective action  301

Collective and connective action may co-occur in various formations within the same ecology of action. It is nonetheless possible to discern three clear ideal types of large-scale action networks. While one is primarily characterized by collective action logic, the other two are connective action networks distinguished by the role of formal organizations in facilitating personalized engagement. As noted above, conventional organizations play a less central role than social technologies in relatively crowd-enabled networks such as the indignados of Spain, the Arab Spring uprisings, or the Occupy protests that spread from Wall Street around the world. In contrast to these more technology-enabled networks, we have also observed hybrid networks (such as PPF) where conventional organizations operate in the background of protest and issue advocacy networks to enable personalized engagement. This hybrid form of organizationally enabled connective action sits along a continuum somewhere between the two ideal types of conventional organizationally brokered collective action and relatively more crowd-enabled connective action. The following section presents the details of this three-part typology. It also suggests that co-existence, layering, and movement across the types becomes an important part of the story.

A TYPOLOGY OF COLLECTIVE AND CONNECTIVE ACTION NETWORKS We draw upon these distinct logics of action (and the hybrid form that reveals a tension between them) to develop a three-part typology of large-scale action networks that feature prominently in contemporary contentious politics. One type represents the brokered organizational networks characterized by the logic of collective action, while the others represent two significant variations on networks primarily characterized by the logic of connective action. All three models may explain differences between and dynamics within large-scale action networks in event-centred contention, such as protests and sequences of protests as in the examples we have already discussed. They may also apply to more stable issue advocacy networks that engage people in everyday life practices supporting causes outside of protest events, such as campaigns. The typology is intended as a broad generalization to help understand different dynamics. None of the types are exhaustive social movement models. Thus, this is not an attempt to capture, much less resolve, the many differences among those who study social movements. We simply want to highlight the rise of two forms of digitally networked connective action that differ from some common assumptions about collective action in social movements and, in particular, that rely on mediated networks for substantial aspects of their organization. Figure 19.1 presents an overview of the two connective action network types and contrasts their organizational properties with more familiar collective action network organizational characteristics. The ideal collective action type at the right side in the figure describes large-scale action networks that depend on brokering organizations to carry the burden of facilitating cooperation and bridging differences when possible. As the anti-capitalist direct action groups in the G20 London summit protests

302  Handbook of digital politics

Figure 19.1

Elements of collective and connective action networks

exemplified, such organizations will tend to promote more exclusive collective action frames that require frame bridging if they are to grow. They may use digital media and social technologies more as means of mobilizing and managing participation and coordinating goals, rather than inviting personalized interpretations of problems and self-organization of action. In addition to a number of classic social movement accounts (for example, McAdam, 1986), several of the NGO networks discussed by Keck and Sikkink (1998) also accord with this category (Bennett, 2005). At the other extreme, on the left side in the figure we place connective action networks that self-organize largely without central or lead organizational actors, using technologies as important organizational agents. We call this type crowd-enabled connective action. While some formal organizations may be present, they tend to remain at the periphery or may exist as much in online as in offline forms. In place of collective action frames, personal action frames become the transmission units across trusted social networks. The loose coordination of the indignados exemplifies this ideal type, with conventional organizations deliberately kept at the periphery as easily adapted personal action frames travel online and offline with the aid of technology platforms such as the Democracia real Ya! organization.15 In between the organizationally-brokered collective action networks and the crowd-enabled connective action network is the hybrid pattern introduced above. This middle type involves formal organizational actors stepping back from projecting strong agendas, political brands, and collective identities in favour of using resources to deploy social technologies enabling loose public networks to form around personalized action themes. The middle type may also encompass more informal organizational actors that develop some capacities of conventional organizations in terms of resource mobilization and coalition building without imposing strong brands and collective identities.16 For example, many of the general assemblies in the Occupy

The logic of connective action  303

protests became resource centres, with regular attendance, division of labour, allocation of money and food, and coordination of actions. At the same time, the larger communication networks that swirled around these protest nodes greatly expanded the impact of the network. The surrounding technology networks invited loose-tied participation that was often in tension with the face-to-face ethos of the assemblies, where more committed protesters spent long hours with dwindling numbers of peers debating on how to expand participation without diluting the levels of commitment and action that they deemed key to their value scheme. Thus, even as Occupy displayed some organizational development, it was defined by its self-organizing roots. Networks in this hybrid model engage individuals in causes that might not be of such interest if stronger demands for membership or subscribing to collective demands accompanied the organizational offerings. Organizations facilitating these action networks typically deploy an array of custom-built (for example, “send your message”) and outsourced (for example, Twitter) communication technologies. This pattern fits the PPF demonstrations discussed earlier, where some 160 civil society organizations – including major NGOs such as Oxfam, Tearfund, Catholic Relief, and World Wildlife Fund – stepped back from their organizational brands to form a loose social network inviting publics to engage with each other and take action. They did this even as they negotiated with other organizations over such things as separate days for the protests (Bennett and Segerberg, 2011). The formations in the middle type reflect the pressures that Bimber et al. (2005) observed in interest organizations that are suffering declining memberships and have had to develop looser, more entrepreneurial relations with followers. Beyond the ways in which particular organizations use social technologies to develop loose ties with followers, many organizations also develop loose ties with other organizations to form vast online networks sharing and bridging various causes. Although the scale and complexity of these networks differ from the focus of Granovetter’s (1973) observations about the strength of weak ties in social networks, we associate this idea with the elements of connective action: the loose organizational linkages, technology deployments, and personal action frames. In observing the hybrid pattern of issue advocacy organizations facilitating personalized protest networks, we traced a number of economic justice and environmental networks, charting protests, campaigns, and issue networks in the UK, Germany, and Sweden (Bennett and Segerberg, 2013).17 In each case, we found (with theoretically interesting variations) campaigns, protest events, and everyday issue advocacy networks that displayed similar organizational signatures: (1) familiar NGOs and other civil society organizations joining loosely together to provide something of a networking backbone; (2) for digital media networks engaging publics with contested political issues; yet with (3) remarkably few efforts to brand the issues around specific organizations, own the messages, or control the understandings of individual participants. The organizations had their political agendas on offer, to be sure, but as members of issue networks, put the public face on the individual citizen and provided social technologies to enable personal engagement through easy-to-share images and personal action frames.

304  Handbook of digital politics

The organizations that refrain from strongly branding their causes or policy agendas in this hybrid model do not necessarily give up their missions or agendas as name-brand public advocacy organizations. Instead, some organizations interested in mobilizing large and potentially “WUNC-y” publics in an age of social networking are learning to shift among different organizational repertoires, morphing from being hierarchical, mission-driven NGOs in some settings to being facilitators in loosely linked public engagement networks in others. As noted by Chadwick (2007, 2013), organizational hybridity makes it difficult to apply fixed categories to many organizations as they variously shift from being issue advocacy NGOs to policy think tanks, to SMOs running campaigns or protests, to multi-issue organizations, to being networking hubs for connective action. In other words, depending on when, where, and how one observes an organization, it may appear differently as an NGO, SMO, INGO, TNGO, NGDO (non-governmental organization, social movement organization, international non-governmental organization, transnational non-governmental organization, non-governmental development organization), an interest advocacy group, a political networking hub, and so on. Indeed, one of the advantages of seeing the different logics at play in our typology is to move away from fixed categorization schemes, and observe actually occurring combinations of different types of action within complex protest ecologies, and shifts in dominant types in response to events and opportunities over time. The real world is of course far messier than this three-type model. In some cases, we see action formations corresponding to our three models side by side in the same action space. The G20 London protest offered a rare case in which organizationally enabled and more conventional collective action were neatly separated over different days. More often, the different forms layer and overlap, perhaps with violence disrupting otherwise peaceful mobilizations as occurred in the Occupy Rome protests on October 15, 2011, and in a number of Occupy clashes with police in the United States. In still other action cycles, we see a movement from one model to another over time. In some relatively distributed networks, we observe a pattern of informal organizational resource-seeking, in which informal organizational resources and communication spaces are linked and shared (for example, re-tweeted), enabling emergent political concerns and goals to be nurtured without being co-opted by existing organizations and their already fixed political agendas. This pattern occurred in the crowd-enabled Twitter network that emerged around the 15th UN Climate Change Conference in Copenhagen. As the long tail of that network handed its participants off to the Twitter stream devoted to the next summit in Cancun, we saw an increase in links to organizations of various kinds, along with growing links to and among climate bloggers (Segerberg and Bennett, 2011). Such variations on different organizational forms offer intriguing opportunities for further analyses aimed at explaining whether mobilizations achieve various goals, and attain different levels of WUNC. In these varying ways, personalized connective action networks cross paths (sometimes with individual organizations morphing in the process) with more conventional collective action networks centred on SMOs, interest organizations,

The logic of connective action  305

and brand-conscious NGOs. As a result, while we argue that these networks are an organizational form in themselves, they are often hard to grasp and harder to analyse because they do not behave like formal organizations. Most formal organizations are centred (for example, located in physical space), hierarchical, bounded by mission and territory, and defined by relatively known and countable memberships (or in the case of political parties, known and reachable demographics). By contrast, many of today’s issue and cause networks are relatively decentred (constituted by multiple organizations and many direct and cyber activists), distributed, or flattened organizationally as a result of these multiple centres, relatively unbounded, in the sense of crossing both geographical and issue borders, and dynamic in terms of the changing populations who may opt in and out of play as different engagement opportunities are presented (Bennett, 2003, 2005). Understanding how connective action engages or fails to engage diverse populations constitutes part of the analytical challenge ahead. Compared to the vast number of theoretically grounded studies on social movement organizing, there is less theoretical work that helps to explain the range of collective action formations, running from relatively crowd-enabled to organizationally enabled connective action networks. While there are many descriptive and suggestive accounts of this kind of action, many of them insightful (for example, Castells, 2000; Rheingold, 2002), we are concerned that the organizational logic and underlying dynamic of such action is not well established. It is important to gain clearer understandings of how such networks function and what organizing principles explain their growing prominence in contentious politics.

CONCLUSION DNA is emerging during a historic shift in late modern democracies in which, most notably, younger citizens are moving away from parties, broad reform movements, and ideologies. Individuals are relating differently to organized politics, and many organizations are finding that they must engage people differently: they are developing relationships to publics as affiliates rather than members, and offering them personal options in ways to engage and express themselves. This includes greater choice over contributing content, and introduces micro-organizational resources in terms of personal networks, content creation, and technology development skills. Collective action based on exclusive collective identifications and strongly tied networks continues to play a role in this political landscape, but this has become joined by, interspersed with, and in some cases supplanted by personalized collective action formations in which digital media become integral organizational parts. Some of the resulting DNA networks turn out to be surprisingly nimble, demonstrating intriguing flexibility across various conditions, issues, and scales. It has been tempting for some critics to dismiss participation in such networks as noise, particularly in reaction to sweeping proclamations by enthusiasts of the democratic and participatory power of digital media. Whether from digital enthusiasts or critics, hyperbole is unhelpful. Understanding the democratic potential and effec-

306  Handbook of digital politics

tiveness of instances of connective and collective action requires careful analysis. At the same time, there is often considerably more going on in DNA than clicktivism or facile organizational outsourcing of social networking to various commercial sites.18 The key point of our argument is that fully explaining and understanding such action and contention requires more than just adjusting the classic social movement collective action schemes. Connective action has a logic of its own, and thus attendant dynamics of its own. It deserves analysis on its own terms. The linchpin of connective action is the formative element of “sharing”: the personalization that leads actions and content to be distributed widely across social networks. Communication technologies enable the growth and stabilization of network structures across these networks. Together, the technological agents that enable the constitutive role of sharing in these contexts displace the centrality of the free-rider calculus and with it, by extension, the dynamic that flows from it; most obviously, the logical centrality of the resource-rich organization. In its stead, connective action brings the action dynamics of recombinant networks into focus, a situation in which networks and communication become something more than mere preconditions and information. What we observe in these networks are applications of communication technologies that contribute an organizational principle that is different from notions of collective action based on core assumptions about the role of resources, networks, and collective identity. We call this different structuring principle the logic of connective action. Developing ways to analyse connective action formations will give us more solid grounds for returning to the persistent questions of whether such action can be politically effective and sustained (Tilly, 2004; Gladwell, 2010; Morozov, 2011). Even as the contours of political action may be shifting, it is imperative to develop means of thinking meaningfully about the capacities of sustainability and effectiveness in relation to connective action and to gain a systematic understanding of how such action plays out in different contexts and conditions. The string of G20 protests surrounding the world financial crisis illustrate that different organizational strategies played out in different political settings produce a wide range of results. The protests at the Pittsburgh and Toronto G20 summits of 2009 and 2010, respectively, were far more chaotic and displayed far less WUNC than those organized under the banner of PPF in London. Disrupted by police assaults and weak organizational coordination, the Pittsburgh protests displayed a cacophony of political messages that were poorly translated in the press and even became the butt of late-night comedy routines. The Daily Show sent a correspondent to Pittsburgh and reported on a spectrum of messages that included: a Free Tibet marching cymbal band; Palestinian peace advocates; placards condemning genocide in Darfur; hemp and marijuana awareness slogans; and denunciations of the beef industry; along with the more expected condemnations of globalization and capitalism. One protester carried a sign saying “I protest everything”, and another dressed as Batman stated that he was protesting the choice of Christian Bale to portray his movie hero. The correspondent concluded that the Pittsburgh protests lacked unity of focus, and turned for advice to some people who knew how to get the job done: members of the

The logic of connective action  307

Tea Party. The Daily Show panel of Tea Party experts included a woman wearing a black Smith & Wesson holster that contained a wooden crucifix with an American flag attached. When asked what the Pittsburgh protesters were doing wrong, they all agreed that there was a message problem. One said, “I still don’t know what their message is”, and another affirmed, “Stay on message and believe what you say”. The Daily Show report cut back to show a phalanx of Darth Vader-suited riot police lined up against the protesters; according to the correspondent, the “one single understandable talking point” in Pittsburgh (Daily Show, 2009). Humor aside, this example poses a sharp contrast to the more orderly London PPF protests that received positive press coverage of the main themes of economic and environmental justice (Bennett and Segerberg, 2011). The challenge ahead is to understand when DNA becomes chaotic and unproductive, and when it attains higher levels of focus and sustained engagement over time. Our studies suggest that differing political capacities in networks depend, among other things, on whether: (1) in the case of organizationally enabled DNA, the network has a stable core of organizations sharing communication linkages and deploying high volumes of personal engagement mechanisms; or (2) in the case of crowd-enabled DNA, the digital networks are redundant and dense with pathways for individual networks to converge, enabling viral transmission of personally appealing action frames to occur. Attention to connective action will neither explain all contentious politics nor replace the model of classic collective action that remains useful for analyzing social movements. But it does shed light on an important mode of action making its mark in contentious politics today. A model focused primarily on the dynamics of classic collective action has difficulties accounting for important elements in the Arab spring, the indignados, the Occupy demonstrations, or the global protests against climate change. A better understanding of connective action promises to fill some of these gaps. Such understanding is essential if we are to attain a critical perspective on some of the prominent forms of public engagement in the digital age.

NOTES 1. The original version of this chapter was published as: W. Lance Bennett and Alexandra Segerberg (2012), The logic of connective action: Digital media and the personalization of contentious politics. Information, Communication & Society, 15(5), 739–768. The authors are grateful for permission from Taylor & Francis (http://​www​.tandfonline​.com) to reprint the article as this chapter. This version has been updated to reflect changes that appear in The Logic of Connective Action (Bennett and Segerberg, 2013). 2. Simultaneous protests were held in other European cities with tens of thousands of demonstrators gathering in the streets of Berlin, Frankfurt, Vienna, Paris, and Rome. 3. US Vice President Joe Biden asked for patience from understandably upset citizens while leaders worked on solutions, and the British Prime Minister at the time, Gordon Brown, said: “the action we want to take (at the G20) is designed to answer the questions that the protesters have today” (Vinocur and Barkin, 2009). 4. See http://​www​.democraciarealya​.es/​.

308  Handbook of digital politics

5. Beyond the high volume of Spanish press coverage, the story of the indignados attracted world attention. BBC World News devoted no fewer than eight stories to this movement over the course of two months, including a feature on the march of one group across the country to Madrid, with many interviews and encounters in the words of the protesters themselves. 6. For example, our analyses of the US Occupy protests show that increased media attention to economic inequality in the USA was associated with the coverage of the Occupy protests (Bennett and Segerberg, 2013). While political elites were often reluctant to credit the occupiers with their new-found concern about inequality, they nonetheless seemed to find the public opinion and media climate conducive to addressing the long-neglected issue. 7. A Google search of “put people first g20” more than two years after the London events produced nearly 1.5 million hits, with most of them relevant to the events and issues of the protests well into 75 search pages deep. 8. We would note, however, that carnivalesque or theatrical expressions may entail strategically depersonalized forms of expression in which individuals take on other personae that often have historically or dramatically scripted qualities. We thank Stefania Milan for this comment. 9. We are not arguing here that all contemporary analyses of collective action rely on resource mobilization explanations (although some do). Our point is that whether resource assumptions are in the foreground or the background, many collective action analyses typically rely on a set of defining assumptions centered on the importance of some degree of formal organization and some degree of strong collective identity that establishes common bonds among participants. These elements become more marginal in thinking about the organization of connective action. 10. While we focus primarily on cases in late modern, post-industrial democracies, we also attempt to develop theoretical propositions that may apply to other settings such as the Arab Spring, where authoritarian rule may also result in individualized populations that fall outside of sanctioned civil society organization, yet may have direct or indirect access to communication technologies such as mobile phones. 11. Routledge and Cumbers (2009) make a similar point in discussing horizontal and vertical models as useful heuristics for organizational logics in global justice networks (see also Robinson and Tormey, 2005; Juris, 2008). 12. We are indebted to Bob Boynton for pointing out that this sharing occurs both in trusted friends networks such as Facebook and in more public exchange opportunities among strangers of the sort that occur on YouTube, Twitter, or blogs. Understanding the dynamics and interrelationships among these different media networks and their intersections is an important direction for research. 13. We have developed methods for mapping networks and inventorying the types of digital media that enable actions and information to flow through them. Showing how networks are constituted in part by technology enables us to move across levels of action that are often difficult to theorize. Network technologies enable thinking about individuals, organizations, and networks in one broad framework. This approach thus revises the starting points of classic collective action models, which typically examine the relationships between individuals and organizations and between organizations. We expand this to include technologies that enable the formation of fluid action networks in which agency becomes shared or distributed across individual actors and organizations as networks reconfigure in response to changing issues and events (Bennett and Segerberg, 2013; Bennett et al., 2014). 14. See http://​www​.15october​.net (accessed 19 October 2011). 15. We wish to emphasize that there is much face-to-face organizing work going on in many of these networks, and that the daily agendas and decisions are importantly shaped

The logic of connective action  309

offline. However, the connectivity and flow of action coordination occurs, importantly, online. 16. We thank an anonymous referee for highlighting this subtype. 17. Our empirical investigations focused primarily on two types of networks that display local, national, and transnational reach: networks to promote economic justice via more equitable North–South trade norms (fair trade) and networks for environmental and human protection from the effects of global warming (climate change). These networks display impressive levels of collective action and citizen engagement and they are likely to remain active into the foreseeable future. They often intersect by sharing campaigns in local, national, and transnational arenas. As such, these issue networks represent good cases for assessing the uses of digital technologies and different action frames (from personalized to collective) to engage and mobilize citizens, and to examine various related capacities and effects of those engagement efforts. 18. Technology is not neutral. The question of the degree to which various collectivities have both appropriated and become dependent on the limitations of commercial technology platforms such as Flickr, Facebook, Twitter, or YouTube is a matter of considerable importance. For now, suffice it to note that at least some of the technologies and their networking capabilities are designed by activists for creating political networks and organizing action (Calderaro, 2011).

REFERENCES Anduiza, E., Cristancho, C., and Sabucedo, J. M. (2014). Mobilization through online social networks: The political protest of the indignados in Spain. Information, Communication & Society, 17(6), 750–764. Bauman, Z. (2000). Liquid Modernity. Cambridge: Polity Press. Beck, U. and Beck-Gernsheim, E. (2002). Individualization: Institutionalized Individualism and its Social and Political Consequences. London: Sage. Benford, R. D. and Snow, D. A. (2000). Framing processes and social movements: An overview and an assessment. Annual Review of Sociology, 26, 611–639. Benkler, Y. (2006). The Wealth of Networks: How Social Production Transforms Markets and Freedom. New Haven, CT: Yale University Press. Bennett, W. L. (1998). The uncivic culture: Communication, identity, and the rise of lifestyle politics. Ithiel de Sola Pool Lecture, American Political Science Association, PS: Political Science and Politics, 31, 41–61. Bennett, W. L. (2003). Communicating global activism: Strengths and vulnerabilities of networked politics. Information, Communication & Society, 6(2), 143–168. Bennett, W. L. (2005). Social movements beyond borders: Organization, communication, and political capacity in two eras of transnational activism. In D. Della Porta and S. Tarrow (eds.), Transnational Protest and Global Activism (pp. 203–222). Lanham, MD: Rowman & Littlefield. Bennett, W. L., Lang, S., and Segerberg, A. (2014). European issue publics online: The cases of climate change and fair trade. In T. Risse (ed.), European Public Spheres: Politics Is Back (pp. 108–137). Cambridge: Cambridge University Press. Bennett, W. L. and Segerberg, A. (2011). Digital media and the personalization of collective action: Social technology and the organization of protests against the global economic crisis. Information, Communication & Society, 14, 770–799. Bennett, W. L. and Segerberg, A. (2013). The Logic of Connective Action: Digital Media and the Personalization of Contentious Politics. New York: Cambridge University Press. Bimber, B. and Davis, R. (2003). Campaigning Online: The Internet in US Elections. New York: Oxford University Press.

310  Handbook of digital politics

Bimber, B., Flanagin, A., and Stohl, C. (2005). Reconceptualizing collective action in the contemporary media environment. Communication Theory, 15, 389–413. Bimber, B., Flanagin, A., and Stohl, C. (2012). Collective Action in Organizations: Interaction and Engagement in an Era of Technological Change. New York: Cambridge University Press. Bimber, B., Stohl, C. and Flanagin, A. (2009). Technological change and the shifting nature of political organization. In A. Chadwick and P. Howard (eds.), Routledge Handbook of Internet Politics (pp. 72–85). London: Routledge. Calderaro, A. (2011). New political struggles in the network society: The case of free and open source software (FOSS) movement. Paper presented at ECPR General Conference, Reykjavik, August 25–27. Castells, M. (2000). The Network Society, 2nd edition. Oxford: Blackwell. Chadwick, A. (2007). Digital network repertoires and organizational hybridity. Political Communication, 24(3), 283–301. Chadwick, A. (2013). The Hybrid Media System: Politics and Power. New York: Oxford University Press. Chesters, G. and Welsh, I. (2006). Complexity and Social Movements: Multitudes at the End of Chaos. London: Routledge. Daily Show (2009). Tea partiers advise G20 protesters. Daily Show, October 1. http://​www​ .thedailyshow​.com/​watch/​thu​-october​-1​-2009/​tea​-partiers​-advise​-g20​-protesters (accessed October 6, 2010). Dawkins, R. (1989). The Selfish Gene. Oxford: Oxford University Press. Della Porta, D. (2005). Multiple belongings, flexible identities and the construction of “another politics”: Between the European social forum and the local social fora. In D. Della Porta and S. Tarrow (eds.), Transnational Protest and Global Activism (pp. 175–202). Lanham, MD: Rowman and Littlefield. Della Porta, D. and Diani, M. (2006). Social Movements: An Introduction, 2nd edition. Malden, MA: Blackwell. Diani, M. (forthcoming). The Cement of Civil Society: Civic Networks in Localities. Cambridge: Cambridge University Press. Earl, J. and Kimport, K. (2011). Digitally Enabled Social Change: Online and Offline Activism in the Age of the Internet. Cambridge, MA: MIT Press. Foot, K. and Schneider, S. (2006). Web Campaigning. Cambridge, MA: MIT Press. Giddens, A. (1991). Modernity and Self-Identity: Self and Society in the Late Modern Age. Stanford, CA: Stanford University Press. Gitlin, T. (1980). The Whole World is Watching: Mass Media in the Making and Unmaking of the New Left. Berkeley, CA: University of California Press. Gladwell, M. (2010). Small change: Why the revolution will not be tweeted. New Yorker, October 4. Granovetter, M. (1973). The strength of weak ties. American Journal of Sociology, 78, 1360–1380. Howard, P. and Hussain, M. (2011). The role of digital media. Journal of Democracy, 22(3), 35–48. Hunt, S., Benford, R. D., and Snow, D. A. (1994). Identity fields: Framing processes and the social construction of movement identities. In E. Laraña, H. Johnston, and J. R. Gusfield (eds.), New Social Movements: From Ideology to Identity (pp. 185–208). Philadelphia, PA: Temple University Press. Inglehart, R. (1997). Modernization and Post-Modernization: Cultural, Economic and Political Change in 43 Societies. Princeton, NJ: Princeton University Press. Juris, J. (2008). Networking Futures: The Movements against Corporate Globalization. Durham, NC: Duke University Press.

The logic of connective action  311

Keck, M. and Sikkink, K. (1998). Activists beyond Borders: Advocacy Networks in International Politics. Ithaca, NY: Cornell University Press. Latour, B. (2005). Reassembling the Social: An Introduction to Actor-Network-Theory. Oxford: Oxford University Press. Livingston, S. and Asmolov, G. (2010). Networks and the future of foreign affairs reporting. Journalism Studies, 11(5), 745–760. Lupia, A. and Sin, G. (2003). Which public goods are endangered? How evolving communication technologies affect ‘the logic of collective action’. Public Choice, 117, 315–331. McAdam, D. (1986). Recruitment to high-risk activism: The case of freedom summer. American Journal of Sociology, 92, 64–90. McAdam, D., McCarthy, J. D., and Zald, M. N. (1996). Opportunities, mobilizing structures, and framing processes: Toward a synthetic, comparative perspective on social movements. In D. McAdam, J. D. McCarthy, and M. N. Zald (eds.), Comparative Perspectives on Social Movements: Political Opportunities, Mobilizing Structures, and Cultural Framings (pp. 1–20). New York: Cambridge University Press. McAdam, D., Tarrow, S., and Tilly, C. (2001). Dynamics of Contention. New York: Cambridge University Press. McCarthy, J. D. and Zald, M. N. (1973). The Trend of Social Movements in America: Professionalization and Resource Mobilization. Morristown, NJ: General Learning Press. McCarthy, J. D. and Zald, M. N. (1977). Resource mobilization and social movements: A partial theory. American Journal of Sociology, 82(6), 1212–1241. McDonald, K. (2002). From solidarity to fluidarity: Social movements beyond ‘collective identity’ – the case of globalization conflicts. Social Movement Studies, 1(2), 109–128. Melucci, A. (1996). Challenging Codes: Collective Action in the Information Age. Cambridge: Cambridge University Press. Micheletti, M. (2003). Political Virtue and Shopping. New York: Palgrave Macmillan. Morozov, E. (2011). The Net Delusion: How Not to Liberate the World. London: Allen Lane. Olson, M. (1965). The Logic of Collective Action: Public Goods and the Theory of Groups. Cambridge, MA: Harvard University Press. Polletta, F. (2002). Freedom is an Endless Meeting: Democracy in American Social Movements. Chicago, IL: University of Chicago Press. Put People First (2009). http://​www​.putpeoplefirst​.org​.uk/​(accessed July 6, 2011). Putnam, R. (2000). Bowling Alone: The Collapse and Revival of American Community. New York: Simon & Schuster. Rheingold, H. (2002). Smart Mobs: The Next Social Revolution. Cambridge, MA: Perseus Pub. Robinson, A. and Tormey, S. (2005). Horizontals, verticals and the conflicting logics of transformative politics. In C. el-Ojeili and P. Hayden (eds.), Confronting Globalization (pp. 208–226). Basingstoke: Palgrave Macmillan. Routledge, P. and Cumbers, A. (2009). Global Justice Networks: Geographies of Transnational Solidarity. Manchester: Manchester University Press. RTVE (2011). Mas de seis millones de Espanoles han participado en el movimiento 15M. August 6. http://​www​.rtve​.es/​noticias/​20110806/​mas​-seis​-millones​-espanoles​-han​ -participado​-movimiento​-15m/​452598​.shtml (accessed September 18, 2011). Segerberg, A. and Bennett, W. L. (2011). Social media and the organization of collective action: Using Twitter to explore the ecologies of two climate change protests. Communication Review, 14(3), 197–215. Shifman, L. (2013). Memes in Digital Culture. Cambridge, MA: MIT Press. Snow, D. A. and Benford, R. D. (1988). Ideology, frame resonance, and participant mobilization. International Social Movement Research, 1, 197–217.

312  Handbook of digital politics

Snow, D. A., Rochford, B. Jr., Worden, S. K., and Benford, R. D. (1986). Frame alignment processes, micromobilization, and movement participation. American Sociological Review, 51, 464–481. Tarrow, S. (2011). Power in Movement: Social Movements in Contentious Politics, 3rd edition. New York: Cambridge University Press. Tilly, C. (2004). Social Movements, 1768–2004. Boulder, CO: Paradigm. Tilly, C. (2006). WUNC. In J. T. Schnapp and M. Tiews (eds.), Crowds (pp. 289–306). Stanford, CA: Stanford University Press. Vinocur, N. and Barkin, N. (2009). G20 marches begin week of protests in Europe. Reuters, March 28. http://​www​.reuters​.com/​article/​2009/​03/​28/​us​-g20​-britain​-march​ -i​dUSTRE52R0TP20090328 (accessed July 9, 2011). Weller, B. (2010). G20 protests in Seoul. Demotix. http://​www​.demotix​.com/​photo/​504262/​ g20​-protests​-seoul (accessed July 9, 2011).

20. Media ecologies, social movements and activism Emiliano Treré

THE RELEVANCE OF MEDIA ECOLOGY THEORY TO STUDY SOCIAL MOVEMENTS AND ACTIVISM In the last decade, scholars interested in digital activism and social movements have started to rely on the media ecology framework to make sense of the complex interactions between activists and media technologies. The power of this conceptual gaze lies in its ability to recognize, explore and disentangle the communicative complexities of activist practises. This framework has allowed a variety of scholars to understand how media and movements mutually shape each other in complex, creative and often unpredictable ways. In this chapter, I will shed light on this emerging area of study, reflecting on its major strengths and limitations, outlining my view on the topic, and zooming in on key priorities for future inquiry. It is important to clarify that the interest in adopting the media ecology lens comes from heterogeneous disciplines and often intersecting areas of study: political science and sociology, media studies and journalism, citizen media and digital activism. Furthermore, this nascent area of study increasingly includes disciplines as varied as Science and Technology Studies (STS), memory studies, data and algorithm studies, arts, humanities, design studies and environmental architecture. In the next sections, I will use the MovAct acronym to indicate the convergence of social movement and activism studies. While there are many overlaps between these two areas, activism studies also include more individual manifestations of activism that are not strictly related to social movements and more organized forms of collective action. The MovAct field has engaged with media ecology theory in several ways that will be addressed in the following lines.

LEVELS OF ENGAGEMENT AND VARIETY OF MEDIA ECOLOGY THEORIES IN MOVACT STUDIES The media ecologies that are addressed in MovAct accounts are often not clearly defined in their constitutive elements, nor it is always outlined which specific ecological tradition (or traditions in some cases) is being used. In general, there are three levels of engagement with media ecology theory in study on movements and activism (Treré, 2019, 2020). At the first level, we find authors who only ‘evoke’ the media ecology lens and draw on this concept without further specifications. At the second level, authors are more explicit regarding the components that constitute media 313

314  Handbook of digital politics

ecologies, but still do not conceptually engage with ecological theories. At the third level, there is instead an open recognition and deeper engagement with theorizations of media ecologies. First-level studies include Darmon’s reflections (2013, p. 1) on portable devices like smartphones and social media that were merged with more traditional mass media channels, thus creating ‘new media ecologies’. In relation to the Arab uprisings, other accounts have pointed to the existence of ‘hybrid media ecologies’ which emerged through the combination of older and newer media technologies (Robertson, 2013; Wilson and Dunn, 2011). Focusing on media activism in the G20 protests in Canada, Poell and Borra (2012, p. 700) have critically addressed the formation of an ‘activist social media ecology’. Similarly, in relation to the Occupy Wall Street movement’s mobilizations in the US, Thorson et al. (2013, p. 421) speak of ‘a loosely bound media ecology’ in which digital material circulated across different social media platforms. These examples do not usually provide more details regarding the composition of said media ecologies nor dig deeper into the implications of deploying such a concept. Second-level accounts display a more articulated appreciation of the elements that define these media ecologies. For instance, Srinivasan and Fish used the concept to describe the Kyrgyzstan uprisings of 2010, where a multiplicity of digital platforms was used in combination with low-tech media channels, inspiring the establishment of community networks and grassroots coordination ‘through the re-mediation of messages via posters, megaphones, and word-of-mouth’ (2011, p. 3). The authors’ exploration of the media ecology shows activists’ capacity to skilfully navigate these media ecologies to turn their local protest narratives into transnational discourses informing multiple publics. This was possible also due to mainstream media networks such as CNN, Free Speech TV and Al Jazeera in rebroadcasting the content of activists’ citizen media. In particular, Al Jazeera’s role was critical in erecting alliances with social media activists and its use of digital media to share information through live Internet streams and the rebroadcasting agreement with US-based non-profit satellite networks. Similarly, in their study of Egypt’s Tahrir Square protests, Tufekci and Wilson (2012) advocated for the need to move beyond a reductive focus on social media revolutions and specific platforms and urged instead to consider the connectivity infrastructure of these events as a complex, intermeshed media ecology. This ecology, the scholars illustrate, is constituted by three interrelated elements: satellite TV channels such as Al Jazeera and its role in the formation of a new kind of public sphere in the Arab world; social media platforms such as Facebook and Twitter, with their capacity to modify the infrastructure of social connectivity; and mobile devices, with their capabilities for dispersed communication and for cultivating practices of citizen journalism.

Media ecologies, social movements and activism  315

FOUR KEY MEDIA ECOLOGY THEORIES MOBILIZED BY MOVEMENT AND ACTIVISM STUDIES Third-level studies overtly engage with diverse media ecology theories. The assortment of media ecology approaches is such that it has been at the centre of dedicated literature reviews (see for example: Lyle et al., 2020; Wang et al., 2017) whose objective has been solely to sum up the contributions, limitations and future directions of this archipelago of theories. Yet, in my own reviews of the literature (Treré, 2019, 2020), I have found that scholars interested in digital activism and social movements have so far mainly drawn on four conceptual lenses, namely medium theory, information ecology, communicative ecologies and Guattari/Fuller’s understanding of media ecologies. In the next lines, I will briefly introduce them and subsequently examine MovAct studies that have relied on each of these theories. Medium theory, also known as the Toronto school of media ecology, is linked to influential thinkers such as Marshall McLuhan, Harold Innis, Neil Postman and – more recently – Lance Strate, Joshua Meyrowitz and Carlos Scolari. This school approaches media as environments constituted by technologies that coexist and co-evolve and are traversed by processes of survival and extinction. Media ecology theorists are interested in how societies change following the introduction of a new medium and has therefore developed a major focus on cultural and comparative history (Schofield Clark, 2016). According to Strate (2016), medium theory’s understanding of media includes a much broader array of phenomena than other research traditions. Furthermore, he argues, this school is characterized by an interest in media biases, effects and environments. In their examination of the role of citizen media, Kahn and Kellner (2004) refer to the media ecology tradition of the Toronto School and its vision of media as environments. Their aim is to expand the media ecology concept to include newer technologies, while at the same time reconceptualize media ecologies from a critical and reconstructive standpoint. The two scholars are critical of corporate and mainstream media uses of technology and advocate a reconstructive approach that encourage technological appropriations that advance social and political justice. In contrast to these macro narratives of media change that define the Toronto School, information ecology theory adopts a micro, practice-based approach, conceiving an ecology as ‘a system of people, practices, values, and technologies in a particular local environment’, placing the emphasis ‘not on technology, but on human activities that are served by technology’ (Nardi and O’Day, 1999, p. 49). Authors such as Treré (2011, 2012), Barassi (2015) and Barranquero and Barbas (2022) draw on this perspective to connect activists’ practices, imaginaries and cultures to the material affordances of technologies. Their research displays how protesters select their technologies informed by their perceptions of the risks of surveillance and commodification that are inherent in corporate digital platforms. Yet, activists often use corporate social media in critical, creative and unexpected ways. For example, they concomitantly rely on networks of online and offline autonomous media that, despite their more limited capabilities, can safeguard their anonymity and

316  Handbook of digital politics

protect their digital rights without commodifying their actions. In their exploration of the slow media activism of the Pensioners’ Movement, Barranquero and Barbas (2022) blend Nardi and O’Day (1999) and Treré (2019) to foreground the pragmatic, non-idealized approach to communication technologies within Spanish elderly’s hybrid media practices. The communicative ecology perspective sheds light on the exploration of the local context of communication (Hearn and Foth, 2007) and understands the ecology as a milieu of agents connected in several ways by exchanges of mediated and unmediated forms of communication, along technological, social and discursive layers (Tacchi et al., 2003), encompassing the totality of technologies in which people are immersed (Ito, 2009). Peeples and Mitchell’s (2007) study of the protests around the 1999 WTO summit relies on this conceptual lens (Tacchi et al., 2003). The scholars illuminate the three interconnected layers of the media ecology to make sense of the organizational dynamics within activist networks and articulate the communication themes that transpired from the discussions among activists. Drawing on the work of Félix Guattari, Fuller has criticized the use of the environment metaphor of medium theory because it suggests ‘a state of equilibrium’ casting media ecologies as static. In contrast, he defines them in terms of a ‘dynamic interrelation of processes and objects, beings and things, patterns and matter’ (Fuller, 2005, p. 2). Feigenbaum et al. (2013) have borrowed the language of media ecology to explain the multiple relations among social actors, things and environmental conditions in the context of protest camps. Guattari’s focus on the political value of media ecologies allows these authors to go beyond a mere environmental conception of media ecologies and locate the social and the political at the centre of ecological thinking. These scholars suggest that an ecological standpoint can transform the ways in which protesters themselves think about their own positions and interactions within the media ecology. This, the authors maintain, can allow them to ‘navigate the ways in which social movement ideologies are exchanged and carried into the reproduction of protest camps’ infrastructures and practices’ (Feigenbaum et al., 2013, p. 72). Hence, these media ecologies appear as multifaceted environments where activists are able to create, invent and experiment with media technologies. At a first glance, it might seem that there are more differences than resemblances among these ecological traditions. Medium theory in particular has undergone major critique for its technological determinism by authors such as Williams (1974), the proponents of the ‘Social Shaping of Technology’ approach (MacKenzie and Wajcman, 1999) and – more recently – mediatization theory (Hepp, 2013). It would thus seem to be out of place among conceptual lenses that are more interested in studying socially situated media practices, engagements and environments. Yet, a deeper scrutiny reveals that medium theory shares with the other approaches an attention to the materiality of technology and a holistic gaze towards media interactions and evolution. As I have argued extensively (Treré, 2019, 2020), instead of stubbornly obsessing over the variances between these ecological frameworks (or striving to find a definitive ‘winner’ among them that will somehow supplant the others), we should focus on productively acknowledging their overlaps and rely on their com-

Media ecologies, social movements and activism  317

bined strengths in our research endeavours. While the emphasis given to the macro or micro dimension and the political value attributed to the ecologies vary across these views, the ecological gaze that emanates from them urges us to go beyond specific media manifestations. It pushes us to recognize the communicative complexities and multiplicities of social, cultural, and political phenomena by exploring variable constellations, entanglements, interconnections and evolutions of media and people across time and space. In other words, this gaze allows scholars to navigate how activists pursue social justice and political change ‘by any media necessary’ (Jenkins et al., 2016). It is this holistic urge that has inspired MovAct scholars to rely on the media ecology gaze in their assessments of the nexus between media and movements. In the next section, we will review the major findings of this nascent area of inquiry.

MAJOR FINDINGS OF STUDIES ADOPTING THE MEDIA ECOLOGY GAZE The media ecology gaze has unlocked the gates of digital activism and social movement research allowing the inclusion of a broader variety and multiplicity of practices, actors and technologies. This approach has pushed the boundaries of which sociotechnical phenomena are part of the academic conversation about media and activism. Inspired by media ecology theories, scholars have reflected on the role and reconfigurations of independent documentaries in contemporary activism (Lekakis, 2017), the enduring relevance of bodies in recent protests (Boler et al., 2014) and even conceptualized food itself as a medium (Giraud, 2017). This gaze has contributed to advance research on what Feigenbaum (2014) has called the ‘other media’ of activism. This comprises for instance the role of tents within protest camps or the ‘offline performative acts of memory and protest’ in recent research that relates memory, activism and digital environments (Dufays et al., 2021, p. 71). The malleability of media ecology theory has allowed the flourishing of interdisciplinary research that looks at the media–movement nexus with new eyes, challenging the constrictions of conventional social movement studies, and establishing bold connections with – among others – memory and film studies, feminist perspectives, and arts and humanities. Another of its contributions has been moving the attention within accounts of social movements’ technological appropriations from a fixation on new media to the study of the interrelations between old and new media. Hence scholars have started to shed light on the ways old media are being reconfigured by activists in their protest-related practices and used alongside new media, but also zoomed in on the enduring relevance of old media in specific socio-political contexts. For example, in his study of the Italian student movement Anomalous Wave, Treré (2012) examined the key political relevance of mailing lists used by student collectives. While various Italian newspapers and academics were disproportionately focused on the revolutionary potential of newer social media, Treré’s study illustrates how a first-generation digital tool played a role in both the organization and the construction of the col-

318  Handbook of digital politics

lective identity of the movement far greater than platforms such as Facebook and Twitter. Similarly, Bonini (2017) shows the significance of radios in the protests that took place in Turkey in 2013. The Italian scholar studied the role played by Açık Radyo, the only independent and listener-supported radio station based in Istanbul during the Gezi Park mobilizations. He concludes that radio has not lost its value as a powerful alternative media but has instead seen its function reconfigured within an ever-expanding media ecology. Açık Radyo merged with social media to continue amplifying radical political discourses while at the same time enabling activists’ networks to flourish. Another example is the comparative ethnographic study of the media ecologies of various political organizations in Spain and the UK carried out by Barassi (2013). Barassi stressed the enduring political significance of print magazines. Just like mailing lists and the radio in previous examples, these traditional forms of alternative media are not becoming extinct, but rather redefining their role and competing with newer digital platforms in the new media ecologies that define contemporary activism. As it emerges from these analyses, studies drawing on the media ecology gaze are generally interested in looking at transitions, displacements, abandonments, adjustments and reconfigurations of media technologies and practices. Another clear example of this is Giraud’s (2014) analysis of the changes that happened to Indymedia, one of the most emblematic examples of radical media network that emerged in the late 1990s and became closely related to the global justice movement. Due to the spread of corporate social media, Indymedia centres declined in the 2010s, but Giraud shows that Indymedia has not entirely disappeared and continues to fulfil an archival function within an expanding media ecology characterized by new digital platforms that are now deployed for coordinating more pressing political actions. This shift towards the appreciation of the dynamic interactions between older and newer media practices has also spurred a renewed interest in the concept and the implications of the notion of hybridity in both conventional and unconventional forms of media politics (Ardizzoni, 2015; Chadwick, 2017; Dahlberg-Grundberg, 2016; Dennis et al., 2016; Iannelli and Giglietto, 2015; Jenkins et al., 2016; Russell, 2017; Treré, 2019: Treré and Yu, 2021). Studies adopting the ecological gaze can be situated within what Iannelli has called ‘hybrid politics’ research (2016). This emerging and multi-layered area of investigation criticizes the fetishization of technological novelty and traditional conceptions based on a single, dominant media logic in political newsmaking (see Altheide and Snow, 1979). Authors who are part of this academic conversation favour instead a holistic standpoint that maps open and often unpredictable interrelations between multiple logics, formats, outlets, and actors in diverse socio-political contexts. Relatedly, another major contribution of the ecological gaze has been to push MovAct studies away from a fascination with the online sphere towards the analysis of the hybrid, dynamic interplay between physical and digital spaces. Relevant studies of activism in the early 2000s were specifically interested in the potential of the digital realm and the emergence of online repertoires of contention. For instance, McCaughey and Ayers (2003), in an important edited collection on cyber-activism,

Media ecologies, social movements and activism  319

expressed their worry about the possible disappearance of the body in the new forms of online protest. Their reflection was aligned with other researchers’ focus on the virtuality of contention, and a concomitant declining attention to physical locations, social contexts, and the embodied experiences of social movements. This enchantment with virtuality was of course not confined to the realm of activism but resonated with a broader context of society’s fascination with virtual reality and its imaginary of infinite possibilities beyond the limiting constraints of bodies and the material world. This imaginary shaped by sci-fi narratives and corporate discourse permeated also academic reflections on the new horizons of cyber-protest feeding what social movement scholar Merlyna Lim has called the ‘fallacy of spatial dualism’ (Lim, 2015, p. 118). This fallacy holds that the physical and digital spaces are distinct and can thus be appraised separately. It has infused most accounts of activism in the early 2000s, casting protest activities and collective action as practices that are somehow progressively separated from the materiality of physical spaces. New research inspired by the ecological gaze does not look at the online and offline realms as separate but aims instead to examine the complexities of ‘cyber-urban space’ (Lim, 2015), that is the interweaving between digital and physical spaces (Barranquero and Barbas, 2022; Farinosi and Treré, 2010; Lekakis, 2017; Lim, 2015, 2018; Treré, 2018). It asks: how do contemporary movements and protesters navigate this space as a socially constructed environment ‘produced in the interaction of, and within a continuum of, online and offline relations’ (Lim, 2015, p. 118)? Here, the focus lies on the specific ways in which social actors make sense of this interplay across different political contexts, along with the motivations and factors that drive activists to choose specific constellations of technologies (Liu, 2021). As Foust and Hoyt (2018, p. 48) observe, ‘media ecology can be understood as the entanglement of the geosphere (material landscape) and the infosphere (symbolic action and flows of information)’. This focus aligned with what Ardizzoni calls ‘matrix activism’, a conceptual framework that ‘allows us to explain the hybrid nature of new forms of dissent and resistance, as they are located at the intersection of alternative and mainstream, nonprofit and corporate, individual and social, production and consumption, online and offline’ (2015, p. 1086). Finally, another main contribution of the ecological gaze has been to show the ambivalences and ambiguities of digital activism in relation to the appropriation of corporate social media platforms. For instance, Harlow (2016), in her ethnography of the Salvadoran group Activista and its Todos Somos Agua campaign, demonstrated that online social media – Facebook in particular – were reconfigured as a form of alternative activist media in El Salvador. Harlow illustrates that Facebook offered a space that allowed people with non-mainstream views to voice an opinion and share news about mining, water contamination and other social issues that the public would otherwise never learn about. Her interviewees saw Facebook as a reclaimed media territory for youths, whose voices are normally excluded by more conventional media. Hence, her analysis shows that through a media ecology lens we can better examine how, notwithstanding enduring digital inequalities, social media can be (re)appropriated in non-hegemonic ways. These findings echo Treré’s analysis

320  Handbook of digital politics

of the technopolitical practices and imaginaries of the 15M movement in Spain (2019), where corporate digital platforms like Twitter were skilfully appropriated by protesters to maximize their visibility and spread their narratives generating a new type of algorithmic resistance. Yet, 15M activists continued to concomitantly build their own autonomous networks, software and infrastructure that coexisted alongside their tactical corporate appropriations. Through ecological explorations, these studies display the ambiguities in the cohabitation of alternative and hegemonic infrastructures and the counterhegemonic uses that can emerge in the context of digital protest (for a similar analysis in the Mexican political context, see Pool, 2022). They defy taken for granted and Manichean assumptions on the nature of contemporary contention, favouring instead the scrutiny of the historically defined and context-dependent contradictions of digital activism and social media power.

CHALLENGES AND AREAS OF DISAGREEMENT How can we define a media ecology and what elements constitute it? As my previous analysis of MovAct studies’ engagement with the term reveals, this simple yet important point is often not clearly articulated, especially in those accounts that rely on a more superficial reading of the ecological metaphor. Some authors include a broader variety of media practices, while others more specifically focus on online or digital-only ecologies. An evident risk of superficial ecological accounts is that the media ecology can be merely evoked but not carefully disentangled. This is a tendency that I have also witnessed in the essays of students or early career researchers that rely on this approach for the first time. Sometimes the ecological gaze is only used to list an array of technologies but there is no further engagement with the motivations of the activists, the evolving entanglements of human and non-human actors, the role of a specific media practice within the ecology and in relation to a specific socio-political context, the affordances and overall architecture of digital platforms, etc. An ecological exploration should always be thick and deep, otherwise it runs the risk of becoming only a declaration of variety which can only represent a necessary first step but cannot substitute a more profound investigation. The opposite risk can also be an issue, that is seeing socio-material complexity in activist practices even where there is none (Rodríguez, 2017). For instance, some collectives may rely exclusively on the production of video and adopt no more than a few media outlets, e.g., a community radio, a video collective, etc. In these cases, the media ecology lens may not represent necessarily the best option in our conceptual toolbox. Nadler (2019) has been a strenuous critic of the use of the metaphors of news ecology and ecosystems in the field of digital journalism. According to him ‘the ecosystem metaphor has come to be used in ways that suggest news systems take shape through a spontaneous, self-ordering principle associated with ecological systems. … The metaphor can obscure the political choices that make it possible for societies to build digital media systems reflecting more or less egalitarian and democratic values’ (2019, p. 825). In MovAct studies, this risk has been mostly avoided by stressing

Media ecologies, social movements and activism  321

the political nature of such media ecologies and grounding their inspections in the concrete social contexts where they arise and unfold. However, it is a danger that scholars should be aware of, and that thin ecological accounts may encounter if they do not drill down sufficiently into the political dynamics of hybrid media systems.

MY THINKING AS A SCHOLAR ON THE TOPIC: A FLEXIBLE AND IMPURE ECOLOGICAL GAZE I have been developing a media ecology approach to social movements and activism during the last fifteen years, first during my PhD research that started in 2008 (Treré, 2011) and then through several projects in different countries. I view the media ecology as a gaze, an invitation to look at the complexities at the intersection between politics and media technologies in a holistic way. I have come to the conclusion that the variety of approaches in this area (and, sometimes, the creative chaos that results from that) is somehow unavoidable, and perhaps even welcome. As scholars, what we can do is outline a set of conceptual recommendations and empirical lessons from the field, but I do not think that we should be too prescriptive. The risk is that being excessively rigid in terms of what elements should be part of the ecology and which theories should be included in the conversation will hinder the development of original and unexpected accounts; yet, this is precisely what the media ecology lens wants to promote and encourage. Alternative media scholar Alfonso Gumucio Dagron (2007) has criticized those academics who obstinately focus on the supposed purity of citizen media stressing instead their impurity and richness. He adds that an exact terminological labelling of these experiences often serves exclusively academic purposes but can risk excluding other projects that may not fit into these blueprint definitions. While some authors advocate for clearer delimitations of phenomena such as alternative media and digital activism, in line with Gumucio Dagron I consider that this nascent area of inquiry has benefited greatly from its flexibility and openness of interpretations that have contributed to disclose new horizons of inquiry. One of them is the exploration of disconnective practices and the ways activists purposively use disconnection as a form of resistance (Kaun and Treré, 2020; Natale and Treré, 2020) in the context of our media saturated societies. This kind of research defies conventional understandings that necessarily conceive activists as over-connected geeks and connectivity as an inherent force for the good. I am not sure that this kind of reflection would have emerged had we been too specific in terms of what should be included within the ecologies. What should instead be improved is our engagement and contribution to media ecology theory and the development of cross-cultural research in this field that informs these understandings. How and why can we compare different ecologies across time and space and how to account for variations require some degree of mapping and specification, even if we should always remain open to differences and variations and be prepared to be surprised and amazed by new findings.

322  Handbook of digital politics

KEY PRIORITIES FOR FUTURE RESEARCH INTO THE MEDIA ECOLOGY GAZE AND ACTIVISM Inspired by the work of John Dewey, media ecology studies have considered the environmental consequences of media technologies (Lopez, 2012; Maxwell and Miller, 2012). In her review of media ecology theory, Schofield Clark (2016) shows that these accounts have challenged the myth of clean technologies, exposing issues of technological waste. The Toronto school never really drilled down into the environmental consequences of the media, but new studies in the media ecology field advocate for a ‘materialist media ecology’ or a ‘deep media ecology’ (Arroyave-Cabrera and Miller, 2017) where the interrelations with the natural world are at the centre (see also Oricchio, 2021). Media ecology approaches in MovAct studies need to do the same and start to seriously pay attention to the interdependence between humans, technologies and the environment. A promising attempt to combine technopolitics, media ecologies and ecosocial struggles in a context of communicative hybridity is represented by a recent study of #SOSPuebloShuar on Twitter (Vanegas Toala et al., 2020). This is a key area of future research that could for example analyse environmental social movements and their interactions with media while at the same time scrutinizing the environmental consequences (and awareness) of such interventions. Another area of inquiry that is being developed looks not only at the media ecologies that activists create, but also at the changes that social movements provoke in the creation, evolution and development of broader media ecologies, intended as interactions between different media outlets. A clear example of this is the case of the 15M movement in Spain. As Flesher Fominaya and Gillan (2017, p. 391) show, the Spanish movement not only relied on a multidimensional ecology of technologies, ‘but provided a support base and impetus for the development of various critical media initiatives that attempted to put into practice alternative media business models … While some such initiatives existed prior to 15-M … the supply of and demand for independent critical media increased in a virtuous circle … thus altering the media ecology of political communication in Spain’. This opens up the field to examining the contributions of social movements (and their media ecologies) in the establishment and flourishing of new ecosystems of alternative media and in the subsequent alteration of wider hybrid media systems (Barbas and Treré, 2023). Another promising area of research is constituted by the alliances between human and non-human actors within media ecologies. Given the increasing importance that algorithmic power has for contemporary digital and data activists (Milan, 2015), the study of the mutual shaping between social movements and algorithms (Treré, 2019) should be a priority for scholars and academics alike. To better grasp the communicative complexity of social movements and activism, all future media ecological endeavours will have to develop interdisciplinary and multidisciplinary research and establish fruitful collaborations with disciplines as varied as critical cartography, architecture, arts, design, disconnection studies and many more. The malleability of this gaze needs to be preserved and nurtured to capture the technolog-

Media ecologies, social movements and activism  323

ical developments of the future in the exciting and ever-changing area of inquiry that is unconventional politics.

FURTHER READING On Media Ecology Theory Schofield Clark, L. (2016). Media ecology theory. In K. B. Jensen and R. T. Craig (eds.), The International Encyclopedia of Communication Theory and Philosophy. Hoboken, NJ: Wiley Blackwell. Scolari, C. A. (2012). Media ecology: Exploring the metaphor to expand the theory. Communication Theory, 22(2), 204–225. Scolari, C. A. (forthcoming). On the Evolution of Media: Understanding Media Change. London: Routledge. Strate, L. (2016). Media ecology. In K. B. Jensen and R. T. Craig (eds.), The International Encyclopedia of Communication Theory and Philosophy. Hoboken, NJ: Wiley Blackwell.

On Media Ecologies, Social Movements and Activism Barassi, V. (2015). Activism on the Web: Everyday Struggles against Digital Capitalism. New York: Routledge. Foust, C. R. and Hoyt, K. D. (2018). Social Movement 2.0: Integrating and assessing scholarship on social media and movement. Review of Communication, 18(1), 37–55. Mercea, D., Iannelli, L., and Loader, B. (2016). Protest communication ecologies. Information, Communication & Society, 19(3), 279–289. Treré, E. and Mattoni, A. (2016). Media ecologies and protest movements: Main perspectives and key lessons. Information, Communication & Society, 19(3), 290–306. Treré, E. (2019). Hybrid Media Activism: Ecologies, Imaginaries, Algorithms. London: Routledge. Treré, E. (2020). Media ecologies. In L. Pérez-González, B. Blaagaard, and M. Baker (eds.), Routledge Encyclopaedia of Citizen Media. London: Routledge.

REFERENCES Altheide, D. L. and Snow, R. P. (1979). Media Logic. London: Sage. Ardizzoni, M. (2015). Matrix activism: Media, neoliberalism, and social action in Italy. International Journal of Communication, 9, 18. Arroyave-Cabrera, J. A. and Miller, T. (2017). Da ecologia de mídia à ecologia profunda de mídia: Esclarecer a metáfora e visibilizar o seu impacto no meio ambiente. Palabra Clave, 20(1), 239–268. Barassi, V. (2013). When materiality counts: The social and political importance of activist magazines in Europe. Global Media and Communication, 9(2), 135–151. Barassi, V. (2015). Activism on the Web: Everyday Struggles against Digital Capitalism. New York: Routledge. Barbas, Á. and Treré, E. (2023). The rise of a new media ecosystem: Exploring 15M’s educommunicative legacy for radical democracy. Social Movement Studies, 22(3), 381–401.

324  Handbook of digital politics

Barranquero, A. and Barbas, Á. (2022). The slow media activism of the Spanish pensioners’ movement: Imaginaries, ecologies, and practices. International Journal of Communication, 16, 25. Boler, M., Macdonald, A., Nitsou, C., and Harris, A. (2014). Connective labor and social media: Women’s roles in the ‘leaderless’ Occupy movement. Convergence, 20(4), 438–460. Bonini, T. (2017). Twitter or radio revolutions? The central role of Açık Radyo in the Gezi protests of 2013. Westminster Papers in Communication and Culture, 12(2), 1–17. Chadwick, A. (2017). The Hybrid Media System: Politics and Power, 2nd edition. Oxford: Oxford University Press. Dahlberg-Grundberg, M. (2016). Technology as movement: On hybrid organizational types and the mutual constitution of movement identity and technological infrastructure in digital activism. Convergence, 22(5), 524–542. Darmon, K. (2013). Introduction: Protest in the new media ecology. Networking Knowledge: Journal of the MeCCSA Postgraduate Network, 6(3), 1–2. Dennis, J. W., Chadwick, A., and Smith, A. P. (2016). Politics in the age of hybrid media: Power, systems, and media logics. In A. Bruns, G. Enli, E. Skogerbø, A. Olof Larsson, and C. Christensen (eds.), The Routledge Companion to Social Media and Politics (pp. 7–22). New York: Routledge. Dufays, S., Zicari, M., Mandolessi, S., and Cardoso, B. (2021). Twitter as a mnemonic medium from an ecological perspective: Ayotzinapa and the memory of Tlatelolco in Mexico. History and Memory, 33(2), 46–79. Farinosi, M. and Treré, E. (2010). Inside the “People of the Wheelbarrows”: Participation between online and offline dimension in the post-quake social movement. The Journal of Community Informatics, 6(3). Feigenbaum, A. (2014). Resistant matters: Tents, tear gas and the “other media” of Occupy. Communication and Critical/Cultural Studies, 11(1), 15–24. Feigenbaum, A., Frenzel F., and McCurdy, P. (2013). Protest Camps. London: Zed Books. Flesher Fominaya, C. and Gillan, K. (2017). Navigating the technology-media-movements complex. Social Movement Studies, 16(4), 383–402. Foust, C. R. and Hoyt, K. D. (2018). Social Movement 2.0: Integrating and assessing scholarship on social media and movement. Review of Communication, 18(1), 37–55. Fuller, M. (2005). Media Ecologies: Materialist Energies in Art and Technoculture. Cambridge, MA: MIT Press. Giraud, E. (2014). Has radical participatory online media really “failed”? Indymedia and its legacies. Convergence, 20(4), 419–437. Giraud, E. (2017). Displacement, “failure” and friction. In T. Schneider, K. Eli, C. Dolan, and S. Ulijaszek (eds.), Digital Food Activism. London and New York: Routledge, 130–150. Gumucio Dagron, A. (2007). Call me impure: Myths and paradigms of participatory communication. In L. Fuller (ed.), Community Media: International Perspectives (pp. 197–207). New York: Palgrave Macmillan. Harlow, S. (2016). Reconfiguring and remediating social media as alternative media: Exploring youth activists’ digital media ecology in El Salvador. Palabra Clave, 19(4), 997–1026. Hearn, G. N. and Foth, M. (2007). Communicative ecologies. The Electronic Journal of Communication, 17(1–2). http://​eprints​.qut​.edu​.au/​8171/​1/​8171​.pdf. Hepp, A. (2013). Cultures of Mediatization. Cambridge: Polity Press. Iannelli, L. (2016). Hybrid Politics: Media and Participation. London: Sage. Iannelli, L. and Giglietto, F. (2015). Hybrid spaces of politics: The 2013 general elections in Italy, between talk shows and Twitter. Information, Communication & Society, 18(9), 1006–1021.

Media ecologies, social movements and activism  325

Ito, M. (2009). Hanging Out, Messing Around, and Geeking Out: Kids Living and Learning with New Media. The John D. and Katherine T. Macarthur Foundation Series on Digital Media and Learning. Cambridge, MA: MIT Press. Jenkins, H., Shresthova, S., Gamber-Thompson, L., Kligler-Vilenchik, N., and Zimmerman, A. (2016). By Any Media Necessary: The New Youth Activism. New York: New York University Press. Kahn, R. and Kellner, D. (2004). New media and internet activism: From the “Battle of Seattle” to blogging. New Media & Society, 6(1), 87–95. Kaun, A. and Treré, E. (2020). Repression, resistance and lifestyle: Charting (dis)connection and activism in times of accelerated capitalism. Social Movement Studies, 19(5–6), 697–715. Lekakis, E. J. (2017). Alternative media ecology and anti-austerity documentary: The Greekdocs archive. Journal of Alternative & Community Media, 2(1), 28–44. Lim, M. (2015). A cyberurban space odyssey: The spatiality of contemporary social movements. New Geographies, 7, 117–123. Lim, M. (2018). Roots, routes, routers: Communications and media of contemporary social movements. Journalism & Communication Monographs Series, 20(2), 92–136. Liu, J. (2021). Technology for activism: Toward a relational framework. Computer Supported Cooperative Work (CSCW), 30(5), 627–650. Lopez, A. (2012). The Media Ecosystem: What Ecology Can Teach Us About Responsible Media Practice. Berkeley, CA: Evolver Editions. Lyle, P., Korsgaard, H., and Bødker, S. (2020). What’s in an ecology? A review of artifact, communicative, device and information ecologies. Proceedings of the 11th Nordic Conference on Human-Computer Interaction: Shaping Experiences, Shaping Society (pp. 1–14). MacKenzie, D. and Wajcman, J. (1999). The Social Shaping of Technology, 2nd edition. Buckingham: Open University Press. Maxwell, R. and Miller, T. (2012). Greening the Media. New York: Oxford University Press. McCaughey, M. and Ayers, M. (2003). Cyberactivism: Online Activism in Theory and Practice. New York: Routledge. Milan, S. (2015). When algorithms shape collective action: Social media and the dynamics of cloud protesting. Social Media + Society, 1(2). Nadler, A. (2019). Nature’s economy and news ecology: Scrutinizing the news ecosystem metaphor. Journalism Studies, 20(6), 823–839. Nardi, B. and O’Day, V. (1999). Information Ecologies: Using Technology with Heart. Boston, MA: MIT Press. Natale, S. and Treré, E. (2020). Vinyl won’t save us: Reframing disconnection as engagement. Media, Culture & Society, 42(4), 626–633. Oricchio, S. (2021). Digitization of ecology and ecologization of media: Going beyond ICT environmental impact. TECNOSCIENZA: Italian Journal of Science & Technology Studies, 12(1), 99–116. Peeples, J. and Mitchell, B. (2007). No mobs – no confusions – no tumult: Networking civil disobedience. Electronic Journal of Communication, 17(1–2). www​.cios​.org/​EJCPUBLIC/​ 017/​1/​01714​.HTML. Poell, T. and Borra, E. (2012). Twitter, YouTube, and Flickr as platforms of alternative journalism: The social media account of the 2010 Toronto G20 protests. Journalism, 13(6), 695–713. Pool, M. C. L. (2022). Ecología comunicativa del movimiento# YoPrefieroElLago: un abordaje desde los medios privados, independientes y alternativos. Comunicación y Sociedad, 19, 1-31. Robertson, A. (2013). Connecting in crisis: “Old” and “new” media and the Arab Spring. The International Journal of Press/Politics, 18(3), 325–341.

326  Handbook of digital politics

Rodríguez, C. (2017). Studying media at the margins: Learning from the field. In V. Pickard and G. Yang (eds.), Media Activism in the Digital Age. New York: Routledge, 49–62. Russell, A. (2017). Journalism as Activism: Recoding Media Power. New York: John Wiley & Sons. Schofield Clark, L. (2016). Media ecology theory. In K. B. Jensen and R. T. Craig (eds.), The International Encyclopedia of Communication Theory and Philosophy. Hoboken, NJ: Wiley Blackwell. Srinivasan, R. and Fish, A. (2011). Revolutionary tactics, media ecologies, and repressive states. Public Culture, 23(365), 505–510. Strate, L. (2016). Media ecology. In K. B. Jensen and R. T. Craig (eds.), The International Encyclopedia of Communication Theory and Philosophy. Hoboken, NJ: Wiley Blackwell. Tacchi, J., Slater, D., and Hearn, G. N. (2003). Ethnographic Action Research: A User’s Handbook. New Delhi: UNESCO. http://​eprints​.qut​.edu​.au/​archive/​00004399/​. Thorson, K., Driscoll, K., Ekdale, B., et al. (2013). YouTube, Twitter and the Occupy Movement: Connecting content and circulation practices. Information, Communication & Society, 16(3), 421–451. Treré, E. (2011). Social movements and alternative media: The “anomalous wave” movement and the ambivalences of the online protest ecology. Unpublished Doctoral dissertation, Udine, Italy, University of Udine. http://​dspace​-uniud​.cilea​.it/​handle/​10990/​306. Treré, E. (2012). Social movements as information ecologies: Exploring the coevolution of multiple internet technologies for activism. International Journal of Communication, 6, 2359–2377. Treré, E. (2018). Nomads of cyber-urban space: Media hybridity as resistance. In M. Mortensen, C. Neumayer, and T. Poell (eds.), Social Media Materialities and Protest: Critical Reflections (pp. 42–55). London: Routledge. Treré, E. (2019). Hybrid Media Activism: Ecologies, Imaginaries, Algorithms. London: Routledge. Treré, E. (2020). Media ecologies. In L. Pérez-González, B. Blaagaard, and M. Baker (eds.), Routledge Encyclopaedia of Citizen Media. London: Routledge. Treré, E. and Yu, Z. (2021). The evolution and power of online consumer activism: Illustrating the hybrid dynamics of “consumer video activism” in China through two case studies. Journal of Broadcasting & Electronic Media, 65(5), 761–785. Tufekci, Z. and Wilson, C. (2012). Social media and the decision to participate in political protest: Observations from Tahrir Square. Journal of Communication, 62(2), 363–379. Vanegas Toala, Y. V., Medina Bravo, P., and Rodrigo Alsina, M. (2020). Technopolitics, connective action and convergent activism: Emerging communication practices from ecosocial struggles. IC: Revista Científica de Información y Comunicación, 17, 505–532. Wang, X., Guo, Y., Yang, M., Chen, Y., and Zhang, W. (2017). Information ecology research: Past, present, and future. Information Technology and Management, 18(1), 27–39. Williams, R. (1974). Television: Technology and Cultural Form. London: Fontana. Wilson, C. and Dunn, A. (2011). The Arab Spring: Digital media in the Egyptian revolution – Descriptive analysis from the Tahrir data set. International Journal of Communication, 5, 1248–1272.

21. E-petitioning and changing state–citizen engagement Scott Wright and Ariadne Vromen

INTRODUCTION Electronic petitioning – or ‘e-petitioning’ – has become a mainstream mechanism for political engagement for many citizens. At a purely quantitative level, e-petitions have clearly been successful across a number of jurisdictions (e.g. government sites such as Downing Street e-petitions, We the People and parliament sites such as the German Bundestag, Scottish Parliament – Wright, 2012; Bochel and Bochel, 2017), for individual organizations (Chadwick and Dennis, 2017), and via not-for-profit (e.g. change.org or 38 Degrees) petition and campaign platforms that typically sit beyond the nation state and have grown to prominence in recent years (Karpf, 2016; Vromen, 2017). The Hansard Society’s audit of political engagement argues that they are the ‘single most important route to engage the public that Parliament currently has at its disposal, apart from direct contact with a representative’ (Hansard Society, 2016, p. 28). Tens of thousands of petitions are created on government-led e-petition platforms every year (far more than were submitted by paper – Bochel, 2013, p. 803), with many more hosted on petition platforms. E-petitions are seen as a way to modernize the centuries-old and relatively unchanged right to petition political institutions, with the platforms creating what is arguably a novel form of political participation that provides a fresh challenge to institutional petition systems (Halpin et al., 2018). However, controversy about this means of engagement persists. Success or impact of e-petitioning is about more than the volume of signatures received, or the volume of petitions created; and varies from platform to platform, issue to issue, and differs between petition creators and the people running e-petitioning systems. Success might be measured in terms of agenda setting, or the policy impact, or an increased number of, or more deeply engaged, supporters, for example. The ‘value’ to democracy of e-petitions remains contested. Critics lament e-petitions as little more than ‘slacktivism’ or ‘clicktivism’, which have limited, if any, impact on politics (Morozov, 2009; Shulman, 2009). For others, they are undemocratic and may encourage policy to benefit better-organized or wealthier social groups (see Wright, 2015, 2016). The reality is that e-petition systems are diverse, and shaped by their socio-political contexts and technical design and affordances. This chapter seeks to assess these different affordances, drawing on research from across government and non-governmental e-petition platforms. The chapter identifies key issues in current research on e-petitions and how findings cut across the 327

328  Handbook of digital politics

different platforms. The chapter argues that e-petitions have the potential to impact upon policy and empower individuals, but that how e-petitions are institutionalized is crucial, and there are a number of challenges that can serve to limit their impact with these varying between institutional and independent e-petitioning systems. By canvassing e-petition platforms broadly to include both those that are government-led and housed on parliamentary or affiliated websites, as well as non-government platforms that are associated with digital campaigning organizations and have emerged over the last decade or so, we also recognize that there is variation in terms of e-petitions started by citizens versus those initiated by organizations.

AGENDA-SETTING AND MODERATION The ability to set or control the agenda is considered crucial in many theories of democracy (Smith, 2009). In this context, agenda-setting works in two directions. First, there is the question of who gets to choose the topics of petitions, and whether e-petitions lead to different issues making the agenda. We might expect the more egalitarian and open space of e-petitioning, in which people can use e-petition systems to ‘reach in’ to decision-making processes (Bochel and Bochel, 2017, p. 696), to facilitate different kinds of policies and issues to make the agenda rather than more formal spaces where vested and wealthier interests can dominate. However, research suggests that they broadly follow similar patterns with people putting forward particularistic policy proposals rather than ones with broader consequences, and postmaterialist rather than redistributive ones (Hersh and Shaffner, 2018). They also tend to target government and a broad range of public policy issues, rather than consumer or non-political issues (Halpin et al., 2018) – though when they do target commercial entities they can be successful (see more below – Minocher, 2019). Second, there is the question of how petitions are managed by the host, whether it be coaching, editing, or the rules (be they formal, or driven by algorithms) about which petitions are promoted or placed prominently (Karpf, 2016). Third, there is the question of whether and how e-petitions actually shape the agenda of governments and other political actors. On the latter point, moderation may be one part of a wider process in which a petition committee (or similar) seeks to strengthen the engagement with petitions and petitioners (Asher et al., 2019). E-petitions are managed in myriad ways by their controllers. Most petition systems initially allow people to create their own petitions. A moderator will then review the petition, and they may publish the petition as it is, delete the petition against a set of criteria (which are now generally public), or, in some cases, ask (or at least suggest to) the creator to edit the petition as part of a curation process (Scottish Parliament, 2010, p. 6; Karpf, 2016). In the government-led systems, such as Downing Street, We the People and Direct. gov, decisions over what petitions to accept are typically made by individual civil servants, though some also include a user-led flagging system. Government-led moderation can create controversy, as it is sometimes perceived as unjustified censorship (Wright, 2006, pp. 558–560). However, the issues are more to do with

E-petitioning and changing state–citizen engagement  329

under-resourcing, limited training, and poorly designed and advertised rules than a ‘government plot’ to stop criticism (Wright, 2012). When a petition system is perceived as well run it can increase trust (Carman, 2010), but if it is perceived as poorly run and lacking transparency it can have a negative effect (Escher and Riehm, 2016, pp. 20–21). Some legislatures take it very seriously, with significant staff (e.g. around 80 in Germany – Lindner and Riehm, 2011, p. 9), and these arguably have greater potential to set and influence the agenda (Jungherr and Jürgens, 2010). Another way in which governments can shape the agenda of petitions is by defining how many signatures must be received to get an official reply, and how long is given to achieve this. If the bar is set too low or too quickly, petition systems can become swamped and little meaningful can happen (or petitions are simply ignored – Wright, 2012, 2016), too high and it can put people off or become the preserve of better organized interest groups. Many of these problems persist across other institutional or strong petition systems (Karpf, 2016, p. 78). The different non-governmental e-petition platforms and organizations also have a wide variety of moderation processes and approaches to agenda-setting as part of what are often ‘dramatically different [operating] philosophies’ – it is an industry, and platforms such as MoveOn and Change.org ‘combine digital listening and experimentation to help determine which petitions go viral and how those viral moments are transformed into long-term movements’ via analytic activism (Karpf, 2016, pp. 60–61). However, organizational interests (e.g. growing their email list, perceived social value, potential for impact) might influence which petitions get the most attention. For digital campaigning organizations like GetUp, Avaaz and MoveOn that act more as interest groups, they usually make decisions on campaign and petition issues and subsequently appeal to their memberships to join in or sign on (Vromen, 2017). Other platforms, such as Change.org or Care2, encourage citizens to start their own petitions on issues that matter to them, aimed at a target of their choice. For Change.org, deciding which petitions to allow, and particularly to support, has proved controversial internally, with a shift from more progressive language focusing on the common good to openness and neutrality (Karpf, 2016, p. 66). But, as Karpf notes, this does not mean they do not take positions. Rather, Change. org focuses on smaller, personal, and more winnable issues (that might involve collaboration with decision-makers) that can generate growth, or approaches larger issues on such terms. This ideological neutrality (within limited parameters) is different from social justice-focused groups such as Avaaz, MoveOn, 38 Degrees and GetUp. The Australian organization GetUp, for example, emphasizes strategy and deliberation in making campaign choices. The campaign staff chose to prioritize a marriage equality campaign, which was not listed in the top ten priorities from its membership survey, because it was considered an important issue to campaign on. This long-term campaign, which resulted in legislative change in Australia in 2017, became one of their most high-profile successes (Rugg, 2019). GetUp has trialled different ways to engage and facilitate action by its supporters, starting with its Campaign Ideas Forum (CIF), then shifting to the Communityrun.org platform which allowed individuals to

330  Handbook of digital politics

develop their own campaigns, and now Campaigns By Me – the latter are separate from GetUp. The choices of the organization’s campaign staff and executive are still key here, but the philosophy and organizational logics that inform those choices are very different. The media are central to agenda-setting with e-petitions – in the early days, some journalists monitored petition websites looking for stories (Wright, 2012), but increasingly petition creators – or the platforms themselves – are contacting journalists and the petition is a media object (Karpf, 2016, p. 60) with campaigners exploiting a hybrid media system (Chadwick, 2013; Wong and Wright, 2018, 2020). Arguably the pre-eminent example today is Change.org. It was founded in 2007 in San Francisco by Ben Rattray as a commercial, for-profit B Corporation, with income driven by revenue from advertising or sponsorship of petitions, originally by campaigning organizations interested in signer data such as emails, though over time this also became via campaign donations from individuals. The original idea of Change. org was to give individuals their own capacity to start campaigns via petitions. David Karpf (2016, p. 64) focuses on the media driven sharing logic of large-scale petition platforms like this, and describes Change.org as using a ‘reverse mullet’ business model: that is there is a user generated ‘party in the front’ starting petition, with a lead generation business operation behind the scenes. Most of the staff working in these organizations are experienced political campaigners, and Change.org now has small bases in countries all around the world. Karpf (2016, p. 62) cites a senior Change staffer as saying that over half of the staff are media specialists, and they ‘promote, shepherd, and support citizen petitions that can be packaged into stories in the nonpolitical media’ such as The Ellen DeGeneres Show. In September 2021 Rattray stepped down as CEO, and the platform officially transitioned to become fully not for profit, governed by the Change.org Foundation (now chaired by Rattray), underwritten by equity donations from 50 leading entrepreneurs and philanthropists.

PARTICIPANTS’ PERSPECTIVES Surprisingly few studies of e-petitions have focused on surveying the people who signed the petitions, to understand why they chose to act, what they thought about the process and how impactful the petition was. In particular, scholars analysing patterns of participation have identified ‘super-participants’: people who create and/or sign dozens of petitions (Jungherr and Jürgens, 2010; Cotton, 2011, p. 36; Wright, 2012; Halpin et al., 2018). Closer analysis of these participants would be helpful to better understand what motivates them, whether they reflect the broader population, their links to formal organizations, and how they operate (for example, whether they write to journalists). A recent survey looked at those who sign and share online petitions in Australia and Germany (Vaughan et al., 2022), finding that while in the past petition signing had tended to skew to more educated and younger citizens, signing has now become much more of a mainstream form of political engagement. What mainly determine the tendency to sign, and importantly share and promote petitions

E-petitioning and changing state–citizen engagement  331

via social media, are the norms of citizenship and understandings of democracy that people have. The more people are committed to the idea of sharing information and ideas about politics the more likely they are to sign to petitions and share them in their social media networks. For citizens this distribution of information via petition signing is not only about news or political information that is readily shared across social media platforms. It is real-time ‘social information’ that is shared on social media platforms: the feedback, publicness, and the knowledge accumulated of what other people in the network are doing and saying online (Margetts et al., 2015).

IMPACT Research into the impact of e-petitions has produced mixed results. The first key debate has been over how to measure the impact of e-petitions. This should arguably differ between government-led systems and the petition platforms. One perspective has been to focus on whether they have substantive impacts on public policy (Shulman, 2009 – focusing specifically on mass e-mail campaigns), while others argue that this is an unduly narrow analytical frame that captures neither the rationale for many organizations in using e-petitions, nor how e-petitions can sit alongside a much wider range of campaigning activities (Karpf, 2010, 2016). Leston-Bandeira (2019, p. 420) usefully distinguishes between four functions or roles that e-petitions perform – linkage, campaigning, scrutiny and policy. Linkage refers to roles such as education, public engagement and political participation. Campaigning refers to functions such as recruitment, dissemination and mobilization. Scrutiny refers to roles such as agenda-setting and evidence gathering. Finally the policy category includes the review, improvement and influence on policy alongside policy change. This broadly mirrors Rosenberger et al.’s (2022) finding that e-petitions support voice in parliament, public mobilization and links with constituents – though it is worth noting that politicians themselves are often concerned about the potential impact of e-petitions (Wright, 2012; Matthews, 2020). The second key debate is more normative: just how much influence should e-petitions be allowed to have within a democratically elected representative democracy? While participating through e-petitions might be easier, participation is still unequal (Wright, 2012; Puschmann et al., 2017; Halpin et al., 2018, p. 439) and this extends to social media debates of e-petitions too. Ultimately the silent majority is far larger than the more vocal minority of people who have signed any single petition. For example, after 1.8 million people signed an anti-road-pricing petition, Blair (2007) argued that: ‘it’s not possible, wise or healthy for politicians to try and sweep [such views] under the carpet’ and argued that the response to the e-petition ‘shows that my government is listening’. However, Blair (2007) was not promising to drop the policy, but to have a public debate with a view to convincing people of the need for change, because ‘road charging is surely part of the answer’. Less than a year later, however, road pricing was quietly shelved – with many people crediting the e-petition. But is this enhancing democratic legitimacy and accountability? Road

332  Handbook of digital politics

pricing was included in Labour’s 2005 General Election manifesto, on which the party was democratically elected (Labour Party, 2005, p. 25). Yet when the policy was dropped, the then Transport Minister, Lord Adonis, is reported to have stated that this was because road pricing lacked a democratic mandate (Wright, 2009). Moreover, representative survey evidence indicated that there was public support (61 per cent) for road pricing, if the revenue was invested in public transport (Ipsos MORI, 2010). The concern for political elites is that e-petitions (and more direct forms of democracy in general) could lead to reactionary, lowest common denominator policy-making where tough policy choices are avoided or get dropped because of mob rule (Matthews, 2020). Fair assessment of the impact of e-petitions is difficult because it is hard to disentangle what impact the e-petition had when many other activities are typically happening; government-led e-petitions can become a political football amongst politicians; and non-governmental e-petitions have a vested interest in arguing that their actions have been successful, not least because they want to encourage people to keep acting, and they are in competition with each other for donations and support. Government-led e-petitions are often criticized for being little more than public relations exercises, designed to garner large e-mail databases that can be used to lobby people (Snider, 2013). The Queensland state-level e-petition system, for example, has been found to have minimal impact: ‘little, if any, action is taken on individual petitions other than a ceremonial reading followed by the placement of a printed copy on the relevant minister’s desk’ (Belot, 2013). Criticisms have also been made of some non-governmental organizations for choosing e-petition topics that they believe will recruit new supporters (adding to their email list), rather than their being considered the most socially important or impactful campaign (Wright, 2015, 2016). There is also a danger that e-petition systems can become exploited by extremist groups (Dumas, 2021). Starting with the impact of e-petitions on public policy, research to date has indicated several cases where there has been an apparent impact. Cotton’s (2011, p. 38) analysis of Scottish e-petitions discovered that ‘12.7% of E-petitions were closed as a result of the issues raised being implemented, [indicating] that E-petitions do have the ability to affect policy formulation, that the Scottish Parliament takes E-petitioning seriously, and that E-petitions have the ability to become or change laws’. Cotton then looked at specific cases, such as a petition to provide cancer drugs on the National Health Service (NHS), submitted by the partner of a cancer sufferer who was self-funding a key treatment. This petition received a relatively modest 632 signatures. However, the Scottish system does not just focus on numbers, and the committee invited the petition creator and her husband to give evidence before the committee. On the basis of their evidence, the committee launched an inquiry, which led to the Better Cancer Care report and significant changes to the funding of cancer treatment. The committee’s closing response is worth highlighting at length: I am sure that the committee would like to reflect not only on the positive actions of the Scottish Government but on the indispensable input of the petitioner, Tina McGeever, on

E-petitioning and changing state–citizen engagement  333

behalf of her husband, the late Mike Gray. Without the petitioner and the energy of both individuals directly involved, we would not be seeing the real improvements that I am sure the petition will effect throughout Scotland in respect of patients accessing newly licensed medicines, in the process for considering objectively individual patient treatment requests and in the arrangements for the combination of care that is available to patients. Finally, we should reflect on the fact that all of those real improvements for people throughout Scotland have been effected through the simple process of lodging a petition. The petitioner should take great pride in that. (Scottish Parliament Public Petitions Committee, 2011 cited in Cotton, 2011)

On the basis of his analysis, Cotton (2011, p. 48) concludes: It is clear that the E-petition process has allowed individuals such as Tina McGeever, Rajiv Joshi, Deryck Beaumont, and hundreds of others to participate in policy formulation with the Scottish Parliament. It has facilitated public debate with the Parliament, and given a new outlet for citizens and groups to voice their grievances and concerns. Increasing public participation in the democratic process was one of the goals of the new Scottish Parliament, and the development and use of its E-petitioning system has fulfilled this function. The case studies have demonstrated the tangible changes in policy that E-petitioning can bring, and further cements its critical role in creating a participative Parliament.

Moving away from Scotland, Wright (2012) cites a range of examples from Downing Street e-petitions, such as the dropping of the fledgling road pricing policy proposal, stopping hospitals from using revenue-producing premium phone lines for contacting them, and a national holiday to commemorate troops. Other successful petitions were not specifically policy-related, making it easier for the government to act, such as a pardon for the famous scientist and code-breaker, Alan Turing. Birrell (2012, pp. 121–122) cites some examples from the Welsh Assembly, including a petition that led to increases in funding for disabled children, and one that led to the adoption of a mandatory charge for plastic bags. Similar to Scotland, petitions typically lead to debates and reports, which can then influence policy. Taking a broader approach, Blumenau (2021, p. 911) finds that when constituent opinion expressed in e-petitions is strong, MPs are more likely to support the issue in Parliament, suggesting ‘stronger representational ties between citizens and politicians’. As shown above, there are a number of examples of e-petitions that have impacted upon policy. However, it must be recognized that the vast majority of e-petitions have little or no impact. While with the Downing Street model several thousand official responses were made, content analysis has shown that these were largely dismissing or correcting e-petitioners, with no evidence of policy impact (Wright, 2012). Perhaps unsurprisingly, petition creators were often very negative about the official replies. While this is not a policy impact, it is a form of political engagement. Interestingly, many petition creators had a positive view about the impact of their petition, precisely because they did not expect it to have an impact and measured success with a lower, broader bar (Wright, 2015). In fact, several petition creators felt that the government was more open and responsive privately than what was communicated in the official replies. It is also worth noting that a UK parliamentary

334  Handbook of digital politics

committee decides whether a petition that receives the required 100,000 signatures is actually debated in Parliament, and Bochel (2013, p. 804) found that only five of the eight petitions that achieved 100,000 signatures had been debated. Moreover, the actual debates often did not match the text of the petition. In perhaps the most famous example, a petition asking for those convicted of participating in the 2011 London riots to lose their welfare benefits was debated in Parliament for three hours, but the debate actually failed to cover the substantive issue of cutting benefits (Jackson, 2011). The success of non-government petitions is also often conceptualized as attaining a pre-prescribed goal or threshold of signatories. Using these kinds of measures means a large majority of online petitions are not successful. However, seeing success or impact only in terms of signature thresholds or policy change is limited and may not be the indicator of success that petition creators even seek (Wright, 2015). As noted above, to broaden the idea of success we could also include how non-government petitions: increased publicity, particularly via media coverage; reached a target number of signatures to galvanize support; created a public statement and/or attention to reflect how much people care about an issue; and gained attention of political elites, even if the issue of concern was not directly resolved. Non-governmental e-petitions have been credited with impacting policies (e.g. stopping the closure of BBC Radio 6 Music and reversing a policy to sell off national forests in the UK), but these are exceptions. Recent analysis of a dataset of over 17,000 Change.org petitions found that only 10 per cent of closed petitions were officially announced as a victory in gaining a response or recognition from their target (Halpin et al., 2018), but many more were shared over social media and/or gained legacy media attention. Further, Minocher (2019) has analysed how consumer activists have successfully used Change.org to challenge corporate business practices. There is evidence that the novelty of using e-petitions to challenge business may produce a higher level of success than those that target government alone.

CONCLUSION While e-petitions remain controversial, empirical analysis suggests that they are popular with citizens and have had impacts on public policy. However, there are significant variations between e-petition platforms, and how they are designed and institutionalized is crucial. For example, well-resourced, substantive e-petition systems such as in Scotland have had clear impacts. However, even descriptive e-petition platforms, such as Downing Street and direct.gov, have had an influence; although there remain concerns about the normative desirability of this, not least because of inequality of access. A key difference is that the Scottish system is fully embedded into the parliamentary process, and petitions are taken forward by a panel of Members of the Scottish Parliament (MSPs) who understand how the system works. This also seems to make it easier to identify discrete examples where e-petitions have made a difference. Nevertheless, it must be recognized that the majority of e-petitions

E-petitioning and changing state–citizen engagement  335

make no impact. However, petition creators were often still quite positive about their petitions because they were using broader definitions of success.

PRIORITIES FOR FUTURE RESEARCH E-petitions continue to change rapidly. As this chapter has highlighted, a number of important studies of e-petitions now exist, and there has rightly been a focus placed on platforms such as Change.org. There remain, though, several gaps in the literature that could be usefully targeted in future research. Transparency, Apps and Big Data One relatively new field is apps and visualization widgets that can be used to help make sense of, and promote, e-petitions. The best example of this was featured on the now defunct White House e-petition platform, We The People, which had an API and Hackathon to allow people to develop the capacity of the platform (Welch, 2013). However, we have also seen the removal of some APIs and this openness remains in flux. Policy Impact While there has already been some research that has attempted to assess the policy impact of e-petitions, this has been relatively limited in scope, focused on a small number of cases, and has struggled to disentangle what role the e-petition played in successful broader campaigns. There is also a lack of research on how both petition signers, and people who disagree with them, perceive ‘success’ and ‘failure’. Non-Governmental E-Petitions While there has been some research on petition platforms such as Change.org, this remains relatively limited. How such platforms are used to organize and mobilize would be of interest. While some researchers have been able to scrape data from large petition platforms this is now increasingly difficult as organizations such as Change.org close down access to their APIs. It would also be interesting to explore whether the rise of petition platforms has driven change amongst institutional petition systems.

FURTHER READING Asher, M., Leston Bandeira, C. and Spaiser, V. (2019). Do parliamentary debates of e-petitions enhance public engagement with Parliament? An analysis of Twitter conversations. Policy and Internet, 11(2), 149–171.

336  Handbook of digital politics

Leston-Bandeira, C. (2019). Parliamentary petitions and public engagement: An empirical analysis of the role of e-petitions. Policy & Politics, 47(3), 415–436. Leys, C. (1955). Petitioning in the nineteenth and twentieth centuries. Political Studies, 3(1), 45–64. Rosenberger, S., Seisl, B., Stadlmair, J., and Dalpra, E. (2022). What are petitions good for? Institutional design and democratic functions. Parliamentary Affairs, 75(1), 217–237. Wright, S. (2012). Assessing (e-)democratic innovations: “Democratic goods” and Downing Street e-petitions. Journal of Information Technology and Politics, 9(4), 453–470.

REFERENCES Asher, M., Leston Bandeira, C. and Spaiser, V. (2019). Do parliamentary debates of e-petitions enhance public engagement with Parliament? An analysis of Twitter conversations. Policy and Internet, 11(2), 149–171. Belot, H. (2013). Petitions are in vogue, but do they make a difference? Citizen. http://​ www​.thecitizen​.org​.au/​analysis/​petitions​-are​-vogue​-do​-they​-make​-difference​#sthash​ .hKVKGFSZ​.dpuf. Birrell, D. (2012). Comparing Devolved Governance. Basingstoke: Palgrave Macmillan. Blair, T. (2007). The e-petition shows that my government is listening. The Guardian, 18 February. http://​www​.theguardian​.com/​commentisfree/​2007/​feb/​18/​uk​.transport. Blumenau, J. (2021). Online activism and dyadic representation: Evidence from the UK E-petition system. Legislative Studies Quarterly, 46(4), 889–920. Bochel, C. (2013). Petition systems: Contributing to representative democracy? Parliamentary Affairs, 66(4), 798–815. Bochel, C. and Bochel, H. (2017). ‘Reaching in’? The potential for e-petitions in local government in the United Kingdom. Information, Communication & Society, 20(5), 683–699. Carman, C. J. (2010). The process is the reality: Perceptions of procedural fairness and participatory democracy. Political Studies, 58(4), 731–751. Chadwick, A. (2013). The Hybrid Media System. New York: Oxford University Press. Chadwick, A. and Dennis, J. (2017). Social media, professional media and mobilisation in contemporary Britain: Explaining the strengths and weaknesses of the citizens’ movement 38 Degrees. Political Studies, 65(1), 42–60. Cotton, R. D. (2011). Political participation and e-petitioning: An analysis of the policy- making impact of the Scottish Parliament’s e-petition system. Unpublished thesis, University of Central Florida. http://​etd​.fcla​.edu/​CF/​CFH0004083/​Cotton​_ Ross_D_201112_BA.pdf. Dumas, C. (2021). Electronic petitioning as online collective action: Examining the e-petitioning behavior of an extremist group in We The People. Information Polity, 26(4), 477–499. Escher, T. and Riehm, U. (2016). Petitioning the German Bundestag: Political equality and the role of the internet. Parliamentary Affairs, 70(1), 132–154. Halpin, D., Vromen, A., Vaughan, M., and Raissi, M. (2018). Online petitioning and politics: The development of Change.org in Australia. Australian Journal of Political Science, 53(4), 428–445. Hansard Society (2016). Audit of Political Engagement 13: The 2016 Report. London: Hansard Society. Hersh, E. D. and Schaffner, B. F. (2018). Postmaterialist particularism: What petitions can tell us about biases in the policy agenda. American Politics Research, 46(3), 434–464. Ipsos MORI (2010). Support for road pricing if revenues used for public transport. Research archive. http://​www​.ipsos​-mori​.com/​r​esearchpub​lications/​researcharchive/​ poll. aspx?oItemId=234.

E-petitioning and changing state–citizen engagement  337

Jackson, M. (2011). MPs ‘ignore’ riots e-petition in Westminster debate. https://​www​.bbc​ .com/​news/​uk​-politics​-15283837. Jungherr, A. and Jürgens, P. (2010). The political click: Political participation through e-petitions in Germany. Policy and Internet, 2(4), 131–165. Karpf, D. (2010). Online political mobilization from the advocacy group’s perspective: Looking beyond clicktivism. Policy and Internet, 2, 7–41. Karpf, D. (2016). Analytic Activism. Oxford: Oxford University Press. Labour Party (2005). Forward Not Back: The Labour Party Manifesto 2005. London: Labour Party. Leston-Bandeira, C. (2019). Parliamentary petitions and public engagement: An empirical analysis of the role of e-petitions. Policy & Politics, 47(3), 415–436. Lindner, R. and Riehm, U. (2011). Broadening participation through e-petitions? An empirical study of petitions to the German Parliament. Policy and Internet, 3(1), 1–23. Margetts, H., John, P., Hale, S., and Yasseri, T. (2015). Political Turbulence: How Social Media Shape Collective Action. Princeton, NJ: Princeton University Press. Matthews, F. (2020). The value of ‘between-election’ political participation: Do parliamentary e-petitions matter to political elites? The British Journal of Politics and International Relations, 23(3), 410–429. Minocher, X. (2019). Online consumer activism: Challenging companies with Change.org. New Media & Society, 21(3), 620–638. Morozov, E. (2009). The brave new world of slacktivism. Foreign Policy, 19 May. http://​ neteffect​.foreignpolicy​.com/​posts/​2009/​05/​19/​the​_brave​_new​_world​_of​_slacktivim. Puschmann, C., Toledo Bastos, M., and Schmidt, J.-H. (2017). Birds of a feather petition together? Characterizing e-petitioning through the lens of platform data. Information, Communication & Society, 20(2), 203–220. Rosenberger, S., Seisl, B., Stadlmair, J., and Dalpra, E. (2022). What are petitions good for? Institutional design and democratic functions. Parliamentary Affairs, 75(1), 217–237. Rugg, S. (2019). How Powerful We Are: Behind the Scenes with One of Australia’s Leading Activists. Sydney: Hachette. Scottish Parliament (2010). Petitioning the Scottish Parliament: Making Your Voice Heard. Edinburgh. http://​www​.scottish​.parliament​.uk/​PublicI​nformation​documents/​ Petitioning-Eng-250712.pdf. Shulman, S. (2009). The case against mass e-mails: Perverse incentives and low quality public participation in US federal rulemaking. Policy and Internet, 1(1), 23–53. Smith, G. (2009). Democratic Innovations: Designing Institutions for Citizen Participation. Cambridge: Cambridge University Press. Snider, J. H. (2013). Updating Americans’ First Amendment right to petition their government. Paper presented at the Harvard Law School Luncheon, 20 September. Vaughan, M., Vromen, A., Porten-Chee, P., and Halpin, D. (2022). The role of novel citizenship norms in signing and sharing online petitions. Political Studies. https://​doi​.org/​10​ .1177/​00323217221078681. Vromen, A. (2017). Digital Citizenship and Political Engagement: The Challenge from Online Campaigning and Advocacy Organisations. London: Palgrave Macmillan. Welch, C. (2013). White House holding hackathon to enhance ‘We the People’ petition platform. https://​www​.theverge​.com/​2013/​2/​5/​3956076/​white​-house​-holding​-hackathon​-we​ -the​-people​-petitions. Wong, S. C. and Wright, S. (2018). Generating a voice among ‘media monsters’: Hybrid media practices of Taiwan’s anti-media monopoly movement. Australian Journal of Political Science, 53(1), 89–102. Wong, S. C. and Wright, S. (2020). Hybrid media mediation opportunity structure? A case study of the anti-national education movement. New Media & Society, 22(10), 1741–1762.

338  Handbook of digital politics

Wright, R. (2009). Adonis shelves road-price policy. Financial Times, 25 June. http://​www​.ft​ .com/​intl/​cms/​s/​0/​0ebf2b32​-6120​-11de​-aa12​-00144feabdc0. Wright, S. (2006). Government-run online discussion fora: Moderation, censorship and the shadow of control. British Journal of Politics and International Relations, 8(4), 550–568. Wright, S. (2012). Assessing (e-)democratic innovations: “Democratic goods” and Downing Street e-petitions. Journal of Information Technology and Politics, 9(4), 453–470. Wright, S. (2015). Populism and Downing Street e-petitions: Connective action, hybridity and the changing nature of organizing. Political Communication, 32(3), 414–433. Wright, S. (2016). ‘Success’ and online political participation: The case of Downing Street e-petitions. Information, Communication & Society, 19(6), 843–857.

22. From Valencia filters to #BlackOutTuesday: collective action on Instagram Elena Sotelo-Prol

INTRODUCTION This chapter aims to provide a theoretical introduction to digital protest in a new social media era, dominated by visually-driven platforms such as Instagram. With a billion monthly active users, Instagram has established itself over the years as one of the most influential social media platforms globally, attracting a comparatively younger market than other tech giants such as Facebook (We Are Social, 2022). Despite this, the body of theoretical and empirical scholarship on Instagram as a site for politics – both contentious and institutional – is scarce. As such, this chapter aims to provide a critical introduction to the phenomenon of digital protest on Instagram. To do so, the present chapter is divided into four sections. First, we set the scene by reviewing how the changes to the media system since the beginning of the twenty-first century have impacted contentious politics as a phenomenon, and the relevance of digitally enabled protest as an area of inquiry within (digital) political communication. Second, a literature review of online collective action is provided, establishing its evolution and contextualizing contentious politics within the extant scholarship and key debates on digital protest. Afterwards, a theoretical framework for the study of collective action on Instagram is established, exploring how the platform’s network media logic and affordances shape the development of online protest movements. This framework is structured through four key affordances: visibility, visuality, shareability, and performativity. To conclude, this chapter outlines a research agenda for the study of collective action on Instagram, suggesting further avenues of empirical research into this new phenomenon.

LOCATING PROTEST MOVEMENTS IN A NEW DIGITAL ERA Since the turn of the century, the use of social media for political participation has revolutionized our understanding of how protest movements can be undertaken. Across the world, digital technologies have radically transformed the way that activists, politicians, and citizens alike discuss politics, thus creating innovative new opportunities for digital engagement. More than twenty years into the twenty-first 339

340  Handbook of digital politics

century, social media has become a cornerstone in the development of protest movements around the world. From the Arab Spring to March for our Lives, information and communication technologies have played an extremely important role in defining social movements, expanding the boundaries and possibilities of collective action in previously unthinkable ways. Most notably, the use of digital technologies has shaped the development of protest movements by making them global, affording more people across the world the opportunity to engage with contentious politics regardless of where they are (Margetts et al., 2016). This revolution in how protest movements are being carried out in the digital public sphere has been a keen object of study for political communication scholars over the last decade. However, the ways that academics within this field assess online protest movements have diverged considerably, with such analyses ranging from the optimistic to the apprehensive. On the one hand, social media has been celebrated for enabling exciting new avenues for contentious politics. These developments have given rise to a new genre of digital protest – known as connective action – which boasts unique organizational principles and mobilization strategies, ultimately enabling more democratic and accessible political engagement (see Bennett and Segerberg, 2012; Shirky, 2009). On the other hand, some critical scholars have been more cautious in their assessment of digital protest, arguing that these new types of collective action can encourage low effort, low impact political participation (Morozov, 2011). Moreover, attention has been drawn to the ideological gap between online protest movements and the corporate capitalist structures that support them (Leistert, 2015; Poell, 2014). The Black Lives Matter movement and its strive for racial equality can be highlighted as an example of the success that protest movements are currently achieving within the digital media ecology. Although Black Lives Matter as an organization can be traced back to 2013 (Lebron, 2017), the movement gained unprecedented levels of support during May and June 2020, following the murder of George Floyd while in police custody in Minnesota, US. Despite the restrictions in place at the time because of the Coronavirus pandemic, the protests – which denounced police brutality and systemic, institutional violence against black bodies – that took place during those months have been widely recognized as the largest demonstrations in US history (Bolsover, 2020; Buchanan et al., 2020). However, this was not an isolated phenomenon. Instead, these mobilizations sent shockwaves across the world, prompting subsequent demonstrations in countries as disparate as the United Kingdom, Brazil, and Japan, among others. The example of Black Lives Matter demonstrates the sustained relevance and establishment of social media as tools for collective action in contemporary society. In comparison to other protest movements, what made the 2020 Black Lives Matter demonstrations unique were participants’ exceptional ability to overcome the challenges brought about by Covid-19 using digital technologies, thus leading to outstanding volumes of both online and offline political action. During the protests’ most active period, social media platforms served not only as organizational tools, but also as venues of civic discussion, engagement, and identity-building. As such,

Collective action on Instagram  341

Black Lives Matter exemplifies how the specific architecture of network media platforms is now – more than ever – shaping the development of increasingly diverse genres of digital protest, as different technologies are purposefully deployed by participants to achieve different goals (Poell, 2014). It is within this context that a unique collective action cybersphere was able to bloom, as citizens engaged in a global discussion of racism and policing powers, as evidenced by the outstanding volumes of user-generated content being shared across different social media platforms. Between 26 May and 7 June 2020, almost 3.7 million daily tweets were recorded under the #BlackLivesMatter hashtag (Pew Research Centre, 2020). Most notably, another platform rose to notoriety as a space for digital engagement with the movement during this period: Instagram. The use of this platform during the demonstrations was simultaneously influential and controversial, as some activists criticized engagement on Instagram as performative at best, and ‘often unhelpful and meaningless’ (Metta, 2020). Since the academic interest in digitally mediated protest first arose during the early 2010s, following events such as the Arab Spring and Occupy Wall Street, the online environment in which collective action is currently taking place has experienced significant technological, cultural, and structural changes. The Internet is a radically different ecology than it was a decade ago, boasting new digital cultures, social norms, and architectures, with brand new platforms supporting its biggest userbase thus far. Most importantly, and despite all these changes to the digital media ecology, the study of online protest remains grounded in the same platforms as during the first wave of digital activism, as most of the extant scholarship is still dominated by Twitter, Facebook, and YouTube as sites of contentious politics. Despite the sustained influence of such platforms, a need has arisen within political communication scholarship to consider how new information and communication technologies are shaping the development of digitally enabled protest. Moreover, new theoretical frameworks ought to be applied to empirical research to gain fresh insights into how protest is being negotiated and carried out in the contemporary digital public sphere.

THE STRUCTURAL CHANGES TO COLLECTIVE ACTION ON SOCIAL MEDIA In order to better understand the current state of digital protest and how it plays out on Instagram, it is important to briefly review the development of social media and collective action over the last decade, as well as some key observations from the literature on how technological advancement has shaped the deployment of digital protest tools and strategies. In reviewing the development of social media as platforms sustaining collective action, the notoriety of the protest movements which took place during 2011 ought to be acknowledged. Framed in a context of financial recession, increasing worldwide connectivity and rapid technological advancement, this year saw individuals across the world come together in an unprecedented struggle against oppression and

342  Handbook of digital politics

injustice (Nichols, 2011). However, it was not only the vast number of individuals taking part in protest movements that marked a turning point in contemporary understandings of political communication: instead, what widespread engagement with contentious politics during 2011 demonstrated was how emerging network media platforms could be deployed for collective action. Protest movements such as Occupy Wall Street, Indignados and the Arab Spring captured scholarly interest due to their ability to employ network media in subversive ways, challenging their corporate logic and profit-driven political economy, and utilizing them instead as organizational and participatory tools. Such was the influence of digital technologies that commentators, in a clear display of what Gerbaudo (2012) refers to as ‘Twitter fetishism’ (p. 5) hailed them as Twitter or Facebook ‘revolutions’ (Wolfsfeld et al., 2013). However, scholars such as Christensen (2011) argue that such claims ought to be critically examined, as the use of social media for collective action is a much more nuanced phenomenon. Instead, Christensen (2011) argues that the outcomes of digitally enabled protest are not solely positive, as it has considerably contributed to the normalization of digital technologies as tools for ‘state surveillance and repression’ (p. 155). Nevertheless, this first wave of demonstrations has become foundational in the study of digital activism, thus setting the groundwork for the study of collective action in the digital media ecology in terms of both methods and theoretical frameworks. Before delving into how collective action is being negotiated and developed on Instagram, it is important to lay a base understanding of what digitally enabled protest means, and what research during the 2010s has taught us about this phenomenon. First, it is essential to set out a working definition of collective action, and how this concept can be translated into the digital media ecology. Among the scholars examining collective action, Melucci’s (1996) seminal work can be contextualized within the sociological analysis of social movements. As conceptualized by Melucci (1996, p. 20), collective action is: a set of social practices, involving simultaneously a number of individuals or groups, exhibiting similar morphological characteristics in contiguity of time and space, implying a social field of relationships and the capacity of the people involved of making sense of what they are doing.

Although useful insights can be appreciated in Melucci’s (1996) definition, the degree of transformation that collective action has experienced due to the advent of network media calls for an overhaul – at least partial – of scholarly definitions of this phenomenon. Thus, a more updated review of collective action in the digital era will ensue, grounded in recent scholarship and focusing on the newer logic of digital protest on social media. In relation to collective action, the use of social media has challenged the traditional boundaries of what it means to engage in protest movements, enabling more people to creatively participate in politics. Such new mechanisms of participation are located within what Jenkins (2009) refers to as participatory culture, as the use of the Internet

Collective action on Instagram  343

opens up and further democratizes pathways of (political) engagement. According to Margetts et al. (2016), the digitalization of protest movements has allowed citizens to engage in contentious politics through tiny acts of political participation, which they describe as ‘micro-donations of time and effort to political causes’ (p. 35) that require less commitment than traditional forms of contentious politics. However, the extent to which these new participation strategies constitute genuine engagement with protest movements has long been contested within digital politics scholarship. As previously outlined, critics such as Morozov (2011) have argued that digitally enabled protest constitutes a phenomenon known as ‘slacktivism’. Thus, attempts to engage in collective action through social media are not impactful, and ultimately result in citizens believing they are contributing to political change, which in turn reduces their motivation to ‘genuinely’ – understood as offline – engage with protest movements (Kwak et al., 2018; Morozov, 2011). Alternatively, scholars such as Hsiao and Yang (2018), Kwak et al. (2018) and Lane and Dal Cin (2018) have been more optimistic regarding the users for online collective action to mobilize more participants, and even engage previously disaffected citizens. Moreover, the accessibility and ease of engaging with contentious politics on social media can encourage more people to take an active role in politics, with this effect being observed in relation to both online and offline participatory behaviour (Kwak et al., 2018). Regarding commitment, Hsiao and Yang (2018) argue that users engaging with collective action on social media can be equally as committed as their offline counterparts, while providing a less risky avenue for citizens to engage in collective action. In contrast with traditional protest movements, engagement with digitally enabled collective action has lower participation costs, which may encourage citizens to take a more active role in demonstrations. As observed by Bennett and Segerberg (2012) and Margetts et al. (2016), the communication and mobilization costs of collective action have been radically reduced by the use of social media. This change has affected not only individual participants, but social movement organizations as well, who can simultaneously save resources and widen their scope in mobilizing potential through network media (Mundt et al., 2018). However, more critical scholars have pointed out that the cost reduction associated with online collective action has not succeeded in democratizing the accessibility of contentious politics, as the digital divide results in a class-centric gap regarding who gets to engage with digitally mediated protest and benefit from such reduction in costs (Schradie, 2018). Another key contributing element to the increase in the number of people engaging in online collective action is the more flexible nature of social media activism, which is less ideologically restrictive than traditional protest movements. According to Bennett and Segerberg (2012), participation in online protests takes place through personalized action frames. Because this genre of engagement demands less ideological identification and commitment, users can take part in a multitude of social movements on looser terms, which ultimately results in instances of collective action which ought to be understood as ‘an expression of personal hopes, lifestyles, and grievances’ (Bennett and Segerberg, 2012, p. 742). Nevertheless, Milan (2015)

344  Handbook of digital politics

notes that research into digitally enabled protest should not undermine the influence of collective identity in relation to the articulation of online collective action. In an age where information is being constantly shared across the globe, a self-actualizing notion of collective identity is essential to make sense of the endless stream of information framed within digitally mediated protest. As a result, Milan (2015) introduces the concept of ‘cloud protesting’, which refers to ‘a metaphor for a way of connecting individuals in an instance of collective action which is specific to the age of social media’ (p. 893). In contrast to previous conceptualizations of collective action raised by scholars such as Melucci (1996), contemporary understandings of online protest emphasize how social media’s instantaneous communication affordances (Klinger and Svensson, 2015a, 2015b) challenge traditional geographic and time restrictions to collective action. The increasing irrelevance of such limitations has been evidenced by the rise of protest movements on a global scale, such as Black Lives Matter and #MeToo. This phenomenon has been enabled by the fact that online instances of collective action are being developed – at least partly – in an imagined digital space (Milan, 2015). As pointed out by Poell (2014), digitally mediated protest is not a monolithic phenomenon, comprising instead a complex network of interrelated online activities covering virtually the entirety of the digital media ecology. The different platforms across which online collective action spans play different roles within protest movements (Valenzuela et al., 2018), but remain connected through activists’ sharing practices. For instance, Twitter was used as a real-time communication and organizational tool during the 2010 Toronto G20 demonstrations, while YouTube served as a multi-media informational hub covering the action and violence on the streets (Poell, 2014). Most importantly, Poell (2014) points out that social media platforms are ‘not neutral tools, as they are always already entangled in complex techno-cultural and political economic relations, from which they cannot be analytically separated’ (p. 717). Thus, it is important to take into consideration how each platform’s logic and structure shapes the affordances, which in turn influence their use within online collective action. To conclude, a key conceptual debate regarding the nature of digitally enabled protest ought to be addressed. As argued by Bennett and Segerberg (2012), the extent to which collective action has been transformed by the use of social media has resulted in the emergence of a different phenomenon, which they refer to as connective action. In developing this theoretical framework, Bennett and Segerberg (2012) make a key distinction between three types of activist networks, which operate through different logics: collective action, hybrid connective action networks, and self-organizing connective action networks. As previously discussed, collective action relies on individuals constructing a shared collective identity over time, ultimately working towards a common goal with the support of a social movement organization (Bennett and Segerberg, 2012; Mundt et al., 2018). On the other hand, the logic of connective action operates through collaboration and self-expression, as digital media plays the role of the organizing agent, thus connecting participants

Collective action on Instagram  345

and facilitating resource mobilization (Bennett and Segerberg, 2012). However, Milan (2015) stresses that the role of social movement organizations ought not to be undermined within online collective action, as tangible leadership within digitally mediated protest is still significant to ensure a lasting, impactful, and coherent movement. For the purpose of readability, the remainder of this chapter will use the term ‘collective action’ to simultaneously refer to the traditional conceptualization of this phenomenon, as well as the contemporary understanding of ‘connective action’ as raised by Bennett and Segerberg (2012).

COLLECTIVE ACTION ON INSTAGRAM As discussed in the introduction, the latter half of the 2010s saw the emergence of Instagram as an increasingly influential venue for the development of collective action. In particular, employing Instagram to engage in digitally enabled protest constitutes a subversion of the platform’s normative uses. Initially developed as a site to share photography and videos, the evolution of Instagram over the years has seen the integration of new communicative features which have significantly transformed our use and understanding of the platform (Leaver et al., 2020). As such, these emerging platform logics and affordances have provided users across the world with new and ingenious pathways through which to employ Instagram as a resource for contentious politics. Since then, Instagram has been deployed as part of protest movements such as Black Lives Matter and Fridays for Future, with users employing the platform as a mobilization, communication, and even participatory tool (Einwohner and Rochford, 2019; Mundt et al., 2018; Yueng and Tang, 2021). Overall, Instagram ought to be conceptualized as a dynamic and multi-modal platform for collective action, which englobes significantly diverse features that enable the development of contentious politics across the digital media ecology. In order to construct a theoretical framework of how Instagram is currently being employed as a venue for online collective action, it is essential to engage with the scholarly discussions on network affordances and the impact of the material qualities of digital artefacts on digitally enabled protest. As described by Klinger and Svensson (2015a, 2015b), network media logic is conceptualized as the architecture and embedded norms that shape how social media platforms are employed for (political) communication. The logic of network media is strictly distinct from that of traditional mass media, yet remains closely intertwined as both loci coexist within the hybrid media system (Chadwick, 2017). Within the contemporary digital media ecology, network media is articulated through the logic of production, the logic of distribution, and the logic of usage (Klinger and Svensson, 2015a, 2015b). As such, network media affordances can be understood as the features of social media which shape the different ways that users employ the platform (Lievrouw, 2014). However, it must be noted that such discussions of the material affordances of network media have been critiqued in the scholarship for being overtly techno deterministic. However, a more nuanced approach to this debate ought to acknowledge

346  Handbook of digital politics

the two-way relationship between users and digital technology (Lievrouw, 2014), as they both influence each other grounded in processes of co-production which ‘significantly contribute to structure modes of interaction and relationships’ (Milan, 2015, p. 891). Having established the conceptual framework for the analysis of collective action on Instagram, we now focus on articulating the platform’s main network media affordances and their impact on digital protest. As outlined in the introduction, such affordances are as follows: visibility, visuality, shareability, and performativity. Instagram’s Visibility: How ‘Seeing’ Protest Online Contributes to Collective Action Within the context of Instagram’s network affordances, the notion of visibility refers to the platform’s potential for automated content dissemination, a key element of its digital architecture with which users interact and experience when navigating the platform. For instance, the idea of Instagram’s visibility refers to the algorithmic logic through which content is presented to users through their feed, or the content that Instagram shows to users through the ‘explore’ feature. In relation to collective identity, this visibility can translate into more learning opportunities, with digital protest’s potential for mobilization being furthered by the hyper-targeted dissemination of action frames and collective identities. As argued by Milan (2015, p. 895), such visibility within social media ‘indicates the virtual embodiment and online manifestation of grounds and individuals and of the associated meaning, which are … relentlessly negotiated, bolstered, and updated’. However, how the principle of visibility operates within Instagram – as it happens with most other network media household names – is not straightforward to grasp. Ultimately, the visibility of content within Instagram is ruled by an algorithmic logic that is not transparent to users and activists. In relation to this obscure architecture we ought to acknowledge the implicit contradiction in the use of corporate, profit-driven social media platforms for the pursuit of contentious politics (Etter and Albu, 2021). Within this context, participants in collective action are forced to operate on the platform’s terms, partly relinquishing their agency to employ Instagram to pursue collective action, as users are unable to fully control whom their message is going to reach, and on what terms. As a platform, Instagram does not show content in chronological order. Instead, content dissemination – and hence, the visibility of content – is shaped by the platform’s algorithm, thus prioritizing content that it regards of interest for users based on their prior interactions. As such, Instagram considers the user’s previous activity – such as likes, comments, and saved posts – in order to organize the feed according to which content is likely to generate more engagement (McLachlan, 2022; Warren, 2021). In relation to online collective action, the algorithmic visibility through which Instagram operates has two main consequences. On the one hand, Instagram can hinder the development of online protest movements by invisibilizing activists’ posts. Despite the Internet’s promise of democratizing communication and providing

Collective action on Instagram  347

a space for marginalized voices, this is not always the case for discourse that goes against network media’s corporate logic and interests (Dahlberg, 2005). The phenomenon of ‘shadowbanning’ is an example of this issue, as it refers to instances where ‘without notice or explanation, a user’s post(s) is prevented from appearing in different spaces on the platform, making the content much less likely to reach non-followers’ (Cotter, 2021, p. 2). This uneven power dynamic further accentuates the oxymoron of employing corporate and profit-driven platforms such as Instagram to engage in collective action, as a conflict of interest is due to emerge between both, which can ultimately hinder the healthy development of contentious political movements. On the other hand, because Instagram prioritizes content that users have shown prior interest in, this platform can provide users with new learning opportunities, mobilizing them and ultimately connecting them with other activists across the digital media ecology. However, it must be noted that this can also potentially lead to the creation of echo chambers and filter bubbles. Although the extent to which these phenomena are present on social media has long been a highly contested topic of discussion within political communication scholarship (for an in-depth review of the literature on this topic, see Dahlgren, 2021), their potential influence when it comes to collective action on Instagram cannot be overlooked. Ultimately, the interaction with solely like-minded content and individuals on Instagram can result in a skewed and polarizing view of social reality, as participants are reaffirmed in their beliefs by only being shown messages that they already have a prior ideological affinity with (Dahlgren, 2021; Kanai and McGrane, 2021). Although some research has been carried out on how filter bubbles operate within activist circles online (e.g., Kanai and McGrane, 2021), there is a significant gap in research regarding how this phenomenon manifests on Instagram and how it is shaped by the platform’s algorithmically driven visibility logic. Instagram’s Visuality: Multi-Modal Activism in the Digital Media Ecology As a platform, Instagram is heavily focused on sharing audio-visual content, thus locating its visuality as an integral element of its identity. From its very conception, the central logic behind the use of Instagram has been based on sharing images and videos, which to this day remains the platform’s main feature. The predominantly visual elements and affordances of Instagram thus lend themselves to a particular genre of online collective action, one which emphasizes the material aspects of digitally mediated protest in the construction of collective identity. The use of visual elements is not new to the development of protest movements, as Poell (2014) emphasizes how they tend to focus on providing a public spectacle. For instance, banners and dances have long been part of (offline) collective action as a way to attract attention and nurture a shared feeling of collectivity (Einwohner and Rochford, 2019; Milan, 2015). By employing a visual platform such as Instagram for online collective action, users are engaging in a process that weaves together the material and symbolic

348  Handbook of digital politics

elements of protest (Milan, 2015), resulting in the construction of shared meanings and understandings of the movement. Thus, participants employ similar multi-media content and formats to publicly negotiate their engagement with the movement, as well as the terms upon which this participation can take place. In her seminal work on the materiality of social media and its implications for collective action, Milan (2015) stresses how the visual elements of social media cannot be separated from the platform itself, as ‘the latter gives shape to the former, which cannot occur in the same form outside the specific frame of social media platforms’ (p. 890). Within the unique context of Instagram, the visual formats afforded by the platform give shape to the content being shared, and in turn define the semiotic construction of the movement and the processes of collective identity actualization. For instance, the possibility of sharing several images within a single Instagram post – widely known as a ‘grid’ – influences the way activists engage with protest movements on Instagram, as users now tend to share visual-heavy infographic slideshows with their followers. This particular example highlights how the visual elements and appeal of Instagram operate in conjunction with activists’ efforts to spread their discourse, thus leading to the co-construction of meaning and dissemination of action frames. Instagram’s Shareability: Expanding Collective Action’s Mobilization Possibilities Another key element that shapes the development of collective action on Instagram is the platform’s shareability affordances, as content can easily be replicated and disseminated across different networks of users through the platform’s native communication features. As a platform, Instagram revolves around sharing audio-visual content, with such practices including posting on the main feed or Instagram stories, as well as directly sharing posts with other users through private messaging. Overall, these three features can aid in the mobilization of collective action participants, as they allow users to share key information and resources. In particular, the ability to repost content shared by other users in one’s Instagram stories can prove particularly meaningful for online collective action. This feature, which was added in 2016, can aid users in furthering the reach of protest-related messages, thus educating others in their network and potentially mobilizing new participants. Although much research is yet to be carried out in relation to this aspect of collective action on Instagram, Einwohner and Rochford (2019) point to teaching and persuading others as the main purpose behind independent participants sharing content related to the Women’s March on Instagram. Instagram’s Performativity: Symbolically Constructing Meaning in Digital Protest Lastly, the combination of material affordances and symbolic meaning-making analysed by Milan (2015) in relation to collective action on social media points to digitally enabled protest on Instagram showcasing a significant degree of performa-

Collective action on Instagram  349

tivity, which manifests through users actively engaging in doing and ‘showing doing’ (Schechner, 2013, p. 28). It must be noted that protest ought to be conceptualized as an inherently performative process, as it entails participants engaging in intently public-oriented collective action repertoires, which aim to symbolically relay a certain message to an audience (Tilly, 2008; Tilly and Tarrow, 2015). By employing Instagram, users are adding a further layer of performativity to their contentious politics repertoires, as they symbolically engage in both collective action and the negotiation of collective identity through digital, highly visual means. Ultimately, sharing protest-related content on Instagram constitutes a performative effort to showcase users’ engagement in collective action to others in their network, thus actively defining their identity, values, and beliefs (Einwohner and Rochford, 2019). An example of performativity within collective action on Instagram can be observed in how users engaged with the Black Lives Matter movement during the 2020 wave of protests. During this period, as a massive number of users actively participated in the movement through the platform, engagement with Black Lives Matter was – at least partly – motivated by users’ desire to symbolically showcase their commitment to furthering the movement through their Instagram posts. This genre of performative engagement was particularly manifest during #BlackOutTuesday, as users across the globe shared black squares on their Instagram feeds in an effort to raise awareness of racial injustice (Monckton, 2020; Speare-Cole, 2020). With no text accompanying these posts – other than the above-mentioned hashtag – this particular instance of online collective action showcases a process whereby users engage in the symbolic co-construction of meaning and protest discourse, thus establishing a sense of collective identity, belonging, and purpose. Although the performativity of this event has been widely criticized (Bakare and Davies, 2020; Jennings, 2020), the symbolic nature of this example ought not to be understood as inherently bad. Instead, it should be seen as a process of collective negotiation of meaning and identity within digitally enabled protest, which knowingly takes advantage of the unique affordances of Instagram as a platform to symbolically further and construct contentious politics online.

A RESEARCH AGENDA FOR COLLECTIVE ACTION ON INSTAGRAM Throughout this chapter, we have examined how the protest ecology on social media has evolved over the last decade, and how prior research into online collective action can inform contemporary understandings of the use of Instagram for contentious politics. However, the subversion of this platform’s normative uses to further protest movements is a phenomenon which remains very much in its infancy, and it is likely to continue to grow over the coming years. Compared to other social media platforms, the scholarship revolving around collective action on Instagram is relatively scarce. However, the political, cultural, and social impact of such instances of digital protest should not be underestimated. As discussed throughout this chapter, Instagram has

350  Handbook of digital politics

successfully established itself as a legitimate avenue for the development of contentious politics, particularly among younger citizens. Overall, this chapter has aimed to provide a framework for the study of Instagram as a platform for collective action, articulated through four main network affordances: visibility, visuality, shareability, and performativity. Based upon this contribution to the literature, anyone interested in empirically expanding on the current scholarship on this phenomenon might consider the following suggestions. In regard to visibility, more research is required to fully grasp how activist networks are constituted and content is being disseminated across Instagram. For this pursuit, a critical approach to the corporate nature and political economy of digital platforms could prove particularly useful. Additionally, understanding the phenomenon of shadowbanning within the context of collective action stands out as particularly important, especially in relation to its potential to hinder the spread of protest-related messages on Instagram. From a methodological standpoint, the use of digital methods – such as network analysis – can prove particularly useful. Ultimately, empirically examining how different types of interactions influence the spread of protest-related content can be an exciting avenue of inquiry for anyone interested in the algorithmic logic of Instagram. In relation to visuality, qualitative-oriented research can provide valuable insights into the type of content shared on Instagram as part of collective action effort. Particular attention should be given to the multi-modal nature of such messages, and how their format influences reception when compared to other text-based network media platforms such as Twitter. Moreover, more research is required to determine how the platform’s innate materiality shapes the process of collective identity negotiation. In terms of shareability, the scholarship on collective action on Instagram could greatly benefit from research on the dissemination of protest-related content on the platform through two different angles. Firstly, qualitative research could shed light on the sharing practices of users on Instagram, and why they decide to repost content revolving around contentious politics. Secondly, a methodological approach grounded on digital methods such as network analysis might contribute to a better understanding of how activist networks are established and interactions fostered through reposting on Instagram stories. To conclude, the lens of political performance can provide a theoretically grounded approach to the empirical study of collective identity. As a complex phenomenon, further research ought to explore how the different layers of performativity within digital protest on Instagram contribute to the development and actualization of participants’ collective identities.

Collective action on Instagram  351

FURTHER READING Bennett, W. L. and Segerberg, A. 2012. The logic of connective action: Digital media and the personalization of contentious politics. Information, Communication & Society, 15(5), 739–768. Einwohner, R. and Rochford, E. 2019. After the march: Using Instagram to perform and sustain the Women’s March. Sociological Forum, 34(1), 1090–1110. Etter, M. and Albu, O. 2021. Activists in the dark: Social media algorithms and collective action in two social movement organizations. Organization, 28(1), 68–91. Klinger, U. and Svensson, J. 2015. Network media logic: Some conceptual considerations. In A. Bruns, G. Enli, E. Skogerbo, A. Larsson, and C. Christensen (eds.), The Routledge Companion to Bruns, A., Enli, G., Skogerbo, E., Larsson, A. and Christensen, C. (eds.) The Routledge Companion to Social Media and Politics. New York: Routledge. Leistert, O. 2015. The revolution will not be liked: On the systemic constraints of corporate social media platforms for protest. In L. Dencik and O. Leistert (eds.), Critical Perspectives on Social Media and Protest: Between Control and Emancipation (pp. 35–52). London: Rowman & Littlefield. Milan, S. 2015. From social movements to cloud protesting: The evolution of collective identity. Information, Communication & Society, 18(8), 887–900. Yueng, S. and Tang, G. 2021. Instagram and social capital: Youth activism in a networked movement. Social Movement Studies. https://​doi​.org/​10​.1080/​14742837​.2021​.2011189; https://​doi​.org/​10​.1080/​14742837​.2021​.2011189.

REFERENCES Bakare, L. and Davies, C. 2020. Blackout Tuesday: Black squares dominate social media and spark debate. The Guardian, 2 June. https://​www​.theguardian​.com/​us​-news/​2020/​jun/​02/​ blackouttuesday​-dominates​-social​-media​-millions​-show​-solidarity​-george​-floyd. Bennett, W. L. and Segerberg, A. 2012. The logic of connective action: Digital media and the personalization of contentious politics. Information, Communication & Society, 15(5), 739–768. Bolsover, G. 2020. Black Lives Matter discourse on US social media during COVID-19: Positions enacted in a new event. Computers and Society. https://​doi​.org/​10​.48550/​arXiv​ .2009​.03619. Buchanan, L., Bui, Q., and Patel, J. 2020. Black Lives Matter might be the largest movement in US history. The New York Times. https://​www​.nytimes​.com/​interactive/​2020/​07/​03/​us/​ george​-floyd​-protestscrowd​-size​.html. Chadwick, A. 2017. The Hybrid Media System: Politics and Power. New York: Oxford University Press. Christensen, C. 2011. Twitter revolutions? Addressing social media and dissent. The Communication Review, 14(3), 155–157. Cotter, K. 2021. ‘Shadowbanning is not a thing’: Black box gaslighting and the power to independently know and credibly critique algorithms. Information, Communication, & Society. https://​doi​.org/​10​.1080/​1369118X​.2021​.1994624. Dahlberg, L. 2005. The corporate colonization of online attention and the marginalization of critical communication? Journal of Communication Inquiry, 29(2), 160–180. Dahlgren, P. 2021. A critical review of filter bubbles and a comparison with selective exposure. Nordicom Review, 42(1), 15–33. Einwohner, R. and Rochford, E. 2019. After the march: Using Instagram to perform and sustain the Women’s March. Sociological Forum, 34(1), 1090–1110.

352  Handbook of digital politics

Etter, M. and Albu, O. 2021. Activists in the dark: Social media algorithms and collective action in two social movement organizations. Organization, 28(1), 68–91. Gerbaudo, P. 2012. Tweets and the Streets: Social Media and Contemporary Activism. London: Pluto Press. Hsiao, Y. and Yang, Y. 2018. Commitment in the cloud? Social media participation in the Sunflower Movement. Information, Communication & Society, 21(7), 996–1013. Jenkins, H. 2009. Confronting the Challenges of Participatory Culture: Media Education for the 21st Century. Cambridge, MA: MIT Press. Jennings, R. 2020. Who are the black squares and cutesy illustrations really for? Vox. https://​www​.vox​.com/​the​-goods/​2020/​6/​3/​21279336/​blackout​-tuesday​-blacklives​-matter​ -instagram​-performative​-allyship. Kanai, A. and McGrane, C. 2021. Feminist filter bubbles: Ambivalence, vigilance, and labour. Information, Communication & Society, 24(15), 2307–2322. Klinger, U. and Svensson, J. 2015a. Network media logic: Some conceptual considerations. In A. Bruns, G. Enli, E. Skogerbo, A. Larsson, and C. Christensen (eds.), The Routledge Companion to Social Media and Politics (pp. 23–38). New York: Routledge. Klinger, U. and Svensson, J. 2015b. The emergence of network media logic in political communication: A theoretical approach. New Media & Society, 17(8), 1241–1257. Kwak, N., Lane, D. S., Weeks, B. E., Kim, D. H., Lee, S. S., and Bachleda, S. 2018. Perceptions of social media for politics: Testing the slacktivism hypothesis. Human Communication Research, 44, 197–221. Lane, D. S. and Dal Cin, S. 2018. Sharing beyond slacktivism: The effect of socially observable prosocial media sharing on subsequent offline helping behavior. Information, Communication & Society, 21(11), 1523–1540. Leaver, T., Highfield, T., and Abidin, C. 2020. Instagram: Visual Social Media Cultures. Cambridge: Polity Press. Lebron, C. 2017. The Making of Black Lives Matter: A Brief History of an Idea. New York: Oxford University Press. Leistert, O. 2015. The revolution will not be liked: On the systemic constraints of corporate social media platforms for protest. In L. Dencik and O. Leistert (eds.), Critical Perspectives on Social Media and Protest: Between Control and Emancipation (pp. 35–52). London: Rowman & Littlefield. Lievrouw, L. 2014. Materiality and media in communication and technology studies: An unfinished project. In M. Gillespie, P. Boczkowski, and K. Foot (eds.), Media Technologies: Essays on Communication, Materiality, and Society (pp. 21–51). Cambridge, MA: MIT Press. Margetts, H., Peter, J., Hale, S. and Yasseri, T. 2016. Political Turbulence: How Social Media Shape Collective Action. Princeton: Princeton University Press. McLachlan, S. 2022. 2022 Instagram Algorithm Explained: How to Get Your Content Seen. https://​blog​.hootsuite​.com/​instagram​-algorithm/​. Melucci, A. 1996. Challenging Codes: Collective Action in the Information Age. Cambridge: Cambridge University Press. Metta, J. 2020. The problem of performative activism. Al Jazeera. https://​www​.aljazeera​.com/​ opinions/​2020/​7/​20/​the​-problem​-of​-​performativeactivism. Milan, S. 2015. From social movements to cloud protesting: The evolution of collective identity. Information, Communication & Society, 18(8), 887–900. Monckton, P. 2020. This is why millions of people are posting black squares on Instagram. Forbes. https://​www​.forbes​.com/​sites/​paulmonckton/​2020/​06/​02/​blackout​-tuesdayinstagram​-black​ -squares​-black​outtuesday​theshowmus​tbepaused/​?sh​=​23cf3de92794. Morozov, E. 2011. The Net Delusion: The Dark Side of Internet Freedom. New York: PublicAffairs.

Collective action on Instagram  353

Mundt, M., Ross, K., and Burnett, C. M. 2018. Scaling social movements through social media: The case of Black Lives Matter. Social Media + Society, 4(4), 1–14. Nichols, M. 2011. Times names ‘The Protestor’ 2011 Person of the Year. Reuters. https://​www​ .reuters​.com/​article/​us​-time​-person​-i​dUSTRE7BD0​ZB20111214. Pew Research Centre. 2020. #BlackLivesMatter surges on Twitter after George Floyd’s death. https://​www​.pewresearch​.org/​facttank/​2020/​06/​10/​blacklivesmatter​-surges​-on​-twitter​-after​ -george​-floyds​-death/​. Poell, T. 2014. Social media and the transformation of activist communication: Exploring the social media ecology of the 2010 Toronto G20 protests. Information, Communication & Society, 17(6), 716–731. Schechner, R. 2013. Performance Studies: An Introduction. London: Routledge. Schradie, J. 2018. The digital activism gap: How class and costs shape online collective action. Social Problems, 65, 51–74. Shirky, C. 2009. Here Comes Everybody: How Change Happens when People Come Together. London: Penguin. Speare-Cole, R. 2020. Social Media feeds go dark as tens of millions post for Blackout Tuesday in solidarity with George Floyd. Evening Standard. https://​www​.standard​.co​ .uk/​news/​world/​blackout​-tuesday​-instagram​-twittersocial​-media​-george​-floyd​-protests​ -a4457991​.html. Tilly, C. 2008. Contentious Performances. Cambridge: Cambridge University Press. Tilly, C. and Tarrow, S. 2015. Contentious Politics. New York: Oxford University Press. Valenzuela, S., Correa, T., and Gil de Zúñiga, H. 2018. Ties, likes and tweets: Using strong and weak ties to explain differences in protest participation across Facebook and Twitter use. Political Communication, 35(1), 117–134. Warren, J. 2021. This is How the Instagram Algorithm Works in 2022. https://​later​.com/​blog/​ how​-instagram​-algorithm​-works/​. We Are Social. 2022. Digital 2022: Another Year of Bumper Growth. https://​wearesocial​.com/​ uk/​blog/​2022/​01/​digital​-2022​-another​-year​-of​-bumper​-growth​-2/​. Wolfsfeld, G., Segev, E., and Sheafer, T. 2013. Social media and the Arab Spring: Politics comes first. The International Journal of Press/Politics, 18(2), 115–137. Yueng, S. and Tang, G. 2021. Instagram and social capital: Youth activism in a networked movement. Social Movement Studies. https://​doi​.org/​10​.1080/​14742837​.2021​.2011189.

23. Post-Soviet digital democratization experiments: the promise and reality Yuri Misnikov

This chapter explores how civil society exploited digital technologies to democratize post-Soviet Russia in the early years of the twenty-first century. We begin by considering the nature of the transition before the rise of Putin and the subsequent turn away from democracy. We go on to examine the role of digital technologies in the process of democratization. We conclude by referring to one case of a digital civic protest which exemplifies a particular form of post-Soviet civic activism that was voluntary, deliberative and independent from authorities.

POST-SOVIET DEMOCRATIZATION AND THE MEDIA The collapse of communism in Eastern Europe and the former Soviet Union between 1989 and 1991 was a life-changing event for the people of that vast territory. Its global significance as a political and economic system was even greater. The rise of the so-called ‘New Wave’ of democracies prompted Francis Fukuyama (1992) to announce the ‘end of history’ and declare liberal democracy the winner of the ideological competition with communism that spanned over four decades during the ‘Cold War’. It was expected that the socialist planned economy and authoritarianism would be replaced by a market economy and democracy, which would alter ‘the social and political landscape far beyond the countries immediately affected’ (EBRD, 2007, p. 3). However, democratic transition has been neither straightforward nor unproblematic. There is a general consensus that democracy in the former USSR, including Russia, has failed to take root (Wilson, 2005; Shevtsova, 2006; Lebanidze, 2020). The degree of media freedom was the best indicator of the state of transition to democracy in the region. The international media advocacy group Reporters Without Borders collects a wide set of data that comprehensively assesses media freedom worldwide. By 2008, Turkmenistan was rated the ‘least free’ of the ex-Soviet states, while favourable conditions in Ukraine and Moldova earned those countries ‘most free’ status (Reporters Without Borders, 2008). Yet even in those best cases, the mainstream media was significantly less free than the former Warsaw Pact EU member states. The gap between two groups was substantial. With the abandonment of communist ideology, the need to cultivate loyalty – the main role of the media under communism – disappeared (Lefort, 1999). But a new question arose: what should be the role of the media in a society that no longer has 354

Post-Soviet digital democratization experiments: the promise and reality  355

an imposed-from-above ideology to be served by the media? While the role of ideology lost its significance, the ruling elite still preferred to control the media in order to maximize its propagandistic effect to promote their own interests. For example, television in post-Soviet Russia was often manipulated for ‘political necessity rather than the perceived needs of citizens or society in general’ (Oates, 2008, p. 50). The philosophy of democratic neo-liberalism that has dominated the broader democratization discourse prevailed in relation to media transformation. The trend of media ‘homogenization’ and ‘Americanization’, as observed in the West by Daniel Hallin and Paolo Mancini (2004), was applied to post-Soviet transition societies in an effort to rework their media systems according to the Western standard. Barbara Pfetsch and Frank Esser (2004, pp. 4–5) admit that in this regard, the role of political communications and the quality of democracy, even in mature democracies ‘is by no means consistent’. Disparities between established ‘old’ Western democracies and those in transition are even more profound. Colin Sparks (1998, 2000) and John Downing (1996) believe that one of the major weaknesses in the research on the relationship between media, political communication and the post-communist context is the domination of the ‘Westernized’ approach to problems of mediated politics in transition. The problem lies in ignoring the communist legacy in the present-day media and the simplification of its actual diversity and complexity. Sparks (2000) insists that the approach towards media and democratization must be context-sensitive. ‘Contrary to what was widely believed, there was no single, uniform, and monolithic communist media system’, even if existing outlets were state-run and controlled (Sparks, 2000, p. 47). He notes further that the current discourses must shift attention to ‘relations between the media and the mass population’, and away from those between the state and the market, which the previous research had traditionally focused on. Media democratization ‘means breaking the control of those elites over what are necessarily the main means of public speech in large-scale societies’ (Sparks, 2000, p. 47). Downing (1996), having reviewed commonalities and differences between authoritarian and liberal democratic regimes regarding media–power relations, clarifies distinctions between Central European countries and the former USSR. He acknowledges that the change in post-communist media during transition is a movement towards the Western-type ‘business as usual’ (Downing, 1996, p. xiii). The entire post-Soviet legacy should not be dismissed, ignored or reduced to homogeneity. Both at the macro level (media systems) and the micro level (individuals who turn from passive audience members to active political commentators and public discussants), this progression is partly thanks to the communist legacy. As Ellen Mickiewicz (1999) notes in her analysis of Russian television viewers, ‘Soviet-era habits of close reading and creative scepticism have proved remarkably durable and have given people striking analytical capabilities. Russian viewers are no fools. They can spot bias, they detest it, and they do not need a college education to do so. How these extraordinary strategies of ordinary people actually work is hereto unexplored territory’ (Mickiewicz, 1999, p. x).

356  Handbook of digital politics

The capacity of media systems for self-preservation differs substantially, even among countries that have a communist past in common (Splichal, 1994; Kleinsteuber, 2004). Therefore, more thought could be given to find the most relevant methods of media assessment, apart from merely benchmarking against the Western ideal. The correction of the media-based approach relying on such benchmarking is indicative of an ongoing effort in relation to transitional democracies as well (Hallin and Mancini, 2011; Voltmer, 2013). An explosive expansion of Internet access to most people globally – which most vividly revealed itself in the mid-2000s thanks to the affordability of smartphones and the networking utility of social media – has created a radically different media landscape that technically cannot be completely controlled by authoritarian regimes that also need the Internet for effective functioning of their own bureaucracies. The Arab Spring events in 2011–2012 demonstrated that shutting down the Internet was the last resort to curb the protest. Russia’s recent history also demonstrates a desire to control the mainstream media. It started by ‘nationalizing’ the independent and most popular television channel NTV in 2001 one year after Putin’s presidency. Then it shifted to developing strategies and practical drills aimed at disconnecting Russia from the global Internet (i.e., from its Western segment) and controlling social media. Finally, social media became a new target. In one case, the authorities failed to restrict access to Telegram in 2018–2020; in another case, the attempt was more successful resulting in blocking access to Facebook and Instagram in 2022 after the war against Ukraine began. It is interesting that YouTube remains free and the technical access to the global Internet is still unrestricted even in the war conditions – a reminder of the fact that non-democracies need the Internet as well. Otherwise speaking, controlling the mainstream media is less effective without controlling the social media as a radically new medium fit to accommodate a much broader content of human communication, especially from below, thus reaffirming Neil Postman’s broader argument of the role of communication channels in content transmission. Thus, social media might serve as a certain ‘equalizer’ that potentially can correct an inevitable error of assessing media systems based on Western standards. Another way to undertake such a correction may lie in the further development of methods designed to analyse media and communicative practices through the perspective of their discursive character (Steenbergen et al., 2003, Steiner et al., 2005; Misnikov, 2010, 2012, 2013) in line with the Habermasian tradition of deliberative democracy, as a new democratic model alongside liberalism and republicanism. This approach is based on the assumption that contrary to the differences in the traditional mainstream media between democracies and non-democracies, the gap in human behaviour online, including in everyday political talk and protest mobilization, does not differ so fundamentally. After all, people’s desire to discuss politics and act as civic activists is a precondition for democracy. Complementing the available media assessment metrics by adding to it the discourse aspect enabled by new media would enable a more realistic and accurate comparison of countries across the diverse spectrum of political regimes. This ‘equalizer’ would also address in a better manner the interactivity aspect of media discursiveness. Until now the concept of interactivity,

Post-Soviet digital democratization experiments: the promise and reality  357

with its respective analytical methods, remains largely technical and ambiguous without taking a closer look at content and medium sides of interactive communication, including its dialogical dimensions.

DIGITAL DEMOCRATIZATION? Between 2000 and 2008 the number of Internet users in Russia increased by 1,560 per cent, which was the fastest growth not only in Europe, but worldwide. The speed of Internet penetration in the post-communist transition societies was among the fastest in the world. The total number of all Internet users in the two transition groups rose from 6.7 million in 2000 to 108.2 million in eight years, constituting a 1,615 per cent increase compared with the world average of 403 per cent during the same period. Contrary to poor democracy indicators, the Commonwealth of Independent States (CIS) region of the former Soviet Union demonstrated even faster growth of 2,099 per cent (compared with 1,245 per cent in the other European group). Each year the number of Internet users in this region more than doubled. If in 2000, the new EU member states had one million more Internet users than in the former USSR, in 2008 there were 14 million more users in ex-Soviet republics than in their European counterparts (60.9 million versus 47.3 million). The fast growth rate can be attributed to the initially low number of Internet users in former Soviet States compared to the more technologically advanced European transition societies; and that citizens in ex-Soviet states were quick to adopt the new technology as it became available, rapidly increasing Internet penetration throughout the region. In some cases (for example in Russia and Belarus) the penetration rate is on par with some new EU states (for example, with Romania and Bulgaria). Russia leads the post-communist space in Internet users. The number of regular Internet users grew exponentially in Russia, from 3 million in 2000 to 45 million in 2008, constituting nearly three-quarters of all Internet users in the ex-USSR space. Interestingly, Ukraine, traditionally the most democratic country in the region, had one of the lowest Internet penetration rates (just over 11 per cent). Thus, a high Internet penetration rate is not a necessary condition for successful democratization. While this growth in Internet use did not directly cause radical socio-political and ideological change, the role of digital information and communication technologies should be acknowledged as an important facilitating factor, especially in relation to civil society empowerment in non-democracies (Bach and Stark, 2003, 2004). The Russian Internet (or Runet) attracted millions of visitors every day. According to the LiveInternet agency, the very popular news portal Lenta.ru attracted half a million daily visitors, while Izvestia’s popular Internet forum enjoys over 100,000 visits per day. Runet’s monthly audience exceeds one billion monthly visits, of which almost one-tenth are interested in news (those seeking news content account for 8.6 per cent of all visitors, making it the fifth most popular reason people access the site); a group of websites on politics, state and society receive 36 million visits each month

358  Handbook of digital politics

(10th place with 3.2 per cent of hits among other types of media resources). Many who read news online also comment on what they have read and participate in online discussions. Overall, online news in Russia serves as a catalyst for political discussion, as it does in the West (Allan, 2006). Since the 2000s, the number of Internet users in Russia and the CIS region continued to rapidly increase. For example, the percentage of the population (aged 15 years and above) using the Internet in the CIS region reached 82 per cent, just two percentage points below Europe, whereas the Russian Federation reached the level of 85 per cent in 2020 (ITU, 2022; World Bank and ITU, 2022). The rapid expansion of Internet services in Russia enabled the number of bloggers to double annually for several years (the pace has recently slowed). According to Yandex, the operator of Russia’s leading search engine (Yandex, 2008), as of November 2007, there were 3.8 million Russian-language blogs. Around one-third of these blogs were active, that is, they contained at least five entries and were updated at least once in the last three months. In 2008 there were some 160,000 active bloggers on LiveInternet and 230,000 on LiveJournal. In addition, there are many other blog-hosting platforms and websites. However, more important than quantity are the qualitative properties of blogs. For instance, Yevgeny Gorny (2006, 2007), Henrike Schmidt and Katy Teubener (2005) describe blogging in Russia on LiveJournal as a distinctively socio-cultural and psychological phenomenon, often linked with identity construction and fragmentation. As a social activity it has acquired a reputation as a quality outlet for social mediation and interpersonal interaction, which, it has been argued, is a very different experience than blogging on LiveJournal’s English-language site. Russian blogs are characterized by higher interactivity, tighter connectivity and a stronger orientation towards news dissemination and discussion (in politics, literature, etc.) than their Western counterparts. Moreover, even the term ‘friend’ has a different, more real-life meaning, as virtual interaction often leads to face-to-face meetings (at least for those living in the same city, such as Moscow (Gorny, 2007, pp. 111–113). Overall, the Russian Internet is considered culturally special (Schmidt and Teubener, 2005; Rohozinski, 1999), and closely linked with Russian culture and social psychology at large. Internet forums are more egalitarian, highly accessible and have very large audiences which are impossible to assess. For ordinary Russians, forums are a common outlet for communication and political discussion and have become a signature service of both large national and small local newspapers, information agencies and media portals. Forums can be very democratic and devoted to broad themes of general public interest. For example, Izvestia.ru, one of the oldest Russian newspapers (News), hosts a news forum that attracts a broad national audience. Forums can also be elitist, e.g. specifically oriented political discussions such as the forum Kreml. org, which serves a narrower national membership base. They also appeal to the small populations of provincial towns, such as CityK (Kondopoga town), or urban districts, e.g. Lubli.ru, which discusses local issues related to Moscow’s Lublino district.

Post-Soviet digital democratization experiments: the promise and reality  359

The popularity of blogging in Russia can be better understood when one considers the fact that President Medvedev himself is a blogger who admitted that he regularly spends time on the Internet. His blog contained 54 entries (as of 15 February 2010), of which 24 received over 1,000 comments each; eight entries received over 2,000 comments each, and three entries attracted over 4,000 comments. The number of comments a particular blog entry attracted can be used as a measure to determine the topic’s public significance. It was not unusual for comment threads to be active months after an entry was published. Typically, these comments are not isolated one-way messages relating solely to the subject matter; rather, they invite interaction from the Internet community at large, and often lead to full-fledged discussions. On 10 September 2009, President Medvedev intentionally used his official blog to launch an online preview of the draft of his planned annual address before the Parliament (Duma). It was the first time in Russia’s history a leader had asked for direct input from the public online. The article, entitled ‘Russia, Forward!’, was only available online and was published with the intention of drawing comments from the online community of ordinary citizens. The feedback was very enthusiastic, with over 4,677 comments sent directly to the blog. As of 15 February 2010, more responses were continuing to arrive. Almost every major online media outlet published the article, therefore the overall amount of feedback is next to impossible to quantify. For example, 2,609 comments alone were posted on the website of just one online newspaper, Gazeta.ru. This initiative was energetically discussed on forums and blogs by a wide audience. Most important was not that the President approached citizens for advice, but that they demonstrated a genuine sense of democratic duty in their reaction. The impact of new technologies on undemocratic societies is unclear and depends on the assumptions and interests of the investigator. Those studies that seek to explain such impacts do so primarily through the lens of technology (rather than, for example, civic activism) and are typically sceptical (Rohozinski, 1999; Fossato and Lloyd, 2008). Following Norris’ reinforcement model, it is believed that new media strengthens existing non-democratic institutions, just as it reinforces liberal institutions in Western democratic polities. However, there is equally convincing evidence that civil society is more creative and effective in using new media for its own purposes. Online journalism has also benefited from new communication possibilities in the absence of independent mainstream media (Arutunyan, 2009). There are many interesting examples of online political and civic activism in all ex-Soviet countries, where new digital media serves as an intermediary that brings together people and ideas (including public events and policies). Civic activism has effectively used the Internet to raise, discuss and act upon important public issues. One such example was the Black Flag (Chernyj Flag) initiative organized from below by an ordinary person who was an accidental witness to an automobile accident in St Petersburg involving a police car. This individual organized other witnesses (plus 500 additional supporters in 20 cities) to seek justice when the victim was wrongly accused of the accident.

360  Handbook of digital politics

The most impactful and truly symbolic use of Internet civic activism is demonstrated through two YouTube video clips addressed to Prime Minister Putin by a policeman named Major Alexey Dymovskiy in November 2009. He accused the police force of Novorossiysk (a city on the Black Sea coast) of corruption and appealed to Putin to change police management and reform the Russian police. This video posting – watched over 1 million times – was undoubtedly the most remarkable public event of the year, with long-lasting consequences. Dymovskiy’s videos were quickly imitated by other policemen from all over Russia, who shared their social and professional grievances. This in turn ignited serious debates about the Russian police, both in the mainstream media and especially on the Internet. This media onslaught eventually resulted in the passage of laws to reform the Russian police system, effective 1 March 2011. Despite a prevailing negative attitude of ordinary Russians towards the police because of its corruption, it is safe to assume that the reform would not have happened so fast without Dymovskiy’s video clips and the nationwide public debate they inspired. It was not the new media that forced the government to react, but citizen activists using both offline and online media outlets. This unique (for Russia) phenomenon was dubbed the ‘Dymovskiy Syndrome’, and it will be remembered for years to come. The progression of media coverage – as a Fairclough media event – is illustrated on the influential Slon.tu web site, one of thousands of outlets that followed the story of ‘Dymovskiy Syndrome’. A full-fledged multi-media web page devoted to Dymovskiy appeared on 20 November 2009, that contained a lead article with six tags referring to his video address, the Police Ministry, the President, Prime Minister, and the Minister of Interior and Politics. The page gave a full account of the situation for anyone seeking relevant information. In addition, the lead article could be exported to any blog. It was supplemented by 15 additional articles, and each was commented on by readers. Such coverage of major and minor events of public importance was and is typical for the Russian new media today. Its impact on public affairs is substantial. Whatever were the real reasons behind this and other follow-up postings from other policemen, the impact was very significant in at least three ways: firstly, it demonstrated the real power of new media in Russia; secondly, it brought attention to a serious social problem within the Russian police force, which was known but did not get wider public exposure; and, thirdly, this case demonstrated how active citizens can use new media to work for the public good. The above observations reaffirm a conclusion drawn by Coleman and Kaposi (2009, p. 23) that even if it is difficult to prove a statistically significant correlation between Internet penetration and democratic progress one can ‘at least suggest that new media might contribute to an atmosphere of democratic openness’. This change has happened in Russia; it would certainly be a different country without the openness brought not so much by the Internet itself, but by the way it has been used by citizens for both online and offline activism.

Post-Soviet digital democratization experiments: the promise and reality  361

A CASE STUDY OF DIGITALLY ENABLED DELIBERATIVE PROTEST IN POST-SOVIET RUSSIA IN THE 2000s The case study we describe here was an interregional protest that started in 2005 and continued until the end of 2006. It exemplifies the extent to which digital media opened up a potentially democratic space in post-Soviet Russia in which citizens used various forms of computer-mediated communication instruments and forms to express themselves as individuals and discuss politics, including to mobilize civic activism, present arguments and respond to them in a dialogical manner. In accordance with Habermas’ discourse ethics theory, participants in our case routinely demonstrated their desire to make valid points, supported their reasoning with facts, and strove to convince one another of the ‘rightness’ of their positions. The effectiveness of argumentation depended not only on the quality of the arguments, but on their interpretation and acceptance by other participants. The overall intelligibility and rationality of citizens’ online debates was generally high, but to a large degree it was determined by the discussion purpose. When the discourse goal was information sharing rather than discussion, its deliberative quality was visibly lower. The collective performance of the discussants was more important than the discursive behaviour of individual participants. The quality of deliberation is well defined by the degree of discursive dialogism. For example, argumentative reasoning was in effect a dialogical act spread among several messages written by different authors. In other words, it was a dialogically distributed argumentation. Online discussions on Internet forums were especially popular. The interregional non-governmental organization Freedom of Choice (Mezhregionalnaya Obschestvenaya Organisatsiya ‘Svoboda Vybora’ (http://​www​.19may​.ru/​forum/​ index​.php)) established its web forum to discuss, organize and coordinate protest actions among motorists (car owners). The selection of Freedom of Choice’s forum as a case study was determined by two main factors. One was the inseparability of its online discussions with offline actions in real life. The other reason this forum was chosen was the diversity and intensity of its discussion threads. Between 2005 and September 2010 Freedom of Choice’s online forum accumulated 146,000 messages posted to 8,700 discussion themes. The Freedom of Choice discussions were not dependent on the news media, and their role in generating debate was insignificant: only 5 per cent of all discussion threads fell under the news category. Issues of internal interaction and organizational activities accounted for 40 per cent of all discussion themes; common problems and mutual support were equally important, constituting 35 per cent of all posts. Distribution of posted messages followed a similar pattern: 37 per cent belonged to the internal interaction and organizational matters, 41 per cent related to mutual support and common problems and just 1 per cent to news. Posting activity was further analysed in terms of its intensity and significance to define the sample selection. The significance of each of the 45 discussion topics was measured by comparing their ability to attract comments (percentage of mes-

362  Handbook of digital politics

sages posted to the topic’s threads) with the capability to generate new discussion issues (percentage of threads opened for discussion). The resulting indicator – a message-to-theme ratio – reflected the relative ‘weight’ of each theme. Higher ratios indicated themes of higher significance (i.e. ‘weight’). The analysis suggested that the theme group titled ‘Organization’s Activities’ contained the most popular discussion themes; the group was well represented, accounting for one-fifth of all posts, and thus was the most suitable group in the Motorists’ forum to serve as the sample’s base. In terms of discursive qualities, the Motorists’ online conversations were focused and disciplined in their efforts to address specific problems. They were also dialogically responsive and friendly. The problem-solving nature of the posts/discussions engendered mutual respect among the participants. Otherwise, it would be impossible to solve the problems associated with protest organization. There was a greater sense of cohesiveness in this community of like-minded citizens as they worked together on their common goal of protecting the rights of Russian motorists. Another special quality of the Motorists’ forum was its moderate level of disagreements, although participants still demonstrated strong argumentation. While some posts were slightly rude, they were never offensive. The leader of the movement played the important role of forum moderator, whose interventions contributed to the smooth, civil and pragmatic discussions. The widespread use of both commissives and directives was a good indicator of the ‘mobilizational’ character of the Motorists’ discourses (Rolf, 1990). We can make several observations about this case study which cast light upon the affordances of digital media in a democratizing society. But more than that, these observations say something important about the potential for meaningful civic deliberation within digital spaces. The first observation is that it was the creative use of the Internet’s discursive features by protesters that enabled them to prepare and stage street actions on a large scale across Russia; that is, it was not only the fact that street actions would be simply impossible without the Internet, but it was about the discussion of such actions that made them possible in the first place. Secondly, it was a protest by citizens, rather than ‘digital elites’, as lay motorists who pursued a political agenda of social justice by defending the civic rights of all those who owned and drove cars. Thirdly, it was a protest in a context when the Russian authorities were starting to learn how to handle such digitally enabled protests. The attack of the Russian army on Ukraine on 24 February 2022 signalled the end of the country’s relatively mild version of political authoritarianism (the process had been underway since the annexation of Crimea in 2014). The war ended any opportunity for a peaceful civic protest in Russia whose citizens will have to learn again how to do it in future, if and when the country returns to more democratic politics. However, around the mid-2000s, political activism in Russia, including civic protest, as the manifestation of opposition to the ruling order was allowed to a large extent, both online and offline. The Motorists’ case also differs from further large protests that emerged in 2011–2012

Post-Soviet digital democratization experiments: the promise and reality  363

falling under the umbrella of electoral authoritarianism – a part of the broader concept of political contention, ‘when institutional contention and non-institutional contention have occurred together, often in the context of attempted regime change after fraudulent elections’ (Dollbaum, 2020, p. 2). As Mischa Gabowitsch puts it: in most Russian cities, despite modest absolute numbers, the fair-election and anti-Putin protests were the largest since perestroika and the break-up of the Soviet Union, even though in the years before 2011 there had been numerous grassroots protests, two major regional, several interregional and at least one countrywide protest movement … The March of Millions, with its thousands of participants from across Russia, was a turning point in the protests. Coming, as it did, the day before Putin’s formal return to the presidency, the brutal police response made the limits of contention abundantly clear. (Gabowitsch, 2017, p. 26)

Fourthly, this was not simply a space of political mobilization. Participants were typically critical towards one another. However, they were also willing to recognize and accept other positions, although the number of instances of agreement was noticeably smaller than those of disagreement. This may be an indication of the existence of substantial divisions among discussants in terms of shared norms and values. While certain exchanges were ruder than others, they never were uncivil (incivility was usually protested by participants). The prevailing tone was neutral; that is, neither civil nor uncivil. Well-organized exchanges with clear objectives between participants who share certain values were significantly more civil and consensual. Clearly articulated politeness, for example, was usually deliberate and aimed to attract additional attention or make a point stronger and more convincing. The prevailing tone was highly expressive, with discussants actively using emoticons, Internet slang, shorthand and other textual and visual instruments provided by the forum interface. As far as discussion style is concerned, it is often argued that online debates can be compared to everyday face-to-face conversations in terms of their fluid and unorganized nature. My research on this case study confirmed that this resemblance does in fact exist. However, online discourses are not tantamount to offline talk due to the impact of such virtual properties as anonymity, lack of social clues and unpredictability. For example, the lack of social clues in online discussions (which are normal in face-to-face interactions) prompts discussants to look for alternative methods of expression and, most importantly, to be attentive and dialogical when relating to others. In an anonymous virtual environment participants are not obliged to respond unless the message merits a reaction. There were no visible signs in the message content that indicated that individuals were either forced to take part or denied the right to participate in online discussions. Each discussant participated freely and of their own accord. The discussion forum status assigned to all discussants following their contributions also did not reveal any inequality between participants with higher and lower statuses. Citizens’ online discourses cannot be properly assessed using the same approaches developed to analyse offline deliberative practices, where discussions are usually smaller in scale and where the personal characteristics of discussants can significantly influence the quality of discourse. Even large face-to-face deliberations are

364  Handbook of digital politics

smaller than many regular online discussions. Accordingly, such debates require different analytical assessment tools to take proper account of the discourse scale and deliberative qualities. Online discussions are not qualitatively inferior to offline deliberations; rather, they are qualitatively different. The legacy of the ‘procedural’ model of deliberation, elaborated in relation to the offline world, still influences the approaches to studying online discourses. Habermas himself has effectively abandoned his initially strict ‘critical presuppositions’ and has substantially revised his original definition of the ‘ideal speech situation’. The imposition of overly strict ‘procedural’ conditions, which might be justifiable for offline institutionalized policy discourses, would be damaging for online discussions. Fifthly, the Motorists’ case study reflects upon interactivity as an intrinsically online quality, as has been comprehensively studied in relation to online research (Rafaeli and Sudweeks, 1997; Sudweeks et al., 1998). The empirical evidence from the Motorists’ protest suggests that while interactivity is central to online discourses, it should be viewed less technically and more dialogically. There are limits to the extent of interactivity. Realistically, each post can only reference a limited number of previous messages (in most cases, only the most recent ten to fifteen messages are cited in a given post). Therefore, interactivity cannot be gauged based on the amount of content either directly quoted from previous posts or indirectly referenced. Interactivity alone cannot reflect the discursive complexity of online discussion and must be supplemented by the dialogical dimension of interaction. Whereas most of the posted messages were impersonal, the percentage of interpersonal communications was substantial. Accordingly, the dialogic character of discussion was strong too, given that interpersonal posts represent an important aspect of dialogicality. Discussions were sufficiently dialogical thanks to the high degree of interactive responsiveness that was motivated by the need to validate claims made by participants. Interactivity is thus better defined as a dialogical way of accepting and rejecting positions as certain truths by agreeing or disagreeing with them.

CONCLUSION The case described in this chapter poses a question about the role of its historical context that has both global and local dimensions. The global aspect is informed by the curious coincidence of two radically different historical processes. One was epitomized by the ‘end of history’ metaphor pointing to the disappearance of political ideology based on the communist ideals offered by the Soviet Union. The rise of the so-called ‘New Wave’ of democracies meant declaring the liberal variant of democracy the winner of the ideological competition with communism. The term ‘transition’ to liberal democracy became a symbol of this geopolitical shift. Its focus almost automatically meant building market institutions as the transition’s first step, as evidenced by the establishment in 1991 of the European Bank for Reconstruction and Development (EBRD) as a specialized regional institution to finance market economies in transitional countries. It assumed that strong market institutions would

Post-Soviet digital democratization experiments: the promise and reality  365

inevitably bring about democracy. However, by the mid-2000s there was a general consensus that democracy in Russia had failed to take root (Shevtsova, 2006; Wilson, 2005; Coleman and Kaposi, 2009). The other global force was the emergence of the Internet amidst this economic transition. This was not about unidirectional transfer of information from senders to receivers but about diversity of content, networking, discussion and community development. The Motorists’ protest was a local phenomenon confined to several Russian regions. It was ‘an example of successful grassroots internet activity’ that evolved into one of the country’s largest grassroots organizations, tackling issues well beyond the right to drive on the ‘wrong’ side of the road: ‘it became a genuine social movement’ (Fossato and Lloyd, 2008, p. 41). It was a protest based on conversation and discussion symbolizing the emergence in Russia in the first decade of the 2000s of a new, more discursive society whose members want to discuss issues publicly and dialogically. The case presented here is a manifestation of how digital media’s affordances encouraged deliberation, information sharing and reciprocity leading eventually to forms of networked citizenship (Coleman, 2001; Coleman and Goetze, 2001; Lee et al., 2022). Conceptually, Habermas’ model of democracy as a deliberative paradigm that legitimizes procedures of opinion and will formation among equal citizens captures the essence of a discursive society, if we extend its reach beyond argumentation alone (Habermas, 2005, 2006). On the other hand, the actually functioning normative liberal democratic model (effective in protecting liberties of private citizens) could not adapt, as yet, to the changed networked reality by accommodating relevant participation-based republican and deliberative democratic ideals that are fundamentally discursive The primary focus on building market institutions as the essence of transition to democracy in Russia might not have been a sufficient condition to make the entire transition democratically successful. As the Motorists’ case reveals, other global forces were already at play generating a new networked citizen that needed an equally effective normative protection. As future events were to show, the Russian state focused its effort on suppressing this network of citizens, for example, by passing in 2019 the law on the sovereignty of the Russian Internet. That has certainly contributed to the eventual failure of democracy in Russia and, very likely, to the war with Ukraine.

FURTHER READING Chaisty, P. and Whitefield, S. (2013). Forward to democracy or back to authoritarianism? The attitudinal bases of mass support for the Russian election protests of 2011–2012. Post-Soviet Affairs, 29(5), 387–403. Demydova, V. (1921). Alexei Navalny and protests in Russia: Growth of online activism under the authoritarian system. Gaziantep University Journal of Social Sciences, 20(4), 1970–1987.

366  Handbook of digital politics

Kravets, D. and Toepfl, F. (2022). Gauging reference and source bias over time: How Russia’s partially state-controlled search engine Yandex mediated an anti-regime protest event. Information, Communication & Society, 25(15), 2207–2223. Lonkila, M. (2008). The internet and anti-military activism in Russia. Europe-Asia Studies, 60(7), 1125–1149. Oates, S. (2013). Revolution Stalled: The Political Limits of the Internet in the Post-Soviet Sphere. New York: Oxford University Press. Petrov, N., Lipman, M., and Hale, H. E. (2014). Three dilemmas of hybrid regime governance: Russia from Putin to Putin. Post-Soviet Affairs, 30(1), 1–26. Pietila, J. (2008). Media use in Putin’s Russia. Journal of Communist Studies and Transition Politics, 24(3), 365–385. Semetko, H. A. and Krasnoboka, N. (2003). The political role of the internet in societies in transition: Russia and Ukraine compared. Party Politics, 9(1), 77–104.

REFERENCES Allan, S. (2006). Online News: Journalism and the Internet. Maidenhead: Open University Press. Arutunyan, A. (2009). The Media in Russia. Maidenhead: Open University Press. Bach, J. and Stark, D. (2003). Technology and Transformation: Facilitating Knowledge Networks in Eastern Europe. UNRISD Technology, Business and Society. Programme paper number 10. Geneva: UNRISD. http://​ www​ .unrisd​ .org/​ 80256B3C005BCCF9/​ (httpPublications)/​35​352D4B0785​18C0C1256B​DF0049556C​?OpenDocument. Bach, J. and Stark, D. (2004). Link, search, interact: The co-evolution of NGOs and interactive technology. Theory, Culture & Society, 21(3), 101–117. Coleman, S. (2001). The transformation of citizenship? In B. Axford and R. Higgins (eds.), New Media and Politics. London: Sage, pp. 109–126. Coleman, S. and Goetze, J. (2001). Bowling Together: Online Public Engagement in Policy Deliberation. London: Hansard Society. Coleman, S. and Kaposi, I. (2009). A study of e-participation projects in third-wave democracies. International Journal of Electronic Governance, 2(4), 302–327. Dollbaum, J. M. (2020). Protest trajectories in electoral authoritarianism: From Russia’s “For Fair Elections” movement to Alexei Navalny’s presidential campaign. Post-Soviet Affairs, 36(3), 192–210. Downing, J. (1996). Internationalizing Media Theory: Transition, Power, Culture. London: Sage. European Bank for Reconstruction and Development (EBRD) (2007). Life in Transition Report: Survey of People’s Experiences and Attitudes. London: EBRD. Fossato, F. and Lloyd, J. (2008). The Web That Failed: How Opposition Politics and Independent Initiatives Are Failing on the Internet in Russia. Oxford: University of Oxford, Reuters Institute for the Study of Journalism. Fukuyama, F. (1992). The End of History and the Last Man. London: Penguin. Gabowitsch, M. (2017). Protest in Putin’s Russia. Cambridge: Polity Press. Gorny, Y. (2006). A creative history of the Russian Internet. DPhil thesis, Goldsmiths College, University of London. Gorny Y. (2007). Russkiy LiveJournal: Vliyaniye kulturnoy identichnosti na razvitiye virtualnogo obschestva (Russian LiveJournal: Impact of cultural identity on the virtual community development). Russian-Cyberspace.org. Habermas, J. (2005). Concluding Comments on Empirical Approaches to Deliberative Politics. Acta Politica, 40, 384–392.

Post-Soviet digital democratization experiments: the promise and reality  367

Habermas, J. (2006). Political communication in media society: Does democracy still enjoy an epistemic dimension? The impact of normative theory on empirical research. Communication Theory, 16(4), 411–426. Hallin, D. C. and Mancini, P. (2004). Americanization, globalization, and secularization: Understanding the convergence of media systems and political communication. In B. Pfetsch and F. Esser (eds.), Comparing Political Communication: Theories, Cases, and Challenges. Cambridge: Cambridge University Press, pp. 25–44. Hallin, D. C. and Mancini, P. (eds.) (2011). Comparing Media Systems Beyond the Western World. Cambridge: Cambridge University Press. ITU (2022). Global Connectivity Report. International Telecommunication Union (ITU), Geneva, Switzerland. https://​www​.itu​.int/​itu​-d/​reports/​statistics/​global​-connectivity​-report​ -2022/​. Kleinsteuber, H. J. (2004). Comparing mass communication systems: Media formats, media contents, and media processes. In B. Pfetsch and F. Esser (eds.), Comparing Political Communication: Theories, Cases, and Challenges. Cambridge: Cambridge University Press, pp. 64–86. Lebanidze, B. (2020). Russia, EU and the Post-Soviet Democratic Failure. Dordrecht: Springer. Lee, F. L. F., Liang, H., Cheng, E. W., Tang G. K. Y., and Lee, S. (2022). Affordances, movement dynamics, and a centralized digital communication platform in a networked movement. Information, Communication & Society, 25(12), 1699–1716. Lefort, C. (1999). Complications: Communism and the Dilemmas of Democracy. New York: Columbia University Press. Mickiewicz, E. (1999). Changing Channels: Television and the Struggle for Power in Russia (revised and expanded edition). Durham, NC and London: Duke University Press. Misnikov, Y. (2010). Discursive qualities of public discussion on the Russian Internet: Testing the Habermasian communicative action empirically. In F. De Cindio, A. Macintosh, and C. Peraboni (eds.), From E-Participation to Online Deliberation. Proceedings of the fourth international conference on online deliberation, OD2010. Leeds, UK, 30 June–2 July. University of Leeds and University of Milan, pp. 60–74. Misnikov, Y. (2012). How to read and treat online public discussions among ordinary citizens beyond political mobilisation: Empirical evidence from the Russian-language online forums. Digital Icons: Studies in Russian, Eurasian and Central European New Media, 7, 1–37. Misnikov, Y. (2013). You say ‘yes’, I say ‘no’: Capturing and measuring public opinion through citizens’ conversation online. Russian-Language LiveJournal Blogging Platform. In ePart 2013 proceedings. Volume 8075 of the Lecture Notes in Computer Science series, Springer. Oates, S. (2008). Introduction to Media and Politics. London: Sage. Pfetsch, B. and Esser, F. (2004). Comparing political communication: reorientations in a changing world. In B. Pfetsch and F. Esser (eds.), Comparing Political Communication: Theories, Cases, and Challenges. Cambridge: Cambridge University Press, pp. 3–24. Rafaeli, S. and Sudweeks, F. (1997). Networked interactivity. Journal of Computer-Mediated Communication, 2(4). Reporters Without Borders (2008). Freedom of the Press Worldwide in 2008: Annual Report. Paris: Reporters Without Borders. Rohozinski, R. (1999). Mapping Russian Cyberspace: Perspectives on Democracy and the Net. United Nations Public Administration Network (UNPAN). Rolf, E. (1990). On the concept of action in illocutionary logic. In A. Burkhardt (ed.), Speech Acts, Meaning and Intentions: Critical Approaches to the Philosophy of John R. Searle. Berlin: de Gruyter, pp. 147–168.

368  Handbook of digital politics

Schmidt, H. and Teubener, K. (2005). “Our RuNet?” Cultural Identity and Media Usage. Russian-Cyberspace.org. Shevtsova, L. (2006). Russia’s Ersatz Democracy: Current History. Washington, DC: Carnegie Endowment for International Peace. Sparks, C. (1998). Communism, Capitalism and the Mass Media. London: Sage. Sparks, C. (2000). Media theory after the fall of European communism: Why the old models from the West won’t do any more. In J. Curran and M.-J. Park (eds.), De-Westernizing Media Studies. London and New York: Routledge, pp. 35–49. Splichal, S. (1994). Media beyond Socialism: Theory and Practice in East-Central Europe. Boulder, CO: Westview Press. Steenbergen, M. R., Bächtiger, A., Spörndli, M., and Steiner, J. (2003). Measuring political deliberation: A discourse quality index. Comparative European Politics, 1(1), 21–48. Steiner, J., Bächtiger, A., Spörndli, M., and Steenbergen, M. R. (2005). Deliberative Politics in Action: Analyzing Parliamentary Discourse. New York: Cambridge University Press. Sudweeks, F., McLaughlin, M., and Rafaeli, S. (eds.) (1998). Network and Netplay: Virtual Groups on the Internet. Cambridge, MA: MIT Press. Voltmer, K. (2013). The Media in Transitional Democracies. Cambridge: Polity Press. Wilson, A. (2005). Virtual Politics: Faking Democracy in the Post-Soviet World. New Haven, CT and London: Yale University Press. World Bank and International Telecommunication Union (ITU) (2022). World Telecommunication/ICT Indicators Database: Individuals Using the Internet. https://​data​ .worldbank​.org/​indicator/​IT​.NET​.USER​.ZS​?locations​=​RU. Yandex (2008). Blogosfera rossiyskogo interneta. Moscow: Yandex.

PART V POLITICAL PARTIES, LEADERS AND GOVERNANCE

24. The digital performance of populism Thomas Wellings and Lone Sorensen

INTRODUCTION The ‘populist zeitgeist’ identified by Cas Mudde almost two decades ago (2004), like other forms of politics, increasingly manifests itself in digital media alongside more traditional forms of expression. As research into digital populism is growing, evidence is emerging that online media favour the mediation of populism above most other forms of institutional politics. Existing studies tend to identify either the physical or the ideational dimension of digital media as that which uncannily augments populist communication. With respect to the physical dimension – the materiality of digital media technologies – there is a fortuitous fit between populism and the ways in which users interact with digital media artefacts and their design. For example, the circumvention of traditional media gatekeepers gives populists the freedom to use language that breaches the norms of political discourse and spread ideas that might otherwise be moderated, disputed or omitted from news discourse (Engesser et al., 2016). In the ideational dimension, publicly shared ideas of social futures are associated with social media and are often reinforced by platform owners (Van Dijck, 2013). For example, anti-elitist people power, emancipation (Gerbaudo, 2014) and the sense of community that Nick Couldry describes as ‘the myth of “us”’ (Couldry, 2015) resonate with populist ideas. In this chapter, we argue that both such imaginaries and the materiality of digital media technologies are important to consider as influences on the populist meaning-making process. This makes digital media technologies integral to the political performances of populist parties and leaders. In this chapter we focus on the ways in which populist leaders strategically use digital media and how such strategic performances relate to legacy media and their gatekeepers. For example, populists may use digital media to cast doubt on information gatekeepers to boost their own credibility, which they also enact symbolically through performances of directness and claims to public emancipation. They may also use digital media to bypass information gatekeepers or alternatively to gain the attention of the legacy media. Research on digital populism has burgeoned in recent years, but we lack a theoretical overview and there are empirical gaps. Below we first set out our approach to populism and, within this, to populist political performances that integrate the imaginaries and materiality of media technologies. We then review the literature on digital populism as it relates to the performances of populist leaders and arrange it into four types of populist mediated performances developed by Lone Sorensen (2021, Chapter 12): performances to and within media and performances of and about mediation. We thereby use secondary literature to further develop and flesh 370

The digital performance of populism  371

out these types of digital populist performances to encourage an emergent conceptual framework and to pinpoint empirical gaps in the scholarship on digital populism. While we focus on top-down populism, and more specifically on populist leaders’ relationship with legacy media through digital platforms, we acknowledge that work on populist citizens, whether as audiences or content producers, is no less important in the context of digital media and similarly demands further attention.

POPULIST PERFORMANCE AND DIGITAL MEDIA TECHNOLOGY Literature remains split on the classification of populism as an ideology, as propounded by, for instance, Cas Mudde (2004), and as a performance, as championed by scholars such as Benjamin Moffitt (2016). While these approaches are divided in their emphasis on what populism is and what populism does, both agree on most of the core characteristics of the phenomenon. Populists: ● articulate ‘the people’ ambiguously as a disadvantaged, unheard, sovereign, morally decent majority, and a bounded community unified by their idea of a heartland (Abts and Rummens, 2007; Albertazzi and McDonnell, 2008; Brubaker, 2020, p. 49; Canovan, 1999, p. 5; Mény and Surel, 2002, pp. 6–7; Mudde and Rovira Kaltwasser, 2013; Norris and Inglehart, 2018; Taggart, 2000); ● portray the elite as immoral, opposed to and actively deprivileging the people (Aalberg et al., 2016; Albertazzi and McDonnell, 2008; Brubaker, 2017, p. 363; Mudde, 2007; Mudde and Rovira Kaltwasser, 2013, pp. 502–504); ● represent themselves as one of the people, as non-elite (Albertazzi and McDonnell, 2008; Brubaker, 2017; Canovan, 2005; Ostiguy, 2020; Taggart, 2000) and as able to restore sovereignty to the people through their enlightenment (Abts and Rummens, 2007; Canovan, 2005; Wirth et al., 2016); ● use disruption to signal outsider status and the illegitimacy of institutional or elite-driven norms (Bucy et al., 2020; Moffitt, 2016; Sorensen, 2018, 2021), and evoke a crisis (Moffitt, 2016; Taggart, 2000) to justify such disruptive acts; ● portray ‘the others’, identified through cultural difference, as a threat to the people’s sovereignty or way of life (Brubaker, 2017, pp. 362–364). In this chapter we acknowledge the importance of both the ideational and stylistic approaches but stress the centrality of meaning-making in our approach to populism. While we appreciate that certain ideas delineate populism from the big ideologies it sometimes attaches itself to (Mudde and Rovira Kaltwasser, 2013), we argue that the process of contextualizing and giving meaning to these ideas in a way that is felt by and resonates with the lived experience of a given audience is what gives populism its strength and traction (Freeden, 2017). We therefore define populism as a communicative process that involves all of the above features in their capacity as ideas that

372  Handbook of digital politics

are given meaning through political performance (see also Sorensen, 2021, Chapter 3 for a more elaborate argument). This communicative process involves a strong element of storytelling. As omnipotent narrators of political reality, populists position themselves on the outside of, and as looking in on, establishment politics, which they present as morally flawed. This meta-political narrative is performed through symbolic action through which actors, ‘embedded in collective representations and working through symbolic and material means, implicitly orient towards others as if they were actors on a stage seeking identification with their experiences and understandings from their audiences’ (Alexander and Mast, 2006, p. 2). In the case of populist performers, the primary symbolic vehicle is the disruption of the norms of establishment politics (Aiolfi, 2022; Moffitt, 2016, p. 44; Sorensen, 2021, pp. 58–59, 139–143). These disruptive performances signal the illegitimacy of the norms that are their object of disruption and that govern the practice of the establishment’s purportedly deceitful politics. The populist process of communication thus involves both an ideational and a performative component. We further suggest that the media through which the populist performance is imparted are important and integral parts of strategic populist meaning making itself. In other words, media should not only be seen as necessary resources for the transmission of populist messaging, and thereby external to its performance. They are also anticipated and used by populist performers to shape the meaning of the message prior to transmission and should therefore be considered as part of the performance itself. This becomes apparent when we consider media as material and ideational resources in a performance. The remainder of this section discusses each of these types of resources in turn. Jeffrey Alexander’s model of social performance identifies two types of material resources that performers use. The first is the means of symbolic production (Alexander, 2004a, p. 532), such as props and spaces, that allow actors to symbolically project. Digital media technologies are among the objects that are used by strategic populist performers to create meaning. We are specifically concerned with the materiality of digital media as means of symbolic production; that is, with ‘the physical character and existence of objects and artefacts that make them useful and usable for certain purposes under particular conditions’ (Lievrouw, 2014, p. 25). Many scholars have adopted the concept of affordances to describe broadly ‘what material artifacts such as media technologies allow people to do’ (Bucher and Helmond, 2018, p. 235). However, as means of symbolic production, we are less concerned with the instrumental use of media artefacts as the practical means of achieving a certain outcome and more with what such a usage implies or represents in broader terms, for example by enacting obedience or subversiveness. For this purpose, Engin Isin and Evelyn Ruppert introduce the concept of ‘digital acts’, which: involve conventions that include not only words but also images and sounds and various actions like liking, coding, clicking, downloading, sorting, blocking and querying. If Austin showed how we do things with words, we also try to show how we do words with things. (2015, p. 13)

The digital performance of populism  373

In populism studies, populist politicians’ use of digital features such as the recirculation of content created by citizens have, for example, been interpreted as performances that convey that ordinary citizens’ voices matter (Baldwin-Philippi, 2019, pp. 381–382). Populists can, in this way, use digital media as means of symbolic production to enact their idea of ‘the people’ and its role in democratic politics. The other material resource in a political performance is the performer’s body itself. Embodied actions – such as gesture or facial expression – symbolically convey meaning to an audience. In digital performances, embodiment does not lose its significance because a performance is virtual. It may be transposed to multimedia formats that afford some and not other aspects of real-life embodiment and shape them in certain ways. However, while digital performances lose certain embodied features and may therefore in a sense distance the performer from their audience, they also gain new ones through the technology they act through and within. Marshal McLuhan famously referred to media in this sense as ‘the extensions of man’ (1994). The varied aesthetic practices involved in the creation of memes, for example, rely on media materiality to convey meaning in ways that bodies cannot: in spite of, or perhaps by virtue of, their anonymity, memetic performances signal membership of a subculture and meaning through transmissibility (Venturini, forthcoming). Some features of media materiality therefore also bring performers and audience closer together by eroding the distinction between them. The contribution of materiality to the construction of the social is not a unidirectional process. Avoiding both social and technological determinism, the perspective of co-production couples the appreciation of the importance of media materiality in shaping the social order with the idea of technology itself as socially constructed. Sheila Jasanoff describes this interplay: [T]he ways in which we know and represent the world (both nature and society) are inseparable from the ways in which we choose to live in it. Knowledge and its material embodiments are at once products of social work and constitutive of forms of social life; society cannot function without knowledge any more than knowledge can exist without appropriate social supports. (2004, pp. 2–3)

She further argues that technology in particular embeds, and is embedded in, the social rather than being a mirror of reality (2004, p. 3). When political performers use media, they use them as material artefacts but also rely on the social ideas embedded within them. This in turn brings us to media as ideational resources of political performance. Any social performance, argues Alexander, relies on systems of collective representations, which are the mythological and ideational resources shared by a community and used by actors as symbolic reference points (Alexander, 2004b, p. 530). Such resources are often specific to the cultural context of a given locality, time or community, but they can also be specific to the media that performers use. Sheila Jasanoff’s concept of sociotechnical imaginaries is useful to describe the collective representations that users associate with digital media technologies. She defines these as ‘collectively held, institutionally

374  Handbook of digital politics

stabilized, and publicly performed visions of desirable futures, animated by shared understandings of forms of social life and social order attainable through, and supportive of, advances in science and technology’ (Jasanoff, 2015, p. 4). While the concept of sociotechnical imaginaries refers to future-oriented and normative visions, populist collective representations are often said to reside in the past. Populist actors construct the heartland, which Paul Taggart (2000, pp. 95–98) identifies as a key characteristic of populism. The heartland represents an imaginary and idealized past that populists articulate through emotional appeals that they anchor in collective representations shared by the audience. Yet the populist heartland is also a vision of the future – a golden past that ‘the people’ will return to under the populist leader’s guidance, as reflected in Donald Trump’s slogan of ‘Make America Great Again’ and the Brexit Leave campaign’s ‘Take Back Control’. In their use of digital media, populists therefore may tap into popular conceptions of media technology in their construction of the heartland and thereby lend imaginaries new meaning. As Paolo Gerbaudo (2014) argues, so-called ‘Populism 2.0’ integrates techno-utopian assumptions of Web 2.0 with ideas of more direct, participatory democracy. In a more specific cultural context, the South African populist party the Economic Freedom Fighters (EFF) connects this same sociotechnical imaginary to the common narrative of a second liberation from (neo-)colonialism, an escape from the clutches of economic imperialism aided by digital media technologies (Sorensen, 2021, Chapter 11). Common public assumptions about the role that digital media can and should play in society, often solidified by platform owners’ financially opportune narratives (Van Dijck, 2013), thus become additional resources for populist performers as they meld with populist ideas. Media technology – whether in its imaginary or material dimension – is not the only aspect of digital media that has the potential to substantially shape the representation of reality. Roger Silverstone further identifies audiences and mass media institutions as other such ‘sites of mediation’ (Silverstone, 2005, p. 189). With respect to the latter, journalistic gatekeepers still have an important role to play in the mediation of content in the hybrid media system as they amplify and sustain attention on an issue in ways that short-term eruptions on social media platforms do not (Langer and Gruber, 2021). Moreover, they are often part of the audience to which political actors appeal on social media platforms and thereby influence what content is presented to a broader audience and how. From the perspective of the strategic populist politician, material and ideational resources – in the form of media technologies and imaginaries – can be used to shape the performance of populist ideas. Populists use these performative resources of mediation to simultaneously appeal to and limit the distortion or disregard by media gatekeepers and audiences by shaping their content accordingly.

The digital performance of populism  375

FOUR TYPES OF POPULIST MEDIATED PERFORMANCE Building on the framework of Lone Sorensen (2021, Chapter 12), this section outlines four types of populist mediated performance, which we apply to digital media. First, performances within digital media are performances that make use of the affordances of digital media for utilitarian purposes. This allows populists to distribute their message, for example by bypassing information gatekeepers or by gaining their attention. Second, performances to the media are digital performances directed at an audience of media institutions or gatekeepers (thus overlapping with performances within digital media). Third, performances of mediation refer to symbolic performances of, for instance, directness and intimacy that attach additional meaning to the populist performance. Fourth, performances about mediation seek to cast doubt on the legitimacy of information providers. Within these four partly overlapping categories, this section addresses the digital performances of populist leaders, drawing upon previous literature and presenting contextual examples. Performances within Digital Media Performances within digital media refer to instances in which populists use the affordances of digital media for practical purposes to distribute their message, more or less directly, to their followers (Sorensen, 2021, p. 259). Performances within media more broadly may include opinion pieces in newspapers, news interviews or radio shows. We here focus on performances within digital media, such as social media posts by populist leaders. The affordances of digital media platforms have granted political leaders the ability to garner their own following, thus facilitating performances that avoid the gatekeeping and editorializing of legacy media (Ernst et al., 2019a, p. 5). This has led to ‘political actors no longer needing journalists to get their message out, they can simply produce media content and circulate it to large audiences themselves’ (Crilley and Gillespie, 2019, p. 174). The ability to bypass information gatekeepers is significant for populist leaders as they often construct legacy media as conduits of the views of ‘the elite’ in performances about mediation (see below). Digital platforms such as Facebook and Twitter have in recent years become almost synonymous with populist communication (Schürmann and Gründl, 2022, p. 1). To quote Donald Trump after his 2016 US presidential victory, ‘I think that maybe I wouldn’t be here if it wasn’t for Twitter … I have close to 100 million people watching me on Twitter … I have my own form of media’ (Demata, 2018, p. 67). Trump is far from the only notable example of a populist politician using digital media to bypass legacy media. For instance, Prime Minister Narendra Modi of India and President Jair Bolsonaro of Brazil extensively used social media for such purposes during their respective election campaigns in 2014 and 2019 (Cesarino, 2020, p. 407; Sinha, 2017). The material properties of digital media, in the form of networked flows of information, facilitate this opportunity for populist content to spread to a broad audience. In digitally networked spaces, emphasis is placed on the degree of engagement with

376  Handbook of digital politics

content (Blassnig et al., 2019). Pablo González-González and colleagues (2022, p. 3), for example, highlight the importance of users sharing populist content. The spread of populist material through the implicit or explicit endorsement in social networks can potentially positively impact users’ attitudes towards the original message by means of ‘network trust’, an ‘accumulated perception of “personalized”, individual trust’ (Quandt, 2012, p. 14) as distinct from the thinner, more generalized institutional trust that people place in legacy media. In addition, such spaces enable populist leaders to establish interactive relations between senders and receivers (Mazzoleni and Bracciale, 2018, p. 3), challenging the top-down forms of communication associated with legacy media. Despite this, Silvio Waisbord and Adriana Amado (2017, p. 1342) find that populist leaders communicate on digital media platforms in a top-down manner that is ‘not essentially different from the “hegemonic” political communication style they often criticise’. This finding is reiterated by Benjamin Moffitt (2018, p. 37) who suggests that populist online communication is often one-sided, with little interaction with ‘the people’ populists claim to have a direct connection with. Instead, populist performances within digital media are primarily symbolic in their association with imaginaries of directness and emancipation (Sorensen, 2021, Chapter 11). We discuss this below in the context of performances of and about mediation. Performances to the Media Populists do not only use digital media to bypass gatekeepers but also as a means of targeting journalists. In this section, we discuss populist performance to the media, which refers to performances on digital media that are aimed at legacy media in the hope of further coverage. Despite the advantages of communicating directly with citizens on digital platforms, legacy media coverage has the benefit of amplifying the populist message to a wider audience, particularly those who do not use social media. Further, legacy media coverage can amplify the salience of issues and sustain attention (Langer and Gruber, 2021, p. 313). This is important because digital spaces are often highly fragmented, and issues generally only receive a limited degree of attention for a limited amount of time. Populist digital performances, therefore, highlight features of populist messaging that are likely to draw the attention of legacy media. Frank Esser and colleagues (2017) identify a stylistic compatibility between media logic – that is, the processes and routines that govern legacy media production (Altheide and Snow, 1979) – and populist performance. This concord makes legacy media vulnerable to populist messaging without the conscious intention of the journalist. Specifically, this compatibility is inherent in the emotional, negative and dramatized discourse of legacy media (particularly tabloid media), which is also characteristic of populism. For example, Silvio Waisbord and Adriana Amado (2017, p. 1330) demonstrate how populist leaders in Latin America used social media posts to strategically influence the media and public agenda.

The digital performance of populism  377

Especially the disruptive characteristic of populist performance enables populism to form a functional relationship with legacy media. As Sorensen (2021, p. 259) argues, populist performances to the media can be ‘directly or indirectly addressed to media through live disruptions, press briefings, interviews or social media posts that remediate disruptive performances or other content’. Populist digital performances may therefore aim to create media spectacles so as to encourage legacy media institutions to advance the symbolic message inherent in their disruption of seemingly illegitimate norms and behaviours. For example, Douglas Kellner (2017, pp. 1–2) argues that Donald Trump’s 2016 social media use became fodder for legacy media, ‘creating daily spectacles of political attack, insulting and negatively defining opponents, thus helping to construct daily media events through which he was able to define the news agenda’. In this way, populists are able to use social media to ‘occupy an ambiguous position of deriding the role of the media in politics, which bolsters their self-connected authenticity, whilst catering to the media’s needs and wants’ (Sorensen, 2021, p. 200). When legacy media remediate the digital performances of populist leaders, they often play an interpretive role (Krämer, 2014, p. 49; Wettstein et al., 2018, p. 479). Such interpretation and recontextualization mostly presents populism in a negative light (Wettstein et al., 2018). However, there is evidence that negative coverage remains beneficial to populists (Esser et al., 2017, p. 366), especially when it concerns populist performances of mediation (see below) that delegitimize legacy media themselves (Caplan and Boyd, 2018, p. 67; Ernst et al., 2016). During the 2016 European Union referendum in the UK, Tong and Zuo (2021, Chapter 3) demonstrate a notable example of this interpretive role of legacy media on Twitter. They show that legacy media mainstreamed populism within the discussion of Brexit on Twitter as they shared content that conveyed the stance that Britain should exit the European Union. Specifically, they found that the news organizations gave the populist claims of Nigel Farage and Boris Johnson extensive attention and publicity, often directly citing their tweets. They highlight how the tweets of right-wing Leave-supporting newspapers uncritically and favourably quoted populist politicians ‘giving legitimacy to populist claims’ (Tong and Zuo, 2021, p. 44). Moreover, they demonstrate that such claims lacked legitimate challenge in the digital media posts of pro-Remain news organizations. In this example, the news published by pro-Leave legacy media legitimized and boosted the attention of the populist messaging within the platform. As digital media platforms have allowed groups such as alternative media and citizen-journalists to challenge the legitimacy of the legacy media (Jungherr et al., 2019), we also need to consider populist performances that may be aimed at non-traditional gatekeepers. On digital media platforms, there is evidence to suggest that populist performance has a particular resonance with alternative media organizations. For example, Philipp Müller and Anne Schulz (2021) demonstrate that alternative media regularly amplified the messages of the Alternative for Germany (AFD) on social networking sites. Alternative media may play a significant role within the digital media performance of populist leaders, particularly considering their considerable digital reach (Rae, 2021, p. 1121).

378  Handbook of digital politics

Performances of Mediation Populist performances of mediation refer to performances in digital media where the mediation of the performance itself is symbolic. This type of performance chiefly relies on sociotechnical imaginaries, for instance of directness and intimacy, to attach meaning to the populist performance. For example, a populist may send a tweet that addresses an imaginary audience of ‘the people’ directly and thereby prompt association with the imaginary of directness and disintermediation, even though the tweet remains mediated by the technology and its norms of use, the political economy of platform ownership, algorithmic gatekeeping, and so on. The digital medium thereby becomes the message through a symbolic performance. Benjamin Moffitt (2018, p. 35) puts it that ‘much of [populists’] appeal relies on the appearance of “directness” and being in touch with “the people”, and the tools of social media more easily play into, reinforce and amplify this appearance’. One way in which this directness is constructed is through a symbolic discourse of emancipation. Populist digital performances of emancipation suggest that social media may facilitate communication that can bypass legacy media and their perceived elite bias (Ernst et al., 2019b, p. 170). Populist leaders are able to seemingly directly address the interests and views of the virtuous ‘people’. However, the construction of the ‘people’ is purely symbolic, as is the directness. As Benjamin Moffitt and Simon Tormey (2014, p. 389) highlight, when ‘populists claim to speak in the name of “the people”, they are attempting to bring a subject called “the people” into being: they produce what they claim to represent by covering up the aesthetic gap and claiming to have direct, immediate contact with “the people”’. Two examples are the Twitter practices of Donald Trump during the 2016 US presidential election and Jeremy Corbyn during the 2017 UK general election. Angelos Kissas sees them as enacting ‘the people’ as a ‘concrete political subject’ through an ‘emotionally driven recontextualisation of certain, pre-existing, meanings of a people and its enemies’ (2020, pp. 268–269). The construction of the ‘people’ allowed both leaders to form a symbolic relationship of directness with their intended audience. While both politicians differed in their ideological stance and background, their discourse shared a longing to represent the views of the ‘left-behind’ (Kissas, 2020, pp. 273, 276). Digital performances of mediation can pertain to populists’ self-representation as well as their representation of ‘the people’. With respect to the former, populist leaders often enhance their mediating ability through a performance of ordinariness and authenticity, in which the populist leader is viewed as ‘one of us’ (Bracciale et al., 2021, p. 1477). A notable example of ordinariness can be observed in Bolsonaro’s use of Instagram, which seeks to construct an image of ‘an ordinary man, extraordinarily occupying the presidency’ (Mendonça and Caetano, 2021, p. 212). Ordinariness and authenticity are often constructed using symbols and gestures that resonate with the common person. On Instagram, Bolsonaro regularly posted images of himself doing household chores, family scenes, meals and playing with his children (Mendonça and Caetano, 2021, p. 221). In doing so, the populist leader can demonstrate that he is relatable to the citizenry he claims to represent.

The digital performance of populism  379

As well as using representations of ordinariness on digital platforms, performances of mediation can also convey the authenticity of the populist leader through the symbolic use of media technology. A common method of populist authenticity is the use of the live stream, which is a performed act of directness that may allow the populist to symbolically present their ‘true self’. In the case of Bolsonaro, Leticia Cesarino (2020, p. 423) analyses the ways in which he used live streams as performances of rehearsed spontaneity. The symbolic use of live streaming technology granted Bolsonaro the ability to carefully manage his displayed persona and character due to a lack of institutional constraint (Tamaki and Burni, 2020, p. 119). Performances about Mediation Performances about mediation refer to instances in which populists use digital media to cast doubt on the legitimacy of legacy media and the information they provide. Such anti-media discourse is a common feature of online populist movements (Gerbaudo, 2018, p. 749), with digital media platforms offering populists a convenient avenue to attack legacy media (Panievsky, 2021, p. 3). In performances about mediation, populists typically present a democratic imaginary that denounces elite rule as a detriment to the common good (Blokker, 2019, pp. 541–542), with legacy media as the mouthpiece of such elite groups (Waisbord and Amado, 2017, p. 1331). Specifically, anti-media populism often centres around the notion that ‘intermediaries stand in the way of the implementation of the popular will’ (Holtz-Bacha, 2021, p. 225). This has the potential to increase the level of distrust towards legacy media organizations (Hameleers, 2022, p. 9), thereby undermining their mediating ability and granting populists a perceived monopoly on the truth (Sorensen, 2021, p. 211). Much research on anti-media populism articulated through digital media is concerned with countries in which politicians and journalists typically have a high level of interdependence. Arjen van Dalen (2021) argues that, in such systems, digital media have accelerated populist politicians’ ability to challenge the genuinely democratically problematic nature of this interdependence. Populist politicians often use digital platforms to undermine the legitimacy of legacy media. In turn, legacy media often respond by questioning the legitimacy of populist politicians. For example, Hanfu Zhang and colleagues (2020, p. 1277) find that Donald Trump used digital media to draw a picture of ‘corrupted establishment politicians colluding with the mainstream media to betray the interest of the people’. Moreover, Sorensen (2021) highlights how the tweeting practices of the United Kingdom Independence Party (UKIP) and the Economic Freedom Fighters (EFF) sought to expose the close relationship between the political elite and legacy media and present the latter as ‘biased and unobjective’ (2021, p. 244). Anti-media populism can, however, also be observed within countries that do not enjoy the same degree of media–politics interdependence. While the reasons for anti-media populism are not fully understood within this context, it may be surmised that populists, similarly to those in countries with a greater level of media-politics interdependence, seek to eliminate competitors in their desire for a monopoly on

380  Handbook of digital politics

truth-telling. As Silvio Waisbord and Adriana Amado (2017, p. 1338) demonstrate, Ecuadorian President Rafael Correa regularly used Twitter to attack journalists, often using vitriolic labels such as ‘cowards’, ‘liars’ and ‘corrupt’ to question their expertise. Philippine President Rodrigo Duterte used social media to attack legacy media and cast doubt on their role as information providers, using expressions such as ‘fake news’, ‘presstitute’ and ‘bayaran’ (a term that suggests the elite pays journalists) (Ragragio, 2020, p. 5). Indian Prime Minister Modi has used similar language on digital media to attack journalists and legacy media organizations (Sinha, 2017). Populist performances about mediation may also offer legitimacy to a particular media organization or journalist. For example, Donald Trump regularly tweeted about Fox News correspondent Sean Hannity (Al-Rawi, 2019, p. 700; Ott and Dickinson, 2020, p. 633) whom he designated as a trustworthy information source, distinguishing him from other legacy media sources that Trump deemed to be unable to accurately mediate his message (for example, Ross and Rivers, 2018, p. 6). Hannity in turn used his position at Fox News to amplify several of Trump’s challenges to legacy media (Young, 2020, p. 9) and to support Trump through his social media usage (Cornfield, 2017, p. 235). Hybrid and Overlapping Performances within and to Media and of and about Mediation The categories of populist performance outlined above typically overlap, and the distinctions between them are primarily analytical. For example, a populist may seek to demonstrate their own truth-telling faculties through performances of mediation on digital media whilst simultaneously discrediting the truth-telling faculties of legacy media in performances about mediation. Equally, they may simultaneously enact performances within and to the media by using digital media to appeal to an audience of journalists. Populist mediated performances are also rarely digital-only but rather take the form of assemblages of acts in a hybrid media system (Chadwick, 2013) in which digital acts complement and support other mediated formats, and vice versa. Hybrid mediated performances should not be considered merely as the coexistence of differently mediated acts but as assemblages of individual acts that interact with one another’s forms of mediation in the media ecology (Sorensen, 2021, p. 263). We end this section by offering some contextual examples of such hybrid performances. The digital media strategy of UKIP is one example of simultaneous performances within and to the media that also integrates digital and non-digital forms of mediation. Laura Alonso-Muñoz (2020, p. 512) finds that UKIP sought to augment the presence of its candidates’ performances in legacy media broadcast interviews, using digital media to encourage users to tune into the legacy media broadcast. When the broadcast was live, UKIP selectively shared excerpts of their candidate’s interviews to the media (for example, a television debate in which Nigel Farage was a participant) in social media posts (performances within media). In these posts, they interacted with the original performances by adding context and fragmenting their position on issues, with each tweet containing only one subject. This hybrid media strategy allowed

The digital performance of populism  381

UKIP to concentrate attention on objectives pursued by the party or its position on a particular issue. In doing so, they could also limit the impact of media coverage of trivial issues (Alonso-Muñoz, 2020, p. 513). Another example of overlapping types of mediated performance is the tweeting patterns of Donald Trump. Ramona Kreis (2017, p. 614) demonstrates how Trump used social media during his presidency as a platform to critique the motives of the mainstream media, whilst simultaneously portraying himself as close to the people. In doing so, Trump was able to ‘corroborate his constructed position of an outsider and legitimate representative of the people, distancing himself from the establishment’ (Kreis, 2017, p. 614). Trump thereby enacted simultaneous performances of and about mediation. He was able to present a symbolically direct message of emancipation while raising doubts about the ulterior motives of legacy media platforms.

CONCLUSION In this chapter, we have outlined an analytical scheme for approaching populist digital performance. The premise of our argument is that populist performances represent ideas through embodied action. Such performance is a highly contextual practice. Populism, therefore, draws on both its mediated and political contexts: it intersects with, embeds itself in and intertextually references both media and democratic imaginaries to communicate its ideas in a way that resonates with audiences. It manifests these through embodied symbolic action that is constrained and resourced by the given political and democratic norms and systems and the materiality of media. Digital performances of populism represent a particular type of intersection between the media and political ecologies. Firstly, populism harmonizes with sociotechnical imaginaries, which it can accordingly harness to enhance its message but also to reinterpret established democratic imaginaries. Secondly, by its nature, populism transgresses political norms. Digital performances of populism then legitimize populism’s disruptive nature by evoking sociotechnical imaginaries through the instrumental but primarily symbolic use of the materiality of digital media. Digital performance thereby serves to enhance and legitimize the populist project in unique ways. Building on the framework developed by Sorensen (2021, Chapter 12), this chapter set out four types of populist digital performance: performances within and to the media, and performances of and about mediation. While the performances often intersect and should not be considered mutually exclusive, we hope that these analytical categories can inform further research into digital and hybrid mediated articulations of populism. To aid this purpose, we sifted and organized existing research on digital populism accordingly and illustrated the nature of the categories with examples of populist digital performance from a range of political contexts. In the final part of this conclusion, we summarize the four categories as they pertain to digital populism.

382  Handbook of digital politics

Populists often use digital media to conduct performances within media, which refers to instances in which populists use digital media platforms instrumentally to bypass information gatekeepers. We argue that the material affordances of digital media have granted populists the ability to (seemingly) directly disseminate their messages to their followers. This perceived direct connection may facilitate performances of authenticity, sincerity and self-connection (Zummo, 2020). We point to examples of populist digital media use from Latin America, India and North America. However, research demonstrates that the effect of such performance is illusionary, with populists typically utilizing digital media to communicate in a top-down manner (Waisbord and Amado, 2017, p. 1342). At the same time, the digital media strategies of populists often incorporate performances to the media. Such performances use digital media with the goal of remediation. In other words, they perform to gain the attention of information gatekeepers such as journalists who can amplify the populist message. This amplification is partly through the prominent position of legacy media organizations on digital media platforms and partly through more traditional outlets. Pointing to literature exploring the 2016 EU referendum, we argue that legacy media often play an interpretive role in populist messaging. Moreover, we briefly suggest that populists may also perform to alternative media, which often have an affinity with populism. Populists also use digital media to conduct performances of mediation. This refers to instances in which populists evoke the imaginaries of directness and emancipation to attach meaning to their political performance. We point to examples from the US, UK and Brazil to highlight the significance of such performance. We also demonstrate how the construction of a ‘people’ and the self-represented ‘ordinariness’ of the populist leader are central to attaching meaning to such populist articulations. Finally, populists often use digital media to conduct performances about mediation, in which they seek to cast doubt on the legitimacy of information gatekeepers. Populist performances about mediation typically seek to discredit the truth-telling capacity of the legacy media, claiming that such groups represent ‘elite’ opinion. We demonstrate how populist leaders, in countries that have varying levels of interdependence between politics and the media, use digital media for the purpose of discrediting legacy media. Moreover, research also shows that populist leaders may designate a particular news organization or journalist as uniquely capable of relaying their news. In developing these categories of populist performance on digital media platforms, we aim to enable an understanding of the ways in which digital populism works and the role it plays in democratic processes. For example, the approach enables a more nuanced understanding of the ways in which populist leaders use and integrate the symbolic and material resources of digital media technologies to construct a certain story about politics and to disseminate this story to several distinct audiences (citizens as well as gatekeepers). Populist leaders use digital media technology as both a tool in a utilitarian sense and as a symbolic resource that establishment politics is unable to make use of to the same effect since many digital imaginaries are inherently anti-establishment.

The digital performance of populism  383

In understanding populist performance on digital media platforms, future research may also wish to consider the role of populist citizens and users, which we have not addressed here. Moreover, we have only briefly considered alternative media, which often have primarily a digital presence. These organizations are relevant as both populist actors in their own right – what Benjamin Krӓmer terms ‘media populism’ (2014) – and as the audience of populist political leaders in their performances to the media. Finally, we appeal for a more fine-grained analysis of individual digital platforms and their specific materiality and imaginaries. While we have considered digital media in its broadest understanding, the categories outlined within this chapter may be applied to specific platforms to gain a clearer understanding of the interplay between digital media technologies and populist performance.

FURTHER READING Cunningham, A. C. (2017). Populism in the Digital Age. New York: Greenhaven Publishing. Gerbaudo, P. (2018). Social media and populism: An elective affinity? Media, Culture & Society, 40(5), 745–753. Schroeder, R. (2021). Digital Media and the Globalizing Spread of Populism. In D. Y. Jin (ed.), The Routledge Handbook of Digital Media and Globalization (pp. 179–187). New York: Routledge. Sorensen, L. (2021). Populist Communication: Ideology, Performance, Mediation. Cham: Palgrave Macmillan.

REFERENCES Aalberg, T., Esser, F., Reinemann, C., Strömbäck, J., and De Vreese, C. (eds.) (2016). Populist Political Communication in Europe. New York: Routledge. Abts, K. and Rummens, S. (2007). Populism versus democracy. Political Studies, 55, 405–424. Aiolfi, T. (2022). Populism as a transgressive style. Global Studies Quarterly. https://​doi​.org/​ doi​.org/​10​.1093/​isagsq/​ksac006. Al-Rawi, A. (2019). Gatekeeping fake news discourses on mainstream media versus social media. Social Science Computer Review, 37(6), 687–704. Albertazzi, D. and McDonnell, D. (eds.) (2008). Twenty-First Century Populism: The Spectre of Western European Democracy. Basingstoke: Palgrave Macmillan. Alexander, J. C. (2004a). From the depths of despair: Performance, counterperformance, and ‘September 11’. Sociological Theory, 22, 88–105. Alexander, J. C. (2004b). Cultural pragmatics: Social performance between ritual and strategy. Sociological Theory, 22, 527–573. Alexander, J. C. and Mast, J. L. (2006). Introduction: Symbolic action in theory and practice. In J. C. Alexander, B. Giesen, and J. L. Mast (eds.), Social Performance: Symbolic Action, Cultural Pragmatics, and Ritual (pp. 1–28). Cambridge: Cambridge University Press. Alonso-Muñoz, L. (2020). The ‘more is more’ effect: A comparative analysis of the political agenda and the strategy on Twitter of the European populist parties. European Politics and Society, 21(5), 505–519. Altheide, D. L. and Snow, R. P. (1979). Media Logic. Beverly Hills, CA: Sage Publications. Baldwin-Philippi, J. (2019). The technological performance of populism. New Media & Society, 21, 376–397.

384  Handbook of digital politics

Blassnig, S., Ernst, N., Engesser, S., and Esser, F. (2019). Populism and social media popularity: How populist communication benefits political leaders on Facebook and Twitter. In D. Taras and R. Davis (eds.), Power Shift? Political Leadership and Social Media (pp. 97–111). New York: Routledge. Blokker, P. (2019). Populism as a constitutional project. International Journal of Constitutional Law, 17(2), 536–553. Bracciale, R., Andretta, M., and Martella, A. (2021). Does populism go viral? How Italian leaders engage citizens through social media. Information, Communication & Society 24(10), 1477–1494. Brubaker, R. (2017). Why populism? Theory & Society, 46, 357–385. Brubaker, R. (2020). Populism and nationalism. Nations and Nationalism, 26, 44–66. Bucher, T., and Helmond, A. (2018). The affordances of social media platforms. In J. Burgess, A. E. Marwick, and T. Poell (eds.), The SAGE Handbook of Social Media (pp. 233–253). London: Sage Publications. Bucy, E. P., Foley, J. M., Lukito, J., Doroshenko, L., Shah, D. V., Pevehouse, J. C., and Wells, C. (2020). Performing populism: Trump’s transgressive debate style and the dynamics of Twitter response. New Media & Society, 22, 634–658. Canovan, M. (1999). Trust the people! Populism and the two faces of democracy. Political Studies, 47, 2–16. Canovan, M. (2005). The People. Cambridge: Polity Press. Caplan, R. and Boyd, D. (2018). Who’s playing who? Media manipulation in an era of Trump. In P. J. Boczkowsk and Z. Papacharissi (eds.), Trump and the Media (pp. 60–69). Cambridge, MA: MIT Press. Cesarino, L. (2020). How social media affords populist politics: Remarks on liminality based on the Brazilian case. Trabalhos em Linguística Aplicada, 59, 404–427. Chadwick, A. (2013). The Hybrid Media System: Politics and Power. New York: Oxford University Press. Cornfield, M. (2017). Empowering the party-crasher: Donald J. Trump, the first 2016 GOP presidential debate, and the Twitter marketplace for political campaigns. Journal of Political Marketing, 16, 212–243. Couldry, N. (2015). The myth of ‘us’: Digital networks, political change and the production of collectivity. Information, Communication & Society, 18, 608–626. Crilley, R. and Gillespie, M. (2019). What to do about social media? Politics, populism and journalism. Journalism, 20(1), 173–176. Demata, M. (2018). ‘I think that maybe I wouldn’t be here if it wasn’t for Twitter’: Donald Trump’s populist style on Twitter. Textus, 1. https://​doi​.org/​10​.7370/​89446. Engesser, S., Ernst, N., Esser, F., and Büchel, F. (2016). Populism and social media: How politicians spread a fragmented ideology. Information, Communication & Society, 20, 1109–1126. Ernst, N., Engesser, S., and Esser, F. (2016). Switzerland: Favourable conditions for growing populism. In T. Aalberg, F. Esser, C. Reinemann, J. Strömbäck, and C. De Vreese (eds.), Populist Political Communication in Europe (pp. 151–164). New York: Routledge. Ernst, N., Blassnig, S., Engesser, S., Büchel, F., and Esser, F. (2019a). Populists prefer social media over talk shows: An analysis of populist messages and stylistic elements across six countries. Social Media + Society, 5(1), 2056305118823358. https://​doi​.org/​10​.1177/​ 2056305118823358. Ernst, N., Esser, F., Blassnig, S., and Engesser, S. (2019b). Favorable opportunity structures for populist communication: Comparing different types of politicians and issues in social media, television and the press. The International Journal of Press/Politics, 24(2), 165–188. Esser, F., Stępińska, A., and Hopmann, D. N. (2017). Populism and the media: Cross-national findings and perspectives. In T. Aalberg, F. Esser, C. Reinemann, J. Strömbäck, and C.

The digital performance of populism  385

De Vreese (eds.), Populist Political Communication in Europe (pp. 365–380). New York: Routledge. Freeden, M. (2017). After the Brexit referendum: Revisiting populism as an ideology. Journal of Political Ideologies, 22, 1–11. Gerbaudo, P. (2014). Populism 2.0: Social media activism, the generic Internet user and interactive direct democracy. In D. Trottier and C. Fuchs (eds.), Social Media, Politics and the State: Protests, Revolutions, Riots, Crime and Policing in the Age of Facebook, Twitter and YouTube (pp. 67–87). New York: Routledge. Gerbaudo, P. (2018). Social media and populism: An elective affinity? Media, Culture & Society, 40(5), 745–753. González-González, P., Marcos-Marne, H., Llamazares, I., and Gil de Zúñiga, H. (2022). The informational consequences of populism: Social media news use and the ‘news finds me’ perception. Politics and Governance, 10. https://​doi​.org/​10​.17645/​pag​.v10i1​.4772. Hameleers, M. (2022). Empowering the people’s truth through social media? (De)legitimizing truth claims of populist politicians and citizens. Politics and Governance, 10(1). Holtz-Bacha, C. (2021). The kiss of death: Public service media under right-wing populist attack. European Journal of Communication, 36(3), 221–237. Isin, E. and Ruppert, E. (2015). Being Digital Citizens. London: Rowman & Littlefield. Jasanoff, S. (2004). States of Knowledge: The Co-Production of Science and the Social Order. New York: Routledge. Jasanoff, S. (2015). Future imperfect: Science, technology, and the imaginations of modernity. In S. Jasanoff and S. Kim (eds.), Dreamscapes of Modernity: Sociotechnical Imaginaries and the Fabrication of Power (pp. 1–33). Chicago: University of Chicago Press. Jungherr, A., Schroeder, R., and Stier, S. (2019). Digital media and the surge of political outsiders: Explaining the success of political challengers in the United States, Germany, and China. Social Media + Society, 5(3), 1–12. Kellner, D. (2017). Preface: Guy Debord, Donald Trump, and the politics of the spectacle. In M. Briziarelli and E. Armano (eds.), The Spectacle 2.0: Reading Debord in the Context of Digital Capitalism (pp. 1–14). London: University of Westminster Press. Kissas, A. (2020). Performative and ideological populism: The case of charismatic leaders on Twitter. Discourse & Society, 31(3), 268–284. Krämer, B. (2014). Media populism: A conceptual clarification and some theses on its effects. Communication Theory, 24, 42–60. Kreis, R. (2017). The ‘tweet politics’ of President Trump. Journal of Language and Politics, 16(4), 607–618. Langer, A. I. and Gruber, J. B. (2021). Political agenda setting in the hybrid media system: Why legacy media still matter a great deal. The International Journal of Press/Politics, 26(2), 313–340. Lievrouw, L. A. (2014). Materiality and media in communication and technology studies. In T. Gillespie, P. J. Boczkowski, and K. A. Foot (eds.), Media Technologies: Essays on Communication, Materiality, and Society (pp. 21–51). Cambridge, MA: MIT Press. Mazzoleni, G. and Roberta Bracciale, R. (2018). Socially mediated populism: The communicative strategies of political leaders on Facebook. Palgrave Communications, 4(1), 1–10. McLuhan, M. (1994). Understanding Media: The Extensions of Man. Cambridge, MA: MIT Press. Mendonça, R. F. and Caetano, R. D. (2021). Populism as parody: The visual self-presentation of Jair Bolsonaro on Instagram. The International Journal of Press/Politics, 26(1), 210–235. Mény, Y. and Surel, Y. (2002). The constitutive ambiguity of populism. In Y. Mény and Y. Surel (eds.), Democracies and the Populist Challenge. Basingstoke: Palgrave Macmillan. Moffitt, B. (2016). The Global Rise of Populism: Performance, Political Style, and Representation. Stanford, CA: Stanford University Press.

386  Handbook of digital politics

Moffitt, B. (2018). Populism 2.0: Social media and the false allure of ‘unmediated’ representation. In G. Fitzi, J. Mackert, and B. S. Turner (eds.), Populism and the Crisis of Democracy, vol. 2 (pp. 30–46). New York: Routledge. Moffitt, B. and Simon Tormey, S. (2014). Rethinking populism: Politics, mediatisation and political style. Political Studies, 62(2), 381–397. Mudde, C. (2004). The populist zeitgeist. Government and Opposition, 39, 542–563. Mudde, C. (2007). Populist Radical Right Parties in Europe. New York: Cambridge University Press. Mudde, C. and Rovira Kaltwasser, C. (2013). Populism. In M. Freeden, M. Stears, and L. T. Sergeant (eds), The Oxford Handbook of Political Ideologies (pp. 493–512). Oxford: Oxford University Press, Müller, P. and Schulz, A. (2021). Alternative media for a populist audience? Exploring political and media use predictors of exposure to Breitbart, Sputnik, and Co. Information, Communication & Society, 24(2), 277–293. Norris, P. and Inglehart, R. (2018). Cultural Backlash: The Rise of Populist-Authoritarianism. New York: Cambridge University Press. Ostiguy, P. (2020). The socio-cultural, relational approach to populism. Partecipazione e Conflitto, 13, 29–58. Ott, B. L. and Dickinson, G. (2020). The Twitter presidency: How Donald Trump’s tweets undermine democracy and threaten us all. Political Science Quarterly, 135(4), 607–636. Panievsky, A. (2021). The strategic bias: How journalists respond to antimedia populism. The International Journal of Press/Politics, 27(2), 19401612211022656. https://​doi​.org/​10​ .1177/​19401612211022656. Quandt, T. (2012). What’s left of trust in a network society? An evolutionary model and critical discussion of trust and societal communication. European Journal of Communication, 27, 7–21. Rae, M. (2021). Hyperpartisan news: Rethinking the media for populist politics. New Media & Society, 23(5), 1117–1132. Ragragio, J. L. D. (2020). Framing media populism: The political role of news media editorials in Duterte’s Philippines. Journalism, 23(6). https://​doi​.org/​10​.1177/​1464884920959505. Ross, A. S. and Rivers, D. J. (2018). Discursive deflection: Accusation of ‘fake news’ and the spread of mis- and disinformation in the tweets of President Trump. Social Media + Society, 4(2). https://​doi​.org/​10​.1177/​2056305118776010. Schürmann, B. and Gründl, J. (2022). Yelling from the sidelines? How German parties employ populist and crisis-related messages on Facebook. Political Research Exchange, 4(1), 2021095. https://​doi​.org/​10​.1080/​2474736X​.2021​.2021095. Silverstone, R. (2005). The sociology of mediation and communication. In C. Calhoun, C. Rojek, and B. Turner (eds.), The SAGE Handbook of Sociology (pp. 188–207). Thousand Oaks, CA: Sage Publications. Sinha, S. (2017). Mediatized populisms – fragile hegemony: Social media and competitive electoral populism in India. International Journal of Communication, 11. Sorensen, L. (2018). Populist communication in the new media environment: A cross-regional comparative perspective. Palgrave Communications, 4, 48. https://​doi​.org/​10​.1057/​s41599​ -018​-0101​-0. Sorensen, L. (2021). Populist Communication: Ideology, Performance, Mediation. Cham: Palgrave Macmillan. Taggart, P. (2000). Populism. Buckingham: Open University Press. Tamaki, E. R. and Aline Burni, A. (2020). Populist communication during the Covid-19 pandemic: The case of Brazil’s President Bolsonaro, Partecipazione e Conflitto, 14(1). Tong, J. and Zuo, L. (2021). The Brexit Referendum on Twitter: A Mixed-Method Computational Analysis. Bingley: Emerald Group Publishing.

The digital performance of populism  387

Van Dalen, A. (2021). Rethinking journalist–politician relations in the age of populism: How outsider politicians delegitimize mainstream journalists. Journalism, 22(11), 2711–2728. Van Dijck, J. (2013). The Culture of Connectivity: A Critical History of Social Media. New York: Oxford University Press. Venturini, T. (forthcoming). Online conspiracy theories, digital platforms and secondary orality: Toward a sociology of online monsters. Theory, Culture & Society. Waisbord, S. and Amado, A. (2017). Populist communication by digital means: Presidential Twitter in Latin America. Information, Communication & Society, 20(9), 1330–1346. Wettstein, M., Esser, F., Schulz, A., Wirz, D. S., and Wirth, W. (2018). News media as gatekeepers, critics, and initiators of populist communication: How journalists in ten countries deal with the populist challenge. The International Journal of Press/Politics, 23, 476–495. Wirth, W., Esser, F., Wettstein, M., Engesser, S., Wirz, D., Schultz, A., Ernst, N., Büchel, F., Caramani, D., Manucci, L., Steenbergen, M., Bernhard, L., Weber, E., Hänggli, R., Dalmus, C., and Schemer, C. (2016). The appeal of populist ideas, strategies and styles: A theoretical model and research design for analyzing populist political communication. Working Paper No. 88, Challenges to Democracy in the 21st Century. National Centre of Competence in Research (NCCR), Zurich. Young, D. G. (2020). Irony and Outrage: The Polarized Landscape of Rage, Fear, and Laughter in the United States. New York: Oxford University Press. Zhang, H., Afzaal, M., and Liu, C. (2020). American populism in digital era: Strategies of manipulation in Donald Trump’s election tweets. Revista Argentina de Clínica Psicológica, 29(3), 1273–1280. Zummo, M. L. (2020). Performing authenticity on a digital political stage: Politainment as interactive practice and (populist?) performance. Iperstoria, 15. https://​doi​.org/​10​.13136/​ 2281​-4582/​2020​.i15​.589.

25. Political communication about data Brendan Lawson

Most of the chapters in this book engage with digital politics through the lens of digital media. That is, the way political communication exists within a digital media ecosystem. This chapter takes a different tack. It is more concerned with the way political communication is connected to digital data. This relationship has gained lots of attention over the last five to ten years. Much of the focus has been on political communication using digital data, such as the automatic production of propaganda by computers or the way voters can be organized, evaluated and targeted. These are important efforts, documented in the first section of this chapter. But what about the political communication about digital data? Covid-19 has drawn attention to the way politicians draw from digital datasets to communicate numbers to the public and the news media (Corner, 2021). Case numbers, hospitalizations and deaths were used to describe the way Covid-19 affected public health, whilst Gross Domestic Product (GDP), the number of jobs supported by furlough schemes and the financial support offered to the business sector were deployed to represent the economy. Used in combination, these two sets of figures became highly-charged in the political arena. A country that oversaw low levels of cases, hospitalizations and deaths and had GDP growth stabilize in 2020/2021 were deemed as nations operating successful Covid-19 responses. Those nations who witnessed the opposite were considered to have handled the pandemic badly (Lu et al., 2021). The deluge of data in political communication means we need to attend to the different way politics talks about the quantitative. Such a task cannot fall into a naïve empiricism that sees the use of reliable and accurate numbers by politicians as the end goal. Yes, we need numbers to be technically sound. But we also need to acknowledge the way politicians will use accurate and reliable numbers in selective, partial, and politically motivated ways. Therefore, there needs to be an emphasis on numbers as rhetorical devices. This is the focus of the second section of the chapter. But there is an important conversation beyond statistical rhetoric. Statistics are derived from datasets (often digital ones), which themselves play a particular role in political communication. As the third section highlights, the notion of open data during the early stages of the pandemic served to render the government more transparent, accountable and trustworthy whilst simultaneously frame what was important about coronavirus. The final part of this chapter focuses on processes that transform data, drawing on the example of the A-level standardization model. As students in their final year of secondary education could not be tested due to the pandemic, a standardization model was constructed to award grades based on historical data and expected student 388

Political communication about data  389

performance. But awarding grades in this way opened the door to criticism. To counter this, the algorithm underpinning the model was tied to notions of fairness and objectivity. Before we dig into these three aspects of political communication about data, let’s return to political communication using data. Much of the work on political communication and data centres on computational propaganda – a sub-discipline of political communication that points to the relationship between technological developments and the form, purpose and type of propaganda on digital platforms.

COMPUTATIONAL PROPAGANDA Perhaps the most salient type of computational propaganda is ‘algorithmically generated political content’ on social media (Howard et al., 2018, p. 84). Social media bots make use of vast banks of ‘personal records from across media properties, organizational forms, and international borders’ to automatically create content online (Howard et al., 2018, pp. 84–85). For many within political communication, this offers a threat to the core values of democratic politics. Informative, accurate and relevant information from a trusted source is replaced by misleading information from a piece of nefarious technology. The fears of computational propaganda are best captured in the work on the Russian state-supported Internet Research Agency (IRA). It was this organization that formed the focal point of a 103-page report from New Knowledge, upon request from the United States Senate Select Committee on Intelligence (SSCI) (DiResta et al., 2019). The report emphasized the scale of IRA’s operations – pointing to 10.4 million tweets, 77 million engagements on Facebook and 187 million engagements on Instagram. These extensive activities were centred on specific strategies: recruiting Black Americans as assets, voter suppression, sowing literal division, pro-Trump operations and anti-Clinton operations (DiResta et al., 2019, pp. 9–12). For many, the role of IRA in the 2016 US presidential election operates as the touchstone for the threat of computational propaganda (Bail et al., 2020). Whilst the exact role and strategies of the IRA during the run up to polling are contested, it is often taken for granted that this type of computational propaganda had a considerable effect on how people voted (Golovchenko et al., 2020). But others argue that such a cause–effect interpretation of the influence of social media content on the public is too simplistic and misguided. This reflects a broader sentiment within media studies that emphasizes the rejection of the simple media effects model – one that argues against a cause–effect explanation of how media messages influence those ‘exposed’ to such communication (see the work of Newton, 2006). The work of Anderson (2021) sits within this literature. Without delving too deeply into his argument, he would implore us to consider computational propaganda as something that is produced (this is undeniable) but does not automatically have some widespread effect on the behaviour of individuals. In pushing us towards this conclusion, he also emphasizes the way that data, algorithms, automation and artifi-

390  Handbook of digital politics

cial intelligence intersect with political communication in ways that extend beyond the algorithmic construction of political content. For example, Benoit Dillet (2020) argues that the algorithmic nature of digital communication means that orators of political rhetoric no longer have the news media or the public as their audience. Instead, they see their audience as the algorithm itself. For Dillet (2020) this is changing the form and content of political rhetoric itself. This type of reactionary political communication marks a break from the discussions above but is still within a sort of cause–effect nature of data meets political communication. So, we need to understand the way data effects communication but also communication about data. How do politicians use statistics to support their arguments? How do officials lean on data to build political projects? How do notable actors utilize broad understandings of algorithms and artificial intelligence? Such a focus is rarely addressed in political communication, but the fruits of doing so can be observed in those working outside of politics (Bucher, 2017; Fletcher and Nielsen, 2019). Let’s start with the most obvious manifestation, and the one that has received the most attention in the literature: the way politicians use specific numbers to communicate.

STATISTICAL RHETORIC Contemporary statistical rhetoric involves deriving specific numbers from digital databases. When a politician uses the annual rise in GDP for a speech, for example, they derive this number from a digital database about GDP itself. This means that whilst not all work within statistical rhetoric is focused on digital data, the key take aways from this collection of literature can be used here. At the core of the work on statistical rhetoric is the relationship between truth, politics and numbers. This dynamic is often positioned between two points of contention. On the one hand, you have politicians who use numbers as objective, undeniable facts in their communication to the public. These numbers are then used to construct a single, empirically-validated reality about which politicians talk. On the other hand, there is a wealth of literature that emphasizes the contingent and political nature of quantitative knowledge. This work pushes us to see the way quantitative information can be used by different actors to underpin a set of ‘interpretive and culturally contingent’ realities (Lawson and Lovatt, 2020, p. 113). In other words, the statistics are deployed strategically as rhetorical devices. The first rhetorical goal by political communicators is to double down on the truthfulness and undeniability of the numbers they are communicating. This is generally achieved through a process of ‘de-historicization’ (Lawson and Lovatt, 2020). Such a strategy involves omitting any discussion of sample sizes, analyses, time frames and motivations for the production of the number in question. Through erasing the history of these numbers, they can function as an ever-present fact that describes what is currently happening.

Political communication about data  391

From this foundation of certainty, a range of rhetorical strategies are then adopted by politicians. In his account of statistics in the Puerto Rican Pension Reform Report from 2013, Aviles (2016) identifies three ways numbers functioned rhetorically. The first was a process of metonymy: where the whole of an object is referred to by one of its parts. The totality of the Puerto Rico pension system – in all of its complexity – was reduced to a set of indicators about one aspect of the system: financial solvency (Aviles, 2016, p. 52). The second involved prosopopoeia: where a person is able to speak as another person or object. The author of the report regularly argued that ‘the numbers speak for themselves’ (Aviles, 2016, p. 54). In doing so, they positioned themselves as mere conduits of information that could, in fact, speak with their own voice. Such a strategy is intimately linked with the notion of de-historicization detailed above – the more a number is considered objective, the more it can speak for itself. Beyond ideas of the scientific or the rational, Aviles is also keen to stress the importance of pathos – the appeal to emotions. Here numbers are used emotively and dramatically to create what he calls a ‘statistically designed impending calamity’ (Aviles, 2016, p. 58). The list of different rhetorical strategies becomes much longer when we examine literature outside of political communication (Roeh and Feldman, 1984; McCloskey, 1987; John, 1992; Porter, 1995; Van Dijk, 2000; Kilyeni, 2013; Katchergin, 2015; Koetsenruijter, 2018). But what of the different types of numbers in statistical rhetoric? It is important to recognize that whole numbers, fractions, percentages and ratios all have a sort of mathematical syntax that dictate how and when they can be used. But these numbers are not just governed by their mathematical logic, they also take on meaning through their cultural significance. In a recent article, Billig (2021) introduced us to the idea of ‘semi-magical round numbers’ in the political communication of Covid-19. He argues that these numbers do not just indicate quantity but a certain quality too. In the context of the pandemic, he points to the Conservative government’s target of conducting 100,000 tests per day by the end of April in the UK. This number was a precise one – 100,000 tests were to be carried out – but also operated in an iconic fashion to emphasize the success of the government responses to the crisis (Billig, 2021, p. 9). Whilst there was no justification for why 100,000 should be the target, and there was notable sleight of hand to ‘achieve’ this target, Billig (2021, p. 13) encourages us to consider how rounded numbers function ‘as semi-magical markers of qualities’. Taken together, we can see an interplay between the power that politicians have when communicating numbers and the power of specific numbers to function as ‘semi-magical’. Only examining statistical rhetoric, however, would limit our scope of analysis. We can move away from the communication of numbers derived from digital data towards the communication of this digital data. More specifically, the way politics has positioned the push towards open government data (OGD).

392  Handbook of digital politics

OPEN DATA DISCOURSES Over the past 20 years, European and North American governments have transitioned through three distinct phases of data policy: first keeping their data largely private, then allowing data to be accessed through freedom of information requests (FOIs), and now actively publishing datasets on digital data portals. This final stage of data policy rests on the ability of twenty-first century technologies to both store vast banks of data and allow this data to be accessed and analysed by members of the public. This digital data is often framed within the idea of OGD ecosystem, where the data producers (the government), data users (journalists, businesses, etc.) and data beneficiaries (the public, customers, etc.) coexist (Lawson, 2022). Stepping away from a structural argument, we can think of open government as doing a certain type of ‘work’ politically (Kennedy et al., 2016). As with the section above, this can occur at a specific rhetorical level but is probably better understood as specific communicative strategies that tap into a broader discourse surrounding ‘openness’. This can be demonstrated through an example from the pandemic. Transparency, Accountability and Trustworthiness In the early stages of the pandemic in the UK, government communication was characterized by a vagueness in relation to numbers. This manifested in Boris Johnson’s metaphors of ‘flatten the peak’ or ‘squash that sombrero’ when referring to the data visualizations of cases, hospitalizations and deaths over time. The press conference from 27 March 2020, however, marked a turning point. Michael Gove opened a press conference with a series of numbers describing the current outbreak (extract 1) and, during the Q&A session with journalists, also outlined the opening up of data by the government (extract 2): Extract 1: 113,777 have now been tested for the virus, of those 14,543 have tested positive, an increase in the last day of 2,885. The best scientific analysis now is that the rate of infection has been doubling every three to four days. And of those who have contracted the virus, 759 have sadly died. (AFP, 2020) Extract 2: One of the features of today’s briefing was the set of numbers I and Simon shared with you. And of course we will share figures, share data in a spirit of transparency so people can see we are doing everything we can. (AFP, 2020)

The emphasis on specificity and openness marked a distinct communicative shift from vagueness and secrecy. But this was a change in political communication, not a policy on the data itself. The government had been publishing the data referred to above on their website throughout March 2020 (UK government, 2021). In fact, they first published statistics on those tested for Covid-19 on 24 January. Therefore, we can observe how this opening-up of data was a strategic rhetoric shift towards openness rather than the first time the government was revealing hidden data about the crisis. So, we can think of this strategy having certain goals.

Political communication about data  393

The first set of goals centre on a set of concepts that are often used interchangeably: transparency, trustworthiness and accountability. Each term, however, refers to slightly different aspects of data openness. Transparency centres on the process of opening up data – it focuses on the way governments actively take ‘hidden’ data and open it up to the public. Obviously, this process is a technical one – taking data, cleaning it, providing labels that are intelligible to the public, protecting certain sensitive information, organizing it on an open data portal, and so on – but it also operates discursively too. It emphasizes how the government is shifting from governance through opacity to governance through transparency. Such a shift is then linked to two related notions. By revealing data that was previous hidden, the government can position itself as more trustworthy as it says ‘look, we have nothing to hide’. But if you do not trust the government at face value, you can also hold it accountable by exploring and analysing the open data to see if they are actually doing a good job. We can see how political rhetoric will often make use of these three discourses at the same time. Michael Gove emphasizes that this press conference was characterized by ‘a spirit of transparency’ – marking a break from the opacity of previous ones. But he also attempted to position this transparency as automatically equating to accountability by explaining that the data shows ‘we are doing everything we can’. In other words, Gove is saying ‘you do not need to analyse the data, you just need to trust that it shows we are doing well’. Such an emphasis on discourse transparency, trustworthiness and accountability is part of a broader shift within the public sector. This process began in the 1970s with the rise of New Public Management (NPM) that attempted to use quantification, often through things like Key Performance Indicators (KPIs), to make the public sector more accountable (Dunleavy et al., 2006). The emergence of digital technologies from the 1990s onwards marked a shift in how quantification was linked to transparency, trustworthiness and accountability. Being able to store and share large amounts of data with the public can lead to intense criticism of the government. But it can also – as this section has argued – build trustworthiness, accountability and transparency (Bertot et al., 2010; Bannister and Connolly, 2014; De Blasio and Selva, 2016). Framing with Data But open data does not just make the government appear more transparent, trustworthy and accountable, it also performs an important role in framing what is known about phenomena (and, in turn, how they are governed). In his examination of Singapore’s open data project, Stevens argues that open data affects the way that social issues are managed. As Stevens (2019) explains: These apps offer deeply technocratic responses to existing social and political problems. They belong to an imagined future in which information technologies – especially data – will solve social problems.

394  Handbook of digital politics

Often the data provided fits within conventional modes of understanding these social problems. For example, the transport data made available in Singapore focused ‘almost entirely on trains, buses and especially cars’ whereas there was ‘little or no data’ about ‘alternative modes of transportation’ (Stevens, 2019). In this way, data often supports existing modes of governance rather than disrupting them (Dunleavy, 2016; Redden, 2018; Lawson, 2022). We can certainly see how this process of framing worked during the pandemic. And, through exploring framing, we can see how it is a way into critically engaging with the discourses of transparency, trustworthiness and accountability documented above. When we return to the press conference from 27 March 2020, we can see how communicating open data gives the impression that Gove is revealing a hidden reality. He outlines the total number of people who have been tested, those who have tested positive, the daily increase in positive tests, the doubling time and the total number of people who have died. This ‘reality’, however, is incredibly limited. First, we only have quantitative information – no personal stories, images of crowded wards or the smells of disinfected care homes. Second, we have a quantitative snapshot that talks only of national-level information about tests, positive cases, deaths and spread. It is true that this macro-level data centres on the important topics: those falling ill with the virus, the level of testing available, those dying from the virus and the way the virus is spreading through the community. But each indicator serves to frame the pandemic. The news media were fairly good at challenging this openness, generally following two tactics: asking for more openness with the data and challenging the accuracy and validity of the data already made available. Even in the conference from 27 March, journalists were critiquing the supposed openness of the government’s data. This was largely because the data that was being communicated had been available to journalists before the press conference – with many news organizations providing dashboards of the available data on cases, deaths and tests. So, the questions from journalists were about the data that still was not available, such as the data on social distancing measures. This challenge to the scope of openness was part of a broader discourse that said the government was not going far enough. An article in the Guardian interviews Jeni Tennison, vice-president of Open Data Institute (Tennison, 2021). She argued that ‘there is currently no open data available on UK hospitalization rates; no regional age or gender breakdown of daily deaths’. Whilst the author recognizes that not all data should be published like in Singapore, for example, they do point to other initiatives by European governments. On the basic Covid-19 figures, some best practices are developing across the world. Italy has been publishing open data daily on GitHub since the beginning of March, with regional breakdowns, and numbers of people self-isolating, hospitalized and in intensive care. Belgium is providing province-level open data on cases and deaths, broken down by gender and age group, and numbers of people in hospital, ICU, and receiving respiratory support. (Tennison, 2021)

Political communication about data  395

These wide-ranging calls for openness were met with a sort of reactive drip-feed of open data. The request from the journalist on 27 March about data on social distancing was met with data visualizations on 30 March. This involved a line chart of different public and private transport usage from the beginning of the lockdown measures. Except for motor vehicle usage, the level of transport use was generally decreasing substantially (UK government, 2020). As the extent of deaths in care homes from Covid-19 became apparent in April, the government started publishing data on deaths in care homes. This was only made publicly available, however, from 28 April onwards (and included in their press conferences from 29 April). Not only was the government accused of releasing too little data too slowly, the data they did release involved a level of contradiction and confusion. This was most obviously observed in the interpretation of data about deaths from Covid-19. On 29 March, a reporter asked about the discrepancies between Department for Health and Social Care figures for daily deaths and those released by NHS England. Jenny Harries – the Deputy Chief Medical Officer for England – was keen to stress that ‘this is not an issue of transparency’ but one about ensuring that the family know first ‘and all the public parts of the system to consolidate the information and make sure it is accurate’. But such a response seemed to provide little explanation for the differences between the reported deaths from the government and the number of deaths recorded by the ONS on death certificates (Sky News, 2020). In fact, an analysis by Sky News argued that the official figure given by the government under-reported deaths by as much as 22 per cent when compared to the ONS figures. This framing approach allows us to see the holes in the discourses of transparency, trustworthiness and accountability. If data is released to provide a sort of data-framing to the world that is governed, what level of transparency, trustworthiness and accountability is actually achieved? Much like the conversation about statistical rhetoric this tension point means that governments often double-down on discourses of transparency, trustworthiness and accountability that open data affords. And to a large extent, the government’s political communication shift to ‘open data’ was successful in conferring transparency, accountability and trustworthiness. The next section of this chapter, however, is about a political failure.

TRANSFORMING DATA If the first section focused on numbers derived from data and the second examined the opening up of datasets themselves, the third part considers the way data is transformed by algorithms. This process sits at the heart of computation propaganda – where vast amounts of digital data is used to create political messages – but the focus here is about how politicians talk about algorithms. There is a real paucity of literature on this strand of political communication and data, so the final section will be based entirely on an example from the pandemic. Due to the prolonged closure of schools during the first wave of the pandemic, the UK government announced that A-level exams would be cancelled for 2020 (Ofqual,

396  Handbook of digital politics

2020c). In their place, the Office of Qualifications and Examinations Regulation (Ofqual) announced that they would be awarding grades through a ‘standardization model’ that was aimed at making ‘grades as fair as they can be’ (Jadhay, 2020). Underpinning this model was an algorithm that took a set of input data (rankings and historical performance) and produced output data (the students’ grades). The workings of the algorithm are important to understand. Each centre (a college, school or sixth form) would submit, subject by subject, a ranking of all the students in the class. This ranking would then be put into the standardization model that would operate at a subject level (rather than a centre level). For each subject, the model would consider the historical results from 2017, 2018 and 2019 to judge whether the ‘centre assessment grades are more generous or severe than predicted’. This would mean increasing or decreasing the ranking order provided by the teacher. The hope was to keep the 2020 grades in line with averages from previous years (Jadhay, 2020). The two main hopes for this standardization model rested on the rankings and historical data outlined above. We can see this play out in the official documents produced by the government. The teachers submitting their rankings were encouraged to be as ‘objective’ in their judgements as possible. Such was the emphasis upon ‘objective’ judgements that Ofqual produced specific official guidance. The document implored teachers to ‘only take account of existing records and available evidence of a student’s knowledge, skills and abilities in relation to the subject’ (Ofqual, 2020b). But Ofqual were keen to stress that they did not just rely on the objectivity of the teachers’ judgement. The model standardized this initial data using historical grading information. In doing so, the model was promised ‘to bring consistency to teacher judgements across all schools and colleges, and to make sure results are comparable with previous years’ (Ofqual, 2020a). Therefore, the communication around the standardization model emphasized objectivity and fairness. Unlike open data, however, this political project was a resounding failure. On 7 August, the government announced that ‘nearly 40% of A-level results submitted by teachers are set to be downgraded’ (Adams, 2020). In the face of public outrage, the government published a 319-page methodology document on 13 August that revealed the inner workings of the standardization model that was outlined in May (Ofqual, 2020d). Instead of allaying fears, this signalled the end of the algorithm itself. On 17 August, four days after the methodology document was produced, the Education Secretary announced that all students would be awarded the grade predicted by their teacher and all those students who were awarded a higher grade by the algorithm would be allowed to keep that too (Daventry, 2020). This story emphasizes the need to decentre communication. Yes, politicians can use statistics rhetorically or open data as a framing device. But when they talk about quantitative transformations, where existing data is turned into new data, they have to deal with the consequences of these changes. The discourses of objectivity and fairness surrounding the A-level standardization model collapsed because the model itself created a reality that was deemed deeply unfair and subjective. There is only so

Political communication about data  397

much work communication about transformations can do before the transformation itself changes what can be communicated. But the power of these transformations is culturally dependent too (Bucher, 2018). The A-level algorithm was scrapped due to a public outcry that stretched across geographical, class and political boundaries. But what about discriminatory transformations that are not changed because there is no large-scale outcry? Here we can think of the deployment of automatic facial recognition software in law enforcement that consistently mis-recognizes non-white faces or the way opaque credit scoring systems reinforce socio-economic divisions through denying or accepting mortgage applications (Browne, 2015). The discourses surrounding these transformations do not collapse under the weight of the transformation itself.

CONCLUSION So, how can these three ‘ways in’ to political communication about data be used to develop the field? The quantitative must be seen as multi-faceted and inter-linked. Specific statistics, including ‘semi-magical round numbers’, are derived from open or closed datasets – and these datasets can be transformed to create new datasets, from which specific statistics can be created. These different elements of the quantitative have to be set within a specific-general scale of communication. The 100,000 tests per day statistic was deployed with a specific purpose to emphasize the success of the government’s testing programme, whilst the open digital data published and talked about by politicians served to frame the pandemic itself. But for the quantitative to be deployed in this specific fashion, broader discourses needed to be articulated and reproduced too. In respect to open data, this rested on the triad of trustworthiness, accountability and transparency. The A-level standardization model, on the other hand, was set within objectivity and fairness. Within this assemblage of politicians, communication and the quantitative, the chapter provides some critical tools and lessons. First, the rhetorical strategies – de-historicization, prosopopoeia, metonymy and others – can be applied to the way politicians use statistics. Second, framing can be used to assess how much transparency, accountability and trustworthiness emerges from open data projects by governments. Third, the communication about different transformations, e.g. algorithms, models, AI etc., should be set within a culturally-contextualized idea of acceptable and unacceptable effects of the transformation themselves. Each of these ‘ways in’ provide a starting point to explore a political communication system. It is only through their combination, however, that scholars can understand how the quantitative functions. Just as statistics, data and transformations are linked as technical entities, they must be linked in political communication. In doing so, scholars can begin to understand how this tide of data in politics becomes meaningful.

398  Handbook of digital politics

FURTHER READING Aviles, L. A. (2016). The art of public policy statisticians: The case of the Puerto Rico pension reform report. Caribbean Studies, 44(1–2), 47–68. Billig, M. (2021). Rhetorical uses of precise numbers and semi-magical round numbers in political discourse about COVID-19: Examples from the government of the United Kingdom. Discourse & Society, 32(5). https://​doi​.org/​10​.1177/​09579265211013115. De Blasio, E. and Selva, D. (2016). Why choose open government? Motivations for the adoption of open government policies in four European countries. Policy & Internet, 8(3), 225–247. Deringer, W. (2018). Calculated Values: Finance, Politics, and the Quantitative Age. Cambridge, MA: Harvard University Press. Dillet, B. (2020). Speaking to algorithms? Rhetorical political analysis as technological analysis. Politics, 42(1). https://​doi​.org/​10​.1177/​0263395720968060. Lawson, B. T. and Lovatt, M. (2020). Rhetoric of statistics during the NHS ‘winter crisis’. European Journal of Communication, 36(2). Lawson, B. T. (2022). Realizing the benefits of open government data: Journalists’ coverage of the NHS winter crisis, 2016–17. The Information Society, 38(1), 25–35. Stevens, H. (2019). Open data, closed government: Unpacking data.gov.sg. First Monday, 24(4). Woolley, S. and Howard, P. N. (2016). Political communication, computational propaganda, and autonomous agents. International Journal of Communication, 10(9), 4882–4890.

REFERENCES Adams, R. (2020). Nearly 40% of A-level result predictions to be downgraded in England, The Guardian Online. https://​www​.theguardian​.com/​education/​2020/​aug/​07/​a​-level​-result​ -predictions​-to​-be​-downgraded​-england. AFP (2020). REPLAY – 27 March: Downing Street daily coronavirus briefing | AFP, YouTube. https://​www​.youtube​.com/​watch​?v​=​EkH​-dAIUcvk. Anderson, C. W. (2021). Fake news is not a virus: On platforms and their effects. Communication Theory, 31(1), 42–61. Aviles, L. A. (2016). The art of public policy statisticians: The case of the Puerto Rico pension reform report. Caribbean Studies, 44(1–2), 47–68. Bail, C. A. et al. (2020). Assessing the Russian Internet Research Agency’s impact on the political attitudes and behaviors of American Twitter users in late 2017. Proceedings of the National Academy of Sciences of the United States of America, 117(1), 243–250. Bannister, F. and Connolly, R. (2014). ICT, public values and transformative government: A framework and programme for research. Government Information Quarterly, 31, 119–128. Bertot, J., Jaeger, P. T., and Grimes, J. M. (2010). Using ICTs to create a culture of transparency: E-government and social media as openness and anti-corruption tools for societies. Government Information Quarterly, 27(3), 264–271. Billig, M. (2021). Rhetorical uses of precise numbers and semi-magical round numbers in political discourse about COVID-19: Examples from the government of the United Kingdom. Discourse & Society, 32(5). https://​doi​.org/​10​.1177/​09579265211013115. Browne, S. (2015). Dark Matters: On the Surveillance of Blackness. Cambridge, MA: MIT Press. Bucher, T. (2017). The algorithmic imaginary: Exploring the ordinary affects of Facebook algorithms. Information, Communication & Society, 20(1), 30–44.

Political communication about data  399

Bucher, T. (2018). If … Then: Algorithmic Power and Politics. New York: Oxford University Press. Corner, J. (2021). Figuring Covid: A note on media flows and ‘statisticality’. Media Theory, 5(1), 147–158. Daventry, M. (2020). Britain scraps algorithm for student exam grades after outcry over fairness. EuroNews. https://​www​.euronews​.com/​2020/​08/​17/​britain​-scraps​-algorithm​-for​ -student​-exam​-grades​-after​-outcry​-over​-fairness. De Blasio, E. and Selva, D. (2016). Why choose open government? Motivations for the adoption of open government policies in four European countries. Policy & Internet, 8(3), 225–247. Dillet, B. (2020). Speaking to algorithms? Rhetorical political analysis as technological analysis. Politics, 42(1). https://​doi​.org/​10​.1177/​0263395720968060. DiResta, R. et al. (2019). The tactics & tropes of the Internet Research Agency. https://​int​.nyt​ .com/​data/​documenthelper/​533​-read​-report​-internet​-research​-agency/​787lea6d5bafbf19/​ optimized/​full​.pdf​#page​=​1. Dunleavy, P. (2016). Big data and policy learning. In G. Stoker and M. Evans (eds.), Evidence-Based Policy Making in the Social Sciences: Methods That Matter. Bristol: Policy Press, pp. 143–168. Dunleavy, P., Margetts, H., Bastow, S., and Tinkler, J. (2006). New public management is dead—long live digital-era governance. Journal of Public Administration Research and Theory, 16(3), 467–494. Fletcher, R. and Nielsen, R. K. (2019). Generalised scepticism: How people navigate news on social media. Information, Communication & Society, 22(12), 1751–1769. Golovchenko, Y., Buntain, C., Eady, G., Brown, M. A., and Tucker, J. A. (2020). Cross-platform state propaganda: Russian trolls on Twitter and YouTube during the 2016 U.S. presidential election. International Journal of Press/Politics, 25(3), 357–389. Howard, P. N., Woolley, S., and Calo, R. (2018). Algorithms, bots, and political communication in the US 2016 election: The challenge of automated political communication for election law and administration. Journal of Information Technology and Politics, 15(2), 81–93. Jadhay, C. (2020). Making grades as fair as they can be: advice for schools and colleges. UK Government. https://​ofqual​.blog​.gov​.uk/​2020/​05/​15/​making​-grades​-as​-fair​-as​-they​-can​-be​ -advice​-for​-schools​-and​-colleges/​. John, I. D. (1992). Statistics as rhetoric in psychology. Australian Psychologist, 27(3), 144–149. Katchergin, O. (2015). How many learning disabled are there? The rhetorical power of statistics in the Israeli discourse on learning disabilities. Open Journal of Social Sciences, 3(9), 155–166. Kennedy, H., Hill, R. L., and Aiello, G. (2016). The work that visualisation conventions do. Information, Communication & Society, 19(6), 715–735. Kilyeni, A. (2013). The rhetoric of numbers in print advertisements for cosmetics. Scientific Bulletin of the Politehnica University of Timisoara. Transactions on Modern Languages / Buletinul Stiintific al Universitatii Politehnica din Timisoara. Seria Limbi Moderne, 12(1/2), 17–26. Koetsenruijter, W. (2018). Numbers in the news: More ethos than logos? In A. Nguyen (ed.), News, Numbers and Public Opinion in a Data-Driven World. London: Bloomsbury Publishing, pp. 177–190. Lawson, B. T. (2022). Realizing the benefits of open government data: Journalists’ coverage of the NHS winter crisis, 2016–17. The Information Society, 38(1), 25–35. Lawson, B. T. and Lovatt, M. (2020). Rhetoric of statistics during the NHS ‘winter crisis’. European Journal of Communication, 36(2). Lu, G. et al. (2021). COVID-19 in Germany and China: Mitigation versus elimination strategy. Global Health Action, 14(1). https://​doi​.org/​10​.1080/​16549716​.2021​.1875601.

400  Handbook of digital politics

McCloskey, D. (1987). Rhetoric within the citadel: Statistics. Argument and Critical Practice: Proceedings of the Fifth SCA/AFA Conference on Argumentation. Annandale, VA: Speech Communication Association, pp. 485–490. Newton, K. (2006). May the weak force be with you: The power of the mass media in modern politics. European Journal of Political Research, 45(2), 209–234. Ofqual (2020a). GCSEs, AS and A levels: A guide for students in England. https://​www​.gov​ .uk/​government/​publications/​gcses​-as​-and​-a​-levels​-a​-guide​-for​-students​-in​-england​#how​ -your​-results​-were​-calculated. Ofqual (2020b). Guidance for Heads of Centre, Heads of Department and teachers on objectivity in grading and ranking. https://​assets​.publishing​.service​.gov​.uk/​government/​uploads/​ system/​uploads/​attachment​_data/​file/​886921/​Guidance​_on​_objectivity​_in​_grading​_and​ _ranking​_21MAY2020​.pdf. Ofqual (2020c). How GCSEs, AS & A levels will be awarded in summer 2020. https://​www​ .gov​.uk/​government/​news/​how​-gcses​-as​-a​-levels​-will​-be​-awarded​-in​-summer​-2020. Ofqual (2020d). Summer 2020 grades for GCSE, AS and A level, Extended Project Qualification and Advanced Extension Award in Maths. https://​assets​.publishing​.service​ .gov​.uk/​government/​uploads/​system/​uploads/​attachment​_data/​file/​908368/​Summer​_2020​ _grades​_for​_GCSE​_AS​_and​_A​_level​_110820​.pdf. Porter, T. (1995). Trust in Numbers: The Pursuit of Objectivity in Science and Public Life. Princeton, NJ: Princeton University Press. Redden, J. (2018). Democratic governance in an age of datafication: Lessons from mapping government discourses and practices. Big Data & Society, 5(2), 1–13. Roeh, I. and Feldman, S. (1984). The rhetoric of numbers in front-page journalism: How numbers contribute to the melodramatic in the popular press. Text. Interdisciplinary Journal for the Study of Discourse, 4(4), 347–368. Sky News (2020). Number of UK COVID-19 deaths could reach 10,000 within a week. Sky News. https://​www​.youtube​.com/​watch​?v​=​d8iQCdFavzw. Stevens, H. (2019). Open data, closed government: Unpacking data.gov.sg. First Monday, 24(4). Tennison, J. (2021). Why isn’t the government publishing more data about coronavirus deaths? The Guardian Online. https://​www​.theguardian​.com/​commentisfree/​2020/​apr/​02/​ government​-publish​-data​-coronavirus​-deaths. UK government (2020). COVID-19 Press Conference Slides - 03_04_2020. https://​assets​ .publishing​.service​.gov​.uk/​government/​uploads/​system/​uploads/​attachment​_data/​file/​ 878046/​COVID​-19​_Press​_Conference​_Slides​_​-​_03​_04​_2020​.pdf. UK government (2021). Coronavirus data dashboard. https://​coronavirus​.data​.gov​.uk. Van Dijk, T. (2000). Racism: A discourse analytical approach. In S. Cottle (ed.), Ethnic Minorities & The Media: Changing Cultural Boundaries. Philadelphia: Open University Press/McGraw-Hill Education.

26. Regulation of election communication Damian Tambini

INTRODUCTION: REGULATION OF POLITICAL CAMPAIGNS AND THE LESSONS OF TWENTIETH-CENTURY DEMOCRATIC FAILURE Since 2016, social media companies have been criticized for undermining the fairness and legitimacy of election campaigns. The Cambridge Analytica controversy exposed many of the challenges of data driven campaigning (see Box 26.1). According to many commentators the rapid shift of public attention onto social media enabled campaigners and their (often obscure) funders to deploy new, effective forms of targeted propaganda whilst evading the established rules and standards of political campaigns. Campaign and media regulation seek to ensure that elections are clean, fair and transparent whilst maintaining free speech and preventing information control, propaganda, and diminishing the impact of political lying. In response to a previous crisis of democracy, in Europe in the mid-twentieth century1 democratic states established rules and standards for the financing and conduct of political campaigns, as well as media regulation2 to protect values such as accuracy and fairness of reporting, and prevention of manipulation of public opinion through media pluralism.3 The rules-based order of international human rights has institutionalized in liberal democracies these norms for speech and campaign regulation. At a time of rapid media innovation and change in campaign methods, researchers and policymakers are asking once again if these rules are sufficient to protect democracy. Regulation of political campaign communication includes both the general rules that are in place all the time, and some specific rules that apply only during electoral campaigns. These vary from place to place but international standards and election monitoring are leading to some international convergence. Taking the example of the UK, the following provide the main framework: 1. Rights of freedom of expression. Under the (1950) Universal Declaration of Human Rights Article 19 and the (1966) International Covenant on Civil and Political Rights, and regional human rights instruments such as the (1951) European Convention on Human Rights, which is implemented in domestic legislation, everybody is free to impart and receive ideas. Freedom of expression is, however, subject to legitimate restriction if restrictions are necessary, proportionate, and prescribed by law for a legitimate purpose in a democracy. The domestic case law of countries that adhere to such Human Rights Charters, and the standards of the UN Human Rights Council (such as General Comment 34) 401

402  Handbook of digital politics

and the standards of the Council of Europe set out what are legitimate restrictions, and when these are necessary in a democratic society and thus permissible. These legal standards are referred to by various actors that play a role in regulating elections and maintaining trust in them, such as the national Electoral Commissions (Election Management Bodies) and international election monitors such as the Council of Europe, the OSCE and the EC, and on rare occasions, the UN. The United Kingdom broadcast bans on political advertising and on the holding of TV licenses by political parties, religious groups and the state have been upheld by the ECtHR.4 2. Self-regulation and ethical journalism. Those rights of freedom of expression are enjoyed with the knowledge that communication, particularly by powerful media, can have implications for rights other than freedom of expression, and the rights of others. In short, in the context of elections in particular, media power comes with responsibilities. Given the necessity of ethical restraint, democracies have evolved norms of self-regulation of ‘socially responsible’ journalism5 to ensure that journalism respects ethics and fairness, and that the ‘truth-seeking’ nature of journalism is institutionalized and incentivized, even though it may be expensive. Such rights are respected whilst maintaining autonomy of media, particularly from state interference. Ethical journalism is in turn subject to a range of legal and institutional privileges and supports, from tax breaks to liability protections and exemptions which are based on an idea of the social value of ethical journalism which is based on truth-seeking.6 3. Broadcast licensing obligations. By the mid-twentieth century, the potential for private interests to influence public and democratic debate through ownership and control of powerful radio broadcasters was clear. Broadcasting has therefore been subject to licensing obligations not only to maintain a sufficient plurality of providers, but to ensure ‘internal pluralism’ or impartiality or ‘fairness’7 in order to limit the potential for opinion control to undermine democratic legitimacy. In addition, many countries including the UK operate bans on political advertising (for specific parties or campaigns) and on parties or candidates holding broadcast licences. In 2021–2022 these rules were used to ban broadcasts by Russian and Chinese state media in the EU and the UK.8 4. Campaign rules. In addition to these rules that always apply, communication which falls under the definition of campaign communication or political advertising has been subject to regulations that apply only during the campaign period.9 These include limits on donations and advertising spending, labelling and transparency requirements for political advertising, and disqualifications from donations such as bans on foreign donations. Many countries including the UK operate ‘blackout periods’ on political reporting, or broadcast of specific information such as polls on the day of the election or preceding days. Data protection rules have also been increasingly relevant as campaigns gain advantage by using the electoral roll and exploiting other databases for campaign purposes. Election rules are broadly aimed at maintaining the legitimacy of elections by ensuring fairness by limiting the role of money and ensuring a level playing field

Regulation of election communication  403

in the context of changing tactics and tools of campaigning. In most countries the rules on management of elections, including regulation of campaigns and their funding, are monitored and enforced by an independent electoral commission or election management body (EMB). Citizens also have recourse to the courts if they claim election law has been breached. The content of political advertising has not in general been regulated and has been exempted for example from the Advertising Standards Authority code requirements of accuracy.10

BOX 26.1 CAMBRIDGE ANALYTICA AND CAMPAIGN REGULATION The services offered by political marketing consultancy Cambridge Analytica (CA) included database building, management and design, strategy design for election and political marketing campaigns, psychometric profiling, database segmentation, message production, targeting, design and evaluation. The company had a rapid rise during the 2010s, focusing above all on the use of social media as a marketing tool. The methods of CA caused widespread controversy when they came to light. Between 2016 and 2019 a series of investigations revealed that Cambridge Analytica was offering services that breached either the spirit or the letter of election campaign rules in the country they were provided in, or represented new practices that whilst technically legal, undermine the legitimacy of elections. Some, but not all complaints and legal charges were upheld against CA. In 2016 CA worked for the Donald Trump election campaign and according to its critics, for Leave.EU which was a leading campaign for Brexit. In both cases there was a controversy about precisely what was done by CA. Allegations made against Cambridge Analytica included that it had breached the data consent regime by reusing data for which no specific consent had been given; facilitated foreign funding that would otherwise have been illegal; distributed disinformation; used big data for spooky psychometric profiling based on thousands of data points; colluded with hostile foreign actors and facilitated the breach of election spending rules. Each of these allegations was controversial and has been the subject of multiple further investigations by public authorities. Even when complaints under existing rules were not upheld, the Cambridge Analytica controversy brought to public attention new campaign practices which had serious consequences for election legitimacy and raised questions about efficacy of current regulations (Wylie, 2019).

404  Handbook of digital politics

THE NEW PROPAGANDA In the past decade new challenges and controversies have arisen which expose the limitations of campaign regulation. Media Change The second half of the twentieth century was the era of electronic mass broadcast media.11 At the turn of the century 84 per cent of viewers reported that their main source of national and international news was television. Newspapers (52 per cent) and radio (33 per cent) were the next most used sources, while very small numbers mentioned teletext (8 per cent), Internet (3 per cent) and magazines (1 per cent) (Cumberbatch et al., 2000, p. 53). Since then, the Internet has rapidly become the principal source of news for many people. By 2018, more than 64 per cent described the Internet as their main source of news and information. For some demographics, such as the 16–24 age group, the Internet was significantly more used as a source of news than television, and this trend has continued.12 In the context of elections, it is important to note that broadcasting – which is generally regulated for accuracy and balance – has been gradually replaced by more partisan online news for significant numbers of citizens. But it is important to keep these changes in perspective. People continue to use broadcasting sources and the survey data is superficial. Depending on the survey question wording those reporting Internet as a source of news may have been using the Internet to access news covered by the broadcasting code, or the BBC Charter. The total audience for online news was greater than that for TV news before the end of the second decade of the millennium, but online news had supplemented rather than replaced broadcasting. According to the Ofcom News Consumption Report for 2018, 73 per cent of respondents mentioned the Internet as a source of news (Figure 26.1). Many of these also listed television as a source, which was mentioned by 79 per cent of respondents (Ofcom, 2018).13 This shift in news consumption has been replicated across the EU. Online overtook television as the most accessed news source by around 2018 in many European countries, though television which tended to be regulated for balance, remained a dominant source of news. Election campaigns are thus being fought in a dramatically altered news environment. These changing patterns of media consumption also provide changed platforms of access of citizens to party campaign messages, and a paradigm shift in political advertising practices. The spending returns of Election Management Bodies such as the UK Electoral Commission confirm that spending has rapidly migrated from direct mail, broadcasting, press and display advertising to targeted and behavioural social media advertising (Tambini, 2021).

Regulation of election communication  405

Notes: Question: C1. Which of the following platforms do you use for news nowadays? Base: All adults 16+ 2018 – Total = 4618. *Internet figures include use of social media and all other Internet sources accessed via any device. Source: Ofcom News Consumption Survey 2018.

Figure 26.1

A tipping point? The rise of the Internet as a source of news

CONTROVERSIES OF DATA DRIVEN CAMPAIGNS Information about voters, in the form of canvassing returns, basic geographic and demographic data, and information that can be gleaned from the electoral roll has been increasing in importance to party campaigns for many years. But with the rapid proliferation of social media and its accompanying ‘surveillance capitalism’ (Zuboff, 2019) business models, the costs of gathering, storing and processing campaign data rapidly decline as do the number of data points available on each voter. A number of private companies offer marketing tools for the fine tuning and targeting of election messages, such as the surely ironically named ‘Nation Builder’. Campaign consultancies such as Cambridge Analytica advise on how to use the services available. But it is the large social media platforms and in Europe and North America, principally Facebook, that offer the most cost-effective and easy to use message targeting services. A tipping point was reached in the mid-2010s as social media approached ubiquity, and campaigns shifted advertising spend from more traditional means such as direct mail, display and (where it is legal) broadcast advertising, onto social media and mainly Facebook. Scholars such as Daniel Kreiss (2012), Phil Howard (2020), Jacob Rowbottom (2010) and Solon Barocas (2012) were early to spot the dangers inherent in the new forms of data-driven campaigning. Problems deemed inherent to these forms of campaigning include divisiveness, disenfranchisement, discrimination, and chilling effects (Barocas, 2012, p. 33). Many of these have become issues for much wider public debate since these initial observations. Since the mid-2010s

406  Handbook of digital politics

scholarly work has been focused on the following issues in relation to data driven campaigns: Fragmentation Cass Sunstein was the first analyst to bring Internet driven social fragmentation to the attention of the public as a key problem for democratic legitimacy. Alongside related problems such as the decline of mass political parties and the realignment of political cleavages in post-industrial, globalizing economies, the shift to social media has been accompanied by claims of algorithmic driven social fragmentation. Researchers, however, have been divided on the extent to which there remains an effective ‘public square’ of shared truths and a common agenda for debate and will-formation. On one hand analysts such as Sunstein (2018) and Pariser (2011) claim that democratic communication has become too fragmented and on the other experts such as Ross Arguedas et al. (2022)14 respond that this process has been exaggerated. Automation, AI and Human Autonomy Data driven campaigns thus raise direct challenges for existing campaign finance regulation, and data consent rules when databases are reused and those targeted by political advertising have not consented to the use of their personal data for such uses. They also raise some more abstract but potentially more fundamental issues that may in the long term pose a greater challenge to election legitimacy. Whilst the previous paradigm for protecting media plurality and the free circulation of opinion was based on ensuring that in the market as a whole there was a ‘sufficient plurality’ of sources of opinion available,15 data driven targeting campaigns, based on increasing numbers of data points on individual voters, have the potential to create ‘propaganda bubbles’ around individual voters to such an extent that they undermine the autonomy of those voters: their ability to make free choices. The selling point of new campaigning technologies such as those infamously used by Cambridge Analytica is that they are more effective in changing the minds of voters because they were based on reliable knowledge of voter’s preferences and psychological tendencies. Whilst there is evidence that the effectiveness of Cambridge Analytica products for example, was exaggerated in the sales pitch,16 the case has highlighted the potential for new data driven machine learning to undermine the ability of individual voters to think for themselves. Facts and Truth Because social media ads are produced in an ever-greater variety and tested for ‘engagement’ and ‘virality’ rather than for their ideological or truth value, and because targeted campaigns allow for ‘dark ads’ that disappear after viewing and thus cannot be rebutted, it has been argued that these new forms of communication undermine the facticity or veracity of political campaigns (Barocas, 2012). At the

Regulation of election communication  407

same time, mainstream journalism is arguably losing its position as a societal arbiter of truth. John Stuart Mill and John Milton believed that free speech would lead to the emergence of truth. Subsequent philosophers such as Onora O’Neill (2013) have claimed that this is not the case: disciplines dedicated to the emergence of truth such as science often have strong norms of speech control. When it comes to news media, media law and policy have in the past strongly tipped the balance in favour of journalism practices that favour the emergence of truth. This is not the case online. Social media and other intermediaries are based on business models and advertising processes that favour engagement more than truth or the public interest. Fairness and the Role of Money Following the elections and referenda of 2016, allegations were made that existing election regulations had been breached and also that the existing paradigm of election regulation was no longer sufficient.17 Taking the example of the United Kingdom, the Electoral Commission issued a number of fines and warnings in this period relating to use of data and breaching both of finance and consent rules. As well as those breaches, there were aspects that were more difficult to regulate such as complex and derivative uses of data for targeting purposes. Many new practices, such as use of dark ads and situating campaign spending outside the regulated campaign period, were simply not caught by the rules.18 Digital Dominance The latter two have been subject to a new concern with the power of the platforms themselves. Like Standard Oil at the beginning of the twentieth century, the ability of new consolidated corporate powers to extract value at key strategic gateways of networked economies has recently led to concerns that the economic, lobbying power of platforms, particularly when combined with opaque power over opinion formation, was becoming a threat to democracy.19 This presented itself in different ways in the US, where many of these corporations are based, and in Europe and other countries where these corporations are perceived as remote and unaccountable. This concern with platform power led to a rethinking of the role of antitrust and competition law in constraining these corporate giants. An important new approach to antitrust20 which posited a much wider concern about the implications of corporate power for social goods such as trust in democracy became a new orthodoxy in US competition enforcement, embodied in the appointment of Lina Khan as chair of the US Federal Trade Commission. Information Warfare Researchers have presented a strong case that election manipulation via social media is part of a longer-term pattern of ‘active measures’ (meddling) between sovereign states that have been a feature of international relations for some time.21 As Philip

408  Handbook of digital politics

Howard’s work makes clear, whilst it is difficult to measure effects, there is no doubt that there is a burgeoning industry in ‘neuromarketing’ (Howard, 2020, p. 65) and such techniques are used by foreign actors. Whilst the fact of meddling in political information systems in other states is nothing new, and certainly not peculiar to the East, current ‘information operations’ are actively being scaled by military operations in the West, in response to perceived aggression, particularly from China, Russia, North Korea and Saudi Arabia. The Russian doctrine of ‘hybrid warfare’ and the Chinese doctrine of the ‘three warfares’ blur the boundaries of warfare by proposing an approach to warfare which involves a common strategy that includes kinetic, cyber, information, economic, diplomatic and legal warfare.22

ONGOING DEBATES Each of these concerns has been subject to vigorous debate and needs more research. Since Cas Sunstein coined the term ‘the Daily Me’23 to express what he claimed was a trend towards fragmentation and personalization of media diets, and Eli Pariser published the book The Filter Bubble (2011), there has been a concern that this leads to political polarization. The response presented the counter claim that there remains a great deal of common ground in policy and political debates and that claims about social media driven fragmentation have been exaggerated. Oxford researchers reviewed the literature and found no support for the filter bubble hypothesis and found weak evidence for ‘echo chambers’ in social media use patterns.24 The controversy was compounded by the fact that much of the research was focused on one platform (Twitter) and neglected cross-platform effects. Whilst an analysis of Twitter might show a trend towards self-selecting filter bubbles, users of Twitter also use other sources (see Dahlgren, 2019; Ross Arguedas et al., 2022). The new campaign practices involving targeted political advertising appeared to turbo charge fragmentation in a number of ways: firstly the declining cost of data and voter profiling meant that a larger number of political messages could be crafted to fit a wider range of demographic profile segments. The practice of ‘AB testing’ enabled real time testing of the engagement of voters with a political message in a social media environment. Put simply, ads were tested on a particular demographic for engagement, with the weaker performing advertisement withdrawn. This has a number of potential consequences, one of them being that specific tailoring of messages for particular audiences becomes much cheaper and quicker to deploy with the result, in theory, of a more fragmented public sphere. The key issue in a democracy of whether the information environment permits and encourages sufficient autonomy of rational citizens, and therefore legitimates democracy based on the exercise of citizen rationality is an ongoing theoretical, philosophical and empirical controversy. On the one hand many critical theorists, particularly those who base their work on a relativist epistemology for which truth is a consequence of the exercise of power and domination tend to undermine the critical defence of democracy. The theoretical followers of Habermas for example line up in

Regulation of election communication  409

order to support the project of democratic legitimacy and followers of Foucault on the other hand stress the need to undermine it and reveal its weaknesses. But a new grouping of critically engaged democracy scholars began to develop a new theoretical defence of democracy (Landemore, 2020). The question of whether media system change undermines the systemic incentives for truth-seeking media has been focused on the role and function of journalism as a social and legal incentive structure that has evolved to support truth seeking in media. Whilst much of the research on journalism itself focuses on journalism as an activity and the question of innovation and business models (including new private deals with platforms), a wider approach stresses the role of a wider range of public policy incentives (including tax, liability, subsidies and media co-regulation).25 A literature review for the Council of Europe in 202026 showed how the shift from platforms subject to ethical self-regulation to non-regulated or lightly regulated platforms has been marked over recent years. The potential as a result, is of a potential for ‘robo politics’ based more on engagement optimization than rational critical deliberation.27 Campaign finance and the controversial role of advertising has been difficult to research due to the lack of available data, particularly on the content of ‘dark ads’.28 It is clear, however, that in the past decade a decisive and likely permanent shift has occurred in the political economy of campaigning, symbolized by the moment during the 2016 Brexit referendum campaign in which the Leave campaign, which was eventually to win, shifted 100 per cent of its campaign spend onto Facebook,29 reflecting the widely shared view that the targeting functions of social media advertising offer huge benefits when compared to those platforms formerly used, such as display advertising and press. Profiling and targeting is particularly valuable in political advertising, because tiny slivers of the electorate in demographic and geographical terms can be of huge strategic value. New problems in political campaigning include not only new possibilities to evade campaign finance and donation rules through the use of in-kind donations such as databases, but the wider issue of transparency whereby advertisements30 targeted at a narrow niche audience are visible only to them and disappear after distribution. This led to wide and ultimately effective calls for repositories, libraries and transparency on political advertising.31 Finally, the question of foreign interference has been seen as exaggerated by some commentators. Following a wide range of accusations that political advertising, fake news and social media propaganda have been weaponized since the 2016 Brexit referendum and Trump elections, very few of the complaints have achieved the standard of proof requiring criminal charges. The Mueller investigation32 cleared Trump himself of collusion with Russian attempts to interfere with the US election, and allegations of direct Russian funding of propaganda campaigns in the Brexit referendum have yet to be upheld in a UK court (though a court has ruled against the main funder, Aaron Banks’ claim that to suggest foreign involvement was defamatory). There is widespread support for the notion that foreign powers used social media such as Facebook and Twitter to interfere, and sow divisions on wedge issues, but of course little proof that they were successful because the evidence basis and

410  Handbook of digital politics

research techniques to prove such a notion do not exist. Whilst the official stance of Western democracies is that the threat is real, such claims are inherently hard to prove. But as Nigel Inkster put it: ‘How effective that campaign was is neither here nor there. What matters is that it has had, and continues to have, a significant destabilizing effect on US politics and social cohesion, achieved at very little cost.’33 Fundamentally whilst there is no doubt that hostile actors are attempting to interfere in elections34 researchers agree that it is difficult, if not impossible to prove causation between exposure to propaganda and voting intention, which has not stopped people triumphantly declaring that there is no proof that foreign interference had any effect on democratic elections since 2015.

THE RESPONSE: POLICY AND RESEARCH The policy response to these seismic changes in political campaigning has been fragmented and self-contradictory, but there has been a clear political will to take decisive action to protect democracy from these new threats35 through new policy. In 2019 the EU published its Democracy Action Plan, which announced an intention to develop new policy recommendations in three areas: promoting free and fair elections, strengthening media freedom, and countering disinformation. Transparency of political advertising was advanced as a means to protect the legitimacy of election information. In 2021 the EU published a proposal for strict new rules on political advertising – including a ban on political targeting.36 In the UK, where political advertising had arguably played a role in Brexit, the political will for change was more muted. Following the controversial referendum, the independent elections supervisor the Electoral Commission published a report in 201837 which detailed the rapid changes in campaign methods and outlined the need for new ways of reporting spending that reveal more information on donors, whilst also tightening rules on foreign finance of campaigns. When the 2019 election was contested without any major reforms, the EC reported that misleading campaigns ‘undermined trust’.38 As it became clear that campaign spending was rapidly shifting onto sophisticated data-driven targeted advertising on unregulated platforms, the Information Commissioners Office (ICO), which regulates data protection and freedom of information also published several reports alongside its adjudications, calling for tighter regulation of the underlying data, inferred data, and consent. The ICO called in 2018 for Parliament to establish a new statutory code of conduct setting out the obligations of those involved in electoral campaigning.39 In 2020 the Advertising Standards Authority which is a self-regulatory body for the ad industry published an article arguing that political advertising, which had previously been exempt from the wider advertising codes, should be regulated.40 These various authorities agreed that urgent attention was required to make the process more transparent, for example by making clear through data who was funding which ad, who was being targeted by whom, and the privacy implications of using these services. In recognition of the importance of these matters for wider public

Regulation of election communication  411

involvement in democracy the ICO recommended in 2017 that the ICO and the Electoral Commission should ‘conduct an ethical debate in the form of a citizen jury to understand further the impact of new and developing technologies and the use of data analytics in political campaigns’.41 When the UK government introduced legislation to Parliament in 2021–2022, however, it did not address the recommendations of these public bodies. Instead it proposed to introduce voter ID in polling stations. The legislation introduced imprints for political campaign material (which the Electoral Commission had first called for more than a decade previously). It set out some new categories for financial reporting but made no provision for a statutory code on political advertising or the other reforms that had been called for. This highly controversial legislation introduced ministerial roles and codes (a ‘strategy and policy statement’) that compromise the independence of the Electoral Commission. Rather than strengthening the ‘in kind’ rules that govern donation of services to campaigns whose spend is limited, it weakened them.42 Other legislative reforms that may impact campaign regulation include the proposed Broadcasting Act and the Online Safety Bill. It is likely that the UK will be an outlier with stronger regulation in place across the EU. Recent developments in political campaigning have presented some fundamental questions for the Habermasian project of legitimating complex advanced capitalist societies through public deliberation.43 There have been long-acknowledged difficulties with the relationship between mass media and opinion formation. Governance arrangements, including those discussed here, have aimed to prevent the capture of media by state or private interests, and we can see that the shift to digital has triggered a rethink of those rules and policies in order to take account of the changes in campaign techniques described in this chapter and protect democracy and fundamental rights from various forms of hostile actor and private interest capture. So researchers have to contend with a rapidly changing context in which technologies, practices and their socio-legal context are in flux. In order to understand these long-term developments, we should have in mind the tipping points and crisis points that might emerge. In the years after 2015 there has been something of a crisis of legitimacy of liberal democracy itself, reflected in widespread polling evidence expressing rising dissatisfaction in democracy as a political system. The policy responses outlined here seek to make a systemic response to these challenges. Future research is needed above all to track and analyse the changing campaigning ‘play book’ of techniques of data driven campaigning. Researchers have described changing campaign techniques, notably Andrew Chadwick (2017) Phil Howard (2020), and Daniel Kreiss (2012). One important part of the regulatory response in multiple countries has been to establish a legislative framework for obligations of data access for researchers. A key challenge therefore will be to establish data access in a manner and format that enables genuinely independent academic research whilst protecting privacy rights. The issue of fragmentation will continue to be of interest to researchers. Whilst large majorities in those countries served by a free media are likely to continue to access both mainstream and niche media it will be important to analyse in more

412  Handbook of digital politics

detail using cross-platform data how this varies across different groups and whether particular age or other minorities are disengaging from the mainstream. The perils of basing claims about fragmentation on single platform studies are clear and future research based on audience studies, reported media use, as well as the standard audience surveys will be important. The work of academics such as Natali Helberger (2018) on personalized media and its implications for ‘opinion power’ show the importance also of theoretical and normative work on the conceptual and legal framework. As automation continues to transform the news and information process and political campaigns, wider debates about AI ethics and human autonomy will become increasingly relevant. Researchers will need to develop critical and methodologically innovative approaches not only to the practices of campaigners but their implications for the lived experience of citizens and practices of voting, including the audience for political advertising, their active engagement with messages and how this relates to audience segmentation, political geography and electoral strategy. The controversy about post-truth and fakery raises the question about whether deliberative campaigns can be conducted in a realm of agreed facts and whether the alleged truth-seeking function of mediating institutions justifies their privileges. The theoretical cleavage between more relativist epistemologies more sympathetic to Foucault and those adopting a Habermasian approach will continue to divide researchers. In the context of election campaigns themselves there is a continuing need for research on the role of money in elections, the role of new campaign techniques, the data economy and the implications of this for fairness and interference in elections. Other actors engaged in these practices, such as electoral supervisors, election observers and data protection monitors tend to have very limited liberty to conduct original research and they are subject to continuing challenges to their independence. In terms of ‘digital dominance’, researchers have claimed that multi-sided platform markets are particularly prone to lock in, network effects and various economic effects that reinforce tendencies to market concentration and reinforce also the political power of platforms (Barwise and Watkins, 2018). The deployment of new competition laws such as the Digital Markets Act may go some way towards redressing these tendencies through for example promoting interoperability and data access, but given the history of protean strategies by platforms, researchers should focus on how market dominance and regulatory capture evolve in this new era of regulated oligopoly. In an era of global realignment and protracted conflict, the field of ‘information warfare’ will create opportunities both for applied researchers developing new tools for the analysis of ‘computational propaganda’ such bot and troll activity and critical approaches to the role of the state in controlling flows of information. It is hoped that all of these research fields will be conducted with an eye to the wider legal and regulatory context and normative reflection on the liberal democratic project.

Regulation of election communication  413

NOTES 1. 2.

See Humphreys (1994); (Pickard 2015). See in particular the report of the 1947 Royal Commission on the Press, and in the US the Hutchins Commission. In Germany the ‘basic law’ established a commitment to media pluralism, freedom of communication and the prevention of concentrations of ‘opinion power’. See Humphreys (1996); Koltay (2022). 3. See Craufurd Smith and Tambini (2012). 4. See https://​www​.bbc​.co​.uk/​news/​uk​-politics​-22238582. 5. For a discussion of the emergence of the notion of socially responsible journalism in the US see Pickard (2015). 6. See Tambini (2021). 7. All licensed broadcasters are required to be impartial in the UK. Most have some sort of internal plurality requirements in the EU. Until the 1990s, US broadcasters observed the ‘fairness doctrine’ in their news coverage. 8. E. Brogi (2022), ‘Banning Russia’s news outlets in the EU and UK: Is this the right move?’ https://​www​.eui​.eu/​news​-hub​?id​=​banning​-russias​-news​-outlets​-in​-the​-eu​-and​-uk​ -is​-this​-the​-right​-move; Ofcom (2021): Ofcom revoked CGTN’s licence to broadcast in the UK. https://​www​.ofcom​.org​.uk/​about​-ofcom/​latest/​media/​media​-releases/​2021/​ ofcom​-revokes​-cgtn​-licence​-to​-broadcast​-in​-uk. 9. These are established in the Political Parties, Elections and Referenda Act 2000 and the Elections Act 2022 and set out in guidance and codes of the Electoral Commission. See https://​www​.electoralcommission​.org​.uk/​i​-am​-a/​political​-party/​election​-campaign​ -spending. 10. ASA Code, Section 7. https://​www​.asa​.org​.uk/​advice​-online/​political​-advertising​.html. 11. For a longer discussion see Tambini (2021, p. 220). 12. I have explored these somewhat non-controversial trends further in my report for the Council of Europe: Media Freedom, Regulation and Trust (2020). https://​edoc​.coe​.int/​ en/​media/​8212​-media​-freedom​-regulation​-and​-trust​-a​-systemic​-approach​-to​-information​ -disorder​.html. 13. See https://​www​.ofcom​.org​.uk/​research​-and​-data/​tv​-radio​-and​-on​-demand/​news​-media/​ news​-consumption. 14. See https://​reutersinstitute​.politics​.ox​.ac​.uk/​echo​-chambers​-filter​-bubbles​-and​ -polarisation​-literature​-review. 15. See Craufurd Smith and Tambini (2012); Helberger (2018). 16. Chris Wylie (2019) describes an insider view of Cambridge Analytica. 17. See Tambini (2018). 18. See Electoral Commission (2019). Report: Digital campaigning – increasing transparency for voters. https://​www​.electoralcommission​.org​.uk/​who​-we​-are​-and​-what​-we​ -do/​changing​-electoral​-law/​transparent​-digital​-campaigning/​report​-digital​-campaigning​ -increasing​-transparency​-voters. 19. See Wu (2018); Moore and Tambini (2018). 20. For example see Khan (2018); Wu (2018). 21. See Rid (2020); Moore (2018); Howard (2020). 22. See Clack and Johnson (2021), especially Chapters 8 and 11. 23. See for example Sunstein (2018). 24. See Ross Arguedas et al. (2022). 25. See Koltay (2022). 26. See Tambini (2021). 27. Tambini (2016): https://​www​.theguardian​.com/​commentisfree/​2016/​nov/​18/​robopolitics​ -social​-media​-traditional​-media​-dead​-brexit​-trump. 28. See Tambini (2018).

414  Handbook of digital politics

29. See Tambini (2018) and Moore and Tambini (2018). See also Tambini (2016): https://​www​.theguardian​.com/​commentisfree/​2016/​nov/​18/​robopolitics​-social​-media​ -traditional​-media​-dead​-brexit​-trump. 30. See Wardle (2017). 31. See the Perugia Declaration on political advertising. https://​globa​lfreedomof​expression​ .columbia​. edu/​p ublications/​p olicy​- brief​- overview​- italian​- online​- offline​- political​ -communication​-regulation/​. 32. See https://​intelligence​.house​.gov/​social​-media​-content/​. 33. Inkster (2021, p. 283). 34. ‘Russians Again Targeting Americans With Disinformation, Facebook and Twitter Say’, New York Times, 1 September 2020. https://​www​.nytimes​.com/​2020/​09/​01/​technology/​ facebook​-russia​-disinformation​-election​.html. 35. See the recommendations of Venice Commission, Election Monitors, the Council of Europe. Domestic UK bodies such as the Information Commissioners Office and the Electoral Commission have made detailed recommendations for reform. In some countries such as France, specific and more direct legal changes were made rather more quickly. The French law on manipulation of elections was passed in 2019. Most of the early reforms were voluntary. 36. See https://​inforrm​.org/​2021/​11/​30/​new​-eu​-rules​-on​-political​-advertising​-here​-you​-go​ -read​-the​-fine​-print​-judit​-bayer/​. 37. Electoral Commission. Digital Campaigning – increasing transparency for voters. https://​ www​.electoralcommission​.org​.uk/​who​-we​-are​-and​-what​-we​-do/​changing​-electoral​-law/​ transparent​-digital​-campaigning/​report​-digital​-campaigning​-increasing​-transparency​ -voters. 38. See https://​www​.bbc​.co​.uk/​news/​uk​-politics​-52366099. 39. See Democracy Disrupted? Personal Information and Political Influence, ICO, 11 July 2018. 40. Advertising Standards Authority: ‘We think political advertising should be regulated’. 3 June 2020. https://​www​.asa​.org​.uk/​news/​we​-think​-political​-advertising​-should​-be​ -regulated​.html. 41. Information Commissioner’s Office (2017) Recommendation 6. 42. See the Government’s summary: https://​www​.gov​.uk/​government/​publications/​elections​ -bill​-2021​-summary​-factsheet/​elections​-bill​-2021​-summary​-factsheet. 43. For a discussion see Landemore (2020).

FURTHER READING Coe, P. (2021). Media Freedom in the Age of Citizen Journalism. Cheltenham, UK and Northampton, MA, USA: Edward Elgar Publishing. Howard, P. (2020). Lie Machines: How to Save Democracy from Troll Armies, Deceitful Robots, Junk News Operations, and Political Operatives. New Haven: Yale University Press. Koltay, A. (2015). The concept of media freedom today: New media, new editors and the traditional approach of the law. Journal of Media Law, 7(1), 36–64. Kreiss, D. (2012). “Yes we can (profile you)”: A brief primer on campaigns and political data. Stanford Law Review Online, 64(70). Tambini, D. (2021). Media Freedom. Cambridge: Polity Press.

Regulation of election communication  415

REFERENCES Barocas, S. (2012). The price of precision: Voter microtargeting and its potential harms to the democratic process. Proceedings of the First Edition Workshop on Politics, Elections and Data, pp. 31–36. New York. Barwise, P. and Watkins, L. (2018). The evolution of digital dominance. In M. Moore and D. Tambini (eds.), Digital Dominance: The Power of Google, Amazon, Facebook and Apple (pp. 21–49). New York: Oxford University Press. Chadwick, A. (2017). The Hybrid Media System. New York: Oxford University Press. Clack, T. and Johnson, R. (eds.) (2021). The World Information War: Western Resilience, Campaigning and Cognitive Effects. New York: Routledge. Craufurd Smith, R. and Tambini, D. (2012). Measuring media plurality in the United Kingdom: Policy choices and regulatory challenges. Journal of Media Law, 4(1), 35–63. Cumberbatch, G., Wood, G., and Littlejohns, V. (2000). Television: The Public’s View 2000. ITC Research Publication. Dahlgren, P. M. (2019). Selective exposure to public service news over thirty years: The role of ideological leaning, party support, and political interest. International Journal of Press/ Politics, 24(3), 293–314. Helberger, Natali (2018). Challenging diversity: Social media platforms and a new conception of media diversity. In M. Moore and D. Tambini (eds.), Digital Dominance: The Power of Google, Amazon, Facebook and Apple (pp. 152–175). New York: Oxford University Press. Howard, P. (2020). Lie Machines: How to Save Democracy from Troll Armies, Deceitful Robots, Junk News Operations, and Political Operatives. New Haven: Yale University Press. Humphreys, P. (1994). Media and Media Policy in Germany: The Press and Broadcasting since 1945. Oxford: Berg. Humphreys, P. (1996). Mass Media and Media Policy in Western Europe. Manchester: Manchester University Press. Inkster, N. (2021). War in the age of uncertainty. In T. Clack and R. Johnson (eds.), The World Information War: Western Resilience, Campaigning and Cognitive Effects. New York: Routledge. Khan, L. (2018). Amazon: An infrastructure service and its challenge to current antitrust law. In M. Moore and D. Tambini (eds.), Digital Dominance: The Power of Google, Amazon, Facebook and Apple (pp. 98–130). New York: Oxford University Press. Koltay, A. (2022). On the distinctiveness of press freedom: The case of the social responsibility and privileges of the press. Communications Law, 27(1), 9–23. Kreiss, D. (2012). “Yes we can (profile you)”: A brief primer on campaigns and political data. Stanford Law Review Online, 64(70). Landemore, H. (2020). Open Democracy. Princeton, NJ: Princeton University Press. Moore, M. (2018). Democracy Hacked. London: One World. Moore, M. and Tambini, D. (eds.) (2018). Digital Dominance: The Power of Google, Amazon, Facebook and Apple. New York: Oxford University Press. O’Neill, O. (2013). Media freedoms and media standards. In N. Couldry, M. Madianou, and A. Pinchevski (eds.), Ethics of Media (pp. 21–38). Basingstoke: Palgrave Macmillan. Ofcom (2018). News Consumption Survey. Pariser, E. (2011). The Filter Bubble: What the Internet Is Hiding from You. New York: Penguin. Pickard, V. (2015). America’s Battle for Media Democracy: The Triumph of Corporate Libertarianism and the Future of Media Reform. Cambridge: Cambridge University Press. Rid, T. (2020). Active Measures: The Secret History of Disinformation and Political Warfare. London: Profile Books.

416  Handbook of digital politics

Ross Arguedas, A., Robertson, C., Fletcher, R., and Nielsen, R. (2022). Echo Chambers, Filter Bubbles, and Polarisation: A Literature Review. Report. Reuters Institute for the Study of Journalism. https://​reutersinstitute​.politics​.ox​.ac​.uk/​echo​-chambers​-filter​-bubbles​-and​ -polarisation​-literature​-review. Rowbottom, J. (2010). Democracy Distorted: Wealth, Influence and Democratic Politics. Cambridge: Cambridge University Press. Sunstein, C. R. (2018). #Republic: Divided Democracy in the Age of Social Media. Princeton, NJ: Princeton University Press. Tambini, D. (2018). Social media power and election legitimacy. In M. Moore and D. Tambini (eds.), Digital Dominance: The Power of Google, Amazon, Facebook, and Apple (pp. 265–293). New York: Oxford University Press. Tambini, D. (2021). Media Freedom. Cambridge: Polity Press. Wardle, C. (2017). Information Disorder: Towards an Interdisciplinary Framework for Research and Policymaking. Council of Europe. https://​rm​.coe​.int/​information​-disorder​ -toward​-an​-interdisciplinary​-framework​-for​-researc/​168076277c. Wu, T. (2018). The Curse of Bigness. Columbia Global Reports. Wylie, C. (2019). Mindf*ck: Inside Cambridge Analytica’s Plot to Break the World. London: Profile Books. Zuboff, S. (2019). The Age of Surveillance Capitalism: The Fight for a Human Future at the New Frontier of Power. New York: PublicAffairs.

PART VI JOURNALISM, APPARENT JOURNALISM AND MEDIA INSTITUTIONS

27. Social media as resources for journalistic struggle in politically restrictive settings Banafsheh Ranji

INTRODUCTION The significance of the topic of social media and journalism in restrictive media settings lies in the general conditions for the media field in these settings, and the positioning of the field in relation to the economic and political fields. When it comes to the discussion of media and journalism in certain settings that tend to be referred to as ‘non-democratic’ and ‘non-Western’ contexts, the existing narratives tend to reduce the whole reality to the influence of political systems. This tendency is evident in the very terms that are used to refer to these contexts such as ‘non-democratic’ and ‘authoritarian’ (as opposed to ‘democratic’). To be clear about my positioning, I avoid (to the best of my ability in practising reflexivity) using a binary terminology as I believe these terms are generalizations and simplifications, and are therefore imprecise representations of diverse social realities. I rather adopt a continuum approach and use the terms that denote relative meanings, such as ‘less/more restrictive countries’, ‘less/more democratic countries’, and ‘relatively restrictive countries’ to convey the idea that social reality, including the realities of media and journalism, in multiple contexts is situated on a spectrum with many strands, rather than on two extreme points. While a large part of scholarly work and the available popular accounts that are produced by human rights organizations and mainstream media have focused on political repression in countries with less-than-democratic political systems, the real condition of media and journalism in these countries is, in fact, multifaceted consisting of a range of interlinked micro-, meso-, and macro-level factors. Journalism scholars have been preoccupied with political restriction as the primary macro-level force that exerts influence on journalism in these contexts (e.g., Reich and Hanitzsch, 2013; Hanitzsch and Mellado, 2011), and state censorship is addressed as the main cause of self-censorship among journalists in these contexts. ‘Self-censorship’ is therefore mainly defined as the outcome of the anticipation of risks and the conformity of the news organization to political interference in order to avoid coercive state repression (Zhang, 2014; Ahmed, 2012; Skjerdal, 2008). In addition, the existing research on the political control of media has largely failed to provide a fair assessment of the political environment in which journalists work in restrictive settings because it has disproportionately focused on tangible forms of control, such as restrictive rules and regulations, the closure of media, and arrests of journalists. Only a few studies have addressed intangible forms of obstacles such as the climate 418

Social media as resources for journalistic struggle  419

of uncertainty over the boundaries of acceptable coverage (Stern and Hassid, 2012; Tapsell, 2012). In fact, the state-imposed restrictions do not manifest themselves only in obvious forms, and a notable example is China where there is no press law and authorities’ decisions about whether and how an event should be covered is made on an event-by-event basis (Kuang, 2020). Moreover, the political field, even in the most obvious politically restrictive contexts, is not a homogeneous entity, and does not only consist of state actors, but also various actors located in different positions with antagonistic interests who confront each other. In Iran, for instance, political actors with dominance over the economic power have varying degrees of influence on different parts of the media field. Aside from the state-owned media in the country, the main privately owned media organizations are associated with political factions (Shahidi, 2007). The journalistic field, even in the most obvious politically controlled settings, is influenced by the economic field that is often fused with the political field. In Hong Kong, the connection between advertisers and the public sector gives the state the power to deter businesses from advertising in certain media, which are often critical outlets, and therefore cut their sources of income (George, 2020). The press environment in Turkey is marked with political parallelism, clientelism, and the concentration of ownership in the hands of business people who have close political or economic relationships with the government (Yeşil, 2016). In Russia, the authorities put economic pressures on critical news outlets and private media through tax audits (Becker, 2004). Chinese journalists are not only prone to the pressures from the state, but also are exposed to the influence of the market, which started to emerge in the early twenty-first century resulting in the closure of some newspapers and the economic precariousness of journalists (Tong, 2019). In the Iranian context, the economic power and the political field overlap, and the state is the most powerful source of funding for the press. This has resulted in an unequal competition between state-owned and privately owned media, as the former take advantage of better financial resources and advertising revenue (Khiabany, 2009). The economic reality of the press market is significant in understanding the financial uncertainty and precarious working conditions in which some privately owned news organizations and critical journalists live in countries such as Tajikistan (Gross and Kenny, 2011), Uganda (Skjerdal, 2011), Iran (Shahidi, 2007), China (Tong, 2019), and Hong Kong (Chan et al., 2012). While external influences that stem from the political field have received massive attention, evidence from different countries shows that even in the most obvious restrictive contexts, meso-level forces (interlinked with macro-influences) exert their particular influence on journalistic conduct. Autonomy from the state is, in fact, not the only determinant of journalistic autonomy because, as Altschull (1997) puts it, even in contexts such as the Soviet Union, ‘the content of the press is directly interrelated to the interests of those who finance the press’ (p. 261). The inconsistency between media organizations’ private interests and journalistic ideals has been a common trend across the world (Soloski, 1989; Bantz, 1997). The situation is no different in politically restrictive settings such as Turkey (Arsan, 2013), Indonesia

420  Handbook of digital politics

(Tapsell, 2012), Iran (Ranji, 2020), and China (Tong, 2011), where media organizations’ top executives tend to both preserve economic interests and avoid political risks. In China, for instance, the journalists’ stories are manipulated ‘according to a newspaper’s overall position, underlying ideology and understanding of the socio-political situation’ (Tong, 2011, p. 594). In Chinese media with a ‘more liberal political stance’, journalists enjoy more autonomy, and in more conservative media, journalists face a more limited scope of action, as these media ‘give priority to political safety, are keen to stay congruent with authority’ (Tong, 2011, p. 610). News organizations can therefore mediate, affect, intensify or weaken, external influences. It is now well established that there is a gap between professional ideals and actual practices of journalists across different countries due to the influence of the myriad factors mentioned above (Mellado and van Dalen, 2017; Tandoc et al., 2013). However, studies have suggested that journalistic ideals, their role conceptions, and perceptions of role enactment, may function as a counteracting and driving force for journalists to struggle against the constraints stemming from political forces, commercialization and organizational forces (Hassid, 2016; Ogongo-Ongong’a and White, 2008; So and Chan, 2007). Even in the countries where journalism is highly influenced by the logic external to the field, journalists’ orientations, aspirations, motivations, senses and values exert their own particular influence on journalistic practices (Ranji, 2020). Given the circumstances outlined so far, the following questions arise: How do online communication platforms and, more specifically, social media affect journalists’ power in their struggle with the constraints? What are the mechanisms through which journalists’ assets in online communications platforms operate and provide them with higher chances of success in their struggle? This chapter is devoted to those aspects of social media that are significant in relation to journalism in more restrictive environments. I identify some blind spots, address the gap in the existing knowledge, highlight some of the directions that the research is pressing towards, and set out recommendations for future research. An overview of the existing research on social media and journalism in restrictive media settings shows that this branch of literature has yet to coalesce into a coherent body of work, mainly because most of the existing works comprise descriptive studies that do not systematically frame inquiries based on a combination of theoretical perspectives and empirical material. The fragmented research studies on social media and journalism in countries such as China, Iran, Turkey, Uganda, Russia, and other restrictive environments have rarely integrated the discussions into the central concepts in the field of journalism studies such as role conception, role performance, journalistic autonomy, and professionalism. The existing accounts also suffer from the same ontological limitation that I addressed in the opening paragraphs of this chapter: the assumption that political factors explain the whole reality of media in the contexts that are deemed ‘authoritarian’, has largely remained unchanged in recent research areas of journalism and social media. Consequently, other factors such as economics, and micro- and meso-processes are neglected in accounts of the dynamics of social media in restrictive contexts.

Social media as resources for journalistic struggle  421

ONLINE PLATFORMS AND SOCIAL MEDIA Central to the discussions about social media and journalism in restrictive settings is the function of online platforms as alternative domains providing the content that is censored and under-presented in the mainstream media. The existing research argues that online platforms have facilitated an alternative space for various groupings to give themselves a voice, and have taken on the monopoly of the state over the means of communication. A great deal of this branch of literature has focused on cyber dissent, citizen journalism, and the role of social media in fostering public deliberation, mobilization, and interaction among like-minded actors in restrictive media settings such as Turkey (Ataman and Çoban, 2018), China (Yang, 2003), Iran (Sreberny and Khiabany, 2010), and Syria (Wall and El Zahed, 2015). A large number of these studies deal with ‘unprofessional’ actors, namely citizen activist journalists, as the alternative media practitioners, and focus on the content that is produced outside the institutional logic of journalism (e.g., Harcup, 2011; Atton, 2002; Downing, 2001). When it comes to investigating social media in the context of professional journalism, several studies refer to online media as alternative tools facilitating journalists to publish stories that could not be published in the established media due to politically imposed restrictions in contexts such as Malaysia, Singapore, Russia, Venezuela, and Iran (Opgenhaffen and Scheerlinck, 2014; Verweij and Van Noort, 2014; Karlekar and Radsch, 2012; Skjerdal, 2011). Online communication platforms are also said to function as a gateway for journalists to access information that is not normally accessible due to the suppression of information in restrictive environments (Ranji, 2021; Namasinga Selnes and Orgeret, 2020). Early research on online platforms and journalism in controlled environments focused on alternative journalism and blogs. Recent studies have shown that journalists in restrictive contexts, similarly to their counterparts in other countries, mainly use Facebook, Twitter, and Instagram with the exception of countries with a more centralized Internet system such as China where Weibo is one of the leading social media platforms. Social media are considered as platforms through which journalists can contest the power of the state and mainstream media. Diamond (2010) discusses that while Malaysia’s state exercises censorship, the Internet has become ‘an opening for online journalism in Malaysia’ mostly covering ‘taboo subjects as corruption, human-rights abuses, ethnic discrimination, and police brutality’ (p. 72). An investigation into the usage of social media platforms by South African journalists (Verweij and Van Noort, 2014) notes that ‘Twitter, as a journalistic network, strengthens the independence of public debate in a situation where the government aims at more control’ (p. 112). A few studies, which are mainly conducted in the Chinese context, have conceptualized the discussion on social media and journalism by building on agenda-setting theory. These studies focus on Weibo indicating that the Internet influences the flow of agenda in restrictive settings, and increases the chance of media coverage around sensitive issues (Luo, 2014; Sullivan, 2014; Wu et al., 2013). While scholars have focused on social media as platforms from which journalists are able to challenge the state in restrictive settings, my research on Iranian journalists’ use of social media shows that

422  Handbook of digital politics

it is also a domain from which to contest organizational restrictions (Ranji, 2021). Online platforms have therefore influenced the power relations, not only between journalists and political actors, but also between journalists and the dominant actors within the journalistic field such as media owners. While most research considers the involvement of the market irrelevant to the online dynamics in politically restrictive environments, few studies have shed light on the impact of the market and commercialization in these contexts. In the literature, the authoritarian control of the Internet or what MacKinnon (2010) calls ‘networked authoritarianism’, is regarded as a feature of digital communication in controlled communication settings and is understood in contrast to ‘networked commercialization’, which is identified as the digital media logic in the ‘advanced democracies’ (see, for instance, Chan, 2019). This argument may be challenged by reference to the close connection between the state and the market in some politically restrictive contexts. In China, for instance, the market contributes to the consolidation of the state’s power through information surveillance as it encourages the state to integrate new media and digital technologies, media resources and other social forces into the formation of state repression (Hou, 2020; Stockmann, 2013). The state power over online communications platforms is therefore not limited to political power in the politically restrictive context. Tong (2019) provides a counterargument to the idea of digital media as a liberating force, and argues that the capitalization of digital platforms in China, and the pursuit of entertainment and capital has resulted in taming of critical and investigative journalism. Although online platforms have been used by Chinese critical journalists as a ‘self-publishing’ platform for practising investigative journalism (Svensson, 2017), there is a growing distribution of infotainment content that gets funded by advertising (Tong, 2019). The economic pressures that independent journalists are experiencing in China may make them opt for infotainment content that is both politically safe and economically beneficial, which consequently makes them retreat from pushing reporting boundaries (Tong, 2019). A recent and growing body of research has addressed the connection between journalistic role performance and social media. Not only are the roles that journalists perform on social media and print media different (Mellado et al., 2021), but also their role performance varies across social media platforms (Hermida and Mellado, 2020). Different dimensions of social media including structure and design, aesthetics, genre conventions, rhetorical practices, interaction mechanisms and intentionality shape different journalistic roles and practices (Hermida and Mellado, 2020). The relation between journalistic role conceptions and their use of Facebook in Vietnam is examined by Vu et al. (2020) who argue that journalists’ use of social media is connected to their role conception of interpretive and populist-mobilizer functions of the news media. Stern and Hassid (2012) show that the role conceptions of those Chinese journalists who embrace micro-blogging to push the boundaries of acceptable coverage and report on sensitive issues correspond to those of advocate journalists who see the Internet as a tool for solving China’s social problems. Despite a few works, the area of journalistic roles and social media has remained understudied in most of the restrictive countries.

Social media as resources for journalistic struggle  423

DOMAINS OF CONSTANT STRUGGLE There is disagreement among scholars on whether social media facilitate democratic changes by providing a free domain for journalists. On the one hand, there are arguments that identify social media as ‘net positive’, to put it in Lewis and Molyneux’s (2018) words, and refer to the liberating possibilities of the Internet. This branch of studies addresses a positive change as the consequence of journalists’ use of social media; a change for democracy and development (e.g., Namasinga Selnes and Orgeret, 2020). Diamond (2010) refers to social networking websites as ‘liberation technology’ enabling different actors including journalists to challenge authoritarian rules (p. 71). Skjerdal offers the optimistic argument that the Internet has offered Ethiopian journalists the transformation from ‘an oppressive media environment’ to ‘a free atmosphere with new media opportunities’ (2011, p. 729). Namasinga Selnes and Orgeret (2020, p. 393) talk about the democratization possibilities of social media in Uganda: The sourcing practices of journalists offer an example of how social media may have the potential to facilitate democratisation in Uganda, expanding the range of opinions in the news and make way for the diversity of voices that can better serve democracy.

On the other hand, the Internet and, in particular, social media, are said to be one of the easiest ways for authorities to track journalists, in countries such as Iran, China, Cuba, and Nigeria, where journalists get into trouble for what they publish on social media (Akinfemisoye, 2014; Kyriakopoulou, 2011; Diamond, 2010; MacKinnon, 2010; Goode, 2009). Journalists embrace social media under the control of not only the authorities, but also news organizations (Ranji, 2021; Fu and Lee, 2016). While it is often believed that it is only the political authorities that control journalists’ activities on social media, evidence from different countries show that news media managers command journalists to comply with the organizations’ policies and interests in their personal social media accounts (Fu and Lee, 2016). My research has shown that Iranian journalists have faced dismissal, and threats of dismissal by media managers as the result of their activities on social media (Ranji, 2020). Russian journalists, as Bodrunova et al. (2021) argue, consider their news organizations’ interests when posting on social networks, and ‘they seem to be ready to limit themselves in pre-posting information from sources due to perceived conflict of interest’ (p. 2933). Beyond the question of whether or not online platforms provide ‘liberating’ and ‘democratic’ possibilities, a more realistic approach may be looking at these platforms as the domains of constant struggle between journalists and other internal actors such as media managers and external forces such as political actors. Journalists’ practices on social media are, in fact, not an exception to their everyday practices in the established media. Bodrunova et al. (2021) argue that although there are different dominant ‘perceived censors’ for online posting and editorial work among Russian journalists, practices of self-censorship are present both online and offline, which are not only caused by external political pressures, but also by individual journalists’

424  Handbook of digital politics

personal motives. Similarly to their strategies of performing in official media outlets, journalists navigate the restrictions and adopt a set of strategies for their behaviour on social media. In China, journalists use ‘discourse strategies’ to counter online censorship by using ‘transformed words’ or alternative expressions for those expressions that are sensitive (Xu, 2015). Iranian journalists tend to avoid crossing certain red lines, and address those issues on social media that allow leeway for discussion in the established media (Ranji, 2021). Journalists in restrictive settings do not often intend to challenge the whole political system on social media, but rather push the boundaries of acceptable coverage (Stern and Hassid, 2012; Ranji, 2021). In fact, by using social media, journalists can expand the boundaries of undiscussed topics only to a certain point, and not entirely. Their strategies of behaviours on social media are often located somewhere between defiance and compliance. Even in the contexts of strongly censored media, online platforms change the power relations between journalists and political actors as well as the dominant actors within the journalistic field. Social media influence power relations between journalists and the forces that restrict their practices and offer potential to journalists to move closer to the autonomous pole of the journalistic field and to sustain their social roles in a controlled media environment (Ranji, 2021). This offers possibilities for relatively more journalistic autonomy. There has been an ongoing contestation between the journalistic field and the political field over constructing reality or as Castells (2007, p. 238) says, ‘the battle over the minds of the people’. In this struggle, online ‘socialized communication’ as the domain of ‘the social production of meaning’ (Castells, 2007, p. 239) functions as a source of journalistic assets through which journalists can exert symbolic power. The Internet opens up the possibility for journalists to define reality from their own perspectives, which reflects the long-standing argument about the role of journalists as gatekeepers in constructing social reality. Online spaces are thus new tools for an old struggle in which journalists attempt to improve their strength through a variety of strategies: the struggle over exerting symbolic power. While it is often assumed that in politically restrictive contexts, those in politics shape the agenda of journalists and dominate the public discourse, journalists also influence the agenda of political actors, which is facilitated by online platforms (Ranji, 2020). The ‘decentralized’ nature of communication on the Internet allows a large number of actors to speak on an issue and thus provides an atmosphere of discussion with less centralized control (Tang and Sampson, 2012; Yang, 2003), which consequently protects journalists against the possible risks of securitizing sensitive issues. In fact, as Tang and Sampson (2012) argue, mobilization of public opinion by journalists on social media opens leeway for more legitimate coverage of sensitive issues in the established media, and amplifies ‘pressure on government officials and offers the potential for administrative intervention’ (p. 468). This consequently enables journalists to influence socio-political processes and expand not only the limits of acceptable coverage within the journalistic field but also the limits of public discourse.

Social media as resources for journalistic struggle  425

Common sense often identifies the journalists that use social media in restrictive settings as ‘activists’ who do not adhere to the norms of professional journalism. However, evidence from a few studies shows that, in fact, social media and the features of professional journalism are in dynamic coexistence. Social media have enabled journalists to act more professionally to maintain their social recognition and symbolic capital (Ranji, 2020), and conduct creative and critical reporting on sensitive issues in order to maintain their competitive edge (Hassid and Repnikova, 2016). In the process of publicizing sensitive issues through social media, in order to have leverage on an issue and exert their symbolic power, journalists need to be recognized as reliable and legitimate news sources among the public (Ranji, 2020). They should therefore build a relationship of trust with the audience, which is most likely achievable by adhering to the rules of professional journalism such as accuracy. This is why, for instance, Iranian journalists, contrary to expectations, avoid using anonymous identities and show the title ‘journalist’ on their social media pages (Ranji, 2020). In the existing knowledge, there is a tendency to refer to what Castells (2007, p. 248) terms the ‘counter-power’ dimension of social media, yet much less is known about what mechanisms are at work in online communications spaces that shape the conditions for their implementation as ‘counter-power’ platforms. There have been very few empirical investigations into the processes through which journalists’ online strategies become operative. For instance, Stern and Hassid (2012) argue that Chinese journalists draw the attention of the public to a controversial issue by discussing it on social media, and consequently this attention, in turn, may increase pressure on officials to react. My research shows that journalists can push the boundaries of acceptable reporting on social media depending on their power shaped by different resources available. In the contexts where it is risky for journalists to pursue some line of enquiry on their own, journalists’ aggregated network of connection (social capital, to put it in Bourdieu’s words), with various actors including other journalists, activists, and officials on social media is a salient factor that brings into being the possibility of shaping deliberation around controversial issues (Ranji, 2020). Iranian journalists employ social media as sources for shaping ‘news waves’, which refers to having leverage on an issue by involving a variety of actors in talking about the issue and then acting on it. In fact, in a networked context, the status of journalists as elites is contingent ‘on the networked actions of the nonelite’ (Meraz and Papacharissi, 2013, p. 161). As the boundaries over the acceptable coverage are unclear in restrictive environments, journalists’ possibilities of success in publicizing issues through online platforms depend on their familiarity with the political climate, and their assessment of the situation at any time, which is a form of journalistic cultural capital. A topic of discussion on social media cannot exist and develop into the subject of discussion out of nothing or, as Bourdieu (1991) says, ‘ex nihilo’, but it needs to be constructed as a symbolic object influencing people’s minds. A symbolic product is likely to be constructed by journalists when social, cultural, and symbolic capitals are available to them. The dynamics between journalists and other forces or

426  Handbook of digital politics

actors on online platforms play out differently for journalists depending upon where they are situated in relation to different forms of capital that define their power.

CONCLUSIONS AND FURTHER OUTLOOK Reviewing the fragmented literature on journalism and social media in controlled media environments shows that much of the current research has been descriptive and pays particular attention to the politically imposed restrictions. Despite the existence of a large number of scholarly works on social media and journalism in more restrictive environments, most of these accounts have failed to provide a contribution to the theoretical perspectives of the field. This limitation calls for a non-reductionist approach that addresses multiple elements that are at play, both in terms of why journalists embrace social media, and what factors influence their practices on online communications platforms. If we are to systematically address the questions concerning the dynamics of social media and journalism, this requires that scholars pay more attention to a variety of dimensions such as ‘social media’s impact for journalism as an institution, for journalists as individual media workers, for users/ audiences/communities engaged in news, and for the character of public discourse’ (Lewis and Molyneux, 2018, p. 18). This may require following an interdisciplinary research agenda. Scholars should therefore not only lean on theoretical perspectives in the field of media studies, but also should draw from a wide range of disciplines such as political science, sociology, and cultural studies. So far, various aspects of media and journalism across the world have been mainly evaluated in relation to the types of political systems (Zelizer, 2013; Altschull, 1997). Moreover, media studies have remained highly captive to the distinctions between ‘West’ and ‘non-West’ and between ‘democratic’ and ‘non-democratic’/‘authoritarian’ political systems as the principal starting points for understanding the realities of media. This has resulted in constructing an understanding of media and journalism in restrictive contexts as monolithic and stagnant domains, while ignoring the diversity of forms and dynamism that media and journalism have taken at different times. Rather than replicating the same descriptive accounts and making assumptions about the media dynamics based on the political system of countries, further research could benefit from combining theories, explorative perspectives and empirical material. Media scholars who have attempted to dispute ‘Western-centric’ canons (mostly through the idea of ‘de-Westernizing media theory’) have argued that ‘Western’ media theories cannot explain the realities of culturally distinct ‘non-Western’ countries. Some of them have therefore offered models based on a normative Arab-Islamic perspective (Ayish, 2003; Mowlana, 1993). Certain theoretical perspectives have been marginalized in the assessments of media in ‘non-Western’ contexts. An example is that the political economy perspective has been found to be relevant in explaining media logic in the context of ‘Western liberal democracies’ and irrelevant in the context of ‘authoritarian’ countries. This is a limitation to any comprehensive understanding of social media and journalism since this ignores the

Social media as resources for journalistic struggle  427

observable fact that the market, commercialization, and capitalization influence the dynamics of digital communication platforms in politically restrictive settings. This limitation, mostly, stems from the underlying assumption that unlike the countries that are called ‘Western’ and ‘democratic’, the realities of media and journalism in contexts that are deemed ‘non-Western’, ‘authoritarian’, or ‘non-democratic’ can be explained solely by reference to the political pressure and the states’ use of coercion. Evidence suggests that political institutions and the market are in close ties in these settings, and the political power often holds the monopoly over the economic power, and that the media field is dominated by both political and economic elites. Rather than over-focusing on the types of political systems, a much more systematic approach would need to identify how social media interact with other dimensions of the broader media system. Journalists’ social media practices in restrictive settings need theorization based on the arrangement of a range of macro- and meso-level factors, and should be discussed in connection to a variety of influences addressed by scholars in the field of news sociology. In understanding journalism in the age of social media, we need to connect the structure of the economy and the state, and shed light on the patterns of ownership of news organizations as well as the precarious economic conditions in which journalists work. There is also a need to explore how micro-level features of journalistic practice including journalistic ideals and journalistic capital are at work. In fact, as Lewis and Molyneux (2015) point out, a major pitfall, which needs to be avoided in research on social media and journalism is the assumption that ‘social media matters in a singular way, over and above other factors’ (p. 17). Further research needs to be carried out to offer a panoramic overview that critically examines media and journalism from several anchor points even in the contexts where political repression is obviously at work. Much attention has been paid to the topic of digital activism in restrictive settings and the role of activist journalists and netizens. Professional journalists are a rather understudied group, who, often, do not completely cross the boundaries of acceptable reporting and are not openly political activists. Further research could produce interesting findings by examining journalists’ strategies of practice on social media and what these strategies consist of. The existing material pays little attention to the mechanism through which online strategies of journalists become operative. This indicates a need for explanatory frameworks addressing why journalists behave in certain ways on social media (and in particular contexts), and why their online practices are shaped by certain factors (and in which conditions) and not others. What are the macro-, meso-, and micro-factors that provide room for debate around different sensitive topics? If there is contestation between journalists and other actors on social media, how and in which forms does this take place? If there is a disconnection between journalism in established media and journalism on social media in terms of journalistic role performance, how does this disconnection manifest itself? Building on and extending the previous works on social media logic and affordances, there is a need for detailed examination of how structural possibilities of media technologies can influence journalistic performance in relation to controver-

428  Handbook of digital politics

sial issues. For instance, the following questions could be addressed: How do the ways that authorities’ surveillance and filtering operate on different social media platforms, shape the conditions of journalists’ online practices? How do different technological settings and the audience dispositions make a difference in journalists’ online rhetorical strategies? And how do the presence and the network of various actors including political actors, journalists, and activists influence journalists’ performance on social media? Although there is a tendency to talk about social media as ‘free’ public spaces for discussion on sensitive issues, much uncertainty still exists about the relationship between journalists’ online practices and the public discourse in restrictive settings. This indicates a need for a systematic examination of the flow of agenda in restrictive settings and how the public discourse is shaped by competing actors, practices, and meanings. In light of the recent theoretical perspectives on gatekeeping in a networked environment (e.g., Weimann and Brosius, 2017; Meraz and Papacharissi, 2013), further research could conduct multi-methodological studies to develop a more comprehensive picture of the patterns of communications in online spaces in restrictive contexts. It is useful to address how and in what ways diverse actors in various positions with different agendas including activists, journalists, ordinary users, and politicians influence each other and shape the public discourse around an event.

FURTHER READING Bodrunova, S. S., Litvinenko, A., and Nigmatullina, K. (2021). Who is the censor? Self-censorship of Russian journalists in professional routines and social networking. Journalism, 22(12), 2919–2937. Castells, M. (2007). Communication, power and counter-power in the network society. International Journal of Communication, 1(1), 238–266. Chan, J. M. (2019). From networked commercialism to networked authoritarianism: The biggest challenge to journalism. Journalism, 20(1), 64–68. Hassid, J. and Repnikova, M. (2016). Why Chinese print journalists embrace the Internet. Journalism, 17(7), 882–898. Ranji, B. (2021). Shaping news waves and constructing events: Iranian journalists’ use of online platforms as sources of journalistic capital. New Media & Society, 23(7), 1936–1952. Stockmann, D., Luo, T., and Shen, M. (2020). Designing authoritarian deliberation: How social media platforms influence political talk in China. Democratization, 27(2), 243–264. Tong, J. (2019). The taming of critical journalism in China. Journalism Studies, 20(1), 79–96. Weimann, G. and Brosius, H. B. (2017). Redirecting the agenda. The Agenda Setting Journal, 1(1), 63–102. Xu, D. (2015). Online censorship and journalists’ tactics. Journalism Practice, 9(5), 704–720.

REFERENCES Ahmed, A. M. (2012). Official secrecy, self-censorship and political parallelism: A study on the Bangladesh press. Media Asia, 39(1), 23–31.

Social media as resources for journalistic struggle  429

Akinfemisoye, M. O. (2014). Negotiating convergence: ‘Alternative’ journalism and institutional practices of Nigerian journalists. Digital Journalism, 2(1), 62–76. Altschull, J. H. (1997). Boundaries of journalistic autonomy. In D. A. Berkowitz (ed.), Social Meanings of News: A Text-Reader (pp. 259–268). Thousand Oaks, CA: Sage (Original work published 1995). Arsan, E. (2013). Killing me softly with his words: Censorship and self-censorship from the perspective of Turkish journalists. Turkish Studies, 14(3), 1–16. Ataman, B. and Çoban, B. (2018). Counter-surveillance and alternative new media in Turkey. Information, Communication & Society, 21(7), 1014–1029. Atton, C. (2002). Alternative Media. London: Sage. Ayish, M. I. (2003). Beyond western-oriented communication theories: A normative Arab-Islamic perspective. Javnost – The Public, 10(2), 79–92. Bantz, R. C. (1997). News organizations: Conflict as a crafted cultural norm. In D. A. Berkowitz (ed.), Social Meanings of News: A Text-Reader (pp. 123–137). Thousand Oaks, CA: Sage (Original work published 1985). Becker, J. (2004). Lessons from Russia: A neo-authoritarian media system. European Journal of Communication, 19(2), 139–163. Bodrunova, S. S., Litvinenko, A., and Nigmatullina, K. (2021). Who is the censor? Self-censorship of Russian journalists in professional routines and social networking. Journalism, 22(12), 2919–2937. Bourdieu, P. (1991). Language and Symbolic Power. Cambridge: Polity Press (Original work published 1982). Castells, M. (2007). Communication, power and counter-power in the network society. International Journal of Communication, 1(1), 238–266. Chan, J. M. (2019). From networked commercialism to networked authoritarianism: The biggest challenge to journalism. Journalism, 20(1), 64–68. Chan, J. M., Lee, F. L. F., and So, C. Y. K. (2012). Journalists in Hong Kong, a decade after the transfer of sovereignty. In D. H. Weaver and L. Willnat (eds.), The Global Journalist in the 21st Century (pp. 22–35). New York: Routledge. Diamond, L. (2010). Liberation technology. Journal of Democracy, 21(3), 69–83. Downing, J. (2001). Radical Media. Thousand Oaks, CA: Sage. Fu, J. and Lee, A. (2016). Chinese journalists’ discursive weibo practices in an extended journalistic sphere. Journalism Studies, 17(1), 80–99. George, C. (2020). Journalism and authoritarian resilience. In K. Wahl-Jorgensen and T. Hanitzsch (eds.), The Handbook of Journalism Studies (pp. 538–554). London: Routledge. Goode, L. (2009). Social news, citizen journalism and democracy. New Media & Society, 11(8), 1287–1305. Gross, P. and Kenny, T. (2008). Journalism in Central Asia: A victim of politics, economics, and widespread self-censorship. The International Journal of Press/Politics, 13(4), 515–525. Hanitzsch, T. and Mellado, C. (2011). What shapes the news around the world? How journalists in eighteen countries perceive influences on their work. The International Journal of Press/Politics, 16(3), 404–426. Harcup, T. (2011). Alternative journalism as active citizenship. Journalism, 12(1), 15–31. Hassid, J. (2016). China’s Unruly Journalists: How Committed Professionals Are Changing the People’s Republic. London: Routledge. Hassid, J. and Repnikova, M. (2016). Why Chinese print journalists embrace the Internet. Journalism, 17(7), 882–898. Hermida, A. and Mellado, C. (2020). Dimensions of social media logics: Mapping forms of journalistic norms and practices on Twitter and Instagram. Digital Journalism, 8(7), 864–884.

430  Handbook of digital politics

Hou, R. (2020). The commercialisation of Internet-opinion management: How the market is engaged in state control in China. New Media & Society, 22(12), 2238–2256. Karlekar, D. K. and Radsch, C. C. (2012). Adapting concepts of media freedom to a changing media environment: Incorporating new media and citizen journalism into the Freedom of the Press Index. Journal for Communication Studies, 5(1), 13–22. Khiabany, G. (2009). Iranian Media: The Paradox of Modernity. New York: Routledge. Kuang, X. (2020). Self-caging or playing with the edge? News selection autonomy in authoritarian China. SAGE Open, 10(2), 215824402092298. Kyriakopoulou, K. (2011). Authoritarian states and internet social media: Instruments of democratization or instruments of control? Human Affairs, 21, 18–26. Lewis, S. C. and Molyneux, L. (2018). A decade of research on social media and journalism: Assumptions, blind spots, and a way forward. Media and Communication, 6(4), 11–23. Luo, Y. (2014). The Internet and agenda setting in China: The influence of online public opinion on media coverage and government policy. International Journal of Communication, 8(24), 1289–1312. MacKinnon, R. (2010). Networked authoritarianism in China and beyond: Implications for global internet freedom. Paper presented at the conference on Liberation Technology in Authoritarian Regimes, Hoover Institution and Center on Democracy, Stanford University. https://​rconversation​.blogs​.com/​MacKinnon​_Libtech​.pdf. Mellado, C., Humanes, M. L., Scherman, A., and Ovando, A. (2021). Do digital platforms really make a difference in content? Mapping journalistic role performance in Chilean print and online news. Journalism, 22(2), 358–377. Mellado, C. and Van Dalen, A. (2014). Between rhetoric and practice: Explaining the gap between role conception and performance in journalism. Journalism Studies, 15(6), 859–878. Meraz, S. and Papacharissi, Z. (2013). Networked gatekeeping and networked framing on #Egypt. The International Journal of Press/Politics, 18(2), 138–166. Mowlana, H. (1993). The new global order and cultural ecology. Media, Culture & Society, 15(1), 9–27. Namasinga Selnes, F. and Orgeret, K. (2020). Social media in Uganda: Revitalising news journalism? Media, Culture & Society, 42(3), 380–397. Ogongo-Ongong’a, S. and White, R. A. (2008). The shaping of the news values of young journalists in Kenya. African Communication Research, 1(2), 159–184. Opgenhaffen, M. and Scheerlinck, H. (2014). Social media guidelines for journalists. Journalism Practice, 8(6), 726–741. Ranji, B. (2020). Journalistic illusio in a restrictive context: Role conceptions and perceptions of role enactment among Iranian journalists. Journalism, 23(2). https://​doi​.org/​10​.1177/​ 1464884920922026. Ranji, B. (2021). Shaping news waves and constructing events: Iranian journalists’ use of online platforms as sources of journalistic capital. New Media & Society, 23(7), 1936–1952. Reich, Z. and Hanitzsch, T. (2013). Determinants of journalists’ professional autonomy: Individual and national level factors matter more than organizational ones. Mass Communication and Society, 16(1), 133–156. Shahidi, H. (2007). Journalism in Iran: From Mission to Profession. London: Routledge. Skjerdal, T. (2008). Self-censorship among news journalists in the Ethiopian state media. African Communication Research, 1(2), 185–206. Skjerdal, T. (2011). Journalists or activists? Self-identity in the Ethiopian diaspora online community. Journalism, 12(6), 727–744. So, C. Y. K. and Chan, J. M. (2007). Professionalism, politics and market force: Survey studies of Hong Kong journalists 1996–2006. Asian Journal of Communication, 17(2), 148–158. Soloski, J. (1989). News reporting and professionalism: Some constraints on the reporting of the news. Media, Culture & Society, 11(2), 207–228.

Social media as resources for journalistic struggle  431

Sreberny, A. and Khiabany, G. (2010). Blogistan: The Internet and Politics in Iran. London: I. B. Tauris. Stern, R. E. and Hassid, J. (2012). Amplifying silence: Uncertainty and control parables in contemporary China. Comparative Political Studies, 45(10), 1230–1254. Stockmann, D. (2013). Media Commercialization and Authoritarian Rule in China. New York: Cambridge University Press. Sullivan, J. (2014). China’s Weibo: Is faster different? New Media & Society, 16(1), 24–37. Svensson, M. (2017). The rise and fall of investigative journalism in China: Digital opportunities and political challenges. Media, Culture & Society, 39(3), 440–445. Tandoc Jr, E. C., Hellmueller, L., and Vos, T. P. (2013). Mind the gap: Between journalistic role conception and role enactment. Journalism Practice, 7(5), 539–554. Tang, L. and Sampson, H. (2012). The interaction between mass media and the internet in non-democratic states: The case of China. Media, Culture & Society, 34(4), 457–471. Tapsell, R. (2012). Old tricks in a new era: Self-censorship in Indonesian journalism. Asian Studies Review, 36(2), 227–245. Tong, J. (2011). Investigative Journalism in China: Journalism, Power, and Society. London: Continuum. Tong, J. (2019). The taming of critical journalism in China. Journalism Studies, 20(1), 79–96. Verweij, P. and Van Noort, E. (2014). Journalists’ Twitter networks, public debates and relationships in South Africa, Digital Journalism, 2(1), 98–114. Vu, H. T., Trieu, L. T., and Nguyen, H. T. (2020). Routinizing Facebook: How journalists’ role conceptions influence their social media use for professional purposes in a socialist-communist country. Digital Journalism, 8(7), 885–903. Wall, M. and El Zahed, S. (2015). Syrian citizen journalism. Digital Journalism, 3(5), 720–736. Weimann, G. and Brosius, H. B. (2017). Redirecting the agenda. The Agenda Setting Journal, 1(1), 63–102. Wu, Y., Atkin, D., Lau, T. Y., et al. (2013). Agenda setting and micro-blog use: An analysis of the relationship between Sina Weibo and newspaper agendas in China. The Journal of Social Media in Society, 2(2), 8–25. Xu, D. (2015). Online censorship and journalists’ tactics. Journalism Practice, 9(5), 704–720. Yang, G. (2003). The internet and civil society in China: A preliminary assessment. The Journal of Contemporary China, 12(36), 453–475. Yeşil, B. (2016). Media in New Turkey: The Origins of an Authoritarian Neoliberal State. Champaign, IL: University of Illinois Press. Zelizer, B. (2013). On the shelf life of democracy in journalism scholarship. Journalism, 14(4), 459–473. Zhang, Z. (2014). Self-censorship in news production: Findings from reports on the toxic powder scandal. In M. Svensson, E. Sæther, and Z. Zhang (eds.), Chinese Investigative Journalists’ Dreams: Autonomy, Agency, and Voice (pp. 133–157). Lanham, MD: Lexington Books.

28. Fake news and digital politics Bente Kalsnes

INTRODUCTION The rise of fake news is closely connected to politics. The term became popularized and politicized during the 2016 US elections and its global proliferation has created concerns about vulnerabilities of individuals, institutions and society to information manipulations by malicious actors (Quandt et al., 2019). Politicians’ opportunities to disseminate disinformation directly to the public, bypassing the media’s gatekeeping and editorial scrutiny, have increased with the rise of social media, the possibilities they afford and with a weakening of what Graves and Wells (2019, p. 42) call ‘factual accountability’ (Tsfati et al., 2020). Fake news can be seen as an expression of a larger and fundamental shift within the technological and political underpinnings of mediated communication in modern democracies – particularly on how we inform ourselves about the world around us and what sources, platforms and devices we use in this process. The rise of rumours and conspiracy theories through the course of the coronavirus pandemic have also demonstrated how challenging the spread of false stories can be during a health crisis (Brennen et al., 2020). While it might be tempting to disregard fake news as baseless articles most people will discover are fake, there are several examples of the confusion they can create. The story that claimed that ‘Pope Francis shocks world, endorses Donald Trump for president’, became one of the most shared fake news stories during the US election in 2016 (Ritchie, 2016). Misperceptions can thus be one of the consequences of exposure to fake news. Lack of trust is another. Declining public trust in politicians and public institutions has long been a concern in many countries, and the spread of fake news has added additional concerns for this challenge. While prominent media outlets such as The New York Times, CNN, and Buzzfeed News have used the term to designate misleading information spread online, President Donald Trump has used the term as a negative designation of these very same ‘mainstream media’ (Farkas and Schou, 2018). Thus, the term fake news has effectively been weaponized by political actors to attack a variety of news media (Vosoughi et al., 2018). Despite the heightened attention to this phenomenon, important aspects are still only marginally understood (Damstra et al., 2021). The fundamental challenge is the transformations of the public sphere driven by the digital proliferation. Fake news’ connection to digital politics and particular elections is the focus of this chapter. I will now establish a theoretical baseline for the discussion by outlining how I apply the term fake news in relation to digital politics. I differentiate between what we here call a theoretical approach and an empirical approach. The theoretical approach addresses how the term fake news has been defined and applied in the 432

Fake news and digital politics  433

research literature, and the empirical approach addresses how political actors use the term in the ‘real world’, often with strategic intentions. As these approaches do not necessarily align, I have divided them into two separate parts. In the following paragraphs, I will outline how the term fake news has been defined, what role it plays in digital politics and lastly, future avenues for research.

THE ANATOMY OF FAKE NEWS RESEARCH The rise of deliberate disinformation – nowadays often referred to as fake news – has sparked societal as well as academic debates. Previously, the term primarily denoted inaccurate news pieces (often intentionally fabricated), or was used as a more specific term for political satire in the form of staged news shows (Tandoc et al., 2018b). In the field of political communication and journalism, research on fake news can be roughly divided into three sub-areas, as suggested by Damstra et al. (2021). First, there is conceptual work focusing on theoretical clarification, offering fake news typologies and connecting the recent upsurge to the context of changing media landscapes (e.g. Egelhofer and Lecheler, 2019; Søe, 2019; Tandoc et al., 2018b; Waisbord, 2018). Second, a strand of empirical work studies the dynamics, spread, and effects of fake news (e.g. Grindberg et al., 2019; Guess et al., 2018; Pierri et al., 2020; Silverman, 2016), identifying the scope of exposure as well as the most relevant disseminators, and, as far as possible, assessing its impact on citizens’ attitudes and beliefs (see also Tsfati et al., 2020). Finally, there is work focusing on ways to tackle the spread and impact of disinformation (e.g. Hameleers, 2022; Lazer et al., 2017; Walter et al., 2020). In short, fake news has been defined as ‘articles based on false information packaged to look like real news to deceive readers either for financial or ideological gain’ (Tandoc et al., 2018a, p. 674). Other scholars have defined fake news as ‘news articles that are intentionally and verifiably false, and could mislead readers’ (Allcott and Gentzkow, 2017, p. 13) or as ‘complete or partly false information, (often) appearing as news, and typically expressed as textual, visual or graphical content with an intention to mislead or confuse users’ (Kalsnes, 2018, p. 4). The format, the intention to deceive and the political purpose are core features of these definitions that we will return to in the following sections. Historically, fake news has been applied much broader in the research literature to describe everything from news parody and news satire, to native advertising, propaganda, manipulation, and fabrication (Tandoc et al., 2018a, 2018b; see also Kalsnes et al., 2021 for a discussion). Journalism and news is closely linked to knowledge production, as the authority of journalism has historically rested on its ability to establish a robust connection between news and truth. News is supposedly – and normatively – based on truth, which makes the term ‘fake news’ an oxymoron (Tandoc et al., 2018b, p. 140). Modern journalism has been characterized by continuous ‘efforts to claim to provide authoritative reportage of current events’ (Waisbord, 2018, p. 1866). Journalists are expected to deliver independent, trustworthy, accurate and comprehensive information (Kovach and

434  Handbook of digital politics

Rosenstiels, 2007, p. 11). It is still worth mentioning the inherent ambiguity in news’ relationship to reality and truth (Kalsnes, 2019). Michael Schudson (1989) describes the historical controversy between journalists and media researchers over how to define news. Journalists have insisted that they only report the world as they see it – that is, facts – because ‘a responsible journalist never, never, never fakes the news’. Media researchers, on the other hand, argued that journalists ‘construct news’ and make ‘social constructions of reality’ (Schudson, 1989, p. 263). This does not mean that journalists ‘fake the news’ or fabricate the news, but that journalists ‘make the news’. A similar argument is raised by Gaye Tuchman: To say that a news report is a story, no more, but no less, is not to demean the news, not to accuse it of being fictitious. Rather, it alerts us that news, like all public documents, is a constructed reality possessing its own internal validity. A selective reality, rather than a synthetic reality as in literature, news reports exist in and of themselves. They are public documents that lay a world before us. (Tuchman, 1978, p. 97)

Like other buzzwords, ‘fake news’ is semantically confusing (Corner, 2017; Tandoc et al., 2018b). ‘Deceptive news’ or ‘intentional deceptive information’ (Damstra et al., 2021) are similar terms used to describe fake news. In the past, scholars had used ‘fake news’ to describe information that adopted conventional news formats to make satirical commentary, as in the case of late-night television shows, as well as tabloid journalism that walked a fine line between reporting reality and making wild claims (Hartley, 1996). Recently, fake news has primarily been used to refer to content featuring the style of conventional news intended to deliberately misinform. This version of fake news fundamentally refers to fabricated information that astutely mimics news and taps into existing public beliefs to influence electoral behaviour. Understood as information divorced from reality, fake news is not new. Deceitful information wrapped in news packages has a longer history than news consciously produced to represent real events. What is different is the speed, scale and massive proliferation and consumption of false information disseminated on dominant digital platforms such as Facebook and Twitter (Waisbord, 2018). Misinformation and contested truths are constitutive of today’s dynamics of the multilayered, chaotic public communication environment. Egelhofer and Lecheler (2019) have argued that a message should only be studied as ‘fake news’ when it is low in facticity, was created with the intention to deceive, and is presented in a journalistic format. Here I will pay less attention to the level of facticity, but rather go into more details on two crucial characteristics mentioned in the definition outlined by Tandoc et al. (2018b): journalistic format and intention to deceive. The Journalistic Format The appearance of news has been a recurring characteristic of fake news, as alluded to in the term itself. Article features such as a headline, text body and photo are typically used in fake news stories, and in this sense, false information is masqueraded as

Fake news and digital politics  435

news. By mimicking the news format (Lazer et al., 2018, p. 1094), fake news stories try to steal trustworthiness from journalism which is produced according to a professional standard. The news format can also make it easier to spread false or misleading information online, often done unintentionally. Thus, it is important to distinguish between the creation of fake news, which is done intentionally, and its dissemination which may be unintentional (Wardle and Derakhshan, 2017). Intention to Deceive To better differentiate between the terms, researchers suggest distinguishing between false information that has an intention to deceive and that which does not (Tandoc et al., 2018a, 2018b; Wardle and Derakhshan, 2017). While the intention to deceive is low in a genre such as news satire and parody, propaganda has the opposite characteristic, since it has been characterized as a form of communication that aims at shaping public opinion in a way that gratifies the propagandist’s concealed agenda (e.g. Jowett and O’Donnell, 2014). Disinformation is an overarching term that describes an intention to deceive through information. Disinformation can be defined as ‘false, inaccurate, or misleading information designed, presented and promoted to intentionally cause public harm or for profit’ (HLEG, 2018, p. 10). Thus, disinformation includes false or misleading information also outside of the news format or the institution of news, but if disinformation is packaged in a journalistic format, it emerges as fake news. Along the same lines but without an intention to deceive is misinformation, which is defined as incorrect or misleading information that is disseminated unintentionally (e.g. Bakir and McStay, 2018; HLEG, 2018; Lazer et al., 2018; Wardle and Derakhshan, 2017). Fake news can thus be understood as disinformation presented in the news format.

THE POLITICAL ASPECTS OF FAKE NEWS Turning the focus to the more empirical role of fake news in politics, this section will look into how politicians have been associated with the term, both in terms of labelling fake news, producing fake news and creating a legal framework to forbid fake news. Labelling Something or Someone as Fake News There exists a long history of politicians labelling media and journalists as fake or lying, most famously through the Nazis’ use of the label ‘Lügenpresse’, the lying press. The term was used by the German National Socialist Party before and during the Third Reich to discredit journalism as a practice and to undermine public trust in the media institution (Koliska and Assmann, 2021). Similar to the term ‘Lügenpresse’ in Germany, the term fake news has become a potent weapon for a number of political actors all over the world who use it to discredit legacy news

436  Handbook of digital politics

media that conduct critical journalism (e.g. Vosoughi et al., 2018; Egelhofer and Lecheler, 2019), and particularly in the US. In recent times, the term fake news has been used by politicians as a label, often used to attack political opponents and news media. Thus, fake news has become weaponized, and Egelhofer and Lecheler (2019) argue that fake news alludes to two dimensions of political communication: the fake news genre (i.e. the deliberate creation of pseudo-journalistic disinformation) and the fake news label (i.e. the instrumentalization of the term to delegitimize news media). Former president of the United States Donald Trump has been credited with setting the term fake news on the political agenda. Donald Trump politicized the term and used it to discredit established media outlets such as CNN or the New York Times. As reported by the New York Times, in countries where press freedom is restricted or under considerable threat – such as Russia, China, Turkey, Libya, Poland, Hungary, Thailand, Somalia and others – political leaders have invoked fake news as justification for beating back media scrutiny (Erlanger, 2017; Kalsnes, 2018). By suggesting that news cannot be trusted and by labelling it fake news, politicians deliberately undermine trust in journalism and news outlets. When used by political elites, the fake news label – as well as other delegitimizing attacks – might consequently affect citizens’ media perceptions (Ladd, 2012). What makes it such a pressing concern today is that the digital media environment enhances the enormous and rapid spread of information. For example, we know from several studies that Facebook was a key disseminator of fake news stories during the 2016 US presidential election campaign (e.g. Guess et al., 2018). Against this backdrop, some have argued that we have entered ‘a post-truth era’ (Benkler et al., 2018, p. 23) which has provoked renewed academic interest in deceptive information dressed as news. As elite rhetoric is a powerful factor influencing the formation of opinion, political elite attacks on the media have an impact on how citizens perceive such media. Reporters Without Borders has reported that there is a growing habit of authoritarian regimes who systematically brand factual information which does not match the official narrative as fake news. The organization has warned that ‘Predators of press freedom have seized on the notion of “fake news” to muzzle the media on the pretext of fighting false information’ (RSF, 2017) and pointed to how authoritarian regimes have used Donald Trump to justify attacks on local news media. Producing Fake News The phenomenon of politicians as producers of fake news is also something that has been documented recently. Particularly Brazilian President Jair Bolsonaro has been accused and investigated for his denialist stance during the Covid pandemic for conveying misinformation, particularly regarding the symptoms, risks, and cures of the virus, and encouraging risky behaviour (Ricard and Medeiros, 2020). President Bolsonaro’s recurring statements about Covid-19 were one of the main vectors of misleading content, according to Ricard and Medeiros. For example, through his periodic social media live-streamed videos as well as official government channels, Bolsonaro promoted erroneous information about the effects and cures of the virus,

Fake news and digital politics  437

based on unknown data or inconclusive scientific evidence. During the Brexit referendum in the UK, several examples of fake news claims appeared, either as fake fact-checks, misleading health claims or fake leaflets, all produced by politicians (Goldhill, 2019). It has also been demonstrated how politicians have been major amplifiers of fake news made by others. In a study of the spread of misinformation in six countries during the Covid-19 health crisis, one of the findings was that there were many examples of high-level politicians, celebrities, and other prominent public figures pushing false or misleading information via press conferences, official channels, and social media (Nielsen et al., 2020). Similarly, another study found that while the majority of Covid-19 misinformation on social media came from ‘ordinary’ people, most of these posts created less engagement compared to top-down misinformation from politicians, celebrities, and other prominent public figures (Brennen et al., 2020). Misinformation from politicians, celebrities, and other prominent public figures made up just 20 per cent of the claims in the sample but accounted for 69 per cent of total social media engagement. Thus, politicians are able to amplify and spread misinformation to a large audience. As lies spread faster than the truth (Vosoughi et al., 2018), politicians have a particular responsibility to express themselves truthfully and not share false information, as well as to avoid labelling news they do not like fake news. Prohibiting Fake News and Undermining Freedom of Speech Another way the term fake news has been used for political purposes is by criminalizing and prohibiting the production and spreading of fake news by law. For authoritarian leaders who want to impose their version of events, information control is key, and one way to do that is by ‘fighting’ fake news through the justice system. Here are some international examples of how authoritarian regimes are using the fake news label to target critics (RSF, 2017): ● Russia has put a legal ban on the ‘dissemination of false information’. ● In Côte d’Ivoire, insulting the head of state or the dissemination of false news reports may be enough for a journalist to be taken into custody. ● In Madagascar, a new communications code provides heavy fines for infringements ranging from insults to defamation, and refers to the dissemination of ‘false news’. ● In Somalia, the Universal TV channel was suspended for broadcasting false reports alleged to have threatened the stability and peace of the region after it referred to overseas trips by the president. ● The South African government plans to impose a system to control the media in order to meet the ‘challenge’ of ‘fake news’. ● In Egypt, journalists are frequently accused of disseminating false information whenever they criticize the government, or report on sensitive issues that upset it.

438  Handbook of digital politics

The United Nations, along with a group of other organizations,1 have issued a statement (2017) expressing concern at the use of ‘fake news’ for government propaganda and to curb press freedom: Alarmed at instances in which public authorities denigrate, intimidate and threaten the media, including by stating that the media is ‘the opposition’ or is ‘lying’ and has a hidden political agenda, which increases the risk of threats and violence against journalists, undermines public trust and confidence in journalism as a public watchdog, and may mislead the public by blurring the lines between disinformation and media products containing independently verifiable facts.

Russian expansion of laws criminalizing ‘fake news’ during the invasion of Ukraine in February 2022 has demonstrated how damaging such laws are for freedom of speech and independent journalism. The new laws, approved by lawmakers in the state Duma, Russia’s lower house of parliament, threaten those who ‘knowingly’ spread so-called false information about all Russian state bodies operating abroad with fines of 1.5 million roubles (around €13,000) and prison sentences of up to 15 years (Jack, 2022).

CONSUMERS OF FAKE NEWS AND THE ROLE OF IDEOLOGY Turning now to consumers of fake news – what do we know about them and their ideologies? Quite a bit is known about how ideology impacts people’s motivation to believe in fake news. Ideological bias has been identified as one clear characteristic of fake news, and can be found on both left and right leaning websites. Fake news is also shared by people all over the political spectrum. Nevertheless, in an extensive review of the fake news literature, researchers found that intentionally deceptive information tends to be ideologically biased in favour of the right (Damstra et al., 2021). From an American perspective, a study found that strong Republicans engaged with more partisan and unreliable news than strong Democrats did, despite the two groups being exposed to similar amounts of partisan and unreliable news in their Google search results (Robertson et al., 2021). While a majority of the included studies had an American focus, the ideological bias was also found in Italian (Pierri et al., 2020) and German (Zimmermann and Kohring, 2020) studies. In the United States, the political patterns are similar to those found for trust – people on the left worry about what they see as misinformation from individual politicians and the national government, people on the right about what they see as misinformation from news organizations (Brennen et al., 2020). Thus, political engagement outweighs exposure to partisan and unreliable news within Google searches. Research by Tsfati et al. (2020) indicates that the reach of fake news websites is limited to small parts of the population. On the other hand, data demonstrate that large proportions of the public know about notable fake news stories and believe

Fake news and digital politics  439

them. These findings imply the possibility that most people hear about fake news stories not from fake news websites but through their coverage in mainstream news outlets. Several studies have looked into the political leanings of fake news sites, referring to false information intentionally spread, masked as traditional news, to advance political goals. Two well-known examples (Tsfati et al., 2020) are a complex of fake news websites run by teenagers from a small town in Macedonia and a US company called Disinfomedia, owning many sites disguised as serious journalism (including USAToday.com.co and WashingtonPost.com.co). Both operations spread pro-Trump and anti-Clinton fake news stories prior to the 2016 US elections (Allcott and Gentzkow, 2017, p. 217), as did many so-called alternative news websites in the right-wing media ecosystem (Benkler et al., 2018).

KEY PRIORITIES FOR FUTURE RESEARCH IN THIS AREA While fake news might have been understood as deceitful websites based on factless stories, it can also be seen as a symptom of fundamental distrust in society – distrust in journalism as an institution, as well as in the political and judicial system. Four years of fact-checking Donald Trump’s alleged 30,573 lies as president did not stop 74 million Americans from voting for him again (Hansen, 2021). The global Covid-19 health crisis has demonstrated that disinformation and fake news about treatments and remedies can have serious health consequences. High-level politicians bear a special responsibility for ensuring that they communicate clearly, accurately, and honestly about such pandemics. If they do not, the consequences can be severe, as people will misunderstand the situation and put themselves, their loved ones, and their communities at greater risk. Despite fake news and disinformation having become frequently used buzzwords in the last few years, also in academic studies, several aspects of the connection between fake news and political communication are not understood well enough. Here, I will suggest three key priorities for the future research of fake news. First, as suggested by Egelhofer and Lecheler (2019), studies need to investigate the fake news label’s impact on media perceptions outside of the US, as concern about the usage of the term fake news is growing around the world, even more so in Austria (56 per cent) and Bulgaria (53 per cent) than in the US (48 per cent) (Newman et al., 2018, pp. 37–38). Here, a possibility could be to test how politicians’ statements concerning media outlets’ credibility might increase levels of trust in these outlets for citizens who support these politicians. Secondly, more research is needed into what types of corrections work and in which context. As false information is easily spread, particularly on social media platforms such as Facebook and YouTube, and even more so on closed communication communities such as WhatsApp and Parler, more knowledge about different types and formats of corrections is needed.

440  Handbook of digital politics

Thirdly, studies should look more carefully at the role of social media platforms, and their moderation practice in regard to false and deceptive information. Platforms moderate and block content through human moderators and machine learning, but particularly the lack of transparency and insight into the moderation practices have been heavily criticized. Thus, we need more insight into how the moderation operations are put into practice and how they impact digital spheres and freedom of speech in different countries.

NOTE 1.

The Organization for Security and Co-operation in Europe, the Organization of American States, and the African Commission on Human and Peoples’ Rights.

FURTHER READING Albright, J. (2017). Welcome to the era of fake news. Media and Communication, 5(2), 87–89. Egelhofer, J. L. and Lecheler, S. (2019). Fake news as a two-dimensional phenomenon: A framework and research agenda. Annals of the International Communication Association, 43(2), 97–116. Flynn, D. J., Nyhan, B., and Reifler, J. (2017). The nature and origins of misperceptions: Understanding false and unsupported beliefs about politics. Political Psychology, 38, 127–150. Frankfurt, H. G. (2005). On Bullshit. Princeton, NJ: Princeton University Press. Hochschild, J. L. and Einstein, K. L. (2015). Do Facts Matter? Information and Misinformation in American Politics. Norman: University of Oklahoma Press. Karlsen, R. (2015). Followers are opinion leaders: The role of people in the flow of political communication on and beyond social networking sites. European Journal of Communication, 30(3), 301–318. Lewandowsky, S., Ecker, U. K. H., Seifert, C. M., Schwarz, N., and Cook, J. (2012). Misinformation and its correction: Continued influence and successful debiasing. Psychological Science in the Public Interest, 13(3), 106–131. Nyhan, B., Porter, E., Reifler, J., and Wood, T. J. (2020). Taking fact-checks literally but not seriously? The effects of journalistic fact-checking on factual beliefs and candidate favorability. Political Behavior, 42(3), 939–960. Nyhan, B. and Reifler, J. (2010). When corrections fail: The persistence of political misperceptions. Political Behavior, 32(2), 303–330. Woolley, S. C. and Howard, P. N. (2018). Computational Propaganda: Political Parties, Politicians, and Political Manipulation on Social Media. New York: Oxford University Press.

REFERENCES Allcott, H. and Gentzkow, M. (2017). Social media and fake news in the 2016 election. Journal of Economic Perspectives, 31(2), 211–236. Bakir, V. and McStay, A. (2018). Fake news and the economy of emotions. Digital Journalism, 6(2), 154–175.

Fake news and digital politics  441

Benkler, Y., Faris, R., and Roberts, H. (2018). Network Propaganda: Manipulation, Disinformation, and Radicalization in American Politics. New York: Oxford University Press. Brennen, S., Simon, F., Howard, P. N., and Nielsen, R. K. (2020). Types, sources, and claims of COVID-19 misinformation. Reuters Institute for the Study of Journalism. https://​ reutersinstitute​.politics​.ox​.ac​.uk/​types​-sources​-and​-claims​-covid​-19​-misinformation. Corner, J. (2017). Fake news, post-truth and media–political change. Media, Culture & Society, 39(7), 1100–1107. Damstra, A., Boomgaarden, H. G., Broda, E., Lindgren, E., Strömbäck, J., Tsfati, Y., and 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(14), 1947–1963. Egelhofer, J. L. and Lecheler, S. (2019). Fake news as a two-dimensional phenomenon: A framework and research agenda. Annals of the International Communication Association, 43(2), 97–116. Erlanger, S. (2017). ‘Fake news,’ Trump’s obsession, is now a cudgel for strongmen. New York Times, 12 December. Farkas, J. and Schou, J. (2018). Fake news as a floating signifier: Hegemony, antagonism and the politics of falsehood. Javnost – The Public, 25(3), 298–314. Goldhill, O. (2019). Politicians are embracing disinformation in the UK election. Quartz. https://​qz​.com/​1766968/​uk​-election​-politicians​-embrace​-fake​-news​-disinformation/​. Graves, L. and Wells, C. (2019). From information availability to factual accountability. In J. Katz and K. K. Mays (eds.), Journalism and Truth in an Age of Social Media (pp. 39–57). New York: Oxford University Press. Grindberg, N., Joseph, K., Friedland, L., Swire-Thompson, B., and Lazer, D. (2019). Fake news on Twitter during the 2016 US presidential election. Science, 363(6425), 374–378. Guess, A., Nyhan, B., and Reifler, J. (2018). Selective Exposure to Misinformation: Evidence from the Consumption of Fake News during the 2016 US Presidential Campaign. European Research Council. http://​www​.dartmouth​.edu/​~nyhan/​fake​-news2016​.pdf. Hameleers, M. (2022). Separating truth from lies: Comparing the effects of news media literacy interventions and fact-checkers in response to political misinformation in the US and Netherlands. Information, Communication & Society, 25(1), 110–126. Hansen, E. E. (2021). The financing of journalism solved, let’s move on to 3 bigger problems. INMA. https://​www​.inma​.org/​blogs/​ideas/​post​.cfm/​the​-financing​-of​-journalism​ -theoretically​-solved​-it​-s​-time​-to​-move​-on​-to​-larger​-problems. Hartley, J. (1996). Popular Reality: Journalism, Modernity, Popular Culture. London: Hodder Education. High Level Expert Group (HLEG) on Fake News and Online Disinformation (2018). A Multi-Dimensional Approach to Disinformation. Brussels: European Commission. Jack, V. (2022). Russia expands laws criminalizing ‘fake news’. Politico. https://​www​.politico​ .eu/​article/​russia​-expand​-laws​-criminalize​-fake​-news/​. Jowett, G. and O’Donnell, V. (2014). Propaganda & Persuasion. London: Sage Publications. Kalsnes, B. (2018). Fake news. Oxford Research Encyclopedia of Communication. https://​doi​ .org/​10​.1093/​acrefore/​9780190228613​.013​.809. Kalsnes, B. (2019). Falske nyheter: Løgn, desinformasjon og propaganda i den digitale offentligheten. Oslo: Cappelen Damm Akademisk. Kalsnes, B., Falasca, K., and Kammer, A. (2021). Scandinavian political journalism in a time of fake news and disinformation. In E. Skogerbo, O. Ihlen, N. Norgaard Kristensen, and L. Nord (eds.), Power, Communication, and Politics in the Nordic Countries (pp. 283–304). Gothenburg: Nordicom, University of Gothenburg. Koliska, M. and Assmann, K. (2021). Lügenpresse: The lying press and German journalists’ responses to a stigma. Journalism, 22(11), 2729–2746.

442  Handbook of digital politics

Kovach, B. and Rosenstiels, T. (2007). The Elements of Journalism: What Newspeople Should Know and the Public Should Expect. New York: Three Rivers Press. Ladd, J. M. (2012). Why Americans Hate the Media and How It Matters. Princeton, NJ: Princeton University Press. Lazer, D. M. J., Baum, M. A., Benkler, Y., Berinsky, A. J., Greenhill, K. M., Menczer, F., and Zittrain, J. L. (2018). The science of fake news. Science, 359(6380), 1094–1096. Lazer, D. M. J., Baum, M. A., Grinberg, N., Friedland, L., Joseph, K., Hobbs, W., and Mattson, C. (2017). Combating Fake News: An Agenda for Research and Action. https://​ shorensteincenter​.org/​combating​-fake​-news​-agenda​-for​-research/​. Newman, N., Fletcher, R., Kalogeropoulos, A., Levy, D. A. L., and Nielsen, R. K. (2018). Reuters Institute Digital News Report. https://​reutersinstitute​.politics​.ox​.ac​.uk/​sites/​default/​ files/​digital​-news​-report​-2018​.pdf. Nielsen, R. K., Fletcher, R., Newman, N., Brennen, S. J., and Howard, P. N. (2020). Navigating the ‘Infodemic’: How people in six countries access and rate news and information about Coronavirus. Reuters Institute for the Study of Journalism. https://​www​.politico​.eu/​wp​ -content/​uploads/​2020/​04/​Navigating​-the​-Coronavirus​-infodemic​.pdf. Pierri, F., Artoni, A., and Ceri, S. (2020). Investigating Italian disinformation spreading on Twitter in the context of 2019 European elections. PLoS ONE, 15(1), e0227821. Quandt, T., Frischlich, L., Boberg, S., and Schatto-Eckrodt, T. (2019). Fake news. In The International Encyclopedia of Journalism Studies. Hoboken, NJ: John Wiley & Sons. Ricard, J. and Medeiros, J. (2020). Using misinformation as a political weapon: COVID-19 and Bolsonaro in Brazil. Misinformation Review. Harvard Kennedy School. https://​ misinforeview​.hks​.harvard​.edu/​article/​using​-misinformation​-as​-a​-political​-weapon​-covid​ -19​-and​-bolsonaro​-in​-brazil/​. Ritchie, H. (Producer) (2016). Read all about it: The biggest fake news stories of 2016. https://​ www​.cnbc​.com/​2016/​12/​30/​read​-all​-about​-it​-the​-biggest​-fake​-news​-stories​-of​-2016​.html. Robertson, C. T., Bentele, K., Meyerson, B., Wood, A. S. A., and Salwa, J. (2021). Effects of political versus expert messaging on vaccination intentions of Trump voters. PLoS ONE, 16(9), e0257988. RSF (2017). Predators of press freedom use fake news as a censorship tool. https://​rsf​.org/​en/​ news/​predators​-press​-freedom​-use​-fake​-news​-censorship​-tool. Schudson, M. (1989). The sociology of news production. Media, Culture & Society, 11(3), 263–282. Silverman, C. (2016). Viral fake election news outperformed real news on Facebook. Buzzfeed News. https://​www​.buzzfeednews​.com/​article/​craigsilverman/​viralfake electio n-news-outperformed-real-news-on-facebook. Søe, S. O. (2019). A unified account of information, misinformation, and disinformation. Synthese, 198(6), 5929–5949. Tandoc Jr., E. C., Jenkins, J., and Craft, S. (2018a). Fake news as a critical incident in journalism. Journalism Practice, 13(6), 673–689. Tandoc Jr., E. C., Lim, Z. W., and Ling, R. (2018b). Defining ‘fake news’: A typology of scholarly definitions. Digital Journalism, 6(2), 137–153. Tsfati, Y., Strömbäck, J., Boomgaarden, H., Vliegenthart, R., Lindgren, E., and Damstra, A. (2020). Annals of the International Communication Association. Advance online publication. https://​doi​.org/​10​.1080/​23808985​.2020​.1759443. Tuchman, G. (1978). Making News: A Study in the Construction of Reality. New York: The Free Press. United Nations (2017). Joint Declaration on Freedom of Expression and ‘Fake News’, Disinformation and Propaganda. https://​www​.osce​.org/​files/​f/​documents/​6/​8/​302796​.pdf. Vosoughi, S., Roy, D., and Aral, S. (2018). The spread of true and false news online. Science, 359(6380), 1146–1151.

Fake news and digital politics  443

Waisbord, S. (2018). Truth is what happens to news: On journalism, fake news, and post truth. Journalism Studies, 19(13), 1866–1878. Walter, N., Cohen, J., Holbert, R. L., and Morag, Y. (2020). Fact-checking: A meta-analysis of what works and for whom. Political Communication, 37(3), 350–375. Wardle, C. and Derakhshan, H. (2017). Information Disorder: Toward an Interdisciplinary Framework for Research and Policy Making. Council of Europe. https://​edoc​.coe​.int/​en/​ media/​7495​-information​-disorder​-toward​-an​-interdisciplinary​-framework​-for​-research​-and​ -policy​-making​.html. Zimmermann, F. and Kohring, M. (2020). Mistrust, disinforming news, and vote choice: A panel survey on the origins and consequences of believing disinformation in the 2017 German parliamentary election. Political Communication, 37(2), 215–237.

29. Right-wing alternative news media and digital politics Kristoffer Holt

Politics is increasingly defined by organizations, groups, and individuals who are best able to blend older and newer media logics, in a hybrid system. Power is wielded by those who create, tap, and steer information flows to suit their goals and in ways that modify, enable, and disable the power of others, across and between a range of older and newer media. (Chadwick, 2017) We don’t want to talk to the mainstream media, because they are not fair, they are not balanced, they are too busy making the news as opposed to reporting what’s true. (IrnieracingNews, 2022)

INTRODUCTION In late January 2022, a group of Canadian truckers announced that they were forming a convoy, bound for Ottawa, where they would remain and protest until either Justin Trudeau resigned or the draconian Covid-19 measures were lifted (especially a vaccine pass, mandatory for cross-border truck drivers). Available narratives about what actually happened after that, diverge significantly, but what is clear is that the protest movement had a resounding impact on the political scene in Canada, and forced the ruling Liberal party, as well as the conservative opposition, to position themselves outspokenly in relation to the protests. Alternative news media played an important role in the drama, especially in terms of coverage and dissemination of events from the perspective of the truckers. From the outset, it was clear that the organizers of the convoy acted according to a successful media strategy and gained significant reach in social media, using various hashtags (for example #freedomconvoy, #truckersforfreedom etc.) and posting videos and pictures of people on bridges and roadsides with Canadian flags cheering on truckers who were passing through. The convoy organizers broadcast their own press-conferences on YouTube and Rumble, where they stated their message and took questions from a select number of reporters, notably deliberately excluding those from mainstream news media. A positive picture and enthusiastic narrative of how the freedom protests developed was disseminated through supportive hashtags on social media and the multitude of snippets of on-site footage available there. The trucker convoy, it was soon reported by optimistic supporters, formed the longest convoy of vehicles ever seen in the world during peacetime, and a new Guinness world record was announced.1 Claims about the number of participating trucks varied from a few thousands of trucks to up to fifty thousand. The movement soon spread to 444

Right-wing alternative news media and digital politics  445

other countries: UK, Belgium, Netherlands, Italy, Australia, Israel, and the USA soon had their own hashtags and new convoys were forming. The GoFundMe platform had soon raised about $10 million, and Canada’s President, Justin Trudeau, went into isolation, allegedly because he had been in contact with someone with the Omicron variant of the coronavirus. This description, however, stands in stark contrast to the account of the events as told on other channels. The Canadian Association for Security and Intelligence Studies described the convoy as “a national security threat” because the truckers’ “violent, extremist rhetoric is normalized which threatens the cohesion of Canadian society and increases the potential for escalation to kinetic violence as counter-protest efforts intensify” (West, 2022). President Trudeau, as a result of being unable to resolve the situation, and for refusing to ease Covid restrictions, ended up invoking emergency powers to be able to forcefully remove the protesters and seize their funds. He called the truckers a “fringe minority” with “unacceptable views”, and after the first day of protesting in Ottawa, commented he was “shocked” and “disgusted by the behavior displayed by some people protesting”, asserting that he would not be “intimidated by those who hurl insults and abuse at small business workers and steal food from the homeless. We won’t give in to those who fly racist flags. We won’t cave to those who engage in vandalism or dishonor the memory of our veterans” (Global News, 2022). In mainstream news media, the focus was directed at deflating the alleged number of trucks and in the crowds, the fact that a Nazi flag had been spotted and photographed among the protesters, and that the statue of Canadian icon Terry Fox had been given a sign with the words “mandate freedom” and an upside-down Canadian flag (Fung, 2022). This is an example of a narrative battle, which plays out in different parts of the media landscape, and as an event, this illustrates the importance of alternative media in contemporary digital political life. It is difficult to draw conclusions about winners and losers in this battle. While polling suggested a slight majority (54 percent) not sympathizing with the truckers, a significant minority (46 percent) said they sympathized with the truckers, even though they did not necessarily agree with them (Bricker, 2022). The poor citizens of Ottawa, having to listen to endless honking for days on end were probably not very happy. On the other hand, the fact that a substantial part of the GoFundMe dollars was seized and eventually refunded to donors – never reaching the truckers – could be noted as a win for the truckers’ opponents. Yet again, a new collection was started through the fundraising alternative GiveSendGo, quickly gathering substantial donations, and donations in Crypto-currencies were also made. While the on-site footage from the protests indicates a large turnout during weekend protests, the point is not so much, in this chapter, who was the victor in this stand-off. The truckers and their organization represent an group of people who obviously have strong convictions and demand to be heard. Elite politicians, bureaucrats, and journalists on the other hand, represent a clear opposition, and the trucker convoy shows traits of being a typical populist phenomenon, pitting the real, ordinary people against a self-serving and corrupt elite (Mudde, 2004). The interesting aspect is that the two perspectives and accounts of what was happening differ so completely, and to a point where it

446  Handbook of digital politics

is virtually impossible for most people to ever know for sure what was true at what point and to what extent. Was the Nazi-flag waver representative of the protesters – or was it a false-flag manoeuvre cunningly planned and executed by the ruling class – or just an outlier? How many trucks were actually in the convoy? Were the organizers sincerely concerned about human rights and the plight of working people hit by the restrictions, or merely using the fact that a two-year-long and taxing Covid pandemic provides an opportunity to capitalize on discontent and canalize it for other, sinister purposes? The story of the Canadian truckers highlights the new role that alternative media can play, and the political relevance it has in our digital age: Firstly, it shows that there is no longer a unified entity that can be called “the media” and that can be characterized as representative of powerholders. As Chadwick (2017) described, today’s media is characterized by “hybridity”, giving advantage to those who can master both old and new media in successful ways. The trucker story illustrates this. To be successful, it is undoubtedly still important to communicate in “old” ways to protest: to drive a truck, honk the horn, hold a sign at a rally, demonstrate and chant, speak powerfully from a stage to the supporters or, for that matter, actually gather huge crowds in physical locations. It is not, however, enough anymore. Today, these things must not only be meticulously documented through drone footage or livestreams, or selfies posted on social media, it also needs to be packaged smartly in hashtags and disseminated cunningly, in order to tap into algorithmically operated social media logic. This requires planning, coordination, and dedicated publication outlets such as alternative news media. From the authority side of this story, it is still clearly important to get the message across to the important mainstream media channels, deliver it in a media-smart and sensible way, to own the narrative and speak with authority. But it is not enough anymore. When sweeping descriptions of unacceptable views and bad behavior can be contradicted powerfully through documented peaceful protest (and there is an abundance of testimony to this available, both in more organized alternative news media outlets but also from individual social media users), and mercilessly ridiculed on social media – the narrative risks crumbling dramatically. This new reality has become pivotal for the dynamics of media and politics – and in the digital realm, right-wing actors have seen a chance to make use of the affordances offered in the digital public sphere (Mazzoleni, 2014; Mazzoleni and Vaccari, 2020). In this, alternative news outlets have become important venues, both for harnessing the engagement from individuals through social media, but also for fuelling it.

RIGHT-WING ALTERNATIVE MEDIA AND DIGITAL POLITICS At the heart of academic theorizing about media and their political role in modern democratic society, there has always been concern for the power that it is possible to wield through mass-mediated symbolic representation (Thompson, 1995). This

Right-wing alternative news media and digital politics  447

power is expressed through various processes, such as normalizing viewpoints and preferences, while ostracizing others (Ramasubramanian and Yadlin-Segal, 2017), focusing attention to specific issues while ignoring others, representing acceptability and anathematizing otherness. Ultimately, the spectrum of various expressions and exercises of media power laid bare by media scholarship so far, has to some extent always rested upon the notion of there being a few highly influential and dominant media outlets in a given context, which form the basis for an infrastructure of dissemination, that distribute news content to a mass-audience, thereby constituting a framework for the formation of a shared frame of reference among citizens. Although terminology is still debated, these are commonly referred to as “mainstream media” or alternatively as “legacy media”. The way news is told through such channels will naturally have consequences for what is generally considered common sense and self-evident, legitimately debatable, or illegitimate and deviant (Hallin, 1989). Discussion about the power of media in relation to political life has therefore been related to the concentration of ownership, distributive reach, and capacity for production among a few actors who will be centrally placed and therefore play key roles in the shaping of political discussions, information, and the conditions for participation. The Internet, and the social media, “Web 2.0” (O’Reilly, 2005) revolution added a new layer to this, since it opened up possibilities for participation and visibility in a new way, partly tearing down a long held, and well-established power-concentration among the influential media operations. Because there is power attached to media with a mass-reach, it becomes natural to take an interest in alternative voices, media outlets that provide messaging, information and deliberative space for news, opinions and debaters who represent perspectives, opinions, and identities that for some reason or other tend to be ignored, shunned, or simply not noticed by the major mainstream media outlets. Therefore, alternative media has been on the agenda of media researchers for a long time. Already in the 1970s, the notion of media outlets that go against the mainstream and challenge norms and hegemonies by disseminating counter-narratives, reporting about topics and areas that generally do not get the attention of the major news providers, was of particular interest, especially in the field of critical media studies. This line of research and theory has developed over the decades into a rich field within media studies, generating influential and important contributions to communication studies (Atton, 2015; Coyer et al., 2007; Downey and Fenton, 2003; Downing, 2003; Fraser, 1990; Kenix, 2011). While research on alternative media initially and for the most part since, has been conducted with a specific focus on representation of minorities and underrepresented groups, clearly with an emancipatory normative progressive vision about plurality and inclusion in mind, later years, especially the last decade, have seen a shift in focus where alternative media that do not easily fit into a normative conception of progressive values have gained increasing interest from the scholarly community. While much has been written about the Occupy movement, and their use of alternative media (see for example DeLuca et al., 2012; Fuchs, 2014) it is not self-evidently possible to employ the same theoretical framework and empirical lessons when ana-

448  Handbook of digital politics

lyzing the media presence of movements like PEGIDA in Germany (Holt and Haller, 2017), the nebulous alt-right phenomenon (Nagle, 2017) and as of late, the Canadian truckers for freedom. In recent years, much scholarship has been devoted to untangling this, and to building theory that facilitates a more commonly applicable framework for understanding highly important and consequential but also ideologically and culturally diverse occurrences involving alternative media (Holt et al., 2019).

ALTERNATIVE MEDIA, POLITICS AND THE DIGITAL Alternative media, according to current definitions, are to be understood primarily as alternatives to something that already exists and have enough weight, reach and status to constitute a natural point of reference and to some extent authority, within a given media system (Holt et al., 2019). Alternative media, from this “relational” perspective, are therefore construed as channels for voices and champions of ideas and perspectives that are, for whatever reason, not experienced as heard/present/ visible/represented/respected in the most powerful and wide-reaching media channels in a given societal context. In order to acknowledge the existence of an alternative, therefore, it is necessary to assume the existence of a somewhat powerful and omnipresent set of mainstream media channels which distribute news and views. It is in relation to this mainstream that the alternative becomes interesting, since it will necessarily have to be defined in terms of why, how, to what extent and with what effect it opposes or challenges the mainstream. This is also why alternative media are especially interesting from a media scholar’s perspective; to some extent, they offer avenues for studying the borderlines between what is popularly assumed to be the mainstream construal of what is normal and within the realms of the acceptable, and what is considered as being beyond the pale. In any case, media representatives and organizations who label themselves “alternative” would not do so, if they perceived themselves to be represented or to be tolerated within the realms of the general mainstream.2 The reasons for doing so can vary between different contexts (Heft et al., 2021). The political aspect of this distinction is therefore especially relevant. We are not talking here about blogs offering “alternative” or “ostracized” cupcake recipes, homebrewing tips or knitting-descriptions – although these valuable and important aspects of life can become political and subject of much controversy too. The burning issue is, of course, when a certain number of citizens in a given country or community, experience that important values, identities, beliefs, viewpoints or philosophies they adhere to and that to some extent define who they are, are being dismissed, not accepted, misrepresented or not represented at all within the channels for public discourse that matter, when they feel that the crucial discussions about how to best proceed forward disregards their existence. The BLM movement and the rhetoric surrounding protests sparked by racial injustice in the USA clearly is an example in point, as it channelled and expressed frustrations and experiences of people who feel marginalized in this way (Kilgo and Mourão, 2019). Yet again, since we are focusing

Right-wing alternative news media and digital politics  449

on right-wing alternative media in this chapter, the Canadian truckers also offer us a good example. This role of the marginalized is exactly how the movement’s representatives pitched themselves through their own media channels to the public. They were highlighting the fates of a vulnerable minority among themselves, the unvaccinated truckers, whose livelihoods were affected by undiscriminating policies. They expressed the despair of not being heard or taken seriously and were therefore pushed to means of communication that lie outside the major channels. Whether this is actually true or not, is another question. The important thing is the positioning vis-à-vis the political and media establishment. Alternative media, often in concert with and fuelled by populist sentiments, become relevant as a political force the moment such groups and interests feel forced to seek communication channels outside the mainstream, and to the extent that they manage to have an impact by doing so (Holt, 2019). The BLM movement has been highly successful by resonating not only in alternative media channels and social media but had a strikingly high impact also in mainstream media channels (Freelon et al., 2018). It remains to be seen how effective the protest of the truckers actually was in terms of political results, but one can clearly already see an impact of inspiring other, similar protests around the world, and the attraction among a number of Canadians to show up for protests. Also, more mainstream conservative media, like Fox News Network, has devoted substantial positive attention to the truckers. Thirdly, the digital aspect connects to the previous notions by means of providing new avenues of dissemination and reach for alternative news. The threshold to publicity has been lowered remarkably, and anyone with the means to produce media content can, in theory, reach a large audience and therefore also possibly have a substantial impact. While this was generally hailed as a possible and well needed democratic boost in the early days of the Internet and social media (Jenkins, 2008; Lévy and Council of Europe, 2001), participatory culture has turned out to also involve what Quandt (2018) has called “dark participation”: “Instead of positive, or at least neutral, contributions to the news-making processes, it is characterized by negative, selfish or even deeply sinister contributions” such as trolling, “piggy-backing” on journalistic reputation, and large-scale disinformation (Quandt, 2018). In any case, alternative media, through the affordance of the digital tools available, is both an avenue of participation (be it “dark” or just plain old participation), and mainly becomes politically relevant to the degree that it manages to stir things up through the digital channels.

CULTURE WARS? To properly discuss the emergence and relative impact of right-wing alternative media, it is also necessary to look beyond platforms and channels and place the phenomenon within a broader societal and cultural context. Much could be said about phraseology that acknowledges the existence of a “culture war”, the struggle for dominance between different social and identity groups (Davis, 2018; Fiorina et

450  Handbook of digital politics

al., 2005). There is, however, something crucial in this way of describing our current digital political environment in many Western democratic countries. Right-wing alternative media are not primarily engaged in the strictly political battles about how high taxes should be, or how much should be spent on hospitals, schools, and the military. It is not considered a fringe or bigoted standpoint to prefer lower taxes, for example, and such arguments can be unproblematically represented in mainstream discourse – it is well within the boundaries of what Hallin (1989) has called the “sphere of legitimate debate” and few, if any, run the risk of being de-platformed from the major social media companies because of approvingly tweeting about such matters. The core driving issues of alternative media on the right, lies, in my opinion, closer to the cultural field, touching on issues that concern identity, values, religious beliefs and tradition. This is something that becomes apparent when studying motives and aims among right-wing alternative media representatives (Holt, 2019). Andrew Breitbart, founder of Breitbart News, and highly influential among right-wing circles, described why he involved himself in right-wing alternative news: “The left made me do it! I swear! I am a reluctant cultural warrior” (Breitbart, 2011, p. 10). This quote is revealing and central for understanding right-wing alternative media today. It tells us something important about the motivations behind alternative news media on the right – they represent an oppositional force in the midst of a culture that is almost completely taken over by a progressive left/liberal paradigm. Hollywood, popular culture, the music industry, universities, the mainstream media, schools, Big Tech, and even many religious organizations, are seen as institutions that concertedly push an agenda that threatens to ostracize and delegitimize perspectives that do not fit: the usual suspects are issues like pro-life values, traditional views on marriage, sexuality and gender, immigration, multiculturalism, and religious freedom. The analysis here is that it will be more difficult to champion, for example, traditional family values or sexual morals, if you run the risk of being ridiculed by celebrities or cancelled from participation in cultural events because your views are labelled as unacceptable by important social actors (conference organizers, etc.). This is why Breitbart often talked about politics as existing downstream from culture. The dynamics of cancel culture, according to Breitbart, for long kept conservatives afraid, anxious, and overly careful, which is obviously a losing strategy. Breitbart’s logic behind right-wing alternative media was therefore to build platforms for fighting back, to be able to go on the offensive, and to be able to “tell better stories” (Wood, 2019). This analysis has been highly influential and explains much of the media-induced reinvigoration of the warfare we see today from many right-wing actors in the alternative media sphere. Coupled with the possibilities offered by the hybrid media system (Chadwick, 2017) social media and the Internet made it possible to bypass established gatekeeping thresholds, and reaching large audiences through smart publicity tactics, this partly explains the state of public discourse we are currently in (Buyens and Van Aelst, 2022; Michael, 2017).

Right-wing alternative news media and digital politics  451

DISCUSSION Alternative media should also be viewed from a historical perspective, since the dynamics described in this chapter is not entirely new. In fact, what is today considered to be mainstream publications, in many Western countries, the established legacy newspapers often emerged from a radical and oppositional ideological motivation and have a long history of portraying themselves as alternatives to the dominant powers of the time when they were launched. The Swedish tabloid Aftonbladet, launched in 1830, for example, was from the start a voice for the liberal opposition (like many other European newspapers of record which were launched in the early nineteenth century), and positioned itself against the various royalist publications, subsidized by the Swedish monarch, and claimed to represent the people in the face of tyrannical powers (Lundell, 2002; Nordmark et al., 2001). Such ventures often ran into problems, like censorship and prohibitions, and have kept this oppositional pathos throughout history. It has also served as a legitimizing journalistic mythology, closely related to the democratic function of scrutinizing and holding those in power accountable. The Washington Post’s emblematic slogan, “Democracy dies in darkness”, is a reminder of how this pathos still, to some extent, lives on as part of the identity that defines the journalistic profession throughout the Western world. Even though many such originally oppositional newspapers turned into successful and lucrative businesses, and “the press” eventually became an established institution in society, a foundation of the “public sphere” (Habermas, 1989) and of modern democratic life and culture, this idea of representing the oppressed, of speaking truth to power, has lived on. The right-wing anti-media populist tendency of division between “the people” and “the elites” (often including mainstream journalists as part of an out of touch, privileged and self-serving class together with career politicians and big business) that has become so common today (Fawzi, 2020), is an interesting turn of events that has come to define the struggle for influence through media, especially in the recent decade, along with increased, especially right-wing political activity and success (Krämer and Holtz-Bacha, 2020). Paradoxically, the language used to criticize the media from this perspective, was preceded by a surge in media criticism from the radical left, mainly since the 1970s which saw mainstream media as a tool for preserving the status quo and upholding hegemonic consensus (Chomsky, 1997). From a historical perspective, what we are seeing today is not exactly new, but the roles are different than before. Cultural and political realities change over time, and so do the people who identify as outsiders or claim to represent the marginalized. This has always been to some extent a rhetorical pose, often used by various media and political actors (and not always necessarily true). It proves to be still effective, both politically and culturally in our day. What is new in our current situation, is that the means of production and dissemination of alternative media content is more powerful than ever and is therefore bound to play an increasingly important role in shaping digital politics.

452  Handbook of digital politics

The scholarly interest in aspects of alternative media in relation to digital politics has grown substantially over the last years. Some key works that have contributed to theory and definitions and empirical work which have helped set standards and procedures for studying this phenomenon can be recommended. Firstly, the relationship between media and populism is treated thoroughly and from many perspectives in the anthology Populism and the Media edited by Krämer and Holtz-Bacha (2020). This volume contains illuminating pieces related to our current politically tense media environment, marked by the increasing presence of populist politicians and their different ways of using media, through alternative means and digitally. My book, Right-Wing Alternative Media (2019), along with the co-authored article “Key dimensions of alternative news media” (Holt et al., 2019) represent much of the theoretical foundation for the “relational approach” to the study of alternative media presented above. In addition, Thorsten Quandt’s “Dark participation” (2018), is a key article, signalling an important pathway forward when it comes to understanding and doing research in relation to audience participation in public discourse, which is a central aspect of how alternative media play a role in politics, especially in the digital context. When it comes to the alternative media audiences, Rauch (2015) has laid the groundwork for a nuanced understanding, but there is still a research gap in this area, although interesting new studies are published. One example is Schwarzenegger (2021), who illuminates motivations and experiences of alternative media users. Based on substantial interview material, he shows that there are commonalities, such as a critical attitude and anti-system attitude and a shared desire for community belonging, while also highlighting ambiguities. While there are many important contributions currently being published, more research is essential for a better understanding of alternative media and their role in politics today.

FUTURE RESEARCH The way that right-wing alternative media impacts digital politics is in and of itself an essential future arena for scholarly work. It will be a necessary one to monitor in order to explain future developments at the dynamic and ever-changing intersection of politics, media and digitalization. As stated above, the importance of the Web 2.0 for the establishment of alternative media messaging on the right can hardly be overstated. The battle for dominance on the big platforms is still ongoing, but the tendency seems to be that the more extreme examples (and some not so extreme) are migrating to alternative platforms, such as Gettr, Gab and possibly Donald Trump’s upcoming media enterprise Truth Social. It will be interesting to see if this tendency of people with different worldviews to populate separate platforms will continue and increase. This could signify a new level of infrastructurally manifested polarization and cyber-balkanization of the digital political realm, but one where opponents would be removed from the uncivility of each other (Lee et al., 2019), paving the way for a new sort of dynamic, possibly marked by infighting and further divisions or simply implosion into non-significance. Another possible development could

Right-wing alternative news media and digital politics  453

instead be a further normalization and mainstreaming of alternative media. There are already tendencies indicating that this is the case, for example in Sweden, where both the political parliamentary reality as well as the way media write about immigration have shifted drastically, rendering previously “unacceptable” perspectives in the mainstream more acceptable, and thus the immigration critical alternative media less alternative. Along with this, there are also tendencies of increased professionalization and a move towards legitimization through joining press-clubs and journalist organizations (Figenschou and Ihlebæk, 2019). The extent to which actors representing alternative media appear on mainstream news shows, for example, or are interviewed in the established press, will determine the level of alternativeness in some respects. At some point, the mainstream tent might extend to also include what was previously considered to be outside. What this does to the political condition, in terms of moving goalposts, or by de-radicalizing alternative voices, is an important development to study further. Furthermore, alternative news in this way sparks and further complicates the ever-ongoing discussion about “what is journalism and non-journalism” (Seuri and Toivanen, 2021). In relation to this, an upcoming and possibly increasingly important domain is the Web 3.0, also popularly referred to as the “metaverse” (Mystakidis, 2022). While the immersive experience hailed by Big Tech as the future of the Internet is currently still in its developmental stage, it will be of interest to see if this Metaverse will attract politically motivated alternative media actors, and to what extent this new interface proves to be fertile soil for alternative ventures and “dark participation” (Quandt, 2018). In the early versions of the metaverse (such as Second Life), the issue of radical elements using the platform to organize and proselytize has been an issue (Cole, 2012; Di Pietro and Cresci, 2021). If Meta, according to the vision laid out by Mark Zuckerberg during his “Metaverse reveal” (Meta, 2021), succeeds in becoming an essential part of society, culture and everyday life for citizens, it will also naturally become a new frontier for digital politics and alternative media.

NOTES 1. Although this claim has later been debunked. 2. Of course, it could be argued, that it is possible to use the “alternative” label in order to deceive and create a false sense of victimization. This is, of course, the case in some instances. Nevertheless, I argue that the dynamic that occurs when invoking the polarity between “alternative” and “mainstream”, reveals a de facto existence of such a distinction, in the minds of most citizens. Of course, this borderline looks different in different countries and contexts, but in any given case, there will be some sort of organizing principle in play that draws that line implicitly or explicitly.

454  Handbook of digital politics

FURTHER READING Carpentier, N., Cammaerts, B., and Bailey, O. (2008). Understanding Alternative Media. Maidenhead: Open University Press. Fuchs, C. (2010). Alternative media as critical media. European Journal of Social Theory, 13(2), 173–192. Grossman, E. (2022). Media and policy making in the digital age. Annual Review of Political Science, 25, 443–461. Holt, K. (2020). Populism and alternative media. In B. Krämer and C. Holtz-Bacha (eds.), Perspectives on Populism and the Media: Avenues for Research (pp. 201–214). Baden-Baden: Nomos. Kluge, A. and Negt, O. (2016). Public Sphere and Experience: Analysis of the Bourgeois and Proletarian Public Sphere. New York: Verso Books. Krämer, B. and Holtz-Bacha, C. (eds.) (2020). Perspectives on Populism and the Media: Avenues for Research. Baden-Baden: Nomos. Lee, F., Liang, H., and Tang, G. (2019). Online incivility, cyberbalkanization, and the dynamics of opinion polarization during and after a mass protest event. International Journal of Communication, 13, 4940–4959. Marwick, A. and Lewis, R. (2017). Media Manipulation and Disinformation Online. New York: Data & Society Research Institute. Mayerhöffer, E. and Heft, A. (2022). Between journalistic and movement logic: Disentangling referencing practices of right-wing alternative online news media. Digital Journalism, 10(8), 1409–1430. Nygaard, S. (2019). The appearance of objectivity: How immigration-critical alternative media report the news. Journalism Practice, 13(10), 1147–1163.

REFERENCES Atton, C. (ed.) (2015). The Routledge Companion to Alternative and Community Media. New York: Routledge. Breitbart, A. (2011). Righteous Indignation: Excuse Me While I Save the World. New York: Grand Central Publishing. Bricker, D. (2022). Nearly half (46%) of Canadians say they “may not agree with everything trucker convoy says or does, but ...”. Toronto: IPSOS Global Public Affairs. https://​www​ .ipsos​.com/​en​-ca/​news​-polls/​nearly​-half​-say​-they​-may​-not​-agree​-with​-trucker​-convoy. Buyens, W. and Van Aelst, P. (2022). Alternative media, alternative voices? A quantitative analysis of actor diversity in alternative and mainstream news outlets. Digital Journalism, 10(2), 337–359. Chadwick, A. (2017). The Hybrid Media System: Politics and Power. New York: Oxford University Press. Chomsky, N. (1997). What makes mainstream media mainstream? Z Magazine, 10(10), 17–23. Cole, J. (2012). Radicalisation in virtual worlds: Second Life through the eyes of an avatar. Journal of Policing, Intelligence and Counter Terrorism, 7(1), 66–79. Coyer, K., Dowmunt, T., and Fountain, A. (2007). The Alternative Media Handbook. London: Routledge. Davis, M. (2018). ‘Culture is inseparable from race’: Culture wars from Pat Buchanan to Milo Yiannopoulos. M/C Journal, 21(5). DeLuca, K. M., Lawson, S., and Sun, Y. (2012). Occupy Wall Street on the public screens of social media: The many framings of the birth of a protest movement. Communication, Culture & Critique, 5(4), 483–509.

Right-wing alternative news media and digital politics  455

Di Pietro, R. and Cresci, S. (2021). Metaverse: Security and privacy issues. Paper presented at the Conference: The Third IEEE International Conference on Trust, Privacy and Security in Intelligent Systems, and Applications (IEEE TPS’21). Downey, J. and Fenton, N. (2003). New media, counter publicity and the public sphere. New Media & Society, 5(2), 185–202. Downing, J. D. H. (2003). Audiences and readers of alternative media: The absent lure of the virtually unknown. Media, Culture & Society, 25(5), 625–645. Fawzi, N. (2020). Right-wing populist media criticism. In B. Krämer and C. Holtz-Bacha (eds.), Perspectives on Populism and the Media: Avenues for Research (pp. 39–56). Baden-Baden: Nomos. Figenschou, T. U. and Ihlebæk, K. A. (2019). Challenging journalistic authority: Media criticism in far-right alternative media. Journalism Studies, 20(9), 1221–1237. Fiorina, M. P., Abrams, S. J., and Pope, J. C. (2005). Culture War: The Myth of a Polarized America. New York: Pearson. Fraser, N. (1990). Rethinking the public sphere: A contribution to the critique of actually existing democracy. Social Text, 25/26, 56–80. Freelon, D., McIlwain, C., and Clark, M. (2018). Quantifying the power and consequences of social media protest. New Media & Society, 20(3), 990–1011. Fuchs, C. (2014). Occupy Media! The Occupy Movement and Social Media in Crisis Capitalism. Winchester: Zero Books. Fung, C. (2022). Canadian protesters face investigation after national hero statue defaced, swastikas found. Newsweek, January 31. https://​www​.newsweek​.com/​canadian​-protesters​ -face​-investigation​-after​-national​-hero​-statue​-defaced​-swastikas​-found​-1674551. Global News (2022). Trudeau says ‘fringe minority’ in trucker convoy with ‘unacceptable views’ don’t represent Canadians. https://​globalnews​.ca/​video/​8542159/​trudeau​-says​ -fringe​-minority​-in​-trucker​-convoy​-with​-unacceptable​-views​-dont​-represent​-canadians/​. Habermas, J. (1989). The Structural Transformation of the Public Sphere: An Inquiry into a Category of Bourgeois Society. Cambridge, MA: MIT Press. Hallin, D. C. (1989). The Uncensored War: The Media and Vietnam. Berkeley: University of California Press. Heft, A., Knüpfer, C., Reinhardt, S., and Mayerhöffer, E. (2021). Toward a transnational information ecology on the right? Hyperlink networking among right-wing digital news sites in Europe and the United States. The International Journal of Press/Politics, 26(2), 484–504. Holt, K. (2019). Right-Wing Alternative Media. London: Routledge. Holt, K. and Haller, A. (2017). What does ‘Lügenpresse’ mean? Expressions of media distrust on PEGIDA’s Facebook pages. Politik, 20(4), 16. https://​tidsskrift​.dk/​politik/​article/​view/​ 101534/​150610. Holt, K., Figenschou, T., and Frischlich, L. (2019). Key dimensions of alternative news media. Digital Journalism, 7(7), 860–869. IrnieracingNews (Producer) (2022). Freedom convoy – address to Canadians by Tom Marazzo. YouTube, February 10. https://​www​.youtube​.com/​watch​?v​=​mKMAOFCpVfQ. Jenkins, H. (2008). Convergence Culture: Where Old and New Media Collide. New York: New York University Press. Kenix, L. J. (2011). Alternative and Mainstream Media: The Converging Spectrum. London: Bloomsbury. Kilgo, D. and Mourão, R. R. (2019). Media effects and marginalized ideas: Relationships among media consumption and support for Black Lives Matter. International Journal of Communication, 13, 19. Krämer, B. and Holtz-Bacha, C. (eds.) (2020). Perspectives on Populism and the Media: Avenues for Research. Baden-Baden: Nomos.

456  Handbook of digital politics

Lee, F., Liang, H., and Tang, G. (2019). Online incivility, cyberbalkanization, and the dynamics of opinion polarization during and after a mass protest event. International Journal of Communication, 13, 4940–4959. Lévy, P. and Council of Europe (2001). Cyberculture. Minneapolis: University of Minnesota Press. Lundell, P. (2002). Pressen i provinsen: från medborgerliga samtal till modern opinionsbildning 1750–1850. Lund: Nordic Academic Press. Mazzoleni, G. (2014). Mediatization and political populism. In F. Esser and J. Strömbäck (eds.), Mediatization of Politics: Understanding the Transformation of Western Democracies (pp. 42–57). Basingstoke: Palgrave Macmillan. Mazzoleni, G. and Vaccari, C. (2020). 20 years of political communication scholarship: Accomplishments, changes, and challenges. Comunicazione politica, 21(1), 5–20. Meta (Producer) (2021). The Metaverse and How We’ll Build It Together – Connect 2021. https://​youtu​.be/​Uvufun6xer8. Michael, G. (2017). The rise of the alt-right and the politics of polarization in America. Skeptic, 22(2), 9–18. Mudde, C. (2004). The populist zeitgeist. Government and Opposition, 39(4), 542–563. Mystakidis, S. (2022). Metaverse. Encyclopedia, 2(1), 486–497. Nagle, A. (2017). Kill All Normies: Online Culture Wars from 4Chan and Tumblr to Trump and the Alt-Right. Alresford, UK: John Hunt Publishing. Nordmark, D., Johannesson, E., and Peterson, B. (eds.) (2001). Den Svenska Pressens Historia II (Vol. 2). Stockholm: Ekerlids förlag. O’Reilly, T. (2005). What Is Web 2.0? Design Patterns and Business Models for the Next Generation of Software. http://​oreilly​.com/​web2/​archive/​what​-is​-web​-20​.html. Quandt, T. (2018). Dark participation. Media and Communication, 6(4), 36–48. Ramasubramanian, S. and Yadlin-Segal, A. (2017). Media influence on stigma. International Encyclopedia for Media Effects. Hoboken, NJ: Wiley-Blackwell/International Communication Association. Rauch, J. (2015). Exploring the alternative–mainstream dialectic: What “alternative media” means to a hybrid audience. Communication, Culture & Critique, 8(1), 124–143. Schwarzenegger, C. (2021). Communities of darkness? Users and uses of anti-system alternative media between audience and community. Media and Communication, 9(1), 99–109. Seuri, O. and Toivanen, P. (2021). Fortifying boundaries: The “how and why” of the Finnish media and countermedia from 2014 to 2018. Journalistica, 15(1). https://​doi​.org/​10​.7146/​ journalistica​.v15i1​.125038. Thompson, J. B. (1995). The Media and Modernity: A Social Theory of the Media. Stanford, CA: Stanford University Press. West, C. (2022). Ottawa Freedom Convoy Hot Brief. https://​vtsm​.org/​wp​-content/​uploads/​ 2022/​01/​5TA​-Freedom​-Convoy​-2022​.02​.04​.pdf. Wood, M. M. (2019). On ‘telling better stories’. Cultural Studies, 33(1), 19–28.

30. Research on the political implications of political entertainment Michael A. Xenos

INTRODUCTION Although not always recognized as such, the increasing prominence of political comedy within contemporary political communication is intimately related to the fundamental forces that have given rise to digital politics. Among the first to recognize this were Richard Davis and Diana Owen, who in their 1998 book, New Media and American Politics, identified political entertainment programming (alongside the Internet and other factors) as a key part of the “new media” environment, which they saw as marked by the increasing political relevance of numerous media forms previously not explicitly associated with politics. Thus, during the 2000 US Presidential election, we witnessed not only dramatic advancements in the realm of online campaigning, but also the regular appearances of the major candidates on a variety of late-night and daytime talk show programs as well as the special “Indecision 2000” episodes of The Daily Show with Jon Stewart. But the relationship between digital and humorous politics is more than just a historical coincidence. As Bruce Williams and Michael Delli Carpini point out, albeit in conjunction with a number of other social and cultural factors, the same technological advances that have brought about more obviously digital forms of politics and political communication (increasing bandwidth and channel capacities, new logics of communication production and distribution, and so forth) have effectively “destabilized the media regime of the mid-twentieth century”, and along with it strong distinctions between serious political content and mere entertainment (2011, p. 21). As a result, discussions of “the eroding boundary between news and entertainment” (Williams and Delli Carpini, 2011, p. 21) are closely related with the rise of digital politics. Indeed, as early as 2004, scholars such as Jody Baumgartner were investigating the effects of a growing amount of humorous political content online surrounding political campaigns, and recent research has begun to examine phenomena surrounding user-generated political satire and its increasing circulation through YouTube and social media (Baumgartner, 2013; Rill and Cardiel, 2013). Thus, though it has been a feature of the political communication environment since the days of Aristophanes, political entertainment in the way that contemporary readers know it and as it is commonly explored in communication research, entered the public sphere roughly at the dawn of the digital era. In this chapter I provide an overview of research on political entertainment and political comedy, which for our purposes will include research on media effects 457

458  Handbook of digital politics

associated with various kinds of political humor, as well as so-called “soft news” programming. In doing so I will discuss the principal findings and controversies found within this literature as well as articulate a vision for how I think future research on this topic can best mature and make substantive contributions to the broader fields of communication and political science. I begin by discussing the significance of political comedy and satire in our contemporary communication environment. This will involve consideration of the objective outlines of the increasing place of political humor in political communication in recent history, as well further consideration of the relationship between political entertainment and broader features of the world of digital politics just mentioned. From there, I will move on to a summary of the major findings and controversies found in the literature on political comedy, before proceeding to a discussion of what I consider to be the principal challenges and opportunities facing those interested in conducting future research in this area. To be sure, the growth and development of scholarship on political comedy has thus far been impressive, to the point where researchers no longer need to concern themselves with the legitimacy of this established area of interest within political communication. At the same time, however, there are a number of obstacles and as yet less developed routes on the path ahead. Thankfully, we can reasonably expect a few good laughs along the way.

THE SIGNIFICANCE OF POLITICAL COMEDY The use of humor or entertainment as a way to convey political information and arguments is far from new, but notable examples prior to the 1990s are often little more than isolated points of reference. As Matthew Baum explains in his seminal work on the political relevance and impacts of entertainment media, Soft News Goes to War: Public Opinion and American Foreign Policy in the New Media Age, until the 1980s, most people learned about politics and foreign policy (a particular emphasis of Baum’s work), from newspapers and traditional television news programs (Baum, 2005). For readers with limited first-hand experience of that period, it is worth noting that such a pattern was not simply a matter of viewing preferences. Rather, prior to the 1990s media professionals and the public simply observed much sharper boundaries between entertainment content and discussion of politics and public affairs. Of course such distinctions were always artificial, and emerged from a historically specific set of cultural and economic factors (see Baym, 2009; Delli Carpini and Williams, 2001; Williams and Delli Carpini, 2011 for a more detailed discussion of this history), but in empirical terms, they translated into a world in which “politics” was something primarily encountered only within the context of serious, and distinctly unfunny, media content. In that world, for most people the “main source” of information about politics and public affairs was the nightly television news broadcast, and though scholars debated whether newspapers were more effective in terms of helping individuals learn about complex policy issues, political entertainment of

Research on the political implications of political entertainment  459

the kind we regularly encounter today was virtually absent from such discussions (see for example, Neuman et al., 1992; Robinson and Levy, 1986, 1996). Over the course of the 1990s, however, the political communication environment had begun to change. The emergence of cable and satellite television, as well as the Internet, offered more and more bandwidth for all kinds of communication, political and otherwise. This led to the availability of a host of hybrid and niche media products, and the traditional forms of news and entertainment programming became more difficult to identify as purely one or the other. A number of particularly illustrative early examples of this can again be found in Baum’s work on soft news (2002, 2003, 2005). These examples – the Persian Gulf war of 1991, the US missile strikes on Afghanistan and Sudan in 1998 (which occurred just days after President Clinton’s grand jury testimony regarding allegations of his affair with Monica Lewinsky, spawning many references to the movie Wag the Dog), and the events of the 2000 US Presidential campaigns – all brought into sharp relief the ways in which coverage of political topics such as elections and foreign policy had, in the 1990s, begun to jump the boundaries between news and entertainment erected in a previous era. Whereas previous wars and earlier discussions of national politics were encountered in only limited and designated “hard news” spaces, in the 1990s many Americans began to learn about similar events through programs like Entertainment Tonight and A Current Affair. Writing specifically about foreign policy crises but making a point equally applicable to general political content, Baum described the expansion of political coverage at that time as extending “far beyond traditional network newscasts to include a vast array of soft news programs, including television newsmagazines, ‘human drama’-oriented local newscasts, tabloid or entertainment news programs, as well as daytime and late-night talk shows” (2003, p. 94). In recent years, this expansion has only continued, to the point where programs such as The Daily Show with Trevor Noah, Last Week Tonight with John Oliver, and Full Frontal with Samantha Bee, as well as similar programs in other countries, have become commonly accepted portions of the political communication landscape and many traditional news programs have made significant efforts to inject entertainment into their previously more staid offerings in an effort to compete for viewers in an increasingly crowded media system. Perhaps unsurprisingly, a large number of citizens, particularly those under 30, followed the overflow of political coverage beyond the confines of traditional “hard news” content that began in the 1990s. Looking to research on the growth of soft news in the 1990s and early 2000s, we see that by 2002 Nielsen ratings as well as survey data supported the claim that audiences for soft news programs were “quite large, both in an absolute sense, and relative to network and cable news programming” (Baum, 2005, p. 62). Indeed, in a widely cited report from 2000 that enflamed worries about the potentially negative effects of political comedy on democratic processes, the Pew Research Center for the People and the Press released survey data results showing that 47 percent of US adults aged 18 to 29 sometimes or regularly learned about the presidential campaign from late-night talk shows and political comedy programs (2000). While later research dispelled the myth that such trends

460  Handbook of digital politics

were evidence of young people abandoning traditional news sources in favour of exclusive attention to more entertaining political fare (see Young and Tisinger, 2006), these and more recent data confirm that, in contrast to what held in the “broadcast era”, a nontrivial proportion of the public appears to be increasingly consuming a diet of political content that is strongly laced with humor (Kohut, 2012). Before considering the implications of the rise of “infotainment” programming further, however, it is important to take note of some important considerations that help place these developments in context. Though quite popular, particularly among young people, political entertainment programs have not necessarily resulted in a net expansion of the amount of people attentive to political information and communication. In fact, scholars such as Markus Prior (2007) would argue that the increasing availability of all kinds of entertainment programming (including, but by no means limited to political entertainment) creates a world in which those with little interest in politics have an easier time of avoiding public affairs content in any form, creating significant audience fragmentation based on basic levels of individual interest in politics. Thus despite findings suggesting that political entertainment is often complementary with, or even helps promote, attention to politics and public affairs in other media among the otherwise inattentive (Young and Tisinger, 2006), broader patterns still point to an overall audience for news and politics that is shrinking, rather than growing, relative to the size of audiences interested in media focused purely on entertainment. As Prior concludes in his sweeping analysis of media choice and political engagement from the mid-twentieth century to the early twenty-first century, the “flight from the news by entertainment” caused by increased levels of media choice, “is a more profound influence than the slowing of this flight through infotainment” (Prior, 2007, p. 275). This is of course not to say that the rise of political entertainment and political comedy are not significant. To the contrary, as will be discussed shortly, in some ways these considerations may even make the rise of infotainment all the more significant. It is, however, important not to lose sight of the basic outlines of the existing audiences for political entertainment, especially in terms of their place within broader currents of socio-technical change and other dynamics within contemporary political communication. Even if one takes great care not to overstate the amount of political entertainment being consumed by various sectors of the population, however, its significance as a research topic is found not so much in size of the audiences or user-bases for such content, but in the extent to which these underlying dynamics reflect and speak to a variety of research questions both timeless and contemporary. As will be discussed in more detail in the next section, research on political entertainment has helped shed light on the complex processes by which citizens gain knowledge about political figures and issues, form political opinions and attitudes, and become more actively engaged as citizens in a communication environment that has been indelibly affected by the growth of digital technologies. In some cases, such as research exploring the sense in which political comedy can serve as a counter-current to broader patterns of interest-based fragmentation just discussed, work in this area helps identify, and forces us to consider, patterns that are central to understanding democratic processes

Research on the political implications of political entertainment  461

in a communication environment defined by digital networks. In many other cases, such research simply helps provide a fresh way of looking at timeless questions concerning topics such as persuasion, political norms, and political participation. Across the board, however, because of the sense in which the rise of political entertainment is intertwined with the rise of the digital age itself, research on the implications of humorous political communication will very likely continue to grow well into the future.

MAJOR FINDINGS AND CONTROVERSIES IN SCHOLARSHIP ON POLITICAL ENTERTAINMENT Research on political comedy and political entertainment covers a wide array of possible media effects, and has given rise to significant controversy concerning the normative implications of these effects. Taking the classic typology of media effects offered by Chaffee (1980) as a guide, one may describe research in this area as focused on content that combines humor with political information or arguments (regardless of medium), as exploring effects dealing with cognitive, attitudinal, as well as behavioural outcomes, and as typically conducted at the individual level of analysis. Such effects are not always direct or entirely straightforward, but they often involve core aspects of citizen engagement (Street et al., 2011). Because of this, research in this area has, since the earliest studies reviewed here, involved a vibrant discussion of whether the rise of political entertainment should be interpreted as a boon or bane for democratic processes. In this section, I will review the major findings in this literature concerning the various kinds of effects (cognitive, attitudinal, behavioural), as well as provide a guide to debates over normative questions, such as whether we should consider someone like The Daily Show’s Jon Stewart as a saint of democracy or, as Hart and Hartelius have colourfully declared, a sinner, whose “sins against the Church of Democracy” are “so heinous that he should be branded an infidel and made to wear sackcloth and ashes for at least two years, during which time he would not be allowed to emcee the Oscars, throw out the first pitch at a Yankee’s game, or eat at the Time-Warner commissary” (Hart and Hartelius, 2007, p. 263). Studies of the potential for political comedy exposure to facilitate greater political learning make up one of the largest areas of effects research focused on political entertainment. Most research in this area draws on Baum’s notion of political entertainment as a potential “gateway” to political awareness or information that can help facilitate political learning for individuals who might not otherwise attend to political information in the news (Baum, 2003, 2005). The idea behind this concept is that by “piggybacking” political information onto humorous or entertaining messages, political comedy stimulates individuals who might otherwise remain uninterested in politics to pay attention to and acquire factual information about the topics treated in a particular piece of political entertainment or satire (Baum, 2003, 2005). Along these lines, researchers have documented this “gateway effect” with respect to political comedy with evidence from both experimental as well as survey-based studies

462  Handbook of digital politics

(e.g. Feldman and Young, 2008; Xenos and Becker, 2009). These studies focus on the extent to which political comedy can facilitate learning especially for those low in intrinsic political interest, “contributing to an equalizing effect over time” between political sophisticates and those less interested in politics and public affairs (Young and Tisinger, 2006, p. 116). Other studies have taken a similar tack but focused instead on the ways in which learning is differentially stimulated based on variables such as age (i.e. greater learning among younger individuals) or predispositions toward humor as a preferred mode of communication (which can also enhance learning from comedic content), rather than intrinsic interest in politics (e.g. Cao, 2008; Matthes, 2013). These studies contribute to an optimistic interpretation of political comedy as a positive force in helping to create an informed citizenry. Not all research on the effects of exposure to political comedy on political knowledge, however, supports such a straightforwardly optimistic interpretation. Indeed, a number of scholars have taken a more sceptical approach to these effects, typically through careful scrutiny of the kinds of political knowledge and learning that political comedy may facilitate. For example, in response to Baum’s early work on soft news, Prior argued that while such programming may be effective in getting the attention of people who might otherwise ignore politics and foreign policy topics, “this attention does not translate reliably into a learning effect” (Prior, 2003, p. 162). Pursuing a similar line of argument, Hollander (2005) makes a distinction between mere “recognition” of information as opposed to “recall of actual information”, in interpreting his findings from a study of political entertainment programming and campaign information. Perhaps the clearest articulation of reservations concerning the effects of viewing political comedy on knowledge, however, comes from work by Baek and Wojcieszak (2009) in which item response theory (IRT) is used to show that such effects are mainly found when investigating “widely known, thus relatively easy, political facts and issues” (p. 797). Thus despite the intuitive appeal and large body of empirical support for the basic argument surrounding the gateway hypothesis, there are also good reasons to temper overly strong expectations about the power of political comedy to facilitate political learning. Aside from effects on knowledge, researchers have also examined the power of political humor to affect a variety of attitudes and opinions. Research along these lines has examined the extent to which exposure to political entertainment content has effects on particular opinions about specific political figures or issues, as well as effects on more generalized attitudes such as efficacy and trust. Though a number of studies have explored both kinds of effects simultaneously, given that they involve different theoretical arguments and hold different kinds of implications for democratic processes, I will review each separately. To understand the power of political comedy to affect specific attitudes or opinions, researchers have turned to a variety of concepts borrowed from scholarship on persuasion and media effects, often from an information-processing perspective. One of the central assumptions of this approach is that “getting a joke” is fundamentally an exercise in processing information that takes effort. The classic example of this is seen in the reconciliation of punch lines in simple jokes, but in the case of more

Research on the political implications of political entertainment  463

complicated satire, such as the work of Jon Stewart and Stephen Colbert, it is clear that following the message carefully can require even more mental energy. Based on this assumption, researchers have posited that humor may increase the persuasive power of a political message by reducing one’s ability to counter-argue any political claims or opinions contained in the humorous content with which one does not already agree. This is hypothesized to occur either because “getting the joke” takes up finite cognitive energies, or because simply knowing that something is humorous decreases the motivation for effortful information processing (this is known as a “discounting effect”), but there is slightly more evidence for the latter (Nabi et al., 2007; Young, 2008). In either case, though the mechanisms may differ somewhat, the ultimate effect is for the attitude or opinion held by the hearer of the joke to move in the direction implied by the comedic message, sometimes much later, in what is referred to as a “sleeper effect” (Nabi et al., 2007, p. 49). Other research on the effects of viewing political entertainment content on specific attitudes have explored less direct routes to persuasion or opinion formation. For example, in a study of the effects of US Presidential candidate appearances on the late-night and daytime talk show circuit in 2000, Moy, Xenos, and Hess explored the extent to which attitudes toward major political figures may be indirectly affected by political entertainment programming through priming effects (Moy et al., 2006). Specifically, using survey data collected in a rolling cross-section design (the National Annenberg Election Survey), Moy and colleagues identified a significant increase in the extent to which personality factors (relative to policy or issue concerns) drove public attitudes toward George W. Bush, immediately following his personal appearance on The Late Show with David Letterman, resulting in a temporary increase in overall support for Bush among the somewhat left-leaning Late Show audience (Moy et al., 2006). Other studies have suggested that the effects of viewing political humor on attitudes and opinions may be conditioned by the political predispositions of the viewer. In these studies, political opinions held by individuals who are exposed to political comedy are indeed affected by that exposure, but these effects vary based on political partisanship or ideology such that something more complicated than a simple unidirectional persuasive effect explains changes in attitudes (Baumgartner et al., 2012; Brewer and McKnight, 2015; LaMarre et al., 2009; Xenos et al., 2011). Thus the effects of exposure to political entertainment content on attitudes and opinions can sometimes appear to have simply moved opinions in one direction or another, but in reality this is a more subtle result of complex interactions between the content of a comedic message and viewer predispositions, perceptions, and beliefs. Still more research on the effects of exposure to political comedy on attitudes has focused more on generalized attitudes related to broader democratic processes, as opposed to more distinct opinions about specific individuals or political issues. Most work in this area examines whether political comedy has positive or negative effects on attitudes of political trust and efficacy, and indeed there is certainly a healthy amount of disagreement over this question. On the one hand, scholars such as Baumgartner and Morris (2006), as well as Hart and Hartelius (2007), argue that

464  Handbook of digital politics

political comedy and satire, by creating a steady set of negative messages that often impugn the leadership abilities of major government and political figures, may be doing significant harm to attitudes believed to be fundamental to healthy democratic processes. Specifically they argue that programs such as The Daily Show serve to erode trust in the media and the electoral process, while fostering a general sense of political alienation among viewers. In a relatively rare European study in this vein, for example, Matthes and Rauchfleisch (2013) illuminated how political parody exposure can negatively impact the evaluation of political actors, dubbing the pattern of findings a “Swiss ‘Tina Fey’ effect”. In contrast, other studies have found positive relationships between regular viewership of political entertainment programming and personal political efficacy, understood as one’s sense of confidence in one’s ability to understand and engage with the world of politics, as well as general political trust (e.g. Becker, 2011; Hoffman and Young, 2011; Long et al., 2021). Those with more optimistic expectations regarding the effects of political entertainment on attitudes like efficacy and trust typically argue that political humor, particularly as practiced on programs such as The Daily Show, helps to facilitate good democratic citizenship through modelling critical thinking and cultivating a healthy skepticism among viewers. Though findings are certainly mixed in this area, there is slightly more evidence for positive effects on democratic attitudes, especially when one considers research on participatory behaviors, which will be discussed later. At the same time, however, more recent work has explored possible effects on other kinds of democratic attitudes, beyond the basics of trust and efficacy, with results that further complicate an optimistic interpretation of these effects. In a study examining the effects of exposure to political satire online, for instance, researchers found that participants assigned to view a satirical version of political content on the web were more likely to engage in partisan selective exposure and registered significantly lower levels of political tolerance, as compared to participants assigned to view comparable non-comic materials or (control condition) no materials at all (Stroud and Muddiman, 2013). Thus while the overall pattern of findings has been somewhat mixed, research has documented a number of connections between viewing humorous political content and general attitudes that individuals may hold about political participation, major democratic institutions, and their fellow citizens. Finally, in addition to studying the effects of political comedy on political knowledge and attitudes, researchers have also investigated whether viewing humorous political content has any appreciable effect on concrete political behaviors. In particular, research in this area has focused on various forms of political engagement, including standard forms of electoral participation (such as voting, and involvement with political campaigns), as well as more informal acts of political engagement such as political discussion (Baumgartner and Lockerbee, 2018; Becker and Bode, 2018; Cao and Brewer, 2008; Hoffman and Young, 2011; Moy et al., 2005). As alluded to earlier, research on political comedy and political engagement or participation tends to support the arguments of those who view the rise of political entertainment as beneficial to democratic processes. That is, contrary to concerns such as

Research on the political implications of political entertainment  465

those expressed by Baumgartner and Morris, who cautioned that political comedy viewership “may dampen participation among an already cynical audience … by contributing to a sense of alienation from the political process” (2006, pp. 362–363), work in this area has generally documented positive relationships between political comedy viewing and political engagement (Cao and Brewer, 2008; Hoffman and Young, 2011; Moy et al., 2005). This work suggests that independent of other kinds of effects, one positive effect of political entertainment content is that (similar to traditional news programming) it may help increase individuals feelings of political efficacy, which in turn facilitates greater levels of political participation (Hoffman and Young, 2011). Though arguably the smallest of the three different areas of effects study in the political comedy literature, research on the effects of political entertainment viewership on political engagement is thus in many ways also the most consistent, with a clear set of relatively uniform findings. As one can easily see from the preceding discussion, debates over whether the rise of political entertainment and political comedy is beneficial or detrimental to healthy democratic processes are a common theme encountered in studies spanning the full range of political humor media effects reviewed here. Much like broader debates over the normative implications of the rise of the Internet, conflicting interpretations of what political humor may mean for democracy are often represented as a clash between optimists and pessimists. In the case of the pessimists, whether the roots of critique extend back to the work of Postman (1986), or Patterson (2000), in many ways the basic concern is that political entertainment may detract from the kind of detached and rational approach to citizenship envisioned by progressive era reformers. Along these lines, pessimists worry that political entertainment may draw citizens away from more serious “hard news” and information, while promoting a relatively thin engagement with politics and cynical (or distrustful) attitudes toward major democratic institutions. For their part, optimists tend to focus on positive trends found within the empirical literature (such as “gateway effect” patterns for political knowledge and positive effects on political engagement), while also arguing that in many ways cynicism may be an appropriate or even beneficial ethic in our contemporary political and media systems. As Lance Bennett argued in a “mock trial” of Jon Stewart held at the 2006 meetings of the US-based National Communication Association (the transcripts of which were later reprinted in Critical Studies in Media Communication), “when the prevailing tone of public life is cynical, the best defense and response may be a probing and illuminating form of cynicism” such as that demonstrated and promoted by figures like Jon Stewart (Bennett, 2007, p. 282). As is also true in the case of the wider literature on digital media and digital politics, the frame of democratic optimism versus pessimism is a useful heuristic tool, particularly for introducing new students or researchers to the literature, but as will be discussed in the conclusion of this chapter, a careful accounting of available findings often reveals its limitations. Such limitations notwithstanding, however, this frame forms an important point of orientation within the literature on political entertainment media.

466  Handbook of digital politics

TAKING STOCK AND LOOKING AHEAD: COMMENTARY AND SUGGESTIONS FOR FUTURE RESEARCH Without question, the current wave of literature on political entertainment and political comedy has made significant strides since its beginnings in the 1990s. In particular, this literature has seen progress both in terms of the quantity and quality of empirical studies, as well as the increasing sophistication of the theoretical frameworks used to understand the core dynamics involved in relationships between exposure to political comedy and relevant outcome variables. In a steadily growing body of studies, researchers interested in making sense of the potential political implications of political humor as a broad class of communication forms have drawn on major theoretical frameworks from a variety of social science disciplines, and carefully honed the application of these frameworks to a number of specific questions about the effects of exposure to political entertainment. These developments have transformed what was originally a small group of relatively exploratory investigations into a literature in its own right with increasing potential in terms of being able to offer original insights back to areas of scholarship to which its early studies were indebted. This question of what research on political entertainment may ultimately contribute to broader literatures in communication and political science is central to my own particular perspective on this area of research, and strongly informs my assessment of promising paths for future research. To be clear, extant research on political entertainment has certainly already made significant contributions to these literatures simply by virtue of illuminating important recent developments within our communication environments involving the increasing prevalence of humor and satire in political communication. But given the connections between political entertainment and broader tides of socio-technical change discussed earlier, I believe research on political humor has the potential to offer much more than just an understanding of what happens, all else equal, when we inject humor into a particular piece of political communication. Studies examining the unique interplay between political comedy and social media – such as when key figures like Donald Trump turn to Twitter to directly respond to satirical portrayals – provide a useful example here (Becker, 2018, 2020). Indeed, by virtue of its intimate relationship with underlying tectonic shifts in mediated communication, I think there is a reasonable potential for research on political entertainment to help provide useful insights into the wider world of contemporary political communication from which it has emerged, and that this is a worthy goal for future research in this area. One particularly promising avenue for future work in this area that could help to bring about more of a macro-level discussion is comparative research that explores political entertainment and its effects in a variety of countries and media systems. Like much work within political communication, most of the research reviewed above is focused on the US. But readers familiar with media offerings outside of the US would certainly agree that there are both similarities and differences in terms of the form and function of political entertainment in such countries. For example,

Research on the political implications of political entertainment  467

the international popularity of The Daily Show appears to have inspired a number of programs in other countries that adopt a very similar format and structure. Germans are able to enjoy humorous and irreverent discussions of public affairs topics through Heute-Show, while Pakistanis can turn to The Real News for their own version of satirical news. Though a number of relatively recent studies have begun to appropriately expand the scope of this work to non-US shows and audiences, such examples are still relatively rare (e.g. Boukes, 2019; Matthes and Rauchfleisch, 2013) and truly comparative studies rarer still (Xenos et al., 2018). Combined with insights from previous research in comparative political communication on features of the media systems in different countries (see for example, Esser and Pfetsch, 2004; Hallin and Mancini, 2004), as well as contemporary research on the ways in which profound changes in media systems are being experienced around the world (Bennett, 2000), additional research on how political entertainment impacts individual viewers and broader political dynamics could yield a much more comprehensive understanding of the macro-level implications of political entertainment as a new media form. A related priority for future research on political entertainment and political comedy would be to increase efforts at what has been called “systematic normative assessment” in the broader political communication literature (e.g. Rinke et al., 2013). Going beyond relatively surface level connections between particular processes or outcome variables and seemingly generic concepts drawn from broad gestures to “democratic theory”, systematic normative assessment involves a careful and thorough review of how certain empirical findings may be interpreted differently according to a number of distinct (and plural) democratic theories (Althaus, 2012). Such an approach would enable scholars of political comedy to not only establish the significance of particular criterion variables, such as political knowledge or participation, but to actually arrive at broader, on balance, discussions of how the rise of political entertainment may be affecting political communication as a whole. Indeed, an explicit turn toward more systematic normative assessment in research on political comedy could facilitate, for example, discussions of the extent to which gains in knowledge and participation may or may not be offset by increases in selective exposure and decreases in careful argument scrutiny and political tolerance. In doing so, scholars of political entertainment could find a useful framework for helping existing back-and-forth conversations between “optimists” and “pessimists” to move forward in a more productive direction. In conclusion, the overall assessment of the literature on political entertainment offered here is that while it is important to understand the historical development of work in this area since the 1990s, there is still a great deal of potential for further growth and development of this literature in the future. Thus far, progress has been primarily concentrated on the development of increasingly sophisticated and effective explanations for particular kinds of effects at the individual level. In other words, research in this area has moved from simply treating humorous political content as “just another input variable in a media effects equation” (Young, 2008, p. 133), to developing a mature body of theory-driven studies devoted to exploration of the unique properties of political comedy and the implications of these for

468  Handbook of digital politics

political knowledge, persuasion and attitude change, as well as political engagement. To be sure, work of this kind is important and there are certainly good reasons to further explore and refine scholarship focused on these kinds of effects. On its own, however, progress in this direction still does not provide direct insights into the kind of macro-level questions encountered as one moves beyond the level of the individual media user. In my view, by pursuing such questions research in this area stands the best chance of further expanding its reach beyond the confines of a particular set of media forms, and becoming more integrated with broader conversations in political communication. Though such an approach may seem somewhat daunting, I think that reflection on the connections between political humor and digital politics in general, as well as progress on the specific research priorities articulated above, can both contribute to further advancement in this exciting (as well as entertaining) area of research.

FURTHER READING Baum, M. A. (2005). Soft News Goes to War: Public Opinion and American Foreign Policy in the New Media Age. Princeton, NJ: Princeton University Press. Becker, A. B. (2013). What about those interviews? The impact of exposure to political comedy and cable news on factual recall and anticipated political expression. International Journal of Public Opinion Research, 25(3), 344–356. Becker, A. B. and Bode, L. (2018). Satire as a source for learning? The differential impact of news versus satire exposure on net neutrality knowledge gain. Information, Communication & Society, 21(4), 612–625. Davis, R. and Owen, D. (1998). New Media and American Politics. New York: Oxford University Press. Feldman, L. and Borum Chattoo, C. (2019). Comedy as a route to social change: The effects of satire and news on persuasion about Syrian refugees. Mass Communication and Society, 22(3), 277–300. Long, J. A., Jeong, M. S., and Lavis, S. M. (2021). Political comedy as a gateway to news use, internal efficacy, and participation: A longitudinal mediation analysis. Human Communication Research, 47(2), 166–191. Matthes, J. and Rauchfleisch, A. (2013). The Swiss “Tina Fey effect”: The content of late-night political humor and the negative effects of political parody on the evaluation of politicians. Communication Quarterly, 61(5), 596–614. Williams, B. A. and Carpini, M. X. D. (2011). After Broadcast News: Media Regimes, Democracy, and the New Information Environment. New York: Cambridge University Press. Young, D. G. (2008). The privileged role of the late-night joke: Exploring humor’s role in disrupting argument scrutiny. Media Psychology, 11(1), 119–142. Young, D. (2012). Laughter, learning, or enlightenment? Viewing and avoidance motivations behind the Daily Show and the Colbert Report. Journal of Broadcasting & Electronic Media, 57(2), 153–169. Zoonen, V. L. (2004). Entertaining the Citizen: When Politics and Popular Culture Converge. Lanham, MD: Rowman & Littlefield Publishers.

Research on the political implications of political entertainment  469

REFERENCES Althaus, S. (2012). What’s good and bad in political communication research: Normative standards for evaluating media and citizen performance. In H. Semetko and M. Scammell (eds.), The SAGE Handbook of Political Communication. London: Sage Publications, pp. 97–112. Baek, Y. M. and Wojcieszak, M. E. (2009). Don’t expect too much! Learning from late-night comedy and knowledge item difficulty. Communication Research, 36(6), 783–809. Baum, M. A. (2002). Sex, lies, and war: How soft news brings foreign policy to the inattentive public. American Political Science Review, 96(1), 91–109. Baum, M. A. (2003). Soft news and political knowledge: Evidence of absence or absence of evidence? Political Communication, 20(2), 173–190. Baum, M. A. (2005). Soft News Goes to War: Public Opinion and American Foreign Policy in the New Media Age. Princeton, NJ: Princeton University Press. Baumgartner, J. C. (2013). Internet political ads in 2012: Can humor mitigate unintended effects of negative campaigning? Social Science Computer Review, 31(1), 601–613. Baumgartner, J. C. and Lockerbie, B. (2018). Maybe it is more than a joke: Satire, mobilization, and political participation. Social Science Quarterly, 99(3), 1060–1074. Baumgartner, J. and Morris, J. (2006). The Daily Show effect: Candidate evaluations, efficacy, and American youth. American Politics Research, 34(3), 341–367. Baumgartner, J. C., Morris, J. S., and Walth, N. L. (2012). The Fey effect: Young adults, political humor, and perceptions of Sarah Palin in the 2008 presidential election campaign. Public Opinion Quarterly, 76(1), 95–104. Baym, G. (2009). From Cronkite to Colbert: The Evolution of Broadcast News. New York: Oxford University Press. Becker, A. B. (2011). Political humor as democratic relief? The effects of exposure to comedy and straight news on trust and efficacy. Atlantic Journal of Communication, 19(5), 235–250. Becker, A. B. (2018). Live from New York, It’s Trump on Twitter! The effect of engaging with Saturday Night Live on perceptions of authenticity and the salience of trait ratings. International Journal of Communication, 12, 22. Becker, A. B. (2020). Trump trumps Baldwin? How Trump’s tweets transform SNL into Trump’s strategic advantage. Journal of Political Marketing, 19(4), 386–404. Becker, A. B. and Bode, L. (2018). Satire as a source for learning? The differential impact of news versus satire exposure on net neutrality knowledge gain. Information, Communication & Society, 21(4), 612–625. Bennett, W. L. (2000). Introduction: Communication and civic engagement in comparative perspective. Political Communication, 17(4), 307–312. Bennett, W. L. (2007). Relief in hard times: A defense of Jon Stewart’s comedy in an age of cynicism. Critical Studies in Media Communication, 24(3), 278–283. Boukes, M. (2019). Agenda-setting with satire: How political satire increased TTIP’s saliency on the public, media, and political agenda. Political Communication, 36(3), 426–451. Brewer, P. R. and McKnight, J. (2015). Climate as comedy. Science Communication, 37(5), 635–657. Cao, X. (2008). Political comedy shows and knowledge about primary campaigns: The moderating effects of age and education. Mass Communication and Society, 11(1), 43–61. Cao, X. and Brewer, P. (2008). Political comedy shows and public participation in politics. International Journal of Public Opinion Research, 20(1), 90–99. Chaffee, S. H. (1980). Mass media effects: New research perspectives. In C. G. Wilhoit and H. de Bock (eds.), Mass Communication Review Yearbook, Volume 1. Beverly Hills, CA: Sage, pp. 77–108. Davis, R. and Owen, D. (1998). New Media and American Politics. New York: Oxford University Press.

470  Handbook of digital politics

Delli Carpini, M. X. and Williams, B. A. (2001). Let us infotain you: Politics in the new media environment. In W. L. Bennett and R. Entman (eds.), Mediated Politics: Communication in the Future of Democracy. New York: Cambridge University Press, pp. 160–181. Esser, F. and Pfetsch, B. (eds.) (2004). Comparing Media Systems: Theories, Cases, and Challenges. New York: Cambridge University Press. Feldman, L. and Young, D. G. (2008). Late-night comedy as a gateway to traditional news: An analysis of time trends in news attention among late-night comedy viewers during the 2004 presidential primaries. Political Communication, 25(4), 401–422. Hallin, D. C. and Mancini, P. (2004). Comparing Media Systems: Three Models of Media and Politics. New York: Cambridge University Press. Hart, R. P. and Hartelius, E. J. (2007). The political sins of Jon Stewart. Critical Studies in Media Communication, 24(3), 263–272. Hoffman, L. H. and Young, D. G. (2011). Satire, punch lines, and the nightly news: Untangling media effects on political participation. Communication Research Reports, 28(2), 159–168. Hollander, B. A. (2005). Late-night learning: Do entertainment programs increase political campaign knowledge for young viewers? Journal of Broadcasting Electronic Media, 49(4), 402–415. Kohut, A. (2012). Cable leads the pack as a campaign news source. Pew Center for the People and the Press. http://​www​.people​-press​.org/​files/​legacy​-pdf/​2012​%20Communicating​ %20Release​.pdf. LaMarre, H. L., Landreville, K. D., and Beam, M. A. (2009). The irony of satire: Political ideology and the motivation to see what you want to see in The Colbert Report. The International Journal of Press/Politics, 14(2), 212–231. Long, J. A., Jeong, M. S., and Lavis, S. M. (2021). Political comedy as a gateway to news use, internal efficacy, and participation: A longitudinal mediation analysis. Human Communication Research, 47(2), 166–191. Matthes, J. (2013). Elaboration or distraction? Knowledge acquisition from thematically related and unrelated humor in political speeches. International Journal of Public Opinion Research, 25(3), 291–302. Matthes, J. and Rauchfleisch, A. (2013). The Swiss “Tina Fey effect”: The content of late-night political humor and the negative effects of political parody on the evaluation of politicians. Communication Quarterly, 61(5), 596–614. Moy, P., Xenos, M., and Hess, V. K. (2005). Communication and citizenship: Mapping the political effects of infotainment. Mass Communication and Society, 8(2), 111–131. Moy, P., Xenos, M., and Hess, V. (2006). Priming effects of late-night comedy. International Journal of Public Opinion Research, 18(2), 198–210. Nabi, R. L., Moyer-Gusé, E., and Byrne, S. (2007). All joking aside: A serious investigation into the persuasive effect of funny social issue messages. Communication Monographs, 74(1), 29–54. Neuman, W. R., Just, M. R., and Crigler, A. N. (1992). Common Knowledge: News and the Construction of Political Meaning. Chicago, IL: University of Chicago Press. Patterson, T. E. (2000). Doing Well and Doing Good: How Soft News and Critical Journalism Are Shrinking the News Audience and Weakening Democracy – and What News Outlets Can Do About It. Research Report. Cambridge, MA: Joan Shorenstein Center on the Press, Politics, and Public Policy. http://​www​.uky​.edu/​AS/​PoliSci/​Peffley/​pdf/​47​5Patterson​ Softnews(1​-11​-01)​.pdf. Postman, N. (1986). Amusing Ourselves to Death: Public Discourse in the Age of Show Business. New York: Penguin. Prior, M. (2003). Any good news in soft news? The impact of soft news preference on political knowledge. Political Communication, 20(2), 149–171. Prior, M. (2007). Post-Broadcast Democracy: How Media Choice Increases Inequality in Political Involvement and Polarizes Elections. New York: Cambridge University Press.

Research on the political implications of political entertainment  471

Rill, L. A. and Cardiel, C. L. B. (2013). Funny, ha-ha: The impact of user-generated political satire on political attitudes. American Behavioral Scientist, 57(12), 1738–1756. Rinke, E. M., Wessler, H., Löb, C., and Weinmann, C. (2013). Deliberative qualities of generic news frames: Assessing the democratic value of strategic game and contestation framing in election campaign coverage. Political Communication, 30(3), 474–494. Robinson, J. P. and Levy, M. R. (1986). The Main Source: Learning from Television News. London: Sage Publications. Robinson, J. P. and Levy, M. R. (1996). News media use and the informed public: A 1990s update. Journal of Communication, 46(2), 129–135. Street, J., Inthorn, S., and Scott, M. (2011). Playing at politics? Popular culture as political engagement. Parliamentary Affairs, 65(2), 338–358. Stroud, N. J. and Muddiman, A. (2013). Selective exposure, tolerance, and satirical news. International Journal of Public Opinion Research, 25(3), 271–290. Williams, B. A. and Carpini, M. X. D. (2011). After Broadcast News: Media Regimes, Democracy, and the New Information Environment. New York: Cambridge University Press. Xenos, M. and Becker, A. (2009). Moments of Zen: Effects of The Daily Show on information seeking and political learning. Political Communication, 26(3), 317–332. Xenos, M. A., Moy, P., and Becker, A. B. (2011). Making sense of The Daily Show: Understanding the role of partisan heuristics in political comedy effects. In A. Amarasingam (ed.), The Stewart/Colbert Effect: Essays on the Real Impacts of Fake News. Jefferson, NC: McFarland, pp. 47–62. Xenos, M. A., Moy, P., Mazzoleni, G., and Mueller-Herbst, J. (2018). Political entertainment in comparative perspective: Exploring the applicability of the gateway hypothesis across media systems. In J. C. Baumgartner and A. B. Becker (eds.), Political Humor in a Changing Media Landscape: A New Generation of Research. Lanham, MD: Lexington Books, pp. 185–206. Young, D. G. (2008). The privileged role of the late-night joke: Exploring humor’s role in disrupting argument scrutiny. Media Psychology, 11(1), 119–142. Young, D. G. and Tisinger, R. M. (2006). Dispelling late-night myths: News consumption among late-night comedy viewers and the predictors of exposure to various late-night shows. The Harvard International Journal of Press/Politics, 11(3), 113–134.

Index

discourse architecture and 228 filter bubbles and lower levels of 98 online political discourse 66–9 Agenda 2063 (AUC and OECD) 22, 23 agenda-setting 110, 144, 188, 218, 328–30, 421 Agur, C. 50 Ahn Hee-jung see Kim-Ahn case Al Jazeera 314 Alexander, J.C. 78, 372, 373 algorithms 8, 210–23, 243–4 and accountability 222–3 bias 49, 210–12, 264 choice-reinforcing 97 chronology, algorithmic filtering 238 content moderation 261, 263 decision-making 210, 211, 216, 221 definitions 211–12 exposure to political content 94, 95 influence 231, 246 mediation of events 204, 207 news consumption 172 overriding 238 political participation 237 power 216–21, 347 ranking function 249 transforming data 395–7 understanding of 211 visibility of posts 346, 347 Allan, S. 38 Alonso-Muñoz, L. 380, 381 Alterman, J.B. 50 alternative media 272 see also right-wing alternative media alternative politics 77, 79, 81 Altschull, J.H. 419, 426 Alvares, C. 88 Amado, A. 376, 379, 380, 382 ambivalence 5–6, 8, 9, 12–13 Anderson, C.W. 389 Animal Crossing: New Horizons (ACNH) 174, 175 Anomalous Wave 317 anonymity 49, 53, 148, 174, 185, 226, 229, 239, 315, 363, 373 anti-media populism 379, 451

accidental exposure, to political news and disagreement 95–6, 104 accountability algorithmic 222–3 data openness 388–9, 392–3, 397 democratic 272–84 factual 432 social media and 265–6 Açik Radyo 318 actor network theory (ANT) 201, 216, 300 Adonis, Lord 332 Advertising Standards Authority 403, 410–11 aesthetic politics 32–3, 36, 37–8, 41 affect affective publics 67 dimensions of politics 32 affective publics 67 affinity groups, and youth engagement 182–4 affordances 97, 103, 104, 226, 239 creative adaption 82 digital protest 8, 339–42, 346, 348–50 exposure to political content 92–7, 103, 104 political talk 110–13 proliferation of 234 user behaviour 96 user–technology relationship 226, 228–9 youth engagement 185 Africa conflict and social media 260–64, 266 de-Westernization 16–26 fake news 435–7 journalism in 421, 423 role of social media 53 Aftonbladet 451 Aganaktismenoi movement 230–31 age, and social media use 163 agency algorithmic 216–17, 219 civic 87, 88 data-driven events 206 digital architectures and 229–31 digital protest and relinquishing of 346 472

Index  473

anti-road pricing petition 331–2 Applebaum, A. 60 appointment television 161–2 apps, and e-petition research 335 Arab Spring (2011) 50, 158, 199, 290, 291, 300, 342, 356 Arafa, M. 50 Ardizzoni, M. 318, 319 Arguedas, R. 97, 98, 214, 406, 408 argument in online discussion forum 361 use of humour to convey 458 argument visualization (AV) 146, 147 Armstrong, C. 50 artificial intelligence (AI) 242–54, 269 applications 244–8 bias 244 content moderation 261 and deepfakes 11 human involvement 244 introduction to 242–4 military applications 246, 248 narrow systems 243 in politics 244–5, 253–4, 390 logistical assistance 245–6, 249–50 see also machine learning (ML); machine translation (MT) Asmolov, G. 48 assistive technology 184 asynchronous deliberation 145 Atintande, M. 21 Atkinson, P. 2 attention economy 8, 112, 219 attitude-congruent sources and content 93 attitudes, effects of humour on 462–4 audience research and audiences alternative media 452 as sites of mediation 374 online attention and policy change 112 role of social media in authoritarian regimes 52–3 soft news programs 459 symbolic politics 33, 34 Australia 65, 329, 330, 445 authentication, online deliberative space design 148 authenticity construction of political 9–13 populist performance of 10–11, 378–9 authoritarian regimes in Africa 17, 18 diasporic digital politics 64–5

fake news 436, 437 lifestyle and consumerist politics 67–9 reaction to digital activism 19 social media in 45–53 use of Internet as tool for control 83, 274 see also individual regimes authoritarianism 16, 65, 422 auto pilot (E-Liberate) 147 autocracy 45 automation 406, 412 autonomy 402, 406, 412, 419, 420, 424 autoregressive integrated moving-average (ARIMA) model 129 Avaaz 329 Aviles, L.A. 391 Ayers, M. 318 Azerbaijan 47, 51 Baek, Y.M. 462 Baidu 49 Bail, C. 238 Bakshy, E. 94, 99 Barassi, V. 315, 318 Barbas, Á. 315 Barberá, P. 100 Barnhurst, K.G. 33 Barnidge, M. 100, 160 Barocas, S. 405, 406 Barranquero, A. 315, 316, 319 Bast, J. 36 Battle of Seattle (1999) 296 Baum, M.A. 458, 459, 461, 462 Baumgartner, J.C. 457, 463–5 Be water 49 Beck, U. 181 behaviour affordances and user 96 algorithms and 219–21, 246 effects of humour on 465–6 online deliberative 140–41 technical protocols and 228 Being Digital 3 Belarus 46, 47, 51, 83, 264, 357 Benkler, Y. 64, 83, 299, 436, 439 Bennett, W.L. 12, 49, 63–8, 77, 113, 170, 181, 201, 288, 291, 294, 302–5, 307, 340, 343–5, 465, 467 Bernal, V. 65 Better Cancer Care report (2011) 332 Biden, J. 105, 174, 307

474  Handbook of digital politics

big data 112, 204, 220, 246, 253, 254, 335, 403 Billig, M. 391 Bimber, B. 213, 291, 296, 298, 299, 303 Birrell, D. 333 Black Flag initiative 359 Black Lives Matter (BLM) 39, 112, 126, 168, 176, 186, 188, 214, 278, 340, 341, 344, 345, 349, 448, 449 Black youth, political engagement 182 black-box society 217, 237 BlackOutTuesday 339, 349 Blair, T. 331 Blank, G. 98, 281 Bleiker, R. 32, 40 bloggers/blogging 83, 275, 277, 280, 282, 283, 304, 358–9, 423 Blumenau, J. 333 Blumler, J.G. 9, 61, 138, 146 Bochel, C. 327, 328, 334 Bode, L. 96, 104, 113, 130, 172, 183, 185, 213, 464 Bodrunova, S. 45, 50, 51, 423 Bolsonaro, J. 10–11, 36, 375, 378, 379, 436 Bolsover, G. 48 Bonini, T. 318 Borra, E. 314 bots 69, 148, 164, 203, 232, 238, 248, 272, 389 Bourdieu, P. 425 Bozdag, Ç. 51 Bradshaw, S. 64 branded political consumerism 68 Breitbart, A. 450 Breuer, A. 49, 50 Brexit 158, 237, 374, 377, 403, 409, 410, 437 bridging networks 293, 297, 298, 303 broadcast feeds 229, 236–8 broadcast licensing obligations 402 brokered coalitions/ networks 299, 301, 303 Bruns, A. 97, 215, 216 BTS Army 175 Bucher, T. 94, 220, 234, 372, 390, 397 Bucy, E.P. 34, 39, 40, 114, 371 Burke, E. 273 business organizations, and Fifth Estate 281–2 cable television 92 Caetano, R.D. 36

Cambridge Analytica 162, 401, 403, 405, 406 campaigning e-petitions 331 see also election communication Canadian Association for Security and Intelligence Studies 445 cancel culture 450 capitalism, global 68, 69 carnivalesque 38–9, 294, 308 Carr, N. 84 Castells, M. 61, 66, 67, 69, 70, 83, 170, 218, 273, 277, 291, 295, 305, 424, 425 censorship 45, 65, 418, 421, 424 circumvention/bypassing 7, 70 during conflict 260, 264, 266 of the Fifth Estate 282 government-led moderation perceived as 328 meta-information censorship 49 use of social media for 46, 48, 49 Central Africa 22–3, 26 Ceron, A. 161 Cesarino, L. 375, 379 CGTN 61 Chadwick, A. 277, 304, 411, 446 Chaffee, S.H. 461 Change.org 327, 329–30, 334, 335 character limits 144, 228–31, 236 chatbots 245 China consumerist politics 68 control of social media 8, 46–9, 282 crackdown on political discussion 59–60, 63, 70 crackdown on protests and dissidents 49 diasporic digital politics 64–5 global digital activism 67 Internet 70, 83 journalism 419–26 misinformation/disinformation 61, 63, 64, 436 Christensen, C. 324 citizen engagement with political information 93 two-screening practices and offline 159 under authoritarianism 51 see also Internet, as a civic space; political engagement citizen-consumers 68, 69 citizens dogmatic and inflexible 137, 148

Index  475

global digital 66–71 under-informed, unconfident 137–8, 149 citizenship e-petitioning 331 global 69 online diasporas and 65 political humour and 464 young people and 181 see also global citizenship; networked citizenship civic cultures 47, 88 civic education 187–8 civic identity 87, 169, 181 civil society 78–80 Clark, S. 322 Clark, T. 110 clicktivism 51, 86, 159, 186, 279, 306, 327 cloud protesting 344 Clubhouse 59 co-optation 45–9, 65, 158 co-production 197, 199, 200, 281, 299, 346, 373 coercion 65, 218–20, 296, 427 cognitive overload 93 Cohen, J. 200 Coleman, S. 9, 150, 360 collaboration and communicative power 277 in presentation of media events 198 in research 25–6, 40–41 research on 283 collaborative network organizations (CNOs) 281–2 collective action 143, 287, 291, 295–301 social media 201, 341–5 communicative activism 50 collective action frames 289, 293–5, 297, 302 collective action networks 301–5 collective identities 38, 87, 113, 287, 289, 295, 297, 317–18, 344, 346, 347, 348 collective representations 372–4 commercialization, and journalism in restrictive settings 422, 427 communal family system 24 communication in action networks 287, 288–9 in conflicts 261 decentralized 424 fragmentation of democratic 406 pluri-directional 24

see also political communication; shutdowns communication infrastructure, Africa 23 communitarian mode of discussion 141, 227 community, and civic engagement 183 comparative research 53, 113–31, 467 computational propaganda 48, 64, 389–90, 395, 412 computational social science 111, 113, 130–31 computer vision 114, 246 conflicts, and social media 259–68 content moderation see content moderation control, oversight and intervention 264–7 Connect with Bernie app 173 connective action 12, 67, 220, 287, 295–301, 340, 344 networks 297, 301–5, 344 personalized communication 291–5 and sharing 306 under authoritarianism 49 connectivity, network structure 235–6 Conroy-Krutz, J. 24 conspiracy theories 100, 128, 130, 432 consumerist politics 67–9 content convergence 82 of discourse, technical protocols and 229 filtering 47, 85, 95 prioritization 347 visibility 172, 346–7 content moderation 7, 8 during conflict 260–62 online deliberation 146 political discussion 227 transparency of 440 contrarian clubs 96, 101, 103 control 7, 45–9, 53, 83, 264–7, 274, 280, 282 conventional organizations, and protests 288, 292, 301, 302, 303 cooperation 22 corporate social media, platform appropriation 319–20 Correa, R. 380 corruption 19 costs, of collective action 343 Cotton, R.D. 332, 333 Couldry, N. 19, 198–9 Courtois, C. 159

476  Handbook of digital politics

Covid-19 pandemic misinformation and disinformation 436–7, 439 political communication 279, 388–9, 392–5 and social media 170 and two-screen viewing 156–8 visual representations 40 Crawford, K. 251 creative ambivalence 12–13 criminalization, hate speech and disinformation 263, 268, 437–8 cross-platform analysis 128–9 cross-platform integration 117 crowd-enabled networks 301, 302, 304 cultural engagement, Gen Z 174–6 culture and online political expression 128–9 and Russian internet 358 and two-screen viewing 159–60 see also cancel culture; civic cultures; participatory cultures; popular culture culture studies, visual 33–4 culture wars, alternative media 449–50 cyber-optimism 5, 6, 13 cyber-pessimism 5, 6, 13 cyberwarfare 248 Dahl, R.A. 92 Dahlgren, P. 88 Daily Show, The 306–7, 457, 461, 464, 467 Dal Cin, S. 343 Dalen, A. van 379 Damstra, A. 433 dark ads 406, 407, 409 dark participation 449, 452, 453 Darmon, K. 314 data see digital data data protection rules 402 data-driven campaigns 401, 404–8 datafication 202–7, 211, 238 Davis, R. 457 Dayan, D. 197–8, 205 Dean, J. 37, 85 DebateGraph 146 DC Inside, political expression (case example) 115, 119, 122, 128 De Cindio, F. 143, 148 decision-making 210, 211, 216, 218, 243, 262, 328 decoloniality 19–20, 253

deepfakes 11, 31, 247–8 deliberation cross-cutting debate, online deliberation 149 democracy 82, 111, 138, 139, 279, 356 digital protest 361–5 facilitation, online 146–7 hybrid online 149 norms 148 principles of 138 procedural model 364 two-screen viewing and 159 see also online deliberation Deliberative Community Networks (Open DCN) project 144–5 Deliberatorium 146, 147 Della Porta, D. 50 Delli Carpini, M.X. 457 Del Vicario, M. 100 Deme interface 144 Democracia real YA 288, 302 democracy African countries 17, 18 artificial intelligence and 252 direct versus representative 272–3 e-petitions 327–8, 330–31 effect of political humour on 465 exposure to political information 92 the Fifth Estate and 279–80 homophilic political communication as challenge to 140 and the Internet see Internet legitimacy 409 post-Soviet 354–7, 364 two-screen viewing and 159–60, 164 see also deliberative democracy; digital democracy Democracy Action Plan (EU, 2019) 410 democratic backsliding 157 democratic fatigue 18 design, power through algorithms 219 Dewey, C. 207 de-Westernization, Global South perspective 16–26 culture-centered approach 19–20 defined 17–19 digital transformation and e-colonialism 20–22 locally based digital solutions 24–5 physical challenges and divides 22–4 de-Westernizing 16–26 D’Heer, E. 159

Index  477

Diakopoulos, N. 217, 222 Diamond, L. 421, 423 Diani, M. 300 diasporic digital politics 38, 62, 64–5 digital activism adaptation to restrictive environments 49–52 African countries 17 authoritarian reactions to 19, 45, 46 global 66–7 post-Soviet countries 359–60 relevance of media ecology 313 social media 45–6, 87, 200 visual 30–35, 38–40 see also collective action; connective action; digital protest; individual movements; MovAct studies digital architectures defining 228–9 design, and shaping of political processes 226 implications for democracy 6–7 mapping for politics 232–4 political agency 229–31 political participation 234–8 and user activity 231–2 digital authoritarianism 16 see also authoritarian regimes digital data algorithmic processes 212 citizens’ information diets 98–100, 105 and communication see political communication theft 85 and user behaviour 111 see also big data; trace data digital democracy (e-democracy) 17–18, 138, 279 direct democracy 4, 272 opposition to 272–3 post-Soviet 357–60 digital dictatorship 8 see also authoritarian regimes, digital authoritarianism digital diplomacy 62–4 digital divides 16, 24, 282, 343 digital literacy/illiteracy 8, 16, 20, 24, 184, 185 Digital Markets Act 412 digital natives 168, 169 digital revolution, Africa 24–5 Digital Services Act 265

digitally networked action (DNA) 290, 291, 295, 299 see also connective action D’Ignazio, C. 253 Dillet, B. 390 Dimitrova, D.V. 173 diplomacy, migration 65 disabled youth, and civic engagement 184 discourse digital protest, Russia 361, 362, 364 search engines and shaping of 49 technical protocols and content of 228–9 two-screen viewing and political 160 strategies, of journalists in restrictive settings 424 discourse architecture 227–8 discrimination 183, 210, 212, 218, 261, 405 discussion forums 141, 361–5 see also forums, political expression disinformation 8, 10, 11, 21, 27, 182, 211, 212, 213, 217, 264–5, 293, 404, 436, 439, 450–53 criminalization 263, 267–8 during conflict 260 Facebook removal of accounts and pages tied to 24 spread/dissemination 18, 49, 104, 105, 212, 248, 261 state-sponsored transnational 63–4 disruptive political events 196 disruptive populist performances 372, 376–7, 379 distrust 24, 379, 439 Downing, J. 355 DREAMer protests 168 Dubois, E. 98 Duncombe, C. 63 Durkheim, E. 110 Duterte, R. 380 Dwyer, T. 85 Dymovskiy, A. 360 e-colonialism 20–22 e-commerce 23 E-Liberate 147 e-petitioning 327–55 agenda-setting and moderation 328–30 Downing Street 36–7, 327, 328, 333 interactive 147 NGO 327–8, 331–5 impact 331–4

478  Handbook of digital politics

linkage 331 participants, e-petitions 330 scrutiny 331 echo chambers 97–111, 214, 281, 347, 408 and political talk 112 prevalence in digital media 98–101 prevalence and effects on small groups 104–5 ecology approach to communication communicative ecology theory 315–16 economic field, influence on journalism 419, 422 Economic Freedom Fighters (EFF) 374, 379 economic globalization 170, 291 economic power 164, 262, 419, 427 ecosystem metaphor, digital journalism 320–21 efficacy, effects of humour on attitudes to 463–5 Egelhofer, J.L. 434, 436, 439 Egyptian Revolution (2011) 45, 51, 197, 199, 200, 204, 314 Einwohner, R. 348 election communication see also campaigning algorithms and 220 campaign regulation controversies of data-driven campaigns 405–8 digital architectures and 231 ongoing debates 408–10 policy and research 410–12 social media and defining of reality 13 use of pop culture 173–4 election management bodies (EMBs) 402, 403 Electoral Commission 410, 411 elite democracy 76, 79 elites elite-level communication, in conflicts 261 journalists as 425 political discussion online 67 see also political elites Elsberg, D. 278 embodied actions 30–35, 38, 373, 381 embodied AI representations 243 emotional contagion 63 empowerment 3, 83, 87, 122, 149, 272, 274, 281, 282, 328, 357 enclave deliberation 150 ‘end of history’ consensus 157, 354, 364

entertainment social media use, Africa 18–19 see also humour; political entertainment entrepreneurship, digital 20, 22, 23 Esser, F. 355, 376 ethical journalism 402 Eurocentrism 18, 19, 253 Euromaidan protests 51–2 Europe/European Union alternative politics 81 democratic backsliding 157–8 fake news 436, 439 content moderation 264–5 news sources 213, 404 personalized action frames 293 right-wing alternative media 450–53 understanding of algorithms 211 European Information Society Forum 3–4 event television 157, 162, 176–7 events catalytic events, mediation of 202 broadcast events, and online discussion 158–9 digital data, mediation of events 197, 202–7 global iconic196 immediacy of 219 everyday politics 37, 39, 66–9 explainable artificial intelligence (XAI) 247 exposure to political news and knowledge 295–6 Facebook 12 algorithms and power 216–8 blocked access, Russia 356 challenge to information practices 281 citizens’ information diets 98–100 civic engagement/participation 87, 184 closure of APIs 162 computational propaganda 389 content moderation 8, 262–3 digital activism 6, 49–50, 67, 87, 199, 200, 230, 314, 319 digital architecture 229–31, 233, 236, 237 echo chambers 103 election communication 405 exposure to political content 96, 172 fake news 436 journalism in restrictive settings 422, 427 negativity and user engagement 217

Index  479

as a news source 213 online deliberation 149 personal profiling 85 political communication/discussion 38, 60, 63, 85 political expression (case example) 115, 117, 119, 120–21, 122, 128 populist communication 375 removal of accounts tied to disinformation 24 role in conflict 259 user base 156, 162–3 Facebook Oversight Board 265 facts 202–7, 406 fairness 396, 401, 402, 407, 412 fake news 248, 432–40, 436–9 see also disinformation consumers of 438–9 intention to deceive 435 journalistic format 434–5 labelling 435–6 in politics 432–40 production 247, 436–7 prohibiting 437–8 weaponization of 432, 436 fandom 173–5 far-right activist networks 67 Farage, N. 377, 380 Feezell, J.T. 215 Feigenbaum, A. 316, 317 15M mobilization 289–90, 320, 322 Fifth Estate 272–84 communicative power 275–8, 280, 282 communicative strategies 275–8 implications of 282–3 networked institutions and individuals 278–81 50 cent party 48 filter bubbles 97–8, 101–4, 214, 347, 408 filtered chronology 237–8 Fish, A. 314 Flaxman, S. 100 Flesher Fominaya, C. 322 Fletcher, R. 99 For Fair Elections (2011–2012) 45, 46, 50 Ford-Kavanaugh case, online political expression 117–22, 125 foreign interference, election campaigns 407, 409 forums, political expression (case example) 115, 117, 119, 122, 125, 128 see also discussion forums

Foucault, M. 409, 412 4Chan 37, 456 Fourth Estate 273, 280–82 see also Fifth Estate Foust, C.R. 319 fragmentation 406 fostered societal 48 media channels 92–93 of contemporary media environments 105 of publics 97 political 104 political polarization and 94 on single platform studies 412 social 406 structural 291 of Western hegemony 26 framing with data 393–5, 397 France 160, 211, 293 Freedom of Choice, Motorists’ forum 361–5 freedom of expression/speech 47, 401–2, 407, 437–8 Freelon, D. 137, 141, 149, 228 Frisch, N. 50 Fuchs, C. 83 Fukuyama, F. 354 functionality of digital media 39, 232–3, 236–7, 254 Furie, M. 37 G7 Summit (Biarritz, 2019) 20 G20 Meltdown 294 G20 protests 287–8, 294, 301–2, 304, 306, 314, 344 Gabowitsch, M. 363 Gagliardone, I. 259 gamified apps, political participation 237 gaming cultures 175 gatekeepers/gatekeeping 32, 59, 61, 160, 278, 370, 374, 375, 376, 382, 424, 428, 432 Gen Z 168–76, 181–9 cultural engagement 174–6 formal politics 173–4 late modern society 169–71 news use 171–3 online political discussion 60 political socialization and civic education 187–8 protest politics 168 skills and online practices 184–7

480  Handbook of digital politics

socio-economic status and affinity groups 182–4 gender 36, 125, 128–9, 160, 183 Gerbaudo, P. 38, 200, 342, 374 Germany 215, 224, 303 GetUp 329 Gezi Park protest (2013) 45, 46, 318 Giddens, A. 181 Gillan, K. 322 Giraud, E. 318 Gjesvik, L. 64 global digital activism 66–7 see also connective action; digital activism global politics 32, 40, 65–6 Global South 16–26, 34, 262 glocalization 114 GoFundMe 445 González-González, P. 376 Google 13, 49, 84, 204–5, 207, 245, 249, 281 Gorny, Y. 358 Gove, M. 392–3 government, the Fifth Estate and 279–80, 282, 283 government-led e-petitions 332–4 government-led moderation 328 see also moderation GPT systems 247 Grabe, M.E. 39 Graber, D.A 32 Graham, A. 2 Graham, T. 139, 149 Granovetter, M. 303 graphical user interface (GUI) 236 Graves, L. 432 Great Firewall of China 70, 282 Griffin, N. 155, 164 Grimm, J. 51 Guattari/Fuller understanding of media ecologies 316 Guess, A.M. 100 Gumucio Dagron, A. 321 Habermas, J. 78, 79, 356, 361, 364, 365, 408, 412 Haßler, J. 231 Hahn, K.S. 20 Hallin, D.C. 355, 450 Hannity, S. 380 hard news 99, 213, 459, 460, 465 Harders, C. 51

hardware 236 Harlow, S. 319 Harries, J. 395 Hart, R.P. 462, 463 Hartelius, E.J. 462, 463 hashtag activism 67, 105, 113, 125, 128 see also digital activism, connective action hashtags 36, 130, 184, 228, 230 Hassid, J. 422, 425 hate speech 8, 10, 18, 24, 259–64, 267 Haugen, F. 259 Hautea, S. 39 health/health care 250, 277, 332–3 heartland (populist) 374 Helberger, N. 412 Hellmeier, S. 47 Helmond, A. 234 Heo, Y.-C. 159 Hepp, A. 198–9 Hess, V.K. 463 high-choice media environments 92–3 Hindman, M. 83 historic events, mediation of 196–207 see also events Hollander, B.A. 462 Holtz-Bacha, C. 452 homophily 95, 96, 97, 99, 105, 140, 214 Hong Kong 49, 50, 61, 168, 176, 419 Hong, S.H. 211 Hoover, S.M. 197 Howard, P.N. 48, 64, 405, 408, 411 Hoyt, K.D. 319 Hsiao, Y. 343 human rights, and freedom of expression 401–2 Human Rights Watch 11 humour and political dissent 51 see also political comedy hybrid action networks 187, 301, 303–4, 344 see also digital activism, connective action hybrid media systems 274, 314, 318, 330, 345, 374, 446, 450 performances 380 Iannelli, L. 318 identity(ies) AI and marginalized 254 construction 11, 358 consumerist politics 68

Index  481

peer discussion and influence on 187–8 symbolism and 38 user authentication and trust 147 see also civic identity; collective identities ideology 93, 97, 100, 112, 438–9 image management/politics 30–31, 35–6, 38 imaginaries, digital 3 impact on undemocratic societies 359 incidental exposure, to news and political content 92, 95, 171–2, 184 Indigenous youth, political participation 183 Indonesia 48, 282, 419 Indymedia 318 inequalities among Gen Z 168 automated decision-making and 210 conflict, content moderation and 262–4 digital, Africa 23–4 online deliberative space design 142 in political knowledge 92 transnationality of political discussion 59 two-screening networks and 160 information Internet and 84 provision, deliberative online spaces 144–5 see also disinformation; false information; misinformation Information Commissioners Office (ICO) 410–11 information disorder 24 information ecology theory 315 information gap 94, 104 information interventions 265–8 information overload 170, 213 information production and dissemination, AI in 247–8, 251 information seekers/avoiders 92–4 information warfare 62–4, 407–8, 411 informed discussion tool (OpenDCN) 144–5 infotainment 422, 460 Inkster, N. 410 Innis, H. 315 Instagram blocked access, Russia 356 collective action 345–9 computational propaganda 389 digital architecture 229–31, 233, 236, 238–9 digital protest 339, 341

disability and civic engagement 184 journalists’ use in restrictive settings 421 as a news source 172 political expression (case example) 115, 117, 121–2, 128, 130–31 populist use of 375 visual political communication 36 institutional media 61, 69 institutional power 142, 164 see also political power, power/power relations institutions deliberative initiatives, and authentication 148 and the Fifth Estate 284 networked 278–81 two-screen viewing 160 interdisciplinary research 31, 33, 318, 426 international media 61 Internet and accountability 272 Africa 22–4 Chinese 70 as a civic space 76–88 contingencies as dynamic configurations 86–8 enthusiasts and sceptics 83–4, 87 significance for evolving democracy 76–80 control/suppression 45–8, 53 and democracy Fifth Estate 282, 283 freedom 46 new actors in media events 199 as a news source 404 political role 274–5 post-Soviet states 357–8 regulation 85 shutdowns 8, 17, 260, 263, 264, 268, 356 transformation of practices 85–6 Internet Research Agency (IRA) 48, 389 interpersonal communication 111, 112, 159, 358 interpersonal networks 220, 288, 299 investigative journalism 422 Iosifidis, P. 17 Iran 419–21, 423–5 Isin, E. 372 issue advocacy organizations 289, 301, 303 item response theory (IRT) 462

482  Handbook of digital politics

Iyengar, S. 92, 93 Jamieson, K.H. 97 Jasanoff, S. 373 Jenkins, H. 175, 342 Jiang, X. 112 Johnson, B. 36–7, 377, 392 Jones, Q. 227 Joo, J. 34, 39, 40 journalism 212 AI applications 247 digital architectures and influence on 231 ecosystem metaphor 320–21 ethical 402 Internet as marginalizing high quality of 280–81 and knowledge production 433 and social media in restrictive settings 418–28 see also news journalists boundary pushing 424, 425 criticism of OGD during pandemic 394 as gatekeepers 374–5, 424 interdependence between politicians and 379–80 labelled as fake or lying 435–6 populist targeting of 376–7, 379 role performance 423 tracking 423 visual political communication 38 K-pop fandom 175 Kahn, R. 315 Kaposi, I. 360 Karigambe-Sandberg, T. 18 Karlsson, M. 141 Karpf, D. 159, 329–30 Katz, E. 110, 111, 197–8, 205 Kavanaugh, Brett see Ford-Kavanaugh case Kazakhstan 46 Keck, M. 302 Keen, A. 84 Kellner, D. 315, 377 Kim, E.M. 148–9 Kim-Ahn case, online political expression 114–31 King, G. 47 Kissas, A. 378 Klein, L.F. 253 Klinger, U. 214, 345

Koçer, S. 51 Korean #MeToo activism, online political expression (case example) 127–9 Krämer, B. 383, 452 Kranzberg, M. 176 Kreis, R. 381 Kreiss, D. 405, 411 Kwak, N. 343 Kyrgyzstan uprisings (2010) 314 Lalancette, M. 36 Lane, D.S. 343 language detection 262–3 large-scale action networks 287, 290 organizational dynamics 287, 298, 300–306 see also collective action networks; connective action networks; digital activism Latin America 19, 53, 376, 382 Latour, B. 111, 300 Laws of Imitation, The 111 leadership 36, 221, 345, 464 leadership-critical approach, to public communication 47 leaking information 278 Lecheler, S. 434, 436, 439 legacy media see traditional media legitimacy of alternative media 453 democratic 409 of election campaigns 401 of information providers, populist questioning of 375, 379, 382 of populist leaders, questioning of 382 Leston-Bandeira, C. 331 Lewis, S.C. 423, 427 Leyva, R. 187 liberation technology 6–8, 423–7 Lievrouw, L.A. 82 lifestyle politics 67–9, 181 Lim, M. 319 linguistics 114, 129–31, 246–50, 262 Litvinenko, A. 45, 47, 48 LiveJournal 358 Lokot, T. 48, 50 London G20 protests 287–8, 292–4, 301, 304, 307 Lonkila, M. 50 Lopez, I. 114 los indignados 287–9, 293, 300, 302, 307, 342

Index  483

low-choice media environments 213 Lügenpresse 435 Lukito, J. 48 Lupia, A. 299 machine learning (ML) 129–30, 215, 243, 250, 406 see also artificial intelligence (AI) machine translation (MT) 114, 245, 249–50 see also artificial intelligence (AI) Macintosh, A. 146 MacKinnon, R. 85, 422 MacMillan, I 218 Maier, D. 114 mainstream media 87, 102, 273, 281, 411, 446–7 appearance of alternative media actors in 453 control, Russia 356 digital architectures 231 protest narratives 314 Malaysia 421 Mancini, P. 355 Mansbridge, J. 148 Margetts, H. 343 market, and journalism in restrictive settings 421–2, 427 market institutions, democracy and 364–5 Martini, F. 214 mass media see mainstream media mass self-communication 69 mass-based appeals, in conflicts 261 materiality digital media and populism 370, 372, 375, 381, 382 digitally mediated protest 341, 343, 344, 345 interwovenness of sociability and 5 Matthes, J. 18, 464 Mattoni, A. 50, 87 McCarthy, J. 242 McCarthy, J.D. 297 McCaughey, M. 318 McGarry, A. 34 McLuhan, M. 30, 315, 373 MeToo movement 114, 115, 200, 214, 278, 344 see also online political expression Medeiros, J. 436 media as an extension of ourselves 30–31, 41, 373

fake news label 432, 435 post-Soviet democratization 354–7 revolutions in technology 80 see also alternative media; digital media; mainstream media; social media Media, Culture and Society 198 media ecology digital revolution, Africa 25 social movements and activism 313 see also ecology approach to communication media events 196–207 actors 199, 200, 201, 202, 204, 205, 206 consensus and debate about 197–9 in context of social media and data 199–201 and datafied facts 202–5 operationalizing 201–2 Media Events: The Live Broadcasting of History 197 media populism 383 see also populism mediation, digital data 196, 202–5, 206–7 mediatization theory 316 medium theory 315–17, 322 Medvedev, President 359 Mejias, U.A. 19 Melucci, A. 342 memes 37–8, 51, 186, 292–3, 294, 373 memetic weaponization 38 memory, asynchronous conversation and 145 memory narratives 48 Mendonça, R.F. 36 Merton, R.K. 12 messaging apps 101, 102, 105, 145, 236, 238 see also individual apps Messenger 184 Meta 264 metaverse 453 Meyrowitz, J. 315 Mickiewicz, E. 355 Middle East and North Africa (MENA) 38, 47, 280 Milan, S. 343–6, 348 military AI applications 246, 248 Mill, J.S. 407 millennials 168, 169, 170 see also political participation/young people Milton, J. 407

484  Handbook of digital politics

Minocher, X. 334 misinformation 280–81, 434 see also fake news authoritarian regimes 61 during Covid-19 pandemic 436 echo chambers and spread of 104 social media and 19, 158 visual 34 Mitchell, B. 316 Mitchell, W.J.T. 33, 40–41 Mitts, T. 231 Moats, D. 201–2, 204, 206 mobile data costs 23 mobile phones 20, 22, 23, 39, 87, 275–8 mobile technologies 23, 82 mobilization 235 digital architecture 236 e-petitions and 331 social media and 7–8, 45, 46, 87 see also 15M mobilization; counter-mobilization; resource mobilization/theory moderation agenda-setting and 328–30 content 7, 8, 261, 263 e-petitions 327–8 government-led 332–4, 328 online deliberation 146–7 oversight, social media 264–5 platform 440 see also content moderation modernity, late 62, 66, 68, 169–70, 291, 295, 305 Moffitt, B. 371, 376, 378 Molyneux, L. 423, 427 money, role in campaigns 407, 412 Monnoyer-Smith, L. 142 Moreno-Almeida, C. 38 Morgan, M.S. 203 Morozov, E. 83, 274, 343 Morris, D. 4 Morris, J.S 463, 465 Mortensen, M. 38 Moss, S. 140 Motorists’ forum 361–5 MovAct studies 313–14 MoveOn 329 Moy, P. 463 Mudde, C. 370 Mueller investigation 409 Müller, P. 377 multi-disciplinary research 40

multi-modal activism 347–8, 350 multi-modal analysis, political expression 130–31 multi-modality, political discussion online 61–2 multinational corporations 68 multivariate time-series analysis 129 Musa, B.A. 24, 25 Myanmar 6–9, 19, 259, 260, 262, 263 Nadler, A. 320 Naím, M. 218, 221 Namasinga Selnes, F. 423 Nardi, B. 316 Nate Pann, political expression (case example) 115, 118, 119, 122, 131 National AI Strategy 52 National Security Agency (USA) 84 natural language generation (NLG) 247, 251 natural language processing (NLP) 113, 115, 129–30, 250 Navalny, A. 50 negativity, and user engagement 217 Negroponte, N. 3–4, 94 neoliberalism 77 network media logic 375 network structure 235–6 networked authoritarianism 422 networked citizenship 365 networked commercialization 422 networked individualism 170 see also Fifth Estate networked institutions 278–81 networks/networking communicative power 277 late modern society 169–71, 291 see also actor network theory; interpersonal networks; large-scale action networks; virtual private networks Neumayer, C. 38 NeverSeconds 276 New Knowledge 389 New Media and American Politics 457 New Public Management (NPM) 393 ‘New Wave’ of democracies 354, 364 new waves, in journalism 425 news AI and creation of 245 digital architectures and presentation of 231

Index  485

eroding boundaries between entertainment and 457, 458 Gen Z 170–72 media see alternative media; mainstream media; traditional media sources 92, 93, 171, 211, 212, 290, 404–5 see also fake news; journalism News Consumption Report (Ofcom, 2018) 404 ‘news finds me’ attitude 171, 213 newsfeeds 94, 221 newspapers 110, 171, 202, 281, 404 NGO e-petitions 327–8, 331–5 NGO networks 302, 303 niches, online deliberation 150 Nicoli, N. 17 Noble, S.U. 210 non-simultaneous two-screen viewing 156–7, 161–2 normalization 5, 38, 157, 184, 186, 342, 445, 447, 452–3 norms, populism and transgression of political 381 Norris, P. 5 Obama, B. 63, 81, 158 objectivity, A-level standardization model 396–7 obligation 218 Ocasio-Cortez, A. 30–31, 174 Occupy movement (2011) 81, 290, 292, 300, 301, 302, 307, 319, 342 O’Day, V. 349 Ofcom 61, 404, 413 Office of Qualifications and Examinations Regulation (Ofqual) 396 offline engagement, Gen Z 187 Olson, M. 296, 297, 299 on-demand media, expectation of 170 O’Neill, O. 407 online deliberation 137–50 cross-cutting debate 149 design of spaces for 140–43 division of labour and functions 148 future research 148–50 quality of deliberative practice 139–43 online political discussion 59–70 broadcast events and 158–9 convergence or decoupling 69–70 discourse architectures 227–8

Facebook as a site for 85 rise of (global) digital citizens 66–9 Russia 59, 357–8 state actors as protagonists in curating 62–5 transnational and multimodal 61–2 see also discussion forums online political expression computational and comparative approaches 113–14 framework for research 129–31 Gen Z 187–8 MeToo and global activism (study) 114–15 and political action 146 power of social media 16–17 quality of 110–11 shifting forms of 112–13 opacity, algorithmic 217, 219, 222 open government data (OGD) 391–5 opinion formation 111, 160, 213, 411, 412, 436, 463 opinion leadership 103–4, 110, 138 opinion reinforcement 215 oppositional media 52 oppositional pathos, legacy media 451 Orange Revolution (2005) 50 ordinariness, populist performance of 378, 382 organization asynchronous conversation and problem of 145 network structures of 170 organizational convergence 82 organizers, of media events 197, 198 Orgeret, K. 423 originating information 276 ‘other media’ of activism 317 out-groups, populist politics 164 oversight, social media content moderation 264–5 Owen, D. 457 Oxford Internet Institute 48 pack journalism 281 Paglen, T. 251 paid commenters 48–9 Paltemaa, L. 49 Papacharissi, Z. 85, 86, 110–11 Papakyriakopoulos, O. 214 Pariser, E. 97, 406, 408 Parkinson, J. 148

486  Handbook of digital politics

part-of-speech approach, data analysis 130 participant authentication, online deliberative space design 147–8 participants, e-petitions 330 participation latent 235 see also dark participation; political participation participatory cultures 174, 175, 342–3, 449 participatory politics, memes and 37–8 partisanship 143, 149, 215, 238, 404, 438 Partygate scandal 36 Pasquale, F. 217 pathos 391, 451 Patterson, T.E. 465 Payne, M. 276 Pearce, K.E. 51 Peeples, J. 316 Pensioners’ Movement 316 Pepe the Frog 37–8 Peraboni, C. 148 perfect information arena 4 performance/performativity of authenticity 11 digital acts 203, 412, 421 of Instagram, in digital protest 348–9, 350 visual politics 31–2, 35 see also populist performance; role performance; social performance persistence 226 personal action frames 291–5, 299, 302, 303 personal information, selling of 84 personal preferences, visibility of news content 172 personal profiling 85 personality, and opinion formation 463 personalization of politics 30–31, 36, 66, 86, 212, 214, 412 persuasion 2, 218, 219, 220, 462–3 Peters, C. 38 Pew Research Center 171, 459 Pfetsch, B. 355 physical-digital spaces, interplay between 318–19 piggybacking, political information onto humour 461 Pingree, R.J. 145, 146 Pittman, M. 161–2 Pittsburgh G20 protest 287 platforms

activism and appropriation of corporate 319–20 affordances see affordances authoritarian resilience 49 business model 8 civic participation 87 dissemination of hate speech and disinformation 18 echo chambers and filter bubbles 101–4 incitement of violence 8 journalism in restrictive settings 421–2 participatory culture, Gen Z 174–5 political creativity 173 and power 217–18, 407, 412, 422 and visuality 347 see also digital architectures; individual platforms plurality 273, 402, 406, 413 Poell, T. 314, 344, 347 Pokémon GO 176 polarization algorithms and 215 alternative media and 453 country-level systemic characteristics, and 109 echo chambers and 98–100, 104 exposure to political information and 92, 93, 97 horizontal 94, 97, 104 online political expression (case example) 122, 128–9 personalization of media diets and 408 political discussion 137, 139 police brutality 38, 39, 175, 187, 340, 421 policy e-petitions and 331–4 peer-to-peer deliberation 143 policy critical approach, to public communication 47 political action, decoupling of knowledge and 172–3 political advertising 173, 402–4, 406, 408–12 political authenticity 9–12 political campaign efforts 173–4 political comedy increasing prominence 457–8 political learning and knowledge 461–2 significance of 458–61 and social media, research 466 political communication change in environment of 459

Index  487

complexity of process 9 and data 388–97 fake news research 432–3 homophilic 140 mobilization of affect 13 populist 372, 376, 381 reciprocal 70 social media and 85 two-screen viewing and 155 visual 31–3 see also discourse; election communication; fake news; propaganda political content, personal action frames 292 political discussion citizen engagement with 92 civil society and the public sphere 78–80 effect of humour on 464 polarization 137, 138 see also online deliberation; online political discussion political economy 84–5, 426 political elites blocking Internet access, Africa 17 co-opting of digital networks 12 e-petitions and 331 fake news 435, 436 and journalism 426 populism and 379, 380, 452 post-Soviet democratization 354–7 political engagement effect of humour on 464–5 fragmentation and 291–2 and media choice 360 social media and 16 see also e-petitioning; political participation political entertainment 457–68 comedy see political comedy future research 466–8 research findings and controversies 461–6 political figures, effects of humour on attitudes to 463–4 political information access to 7, 274 accidental and deliberate exposure 95–6, 104 algorithms 213 citizen engagement with 92

incidental exposure 92, 93, 95, 171–2, 184 selective exposure 93–5 use of humour to convey 458, 461 see also echo chambers; filter bubbles political knowledge 35, 45, 92, 172, 184, 186, 212, 222–3, 231, 235, 462–3, 467 political learning 92, 159, 164, 462–3 political participation algorithms and 213 defining 234–5 digital architectures 234–9 effect of humour on 464–5 elite and republican models of democracy 76, 79 Gen Z 182–3 horizontal trust and 88 social media 339–40 two-screen viewing 155 young people 17, 173–4, 175–6, 181–2 political power 2, 220, 392, 412, 427 political socialization 183–4, 187–8 political talk see also authoritarian regimes among Gen Z 174 communication research 110–13 frequency of, and experience of echo chambers 103–4 politician-focused visual imagery 35–7 popular culture 32, 88, 173–4, 175, 186, 450 populism 370–83 in-groups, populist politics 164 materiality of digital media 370, 373, 381, 382 technological 63 and two-screen viewing 163 visual politics 37–8 Populism and the Media 452 populist performance digital media and 409, 410–21, 423 four types of mediated 375–81 post-broadcast democracy 92, 116 post-Soviet democratization 354–65 post-truth 412,436 Postman, N. 315, 356, 465 posts/posting datafied responses 238 digital activism, Russia 427 length of, and user behaviour 96 power/power relations algorithms and 216–21, 222, 346–7

488  Handbook of digital politics

de-Westernization and 17–18 of digital media 200, 240–41 of digital platforms 218, 407, 411–12, 422 Internet as a civic space 79, 83, 84 journalism in restrictive settings 424 of mass media 446–7 in mediation of events 205–6 of political humour 462 of statistical rhetoric 391 transnational political discussion 69–71 two-screen viewing and 161 of the visual 31 see also communicative power; economic power; empowerment; political power; symbolic power predictive analytics 246 preference shifting, online deliberation 140, 148 presidential elections 38, 105, 155, 158, 162, 173, 221, 242, 389, 436, 457 press, and the Fifth Estates 280–81 press freedom 47, 436 print media 171, 318 Prior, M. 92, 94, 460, 462 privacy 84, 85, 229, 233, 235–6, 281, 300, 410–11 pro-China narrative 60, 63 profiling 85, 408, 409 programmers, power through algorithms 220 propaganda authoritarian 45, 47, 48, 52, 53, 59, 437 computational 48, 64, 389, 412 prosopopoeia 391, 397 protest movements 287–8, 341–3 see also digital protests; individual movements protest politics, young people 168 protests, digital 287 digital architectures and 7, 229–31 embodied (visual) 30–31 networked individuals 279–80 Russia, in the 2000s 361–4 see also collective action; connective action; digital activism; protest movements public sphere 78–80, 82, 84, 137–8 Put People First (PPF) 287–90, 292–4, 301, 303, 307, 308 Quandt, T. 449, 452 quantification 393

queer youth, and political participation 183 Question Time (BBC, 2009) 155, 164 question-answer machines 203–4 radicalization 105, 158, 215–16 radio 266, 318, 404 Rafaeli, S. 227 rationality 77, 84, 103, 144, 149, 361, 408 Rattray, B. 330 Rauch, J. 452 Rauchfleisch, A. 464 Raynauld, V. 36 Red Sunday (Thailand) 38 Reddit 37, 48, 115, 118–20, 122, 131, 172, 227, 230, 236 redistribution affordances 96 regulation of the Internet 85–6 of political communication 401–3, 412 reinforcement machine learning 243 relevance, algorithmic filtering 237–8 Reporters Without Borders 354, 436 representative democracy 272–3, 279, 331 republican democracy 77, 79 resistance 8, 25–6, 67 see also digital protest resource mobilization/theory 48–9, 297, 298, 302, 308 Reuter, O.J. 50 Reuters Digital News Report (2021) 211 reverence, in mediation of events 197 reverse mullet business model 330 Rhee, J.W. 147–8 rhetoric of digital politics 2–13 Ricard, J. 436 right-wing alternative media 452–3 culture wars 449–50 historical perspective 451 political aspect 446–7, 448–9, 452–3 populist performances 377–8, 382 trucker convoy protests and relevance of 444–5 Right-Wing Alternative Media 446, 449–50, 452 Robert’s Rules of Order 147 robots 243, 247 Rochford, E. 348 Rorty, R. 5 Rosenberger, S. 331 Rowbottom, J. 405 RT 61

Index  489

rule systems see regulation; Robert’s Rules of Order; technical protocols Rumble 444 Runet 357 Ruppert, E. 372 Russia decline in press freedom 47 digital activism/protest 49–52, 361–4 digital democratization 357, 358 fake news 436, 437–8 journalism in 420–22, 425 media manipulation/control 355–6 political discussion 59, 358, 359 social media 45, 47–8, 50–52 see also Ukraine War Sack, W. 227 Sampson, H. 424 Sanders, B. 173–4 Scannell, P. 197 Schedler, A. 47 Schia, N.N. 64 Schill, D. 33–4 Schmidt, H. 358 Schraube, E. 5 Schudson, M. 434 Schulz, A. 377 Schwarzenegger, C. 452 Scolari, C. 315 Scotland 332–4 screen culture, and communication 36–7 scrutiny, e-petitions and 331 search engines 49, 94, 203, 249, 263, 274 searchability 235–6 searching 276 Segerberg, A. 343, 344, 345 selective avoidance 96 selective exposure, to political content 93–5, 112 self-branding 9–11 self-censorship 45, 51, 418–19, 423 self-exposure, and political authenticity 9–10 self-organizing action networks 301, 302, 344 self-questioning, and quality deliberation 140 self-regulation, election campaigns 402 self-representation 11, 35–7, 378 semantic data 204 semi-magical round numbers 391, 397 Senate Select Committee on Intelligence (SSCI) 389

sentiment analysis, in forecasting 250 sexual minorities, and political participation 183 shadow banning 184, 219, 347, 350 shareability 337, 348 shared community, platform affordances and 185–6 shared public knowledge 85 shared values, media events and 198 sharing 51, 78, 83, 299, 304, 306, 308, 376 Shifman, L. 37 Shirky, C. 83 shutdowns 8, 17, 260–61, 263, 264, 268, 356 Signal 229, 236 Sikkink, K. 302 Silverstone, R. 374 Sin, G. 299 Singapore 48, 383–4, 421 Sinpeng, A. 48 skills, and civic engagement 184–8 Skjerdal, T. 423 slacktivism 86, 186, 187, 279, 327, 343 Snapchat 172, 230, 233, 235, 238 social constructionism 373, 434 social contract, online deliberative space design 143, 146 social grouping, online deliberative space design 143 in authoritarian regimes see authoritarian regimes collective action 341–5 Covid-related information 170 exposure to politics and offline engagement 186–7 global digital activism 66–7 Global South 16–18 impact on digital politics 250 journalism see journalism mediation of events 199–201 and misinformation 158 news use 171–2 personal action frames spread through 292–3 platforms see platforms political communication 35, 36, 40, 156, 175, 389 political participation 183, 339 research, use of AI in 250 role in conflict see conflict trust in 211 youth civil disobedience 182 social movements

490  Handbook of digital politics

and algorithms 213, 219–20 use and role of social media 8, 201 see also MovAct studies social performance 372–3 social reality(ies) 114, 347, 418, 424, 434 social relations 85–6, 197 Social Shaping of Technology 316 social ties 142, 148 see also strong ties; weak ties socialized communication 69, 70, 424 socio-economic status (SES), and youth engagement 182–4 sociotechnical imaginaries 373–4, 378, 381 soft news 458, 459, 462 solo sphere 86 Sombatpoonsiri, J. 38 Soninke web 20 Sorensen, L. 370, 375, 377, 379, 381 South Korea deepfakes in presidential election 247–8 see also online political expression, MeToo and global activism space of flows 273–4 Sparks, C. 355 specificity, visual digital politics research 41 speech analysis, AI and 250–51 Splendore, S. 161 Srinivasan, R. 314 Srinivasan, S. 16 standardization model (A-level) 388, 396–7 statistical rhetoric 388, 390–92 Stern, R.E. 422, 425 Stevens, H. 393 Stewart, J. 457, 461 stimulus/response relationship, broadcasts and online discussion 158–9 Stockmann, D. 53 ‘storm the banks’ events 294 storytelling 184, 243, 251, 372, 382, 450 Strate, L. 318 streaming 157, 161, 174 Street, J. 146, 227 strong authentication 148 strong ties 98, 105, 233, 288, 305 Structural Topic Modeling (STM) 125–30 subjectivities, reconfiguration of 13 Sudworth, J. 60 Suk, J. 113–14 Sun, W. 65 Sunstein, C.R. 97, 150, 406, 408 supervised machine learning 130, 243 supported media 236

suppression 17, 18, 45, 46, 49, 82, 184, 365, 389, 421 surveillance 65, 82, 84, 85, 321–2, 405, 422, 428 surveys, citizens’ information diets 98, 101 Svensson, J. 219–20, 345 symbolic capital 425 symbolic meaning construction 348–9 symbolic performance 372, 375, 378, 379, 375, 378 symbolic politics 33–4, 37–8, 41 symbolic power 424, 425 synchronous conversation 145 Synthesio 115 Szakonyi, D. 50 Taggart, P. 374 Tandoc Jr., E.C. 434 Tang, L. 424 Tarde, G. 110–12 targeting, in political advertising 409 technical protocols 228–9, 232, 236 techno-fix, and the Internet as a civic space 80 technological convergence 81–2 technological determinism 3, 80, 274, 316 Tefertiller, A.C. 161–2 Telegram 50–53, 212, 217, 229–31, 236, 356 TeleParty 157, 162 televised election debates 155–6 television counter-mobilization 51 exposure to political information 92–3 as a news source 171, 404, 405 political communication 32, 40 post-Soviet Russia 354, 355, 361 protest mobilization 50 two-screen viewing 155–9, 161–2 temporality 50, 202 see also chronology; time-series analysis Tennison, J. 394 Tenove, C. 64 terrorism 21, 105, 199, 205, 229, 231, 236, 262 Teubener, K. 358 Tewksbury, D. 95 texts/texting 87, 280 theoretical clarification, fake news research 433 third space(s) 67–8, 143–4 Thompson, J.B. 35–6

Index  491

Thorson, K. 314 Thunberg, G. 276, 277, 279 TikTok algorithm 243 authoritarianism, activism and research 53 diasporic digital politics 64 digital architecture 228, 237 global digital activism 66–7 incitement of violence, Myanmar 8 news use 171–2 visual activism 34 youth engagement 176, 182, 186–7 Tilly, C. 288 time-series analysis 177, 129 Todos Somos Agua campaign 319 Toepfl, F. 47, 48 toeslagenaffaire 254 Tong, J. 377 top-down communication 376, 382, 437 topic analysis, in forecasting 250 NGO e-petitions 332 and online deliberative quality 143 online political expression (case example) 116–17, 125–9 Tormey, S. 378 Toronto protests 293, 306, 344 Toronto School of media ecology see medium theory trace data 95, 98, 105, 132, 239 traditional media (legacy media) activism and reconfiguration of 317–18 and the Fifth Estate 277, 279–80 humour in news programmes 458–9 populism and 376–8, 379, 382 transborder activism 52 transforming data 395–7 transmedia 82 transnational authoritarianism 65 transnational events, digital media and creation of 200 transnationality, political discussion online 59–70 transparency data openness 388–9, 392–5 platform content moderation and 440 political advertising 409–11 Treré, E. 230, 315–17, 319 trucker convoy protests 444–6, 449 Trudeau, J. 36, 444, 445 Trump, D. 105, 113, 163, 409

Cambridge Analytica and 2016 election campaign 406 and fake news 432, 435, 436 Twitter use 63, 116, 121, 125–6, 163–4, 275, 375, 377, 378, 379, 381 visual politics 21 trust 415 data openness 388–9, 392–5, 397 digital networking mechanisms 300 e-petitioning and 327 effect of political humour on 463–4 fake news and 432, 434 horizontal 88 journalism and 425 network 235–6 online deliberation 148 and participation 88 quality of online deliberation 142 in social media as a news source 211 truth-seeking 402, 409, 412 truth-telling 13, 380, 382 Tsfati, Y. 438 Tuchman, G. 434 Tufekci, Z. 215, 314 Tumblr 183, 290 Tunisia 45, 49, 61 Turkey 45, 46, 51, 419–21, 436 Turkmenistan 47, 354 Tweets and the Streets 200 Twitch 174, 176 Twitter diasporic digital politics 63 digital activism 6–7, 49–52, 67, 199, 201, 231, 288–9, 313, 319–20, 344–5 digital architecture 229–31, 233, 237, 238, 239, 240 echo chambers and filter bubbles 103, 112, 408 journalists’ use in restrictive settings 422 news use 171 online deliberation 144, 149 political discussion 71–2, 76 political expression (case example) 115, 116, 117, 119–23, 128 political talk 111–14 populist communication 375, 379 propaganda 48 user base 156–7, 163 youth engagement 186 Twitterbots 248

492  Handbook of digital politics

two-screen viewing household interaction 159 research 158–60, 161–4 Ukraine 50–51, 353–4, 357 Ukraine War 45, 244–5, 259, 260, 264, 280, 362–3 Umbrella protests (2014) 50 Unchat 144, 145, 147 Understanding Media: The Extensions of Man 30 unfiltered chronology 238 unfriending 48, 96, 172 artificial intelligence 252 Brexit 158, 377, 409 campaign regulation 401–3, 410–11 Facebook and Twitter user base 162–3 United Nations (UN) 265, 268, 438 Chinese diasporas 65 Gen Z 168–9 storming of Capitol 113, 158 two-screen viewing 155–6 see also online political expression, MeToo and global activism univariate time-series analysis 129 unsupervised machine learning 130, 243 Usenet 227 user-generated content (UGC) 93, 173, 175, 282, 457 user-technology relationship 226, 229–30 Uyghur Muslims 59–60, 65, 66 Vaccari, C. 159 vanity metrics 9, 11 Veneti, A. 34 vertical polarization 94, 104 video games, and political participation 175–6 viral digital activism 67 virtual private networks 7, 8, 280 virtual publics, discourse architecture and 226–7 virtuality, focus on, in research 319–20 visibility 40, 50, 172, 226, 346–7, 350 visual analysis, artificial intelligence 246 visual communication communication studies 33 visual digital politics 30–41 visual political expression 186 visuality, collective action, Instagram 347–8, 350–51 visualization, AI and 278

voices, marginalized and youth engagement 186 Volkmer, I. 70 Vote.Com 4 voter ID 411 voters/voting 161, 173, 183, 246, 389, 405, 406, 408 Vu, H.T. 422 Wagner-Pacifici, R. 200, 207 Waisbord, S. 376, 380 Walsh, K.C. 111 weak authentication 148 weak ties 67, 96, 105, 233, 291, 303 Web 2.0 21, 374, 447, 452 Web 3.0 453 web tracking, citizens’ information diets 98, 100–101, 104 WeChat 65 Weibo 48, 149, 421 Weizenbaum, J. 222 Wells, C. 432 Western hegemony 25–6 Westernization, mediated politics in transition 355 Western media theories 426 Western multimedia conglomerates 21–2, 25 WhatsApp 10, 157, 163, 175, 184, 212, 217, 230, 236 Wikidata 204–5 Wikipedia 84, 199, 202, 203, 204, 205, 207, 281, 299 Williams, B.A. 457 Williams, N.W. 131 Williams, R. 316 Wilson, C. 314 Wizards of the Web 219–20 Wojcieszak, M.E. 100, 462 work, and the Fifth Estate 281–2 Wright, S. 67, 146, 147, 227, 333 WUNC 288, 289, 304 Xenos, M. 463 Xinjiang 59–60, 63, 64, 65, 67, 68 Yandex 49, 51, 358 Yang, Y. 343 young people Africa 18–19, 22 global digital activism 66–7 Internet as a civic space 82–3, 86 political entertainment 460

Index  493

two-screen viewing 155 see also Gen Z YouTube digital activism 51, 52, 200–201, 288–9, 344, 359–60, 444 digital architecture 226–8, 233, 237 echo chambers 103 online deliberation 149 political expression (case example) 115, 117, 119, 121–2, 128, 130–31 political satire 457–8 post-Soviet Russia 354–5 radicalization 215–16

Yu, H. 65 Zald, M.N. 297 Zavadski, A. 48–9 Zeitoff, T. 261 Zelensky, V. 9–11, 248, 275 Zhang, H. 379 Zhang, Y. 111, 112, 130 Zimbabwe 18 Zoom 163, 183 Zuboff, S. 219 Zuckerberg, M. 217, 453