Turning Outrage into Disgust: The Emotional

Basis of Democratic Backsliding in Hungary

Dissertation

Presented in Partial Fulfillment of the Requirements for the DegreeDoctor of Philosophy in the Graduate School of The Ohio State

University

By

Paul Armstrong DeBell, B.A., M.A

Graduate Program in Political Science

The Ohio State University

2016

Dissertation Committee:

Goldie A. Shabad, Advisor

Irfan Nooruddin, Co-Advisor

Anthony Mughan

Kathleen McGraw

c© Copyright by

Paul Armstrong DeBell

2016

Abstract

Once a frontrunner of democratization in post-communist Europe, Hungary is

backsliding. The 2010 election left the country with an unstable and polarized party

system, and the population is quiescent as the Fidesz government dismantles the

institutions of liberal democracy. Indeed, support for the very idea of democracy is

weaker today in Hungary than it was at the transition. Democracy, it turns out, is

not seen to be the only game in town for many Hungarians.

Why would experience with democracy fail to yield support for the ideals and

norms of democracy? Why do party systems in the young democracies of East

Central Europe (ECE) remain highly volatile? Answering these questions requires

attending to voters experiences with and feelings towards their democratic systems.

This dissertation examines the emotional dynamics of political behavior in Hungary,

revealing high levels of popular disgust towards politics driving an active rejection of

competitive multiparty politics and engendering democratic backsliding.

I show that policy constraint from the European Union limits the ability of main-

stream political parties in post-communist Europe to differentiate themselves from

one another concerning many of the key policy issues most important to voters. This

leaves elites with little maneuvering room to make programmatic appeals, increasing

the likelihood that they will leverage the power of populist outrage – a discourse of

ii

alleging real or imagined moral transgressions by political competitors – to differen-

tiate themselves from competitors and inspire political action. Where this vitriolic

discourse elicits anger it causes participation in the form of protest voting, thus ex-

plaining persistent party system volatility. However, this constant stream of vitriol

often elicits disgust. This powerful emotion causes a visceral avoidance of politics

that undermines the accountability mechanism at the heart of democracy and ex-

plains rising disaffection from democracy in Hungary and around the world.

I test this argument with multiple methods and sources of original data. A combi-

nation of Comparative Manifestos data and an original data set of Hungarian political

coverage shows a rise in the use of populist outrage in political communications in

the wake of increased EU constraint and new media technologies. Data from nine

months of fieldwork document the effects of this populist outrage upon the public.

An online experiment and a lab-in-the-field experiment demonstrate the reactions of

both anger and disgust to populist outrage and show that partisanship is the key

determinant of who becomes angry instead of merely disgusted. An original survey of

1,000 Hungarian adults reveals very high levels of political anger and disgust among

the populace and shows that disgusted citizens are less likely to view democracy as

a good system of government for their country. Finally, interviews and focus groups

conducted during the 2014 Hungarian national election show an over-time evolution

of anger at political vitriol into disgust, demonstrating that where multiparty com-

petition is equated to a war between immoral, self-interested factions, the popular

response will be a rejection of the system.

iii

This dissertation is dedicated to my grandmother, Mildred Keller DeBell.

You instilled in me the love of politics that inspired this dissertation and exemplify

the persistence that made it possible.

Thank you.

iv

Acknowledgments

So many people deserve my gratitude for making this dissertation a reality. First

and foremost, I would have never made it through (or even to) graduate school without

the unwavering support of my family and friends. My parents Brenda and Stuart,

my sister Sarah, and my Aunt Connie and Uncle John have always encouraged me to

pursue my passions, even when that took me to strange places and involved enigmatic

career choices. My amazing group of friends from home and from William & Mary

were a constant source of joy and inspiration throughout. Last but not least, my boys,

Alex and Reggie, provided constant inspiration, solace, and companionship through

the highs and the lows. Alex, you kept me grounded and laughing every day. Thank

you for that.

Professionally, the outstanding guidance and friendship of my committee members

has had the single greatest effect upon this document and the political scientist that I

am today. Goldie Shabad guided me from the very beginning of this idea to its current

realization, providing steady encouragement at every point along the way as well as

invaluable insight on every aspect of this research. Our lunches to discuss everything

from research to current events to travel plans remain some of my favorite experiences

in graduate school. Irfan Nooruddin far surpassed the title of mentor, playing the

role of everything from research advisor to real-talk career coach to advocate to close

confidant at any given time. Our friendship, and the many friendships he helped me

v

to form over years of his happy hours and conference dinners, is another great gift

of my graduate experience. Tony Mughan provided constant feedback on all aspects

of the project, and my thinking on party politics and partisanship in particular are

much improved because of his input. Finally, Kathleen McGraw played a particularly

important role in helping me sharpen my thinking on the psychological mechanisms

that make up the crux of this research. Thank you to all four of you — your advice

and friendship are a constant thread throughout this document and my career.

More broadly, I am greatly indebted to the support of the Department of Political

Science at the Ohio State University. I always knew that Department Chair Richard

Herrmann and Director of Graduate Studies Bear Braumoeller were in my corner,

and Graduate Program Coordinator Courtney Sanders was always available to help

with her truly unbelievable ability to deal with any concern with a smile. This

work also greatly benefited from the intellectually stimulating environment of the

department, and the Comparative Politics Workshop and Psychology Workshop in

particular were integral to testing out new ideas and turning them into research.

Thank you in particular to Amanda Robinson and Tom Nelson, who lent their time

to these respective workshops and went above and beyond their duties to provide

constant and invaluable feedback.

The most rewarding element of conducting dissertation research was the wonder-

ful colleagues I had the opportunity to work with and befriend along the way. Special

thanks to Chelsea Ihle, Meri-Ellen Lynott, Caronlina Foresman, Matt Hitt, Vittorio

Merola, John Elliott, Lauren Elliott-Dorans, Alex Castillo, Peter Tunkis, Carolyn

Morgan, Margaret Hanson, and Calla Hummel. You are my colleagues, comrades

in arms, and now, close friends that I am lucky to have. My friend and co-author

vi

Jason Morgan deserves special recognition for constant conversation and unwavering

patience in helping me work through methodological and statistical concerns. Fi-

nally, thank you to Jessy Defenderfer, my closest confidant and biggest cheerleader

throughout this seven year process. Your friendship is the reason I’m emerging from

this process (somewhat) sane.

The people of Hungary deserve special thanks. From the specific friends and

colleagues that made this work possible, such as the invaluable aide of Kitti and

Zoli Fekete-Kovats, Nora Lantos, and Levente Littvay, to the numerous participants

in my various studies, to the critical role of Central European and E otv os Lorand

Universities in hosting me during fieldwork, this research is greatly indebted to you.

More fundamentally, it was talking to Hungarians over many years that I learned of

the ideas explained below. They came up with the hypotheses, I simply tested them.

vii

Vita

August 5, 1984 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Born — Fairfax, VA, USA

2006 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .B.A. in Government & Philosophy,College of William and Mary

2011 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .M.A. in Political Science, The OhioState University

2011-present . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Graduate Teaching Associate,The Ohio State University

Fields of Study

Major Field: Political Science

Studies in:

Comparative Politics, Political Psychology

viii

ix

Table of Contents

Page

Abstract . . . . . .

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

ii

Dedication . . . . .

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

iv

Acknowledgments .

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

v

Vita . . . . . . . .

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

viii

List of Tables . . .

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

xii

List of Figures . .

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

xiii

Chapter 1: Introduction — The Politics of Outrage . . . . . . . . . . . . . . . . 1

1.1 Playing with Fire: Emotional Entrepreneurship and

the Uses and Abuses of Outrage . . . . . . . . . . . . . . . . . . . . 4

1.2 Dissertation Overview . . . . . . . . . . . . . . . . . . . . . . . . . 8

1.2.1 Chapter Outline . . . . . . . . . . . . . . . . . . . . . . . . 8

Chapter 2: Emotions, Anti-Politics, and Democratization as the Third Wave Ebbs . 13

2.1 Post-Communist Political Development . . . . . . . . . . . . . . . . 15

2.2 Populist Outrage in the Political Dialogue . . . . . . . . . . . . . . 17

2.2.1 International Constraint . . . . . . . . . . . . . . . . . . . . 21

2.2.2 Party Competition in the Information Age . . . . . . . . . . 23

2.3 The Explanatory Power of Emotion . . . . . . . . . . . . . . . . . . 25

2.4 Anger and Party System Volatility . . . . . . . . . . . . . . . . . . 27

2.5 Disgust and Dedication to Democratic Institutions . . . . . . . . . 30

2.6 The Distinctiveness of Disgust . . . . . . . . . . . . . . . . . . . . . 34

2.7 Disgust or Anger?

The Roles of Partisanship and Time . . . . . . . . . . . . . . . . . 37

x

2.8 Hypotheses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40

2.9 Alternative hypotheses . . . . . . . . . . . . . . . . . . . . . . . . . 43

2.9.1 Economic Strife . . . . . . . . . . . . . . . . . . . . . . . . . 43

2.9.2 Critical citizens and discerning democrats . . . . . . . . . . 45

Chapter 3: The Prevalence of Populist Outrage in East Central Europe . . . . . 48

3.1 Media Technology and Attack Politics . . . . . . . . . . . . . . . . 50

3.2 Populist Outrage . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52

3.3 Populist Outrage in ECE . . . . . . . . . . . . . . . . . . . . . . . 56

3.4 Analysis: Mapping Populist Outrage . . . . . . . . . . . . . . . . . 60

3.4.1 Outrage Dialogue in the Hungarian Political Discourse . . . 65

3.4.2 Adaptation of the ANEW Dictionary . . . . . . . . . . . . . 69

3.4.3 Populism in Party Appeals . . . . . . . . . . . . . . . . . . 73

3.5 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81

Chapter 4: Immediate Emotional Reactions to Outrage: Anger, Disgust, and the

Role of Partisanship . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83

4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83

4.2 Anger and Disgust towards Politics . . . . . . . . . . . . . . . . . . 85

4.3 Survey: Democratic Commitment, Anger, and Disgust among the

Hungarian Public . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86

4.4 Experimental evidence . . . . . . . . . . . . . . . . . . . . . . . . . 89

4.4.1 Pilot Study . . . . . . . . . . . . . . . . . . . . . . . . . . . 89

4.4.2 Emotional Reactions . . . . . . . . . . . . . . . . . . . . . . 92

4.5 Lab-in-the-field Study . . . . . . . . . . . . . . . . . . . . . . . . . 98

4.5.1 Emotional Reactions . . . . . . . . . . . . . . . . . . . . . . 100

4.6 The Role of Partisanship . . . . . . . . . . . . . . . . . . . . . . . . 103

4.7 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106

Chapter 5: Outrage Exhaustion: How Politics Become Disgusting . . . . . . . . 109

5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109

5.2 From Anger to Disgust . . . . . . . . . . . . . . . . . . . . . . . . . 110

5.3 Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115

5.4 Repetition Hypothesis . . . . . . . . . . . . . . . . . . . . . . . . . 120

5.5 The Age Hypothesis . . . . . . . . . . . . . . . . . . . . . . . . . . 125

5.6 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135

xi

6.1 Political Behavior and the Adaptive Functions of Anger and Disgust 143

6.2 The Emotional Roots of Backsliding . . . . . . . . . . . . . . . . . 146

Chapter 6: Disentangling the Political Implications of Disgust from Anger . . . 140

6.3 Evidence from Experiments . . . . . . . . . . . . . . . . . . . . . . 150

6.4 Qualitative Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160

6.5 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161

Chapter 7: Implications and Future Directions . . . . . . . . . . . . . . . . . . . 164

7.1 Research Goals, Methods, and Findings . . . . . . . . . . . . . . . 164

7.2 Limitations and Future Research . . . . . . . . . . . . . . . . . . . 168

7.2.1 Empirical Refinement . . . . . . . . . . . . . . . . . . . . . 169

7.2.2 Of Time, Outrage, and Partisan Predilections . . . . . . . . 172

7.2.3 The Co-occurrence of Disgust and Anger . . . . . . . . . . . 174

7.3 Implications for Scholars and Policymakers . . . . . . . . . . . . . . 177

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183

Appendices 196

A. Content Analysis Supplementary Materials . . . . . . . . . . . . . . . . . 196

A.1 The Seven Sins of the Gyurcsany Coalition Pamphlet . . . . . . . . 196

A.2 Manifestos Data Scales . . . . . . . . . . . . . . . . . . . . . . . . . 198

A.3 Economic Populism Results . . . . . . . . . . . . . . . . . . . . . . 199

A.4 Nationalistic Populism . . . . . . . . . . . . . . . . . . . . . . . . . 204

A.5 Inductive Outrage Dictionary . . . . . . . . . . . . . . . . . . . . . 211

A.6 ANEW Outrage dictionary . . . . . . . . . . . . . . . . . . . . . . 215

B. Experiment and Survey Supplementary Materials . . . . . . . . . . . . . 222

List of Tables

Table Page

4.1 Negative Emotions Towards Politics . . . . . . . . . . . . . . . . . . . 88

4.2 Self-reported outrage following manipulation . . . . . . . . . . . . . . 96

4.3 Strength of Ideology and Emotional Reactions to Outrages (Experiment)104

4.4 Strength of Ideology and Emotional Reactions to Outrages (Survey) . 106

5.1 Focus Group Composition . . . . . . . . . . . . . . . . . . . . . . . . 118

6.1 Anger, Disgust, and Attitudes towards Democracy (OLS models) . . 149

6.2 Ordered Logit Models: Effects of Anger and Disgust on Political Atti-tudes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153

6.3 Moral anger and perceptions of government . . . . . . . . . . . . . . 155

6.4 Emotions driving affect towards the governing party . . . . . . . . . . 158

B.1 Summary Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226

B.2 Summary Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228

B.3 Anger, Disgust, and Attitudes towards Democracy . . . . . . . . . . . 229

B.4 Anger, Disgust, and Attitudes towards Democracy (with Extreme) . . 230

xii

List of Figures

Figure Page

1.1 The Two-Tailed Dog Party . . . . . . . . . . . . . . . . . . . . . . . . 2

2.1 Overview of the theory . . . . . . . . . . . . . . . . . . . . . . . . . . 42

3.1 Frequency of the term “outrage” in domestic political coverage fromthe Hungarian News Agency . . . . . . . . . . . . . . . . . . . . . . . 63

3.2 Proportion of ANEW outrage words out of total per election year . . 67

3.3 Percentage change in ANEW outrage words from previous election year 68

3.4 Proportion of ANEW outrage words out of total per election year . . 72

3.5 Percentage change in ANEW outrage words from previous election year 72

3.6 Overall Populism in Hungary . . . . . . . . . . . . . . . . . . . . . . 75

3.7 Overall Populism in the Czech Republic . . . . . . . . . . . . . . . . 76

3.8 Overall Populism in Poland . . . . . . . . . . . . . . . . . . . . . . . 77

3.9 Overall Populism in the Slovak Republic . . . . . . . . . . . . . . . . 78

3.10 Nationalistic / Anti-EU Populism in Hungary . . . . . . . . . . . . . 80

4.1 Emotional Reactions by Group in the Pilot . . . . . . . . . . . . . . . 93

4.2 Study 2 emotional reactions . . . . . . . . . . . . . . . . . . . . . . . 102

5.1 Disgust & Anger by Age Cohort . . . . . . . . . . . . . . . . . . . . . 133

xiii

5.2 Anxiety and Apathy by Age Cohort . . . . . . . . . . . . . . . . . . . 134

6.1 Democracy is the best political system for a country like ours . . . . 147

6.2 Affect toward pertinent political actors and parties . . . . . . . . . . 156

A.1 The first page of the “7 Sins of the Gyurcsany Coalition” . . . . . . . 196

A.2 Economic Populism in Hungary . . . . . . . . . . . . . . . . . . . . . 200

A.3 Economic Populism in the Czech Republic . . . . . . . . . . . . . . . 201

A.4 Economic Populism in Poland . . . . . . . . . . . . . . . . . . . . . . 202

A.5 Economic Populism in the Slovak Republic . . . . . . . . . . . . . . . 203

A.6 Nationalistic / Anti-EU Populism in the Czech Republic . . . . . . . 204

A.7 Nationalistic / Anti-EU Populism in Poland . . . . . . . . . . . . . . 205

A.8 Nationalistic / Anti-EU Populism in the Slovak Republic . . . . . . . 206

A.9 Anti-Communist/Corruption Populism in Hungary . . . . . . . . . . 207

A.10 Anti-Communist/Corruption Populism in the Czech Republic . . . . 208

A.11 Anti-Communist/Corruption Populism in Poland . . . . . . . . . . . 209

A.12 Anti-Communist/Corruption Populism in the Slovak Republic . . . . 210

xiv

Chapter 1: Introduction — The Politics of Outrage

“We should free ourselves from politics as if from some contagious infection”

– Gyorgy Konrad, Hungarian Political Theorist

The Hallmarks of a Disgusted Public

“Oh I’m going [to vote]. This year I’m either going to write [bad word for male

anatomy] or [worse word for female anatomy] on the ballot, but I haven’t chosen

which yet.” The interviewee laughed, but grew solemn as he confided that he hates

to do this and once took voting very seriously. Though he was not so old to have

voted in sham elections under communism, he remembered the lack of freedom during

his childhood. “But they’re all terrible,” he says forlornly of the options for the 2014

Hungarian parliamentary elections. Defacing the ballot seemed to him to be the best

option, especially as part of a movement of voters voicing their protest in this way —

“there were over 20,000 of us last time, this time there will be many more!”1

Though the Hungarian Election Committee does not release statistics on defaced

ballots, the intention to deface the ballot rather than voting for any political party

was widespread in discussions of Hungary’s 2014 elections. Indeed, a satirical political

party and movement was founded around the idea: the Two-Tailed Dog Movement.

1Interview with male in his 40s, Budapest, January 2014

1

The faux party garnered hundreds of thousands of followers on Facebook, and con-

ducted a full satirical campaign for the parliament, promising two sunsets a day and

free beer for all. At campaign events they would give up cups of water labeled “beer.”

The reference to politics is clear: politicians feed the public lies and leave people with

a raw deal. Figure 1 1 shows one of their campaign advertisements advocating defac-

ing the ballot.

Figure 1.1: The Two-Tailed Dog Party

If elections are meant to confer legitimacy and public control of government, this

is a sure sign of a failure for this to occur in Hungary. This dramatic exit from

politics is a symptom of a broad disdain for politics that is obvious in Hungary and

many other democracies around the world. Polls in Hungary and across the post-

communist world document deep dissatisfaction with democracy and a persistent,

in some respects growing, nostalgia for the past regime. Rates of voter turnout

are on the decline. Hungary and the other countries of East Central Europe that

2

seemed to be front-runners of democratization in the third wave of democratization

are experiencing many symptoms of backsliding. Party systems are volatile and highly

polarized, while public support for the very ideal of democracy is eroding, even in cases

where leaders tighten their grip on power through undemocratic means. These trends

are particularly troubling in the case of Hungary, where popular commitment to and

engagement in democracy is on the decline as the country’s leadership systematically

deconstructs the institutions of liberal democracy. Democracy, a system premised

on representing the people, and dependent upon their support, is jeopardized in the

absence of popular support and participation. Recent years in Hungary provide a

stark case and point of this truism.

Showing up to the polls only to deface a ballot is a confusing proposition from

the perspective of a rational actor. Rational accounts of voting balance the costs of

turning out to vote — gathering enough information to make an informed choice,

physically getting to the polls — with the potential benefits — the increased prob-

ability of the favored party gaining power and governing according to the actor’s

desires and interests. Writing obscenities on the ballot entails the costs of voting

with none of the benefits. It is not a rational act. Nor is it a sign of apathy; a lack of

feelings towards politics could not inspire such an action. Instead, this, like so many

behaviors integral to politics, is a profoundly emotional act and, as such, can only

be understood by attending to the specific emotional response at play. In particular,

this act is entirely consistent with a citizen experiencing a combination of intense

anger and disgust when it comes to their political systems, compelled by anger to do

something, but by disgust to make that something a rejection of the entire system.

3

What is the source of these emotions? This dissertation focuses upon the effects

of a political dialogue consisting disproportionately of outlandish accusations of im-

morality and baseness on the part of elected officials — what I will label as populist

outrage. Politicians work to inspire political action by eliciting anger at their com-

petitors, largely by documenting supposed or actual outrageous offenses they have

committed. Some citizens do become angry because of these outrages, leading them

to take political action (Marcus et al., 2000), often through the protest voting for

unorthodox parties driving party system instability across the region (Pop-Eleches,

2010). However, among many citizens, this stream of populist outrage elicits disgust

towards the entire political system instead of anger towards any particular group

among many. In this dissertation I argue that this political disgust is at the root of

the intense aversion towards politics that is widespread across Hungarian society, and

causes a decrease in both participation in and normative commitment to democracy. I

provide evidence that the powerful emotions of anger and disgust, and the distinction

between them, explains widespread animosity and disengagement on the one hand,

and continued electoral volatility and radicalization in the party system on the other.

1.1 Playing with Fire: Emotional Entrepreneurship andthe Uses and Abuses of Outrage

Political leaders in democratic states are faced with the problem of mobilizing

citizens into voters. Scholars have long known that purpose of most political speech

from parties is less to inform and more to mobilize; as Ostrogorski (1902) pointed out

over a century ago, for the average campaign communication, the “sole object is to

besprinkle the audience with the magnetic party fluid” (391). That is why politicians

often turn to emotions, drumming up enthusiasm, anger, pride, or fear to attract

4

support and participation. Indeed, one of the most important roles of emotion in

politics is to motivate citizens to shake free of apathy and engage with their political

systems. As Carl Jung poignantly stated: “There can be no transforming of darkness

into light or of apathy into movement without emotion” (1938, 32).

This often presents a challenge, as the average citizen is largely apathetic toward

politics and would rather spend her time and energy attending to her more tangible

and immediate obligations and interests (Lippmann 1922; Rosenberg 1954). As both

scholars note, this apathy is relatively rational for the average voter, who knows

her single vote or small donation will not make any difference whatsoever in the

election. However, where this voter begins to feel strongly about a political issue

or candidate, she becomes motivated to vote and perhaps even to engage in some

more involved form of political participation. Thus, candidates will attempt to use

emotional appeals to “fire up” their bases and to convince the unaligned to care and

to support them at the polls. A growing body of work shows that emotions such as

enthusiasm, anxiety, and aversion can powerfully predict various forms of political

participation and information seeking (Marcus et al., 2000; Brader, 2005; MacKuen

et al., 2010; Best and Krueger, 2011; Valentino et al., 2008, 2009, 2011). These works

often view the power of emotions to motivate citizens into political action as a positive

phenomenon. It is easy to agree with this stance, for it seems that any force that

encourages the uninformed masses to search for political information or motivates an

indifferent public to care and act.

Anger is a particularly paradoxical emotion. It is the root of both aggression and

justice seeking, of righteous indignation that leads to closed-minds as well as acts of

selflessness, and of some actions that bring about social progress and others that harm

marginalized groups. Still, much of the work on anger in general, and on moral anger

(or outrage) in particular, paints it in a positive light. Psychologists have applied the

5

concept of anger to the study of political situations, most typically analyzing its effects

upon willingness to participate in politics to fix some morally offensive situation. In

the earliest example of this work, Montada and Schneider (1989) found that moral

anger over societal inequity encourages the advantaged in a society to advocate on the

part of the disadvantaged. More recently, Lodewijkx and his colleagues (2008) found

this response to be strong enough to motivate people to protest. Another study shows

how people can get angry at other groups taking part in violence, and inspire support

collective conciliation (Tagor, Federico, and Halperin 2010). Thomas and McGarty

(2009) found that interactions between groups in developed and developing countries

could increase feelings of efficacy and moral outrage and thereby spur respondents in

developed countries into activism in the name of the disadvantaged abroad. Overall,

it is clear that anger at injustice and “other emotions of outrage” play a key role in

collective action to fix the injustice (van Zomeren, Postones, and Spers 2008), and

are an even better explanation for collective action in these situations than positive

emotions (Becker et al 2011).

In addition to inspiring action, another critical function of emotions in politics is

to bind groups together. Politicians need to unite people based upon shared identities

and goals to attract political support, and emotions particularly those that evolved

to govern human morality, play a critical role in regulating behavior both within and

between groups (Haidt, 2003). McDermott (2010) identifies a process she defines as

emotional entrepreneurship: “the process by which leaders strategically use outrage at

opposition members to cue in-group members to participate in action against the out-

group members who have committed the outrage” (114). Thus, leveraging outrages

can elicit moral anger, binding groups together in opposition to a group that has

committed an egregious trespass and must be punished.

6

The process of using emotional appeals to elicit action and bind people together

into groups will have both positive and negative consequences. Just as an outrage

can turn inaction into the passion to right a wrong, it can cause a misfire with very

undesirable consequences. Indeed, citizens can be “fanned to righteous indignation

for a cause eventually revealed as unworthy” (Giner-Sorolla, 2012, p. 3). Mcdermott

focuses one way in which this emotional entrepreneurship may be damaging to the

public good; where politicians invent outrages to divide groups that might actually

have interests in working together for their own selfish ends, the greater good suffers at

the hands of the private interests of these political elites. Here I will focus on another

dangerous reaction to elite targeting of moral anger: the elicitation of disgust.

While most emotions, including anger, can have pro-social consequences, disgust

lacks any socially beneficial action tendency. Disgust should have no bearing upon

morality (Kelly, 2011), and likely only hurts democratic governance. It is worse than

apathy, which can give way easily to a new emotion. Once disgust is felt towards an

object, however, it is pervasive and enduring. Political elites must therefore walk a

fine line between moving an apathetic populace to action with the use of appeals to

negative emotions and overusing these appeals such that they instead elicit disgust,

either towards politics writ large or towards others within the polity.

The situation in Hungary is one in which the strategy of “emotional entreprenuer-

ship” (McDermott, 2010) has fundamentally backfired. The citizens scrawling the

worst words in the Hungarian language across their ballots have been pushed over

the edge by their political systems. This represents a tragedy of the commons for the

political elites involved. A single instance of leveraging a major outrage against politi-

cal opponents may be very profitable in the short-term. However, if all in the political

arena resort to this tactic, the public becomes disgusted, and this disgust represents

a fundamental barrier to norms of participation and tolerance at the heart of the

7

democratic ideal. In this dissertation I will argue that the elite strategy of targeting

moral anger to inspire passion and action has backfired and caused widespread disgust

towards politics and groups within society, helping to explain Hungary’s democratic

backsliding and its international infamy during the contemporary refugee crisis.

1.2 Dissertation Overview

This dissertation’s main goal is to analyze Hungarian citizens’ emotional reactions

to their political systems in order to better understand how citizens in new democra-

cies come to relate to politics. It is rooted in the tradition pointing to public attitudes

towards politics, or political culture, as a key determinant for the success or failure of

stable democratic politics (De Tocqueville, 2004; Almond and Verba, 1963; Putnam,

1994), and is novel in basing the study of mass orientations towards the polity in the

psychology of emotional response. Indeed, in order to better support the development

of stable democratic polities and the political cultures that reinforce them, we require

advancements to our understanding of the way citizens feel towards politics, why they

feel the way they do, and how these emotions evolve over time to shape long-term

political orientations.

1.2.1 Chapter Outline

Chapter two, “Emotion, Anti-Politics, and Democratization as the Third Wave

Ebbs,” further details the political landscape in Hungary and East Central Europe

and elucidates the elite and media incentives for resorting to a dialogue of populist

outrage. It then presents a series of hypotheses concerning the effects this steady

stream of accusations upon political behavior, linking them to broader trends in the

region. I argue that a common initial reaction will be with moral anger, which can

8

cause polarization and inspire protest voting driving party system realignments. I

distinguish why moral anger is a prominent force in political communications in gen-

eral and in the political dialogue of post-communist Europe in particular. I then draw

on psychological theories of emotions to argue that the stream of political vitriol will

instead produce disgust among many citizens as a function of two factors. First, the

presence and strength of partisan predispositions will predict anger. This is because of

the integral role of attributions of blame to anger, and the increased likelihood that

partisans attribute blame upon witnessing an outrage. Second, constant repeated

exposure to political outrages will stop producing anger in many citizens over time

and instead transition into disgust. Disgust is associated with very different action

tendencies from anger namely, a desire to create as much distance from the offend-

ing situation as possible. Thus, those who feel disgust towards politics will become

politically disengaged.

The third chapter, “The Prevalence of Populist Outrage,” focuses upon the disser-

tation’s main independent variable: populist outrage in political coverage over time.

Numerous scholars document a negative turn in political coverage in general (Geer,

2012), and a rise in outrage in particular (Berry and Sobieraj, 2013). This work

mainly focuses on the US, and little to nothing exists concerning the young democra-

cies of ECE. This is an unfortunate lapse as there are reasons to believe that the turn

for the negative is even stronger in these cases, where the pressures of globalization

and a changing media environment occur in the contexts of weak parties and atomized

civil societies. In particular, I argue that outrageous rhetoric and coverage increased

in the last decade following increased constraints from the European Union and In-

ternational Monetary Fund. To test this I draw upon two sources. First, I compile

an original data set of coverage of Hungarian parliamentary elections from 1994 to

2014 in the Hungarian News Agency. I apply two different dictionaries of populist

9

outrage to these data to document a sharp increase in populist outrage in Hungarian

political coverage that coincides with European Union accession and membership.

Next, I compute several indexes of policy appeals versus populist appeals using the

Comparative Manifestos Data for the Visegrad Four countries: Hungary, Poland, the

Czech Republic, and Slovakia. These measures document a rise in populism in these

countries over the past 15 years.

The fourth chapter, “Immediate Emotional Reactions to Outrage: Anger, Dis-

gust, and the Role of Partisanship,” draws upon multiple sources of original data to

document Hungarian citizens’ reactions to political vitriol and attack style politics,

and then demonstrates that the strength and presence of partisan predilections is

a key determinant of whether a citizen feels anger or disgust in response to these

appeals. Results from an original survey of 1,000 Hungarian citizens show the pres-

ence of high levels of anger and disgust towards politics in general, and to the much

maligned outrages in Hungarian political life in particular. Together with an online

pilot, a lab-in-the-field experiment ties these emotional effects to the political out-

rages. These survey and experimental data demonstrate the importance of partisan

predilections in predicting anger, raising important questions regarding the role and

origin of partisanship in regulating emotional responses.

Chapter five, entitled “Outrage Exhaustion: Disgusting, contemptuous, or hope-

less politics?” turns to the long-term effects of appeals to outrage. Psychological

research suggests that, over time, unresolved anger gives way to the emotions of dis-

gust and contempt. Thus, there is a key distinction between the initial moral anger

felt towards a situation – “outrage motivation” — and the “outrage exhaustion” that

develops over time. The likelihood of this transition to outrage exhaustion depends on

attributes of both the accusation and the message receiver. Regarding the accusation,

allegations of the same outrageous misdeed or class of wrongdoing eventually provoke

10

disgust in lieu of moral anger. When it comes to the message receiver, there are likely

age effects depending on the nature of the political dialogue during the key time of

political socialization. In other words, different cohorts have been “treated” with

different levels of outrage in the political dialogue, such that older citizens are more

likely to be disgusted with politics because of greater exposure to outrage in politics.

Looking at an emotional reaction’s evolution over time is particularly challenging em-

pirically, necessitating multi-methods research. First, data from the European Social

Survey (ESS) show distinct cohort effects in political attitudes and efficacy. Next,

data from four focus groups provide another source of information, as the groups were

divided by age cohorts and the dialogue among the older participants was suggestive

of outrage exhaustion and the associated disgust. Finally, interviews also provide

illustrative evidence. Thus, the qualitative data supplement the ESS results, lending

credence to the argument that repeated anger at various political outrages over time

accounts for differences in political attitudes across age cohorts.

Having shown the rise of political outrage in Hungary and attended to the emo-

tional effects of these outrages as a function of both time and attributes of the citizen,

the sixth chapter — “Disentangling the Political Implications of Disgust from Anger”

— documents the distinct effects these emotions on political behavior and public opin-

ion. In doing so, it grapples with the distinction between these emotions, a continuing

area of exploration in the psychology literature on emotions. I show that, while anger

and disgust co-occur with great frequency, their distinct and contradictory effects

have important consequences in politics. Evidence from the survey and experiments

shows that anger is associated with an impulse to participate while disgust leads to

disengagement. Moreover, a number of attitudinal differences are explored. These

differences are in large part associated with the distinct appraisals that predict anger,

with the attitudinal consequences of anger having to do with blame and intentionality

11

and the consequences of disgust interacting with perceptions of hopelessness. Thus,

anger predicts less satisfaction with particular actors and policies that can lead to

acting to correct these perceived wrongs. Disgust, however, is associated with over-

all low feelings of efficacy and a desire to detach from politics. The most troubling

finding is that disgusted citizens are less likely to feel democracy is the best form of

government for their country, helping to explain widespread popular quiescence in

light of the Hungarian government’s increasingly blatant authoritarian bent.

12

Chapter 2: Emotions, Anti-Politics, and Democratization as

the Third Wave Ebbs

On July 26, 2014, Hungary’s newly reelected Prime Minister, Viktor Orban, de-

livered a speech in which he openly disavowed liberal democracy, promising to “build

a new illiberal state based upon national foundations.”1 Though Orban’s statement

followed years of systematic assault on democratic institutions by his government,

Hungarian society met it with a muted shrug. The lack of public outcry is more wor-

rying than the proclamation itself in this forerunner of post-communist democratiza-

tion, where growing disengagement coincides with drastic party system realignment,

creeping trends of xenophobia and extremism, and utter disdain for the process of

competitive party politics.

Why would citizens in one of post-communist Europe’s most successful cases of

democratization react to overt attacks on democracy with apathy and disinterest?

Hungary is a member of a global cohort of fragile young democracies playing host to

the dual trend of democratically elected leaders tightening their grip on power just as

the populace turns away from the ideal of democracy. Indeed, such democratic back-

sliding is a widespread but understudied phenomenon (Bermeo, 2016). At the onset

of the transition in post-communist Europe, scholars worried that democracy would

1“Egy illiberalis, nemzeti alapokon allo uj allamot folepıteni.” Hungarian origi-nal text at:http://www.kormany.hu/hu/a-miniszterelnok/beszedek-publikaciok-interjuk/a-munkaalapu-allam-korszaka-kovetkezik. English language coverage at:http://www.bloomberg.com/news/2014-07-28/orban-says-he-seeks-to-end-liberal-

democracy-in-hungary.html

13

fail to take root in the region (Jowitt, 1992; Przeworski, 1991), but their concerns that

a lack of popular support would undermine the new system are even more warranted

today, 25 years after the fall of the Berlin Wall. Though democratic institutions are

entrenched in form, public support wanes, as does political participation; turnout has

fallen in every Hungarian election since 2002. More and more, citizens in the region

are alienated from the democratic system, suffering from what Rupnik (2014) calls

“democracy fatigue.” Why would experience with democracy fail to yield support for

the ideals and norms of democracy?

The recent threats to democracy coincide with another troubling phenomenon:

the drastic realignment of the Hungarian party system. Fidesz’s 2010 victory under

Orban’s leadership was remarkable. The party won an absolute majority in votes, and

because of the majoritarian nature of Hungary’s mixed electoral system, surpassed

the two-thirds majority in Parliament required to rewrite the constitution. Two of

the most influential parties of post-communist politics in Hungary — the Hungarian

Democratic Forum (MDF) and the Free Democrats (SzDSz) — failed to surpass the

electoral threshold and were absent from the parliament for the first time since the end

of communist rule. These centrist parties were replaced by two more extreme entrants

to the party system: the green-libertarian “Politics Can Be Different” (LMP) and

Jobbik, a strident radical right party that won a large chunk of the vote. Meanwhile,

the incumbent Hungarian Socialist Party (MSZP) won a paltry proportion of the

vote, ending the pattern of bipolar political competition between Fidesz and MSZP

that had dominated the country for over a decade. Jobbik nearly surpassed MSZP in

vote share, and the Socialists’ defeat led to the fracturing of the left into several weak,

new parties in fierce competition with one another for an ever-dwindling number of

leftist supporters. This instability is not unique to Hungary, which until recently

was considered to be the most stable party system in the region (Olson, 1998; Sikk,

14

2005; Bertoa and Mair, 2010). What explains the recent unraveling of the Hungarian

party system, and what can developments in Hungary tell us about the persistent

instability that characterizes party politics in ECE?

The answer to each of the above questions lies in the particular nature of Hungar-

ians’ democratic experience: a never-ending stream of vitriol in which all sides accuse

political rivals of outrageous offenses. These attacks aim at eliciting anger to drive

political action. But, in reality, a discourse that deals primarily in outrage elicits

disgust rather than anger for many citizens. Disgust leads to a disengagement from

politics so thorough as to cause a withdrawal from the very ideal of democracy. Thus,

when citizens become disgusted with politics, they are unlikely to take action to pro-

tect a system they no longer admire in the first place. Anger, on the other hand, leads

to greater political engagement (Brader, 2006; Marcus et al., 2000) and helps explain

instability by driving the widespread protest voting for unorthodox parties noted in

the region (Pop-Eleches, 2010). The distinction between anger – an emotion often

studied by scholars of political behavior – and disgust – an emotion largely ignored

in political science – illuminates the situation in Hungary and many other faltering

young democracies.

2.1 Post-Communist Political Development

Post-communist democracies defied all but the most optimistic expectations dur-

ing the initial years of transition. Writing for a college textbook on the region,

Valerie Bunce notes that state socialist dictatorships “seemed to be committed to

the destruction of some of the most elementary building blocks of democratic life”

(Bunce, 2010, 37). The simultaneous transition from the authoritarian party state to

pluralistic democracy on the one hand and from centrally planned economics to the

free market on the other seemed likely to be at odds with one another, and scholars

15

and policymakers alike feared that the necessary yet painful economic reforms would

cause an abandonment of the new democratic institutions (Przeworski, 1991). Others

worried that the social pressures unleashed by liberalization in multi-ethnic societies,

where many had been active or complicit in the atrocities of the former regime, would

cause great fragmentation and violence. Indeed, Ken Jowitt argued that some sort of

“liberal authoritarian” middle-ground between the old Leninist regimes and liberal

democracy would be favorable to the “religio-ethnic, militant nationalist, even fascist

regimes that might emerge from the maelstrom,” or at least would be “a more practi-

cal response than the Utopian wish for mass democracy in Eastern Europe” (Jowitt,

1992, 223).

By and large, however, this Utopian wish came to pass. The states of ECE moved

rapidly towards liberal democracy and between 2004 and 2011 most of these young

democracies joined the European Union. Still, despite the unmistakable establish-

ment of democracy in ECE, these states face serious challenges to the quality and

sustainability of their young democratic institutions. Many of these ailments are

shared with older advanced democracies. Widespread dealignment as parties shed

broad membership bases and loyal voters (Dalton and Wattenberg, 2000), crippling

polarization, the rise of the radical right (Mudde, 2007), and growing cynicism and

disaffection (Mair, 2013; Hay, 2007) afflict democracies globally. However, when it

comes to party system instability and the vulnerability of public support for demo-

cratic institutions, the situation in Eastern Europe is much more severe.

The vote-switching that drives party system instability and atrophied voter par-

ticipation in and affect towards the democratic system are each elements of the wider

trend of dealignment driven by dissatisfaction with both specific political parties and

the entire political system (Dassonneville et al., 2015). Accounts of both trends as

16

they stand now are incomplete, and a fuller explanation requires understanding citi-

zens’ orientations towards politics in their country. The present analysis begins with

an exploration of the type of political information voters in Hungary and other new

democracies of ECE experience. As Moehler (2008) notes, “if we want to understand

the effect of mass action on attitudes, we must pay attention to the content of the

messages to which participants are exposed as well as the means they use to interpret

those messages” (7). I argue that an onslaught of vituperative, bellicose political dis-

course drives citizens in ECE toward uniquely high levels of anger and disgust when it

comes to politics. Indeed, the political arena is dominated by populist appeals to out-

rage, consisting of accusations of egregious wrongdoings by all contestants in politics.

While politicians wish to use these outrages to elicit anger and drive action, for many

citizens the emotional response is instead disgust. Whereas anger helps understand

protest voting and party system instability, disgust accounts for the disengagement

from politics undermining support for democracy.

2.2 Populist Outrage in the Political Dialogue

Viktor Orban’s Fidesz party swept into power in 2010 by leveraging populist-

nationalist rhetoric against the incumbent Hungarian Socialist Party. Orban painted

the socialist successor party as an unreformed and unrepentant cadre of crooked com-

munist oligarchs who were rotten in every way.2 He argued that his government’s 2012

constitution was the true revolution away from communism, though political experts

view it as an illiberal document that places restrictions upon the freedom of the judi-

ciary and the media, while also providing easy avenues to future electoral victories for

2That the (ousted) former Socialist Prime Minister — the so-called “lying Prime Minister” —had been caught on tape admitting his party leadership had lied “morning, noon, and night” aboutthe state of the economy to win their second term in office certainly did not help. See Chapter 5 fora full explanation of this scandal and its effects upon politics in Hungary.

17

Fidesz through new voting procedures and gerrymandering. Meanwhile, the radical

right party, Jobbik, found a winning strategy with their own virulent brand of outrage

aimed at Fidesz and the rest of the political establishment, equating the country’s

political elites with global networks of American-Israeli conspirators out to rob Hun-

gary of its sovereignty. These political antics and their widespread coverage are by

no means unique to Hungary, as scandals and outrages constitute the ammunition

of political battles across post-communist Europe. This outrage forms the backbone

of an endless stream of populism that has brought many politicians in the region to

power and destroyed the careers of many others. Ken Jowitt’s (1992) conjecture that

politics in post-communist democracies would be dominated by demagogues rather

than democrats looks to be more true today than during the immediate aftermath of

the transition.

As will be discussed in the following chapter, there is evidence that broad struc-

tural and technological trends are leading to a global increase in attacks and accu-

sations in political news. However, a key contention of this dissertation is that ECE

is remarkable for its negative political dialogue, and that one type of political nega-

tivity — populist outrage — is especially pernicious to democratic politics in these

young democracies. Populist outrage is defined here as a political strategy of attract-

ing support and distinguishing oneself from political opponents by identifying them

as transgressors who have committed some tremendous, blatant atrocity against the

good, everyday people, with whom the speaker identifies. As populism involves pit-

ting a virtuous, homogeneous “people” against some morally-rotten villain (Canovan,

2002), this political strategy is central to the concept. Outrage, or moral anger, is

defined by psychologists as “anger provoked by the perception that a moral standard

— usually a standard of fairness or justice — has been violated” (Batson et al., 2007,

1,272). It is viewed as one of the moral emotions tasked with the evolutionary role of

18

regulating behavior in groups (Haidt, 2003). While a debate over the distinctiveness

of outrage, or moral anger, and the more well-studied personal anger continues in

psychology, here I remain agnostic as to whether outrage constitutes a distinct emo-

tion from personal anger, as the causes and consequences of each type of anger will

be the same regardless. Instead, the focus of this dissertation is on the use of outrage

as a strategy of political discourse, or the appeal to anger through a specific type of

populist appeal involving a violation of morality, fairness, or justice. Therefore, here I

use the term populist outrage to denote this study’s primary independent variable, as

it captures both the “us versus them” nature of this dialogue as well as the presence

of the egregious moral outrage that separates the “us” from the “them.”

Populist outrage is a growing strategy in democracies everywhere, and a central

method of political appeal within the dialogue of post-communist politics in partic-

ular. It is also central to many nationalist appeals; the common accusation that a

minority group receives scarce state resources from hard-working citizens, an immi-

grant group is out to undermine domestic security or the national way of life, or that

multi-national firms scheme to steal the country’s wealth and sovereignty are prime

examples. This flexible strategy is also a central component of anti-incumbent or

anti-establishment rhetoric. When something is wrong with politics, the accuser can

gain political support by eliciting anger about the issue against those in power with

populist outrage, scapegoating anyone from elected officials to bureaucrats, whether

or not the blame is deserved or the issue is as simple as a conscious malicious act

by those in power. Most frequently, however, populist outrage discourse is used in

quotidian anti-elite attacks. These attacks accuse public officials, or contestants for

public office, of wrongdoing for self-serving purposes at the costs to the public will

19

they are meant to protect. The countless accusations that a politician tasked with rep-

resenting constituents has chosen his own private gains at the cost of his constituents’

well-being is populist outrage par excellence.

Why are appeals to outrage so prominent in ECE? One key is that the transition

from communism inherently involved many opportunities for corruption. The totality

and uncertainty of the transformation from a centrally planned economy to a free

market system was rife with opportunities for exploitation, all at a time of great

institutional flux. The privatization of state-owned enterprises into private firms

opened the door for mid to upper level former nomenklatura officials to use their

positions in the old system to become wealthy in the new one. Inevitable short-term

distortions during reforms meant the chance for early winners of reform to hijack the

transition in order to continue to secure rents from these inefficiencies permanently

(Hellman, 1998). As Vaclav Havel noted, “the nomenklatura, who until very recently

were faking concern for social justice and the working class, have cast aside their masks

and, almost overnight, openly become speculators and thieves” (1992, 3). Even when

all parties to a privatization deal worked to make the proceedings as transparent as

possible, deals were often viewed as corrupt (Dunn, 2004). It is no wonder that Fidesz

framed their 2012 constitution as the real transition away from communism — public

perceptions that former communists remained in power and gained enormously from

the transition to the new system are both widespread and justified.

Two broad processes increase the prevalence of accusations of outrages above and

beyond both the standard rough and tumble conflict of democratic politics and the

particular issues of the post-communist legacy. First, intense policy constraint from

supranational organizations in general, and the European Union (EU) in particular,

places boundaries on the programs and policies parties are able offer. This atrophies

lawmaking capabilities at the state level and infringes upon the ability of domestic

20

political elites to differentiate themselves using policy promises. Second, the contem-

porary media environment increases the incentives for political and media elites to

go negative in order to win shares in competitive electoral and media markets. These

factors influence the logic of political competition, creating a situation in which it is

easier for parties to rely upon accusations of outrages to differentiate themselves than

to depend on programmatic appeals.

2.2.1 International Constraint

One key factor accounting for the prevalence of outrage in the politics of ECE is

international constraint that limits the ability of domestic political actors to differ-

entiate themselves using clear policy appeals or stable ideological positions. Third

wave democracies in general are much more beholden to international forces than

states that democratized before them (Kapstein and Converse, 2008), and this is

especially true for the young democracies of ECE. Post-communist democratization

on the doorstep of the European Union (EU) provides an exceptional example of

this dynamic. For those ECE states that stood a chance of entering the Union, EU

membership was equated with the “return to Europe” and represented a near univer-

sal political goal (Vachudova, 2005). Thus, in eligible countries, movement towards

accession became the central valence issue and a key motivator of all public policy

(Vachudova, 2005).

Accession was, however, no small task for these states in transition. The myriad

requirements embodied in the voluminous acquis communautaire set a very diffi-

cult path before them, including over 80,000 individual policy requirements. These

“great expectations” placed upon the states of ECE were much greater than those

placed upon their Mediterranean predecessors and meant that lawmaking was more

21

about implementing EU policies off the shelf than developing distinct party plat-

forms (Grzymala-Busse and Innes, 2003; Innes, 2002; Follesdal and Hix, 2006). This

“eradicated both detailed and ideological debates over many areas of public policy”

within the young democracies of ECE (Grzymala-Busse and Innes, 2003, 64) This

did not end with accession, as the EU has developed great authority in a number

of issue areas that typically make up the axes of party competition. Parties have

often been unable to offer policy alternatives on crucial issues, and in many key do-

mains, their positions have become indistinguishable from both the positions of the

EU and their competitors (Mair, 2013; Nanou and Dorussen, 2013). This makes it

difficult for parties to attract support by offering distinct platforms, and explains

why parties in post-communist Europe are ideologically inconsistent in terms of pol-

icy output (Tavits and Letki, 2009) and “vacuous and fickle when it comes to policy

content”(Carey and Reynolds, 2007, p. 271).

Thus, parties have little incentive to provide policies according to a consistent

ideology, and instead largely take cues from Brussels. This gap is particularly prob-

lematic for party system institutionalization in light of work by Chhibber and Kollman

(2009) showing that parties compete and coordinate at the level at which real pol-

icy is made. As much of the key policy power is actually at the level of the EU,

country-level party competition — where national voters can assert meaningful ac-

countability — are less consequential than decisions from Brussels, where voters have

no real say. This is doubly true of the Union’s smaller states like Hungary, who can

expect little influence in European institutions either through their national govern-

ment or through European Parliamentary elections. A lack of electoral control over

the European-level policymaking that matters most to domestic citizens opens the

opportunity for populism, as ”feelings of powerlessness, or not being able to voice

dissatisfaction effectively, or of not being able to make oneself heard, are all fertile

22

ground for populist parties” (Meny and Surel, 2002, p. 11). Indeed, populism scholars

argue that the “European project” increases the opportunities for populist appeals

by widening the gap between citizens and those making the policies that govern them

(Taggart, 2002). As Peter Mair notes, ”whether through globalization, internation-

alization, or even the more local Europeanization, the capacity of governments to

exert political control, and hence the capacity of popular democracy to mandate

government action, has now become severely limited” (Mair, 2002, p. 83).

2.2.2 Party Competition in the Information Age

Eastern European parties emerged in the contemporary era of intense media cov-

erage and pervasive information technology. This carries implications for the tone

and content of political information as well as for the nature of party competition.

The medium of television is inherently more evocative of powerful emotions than

print or radio, and political conflict broadcast directly into the living room violates

norms of social distance and civility (Mutz, 2007). Additionally, in the age of plenti-

ful cable channels, competition for ratings is fierce and advertising revenue dictates

programming decisions. This shifts the target of political programming to a more

entertainment-based paradigm (Prior, 2005). One element of this shift is a move

towards framing politics as a “horse-race” between competitors instead of a market-

place of ideas, which drives cynicism towards politic (Cappella and Jamieson, 1997).

Consuming the soft news, or entertainment based news programming, has the same

effect(Boukes and Boomgaarden, 2015), while viewing substantive news can have the

opposite effect, decreasing cynicism among viewers (Adriaansen et al., 2010). An-

other element of this shift is the move towards equating political debate with any

other form of market commodity. In his analysis of growing political disaffection,

Hay (2007) argues that the modern electoral campaigns have undergone a process of

23

marketization at the cost of policy content. Thus, “principled advocacy and defense

of consistently articulated policy platforms by political parties” has been replaced by

“candidate branding” (Hay, 2007, p. 56).

Importantly, the new entertainment media paradigm also makes political nega-

tivity in general, and outrage in particular, more frequent. Increased competition

and the fight for ratings puts a premium on attracting attention and viewers, and a

salacious scandal drives more interest (and profit) than tedious policy debate. In the

vocabulary of the modern media guru, political outrages contain enormous “pop,” the

elusive secret ingredient of successful marketing, whether the product is dish soap or

a political candidate (Berry and Sobieraj, 2013). Indeed, participants in one study on

incivility found uncivil interactions between competing candidates to be significantly

more entertaining and engaging than civil interactions, even though incivility had

deleterious effects upon their political trust (Mutz and Reeves, 2005). Politicians are

aware of the fact that political attacks attract more attention from news outlets and

on social media, and leverage them to get free coverage (Geer, 2012) and attract in-

terest (Ryan, 2012). Recent developments in information technology only exacerbate

these trends, with online media being even more pervasive and with lower barriers to

incivility. One expansive study of Chinese micro-blogging cite Weibo demonstrates

that anger spreads faster than any other emotion across social media (Fan et al.,

2014). These authors find in particular that domestic political problems were espe-

cially likely to lead to anger spreading across social media (pg. 6). Anger’s ability to

inspire action plays an important part in this. A study of a BBC forum shows that in

general negative emotions are stronger predictors of taking part in online discussions

than positive emotions and that, above and beyond other emotions, anger causes

people to take part in the discussion (Chmiel et al., 2011). Moreover, internet media

24

outlets tirelessly pursue click bait to increase their advertising revenue, and a juicy

outrage is likely to spread like wildfire across social media.

The new media market logic, with its emphasis on image and branding, has im-

plications for the very structure of political parties. ECE parties instantly became

“catch-all” parties (Kirchheimer, 1966; Mair, 1996), distancing themselves from spe-

cific ideological commitments or appeals to well-defined social groups in an attempt to

attract the widest support possible. Additionally, the vast majority of parties in East-

ern Europe are internally created (Duverger, 1954), meaning they started as groups

of elites in parliament who worked to attract support in the populace rather than

externally-based parties that rose up from groups in society. Indeed, from the start

post-communist parties were uniquely independent of divisions within the underlying

civil society (Mair 1996). While increasingly pervasive media and the shift towards

catch-all parties and the associated trends of lowered partisanship and dealignment

have been noted in advanced industrial democracies as well (Dalton and Wattenberg,

2000), this trend is doubly powerful in young democracies. In these circumstances,

parties cannot rely upon ownership of particular issues or reputations for representing

distinct segments of society to differentiate themselves from competitors. As Innes

argues, “for all New Labour’s entrepreneuralism, Western Europe’s mainstream par-

ties continue to carry far more historical baggage than successful parties in Eastern

Europe” (Innes, 2002, p. 100).

2.3 The Explanatory Power of Emotion

These factors have left party leaders with a dilemma: they must attract support

at the polls to obtain political power, but they are unable to offer policy differences

on many of the most pressing issues facing the state. In order to solve this problem,

political elites resort to a discourse of political outrage. Following the burgeoning

25

literature on emotions in politics (Marcus et al., 2000; Brader, 2005, 2006; Valentino

et al., 2011; Seawright, 2012), I argue that the anger elicited by these appeals can

drive short-term political action without requiring specific policy prescriptions. At

the same time, however, these appeals elicit the disgust at the root of antipartism and

anti-political sentiment undermining belief in the value of competitive, multi-party

politics.

The extent to which citizens support their democratic institutions and will par-

ticipate to protect them has a lot to do with whether they feel these institutions are

worthy of defense. Likewise, the stability or instability of a party system is rooted

in the interactions between party elites and the opinions, behaviors, and habits of

the voters who ultimately decide the relative strength and standing of parties in a

democracy. Thus, the common thread to understanding the problems facing Hun-

garian democracy is the experience of citizens, and their emotional experiences are a

major part of this. In fact, psychologists view emotion as preceding and guiding cog-

nition in many circumstances (Zajonc, 1980; Damasio, 2005), empirically validating

David Hume’s assertion that “Reason is, and ought only to be, the slave of passions”

(Hume, 2007, 413). Political psychologists have also adopted this view, arguing that

emotion and cognition are “two complementary mental states in a delicate, inter-

active, highly functional dynamic balance” (Marcus et al., 2000, 8). In particular,

Lodge and Taber (2005) provide experimental evidence for what they call “hot cogni-

tion” – the idea that all political objects automatically activate emotional responses

(Taber and Lodge, 2006; Lodge and Taber, 2013). Political parties, politicians, is-

sues, and groups all evoke emotional responses in citizens, and these responses guide

political opinion and behavior. To understand where parties succeed or fail in build-

ing the bases of regular support that anchor party systems, and to ascertain voters’

26

orientations towards their political systems, scholars must attend to the way citizens

feel.

2.4 Anger and Party System Volatility

For most of the period since the fall of communism in Europe, Hungary stood out

as a case of quick democratization and party system stabilization. Partly because of

hybrid, majoritarian-leaning electoral laws, the country saw a rapid winnowing down

of parties, and for the better part of two decades the same four parties competed

for electoral office. The elections of 2010 and 2014 shattered this period of stability.

Renewed volatility in Hungarian party politics and party system instability through-

out ECE are troubling phenomena given the importance of durable, recurring parties

to democratization. For democratic accountability to hold, the same parties must

contest elections so that voters can base their decisions on past experience and party

performance, and so that elites can learn how to effectively campaign, make policy,

and order government. Thus, party system stability is a key factor for democratic con-

solidation (Olson, 1998, 433). Stable political parties channel diverse social forces into

governmental institutions (Huntington, 1968) and legitimize the democratic regime

(Diamond et al., 1989), while volatile parties erect barriers to consistent policy de-

velopment (Haggard and Kaufman, 1995), and democratic accountability (Zielinski

et al., 2005).

Some work on party system institutionalization in ECE posits signs of modest sta-

bilization (cf. Bakke and Sitter (2005)). Recent scholarship shows the primary causal

mechanism in this process to be “democratic maturation” as citizens gain experience

with democratic politics and begin to form ties with their political parties (Tavits,

2005; Tavits and Annus, 2006). However, current events in the region call this view

into question. Some party systems, like Poland, have remained volatile throughout

27

the 25 years of post-communist multiparty politics, while others, like Hungary and

the Czech Republic, are experiencing greater volatility after a brief period of stability.

Bulgaria’s 2014 elections yielded the country’s most fragmented parliament since the

introduction of competitive multiparty politics. Continued and increasing party sys-

tem instability in ECE provide evidence for Scott Mainwaring’s assertion that, when

it comes to party systems in third wave democracies, “institutionalization is not an

inevitable product of time”(Mainwaring, 1998, 80). Instead, constant flux in party

politics may be an enduring reality for third wave democracies, and the new normal

for some advanced industrial democracies as well.

Why do party politics in East Central Europe (ECE) remain highly volatile? Pop-

Eleches (2010) provides a useful theoretical paradigm to explain this phenomenon

in the region so late into the transition, arguing that increasing support for new

and unorthodox parties is the result of protest voting rather than a show of sincere

endorsement of the extreme positions they promulgate. Instead, this is a way to

punish the established parties that have had their chance and have failed to provide.

Examples of this opinion are commonplace in everyday political conversation in the

region. For instance, after detailing a litany of grievances concerning the major parties

in Hungary, one interviewee noted:

I’m afraid that means I’m going to vote for Jobbik. I know they are veryextreme and sometimes that worries me. Still, to be honest, I actuallyagree with a lot of what they say. They point out a lot of the hypocriticaland terrible things that all the politicians do. So when they say certainthings it worries me, but most of what they say I think is true. Theybring a lot of stuff out that the other politicians keep secret. I don’t wantthem to win the election, but I want them to be in Parliament to say whenothers lie.3

3Interview with early 20s female, Szentes, March 2014

28

Taking the dynamics of anger into account illuminates the logic of how this protest

voting works. Indeed, one reason political elites make appeals to outrage and other

forms of anger is that they encourage voting. A growing body of work, particularly

in the paradigm of the Affective Intelligence model, shows that emotions such as

enthusiasm, anxiety, and aversion can powerfully predict various forms of political

participation (Marcus et al., 2000). Generally, these studies find that anger increases

the likelihood of participation but also reliance on stereotyping and biases (Brader,

2005; Valentino et al., 2011; Best and Krueger, 2011; MacKuen et al., 2010). Indeed,

politicians often consciously work to leverage anger to attract attention and drive

participation (Ryan, 2012).

This project contributes to an emerging body of work linking the way people feel

about their political systems to party system volatility. One study links the collapse

of the Venezuelan party system to a sense of a lack of substantive representation by

political parties (Morgan, 2011), while another points to a sense of moral outrage

specifically as a vital explanatory factor in instances of drastic party system change

(Coppedge, 2005). Seawright (2012) argues that anger can lead to party system

collapse by decreasing sensitivity to risk and thereby increasing the vote share for

political outsiders and anti-establishment forces. He argues that mainstream political

parties represent less risky options for governance, while upstarts in the party system

come with more risk. Because anger involves a decrease in risk sensitivity and aversion

(Lerner and Keltner, 2000; Lerner et al., 2003; Fischhoff et al., 2005), a sufficiently

angry populous will be willing to cast their lot with the risky political outsider. This

is a novel argument that the current research builds upon by tracing the source of

political anger. Seawright’s empirical strategy relies upon non-political sources of

political anger, and I advance this theory linking anger and party system instability

29

by showing the origins of political anger: the nature of the political information

citizens receive.

2.5 Disgust and Dedication to Democratic Institutions

Democracy requires public support for and participation in the political system,

and this is particularly important in young democracies lacking in institutionalization.

This was always a key concern in the study of post-communist democratization. The

“Leninist legacy” created great animosity between the public and private spheres and

a great deal of cynicism towards politics in general and political parties in particular

(Jowitt, 1992). Evidence suggests that public regard for the norms of democracy have

fallen even from this low starting point. While increasing unhappiness with politics

is a global phenomenon, trust in political institutions has fallen even more rapidly in

these new democracies than in the advanced industrial democracies (Catterberg and

Moreno, 2006). Wessels and Klingemann (2006) show that trust in political parties

and support for party democracy decreased with experience with democracy in ECE.

They argue that this shows that the “rules of the game” are not fully accepted among

the populace, a key problem for democratic consolidation. Ceka (2012) echoes these

troubling findings, arguing that the more robust the political competition in a post-

communist country, the weaker that country’s public supports democracy. Finally,

Rose and Mishler (1998) provide evidence that negative partisanship — identifying

a party one would never vote for — is far more widespread than any semblance of

party identification in post-communist Europe.

There are many reasons for voters in ECE to become disaffected from their po-

litical systems. As noted above, concerns over rampant corruption during the simul-

taneous economic and political transformations were well-warranted (Hellman, 1998;

Kostadinova, 2012). As Chiru and Gherghina (2012) find in their study of Bulgaria

30

and Romania, voter attachments to political parties are extremely weak in these cases

primarily because of negative perceptions of party performance and widespread cor-

ruption. Moreover, Enyedi and Toka point to widespread anti-party sentiment in the

region, stating “The initial reservation against parties can be explained in terms of the

communist legacy, but the various scandals surrounding party politics have certainly

strengthened negative stereotypes concerning the actual motives of party politicians”

(Enyedi and Toka, 2007, p. 152). Indeed, scandals “generated much cynicism about

the moral integrity of... political parties in general” (Enyedi and Toka, 2007, p. 153).

This is damaging to a young democracy. Indeed, turnout in Hungary has declined

every year since a peak in 2002, and public opinion polls document abysmally low

support for democratic institutions. What is worse, the Orban government’s anti-

democratic constitution and reforms have failed to inspire any meaningful backlash

within Hungary. Despite years of what has been dubbed by political commentators

as “Orbanization” for its striking similarities to “Putinization” and the making of

an illiberal Russian regime, there has been no united front of political resistance or

sustained protest. The most energetic movement in the country is arguably Jobbik,

an avowedly radical movement that is openly critical of the idea of liberal democ-

racy. Even after gutting the independent judiciary, placing restrictions on freedom of

the press, and openly declaring war on the institutions of liberal democracy, Fidesz

received another plurality of the vote in the 2014 elections, albeit with significantly

lower turnout than during their initial election in 2010. Moreover, because of gerry-

mandering the electoral system to drastically favor its candidates,4 Fidesz managed

to turn a 44% plurality among the public into the two-thirds Parliamentary majority

4One interviewee jokingly called this “Fideszes,” which translates to “Fideszing” (Interview withmid-20s male, Szentes, March 2014).

31

needed to continue manipulating the country’s political institutions to meet its every

whim.5

Why would Hungarians tolerate movement away from democracy, and even sup-

port the party responsible? Some support is genuine; Fidesz’s nationalistic, paternal-

istic rhetoric resonates with many voters. Moreover, there is evidence that those who

voted for a victorious party become more “submissive subjects” and allow this party

wide berth and the ability to dismantle democratic institutions (Moehler, 2009). An-

other element surely has to do with the supply of alternatives to Fidesz — the left

remains discredited and fragmented by constant petty infighting, Jobbik scares many

voters, and the new left-libertarian party Politics Can Be Different (LMP) appears

out of touch and ineffective.

Even still, the quiescence of the public requires a deeper explanation that takes into

account the way citizens perceive their political system. The psychological reaction of

disgust, inspired by perceptions of gross violations of morality and purity, fills this gap.

This emotion drives an urge to separate oneself from its source that is automatic and

complete (Rozin et al., 1999; Giner-Sorolla, 2012). Disgusted citizens would display

the exact pattern we observe among many in Hungary, abstaining from elections or

defacing ballots rather than casting a vote for options they see as loathsome, and

having little interest in rallying to defend a political system they feel is rotten.

Disgust evolved to avoid sickness by keeping impurities away from the mouth.

As humans developed more complex social structures, the emotion’s original physical

role expanded to include a socio-moral function that enforces social norms by keeping

moral impurities out of groups (Kelly, 2011). As this emotion guards against con-

tamination and contagion, it is associated with a strong desire to get away from the

offending object, whether it is rotten meat that threatens the body or an unthinkable

5This would prove short-lived as by-elections in 2015 brought leftist and radical right candidatesto parliament and robbed Fidesz of its constitution-making majority

32

moral transgression that threatens the group. The emotion’s evolutionary past means

disgust can easily become associated with an object or behavior. This is because the

adaptive task of disgust — to keep potentially fatal contaminants out of the body —

is fitted with a hair trigger and is extremely prone to false positives. It only takes

eating the wrong sort of berry once for it to become associated with a strong desire

to wretch it up once tasted again, and this is highly adaptive given the life or death

stakes of keeping poisons out of the body. This same mechanism expands readily to

new stimuli. For instance, cigarette smoking has come to elicit disgust among many

in the United States (Rozin and Singh, 1999). Indeed, recent evidence suggests that

disgust is a reaction to a wide range of fairness violations (Chapman et al., 2009),

and that disgust activates parts of the brain associated with moral judgments (Moll

et al., 2005). That disgust is often the reaction to moral violations means it is likely

that a discourse of populist outrage, with is barrage of offensive moral transgressions,

evokes disgust instead of or alongside moral anger for many citizens.

Disgust carries with it a number of troubling ramifications for democratic politics.

On the most basic level, citizens who come to be disgusted by politics will not be

merely apathetic towards their political processes, but will instead actively work to

avoid any interaction with politics at all. As Kelly (2011) notes,

“social norms that co-opt the emotion recruit a kind of aversion, probablyuseful in motivating agents to avoid the types of activities proscribed bythose norms, or motivating them to avoid or shun transgressors. As a by-product, however, such norms and motivations will also be infused withthe other elements that accompany the disgust response, including a senseof offensiveness, contamination, and feelings of revulsion” (134)

Indeed, recent work in American politics shows that many citizens who abstain

from voting are disgusted with politics, seeing politics writ large as a “poisonous

idea” to be shunned at all costs (Vandenbroek, 2012). This is exactly the trend we

33

see growing in democracies around the world, where “to label an activity or process

as ‘political’ is, it seems, invariably to deride and distance oneself from it” (Hay,

2007, p. 5). When citizens become disgusted with the very process of competitive

democratic politics they will feel that the entire political game is immoral, compelling

them to avoid the mere mention of politics, let alone political participation. This

prospect fundamentally undermines the participatory ethos at the heart of democratic

accountability. Even worse, while it is easy for something to become an elicitor of

disgust, it is very difficult for it to become clean again (Kelly, 2011). Thus, for those

who come to feel disgust towards democratic politics, the association will be very

long-lasting. Recent developments in Hungary demonstrate this in the extreme, as the

populace is largely mute and apathetic in the wake of blatant moves by the government

to roll back the institutions of liberal democracy. Disgust helps understand this

problematic trend.

2.6 The Distinctiveness of Disgust

While anger and disgust are distinct in terms of elicitation and behavioral ramifi-

cations, they can be difficult to disentangle from one another empirically (Hutcherson

and Gross, 2011). One reason for this is the enormous overlap in their domains. Both

are “other-condemning” emotions implicated in regulating behavior within and be-

tween groups (Haidt, 2003). Worse still, the various shades of these negative emotions

such as outrage (for anger) and contempt(for disgust) offer even greater potential for

bleeding into one another. This is one reason why these emotions are rarely distin-

guished from one another in studies of political behavior. In the view of the pioneers

of Affective Intelligence Theory (AIT) dominant in political psychology studies of

emotion, both anger and disgust are on the same emotional scale of “aversion,” such

that disgust is a low level of aversion resulting in a lack of action in response to a

34

stimulus, while anger is a higher level of aversion eliciting action (Marcus et al., in

press).

The view that these emotions are different by degree instead of by type is flawed.

While the exact taxonomy of emotion remains an open debate in psychology, most

theoretical paradigms point to both anger and disgust as separate core emotions, or

universally distinct emotional systems deeply rooted in evolutionary history.6 The dis-

tinctiveness of anger and disgust is further supported by neuroimaging work showing

that unique cognitive structures are involved with each. Anger, like other emotions

implicated in responding to threat (fear), is associated with the amygdala, while

disgust, due to its origins in protecting the body from contaminants, is primarily

associated with the gustatory complex(Jabbi et al., 2008; Kelly, 2011). These emo-

tions correspond to different behavioral implications as well. As Giner-Sorolla (2012)

argues, “anger and disgust each stick to their separate jobs” when it comes to the

survival instinct: “anger helps you approach a task aggressively even in the face of

obstacles,” while “disgust in the natural environment keeps you away from things that

might make you ill, like rotten or uncooked flesh”(74) Thus, though both emotions

are correctly identified as aversion responses, their deep evolutionary roots point to

important differences in terms of both elicitation and ramifications.

These emotions do occur together frequently, however. Anger is perhaps most

dangerous when mixed with contempt, a sub-type of disgust.7 Fischer and Roseman

6The most common core emotions are anger, disgust, happiness, sadness, fear, and surpriseas elucidated by Ekman and Friesen (1971) in their cross-cultural studies of facial recognition ofemotions.

7While psychologists continue to debate the line between the two, the distinction between disgustand contempt, if it does indeed exist, is irrelevant to the present research. Some models of emotionposit contempt as a halfway point between anger and disgust, but when applied to social interactions,these two emotions are often distinct and thus lumping them into contempt is not the correct levelof analysis for questions concerning social behavior (Cottrell and Neuberg, 2005, 786) As is thecase with personal anger and moral anger (or outrage), the causes and consequences of disgust andcontempt are the same in the current circumstances. I will use the two words as synonyms.

35

explain that if one is angry it means “it may be worth engaging in an attempt to

change the person’s behavior” but to feel contempt is “to appraise the other person

as unworthy or inferior” and beyond deserving an expenditure of precious energy and

resources to correct (2007, 104). Contempt often comes out of repeated unresolved

anger and is particularly likely given unfamiliarity with and dislike towards the of-

fending group. Because contempt often comes out of chronically unresolved anger,

the two emotions frequently co-occur. This is a dangerous combination, “as it implies

that the inclination to attack co-occurs with the inclination to derogate and exclude

a person who is seen as inferior or bad, which may result in hostile acts that are not

held in check by affection or social relationships” (Fischer and Roseman, 2007, 113).

This is likely the case when partisans are constantly frustrated by political oppo-

nents, or when an oft-maligned minority becomes an object of disgust. The cases of

the long-derogated Roma minority across ECE and the more recent influx of refugees

fleeing political violence are particularly troublesome examples of this trend.

Despite the caveats that these emotions are difficult to measure and may occur

together — issues that will be addressed more fully in the empirical chapters to follow

— the difference between anger and disgust promises to illuminate important political

outcomes. This distinction links the same phenomenon of pervasive populist outrage

in politics to two patterns that can be seen to be at odds with one another. On the

one hand, anger at these moral outrages drives protest voting that undermines party

system institutionalization, a form of political action. On the other hand, those

who are disgusted by the stream of outrage will prefer to exit politics in general,

accounting for waning participation and lack of action in defense of the democratic

system. Moreover, the political domain, where concrete personal interests, abstract

symbols, morals, and values constantly come into conflict, offers a rare opportunity to

distinguish these emotions. Leveraging the unique circumstances of post-communist

36

political development to probe the line between these emotions is an opportunity for

“political psychology true to its name;” that is, research that informs psychology as

well as political science instead of simply applying theories from psychology off the

shelf to understand political outcomes (Krosnick and McGraw, 2002).

2.7 Disgust or Anger?The Roles of Partisanship and Time

The preceding discussion begs the question: what determines whether a citizen

feels angered or disgusted by political outrage? Answering this question involves at-

tending to two distinct functions of emotional response: appraisal and association.

On the one hand, many emotions serve an appraisal function, appraising the envi-

ronment and facilitating an appropriate action (Lazarus, 1991). On the other hand,

emotions also operate as associative learning mechanisms, guiding thought and be-

havior based upon previous experiences, both positive and negative, with objects or

types of objects (Giner-Sorolla, 2012). Anger and disgust are elicited through different

processes; anger typically involves an appraisal of wrongdoing or blame, while disgust

is the result of repeated associations of an object with something nasty (Kelly, 2011;

Giner-Sorolla, 2012). Thus, whereas anger comes about from evaluations, disgust is

better explained as an outcome of associative learning. This is why even when we

know something is not actually a contaminant (chocolate shaped like dog poo), we

can still be disgusted by it if the association is present (Haidt, 2003). As Giner-Sorolla

(2012) notes, “disgust’s learning functions work with the long wisdom of evolution

and culture, not the quick intelligence of the human mind ” (81). Moreover, we learn

what is disgusting from social cues, so the associations that make up disgust are

highly relative (Kelly, 2011).

37

The distinction between the appraisal of anger and the associative origins of dis-

gust points to explanations for when populist outrage discourse elicits anger or dis-

gust. Two factors are most salient in analyzing the connection between outrage and

political behavior. A key distinction is the role of blame: unlike disgust, the ap-

praisal of anger depends on blame (Quigley and Tedeschi, 1996; Petersen, 2010) and

is affected by whether someone is harmed (Gutierrez and Giner-Sorolla, 2007) and

whether the offending act was intentional (Russell and Giner-Sorolla, 2011). Percep-

tions of justice are key (Nugier et al., 2007). One important political proxy for the

attribution blame is the presence and strength of partisan or ideological predilections.

Previous work on emotions in politics has documented that eliciting anger works well

on partisans, strengthening their predispositions and spurring them to action (Mar-

cus et al., 2000; Valentino et al., 2011). Partisans operate with a “perceptual screen”

filtering their understanding of political events (Lodge and Taber, 2013; Westen et al.,

2006). They are therefore more likely to see one side as worthy of blame for either

committing the alleged outrage or for manufacturing it. Meanwhile, those in the

middle without a stake in the matter will be less likely to judge either the accuser

or the accused of blame, instead feeling general aversion that leads to disgust at the

very situation.

Another important factor is the amount of exposure citizens have to a certain

attack, issue, or scandal over time. Repeated exposure to an outrageous accusation

or offender causes moral anger over and over again until eventually it elicits disgust.

At first, a fresh outrage may indeed inspire moral anger in a citizen. This moral

anger is associated with a desire to right the wrong, and as noted above, this can

often be accomplished through voting against the actor responsible for the trespass.

A paradigmatic case is when a citizen is unhappy with the political establishment

38

and the opposition or an unorthodox party leverages moral anger against the govern-

ment to attract support. It is in this way that initial anger can lead to the protest

voting that stymies party system institutionalization (Pop-Eleches, 2010). Over time,

however, if this vote switching does not result in better, cleaner politics, the specific

blame felt towards the original offender becomes generalized to all in the political

system, and any actions to improve the situation seem ever more hopeless. This leads

to generalized disgust towards political processes. Thus, the second variable that will

moderate emotional reactions to political outrages is time: greater, repeated exposure

to populist outrage discourse causes disgust rather than anger.

This over time transition is in line with the findings of Dassonneville et al. (2015),

who take up the question of when political disaffection causes vote switching and when

it causes abstention. They find that disaffection initially leads to vote switching, but

that with time, if this vote switching fails to result in the desired political change,

abstention is the result. This is exactly what the hypothesized route from anger to

disgust over time and thwarted actions predicts. Indeed, this hypothesis also builds

upon theories positing a change in reaction from anger to disgust when anger goes un-

resolved (Roseman et al., 1994). Damasio (2005) outlines a process by which objects

we have previously felt an emotion towards are tagged by a “somatic marker” that is

activated upon exposure to related stimuli in the future. Therefore, in terms of the

psychology of emotions, the transition of anger into disgust is a process of the emo-

tional marker attached to a particular outrageous instance, politician, party, or even

the entire political system evolving over time. Over the long haul, repeatedly elicited

emotional reactions may even evolve into a somewhat stable emotional evaluation

that guides a citizen’s basic political orientation. The evolution of these emotional

responses is adaptive, as emotions give us feedback as to whether to continue toward

a goal, abandon it, or take another approach towards it. As Giner-Sorolla (2012)

39

notes, “Anger may eventually burn out faced with strong and persistent frustration,

which itself is adaptive; people need to see when a task is impossible and let go” (94).

Where prior political action failed to achieve a goal, greater negative emotion is the

result. When a similar stimulus leads to more negative emotion and more action, the

feedback loop returns more and more negative emotion, until behavior changes to no

behavior, and anger changes into disgust.

This dissertation’s analysis of emotional reactions over time constitutes its largest

contribution to work on emotions.8 Scholarship on emotional response examines a

snap shot in lab experiments. However, this does not get at how emotional reactions

work in the real world where they evolve over time. It is crucial to look at emotions

over time, and this is especially true for useful theories of emotions in politics. As

Brader noted in a critique of his own landmark study of emotional reactions to adver-

tisements, “elections are not won or lost on the basis of an effective campaign ad, but

may be won or lost with an effective ad campaign” (Brader, 2005, p. 402). Indeed,

the same message might elicit different emotions over time, or chronic activation of

an emotion might have different effects in the short and long terms.

2.8 Hypotheses

In sum, the theory presented above and represented in Figure 1 produces four

central hypotheses:

• Hypothesis 1 : Politicians resort to a dialogue of populist outrage to overcome

the problem they have distinguishing themselves from one another

8As detailed in Chapter 5, it is also by far the hardest element of the theory to test empirically.

40

– H1a: European Union constraint is a key contributor to this problem of

distinction, and thus predicts an increase in populist outrage in the political

dialogue of ECE

• Hypothesis 2 : Populist outrage elicits anger and disgust among citizens as a

function of whether they attribute blame to the actors involved in the alleged

transgression

– H2a: The strength and presence of partisan predilection increases the

likelihood of the attribution of blame and thus predicts anger

– H2b: Being closer to the partisan and ideological center decreases the

likelihood of the attribution of blame and thus predicts disgust

• Hypothesis 3 : Over time and with repeated exposure to populist outrage and

the actors thought to perpetuate these outrages, the initial reaction of anger

transitions into disgust

– H3a: Initially a given transgression is more likely to elicit anger and the

desire to punish the responsible actor

– H3b: With repetition, this same transgression elicits disgust

• Hypothesis 4 : Anger and disgust have different effects upon political behavior

– H4a: Anger is associated with action to address the offending situation

– H4b: This action impulse will often be expressed as protest voting under-

lying party system instability

– H4c: Disgust is associated with a rejection of politics, political participa-

tion, and the democratic system as a valued goal

41

Fig

ure

2.1:

Ove

rvie

wof

the

theo

ry

42

2.9 Alternative hypotheses

The theory presented above links the dual trends of party system instability and

withdrawal of citizens from politics in ECE to the nature of the political dialogue in

these new democracies, highlighting the emotional responses of anger and disgust as

the key causal mechanisms linking the political dialogue of populist outrage to these

outcomes. These macro-social outcomes are, however, multi-causal by nature. Here

I elucidate the logic of two key alternative hypotheses and demonstrate how the role

of populist outrage and the distinction between anger and disgust nevertheless make

valuable contributions even in light of these alternatives.

2.9.1 Economic Strife

Popular commentary often blames political instability upon economic crises, from

the fall of the Weimar Republic during the Great Depression to the rise of protest

parties in the wake of the 2008 financial crisis. The correlation between economic strife

and political instability is obvious; indeed, Hungary’s own party system realignment

and the dramatic rise in harsh anti-politics are both associated with the country’s

credit crisis and recession in the wake of 2008. Moreover, the link between economic

problems and both party system volatility and political disaffection is reasonable.

We would expect economic problems to make unhappy voters, and unhappy voters

to punish the established parties; indeed, this is the most basic premise of the vast

literature on economic voting (Lewis-Beck, 1988).

Studies of party system instability do find some evidence that economic strife is

associated with higher volatility. Dassonneville and Hooghe (2015) offer one such

study. However, their analysis focuses exclusively on Europe in the wake of financial

crisis, meaning it cannot discern economically-fueled voting from a general trend of

43

dealignment observed before the crisis (Dalton and Wattenberg, 2000). Another such

study is that by Powell and Tucker (2014), who find that economic performance

explains replacement volatility — the replacement of old parties by new entrants into

the party system — such that economic success decreases volatility and economic

problems increase it. However, Crabtree and Golder (ming) replicate these analyses

and show this finding to be fallacious and driven solely by one outlier (Bosnia).

Indeed, their analysis reveals no clear predictor variable for party system instability

in ECE, leading them to conclude that this is a question for which the study of

comparative politics requires fresh theoretical perspective. It is exactly this novel

theoretical bent that the current study provides.

Another key concern is that economic problems decrease support for democracy,

especially in young democracies. This was a major source of anxiety among scholars

and policymakers alike at the onset of post-communist democratization. Przeworski

(1991) provided the most well-formulated statement of this concern, worrying that de-

mocratization and the simultaneous need for painful economic reform would cause the

mass public to abandon democracy as soon as economic hardship became widespread.

Jowitt (1992) went so far as to advocate a “liberal authoritarianism” at the onset of

transition in large part because of this fear that economic distress would lead to

a populist and authoritarian backlash. In the long run, however, these fears were

overstated, and citizens were more adaptive, patient, and able to blame economic

disjuncture on the previous regime than early accounts expected (Rose, 2007).

The links between economic strife and both party system volatility and disaffection

with politics are likely real, but as the debates alluded to above show, they are not

inexorable rules. Moreover, economic arguments preclude anger and disgust from

playing an important role as causal mechanisms. On the contrary, these emotions are

more likely in times of economic duress. Economic hardship increases the potential for

44

political elites to leverage populist outrage to blame political opponents for problems.

Indeed, economic crisis may also increase the efficacy of populist outrage at inspiring

negative emotions among the citizenry. Seawright (2012) provides a model of party

system collapse via anger that includes economic hardship as a precondition. Thus,

the theory presented here remains agnostic on the overall effects of the economy

upon party system stability or political disaffection, but posits that economic strife

can increase both the prevalence and effectiveness of populist outrage and thereby

exacerbate an already difficult situation.

2.9.2 Critical citizens and discerning democrats

Another alternative to the theory presented above challenges the assumption that

unhappy citizens are undesirable from the perspective of accountable democratic gov-

ernance. Indeed, strong negative emotions towards politics do not necessarily mean a

disengagement from politics or a lack of support for democracy. Scholars of emotions

in politics often connect negative affect to desirable behavioral outcomes. For in-

stance, the link between anxiety and the search for novel information leads Valentino

et al. (2008) to ask “Is a worried citizen a good citizen?” in the title of their arti-

cle, while in other work these scholars argue that angry citizens are more likely to

participate in politics (Valentino et al., 2009, 2011).

Other work questions whether dissatisfaction with politics is a bad thing, flipping

the conventional wisdom that disaffection leads to an exit from the political system

on its head by arguing that being skeptical of government is a good attribute of a

democratic citizen. With her “critical citizens” approach, Norris posits that in the

information age citizens expect more from their government and are better endowed

with the resources to pursue their interests (Norris, 1999). The implications are

that new and innovative methods of linking citizens to political decision making are

45

necessary, but not that citizens writ large are exiting politics. In a similar vein,

Moehler’s analysis of the effects of participation in the initial stages of Ugandan

democratization yielded the surprising finding that participation lowered trust in

institutions (Moehler, 2008). However, she argues this is actually a positive finding, as

these “distrusting democrats” have become more skeptical and watchful and therefore

more likely to monitor and act against democratic backsliding.

These sanguine conclusions linking cynicism to increased accountability are con-

tingent upon citizens keeping up the pressure on their leaders over the long term.

After over twenty-five years, however, Hungary and other countries in this cohort

have more disgusted democrats who abandon the very ideal of democracy than dis-

trusting democrats holding their leaders’ feet to the fire. In addition to these time

horizon constraints, there is the matter of the degree of disaffection. In Moehler’ss

research, participation brought trust down from very high to “moderate skepticism,

not outright cynicism” (Moehler, 2008, 187). Thus, higher levels of dissatisfaction,

or lower levels of initial trust, may well lead to a rejection of the system. Other

empirical work fails to find the link between disaffection and critical citizens. In one

study that finds a sharper decrease in trust in political institutions in new democra-

cies than old, Catterberg and Moreno argue that the lack of basic provisions in many

new democracies is evidence that atrophying political trust is “more closely linked

to disillusionment and disaffection rather than to the emergence of a more critical

citizenry” (Catterberg and Moreno, 2006, 33). Therefore, the current work will again

remain somewhat agnostic on whether a degree of critical pressure or discernment in-

creases democratic accountability and will instead focus on the dynamics of disgusted

citizens. As the empirical analyses in the following chapters make clear, disgusted

citizens are not more likely to exert pressure on elected representatives. Instead, they

46

exit the system and make up a necessary if not sufficient requirement for democratic

backsliding.

47

Chapter 3: The Prevalence of Populist Outrage in East

Central Europe

Perhaps the most ubiquitous advertising campaign of the 2014 Hungarian parlia-

mentary election featured members of the opposition in a photo-shopped police lineup,

surrounded by clowns, holding their names as if being booked after arrest. Numerous

pamphlets featuring the image arrived in every mailbox in the country reading “The

seven sins of the Gyurcsany coalition,”1 so named for the former Socialist Party Prime

Minister Ferenc Gyurcsany who had been caught on tape after his reelection admit-

ting to lying about the need for more austerity reforms before the election. Each page

of the pamphlet featured an opposition politician’s photo-shopped mug shot with a

“sin,” or an accusation of wrongdoing. The pamphlet portrays austerity measures

necessitated by the country’s economic situation and European Union economic rules

as crooked betrayals of vulnerable groups in society. This campaign was only one

example of a larger political strategy of portraying political opponents as traitors and

crooks instead of advocates of competing policies that had become the status quo in

Hungary by 2014.

The most casual observer of politics in any democracy will be familiar with similar

allegations of wrongdoing and scandal. Scholars of political communication have

taken note of these attacks, and some studies point to an increase in certain types

1A picture of the pamphlet is in Appendix 1, alongside translations of the seven sins. This adver-tising campaign was so ubiquitous I chose to use it as the basis for the lab-in-the-field experimentdetailed in the next Chapter.

48

of negativity in political debates both from the media and from politicians (Kamber,

1997; Geer, 2012). However, this important work focuses on the case of the United

States, and little exists concerning young democracies around the world. This is an

unfortunate lapse given the possibility that the turn to the negative is even stronger

in some young democracies than in the advanced industrial democracies that are

the focus of the negative campaigning literature. While the effects of these attacks

on the participatory ethos at the heart of democracy may be highly problematic

in an advanced industrial democracy such as the US (Ansolabehere and Iyengar,

2005), a similar increase in political ire in younger democracies, where adherence

to the democratic system is less absolute, may be even more devastating. Post-

communist East Central Europe (ECE), with its heritage of Manichean “us versus

them” communist politics, and where the pressures of globalization and a changing

media environment occur in the contexts of weak parties and atomized civil societies,

may be particularly prone to both the rise of populist political attacks and their

deleterious consequences.

This chapter begins the dissertation’s empirical analysis of the rise of the politics

of outrage in Hungary and the associated responses of anger and disgust by honing in

on the nature of the political information environment in Hungary and ECE. It details

why we might expect greater levels of political negativity in the young democracies

of ECE and introduces a particular type of political attack — populist outrage —

that is hypothesized to be especially pernicious. Evidence from the Comparative

Manifestos Project shows that populism has supplanted policy debate in the Visegrad

countries of Hungary, Poland, and the Czech and Slovak Republics in the context of

European Union accession. A more detailed content analysis of political coverage in

the Hungarian news media during elections shows a rising level of populist outrage

49

dialogue associated with European Union accession and preceding the unraveling of

Hungarian democracy.

3.1 Media Technology and Attack Politics

While the push and pull of politics inherently involves conflict, recent years have

seen an increasingly negative tone of political debate, at least in the United States

(Kamber, 1997; Geer, 2012). There are a handful of unique elements of the American

political experience that contribute to this trend, such as the “permanent campaign”

brought on by the singularly short two-year election cycle in the House of Representa-

tives or the increased role of money in politics following the Citizens United decision.

However, at the heart of most theories of the increase in political negativity is a

global trend: changes in media technology leading to a more fragmented and com-

petitive media landscape featuring more penetrating and pervasive forms of media.

These same technological trends are present in democracies across the globe, so to

the extent that these changes do lead to increasingly negative political dialogue, this

is likely to be the case in democracies beyond the United States.

The central element of this shift is the move towards equating political debate

with any other form of market commodity. In his analysis of growing political dis-

affection, Hay (2007) argues that contemporary electoral campaigns have undergone

a process of marketization, replacing ideological and programmatic principles with

short, snappy tag lines and candidate branding. Whereas in past decades all political

news went through the filter of a handful of major network broadcasters, the ad-

vent of the 24-7 news cycle and the proliferation of countless cable channels have left

broadcasters in a tight race to attract ratings. With 24 hours to fill up with “news,”

stories that once would not have qualified as such, like personal indiscretions of public

servants, become headline news repeated ad infinitum. In the age of plentiful cable

50

channels, competition for ratings is fierce and advertising revenue dictates program-

ming decisions. Content that attracts viewers, listeners, shares, and clicks becomes

the goal for political coverage, and salacious events are more likely to attract atten-

tion. The well-documented negativity bias that people tend to notice and remember

the negative over the positive (Rozin and Royzman, 2001; Ito et al., 1998) only serves

to exacerbates this.

A key driver of the increase in political attacks is therefore the propensity of an

ever more fragmented news media to gravitate towards salacious stories in order to

boost ratings. The same logic applies to candidates and parties vying for votes and

attention in an electoral campaign. Geer (2012) notes a marked increase in negativity

of American presidential campaigns and points to the way that changes in news

coverage affect the use of negative political appeals, arguing that political attack ads

are less about using the ads to persuade voters and more about garnering free coverage

from the media by causing a stir. In other words, the news media is more likely to

cover political attacks, and, knowing this, politicians and their strategists provide

this negativity in their campaign communications and advertisements. Indeed, the

idea that the advertisements and messages politicians put out are as much about

“managing the press” as they are about reaching voters directly is well established

(Lau and Rovner, 2009; Ridout and Franz, 2008).

Another element of the changing media landscape is a shift to a more entertainment-

based paradigm in political coverage (Prior, 2005). Consuming entertainment-based

news programming, or what Prior (2005) calls “soft news,” depresses efficacy and

participation (Boukes and Boomgaarden, 2015), while viewing substantive news can

have the opposite effect, decreasing cynicism among viewers (Adriaansen et al., 2010).

Additionally, the constant need for entertainment and programming in political news

has also shifted the content of a great deal of political programming from substance

51

to updates on the electoral odds of the respective sides. This movement towards

framing politics as a “horse-race” between competitors instead of a marketplace of

ideas drives a “spiral of cynicism” towards politics (Cappella and Jamieson, 1997),

decreasing the different policy proposals presented and further increasing incentives

for campaigns to go negative to frame themselves as the victors of the horse race.

Finally, new technological trends affect the nature of political appeals and news

coverage by providing more persuasive and personal delivery techniques. Television

is by nature more emotionally evocative than print or radio, and this is particularly

true of uncivil dialogue between politicians. When acrimony and accusations are on

the television in the living room, the brain reacts as if a bitter fight is unfolding in real

time, creating intense discomfort as these unwanted guests violate important norms

of civility (Mutz, 2007). Newer forms of political communication on the web are even

more likely to attract negativity, especially from readers and viewers. While several

scholarly studies confirm this (Jansen and Koop, 2006; Blom et al., 2014; Santana,

2014), a cursory glance at the comment threads of a few web outlets purveying po-

litical information is likely enough to support the view that the internet increases

political negativity for many. Taken together, then, there are several reasons to be

concerned about the declining tone of the political information environment due to

the advent of new media technologies and practices across established and young

democracies.

3.2 Populist Outrage

One type of political appeal is particularly likely to attract the attention that

political and media elites so desperately seek in their intense competition for votes

and clicks is populist outrage, a strategy of political appeal that is the central fo-

cus of the current work. In the simplest terms, this dialogue involves identifying a

52

transgressing actor who has committed some tremendous atrocity against the good,

everyday people. This concept builds upon and overlaps significantly with what Berry

and Sobieraj (2013) describe as “outrage discourse,” which they define as “efforts to

provoke emotion responses (e.g. anger, fear, moral indignation) from the audience

from the use of over-generalizations, sensationalism, misleading or patently inaccu-

rate information, ad hominem attacks, and belittling ridicule of opponents;” they

note that “outrage sidesteps messy nuances of complex political issues in favor of

melodrama, misrepresentative exaggeration, mockery, and hyperbolic forecasts of im-

pending doom” (Berry and Sobieraj, 2013, 6). This very broad definition of outrage

is a useful starting point, but here I hone in on a subset of this general discourse

that involves populism and is in line with the psychological correlate of outrage in

the media: moral anger in the individual citizen. Psychologists define moral anger as

“anger provoked by the perception that a moral standard — usually a standard of

fairness or justice — has been violated” (Batson et al., 2007, 1,272). It is the emotion

of “how dare they” so often employed by politicians, and for good reason. It “fires

up the base,” encouraging action through passion and encouraging people to take

a stand for what is right (Brader, 2006; Carver and Harmon-Jones, 2009; Valentino

et al., 2011). While a debate over the distinctiveness of outrage, or moral anger, and

the more well-studied personal anger continues in psychology, here I remain agnostic

as to whether moral anger constitutes a distinct emotion from personal anger, as the

causes and consequences of each type of anger will be the same regardless. Instead,

the focus of this dissertation is on the use of outrage as a strategy of political dis-

course, or the appeal to moral anger and indignation through a violation of morality,

fairness, or justice.

These outrageous accusations are at the heart of many varieties of political attack,

but in particular, they are the essence of populist appeals. Populism involves pitting

53

a virtuous, homogeneous “people” against some morally-rotten transgressor who has

knowingly and treacherously done something to harm the good, everyday people

(Canovan, 2002).2 The accuser implicitly draws a line between this immoral violator

on the one hand and the people on the other, identifying himself with the virtuous

people and vowing to fight for them in the process. The political establishment, the

financial sector, and any definable group can be the target of the accusation. Because

of this malleability, this versatile political strategy exists across many contexts and

is frequently employed by both the left and the right sides of the political spectrum.

Given the role of a violation of morality separating the good, everyday people from

the evildoing other in populism, the outrage appeal and the moral anger it seeks to

evoke are the central mechanisms for most forms of populism. Thus, populist outrage

is defined as a political strategy pitting some transgressor or out-group against the

people, and accusing this transgressing person or group of unfair or unjust acts in an

attempt to inspire moral anger on the part of the message receivers. These appeals

take many different forms. They include accusations that immigrant or minority

groups unfairly take from hard-working citizens, that corrupt politicians violate their

duty to serve the public by pursuing their own selfish interests, and that the EU

and multi-national firms scheme to steal the country’s wealth and sovereignty. As

(Berry and Sobieraj, 2013) note, “Outrage trades in hyperbole, doling out comic book

narratives in which users almost always cast themselves in the role of the intrepid

hero ready to fight for ‘the American people’ ”(47). Outrageous accusations are also

central to many nationalist appeals; the common accusation that an immigrant or

minority group receives scarce state resources at the cost of hard-working citizens,

2I do not mean to imply that this is the only type of populism, as this essentially contestedconcept has multifarious meanings within both the scholarly literature and popular discourse (seeMeny and Surel (2002) for a useful overview). My intention is to isolate populist appeals to outrage– what I label as “populist outrage” – as a very common strand of populism that is particularlyprominent under the current circumstances in Hungary and other ECE states today.

54

or that certain ethno-religions groups are dedicated to destroying the national way

of life, are prime examples. Nationalism and populism often overlap wherever a

virtuous, homogeneous people based upon national identity is said to be threatened

or betrayed by foreign foes or their domestic agents. From Trump’s Mexican rapists to

Putin’s Western Imperialists to the centuries-old Hungarian trope of playing off of the

persistent sense of Hungarian victimization at the hands of international forces out

to destroy the country, this is a tried and true combination across many contexts. 3

Most frequently, however, outrageous discourse is used in the quotidian accusations of

wrongdoing leveraged by contestants for or occupants of public office. The countless

accusations that a politician tasked with representing constituents has chosen his

own private gains at the cost of his constituents’ well-being are examples of populist

outrage par excellence. It is thereby defined as genre of political appeal centering upon

a blatant violation of a widely-valued standard of morality. This is a key strategy in

democracies everywhere, and I argue, a central method of political appeal within the

dialogue of post-communist politics in particular.

Berry and Sobieraj (2013) argue that outrage has increased over recent years due

to the increased marketization of political news even above and beyond negativity

or incivility in general. Their content analysis, detailed below, finds that across

numerous media including newspapers, television shows, talk radio, and blogs, there

is a rough average of a resort to outrage every 90 seconds or so. Of course, this

varies considerably depending on media (worst on TV and Radio), specific hosts,

and ideological orientation(more prevalent on the right than on the left). They also

conduct similar analyses of political media from 1955 and 1975, finding nearly no

outrage, even during these turbulent times in American politics and society. They

3In the case of Hungary, the most common strategy of blending populist outrage and nationalismis playing off of the “Trianon Syndrome”(Gerner, 2007). This refers to the 1920 Treaty of Trianonby which Hungary lost two-thirds of its territory at the end of World War I. This Treaty remains arallying cry with widespread purchase even today, particularly among members of the right.

55

argue that there has been a steep increase in the use of the “outrage genre” in the US

in recent years, and that this has to do with the changing information technology and

media landscape in which media outlets are in an ever tightening race for viewers,

listeners, and readers. They conduct interviews with media analysts, who stress that

media content has to have “pop” — that ineffable attractive element that drives

people to tune in or click the link. The authors note that “pop comes in many forms,

the two most familiar being sex and violence, but carefully negotiated shock is not

far behind” (Berry and Sobieraj, 2013, p. 81). These researchers are not alone in

noting a marked increase in outrage in the media. Indeed, Slate Magazine published

an article declaring 2014 the “Year of Outrage.” 4

3.3 Populist Outrage in ECE

In addition to the broad technological and media trends that drive the increase in

attack politics globally, the young democracies of ECE find themselves in a situation

so as to be particularly prone to populist outrage. One major reason for this is

the contexts of democratization and party system development after state socialism.

The legacy of communism produced highly atomized societies (Przeworski, 1991),

and civil society in post-communist states remained extremely difficult for political

parties to penetrate well into the democratic period (Howard, 2003). Additionally, in

our media-saturated age, parties in ECE worked to penetrate society from above by

appealing to an undifferentiated “people” instead of developing from the bottom up

from social groups such as unions and clerical organizations (Mair, 1996; Innes, 2002;

Katz and Mair, 2009), thereby laying the first step towards populist appeals in the

very nature of political discourse between competitors for public office. Perhaps more

4http://www.slate.com/articles/life/culturebox/2014/12/the_year_of_outrage_2014_

everything_you_were_angry_about_on_social_media.html

56

importantly, the nature of political dialogue during communism involved the constant

language of black and white Manichean struggle between good and evil (Jowitt, 1992).

This means that in many ways the us versus them nature of populist dialogue was

already baked into the post-communist cake.

Another unique aspect of the post-communist legacy is the drastic, forced so-

cioeconomic equality during state socialism that left a “flattened landscape” when it

came to economic issues, muting the socioeconomic cleavage that traditionally divides

party politics across Europe during the critical early years of party system formation

(Zielinski, 2002). Thus, the very starting point of post-communist party competition

meant that parties faced enormous barriers to differentiating themselves from com-

petitors and penetrating civil society in order to form stable bases of support. While

increasingly pervasive media, the shift towards catch-all parties, and the associated

trends of lowered party identification and partisan dealignment are evident in ad-

vanced industrial democracies as well (Dalton and Wattenberg, 2000), this trend is

doubly powerful in young democracies. In these circumstances, parties cannot rely

upon their ownership of particular issues or reputations for representing certain seg-

ments of society to differentiate themselves from competitors. As Innes argues, “for

all New Labour’s entrepreneuralism, Western Europe’s mainstream parties continue

to carry far more historical baggage than successful parties in Eastern Europe” (Innes,

2002, p. 100).

A uniquely high level of international constraint is also an important factor in

influencing the tone of the political dialogue. This constraint limits the ability of

domestic political actors to differentiate themselves using programmatic policy ap-

peals or clear ideological positions. Post-communist democratization on the doorstep

of the European Union (EU) provides an exceptional example of this dynamic. For

57

those ECE states that stood a chance of entering the Union, EU membership repre-

sented a near universal political goal (Vachudova, 2005). However, accession was no

small task for these states in transition. The myriad requirements embodied in the

voluminous acquis communautaire set a very difficult path in front of these states

including thousands upon thousands of individual policy requirements. This meant

that political policy making was more about implementing EU policies off the shelf

than developing distinct party platforms (Innes, 2002; Follesdal and Hix, 2006). This

did not end with successful accession, as the EU has developed great authority in a

number of issue areas that typically make up the axes of party competition. Indeed,

this is the same dynamic that compelled Greece’s radical leftist Syriza to swallow the

bitter pill of center-right austerity programs.

Due to this constraint, young parties in ECE struggling to attract voters by dis-

tinguishing themselves have often been unable to offer policy alternatives on crucial

issues, and in many key domains their positions have become indistinguishable from

both the positions of the EU and their competitors (Mair, 2013; Nanou and Dorussen,

2013). This makes it difficult for parties to attract support by offering programmati-

cally distinct platforms, and explains why parties in post-communist Europe are ideo-

logically inconsistent in terms of policy output (Tavits and Letki, 2009) and “vacuous

and fickle when it comes to policy content”(Carey and Reynolds, 2007, p. 271). The

structural constraint upon politicians in responding to constituents’ demands makes

the situation ripe for populist outrage against the political establishment: “Unable

to deliver on the promises they have given, the structural, functioning conditions of

democratic systems and their representatives become the target of demagogic dis-

course, being accused of betrayal, corruption, unaccountability, and incompetence”’

(Meny and Surel, 2002, 9).

58

These barriers leave party leaders with a dilemma: they must find a way to attract

support without relying upon programmatic or ideological appeals, and appeals to

populist outrage is a key way to do so. Indeed, politicians know they can evoke anger

to increase political interest (Ryan, 2012) and strengthen partisanship (Valentino

et al., 2011). Moreover, moral anger can provide a solution to the problem of differ-

entiation from competitors by activating notions of group boundaries (Haidt, 2003;

Tooby et al., 2006). Thus, political elites can leverage populist outrage in a process of

“emotional entrepreneurship” yielding political support by differentiating a group of

morally upright patriots from their opponents without reliance upon programmatic

or ideological substance (McDermott, 2010).

This chapter focuses on testing the first hypothesis of the dissertation: that pop-

ulist outrage increased in the period surrounding and following European Union ac-

cession. If populist outrage is indeed on the rise in ECE, many troubling implications

for the quality of democracy in these young democracies follow. These appeals likely

drive attitudes towards the groups and actors they are aimed against. Accusations

of corruption, bribery, and state capture among the political elite are ubiquitous in

the region’s political dialogue, and they help explain the region’s strong trends of

anti-political, anti-incumbent, and anti-party attitudes. By constantly angering vot-

ers against all political options, both party system volatility via protest voting and

an eventual disengagement from politics because of disgust with all options may be

the end result. While the effects of political attacks is a huge and contentious field

of scholarly inquiry, a meta-analysis suggests that at the very least they decrease

political efficacy (Lau and Rovner, 2009), which is a particularly harrowing possibil-

ity in post-communist countries that started out with uniquely stunted civil societies

(Howard, 2003). Finally, the efficacy of these appeals against oft-maligned minority

groups is even more troubling. For instance, in Hungary stories of “gypsy crime and

59

terror” 5 are extremely common in the media and in the political dialogue of the

right. These stories frame the Roma minority as taking great resources from society

in the form of social welfare programs and human rights legislation while attacking

and stealing from hard-working citizens who lack such special privileges. This nar-

rative resonates with large portions of the Hungarian population. One recent study

shows increasingly prevalent anti-Roma sentiment among the Hungarian public, and

that these attitudes account for a great deal of the support for Jobbik, the party of

the extreme right (Karacsony and Rona, 2010). In addition to increasing anti-Roma

sentiment, anti-Semitic rhetoric is also associated with rising anti-Semitic prejudice.

A 2012 report by the American Anti-Defamation League analyzing attitudes towards

Jews in Europe shows Hungary to have both the highest absolute level of popular

anti-Semitic attitudes in Europe as well as the greatest negative change over recent

years.6 Recently, Jobbik politicians have even suggested the necessity of making lists

of all Jewish Hungarian Members of Parliament for the purposes of national security.7

3.4 Analysis: Mapping Populist Outrage

Operationalizing and measuring populist outrage dialogue in political discourse

over time presents several empirical challenges. To count as an instance of populist

outrage dialogue, there must be a clear transgressing actor, or “them,” that has com-

mitted an egregious act that violates deeply held notions of morality or common

decency against a virtuous “us.” Differences of opinion, instances of disagreement

about the desirability of certain policies, and heated debate about what causes what

5cigny bnzs

6http://www.adl.org/Anti_semitism/adl_anti-semitism_presentation_february_2012.

pdf

7http://www.euronews.com/2012/12/03/outrage-in-hungary-over-jewish-register-

remark/

60

in society would not count as populist outrage, nor would allegations of honest mis-

takes, mismanagement, or even incompetence. The key is a clear, flagrant, purposeful

violation of morality by a bad guy who knowingly and intentionally harms the people.

Given these criteria, the best measurement approach would be to read, watch,

or listen to massive amounts of political communications, coding for the presence or

absence of such outrageous violations within each. This is precisely what Berry and

Sobieraj (2013) did in the only similar content analysis of outrage discourse. They

focused on the United States and ran a 10 week content analysis of talk radio, blogs,

newspapers, and television programs, employing a team of research assistants to take

in these sources and code them one by one. In order to devise a codebook, each

member of the research team found 20 instances that they thought embodied outrage

discourse and then worked through exactly what was going on in each to inductively

produce a classification scheme. The result was a coding protocol including thirteen

types of outrage.8 Researchers then determined a watching schedule for a variety

of media by choosing days of the week at random for each program or show. They

then marked each instance of outrage discourse and sorted them into their sub-types,

along the way finding an enormous amount of outrage in the media. Indeed, their

rough average is that outrage is used about once for every 90 seconds of political

programming.

Given more modest resources, the primary data source here focuses solely upon

newspaper articles about Hungarian domestic politics. Not only is it possible to go far

back in order to track the nature of the political dialogue as represented in newspapers

over time, but Berry and Sobieraj (2013) found print media to have a somewhat lower

level of outrage discourse compared to television, radio, and web sources. Therefore,

8The thirteen types they found were (in descending order from most to least frequent in theiranalysis): mockery/sarcasm; misrepresentative exaggeration; insulting language; name calling; ideo-logically extremizing [sic] language; emotional display; belittling; verbal fighting/sparring; obscenelanguage; emotional language; character assassination; slippery slope; conflagration

61

while I must limit the analysis to one media format, it is the format most likely to

provide a stringent test of the hypothesis that populist outrage is present and rising

in Hungarian political coverage.

In order to perform this analysis, I collected data from the Hungarian News Agency

(Magyar Tavirati Iroda – MTI). This is a news organization with the public mandate

to inform the citizenry, and its international analogues would be the Associated Press

or Reuters news services. I chose it as a data source for two primary reasons. First,

as is the case with the Associated Press or Reuters, MTI aims to present the facts of

a news event without editorializing, dramatizing, or presenting an argument of any

kind. Thus, these “the facts and only the facts” stories are an especially hard test

for detecting populist outrage in the media as they are rarely emotional. To include

outrageous populist dialogue, this dialogue must be real and present in other news

sources as MTI does not work to break scandals or engage in muckraking, but will only

report on such stories when they have gained prominence in the political dialogue of

the country more broadly. The second reason for using this data source is availability.

The MTI makes all of their news stories available for download on their website going

back to the early 1990s, whereas most other news sources in Hungary have spotty

coverage at best.9 Therefore, using the MTI online archives10 it is possible to sample

news stories from elections dating back all the way to 1990.

9Indeed, in the process of analyzing populist outrage in the Hungarian media I used web scrappingto gather stories by the major newspapers of the right and left but it became apparent that thestories that were kept on their web pages available for scraping were not a complete archive andthat there is no way to find out the logic of which stories remained up. It seems that stories werearchived depending upon idiosyncratic factors, such as the webmaster of a given paper at the time.Coverage is good for the 2010 and 2014 elections, but the farther back in the individual newspaperarchives one goes the worse the coverage gets. A similar problem exists in other web archives ofthese sites such as the Way Back Machine. MTI, on the other hand, keeps meticulous and extensivearchives of every news story they have run since 1988.

10http://archiv1988-2005.mti.hu/Pages/HirSearch.aspx?Pmd=1

62

0

20

40

60

80

100

120

140

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

Outraged

Figure 3.1: Frequency of the term “outrage” in domestic political coverage from theHungarian News Agency

63

Figure 3.1 shows the prevalence of the terms “outrageous,” “outraged,” and “an

outrage” in MTI using their online word search of all domestic news coverage. This

provides a quick feasibility check of this method of testing the hypothesis that populist

outrage is on the increase by using the MTI archives. Though there was no discernible

increase in the average number of articles covering Hungarian domestic political affairs

in these years, there was a noticeable increase in the use of these terms over recent

years. This is particularly interesting given that the most drastic increases in the use

of these terms in Hungarian political coverage directly precedes the 2010 election that

marked the realignment of Hungarian party politics and the beginning of democratic

backsliding in the country. These searches as well as many others using similar words

we would expect to find in stories that feature populist outrage show increases in the

predicted manner.

A real test of the hypothesis that populist outrage increased in the years between

EU accession and the unraveling of Hungarian democracy requires a comprehensive

measurement of the style of populist outrage. Providing this measurement technique

presents several empirical challenges, as there are no existing tools for measuring the

emotional content of language in Hungarian, let alone for assessing the presence of

populist outrage in particular. In order to conduct this test I compiled two compli-

mentary dictionaries of populist outrage words targeted at picking up these appeals

by counting the number of times words commonly used in them appear over time in

the MTI. Each of these dictionaries include their own advantages and disadvantages,

but their problems compliment one another and allow for a multifaceted view of the

use of populist outrage dialogue in the MTI.

64

I apply these dictionaries to a database of articles taken from MTI election years

from 1994 to 201411. I compiled a database of one article every other day for the six

months year leading up to the elections. To ensure the articles were selected randomly

I searched for domestic political coverage for the target date, randomly selecting one

of the search pages listed at the bottom of the page, and then downloading the first

article.12 The resulting database consists of just under 100 articles per election year

for a total of 570 articles.

3.4.1 Outrage Dialogue in the Hungarian Political Discourse

While the hallmark of populist outrage is the presence of the accusation of a

violation of a deeply held tenet of morality, these violations are typically accompanied

by a lexicon of identifier words. Given the importance of morality to the type of anger

that is targeted with populist outrage appeals, many of these words signify immorality

and base actions. Thus, for the first and primary analysis, I computed a dictionary

designed to pick up on the specific language used in making appeals to populist

outrage based upon these accusatory, immoral words. This analysis uses a dictionary

of about 200 words that target the sorts of vocabulary we find in populist outrage. For

original source material I turned to political speeches and advertisements, extracting

common words that might indicate the presence of the accusation of a violation of

morality in the context of Hungarian politics. Having computed this list, I checked

with Hungarian native speakers and adapted it based upon their input. This analysis

is admittedly subjective and prone to researcher bias when compared to the indexes

111994, 1998, 2002, 2006, 2010, & 2014. There was an election in 1990 but given the uniquenessof this campaign as the first free elections following the fall of communism I expect these electionsto be highly exceptional and do not have any expectations regarding them.

12In rare instances the search parameters failed and the first article was actually about a politicaltheme in another country or a non-political topic. In these instances I selected the first article aboutHungarian politics.

65

compiled using other scholars’ coding schemes (see below), but it is the only way to

measure populist outrage with a metric specifically targeted for this purpose. The

full dictionary can be found in Appendix 5.

Measuring these words requires adapting the database such that the words will

register in all of their various forms as they change due to Hungarian grammar. Hun-

garian is an agglutinative language, meaning that sentences are constructed primarily

by adding suffixes and declensions to common word stems. Whereas in English we

utilize word order, prepositions, prefixes, and different spellings, Hungarian typically

relies upon adding more and more endings to give meaning and context. This lends

itself quite well to a dictionary approach, and the words extracted from the ANEW

were therefore parsed to these stems such that the software would detect them and

their variations as they are conjugated and declined in Hungarian sentence structure.

One caveat is worth noting before proceeding to results. Even given the presence

of 200 common populist outrage words, I expect the proportion of these words relative

to all other words in the MTI database to be low. Given the research goal of discrim-

ination from other political news, the particular words included in the dictionary are

the most inflammatory and are an extreme indicator of populist outrage. However,

the accusations at the heart of populist outrage take place mostly by using common

language. Take, for example, the most ubiquitous anti-opposition advertisement dis-

tributed during the parliamentary campaign of 2014, the “Seven sins of the Gyurscany

coalition” described in the the introduction and pictured and translated in Appendix

1. While these statements are paradigmatic cases of populist outrage, including ac-

cusations of immoral violations and stealing from the most vulnerable members of

society, they use obviously outrageous words like “sin” and “stole” rarely. Indeed,

only a .055 proportion of words in these appeals are in the dictionary. Given all of the

other words needed to make meaningful language and the ability to use these words

66

sparingly in order to paint a very nasty picture of a political opponent, a low propor-

tion may be instructive if there is a general trend of change. Moreover, that the MTI

is typically the least emotional news source in the country means that an increase

there is probably matched by an increase in other outlets that likely resort to these

outrageous words and the general appeals much more. Thus, while low proportions

may appear to be weak evidence, given the source material and a dictionary that con-

tains fewer than 200 words of the 100,000+ words in the Hungarian lexicon (Kenesei,

2004), it is changes in the proportion rather than absolute levels that are of concern

here. A more comprehensive dictionary would trade discriminatory power for larger

proportions, which is antithetical to the present research objectives of conservatively

tracking movement over time.

Figure 3.2: Proportion of ANEW outrage words out of total per election year

67

Figure 3.3: Percentage change in ANEW outrage words from previous election year

The analysis was competed using the Yoshikoder software package (Lowe, 2015)

in order to chart the frequency of words in this populist outrage dictionary as a

proportion of total words in the collection of news stories in my database scraped

from the MTI archives. The results from this targeted approach can be found in

Figures 3.2 and 3.3. While in 1994 the proportion was .006, the years surrounding

the 2004 accession saw a marked increase to .009 in 2002 and .01 in 2006. Indeed,

Figure 3.3 shows the largest percent change from the a previous election to be in 2002

as EU accession requirements were going into effect in earnest. However, these data

also show the increase to continue through the most recent election, reaching a height

of .013 in 2014. These proportions, while low, show change in the expected direction

rather drastically. All in all, the proportion of these outrage words in election coverage

in the MTI increased 116.7% between 1994 to 2014.

68

3.4.2 Adaptation of the ANEW Dictionary

In order to construct a dictionary of outrage words that does not rely upon my

own subjective assessments of words or articles, I computed a dictionary of populist

outrage words by translating portions of the Affective Norms in the English Language

(ANEW) dictionary (Bradley and Lang, 1999). The ANEW is a tool widely used by

psychologists to determine the emotional content of written and spoken communi-

cations. It is a comprehensive dictionary of the most commonly used words with

several numeric ratings of the emotionality of these words on three different scales.

These data are based upon commonly understood perceptions of a word’s affective

content and were generated by bringing participants into the lab and presenting them

with different combinations of words in a random order. Participants were instructed

that they would see words and be asked to rank them on three different kinds of

feelings: Happy vs. Unhappy (affect), Excited vs. Calm (arousal), and Controlled vs.

In-control (dominance). The actual scores in the ANEW database are the average

affect, arousal, and dominance scores of all participants for each word on a scale from

1 to 9, with higher scores indicating greater happiness, arousal, and control. Because

outrageous language is best operationlized here as a combination of negative affect

and high arousal, the dictionary of populist outrage is comprised of words from the

ANEW which score below 4.0 on affect and above 4.0 on arousal.13 The complete

ANEW adaption consists of 262 words that met the threshold for negative affect and

high arousal. The list of these words along with translation and scores for arousal

and affect is in Appendix 6.

13As the scale ranges from 1 to 9, drawing the line at 4.0 for both affect and arousal means tobe included in the database the word must have lower affect than it has high arousal. This was adecision made during coding as low affect clearly delineated undesirable words, whereas some wordsat medium levels of arousal, such as coward (arousal = 4.07), corrupt (arousal = 4.67), and greed(arousal = 4.71) would be expected to be integral to populist outrage dialogue.

69

Again, a few caveats are worth mentioning before proceeding to results. First,

the ANEW database was constructed by psychologists to understand the emotional

undertones of words in general, and not in political scenarios. Thus, there are many

words in the database, from “mosquito” to “syphilis” to “gangrene,” that, while out-

rageous in many circumstances, are not commonly used in populist outrage dialogue.

Still, the majority of words in the dictionary are accurate signifiers of populist out-

rage, while some like “betray,” “evil,” and “vandal” are prime examples of populist

outrage. The most significant problem is that these data consist of perceptions of

how native English speakers feel about words in English, which may be different from

how they feel in Hungarian. When this approach has been used in other languages,

scholars have sometimes been able to replicate the study methodology to compute

“ANEW” dictionaries in foreign languages by having native speakers of these other

languages rate words in a manner that replicated the original study methodology.

This is true of the Spanish (Redondo et al., 2007) and Polish (Imbir, 2015) versions

of the ANEW. Unfortunately this is well beyond my resources. On the other hand,

the inductive approach is calibrated specifically for the task at hand, but is open to

criticism of being biased by the researcher towards finding its more dramatic results.

Thus, it is through using both of these methods and a search for common trends that

we can triangulate the relative changes in populist outrage in Hungary from election

to election.

The results from the ANEW analysis of the MTI data archive are pictured below.

Figure 3.4 shows the proportion of words in the article database that are found in

the ANEW outrage dictionary, while figure 3.5 displays the percentage change in this

proportion each year relative to the previous election year. Overall, the proportion of

words is quite low, ranging from .005 in 1994 to .008 in 2014. Thus, the proportions

are even lower than in the inductive dictionary, but this is to be expected given

70

the ANEW was developed to be a very broad measure of the use of emotion in

language, and the 262 words that met the threshold for negative affect and high

arousal thus represent a subset of the negative words that we would expect to find in

populist outrage dialogue.14 Nevertheless, even this rough and conservative measure

presents a clear picture of an increase in outrage over time. The lowest proportions

of outrage language in political coverage leading up to a parliamentary campaign

were 1994 and 1998, both with .005% of the words used represented in the ANEW

outrage dictionary. However, as EU accession negotiations and the associated policy

convergence started and the associated policy convergence took off in earnest, the

largest jump in outrage language took place. In 2002 the proportion of words in

the ANEW outrage dictionary, while still low, jumped nearly 35% relative to the

election in 2002. Accession took place in 2004, and while there was a small decline

in outrage in the 2006 election relative to 2002, by 2010 and 2014 the proportion of

outrage language grew to ever higher and higher levels. Indeed, according to these

data, the proportion of outrage in political coverage increased by 69% between the

1994 elections and the most recent election in 2014.

14Applying the ANEW outrage dictionary to the text of the 7 sins pamphlet that is a paradigmaticcase of populist outrage reveals a .009 proportion of its words are in the ANEW dictionary, whichis right on target with the findings in the MTI database.

71

Figure 3.4: Proportion of ANEW outrage words out of total per election year

Figure 3.5: Percentage change in ANEW outrage words from previous election year

72

3.4.3 Populism in Party Appeals

The argument above predicts that in the contexts of increasingly pervasive and

fragmented mass media technologies and in the run-up to and aftermath of EU ac-

cession, the political dialogue in ECE increasingly features attack politics in general

and populist outrage in particular. This implies both an increase in populism and an

increase in outrageous negativity at the cost of programmatic differences. While the

previous analysis finds evidence for an increase in outrageous language in the media

coverage of political campaigns, it does not measure populism or what parties them-

selves are saying. To provide an analysis of populism in party appeals I draw upon

data from the Manifestos Project (Volkens et al., 2014). The Comparative Manifestos

Project (CMP) data provide information on the content of the platforms of each po-

litical party. CMP coders break each party platform into individual statements and

then categorize each statement as belonging to a specific topic out of an extensive

coding scheme. They then divide the total number of statements in the manifesto

by the number of statements dedicated to each topic. The resulting data measure

the proportion of each party manifesto dedicated to each topic in a given election.

Here these data provide an opportunity to construct a measure of the salience of cer-

tain populist issues coming from parties in their manifestos and to observe this over

time. The working hypothesis is that in the period around European Union accession

populism will have increased.

Scale construction: An overall populism scale consisting of three sub-scales was

constructed. The first scale captures a nationalistic / anti-EU and anti-international

dimension sitting at the heart of many populist appeals. The second scale includes

anti-communist and anti-corruption indicators that are particularly salient in the

wake of the legacy of state socialism in ECE. Finally, the third sub-scale captures

73

economic populism, consisting of vague promises to promote economic policies favor-

ing the common man or to build the infrastructure necessary to benefit the country’s

economy. These three scales were aggregated into an overall scale that captures pop-

ulist appeals coming from parties in the broadest terms. Descriptions of all of the

individual measures included in these scales is in Appendix 2.

Figures 3.6 through 3.9 show the overall populism rates of each major party in

each post-communist election of the Visegrad four countries of Hungary, the Czech

Republic, Poland, and Slovakia. Each individual figure charts the populism rate of

individual parties over time, with x-axes representing the election year and y-axes

showing the percentage of that party’s platform devoted to the topics in the overall

populism scale. While there is great variation from case to case and party to party,

there is a common trend throughout most of the cases: as time went on the salience

of populist topics increased. This is not true unanimously, with parties in Slovakia

being particularly exceptional. However, the broad trend is increasing populism.

74

(a) Fidesz (b) Socialists

(c) Democratic Forum (d) Liberal Democrats

Figure 3.6: Overall Populism in Hungary

75

(a) Civic Democratic Party (b) Christian Democratic Union

(c) Social Democratic Party (d) Communists

Figure 3.7: Overall Populism in the Czech Republic

76

(a) Civic Platform (b) Democratic Left Alliance

(c) Law and Justice (d) Polish Peasants Party

Figure 3.8: Overall Populism in Poland

77

(a) Party of Democratic Left (-2002) Smer (2002-) (b) Movement for a Democratic Slo-vakia

(c) Democratic and Christian Union (d) Slovak Nationalist Party

(e) Hungarian Coalition / Bridge

Figure 3.9: Overall Populism in the Slovak Republic

78

The three subscales lend credence to the general picture presented above. The

weakest movement was with regards to anti-communist populism. Contrary to initial

expectations, the general anti-communist/anti-corruption scale did not show much of

a pattern over time. This is likely due to the decreasing salience of anti-communist

appeals as the transition from the previous system faded into the past. It may also

have to do with the the nature of the Manifestos Data, focusing as they do upon

the officially published party platforms that might not be the place where parties

leverage their specific corruption allegations. The economic populism subscale shows

that, with few exceptions, economic populism in official party manifestos has increased

dramatically. Indeed, for cases like the Czech Republic and Poland, it is increases

in economic populism that make up a majority of the overall increase in populism

documented above. These complete figures are included in Appendix 3. Finally,

nationalistic and anti-EU populism has also risen in recent years across cases. As

Figure 3.10 shows, this is particularly true of Hungary, and of the governing Fidesz

party most of all. Remaining results from this subscale are presented in Appendix 4.

79

(a) Fidesz (b) Socialists

(c) Democratic Forum (d) Liberal Democrats

Figure 3.10: Nationalistic / Anti-EU Populism in Hungary

80

3.5 Discussion

Together, the dictionary and Manifesto Project analyses provide strong evidence

that populist outrage has increased in Hungary and ECE. Both the inductive dic-

tionary of outrage words and the translation of the Affective Norms for the English

Language pick up substantial increases in the level of outrageous dialogue in political

news stories in the Hungarian News Agency, while Manifesto Project data show that

political parties have made more and more populist appeals across ECE. Moreover,

each reveals that the increase took place most sharply in the early 2000s. While

correlation is by no means causation, the logic of international constraint and the

sharp increase in populist outrage measured in both the vocabulary of political me-

dia coverage and the nature of official party appeals suggest that conditionality from

the European Union plays an important role in this process.

As noted in the analysis, each of the data collection and analysis methods pre-

sented above have their own shortcomings. To account for these problems, I conducted

robustness checks to ensure that individual articles that included a large number of

populist outrage terms did indeed count as populist outrage. I selected a random

sample of articles that had a high level of populist outrage terms according to my

adapted ANEW and inductive dictionaries and checked them for both overall outra-

geous tone and for the presence of a violation of a moral standard by a well-defined

transgressor. Confirmatory findings in this robustness check lend further credence to

the analyses, but the subjective nature of the inductive dictionary, translation and

applicability issues with the ANEW dictionary, and low proportions in both analyses

remain important caveats to note. Only through replication and expanding to new

cases can we confirm these results and determine how widespread populist outrage in

81

politics is and what structural factors provide incentives for political elites to resort

to this tactic.

While the European Union represents a uniquely intense level of populist outrage,

there are many reasons to believe that this political tactic is increasingly prominent

across numerous contexts. Moreover, the international constraint that contributes to

elites resorting to outrage is growing. Where foreign aid or vital loans come along

with required policies that are not up for debate by domestic political elites, they

may contribute to a rise in populist outrage. Such is likely the case with economic

and spending policy wherever countries accept loans from the International Monetary

Fund, or where donors or major trading partners require specific policies. By fleshing

out this important causal mechanism at the heart of the democratic deficit, this

research draws attention to the imperative of designing international aid programs

such that national political decisions remain meaningful in the eyes of voters. Indeed,

prominent scholars have pointed to the role of neoliberal institutions and technocratic

leadership in creating a political void (Mair, 2013; Hay, 2007), and this work explains

what moves in to fill this void: populist outrage.

82

Chapter 4: Immediate Emotional Reactions to Outrage:

Anger, Disgust, and the Role of Partisanship

4.1 Introduction

Hungarian parliamentary campaigns of late have been nasty in tone. An infamous

2010 advertisement from Jobbik, the new party of the far-right, offers a vivid illustra-

tion. The ad features wholesome-looking Hungarians swatting an unseen pest from

their faces as the buzz of a mosquito drones on in the background. In one scene, a

young woman walks down the street as a hooded figure lurks behind her and asks:

“Are the gypsy criminals allowed to do whatever they want? Finally, Jobbik party

leader Gabor Vona slaps his arm, killing the pest and ending its buzzing, just as he

declares: “We’ve had enough of this parasitism. If you have as well, on October 3rd

vote for Jobbik.1 Fidesz, the party of the center right that went on to win the 2010

election, also utilized a strategy of vituperative accusation, calling the Socialist gov-

ernment traitors and enemies of the state and arguing that the real transition from

Soviet style rule would only happen when the center-left government was gone.

The strategy of populist outrage exemplified by these appeals was an an enormous

success for both parties. Jobbik stunned political commentators and broke into Hun-

garian electoral politics with almost 17% of the popular vote, becoming the second

largest opposition group in the Parliament and nipping at the heels of the humiliated

1http://www.youtube.com/watch?v=s1TV10CNz90

83

Socialists who, having governed over the preceding eight years, won only 18% of the

vote. Their returns in 2014 were even greater. Jobbik’s strategy, predicated largely

upon accusing the impoverished Roma community in Hungary of egregious moral

outrages, accusing the international community of conspiracies to rob Hungary of its

wealth and sovereignty, and embodied by the haunting campaign slogan “Hungary

for the Hungarians,” is further evidence of what campaign strategists have known for

a very long time: appeals to fear, anger, and other emotions can be a powerful means

of attracting support on election day.

As the previous chapter demonstrated, these appeals rose drastically in Hungary

in the 2000s. Having traced the increasing prevalence of populist outrage in the Hun-

garian political discourse, this chapter examines emotional reactions to these attacks

as a first step to understanding their political consequences for public opinion and

political behavior. Thus, it confronts the dissertation’s second set of hypotheses: that

appeals to populist outrage elicit anger and disgust (H2 ), and that the presence and

strength of partisan and ideological predispositions predicts anger (H2a), while the

lack of the perceptual screens that accompany partisanship and ideological leanings

predicts disgust (H2b). To do so, this chapter introduces several sources of original

data that will illuminate multiple aspects of the emotional and attitudinal effects of

populist outrage throughout the following chapters. First, a survey of 1,000 Hungar-

ian adults demonstrates the presence of large amounts of anger and disgust among

the Hungarian public. Next, two experimental studies — an online pilot and a lab-

in-the-field experiment — show that outrageous accusations of moral transgressions

drive these reactions. These data also demonstrate that the strength and presence of

partisan predilections is a key determinant of whether a citizen feels anger or disgust

in response to these appeals, raising important questions regarding the origin and

role of partisanship and its relationship to emotional response.

84

4.2 Anger and Disgust towards Politics

The central contention of this dissertation is that political elites exploit accusations

of immoral behavior by their competitors to elicit anger and inspire political passion,

but that many citizens feel disgust instead of anger in response. Indeed, given the key

role of disgust in regulating morality, I hypothesize that multiparty politics dominated

by loathsome moral transgressions leads to disgust towards political competition.

As disgust leads to an intense rejection of its source, this possibility merits careful

consideration.

What determines whether a political attack causes anger or disgust? These emo-

tions are elicited through different processes; anger typically involves an appraisal of

threat, wrongdoing, or blame, while disgust is the result of associations with some-

thing nasty or offensive (Giner-Sorolla, 2012). Thus, anger is the result of an evalu-

ation that some actor has committed a transgression, meaning attributions of blame

and intentionality are central to the response (Quigley and Tedeschi, 1996; Petersen,

2010). Conversely, repeated association with something offensive, either to the phys-

ical senses or to deeply held convictions about morality, is sufficient to elicit disgust.

Given these distinct elicitation processes, two factors stand out as key to determin-

ing whether a political outrage will anger or disgust a citizen exposed to it. Because

disgust comes about from repeated associations of an object with unpleasant affect,

the next chapter will evaluate the hypothesis that over time anger at the transgres-

sions at the heart of populist outrage transitions into disgust. This chapter evaluates

a more immediate variable that moderates the relationship between outrageous moral

transgressions and the associated psychological response: the strength and presence

of perceptual screens associated with partisan and ideological leanings. Ever since

the The American Voter (Campbell et al., 1960), we have known that partisans see

85

things in a way that favors their own side. Previous work on emotions in politics

has documented that anger works well on partisans, strengthening their predisposi-

tions and spurring them to action (Valentino et al., 2011). Partisans operate with

a “perceptual screen” filtering their understanding of political events (Westen et al.,

2006) and affecting the way they view all political news, groups, and themes (Lodge

and Taber, 2013). When it comes to populist outrage, they are more likely to see

one side as worthy of blame for either committing the alleged transgression or for

manufacturing it for political gain. Meanwhile, those in the middle without a stake

in the matter will be less likely to judge either the accuser or the accused of blame,

instead feeling a general sense of offense at a morally reprehensible situation that

leads to disgust. When that situation is multiparty political competition, the very

contestation at the heart of democracy becomes the object of disgust.

4.3 Survey: Democratic Commitment, Anger, and Disgustamong the Hungarian Public

An original survey provides unique data on the emotions citizens in Hungary feel

towards their political system, as well as an initial test of whether Hungarian feel anger

and disgust towards politics. In November 2014, Ipsos Hungary used cluster sampling

to recruit a sample of 1,000 Hungarian adults for a survey of, among other things,

“Attitudes towards Hungarian politics and society.”2 This questionnaire was part of

an omnibus survey including demographic indicators and other modules from private

2The author would like to thank the Mershon Center for International Security Studies at the OhioState University for generous support of this data collection. The resulting sample is representativeof the Hungarian populace, but to ensure accurate statistical representation, controls are used inthe models to follow. The survey also includes survey weights based upon the most recent census.All models were estimated both with and without weights. Unweighted models are reported here,but no substantive differences were present in the weighted models, which are reported alongsidecomplete survey sample statistics in the Appendix in Table B.3.

86

businesses.3 Ipsos’s interviewers visited participants’ homes and conducted computer

assisted personal interviews (CAPI), allowing participants privacy in responses on a

laptop while also providing the personal connection and accountability of an in-person

interview.

The survey module consisted of two sets of measures: indicators tapping emotional

responses to politics and indicators measuring political attitudes and self-reported be-

havior. At the beginning of the module, participants were asked to rate how politics

makes them feel in general, and then were asked to do the same to several instances

of outrageous political behavior that varied both by the age of the scandal and the

political side that was implicated.4 These events were incidences that had received

wide coverage either over recent years or in the 2014 campaign. Participants were

asked to rate their dominant emotional reaction to each, choosing only one response

per stimulus from a menu of: happy/pleased; proud/enthusiastic; angry/outraged;

disgusted/repulsed; anxious/afraid; apathetic/ unconcerned or Don’t know / not fa-

miliar with this. The second half of the survey asked a series of questions regarding

political attitudes, such as intention to vote and desire for democracy, which we will

return to in Chapter 6 in order to distinguish the political ramifications of anger and

disgust from one another.

The first step is to evaluate the hypothesis that anger and disgust towards politics

in general, and towards instances of outrageous moral transgressions in the political

realm in particular, exist among the Hungarian public (H2). To do so, emotions

variables were computed by adding the number of times a participant felt a given

emotional reaction towards the various political prompts. For instance, a score of

3The other modules in the particular omnibus survey examined the views consumers hold towardsvarious products and thus is unlikely to have influenced participants’ political evaluations duringthe module reported here.

4There were three events implicating the left and three implicating the right.

87

Table 4.1: Negative Emotions Towards Politics

Instances 0 1 2 3 4 5+ Mean count Towards PoliticsDisgusted 481 166 153 105 50 45 1.27 17.5 %

Angry 323 192 141 121 80 143 2.01 23.1 %Apathetic 387 199 150 91 35 138 1.80 21.4 %

three for the Anger variable indicates that anger was the dominant emotional response

reported by a given participant for three out of the eight prompts. Table 4.1 reports

levels of these variables among the population, sorting individual respondents by the

number of times they reported feeling disgust, anger, and apathy as their dominant

response.

The survey data suggest that anger and disgust are widespread feelings toward

politics. A majority of participants report feeling disgust and anger at least once, and

often many more times than that. Though anger is the most common reaction to the

prompts, disgust is also widespread. Indeed, 10% of the population reports feeling

disgusted at least by half of the prompts. The final column of Table 4.1 isolates the

indicator asking participants to rank their overall feeling towards Hungarian politics.

The trends here are similar to what we see in the total count variables, with 17.5%

reporting feeling disgust over any other emotion when it comes to Hungarian politics

in general, 23.1% more angry than anything else, and 21.4% simply apathetic. 5

Another key consideration is whether disgust is the same thing as apathy. The

data indicate participants who respond that they feel disgusted are not apathetic.

Disgust and apathy are negatively correlated at r = -.303 (Spearman correlation

significant at p <.001), garnering support for H2. Indeed, this is similar to the

5The majority of the participants who did not feel one of these emotions felt anxious (22.8%).8.7% answered “Don’t know” or refused, while only 2.2% responded “Happy/content.”

88

degree of the correlation between anger and apathy at r = -.368 (Spearman corre-

lation significant at p <.001). Finally, while anger and disgust are emotions that

frequently co-occur, self-reported anger and disgust are independent in this sample.

The Spearman correlation between these two emotions is r = .029, which is far short

of statistical significance (p = .368). Therefore, contrary to the view that disgust is

simply a lower level of negative affect than anger, disgust is distinct from anger as

well as from apathy.

4.4 Experimental evidence

The survey demonstrates that there are high levels of anger and disgust among

the Hungarian public when it comes to politics, but it cannot document the source of

this negative affect. For this purpose, two experimental studies, an online pilot and

a lab-in-the-field-experiment, randomly exposed participants in treatment conditions

to populist outrage and trace the subsequent emotional reactions.

4.4.1 Pilot Study

The first experimental study was targeted at understanding the immediate effects

of outrageous political attacks against the targets of the attacks in order to understand

the links between populist outrage, the emotional responses evoked, and associated

democratic attitudes. This study was also designed with the goal of showing that

moral anger, or anger that comes about from a violation of morality as opposed to

some sort of personal goal blockage — is a significant reaction to populist outrage.

It was conducted online in 2013 using participants drawn from a Qualtrics-Market

Cubed panel of Hungarian respondents.6 Panel members, who had a agreed to take

6I would like to thank the Mershon Center for International Security Studies at the Ohio StateUniversity for its generous support of this portion of the project.

89

part in periodic research and marketing surveys for modest compensation, were sent

the recruitment email. Those who chose to participate began the study with a pre-

test in which they answered a series of basic demographic and personality questions,

proceeded by reading one of three articles in the experimental conditions, and finished

the study with a survey including manipulation checks and measures of political

attitudes. The primary goal was to validate the central response of moral anger, or

outrage, to these accusations of moral transgressions and to explore the extent to

which moral anger differs from the more widely studied personal anger associated

with goal-blockage. This pilot was designed to include a single, strong treatment

condition instead of attacks from different sides working to elicit outrage among

distinct ideological groups (see the lab-in-the-field study). Thus, the pilot included

an attack that mirrored the most flagrant examples of populist outrage rhetoric in

the Hungarian media: attacks from the radical right Jobbik party against the (right

leaning) government of Viktor Orban’s Fidesz party.

There were three experimental conditions in the pilot because pre-testing and

preliminary fieldwork suggested that two control conditions were necessary. A pre-test

as well as discussions during fieldwork in Hungary suggested the possibility that the

very mention of politics in any form would elicit negative emotions on the part of many

Hungarian study participants. While this research design problem provides some face

validity to the theory presented above, it erects significant barriers in providing careful

experimental control. Thus, the first condition was a total control condition with a

non-political stimulus. This article had to do with European airline company WizzAir

constructing new training facilities in Budapest. To determine the effects of the mere

mention of politics upon study participants, I compare this non-political control to

a political control that makes political themes and labels salient without including

populist outrage. In this political control condition, participants read an article about

90

the government’s immigration policy in which politicians from the far right Jobbik

party stated that the Fidesz government was letting in too many foreign workers.

The text included accusations of bad policy, but not of outright corruption or lying,

and the language was kept neutral instead of resorting to the inflammatory rhetoric

featured in populist appeals to moral outrage. The third group was placed into the

treatment condition in which members of Jobbik attacked the Fidesz government for

taking bribes from multinational companies in exchange for granting their foreign

workers permits to gain employment in Hungary at a time when many Hungarians

are having trouble finding work. This was meant to violate fairness as the politicians

were lining their own pockets and throwing their constituents under the bus. Using

an economic issue as the stimulus is also beneficial in that it is both highly salient

today and allows for a limited analysis as to the extent that economic self-interest

drives the emotional reaction (which would indicate personal anger) relative to the

violation of fairness (which would indicate moral anger).7

Two major caveats are worth noting. First, an unexpected drawback of this panel

was the enormous attrition rate — approximately half of the participants that started

the survey did not read the manipulation or answered with a response set. While high

attrition is to be expected of any online panel, such a high attrition rate is unusual.

I included only those participants who spent over 30 seconds on the manipulation

in the analysis that follows. This leaves 217 valid responses.8 Second, as this was a

pilot conducted at a theory-building step in this research, this study focused on anger

to the detriment of measuring disgust. Indeed, the results of this pilot led to the

7For full text of these manipulations in English, consult the appendix.

8The differences between participants who completed the study and those that did not are mini-mal and largely random, with the group that did not complete the study tending to be only slightlymore male and young than those that did complete the study. They were statistically indistinguish-able in most respects, including self-reported ideology measured in the pre-test.

91

hypothesis that disgust underlies the link between the Hungarian political dialogue

and mass political behavior in this new democracy.

4.4.2 Emotional Reactions

The first step in the analysis involves ascertaining the effectiveness of the ma-

nipulation in order to determine whether populist outrage does indeed elicit moral

anger among participants. To identify the extent to which this is the case, following

the manipulation participants were asked to indicate how much the article they read

made them feel certain emotions. This manipulation check, which was adapted from

previous studies of emotions in politics (Valentino et al., 2011; MacKuen et al., 2010),

explicitly asked respondents to focus on how they felt while reading the article, not

their political opinions.

Figure 4.1 shows the mean levels of self-reported emotional reactions to the articles

used in each condition along with 95% confidence intervals.9 The control condition did

not elicit moral anger, or outrage, among participants.10 As expected, the political

situation did. However, it did so at a higher than expected level. In retrospect,

this is likely because in an attempt to provide greater control this political control

article also dealt with an accusation of sorts. Though the situation used did not

involve malicious or dishonest conduct, it did accuse the government of sub-optimal

immigration policy, in a country where immigration policy has long been a point of

contention. This helps to account for the large jump in feelings anger in the political

condition relative to the control. The treatment condition did elicit a larger degree

9Participants ranked how strongly they felt each emotion on a scale from 1 (not at all) to 4 (verystrongly)

10In this pilot study, participants were asked to rate both their level of outrage, or moral anger, andanger, or personal anger. This study was partially interested in exploring the distinctions betweenthese types of anger, a direction which the research subsequently moved away from. Here, bothmeasures are included to show the robustness of the anger elicited as well as to demonstrate theimportance of morality to anger.

92

(a) Outrage (b) Anger

(c) Anxiety

Figure 4.1: Emotional Reactions by Group in the Pilot

93

of anger relative to the political condition, and an analysis of variance confirms that

the difference between these two conditions is statistically significant with a p-level

below .05. Again, it is also notable that the political control condition also elicited a

great deal of self-reported outrage.

The treatment condition and the political control also elicited a great deal of self-

reported anxiety. Thus, while the manipulation successfully elicited outrage above

all other emotions, it was not a very clean stimulus as it also caused a host of other

negative emotions among study participants. Interestingly, the differences between

the political control and the treatment condition are statistically significant when

it comes to anger and moral anger, but the differences in self-reported anxiety are

not statistically significant between these groups. Thus, it would seem that the very

mention of politics, and not political accusations per se, increased anxiety among these

participants. This is a recurring theme in the data analyses that follow throughout the

dissertation: general political themes tend to make Hungarian participants anxious,

while populist outrage has the unique distinction of angering and disgusting them.

A series of ordered logit models sheds more light on the dynamics of the emotional

response to the accusation of a moral transgression. Table 2 presents the results of

models aimed at isolating the determinants of this outrage response, along with a

battery of control variables. The manipulation variable, Manip, is an ordered factor

variable coded 0 for the control condition, 1 for the political condition, and 2 for the

treatment condition, clearly emerges as the key determinant of experienced outrage. I

also examine the effects of economic self-interest as measured by a three item pre-test

battery asking respondents to rank the extent to which their family has been affected

by bad economic times.11 A moral foundations battery (Graham et al., 2009) was

included in the short survey that preceded the manipulation in order to measure the

11Specifically, unemployment, decreased wages, and increased difficulty in making ends meet.

94

extent to which perceptions of fairness account for the emotional reaction. This is

because outrage, or moral anger, should, as its name suggest, come about on account

of a violation of morality and not simply because a personal goal has been blocked.

Thus, if accusations of moral transgressions do elicit a unique type of moral anger,

participants who rank higher on the fairness moral foundation scale should be more

likely to respond with outrage, or moral anger. Finally, these models also include

several standard controls: an 11-point left-right ideological self-placement measure,12

political knowledge,13 age, and gender.14

The first model takes into account a series of potential explanatory variables that

might be expected to determine who feels greater moral anger. The manipulation

variable shows that self-reported outrage increased in a linear fashion from the non-

political control, to the political control, to the treatment condition, showing evidence

for the accusation of the moral transgression as a source of moral anger. Those respon-

dents who have felt negatively affected by the economy in recent years also became

more outraged according to this specification. None of the other potential predictor

variables contribute to self-reported outrage in the first model. That ideology does

not hold any explanatory power is surprising given that the manipulation involved an

attack on the right-leaning governing party. This may be partly due to the primary

attacker in the article being Jobbik, a party to the right of the governing Fidesz party,

which is also on the right.

The results of the second model suggest that the manipulations did not elicit moral

anger as a function of economic self-interest. The measure for economic problems and

the manipulation did not work together to predict the outrage response, and instead

12Higher values indicate positioning to the right

13An index constructed from a 7 question factual battery

14All variables were normalized to be between 0 and 1 for inclusion in the models.

95

Table 4.2: Self-reported outrage following manipulation

Variable Outrage Outrage Outrage AngerFull Econ Int. Fair Int. Fair Int.

Manip 3.9(.43)*** 2.98(.71)*** 1.62(1.15) 1.85(1.15)*Econ .84(.46)* -.18(.86) .99(.46)** .75(.44)*Fair .35(.7) -1.5(1.19) -1.14(1.2)

Purity .76(.72)Manip*Econ 1.9(1.26)Manip*Fair 3.35(1.59)** 1.54(1.57)Knowledge .3(.41) .29(.41) .38(.41) .2(.4)Ideology -.43(.65) -.46(.65) -.57(.65) .07(.62)

Religiosity -.9(.53)* -.79(.53) -.84(.54) -.34(.53)Age -.01(.01) .00(.01) -.01(.01) -.01(.01)

Female -.36(.28) -.3(.28) -.28(.28) .03(.27)

N 217 217 217 217AIC 495.8 494.1 492.5 546.9

the manipulation alone is a positive and significant predictor of self-reported outrage.

The third model shows that there is indeed something distinctly moral going on

when participants rate themselves as feeling outraged, as the interaction between

the moral foundation of fairness and the Manip is significant. This demonstrates

that the manipulations elicited greater self-reported outrage, or moral anger, among

those participants who rank fairness as more important in their judgments. This

supports the view that moral anger is the emotion dealing with violations of fairness

and justice, and elucidates the key mechanism linking populist outrage appeals to

the anger response. Finally, it does appear to be the case that respondents mean

different things when they report being angry and outraged, at least in Hungarian and

in this limited sample. As the Anger Fair Interaction model shows, the interaction

between fairness and the manipulation does not predict greater self-reported anger.

96

This informs a debate within the psychology literature on whether moral anger, as

opposed to personal anger, is independent of threats to the person and is instead a

reaction to a violation of morality (Batson et al., 2007). While this debate is somewhat

periphery to the current project, these data do provide compelling evidence to back

up Batson et al. (2007)’s claim that politics, where abstract notions of morality often

conflict with concrete self-interest, is likely a useful domain in which to explore the

distinctions between moral and personal anger.

The pilot study presented an initial empirical test of the link between accusatory,

attack politics and the emotion of moral anger. The appeals did cause anger in the

predicted manner, and there is some evidence in these data that anger can take both

moral and personal forms and that moral anger is the key reaction to accusations of

egregious moral transgressions. There were, however, a number of shortcomings with

this experiment, mainly having to do with the focus purely on the subtle differences

between the interrelated emotions of outrage (moral anger) and (personal) anger

at the expense of looking for signs of disgust. Another key issue was failure to

treat. There was a high rate of participants who did not spend adequate time on

the manipulation article and thus could not be included in the analysis. However,

the drop-out rate was not even across conditions, with those in the outrage condition

being less likely to read the entire manipulation than those in the political control,

and those in the political control being less likely to complete the article than those

in the non-political control. This in and of itself is suggestive that the mere mention

of politics is a turn off for many participants, and that outrageous accusations are

even more of a turn off and likely elicit disgust among many and lead to avoidance.

However, it does complicate the ability to look for differences between subjects in

experimental conditions. Thus, with the lessons from the pilot in hand, we turn to a

lab-in-the-field study.

97

4.5 Lab-in-the-field Study

To address the shortcomings of the pilot and add to the validity of its findings

through replication, I conducted a lab-in-the-field experiment in Budapest drawing

upon a very different sample of Hungarian participants.15 This study was conducted

with the help of the Political Ideology Lab of Budapest’s Eotvos Lorand University

and drew upon undergraduate psychology and education students as participants.

Students came to the psychology lab and were shown by the investigator to a com-

puter where the study was pre-loaded. They then completed the study, which lasted

between 20 and 25 minutes, and then left and received modest cash compensation

upon leaving. The primary purpose of conducting a lab-in-the-field study was to

overcome the problem of failure to treat so prevalent in the pilot. This did work;

out of 304 students who took part in the research, only 5 were removed from the

analysis for failing to spend sufficient time on the manipulation, answering with a

response set, or because they chose to withdraw their data upon learning about the

study’s full purpose and hypotheses in the debriefing. Thus, this study resulted in

299 usable responses. Moreover, even though there were no forced responses for any

questions, participants nearly universally provided answers to every question, indi-

cating that they took the study quite seriously. Therefore, the N is 299 for the figures

and models presented below. The convenience sample of students was largely repre-

sentative of the country at large in terms of ideological makeup (though slightly more

left leaning), but was comprised of approximately two-thirds female participants.

In addition to building upon the pilot by conducting a lab-in-the-field experiment,

this study also included various differences in design meant to add to and improve

upon the pilot. Most notably, the study added an additional treatment condition in

15I would like to thank the Behavior and Decision Making Research Group at the Ohio StateUniversity for their generous support of this study.

98

order to attend to the dynamics of ideology by showing how those sympathizing with

the left and the right react to appeals from the left and the right, respectively. Thus,

there was one treatment condition in which the opposition left attacked the right

(LvR), and another treatment condition in which the right attacked the left (RvL).

This was also beneficial in that it better capitalizes off of the dominant cleavage

in the country — that between the left and the right. While the attack from the

far-right Jobbik party against the right leaning government in the pilot was chosen

because it best emulated the most outrageous of any rhetoric in the political arena,

in retrospect, this likely did not resonate as well with a majority of participants, who

likely discounted the message as being radical because of its source. Given the power

of the political control in the pilot to elicit negative emotions, a political control was

also included in this study alongside a non-political control. The political control in

Study 2, however, was carefully designed not to include accusations, but instead just

to bring up the various party labels and politician names to make politics salient to

participants in this condition.

Another difference from the pilot involved the form of the stimuli in the manipula-

tions. Whereas the pilot included an article mimicking real coverage in a newspaper,

this study proceeded by presenting participants with one statement at a time. Partic-

ipants were told they were going to read individual statements from an article titled

“The 5 sins of ..... .” Within each condition, statements were presented to partici-

pants in random order and they were then asked to rate how credible each statement

was and the strongest emotion they felt towards the statement. This was to ad-

dress two problems from the pilot. First, it addresses the failure to treat indicated

by the large proportion of pilot participants who spent very little time reading the

manipulation and thus had to be dropped from the analysis. By presenting the ma-

nipulation in bits and asking participants to rate their dominant emotional reaction

99

and the credibility of each individual part, participants would be kept more engaged.

Second, the measures after each statement provided valuable insight into both the

overall emotional reaction participants had towards each manipulation as a whole,

and detailed information on each individual component.

In order to provide the greatest amount of experimental control possible, each

condition, even the controls, took the form of an argument against something. The

control condition was “The 5 sins of the Paleo diet,” the political control was “The

5 sins of Hungarian agricultural policy,” the left attacking the right (LvR) treatment

was “The 5 sins of the Orban government,” and the treatment in which the right

attacked the left (RvL) was “The 5 sins of the Gyurcsany coalition.” The statements

used in the populist outrage treatments were taken from actual political attacks dur-

ing the campaign and reviewed by many Hungarians, including political scientists,

psychologists, and laypeople. Indeed, some of the statements attacking the left and

the overall structure of the stimulus mimicked a real piece of negative campaigning —

the “Seven Sins” pamphlet described in Chapter 5. Statements were altered slightly

to mirror one another as closely as possible, swapping out names and particular is-

sues or scandals in the same general argument structure. The complete text of the

back-translated manipulations are in the appendix.

4.5.1 Emotional Reactions

As was the case with the pilot, the first step in the analysis here is to attend

to the emotional effects of the various conditions. Figure 4.2 displays emotional

reactions following the conditions. As one purpose of this study is to take into account

how those on the left and right feel about attacks from both their side and the

other side, these figures show the emotional reactions as a function of self-reported

100

ideology measured before the manipulation.16 Emotional reactions were measured

in two ways. First, just as in the pilot, participants were asked to focus on their

emotions, not thoughts, and rate how strongly they felt specific emotions on a Likert

scale ranging from “Not at all” to “Extremely.” In order to correct for measurement

error, participants also completed another battery in which they rated how strongly

they agreed or disagreed with a list of statements targeting these same emotions but

with metaphors such as “makes me see red” for anger and “leaves a bad taste in my

mouth” for disgust. These were then combined with the explicit measures to create

indexes.17

As the figures show, the treatments were generally efficacious in eliciting negative

emotions, providing support for H2 by showing that populist appeals to outrage elicit

high levels of anger and disgust. As Figures 4.2(a) and 4.2(b) show, outrage and anger

were a function of the attack style politics of the treatment conditions. In neither

the single emotion measures nor the composite anger scale was the political control

statistically distinguishable from the non-political control, while with the exception

of among political centrists, the treatments are much higher. Figure 4.2(c) shows that

disgust was even more powerfully elicited by the treatments than anger. Interestingly,

this also holds for centrists. Thus, while those in the political center were no more

likely to report being fired up and angry, they were more disgusted. Given that

those in the center are typically the least politically aware (Zaller, 1992), it seems

that the attacks turn off those who are already the least engaged.18 This provides

16Participants were asked to place themselves on an 11-point left-right scale. Those in the middlepoint (24% of participants) were coded as center, leaving 34% of participants in the sample identifyingon the left and 42% of participants self-identifying on the right.

17Index construction was verified using factor analysis and each of the scales reported here havean alpha of .75 or above.

18Indeed, in unreported regression estimations including knowledge as a predictor for the disgustresponse, low political knowledge is a strong predictor of feeling disgusted at the attacks in thetreatment.

101

left center right

0.4

0.6

0.8

1.0

Ctrl PoliCtrl LvR RvL Ctrl PoliCtrl LvR RvL Ctrl PoliCtrl LvR RvLCondition

Sel

f−re

port

ed o

utra

ge

Outrage by condition and ideology

(a) Outrage

left center right

0.2

0.4

0.6

Ctrl PoliCtrl LvR RvL Ctrl PoliCtrl LvR RvL Ctrl PoliCtrl LvR RvLCondition

Ang

er S

cale

Anger by condition and ideology

(b) Anger Scale

left center right

0.2

0.4

0.6

Ctrl PoliCtrl LvR RvL Ctrl PoliCtrl LvR RvL Ctrl PoliCtrl LvR RvLCondition

Dis

gust

Sca

le

Disgust by condition and ideology

(c) Disgust Scale

left center right

0.2

0.4

0.6

Ctrl PoliCtrl LvR RvL Ctrl PoliCtrl LvR RvL Ctrl PoliCtrl LvR RvLCondition

Anx

iety

Sca

le

Anxiety by condition and ideology

(d) Anxiety Scale

Figure 4.2: Study 2 emotional reactions

102

initial support for H2a: those who are unaligned and less inclined to follow politics

become more disgusted than angry when they are faced with a political attack. As

disgust is associated with a strong inclination to distance oneself from the source

of disgust, this reaction to political attacks may prompt a vicious cycle in which

those already least inclined to participate in politics become even more turned off.

However, that those on either side of the ideological divide are also highly disgusted

by the populist outrage appeals is an unexpected finding. Finally, participants in the

outrage treatments were no more likely to report being afraid or anxious than they

were in the political control. As opposed to both anger and disgust, fear and anxiety

were not a function of the treatments, but instead of the mere mention of politics.

This replicates the same pattern demonstrated in the pilot data.

4.6 The Role of Partisanship

The evidence to this point shows that high levels of anger and disgust exist among

the Hungarian public, both when it comes to politics in general and to the omnipresent

accusations of moral transgressions in particular. The next question is what deter-

mines whether an attack elicits moral anger or disgust. As argued above, the presence

and strength of partisan and ideological dispositions is one key variable hypothesized

to moderate this relationship. Those who see politics through the lens of partisan

or ideological predilections are more likely to make the attributions of blame and

intentionality key to the anger response. The figures displayed above already provide

some support for this proposition, showing that participants on the left and right

were more likely to report outrage and anger as a result of the treatment conditions,

while those in the middle were not. Meanwhile, participants in the center responded

to the attacks in the treatment conditions with high levels of disgust.

103

Table 4.3: Strength of Ideology and Emotional Reactions to Outrages (Experiment)

Anger Scale Disgust Scale Anxiety Scale

Intercept .28(.03)*** .21(.03)*** .36(.03)Treatment .12(.04)*** .30(.04)*** .09(.05)*Extreme .02(.07) .07(.06) .09(.07)

Treatment*Extreme .24(.10)** .04(.09) .08(.11)Adj. R-sq. .17 .36 .06

N 299 299 299

Table 4.3 provides more evidence that supports these results, presenting models

predicting anger and disgust as a function of experimental condition and ideologi-

cal strength. The Treatment variable is a dummy for inclusion in either treatment

condition (LvR or RvL), while the variable Extreme captures the distance of the

participant from the ideological center by folding their placement on the self-reported

ideological placement scale such that lower values indicate closeness to the ideolog-

ical center and higher values indicate distance towards either the left or the right.

As the dependent variables are the 0–1 scales constructed from multiple indicators,

ordinary least squares (OLS) models are estimated. The results clearly show that

the treatment elicited these negative emotions. However, the key point is that an

interaction of the distance from the center of the ideological spectrum and receiving

the experimental treatment predicts anger but not disgust. Thus, those on either po-

litical side become angry in response to populist outrage relative to those without a

political team, who are left disgusted. These results are robust to including a measure

of ideological placement rather than extremity, as well as to using separate dummy

variables for the two treatments.

104

Finally, we return to the survey data introduced at the beginning of this chapter

for further evidence that partisanship plays a key role in moderating the relation-

ship between accusations of moral transgressions and emotional response. As in the

lab-in-the-field experiment, a variable of ideological extremity/centrism was com-

puted by folding self-placement on the left-right ideological scale so that 0 meant

self-placement precisely in the center and higher values (up to 5), represents greater

ideological extremism on either the left or right. In every light, from a simple Spear-

man correlation19 to multivariate models, being in the center ideologically predicts

greater disgust while those on one or the other side are more likely to exhibit higher

levels of overall anger towards political wrongdoings. Because of the nature of the

emotion variables in the survey, which included many zeros and typically low means,

Table 4.4 presents the results of negative binomial models, but OLS and Poisson

estimations produce identical substantive results.20 These models also include an ab-

breviated need to evaluate scale as this psychological variable has been shown to be

important in predicting a proclivity towards anger (Jarvis and Petty, 1996). The re-

sults show that the total number of times participants reported being angry increases

as a function of ideological extremism. That is, as was the case in the experiment

sample, those farther either to the left or the right were more likely to report anger

as opposed to any other emotion. Meanwhile, the disgust model shows that ideologi-

cal extremism is negatively and significantly associated with choosing disgust as the

dominant emotional response to the political situations presented. This provides sup-

port for the theory that the perceptual screen provided by favoring one side over the

19Spearman correlations between Anger and Extreme are .125(p=.000) when respondents who didnot place themselves on the ideological scale were dropped. The correlation is .185(p=.000) whenthey are included as zeros. Spearman correlations between Disgust and Extreme are .008(p=.803)when non-responses were dropped, and -.067(p=.01) when they are recoded as zeros.

20These models drop the 17 percent of participants who did not place themselves on the ideologicalscale. As with the Spearman correlations, the results are even stronger when these nonresponses arerecoded as zeros.

105

Table 4.4: Strength of Ideology and Emotional Reactions to Outrages (Survey)

Anger Disgust

Extreme .06(.02)** -.08(.03)***Need to Evaluate .09(.03)*** -.02(.03)

Intercept .18 (.14) .5(.180)***AIC 3244 2636N 838 838

other contributes to anger, while its absence means citizens do not make the blame

attributions central to anger and instead become disgusted with politics.

4.7 Discussion

The three sources of data here show that the rise in populist outrage documented

in the previous chapter is responsible for a related rise in anger and disgust towards

politics. The survey showed high levels of anger and disgust towards politics in gen-

eral in Hungary, as well as towards the populist outrage that has become prominent

in everyday political discourse in the country. The two experimental studies demon-

strate that members of the public do react to these attacks with anger and disgust.

Finally, evidence from the survey and the lab-in-the-field study point to partisan

predilections as a key variable moderating the relationship between political outrage

and the emotional response, such that partisans are more likely to be angered while

those in the middle are left disgusted. While these data further our understanding of

the public response to populist outrage, several caveats are worth noting.

First, the studies each present a problem of participant attention, but in opposite

directions. Regarding the participants in the pilot, recruited from a panel that take

online surveys for money, all evidence points to a problem with participants paying

106

very little attention. The attrition rate was very high, many participants did not

spend sufficient time on the manipulation to be included in the analysis, and the

overall time to finish the survey was very short. The lab-in-the-field study presents

just the opposite problem: these participants seemed to take the study quite seriously

and to devote considerable time and thought to their questions. These participants

were university undergraduates taking the study in a lab setting at their university,

and they were greeted and briefed at the beginning of the study by the author — an

American academic — in Hungarian. This seems to have had the effect of making

these participants take the study very seriously, which may help to explain why

emotions had a somewhat weak effect on many of the dependent variables embedded

in the experiment following the manipulations. These participants may have been

especially likely to display signs of social desirability to give a better view of their

country to a foreigner, or at least might have been trying to impress an outsider.

Future studies will need to be constructed with extreme caution in balancing the

need for participants to engage with political stimuli they may often find disgusting

one the one hand, and the need to have participants act naturally and let their

emotions affect their political attitudes as they would in the real world without social

desirability on the other.

The co-occurrence of anger and disgust among many participants in the studies

calls for further research. The hypothesis that ideological predilections determine

anger and disgust was only partially borne out: partisanship does predict anger, but

this finding was driven by a lack of anger among those in the political center, not a

lack of disgust among those on the right and left. Populist outrage elicited disgust

across the board, among those in the middle and those on the ideological extremes.

While a shortcoming of these studies was a lack of measurement of negative emotions

towards groups within society, this is a likely explanation for those on the left and

107

right reporting high levels of disgust. This points to the possibility that citizens are

disgusted by their political opponents, a problematic scenario given that disgust is

a key driver of dehumanization. The language of disgust is nearly always employed

in the process of dehumanization (Miller, 1998), and where disgust accompanies out-

group prejudice, neurological studies show that the regions of the brain associated

with interacting with other people fail to activate (Harris and Fiske, 2006). Worse

still, given the widespread practice among the Hungarian right of targeting their

outrageous accusations at immigrants and the Roma minority, these vulnerable pop-

ulations may be the target of moral disgust. This merits further investigation and

would go a long way in explaining the Hungarian government’s brutal treatment of

refugees attempting to enter the EU from Syria and other conflict zones in 2015, a

policy that increased support for the Orban government.

The role of partisanship is also a serious shortcoming of these analyses. The data

in these studies cannot parse out the extent to which anger drives partisanship, or

partisanship drives anger. While these data offer a snap-shot, over time dynamics

in the acquisition of partisan and ideological predispositions are at play. Future

research to trace the relationship between emotional response and the evolution of

partisanship is necessary in order to increase our understanding of political behavior

in young democracies. It is notable, however, that most of these young democracies

are characterized by very low levels of partisanship. Indeed, even among established

democracies, reliable party identification is on the decline (Dalton and Wattenberg,

2000). Given the large and growing segments of the population that are unaligned,

political disgust is an increasingly important emotion to measure and confront.

108

Chapter 5: Outrage Exhaustion: How Politics Become

Disgusting

“The only thing I’m unsure about concerning the election is whether I’ll write [very

bad word for male genitalia] or [even worse word for female genitalia] on the ballot”

— Interview with male teacher in his 40s, Budapest, January 2014

5.1 Introduction

The interviewee laughed after his defiant statement, but it was a nervous, some-

what forced laugh. His tone changed to remorse as he continued to answer the ques-

tion of whether he would vote in the 2014 Hungarian parliamentary election, assuring

me that he took his civic duties very seriously in the new democratic regime he had

protested to bring about. He and his entire social circle had made politics a central

part of their lives, spending their free time in heated debate and volunteering exten-

sively for political campaigns. However, things had changed for him. “The whole

lot of them [politicians] is absolutely terrible. I’m just so fed up with it I ignore the

entire mess.” He said all of his former activist friends were in the same boat — “we

just can’t stomach it.” They have given up on politics altogether. Far from partic-

ipation, the interviewee admitted that the 2014 election was the first he had pretty

109

much ignored completely. “I can’t even bring myself to hear about politics. Not even

anything as light as Index 1.”

Similar stories of previous engagement and current disaffection were widespread

throughout the public by the time of the 2014 Hungarian election, which would have

the lowest turnout in any Hungarian election of the post-communist era. Having

analyzed the immediate psychological reactions to populist outrage dialogue in the

previous chapter, the analysis here is concerned with the long-term effects. This

chapter tests the hypothesis that constant exposure to populist outrage over time

causes a transition from anger into disgust that explains disgust towards politics and

the rejection of politics evidenced by this interviewee (H3 ). This hypothesis is based

upon psychology research showing that over time unresolved anger gives way to the

emotions of disgust and contempt. In other words, the initial moral anger felt towards

populist outrage that drives action, or “outrage motivation,” eventually turns into

“outrage exhaustion,” disgust and disengagement towards politics, when this anger

is repeatedly unresolved.

5.2 From Anger to Disgust

As the prior chapter demonstrated, attributes of individuals play an important

role in moderating the relationship between accusations of populist outrage and the

nature of the emotional reaction. While partisanship is one central determinant of

the way a person will react to populist outrage, there are other attributes of both

the individual and the instance of populist outrage that will affect the emotional

response. Because populist outrage is, as Chapter Three demonstrates, a persistent

and increasing feature of Hungarian politics over a period of years, time is another

1Index is a news site that features current events with a very entertainment-based twist. USAToday is the closest analogous publication in the United States.

110

key factor that may moderate the emotional response. Indeed, work in neurology

and psychology suggests that emotional reactions to specific stimuli evolve over time

in general, and provide evidence that with time and repeated exposure, much of the

anger towards populist outrage likely evolves into disgust.

Fischer and Roseman (2007) provide a major pillar of this third primary hypothesis

of the dissertation. In their study of the relationship between anger and contempt,

they find strong evidence that repeatedly feeling angry towards a person or object is

the key source of contempt, a subtype of disgust.2 Their work focuses on interpersonal

interactions, and they find that as another person continues to act in a way that

provokes anger, the reaction of anger and the associated approach to confront changes

into contempt and a desire to distance and derogate the other. This is particularly

true after people have acted upon their anger to confront the other but the offender’s

behavior continues unabated and unchanged. As these scholars and others note, this

sort of evolution of emotional response is adaptive; emotions give us feedback as to

whether to continue toward a goal, abandon it, or take another approach towards

it. As Giner-Sorolla (2012) notes, “Anger may eventually burn out faced with strong

and persistent frustration, which itself is adaptive; people need to see when a task is

impossible and let go” (94).

In addition to the adaptive aspect of the evolution of anger into another emotion,

the way that disgust comes about by repeated association also points to an increased

likelihood of disgust given greater exposure and experience with anything provoking

negative affect. Disgust is a highly adaptive response, meant to cause us to distance

2Scholars in the CAD (Contempt-Anger-Disgust) tradition argue that contempt, anger, anddisgust are three separate emotions each governing a separate moral ethic: contempt deals withviolations of the community ethic, anger with violations of the ethic of autonomy, and disgustwith threats to the divinity ethic (Rozin et al., 1999). However, in their empirical demonstration,study participants do not identify contempt as a distinct emotion. Moreover, both the factors thatelicit contempt and contempt’s downstream cognitive and behavioral ramifications are equivalent todisgust for all intents and purposes relevant to this work. Thus, both work on contempt and disgustinform the dissertation’s third hypothesis.

111

ourselves from anything that could harm us. Additionally, it is a completely learned

response dependent upon the environment. Nothing is innately disgusting from the

start; infants are not born finding any particular thing disgusting, but instead learn

to be disgusted by specific objects through experience and observation of how others

react (Kelly, 2011). The adaptive function of keeping us alive where the stakes are

high — one handful of the wrong berries could mean an exit from the gene pool —

means that the disgust system is strongly biased towards false positives. It is relatively

easy for something to become disgusting. Indeed, the act of cigarette smoking has

come to elicit disgust among many in the United States (Rozin and Singh, 1999).

Moreover, disgust has become a critical regulator of moral behavior as well (Haidt,

2003; Kelly, 2011), and behavior viewed as immoral can elicit disgust (Rozin et al.,

2000; Chapman et al., 2009). Finally, when the same stable cause is at the root

of negative affect, appraisal theories find that disgust is more often likely to be the

primary reaction (Chapman and Anderson, 2011).

For these reasons, a steady stream of alleged outrageous moral violations will likely

result in disgust. Constant exposure to outrageous appeals transforms moral anger at

specific actors into disgust at all political parties and the very practice of competitive

politics. To understand how this plays out in the world of politics, imagine a citizen

experiencing negative affect due to some political scandal or accusation. At first,

this fresh outrageous accusation may indeed inspire moral anger in this citizen if

she makes appraisal of intentional, blame-worthy wrongdoing critical to this emotion

(Petersen, 2010; Giner-Sorolla, 2012). “How dare they?” she might think. This

moral anger is associated with a desire to right the wrong (Lerner and Tiedens, 2006;

Carver and Harmon-Jones, 2009), which she can accomplish by voting against the

actor responsible for the trespass (Marcus et al., 2000; Brader, 2006). Now, however,

imagine that once that bad apple of a politician has been replaced, his successor is

112

accused of the same or similar outrageously immoral acts. As this process repeats

itself, not only are the parties discredited one by one as Pop-Eleches (2010) argues in

his explanation of the widespread rise of unorthodox parties in the region, but this

concerned citizen’s negative experiences with politics over and over in the form of

unresolved anger at alleged immorality forms a very likely basis for the associations

key to disgust. This over time transition of emotional response is in line with the

findings of Dassonneville et al. (2015), who take up the question of when political

disaffection causes vote switching and when it causes abstention. They find that

initial disaffection leads to vote switching, but that with time, if this vote switching

fails to result in the desired political change, abstention is the result.

When considering how disgust may supplant anger given repeated negative affect

at the perceived immorality of politics dominated by populist outrage, two more spe-

cific hypotheses may be derived. First, when a specific instance or target of populist

outrage occurs over and over again in a citizens’ political information environment,

this repeated exposure to the same unpleasant stimulus will make this object a likely

cause of disgust for the citizen. Here I will refer to this as the Repetition Hy-

pothesis. While this hypothesis concerns attributes of the object of the accusation,

attributes of the message receiver will also matter. Namely, the degree and nature

of a citizen’s exposure to populist outrage will influence the probability of an angry

versus disgusted reaction. Less exposure to populist outrage will, all things equal, be

more likely to result in anger, while more exposure will lead to disgust. Getting at the

amount and intensity of exposure to populist outrage is a difficult task both theoreti-

cally and empirically, but age serves as a useful, if rough, proxy. As a citizen ages she

gains more experience with politics simply by nature of spending more time living in

the world. I will refer to this as the Age Hypothesis. There are two potential ways

we might observe age as a proxy for a person transitioning from a dominantly angry

113

to disgusted view of politics and populist outrage. First, it may be that over time

and greater exposure to populist outrage there is a straightforward cumulative effect.

In other words, with more time spent observing politics, a citizen may become more

and more disgusted. Second, the nature of the political dialogue at the time of one’s

political coming of age may be influential. Scholars have shown young adulthood to

be the most profound time of political development (Campbell et al., 1960; Plutzer,

2002), so it may be the case that those who were in young adulthood during a time

of “peak populist outrage” will be more likely to feel disgusted than angry. In this

view, it is likely the people who were in their teens and early twenties in the early

aughts, and particularly during the “lying Prime Minister” scandals and protests of

2006 (see below), who would be transitioning to disgust now.

These hypotheses pose viable routes from initial anger felt towards immoral, out-

rageous political antics to eventual disgust as anger-fueled action fails to rectify the

situation. Before moving on to testing these hypotheses it is worth noting that they

are marginal hypotheses; that is, they each isolate an aspect of an individual or polit-

ical object that will influence the emotional response. There are many other aspects

of citizens that will determine how citizens perceive and react to the political infor-

mation they come across. Indeed, as an enormous body of scholarship as well as

the previous chapter demonstrate, the strength and presence of partisan predilections

exerts a powerful influence upon political perceptions in general and in the anger

and disgust response in particular. However, through understanding these time and

exposure related variables we can gain further leverage on the multifaceted process of

emotional response to political information and downstream political ramifications.

114

5.3 Data

The previous chapter, like most research on emotions in politics and in political

behavior generally, looks at the response to populist outrage at one point in time.

While this offers valuable insight and the unquestionable advantage of data availabil-

ity, political interactions do not take place at one and only one time, but unravel

over long periods of repeated interactions. Focusing on a single snap-shot of emo-

tional response to an individual political stimulus offers many insights, but as Brader

(2005) notes in a critique of his own research on how campaigns influence voters’

emotions, “elections are not won or lost on the basis of an effective campaign ad, but

may be won or lost with an effective ad campaign” (402). Much of the research in

this dissertation suffers from the same focus on measurement of static states, but in

this chapter the focus is on examining the dynamic process of emotional response to

populist outrage over long periods of time as people learn about and interact with

the political world. While this over time emphasis may yield enormous insight, the

data challenges are equally large. Indeed, though psychologists posit a transition of

anger into disgust as a function of repetition and time (Fischer and Roseman, 2007;

Giner-Sorolla, 2012), there is little empirical testing of this process, likely because of

these empirical challenges. Thus, through the following analysis I work to use the

unique domain of politics as an opportunity to observe this transition..

In addition to the dynamic nature of these processes, the long period of time

driving the hypothesized transition of anger into disgust presents further challenges.

The ideal data to test the general hypothesis that over time anger towards specific

instances of populist outrage transitions into general disgust towards politics and

causes outrage exhaustion would be panel data measuring how the same people feel

towards politics in general and specific political events over decades. No such data

115

exist. Instead, here I resort to a multi-methods approach, combining quantitative

and qualitative data in an effort to triangulate the dynamics of emotional responses

to populist outrage over time.

First, the quantitative data described in the previous chapter provide important

insight. Namely, the original survey is unique in that it includes questions asking

participants to rate their dominant emotional reactions to politics in general as well

as a number of instances of populist outrage pertinent during the 2014 elections.

While all of these were widely familiar in this context, a number of them were new

accusations, while a few were years old. This allows for a test of the repetition

hypothesis by analyzing whether older scandals are more evocative of disgust and

newer scandals of anger. It also provides data on the age hypothesis as this survey has

unique data on general emotions towards politics alongside demographic indicators

such as age.

The most useful data source for testing the hypotheses related to the transition of

anger inspired by populist outrage into disgust over time is a series of four focus groups

held in Budapest on the eve of the 2014 parliamentary elections.3 These focus groups

are a unique opportunity to analyze political sentiments in a common way people

form political opinions: in groups that replicate and react to public discourse. This is

critical to provide external validity, supplementing experiments and interviews with

data from a context in which participants interact with, influence, and are influenced

by those around them as they are in the real world (Morgan, 1996). Focus groups are

uniquely able to gauge how people form opinions and react to stimuli (Glynn et al.,

1999), making them particularly useful for analyzing how political views evolve over

time.

3I would like to thank the Alumni Grant for Graduate Research and Scholarship at the OhioState University for their generous support of this portion of the research.

116

The four groups were divided by political leaning both to get variation in the way

those on the left and right view politics the same political actors and events differ-

ently, as well as because focus groups typically work best with groups of participants

who by and large agree with one another concerning the subject matter. Strong dis-

agreements can cause conflicts that lead many within the group to shut down. Thus,

two groups were conducted with right-leaning (pro-Fidesz government) participants,

and two other groups were conducted with left-leaning (anti-government / opposition)

participants. We asked potential participants their favored news sources in order to

determine which group to place them in as the major news sources have well-known

partisan leanings. In addition to sorting by political leaning, groups were divided

into older and younger cohorts. This 2 (right versus left) by 2 (young versus old)

study design was meant to make groups relatively homogeneous regarding political

experiences in order to facilitate comfortable conversations that would elicit the most

honest and frank input from participants and allow them to talk about how their

reactions to politics and specific political events evolved over time with like-minded

peers. Moreover, taking age into account was a strategy to gather data to test the

age and socialization hypotheses. One a benefit of the focus group is that the overall

interactions will tend to produce a general consensus-based discussion that drowns

out individual idiosyncrasies and can be compared across groups.

The focus groups were moderated by Gabor Strausz, Ph.D., director of the Bu-

dapest market research firm Straketing and a professor of marketing at Cornivus

University Budapest. Participants were recruited by snowball sampling with the goal

of yielding variation in terms of two attributes: political leaning and age. Partici-

pants were told they would take part in a conversation about how they feel about

Hungarian politics lasting one hour and fifteen minutes. They were compensated

with refreshments during the focus groups and then 3,000 forints (roughly $15) at the

117

end of the groups. The groups were held in the week before the election (late March,

early April 2014) in the psychology building of Eotvos Lorand University in central

Budapest.4

Table 5.1: Focus Group Composition

Group 1 Group 2Younger, Right-Leaning Older, Right-LeaningAge Range: mid 20s-early 40s Age Range: late 40s-early 70sFour Male, Three Female Four Male, Four FemaleGroup 3 Group 4Younger, Left-Leaning Older, Left-LeaningAge Range: early 20s-early 30s Age Range: early 40s-early 70sFour Male, Four Female Four Male, Four Female

Consultation with Dr. Strausz as well as several other researchers versed in fo-

cus group methodology suggested that eight participants is an ideal group size, and

with the exception of one group (younger right-leaning) in which seven participants

were present, all groups ended up with eight participants. As Table 1 shows, the

recruiting was also largely successful at obtaining a balance between male and female

participants within each group. Moreover, using favorite news source as a means

of sorting the groups ideologically was successful. During the groups it was readily

apparent that, though they may have all had many negative emotions to go around

when it came to Hungarian politics, the groups were largely ideologically homoge-

neous. Recruiting was more successful for younger left leaning respondents, and this

group consisted of several university aged participants and completely of people in

4I would like to thank the Political Ideology Lab at Eotvos Lorand University for their extensivehelp in arranging these groups. PhD candidate Nora Lantos in particular provided invaluable counciland logistical support and was also present to assist during every group. This research would nothave succeeded without her.

118

their twenties with one exception. The right-leaning young group tended towards late

20s and, mainly, 30s in age.5 Thus, recruiting based upon which media outlets they

watched was largely successful in sorting the groups ideologically but seemed to have

biased the younger groups to be very young in the case of the left leaning group and

somewhat young to middle aged in the right leaning group.

The moderator followed the same protocol as closely as possible in each group. At

the beginning he introduced me and himself, provided an overview of the purpose of

the research, and then obtained verbal consent for participation and recording while

assuring group members of the complete confidentiality of their identities and re-

sponses. Dr. Strausz then asked each participant to briefly introduce themselves and

say a bit about their level of political interest. From there, the moderator prompted

participants to describe their strongest overall feelings towards politics in general be-

fore moving on to ask their perceptions of Hungarian democracy, both in general and

then specifically in comparison to neighboring countries. After that we distributed a

brief worksheet and gave participants time to write short answers to several prompts

before going around and sharing their answers when interested. The prompts included

the words they would use to describe the parties they most favor and most dislike,

the most serious scandals facing the left and right respectively, and then the things

that have most outraged and disgusted them over the years. Finally, the moderator

read each group a campaign attack against both the right and the left taken from real

campaign materials on both sides. At each stage, the moderator urged participants

to focus on their feelings and how they have changed over time.

Finally, in addition to the focus groups, a series of informal, semi-structured inter-

views supplement the other sources of data. These interviews largely mimicked the

focus group in form. They began with a general discussion of how participants feel

5These are approximations as participants were not inquired about personal attributes like age,income, education, etc.

119

about their political actors and system and why they feel this way. Following that, I

asked detailed questions about specific past and current political appeals and events,

encouraging interviewees to think back upon how they felt about these occurrences at

first, how these feelings evolved over time, and how they contribute to their current

political attitudes and behavioral intentions.

In sum, the combination of the focus groups, survey data, and interviews yields

excellent source material to analyze the evolving way participants feel about politics.

The focus groups in particular provide a unique social context for dynamic conver-

sations about Hungarian politics on the eve of a very high-stakes election, and the

analysis below relies heavily upon these data.

5.4 Repetition Hypothesis

As outlined above, one way in which disgust may evolve towards political objects

is if those same objects are presented and paired with the experience of negative

affect again and again over time. This repetition hypothesis means that when the

same scandal, politician, or party comes up in populist outrage over a long period,

the likely reaction evolves from anger to disgust. This is especially true when it comes

to repeated association with immorality (Rozin et al., 2000; Chapman and Anderson,

2011).

In general, when participants were asked to relate how they felt about a political

scandal at first and then over time, there was a trend of using more and more dis-

gust language. Given the roots of the disgust response in the gustatory complex that

governs eating (Kelly, 2011), the language of nausea is a key marker of disgust. One

interviewee encapsulated a general trend across the focus groups and interviews in

discussing Fidesz’s allegedly corrupt education reforms: “At first I couldn’t believe

even they could do something so wrong to students. The more I think about it,

120

the more it just makes me sick” (Interview with mid-thirties female office manager,

Szazhalombatra, March 2014). Another participant during the first focus group re-

acted the same way when asked how she sees Hungarian democracy. She said that

under the Socialist government from 2002 to 2010, the country had been absolutely

undemocratic, citing police crackdowns of protests after the big 2006 scandal (see be-

low) as an example. She exclaimed “When I saw police working for that government,

beating old men and women, I couldn’t believe my eyes. It makes me sick thinking

about the whole thing even now!” (Focus group participant 1.2). The language of ex-

crement, another lexical hallmark of disgust, was also commonly employed discussing

this as well as other older scandals, but I will spare the reader specific quotations.

That 2006 “lying prime minister” scandal referenced above constitutes the best

case study for the development of disgust over time and repeated exposure. In this

grandfather of all Hungarian political scandals in Hungary, Prime Minister Ferenc

Gyurcsany was caught admitting that his Hungarian Socialist Party (MSZP) had

“lied morning, noon, and night” about the state of the Hungarian economy in or-

der to gain reelection. The specific lies pertained to the need to go through more

austerity measures, which MSZP argued were vital to the long-term improvement

of the Hungarian economy despite their unpopularity among the people and their

Fidesz rivals who would, in Gyurcsany’s estimation, abandon reforms. The tape was

released shortly after MSZP’s razor thin victory over Fidesz, and massive protests

erupted all over the country. Despite calls from Orban’s Fidesz opposition and be-

yond, Gyurcsany and his MSZP government refused to resign, and worked to suppress

the protests with hard-line police action referenced by participant 1.2 above. Even

eight years later, the lying prime minister scandal was still a major aspect of the

political dialogue of the 2014 election, both in the campaign and in everyday political

discussions. Indeed, it came up without prompt or solicitation in all four focus groups

121

and almost every interview. Across groups and interviewees it reliably produced signs

of disgust, from the “gape face” characteristic of disgust (Ekman and Friesen, 1971;

Kelly, 2011) to the language of illness, to reports of this incident making people just

want to give up on politics altogether.

The data from the survey back up the numerous illustrations of disgust from the

interviews and focus groups. Recall from the description in the previous chapter that

at the beginning of this survey module, administered in 2014 to a random sample

of 1,000 Hungarian participants, each participant was asked to rate their dominant

response to Hungarian ‘political life in general’ and then to a short series of infamous

scandals. Participants chose only one emotion per prompt by selecting from a drop-

down menu of “happy,” “proud,” “angry,” “disgusted,” “anxious,” “apathetic,” or

“don’t know / not familiar with this.” Of all of the prompts, the lying prime minister

scandal was the oldest, and it elicited the highest number of disgusted responses, with

over 25% rating this as their top emotion. This is the strongest example of an old

political object repeated in the political information environment ad nauseam until

it increasingly elicits disgust, both in the qualitative and quantitative data.

This widespread disgust felt towards MSZP seems to have inflicted a near mortal

wound upon the party. Their government limped along to fulfill its entire second

term, but it suffered a devastating defeat in 2010. Indeed, participants mentioned

again and again that they could not bring themselves to support MSZP after that, a

trend unanimous among right-leaning participants but also appearing in some among

the left-leaning. Disgust at the lying prime minister scandal and its aftermath appears

to have imparted a permanent uncleanliness upon all of the leaders who were involved

as well as the MSZP party organization itself. That the name MSZP was so tainted

may also help explain the so-far permanent fragmentation of the left in the aftermath

of the scandal. Multiple spin off parties broke away from MSZP and tried to label

122

themselves as something new on the left, though no dominant force has emerged to

replace the socialist successor party that helped to anchor Hungary’s party system for

two decades. This instance seems to be a prime example of the systematic discrediting

of mainstream parties that Pop-Eleches (2010) points to as a root cause for the rise

of unorthodox parties across post-communist Europe.

Two scandals that broke in the run-up to the 2014 elections provide useful foils

to the lying prime minister scandal that had broken eight years prior. The first is the

“Trafik scandal,” refering to the small “trafiks,” or tiny stands that sold cigarettes,

newspapers, and other small conveniences that were ubiquitous in Hungary prior to

new legislation by the Fidesz government. Under the mantle of increasing public

health, the government regulated tobacco sales only to a limited number of licensed

stores, driving these trafiks out of business. The scandal was that it seemed that

either being a long time Fidesz loyalist or someone willing to do the party favors was

required to get a license. Almost everyone I spoke with in Hungary in 2014 had a

story of a kindly little old lady or man who had spent their entire lives working their

trafik on the corner only to be put out on the street as a Fidesz loyalist received a

golden ticket to sell the only cigarettes around. This is a case of populist outrage

par excellence: stealing the livelihood from the country’s numerous trafik owners for

political gain. Throughout the focus groups it was referred to as a major source of

anger. It led to a prolonged angry exchange in the first group, with participant 1.2, an

obvious Fidesz partisan, arguing that no matter how the licenses were distributed it

is a good thing to make smoking harder as many others in the group angrily objected

and said it made them rethink their Fidesz support. The survey results indicate that

this new scandal did elicit predominately anger; 30% of participants rated anger as

their dominant reaction to the trafik scandal (trafik botrany), while only 21% listed

disgust.

123

The other major scandal of the 2014 campaign concerned the left. The so called

Simon deal (Simon ugy) involved the discovery that long-time leftist parliamentarian

Gabor Simon had bank accounts filled with millions of Euros that he could never

have earned with his parliamentary earnings. The accusation was that the MP had

received these funds for sweetheart deals to big business. This scandal broke just

months before the April election and was perhaps the most talked about issue of the

campaign. Again, the accusation at the heart of this scandal is a paradigmatic case

of populist outrage: Simon leveraged his political position to make himself rich with

no regard for the well-being of the people he had taken an oath of office to serve.

Again, this came up in all four focus groups and inspired intense, heated rants about

the lies and corruption of Simon and MSZP. In the survey it yielded the highest level

of anger of any prompt: 32% of participants rated anger as their top emotion to this

scandal, while 20% rated disgust.

While instances of populist outrage may come up again and again even after people

have worked to take action to correct them, another reason participants seemed less

angry and more filled with loathing for the whole process is that they feel that these

accusations are never investigated fully enough for the people to come to any resolu-

tion about who did what. Indeed, there is a pervasive sense that scandal after scandal

breaks but is never resolved and that the facts are buried in endless name-calling and

accusations. Where there are official inquiries or court cases, the proceedings are are

obtuse and riddled with accusations of being manipulated for personal and political

gain. This unforeseen permutation of the repetition hypothesis boils down to disgust

at extensive repetition of accusations without any follow-up providing the details of

who is blameworthy in the first place. Such was the case with the Pal Schmitt faked

doctorate scandal in 2012; the President of the Republic was accused by a major news

magazine of having plagiarized his dissertation from several sources. While this was

124

an enormous story, nobody ever got to the bottom of it, and key documents went

missing.

Another similar case involves accusations of vote buying in by-elections in the

town of Baja. A tape surfaced appearing to show poor Roma citizens being paid for

voting for MSZP. The Fidesz government maintained that the opposition bought off

local voters, but several other sources came forward alleging that Fidesz staged and

planted the tape to smear the opposition. In the end, this large political crisis also

went unsolved. As one participant (2.7) said during Group 2 when a scandal involving

corruption in the Budapest Public Transit came up: “allegedly that’s what happened,

but how would people like us know that?” This exact sentence came up in Group 4

about the Simon deal (participant 4.8), and when asked if the Trafik scandal would

influence anyone’s voting decisions in Group 2, participant 2.8 noted, “For that to

influence us we would need to know if it is correct, and we never will. That’s what’s

revolting.”

5.5 The Age Hypothesis

During an interview with a left-leaning female resident of a small town outside

of Budapest, I learned in great detail the way each major politician in the country

had betrayed the trust of the public. This interview subject was clearly politically

sophisticated. She offered precise details about each politician’s actions during the

EU accession negotiations, the battle over university fees and student loans, and

numerous other specific issues. She detailed who she had chosen to support for each

election and why, often explaining a split ballot and why she had chosen specific

candidates. At each turn, she described the scandals and betrayals that had left

her hopes for competent and honest government dashed. In a pattern I saw again

and again, when it came to the 2014 elections, she had decided not to vote for the

125

first time. “There simply is no party that I could bring myself to support.” Why

not protest vote for a new party? The left-libertarian “Politics Can Be Different”

party seemed to me to be nearly tailored for her varied policy stances. She rolled

her eyes, saying she had not bothered to learn much about the party but what she

had heard led her to expect more of the same from them: shrill scandal and awful

mismanagement.

This once attentive citizen had become completely politically disengaged, but she

was not apathetic. Instead, her experiences with politics over time had led her to

feel a powerful rejection of and queasiness with politics characteristic of disgust. She

was in a way politically passionate, but passionately against the entire Hungarian

political system. Her story was representative of the general findings regarding the

age hypothesis. As citizens gain more and more experience with politics, they seem

increasingly likely to react with disgust. Additionally, the effect is strongest for

people who had once been moved to act — those who had taken part in the protests

after the lying prime minister scandal or student protests and had seen action lead

to no change. However, older participants were not necessarily less angry; instead

of supplanting anger, disgust seems increasingly to accompany anger with political

experience over the years. Thus, while there was scant evidence for a key period of

political socialization leading to disgust, the interview and focus group data point to a

more cumulative effect that plays out over long periods of time experiencing populist

outrage in the political information environment. Indeed, more experienced citizens

in Hungary expressed complete exhaustion when it comes to politics. Having been

outraged again and again, and having seen the new alternatives they’d supported

let them down at every turn, they react with disdain and disgust. In response to

my question as to how the Simon scandal might affect her voting, one interviewee

summed up the general mood of those who had followed Hungarian politics for more

126

than a decade and noted that “it doesn’t affect me anymore. The sad thing is that

we have become so used to it that people in Hungary just accept it as a normal part

of politics.” As of our interview two weeks before the election, she was unsure of who

she would vote for or if she would perhaps abstain from voting for the first time.6

This cumulative effect of age upon political disgust was apparent when comparing

the two younger focus groups to the two older focus groups. In general, the younger

groups had a greater sense of both internal and external political efficacy and felt

they could take action to make changes. Larger proportions of respondents planned

on voting.7 A sense of anger at the injustice of their political system drove these

trends. Right off the bat, when asked about politics they aired displeasure and even

skepticism, but their frustration with politics was unadulterated anger. When asked

to introduce themselves and say briefly how they felt about politics, one participant

(3.2) in the young left-leaning group noted immediately “I’m interested in politics but

don’t always have a formed opinion on the matter. [Moderator: That’s okay what

do you mean?] “Well, the whole thing outrages me a bit.” Other participants in the

young groups mimicked this anger. In this same left-leaning young group (3), the

first person to respond to the moderator’s prompt of how they “feel about politics in

Hungry” responded “unsatisfied.” When pressed, he elaborated that he was angry at

Fidesz, whom he had previously supported, because of corrupt pension reforms, the

trafik scandal, and the current election being rigged to benefit Fidesz. He planned on

voting for one of the opposition parties, though he hadn’t determined which of the

many fragmented leftist parties he might support.

6Interview with late 30s female, Szentes, March 2014

7A caveat: participants in these groups unanimously said they would vote. From actual demo-graphic figures on voting, it is unlikely that they all did as increased age is a predictor of voting inHungary as it is in most countries.

127

This is not to say that the younger groups lacked signs of disdain and disgust. One

participant (3.2) in the younger left-leaning group noted that the state of politics in

Hungary during the current campaign made her just want to ignore the entire thing.

When asked to elaborate, she repeated “just the whole thing” no matter how much

the moderator pressed her for details. This general sense of malaise is a sure sign of

disgust. However, this young person still stated that she would vote, so she has not

yet exited the political system.

Both older groups showed much greater signs of disdain and disgust towards pol-

itics. However, they did not appear any less angry. On the contrary, the emotional

reactions of the older groups were more powerful overall, with these participants be-

ing much more likely to get riled up. This came out immediately in the groups when

the moderator asked participants to say just a bit about themselves and how much

they follow politics. In both younger groups participants typically said they followed

politics and felt that it was important, whereas in the older groups the mood turned

despondent immediately. One participant in the older right-leaning group summed

up the feeling of the whole group when he noted that when it came to politics he

“feels worse about it by the minute” (Participant 2.7). When directly asked by the

moderator what she saw as the big issues in the 2014 campaign, another participant

(2.6) in that same group noted that she “hasn’t been able to bring [herself] to even

listen or read anything about it at all.” At the end of this group the moderator asked

point blank whether the whole thing was too disgusting to deal with and, in sharp

contrast to the relatively empowered and enthusiastic younger pro-government group,

there was a general nodding of agreement. Participant 2.2 took the lead and detailed

how she had been a supporter of the right for decades but couldn’t bear to support

them any longer, to which participant 2.5 replied, “I think that’s the most general

feeling in the country right now.” Talking about the trafik scandal, participants in

128

this right-leaning group didn’t engage in the partisan rejection that vocal participants

in the younger pro-government group had. Instead, when the moderator asked if the

allegations against Fidesz were based in reality, most in the group nodded yes. How-

ever, when the moderator asked whether it would affect their political choices, they

said no. As participant 2.3 noted, “this is just one thing among so many others.”

Interestingly given the right-leaning Fidesz government in power, the right-leaning

older group was by far the most despondent of the lot. Indeed, members of this group

were very disoriented with the country and the Fidesz leadership (with the notable

exception of participant 2.1 who was an obvious Fidesz partisan). One participant

noted that she had never seen a group of people “get so rich so fast” as Orban’s

cronies to general murmurs of agreement and one participant nodding and saying

“terrible, vile.” They were unhappy with “filthy talk” in the media (participant 2.2)

and there was a long general discussion of how the media is too corrupt and full of

liars to give any credit. Though this group was selected based upon media habits that

would generally indicate support for Fidesz, they were much less enthusiastic about

the ruling party than the younger pro-government group. Indeed, when asked to free

associate words and examples about the “party you typically view most favorably,”

participant 2.3 noted that “this is the party I dislike the least, I still don’t like them”

to much murmuring of agreement. This was absent in the younger pro-government

group, where several participants were very strongly in favor of Fidesz.

The left-leaning older group also showed high levels of disgust towards politics.

Indeed, more-so than any other group these participants used the language of disgust

in talking about politics, calling specific politicians and political parties, as well as

politics in general “disgusting,” “repulsive,” “abominable,” and more. However, this

disgust was expressed with a surprising vehemence that can only be labeled anger.

This is likely particularly true of this group because there were a few strong MSZP

129

partisans8 who were obviously extremely angry at the current government, which

influenced the direction of the conversation a great deal. Still, this was true of the

other older group as well; as noted above, these two groups were by far the more

animated on political themes overall. The major theme of the conversation in this

older left-leaning group was the manipulation of the masses by the Orban government;

that they have, in the words of participant 4.2, convinced enough people to “believe

their fairy tale” for their power grab. During long diatribes, multiple group members

railed against instance after instance of their lies and how this made them want to

just shut down. Participant 4.4 summed the group’s consensus up best. When asked

what outraged him most about politics over the past ten years, he replied “the entire

government.” When asked what had disgusted him the most, he again answered

“the entire government, the whole thing, all of them.” Every member of the group

fervently nodded along.

The focus group data show that disgust towards politics does increase over time.

Though age is a very rough proxy for the exposure to populist outrage and disap-

pointment over time that lead to political disgust, the general trend of a cumulative

increase in political disdain with political experience was clear in the focus groups

and interviews. However, disgust does not necessarily replace anger. Instead, it joins

anger and works in combination to fuel an energetic rejection of the system.

Another unexpected finding was the intensity of the disgust for the media itself,

especially among the older groups. The perception that there is no way for concerned

citizens to access the truth about politics is a major mechanism that turns experiences

with the various instances of populist outrage into deep-seated disdain for the entire

process. This was somewhat present in all groups, though it was strongest in the older

groups and in Group 4 (older left wing) in particular. At the start of this group, when

8One of the most vocal participants had several family members who had served as local electedofficials for MSZP in Budapest

130

asked to simply introduce themselves and say “a word or two” about whether they

are interested in politics, two group members went on a long rant about how there

is no way to get good political information because the media is so biased. Towards

the end of this unsolicited venting, participant 4.7 threw her hands up in exhaustion

and noted that “to watch the same piece of news covered by two different channels

you would absolutely have to conclude that there are two different Hungaries.“ When

talking about the Simon scandal allegations against MSZP in the older right-leaning

group, one participant (2.8) noted that, of course, with MSZP out of power they

were hearing a lot of such stories about them. He noted, however, that were MSZP

in power and Fidesz in the opposition, we would instead be hearing predominantly

Fidesz scandals. This sense of a lack of reliable information and of pervasive media

manipulation by the wealthy and politically powerful permeated both older focus

group discussions, and this itself was one of the greatest causes of both physical and

verbal displays of disgust.

In sum, the focus groups provide rough supportive evidence for a slightly modified

version of the age hypothesis: while disgust does not replace anger as a dominant reac-

tion to populist outrage, it grows alongside anger. Given the non-random sample and

the very small size of these groups, this finding is not definitive. There is, however, a

strong sense that those who have been plugged into Hungarian politics for longer have

much more negative and despondent orientations towards politics, even among those

that support the party in power. Of course, within each group there were exceptions.

Participant 4.1 who was just really interested in politics and had been all of her life,

and participant 4.3 had several family members who have served in local Budapest

politics with the Socialist Party over the years. Both of these participants injected a

great deal of anger and desire to act into the older left-leaning groups. However, the

virtue of focus groups is that over the course of the conversation, individual exceptions

131

are largely drowned out in favor of the general mood and experiences of groups as a

whole. In other words, the social aspect of the focus groups ensured that the shared

attributes of group members’ experiences directed the flow of the conversation more

than the individuals’ idiosyncratic attributes, such as having several family members

serving in public office. Moreover, even this exceptional focus group member, unique

for her feelings of efficacy and optimism, muttered “poor thing” as the moderator

explained I was conducting research on Hungarian politics, a reminder that people

and their political perceptions are complicated.

As with the focus groups, the survey data show little evidence of a pattern of

initial anger at a young age being complimented by increased disgust. Again, these

data point to an increased occurrence of both emotions with increased age. The

survey data, however, show a weaker relationship. Figures 1 and 2 show the mean

emotional responses by age cohort. These figures report the number of times partic-

ipants chose a given emotion during the series of political scandals, averaged by age

cohorts. Measuring only the average response to the prompt of how Hungarian poli-

tics in general made participants feel yields the same results. As these figures show,

disgust and anger in particular start out low in the youngest age cohort and gradu-

ally increase over time. Other methods of analysis produce similar, though weaker,

results. Though both anger and disgust are positively correlated with age, the cor-

relations are minuscule and not statistically significant (anger and age are correlated

at .049(p=.120) and disgust and age at.027(p=.389)). Age is not a statistically sig-

nificant predictor of anger or disgust in numerous models predicting these emotions

presented in Chapters Four and Six, as well as many unreported models estimated

during the analysis.

As Figure 2 demonstrates, the young in the general public seem more likely to be

apathetic than any other group, and there is a strong relationship between increased

132

(a) Disgusted

(b) Total Count

Figure 5.1: Disgust & Anger by Age Cohort

133

(a) Anxious

(b) Apathetic

Figure 5.2: Anxiety and Apathy by Age Cohort

134

age and decreased apathy. In hindsight this is less than surprising given the received

wisdom that older citizens are simply more involved in politics. However, one surpris-

ing unexpected result is that anxiety is extremely high among the young. As Figure

2a shows, the 18-24 year old cohort is by far the most anxious when thinking about

politics. Thus, where young survey participants are not apathetic towards politics,

anxiety is the dominant emotion towards both politics in general and specific instances

of populist outrage. The particularly vulnerable state of young people in the weak-

ened Hungarian economy may account for this. This is likely a cross-national trend;

young people face stagnant wages that deeply suppress lifelong earnings globally, to

say nothing of high youth unemployment in places like Spain and Greece.

However, there may also be other developmental aspects to this. This period of

uncertainty in life, along with by such troubling political times makes anxiety the

key among young Hungarians. This calls for future research on the role of anxiety in

political socialization, particularly in the uncertain contexts of young democracies.

5.6 Discussion

The evidence presented in this chapter points to an increased likelihood of disgust

towards politics with repeated exposure to populist outrage over time. This is the case

both as a single instance of populist outrage is repeated as nauseam, and as people are

bombarded with a stream of populist outrage that goes unchanged despite their best

efforts over the course of their lives. The evidence in the qualitative data supporting

both the repetition hypothesis and a cumulative effect regarding the age hypothesis is

robust, however, the survey results are somewhat weaker. These results will therefore

require extensive replication, and continuing to test the specific pathways leading to

political disgust as a function of age, experience, and repetition remains an area of

vital future research. The most rigorous way to do so would be panel data measuring

135

the emotional reactions of the same people over long periods of time. This will be

costly in terms of both time and money, but given the increased scholarly acknowl-

edgement of the importance of emotional processes in determining political opinion

and behavior, there is reason to hope for increased survey data on emotions towards

politics in the future that can help shed more light on how emotional reactions evolve.

Such research would inform both political science and psychology, as both fields lack

sufficient theorizing and data on how emotional processes unfold over time.

The results above, and a number of caveats, point to interesting political implica-

tions and future pathways for research. One key issue is the potential for conflation

of repeated exposure to an instance of populist outrage over time and the severity

of that instance. In other words, perhaps disgust is a reaction to scandal severity

and not repetition. For instance, the lying prime minister scandal of 2006 referenced

above was indeed a mainstay of the Hungarian political dialogue for eight years and it

had led to disgust, but perhaps this disgust was in reaction to the sheer severity of the

original act and not to its incessant repetition. Would a survey in 2006 have found the

same level of disgust towards this story right as it broke? This is a difficult problem

to untangle as these attributes of an instance of populist outrage will always be highly

correlated in the real world: the most serious outrageous accusations or scandals will

get the most coverage and last in the political dialogue the longest. However, while

there are attributes of an instance of populist outrage other than the amount of ex-

posure and time that may lead to disgust, psychology research on this emotion would

not support the expectation that sheer severity of the moral transgression would be

one of them. Given the physical domain of disgust(Kelly, 2011), instances of populist

outrage dealing debasing the body or engaging in sexual indiscretion will likely lead to

disgust from the onset (Haidt, 2003; Vandenbroek, 2012). However, when the moral

transgression has to do with lying for personal gain, the nature of the sin likely elicits

136

anger at first. The lying prime minister scandal is more of the class of scandal as the

anger-inducing Simon dealings that broke during the 2014 election and likely became

disgusting over time through disgust’s associative learning mechanism.

Another important consideration is the role of economic hardship in determin-

ing both emotional reactions to political information and political behavior. The

older groups were much more economically downtrodden than the younger groups.

Numerous unemployed citizens participated in Group 2 in particular, and the sense

of hopelessness at this permeated a great amount of the conversation. As noted in

Chapter Two, however, the idea that economic circumstances also affect emotions

towards politics does not contradict the causal linkage between populist outrage and

anger and disgust towards politics at the heart of the present work. People facing

economic hardship likely react more strongly to populist outrage, and are also more

susceptible to becoming disgusted with politics as legitimate fears of dire economic

circumstances make the often frivolous back and forth of outrage bating seem all

the more base and immoral. Indeed, Seawright (2012) views economic hardship as

a precursor to anger exerting a causal influence on the party system: the anxiety of

economic hard times increases uncertainty and allows anger’s motivation tendency

to be directed towards punishing the establishment and rewarding newcomers to the

party system with votes. Digging deeper into the interconnected relationship be-

tween economic circumstances, emotional reactions to political appeals, and political

attitudes and behaviors remains a key area of future investigation, but the economic

factor works alongside and not against the hypotheses tested here.

One somewhat unexpected major finding of the research on how disgust evolves

over time was the intensity of disgust towards the media as citizens gained more

and more political experience. While this loathing for the media was particularly

pervasive in the older focus groups, it was present to some extent in all the focus

137

groups and most of the interviews. In retrospect, this disdain should not have been

a surprise given the combination of enormous populist outrage in the media just

as Orban’s media law, giving the government enormous influence over the media,

was in full effect. However, the research here suggests not only that the media can

become an object of disgust, but that Hungarian citizens are not passively receiving

populist outrage or any political news. On the contrary, most of the focus group

participants knew they received biased news, acknowledged the salacious nature of

their political dialogue, and intimated that they also understood this to be a strategy

used by politicians to make themselves seem different and better than competitors.

These citizens were not buying it. Here my findings directly mirror work on the

scepticism of Russian citizens as Putin took over their media (Mickiewicz, 2008), as

well as exciting new work on how individual motivation to find accurate information

influences the perception of the political dialogue (Carnahan, 2012).

The most important caveat, however, is that my hypothesis that with time disgust

replaces anger is only partially borne out. These results show that disgust does not

replace anger, but instead supplements anger over increased repeated exposure to

populist outrage and experiences with outrageously immoral politics. These data do

lead me to conclude that disgust is a much slower emotional reaction to evolve than

anger, so there is merit in the distinction between outrage motivation and outrage

exhaustion. However, outrage exhaustion is not purely disgust, but a passionate

mixture of rage and disgust. This co-occurance mirrors findings in the previous

chapter concerning partisanship. To recall, the chapter showed that it is true that

the strong partisans were uniquely angry, but it was not the case that they were

only angry. They were also highly disgusted, most likely at the other side. The

co-occurance of these specific emotions and the complicated way in which several,

often conflicting, emotions may arise and shape thought and behavior in complex

138

and contingent ways is a trend found in other emotion research,9 and requires careful

future theorizing and measurement. I will return to this theme at length in the

concluding chapter of this dissertation.

9See especially work by Giner-Sorolla (2012).

139

Chapter 6: Disentangling the Political Implications of

Disgust from Anger

Though Hungary was once viewed as the most successful case of democratization

in East Central Europe (ECE), in the wake of the Orban government’s new constitu-

tion and power centralization, it is at best a very weak democracy (Herman, 2015),

lacking institutions of democratic accountability in any real sense (Scheppele, 2013).

However, with the exception of scattered and short-lived protests from the left, the

Hungarian public has not exerted pressure upon their leadership to combat this open

and unabashed dismantling of Hungarian democracy. Political participation in gen-

eral is increasingly lackluster in the country. Turnout has fallen in every Hungarian

election since 2002. Indeed, the 2014 election that gave Fidesz another term in office

through a mix of lukewarm support and the manipulation of electoral institutions had

the lowest turnout in the post-communist era of multiparty politics. Radical voices

embodied by Jobbik seem to have a monopoly on political action, and far from mobi-

lizing public pressure to defend democratic institutions, they have waged a constant

campaign for a more authoritarian, less liberal, and more exclusionary Hungary.

Disgust towards politics helps explain this public quiescence towards the Fidesz

government’s power grab. Having shown the rise of populist outrage in Hungary and

attended to the emotional effects of this political strategy as a function of both time

140

and attributes of the citizen, this chapter turns to the distinct effects of anger and dis-

gust upon public opinion. Doing so requires delving into the distinction between these

emotions, a continuing area of exploration in psychology. This work suggests that

disgust and anger are core emotions with unique domains in moral behavior: anger

compels us to confront an offending party to fix a situation, while disgust isolates

and avoids dangerous groups and ideas. I hypothesize that the differences between

anger and disgust can help account for seemingly contradictory trends, from a lack

of political opposition to the government as it dismantles liberal democracy to angry

action that fuels the country’s radical politics and unhinges political institutions.

These differences are in large part associated with the distinct appraisals that

predict anger and the associations that elicit disgust. Anger comes about because of

blame towards an actor who is perceived to have intentionally committed an act of

aggression, while repeated adverse feelings towards an object elicits disgust. Thus,

anger predicts less satisfaction with particular actors and policies that can lead to

taking action to correct these perceived wrongs. Disgust, however, is associated with

a desire to detach from politics and overall low feelings of efficacy. Indeed, disgust

towards politics could corrode support for the very idea of democracy.

This chapter combines many of the data sources discussed in previous chapters

to test these predictions. The survey data show that disgust towards politics equates

to a rejection of the very notion that democracy is a desirable system for Hungary

in the first place, while angry citizens are more likely to have negative evaluations

of how the system works but not of the desirability of the system. The lab-in-the-

field experiment reinforces these distinctions, showing that anger leads to decreased

evaluative support for the government and democratic institutions as they work in

Hungary, but that disgust leads to more generalized feelings of a lack of political

efficacy in the system. Finally, the pilot data demonstrate the role that blame plays

141

in political anger. These findings are significant not only for their demonstration of the

microfoundations of democratic backsliding in Hungary and other similar cases, but

also for documenting the distinctions between evaluations of how democracy works

and normative commitment to the ideal of democracy. Indeed, while both of these

negative emotions detract from popular support for democracy, they do so through

different pathways and through different dimensions of support.

142

6.1 Political Behavior and the Adaptive Functions of Angerand Disgust

Emotional response is a critical component of human behavior, and the brain

reacts to stimuli rapidly and subconsciously with emotions in a way that directs con-

scious cognition and behavior (Zajonc, 1980). This is why a recent turn in political

science acknowledges the need to take emotions into account to fully understand po-

litical thought and behavior (Marcus et al., 2000; Lodge and Taber, 2013). While

the study of emotion is a sprawling field reaching across numerous disciplines, a gen-

eral consensus exists that emotions are evolved neural mechanisms that perform the

adaptive function of providing information about the surrounding environment and

encouraging an appropriate response.1 Thus, emotions can be viewed as mechanisms

that perform many functions, chief among them linking stimuli in the environment

to adaptive thoughts and behaviors that preserve survival. As the study of emotion

progressed, scholars increasingly came to view emotions as distinct modules, often in

competition with one another. Thus, a single situation may elicit a set of emotions,

and these emotions often drive contradictory action imperatives.

Disgust and anger are two such emotions. They are each core emotions with long

evolutionary histories: anger initially evolved to fight off threats through confronta-

tion, and disgust to keep physical impurities out of the body through expulsion and

rejection. Over time both emotions have evolved into critical regulators of group

behavior through the levers of morality (Haidt, 2003; Giner-Sorolla, 2012). Indeed,

morality is felt, communicated, and enacted through emotional systems (Haidt et al.,

1997; Prinz, 2007). Both disgust and anger are “other-condemning” emotions im-

plicated in regulating behavior within and between groups (Haidt, 2003). Thus, the

1See Giner-Sorolla (2012) and Lazarus (1991) for full expositions of this perspective.

143

emotions have an overlapping domain: the condemnation of actors who have violated

some moral standard.

There are, however, important distinctions in their elicitation and the related

behavioral consequences. Anger is most strongly associated with groups that are

perceived to obstruct a goal, while disgust is a reaction to groups that are perceived

to lead to some sort of contamination. (Cottrell and Neuberg, 2005). Thus, blame

and intentions of wrongdoing drive the anger response Quigley and Tedeschi (1996),

while disgust is elicited from repeated associations with physical or moral impurity

(Rozin et al., 1999; Kelly, 2011). Indeed, the very cognitive mechanisms implicated

in each emotional response is different. The disgust response is associative and thus

develops towards an object as a result of repeatedly pairing the object with distasteful

or unpleasant sensations. Anger, however, involves the appraisal of blame and ill

intentions on the part of the transgressing actor. These emotions even differ in their

basic neurological circuitry. Anger, like other emotions implicated in responding to

threat (such as fear and anxiety), is associated with the amygdala. On the other hand,

disgust, due to its origins in protecting the body from contaminants, is primarily

associated with the gustatory complex (Kelly, 2011).

The consequences of disgust upon political behavior differ from the political con-

sequences of anger. The latter is associated with the urge to confront while the former

leads to avoidance and rejection of the offensive object (Roseman et al., 1994; Haidt

et al., 1997; Carver and Harmon-Jones, 2009; Giner-Sorolla, 2012). In other words,

we want to confront and fix a situation that angers us, but we distance ourselves from

objects of disgust. The implication for politics is clear: angry citizens are spurred

to act, while citizens who are disgusted by multiparty politics will want to distance

themselves from politics and the political system as much as possible.

144

These general trends are present in many circumstances in politics, and where they

are research finds both action tendencies. One instance in which anger and disgust

each contribute to public opinion and behavior concerns attitudes towards democracy.

Indeed, the distinction between these emotions may help us understand different

dimensions of attitudes toward democracy. Evaluative satisfaction or dissatisfaction

with the democratic system as it works in the country is the most commonly employed

measure of support. This is, however, but one aspect of a multifaceted cluster of

attitudes. Dating back to Easton (1965), a contrast has generally been made between

“diffuse” support for the regime broadly, and “specific” support for how the regime

functions. Specific support for democracy fluctuates as a function of the economy,

the partisanship of the government, and several other factors based upon the actions

of those in power. Diffuse support is a more stable orientation towards the system

that is less likely to fluctuate (Finkel et al., 1989; Klingemann, 1999). Montero et al.

(1997) and Gunther and Montero (2006) go even further, arguing that in addition to

low specific support, which they call political discontent, or diffuse support, which

they simply label democratic support, political disaffection is a third independent

attitudinal spectrum. This is a “syndrome of alienation and disengagement from

active involvement in the political process” (Gunther and Montero, 2006, 47). In

the conclusion I will return to the ways both anger and disgust influence political

disaffection.

Here I focus upon the more common distinction between diffuse and specific sup-

port. I label Easton’s specific support and Montero et al.’s political discontent sat-

isfaction with how democracy works, and will refer to diffuse support or Montero’s

democratic support as normative commitment to democracy. As satisfaction with

how democracy works involves judgments and evaluations of performance (Kornberg

and Clarke, 1992; Farah et al., 1979), attributions of blame for political outcomes

145

is central to this attitude. Blame drives the anger response, and anger will there-

fore affect satisfaction with how democracy works. Disgust, on the other hand, is

hypothesized to affect the more general orientation towards the system per se, de-

creasing normative commitment to democracy by making the very idea of democracy

unpalatable to many citizens.

6.2 The Emotional Roots of Backsliding

The survey introduced in Chapter Four includes measures that illuminate the role

that anger and disgust play in public opinion in Hungary on a large and represen-

tative sample.2 This survey of 1,000 Hungarian adults included unique measures of

emotional orientations towards politics as well as a number of political attitudes and

(self-reported) behaviors. This unique combination means we have an opportunity to

examine the political correlates of anger and disgust towards politics.

Descriptive data on the key dependent variable of interest — normative commit-

ment to democracy — serves as a useful point of departure for the analysis of the

political ramifications of disgust and anger. The survey asked participants to rate

their agreement or disagreement with the statement “Democracy is the best political

system for a country like ours.” The responses are presented in Figure 6.1.

The data show that there is indeed variation in normative support for democ-

racy in Hungary twenty-five years after the move to competitive multiparty politics.

Though 54.9% of participants at least somewhat agree with the statement, these data

are cause for alarm. Only 14.7% of respondents “completely agree” that democracy

is the best political system for the country. Meanwhile, 20.5% of the sample neither

agrees nor disagrees with the sentiment, showing a sizable portion of the electorate

2For a full discussion of the survey methodology and sample, see the discussion in Chapter 4 andthe Appendix.

146

Figure 6.1: Democracy is the best political system for a country like ours

holding ambivalent views of democracy. Finally, 18.2% of the populace openly report

antagonistic views of the desirability of democracy for the country. This is higher than

the 13% of normative opponents to democracy that Evans and Whitefield (1995) find

in their 1994 Hungarian survey,3 suggesting that, if anything, the subsequent twenty

years of democratic experience lessened normative commitment to democracy. More-

over, this is likely to be a conservative estimate. The question wording opens up

the possibility of acquiescence bias favoring a positive valuing of democracy, while

the format of the survey in which the interviewer was present (though responses on

the laptop were anonymous), runs the risk of social desirability bias in favor of re-

porting a desire for democracy. Thus, while normative commitment for democracy

is widespread in Hungary, at least 40% of Hungarians are ambivalent towards or

outright critical of the idea of democracy.

3Table 1, page 489

147

This survey also included a number of measures that tap into political views and

behaviors, making it a unique opportunity to match emotional orientations towards

politics to their political consequences. Models including these measures as depen-

dent variables are presented in Table 6.1. All of these variables were measured on a

seven-point Likert scale ranging from “strongly disagree” (1) to “strongly agree” (7).

Vote intention and attitudes towards democracy were included in this battery as de-

pendent variables, and sociotropic economic evaluations were included to control for

the possibility that economic hardship, not emotional reactions, accounts for weaker

support for democracy. Thus, the variable Economy taps agreement with the state-

ment that “Over the past years the state of the Hungarian economy has improved,”

with higher values indicating agreement and more sanguine views of the economy.4

The first column of Table 6.1 presents an OLS model predicting agreement with

the statement “If there were to be an election this Sunday, I would definitely vote.”

Education is associated with increased likelihood of voting, which is to be expected

given long-standing findings that education endows citizens with social and cognitive

resources that encourage political participation (Verba and Nie, 1972). Age is another

strong predictor of self-reported turnout, which is in line with a view of voting as a

habit acquired over time and with age (Miller and Shanks, 1996; Gerber et al., 2003).

Positive evaluations of the economy also increase the likelihood that a participant

reports willingness to vote. Meanwhile, Anger is positively and significantly associ-

ated with self-reported vote intention, while disgust is not related to intention to turn

out. This is precisely what we would expect according to H4a: angry citizens are

mobilized to take action. It is also suggestive of the relationship hypothesized in H4c,

as disgust leaves other citizens unmoved to act, though this negative coefficient is

4Alternative specifications including a control for ideological extremism did not change the sub-stantive results and are included in the Appendix in Table B.4.

148

Table 6.1: Anger, Disgust, and Attitudes towards Democracy (OLS models)

Vote Sunday Satisfaction Hungary a Dem the Bestw/ Dem Complete Dem System

Anger .16*** -.06*** -.12*** -.01(.03) (.02) (.02) (.02)

Disgust -.005 -.05* -.06** -.12***(.04) (.03) (.02) (.03)

Age .02*** .001 .00 .004(.00) (.02) (.00) (.003)

Education .30*** .06 -.07 .05(.07) (.05) (.04) (.06)

Female -.02 .16* -.01 .003(.14) (.09) (.08) (.10)

Economy .24*** .67*** .61*** .27***(.04) (.03) (.02) (.03)

Intercept 2.41 1.12*** 2.05*** 3.68***(.33) (.22) (.21) (.26)

Adjusted R-sq. .08 .44 .44 .09N 943 935 941 917

OLS coefficients with standard errors in parentheses.

∗∗∗p < 0.01, ∗∗p < 0.05, ∗p < 0.1

149

extremely small and not significant, meaning that, statistically speaking, this model

does not show the effect of disgust to be different from zero.

The next models predict agreement or disagreement with the statements “I am sat-

isfied with the way that democracy works in Hungary” and “Hungary is a completely

democratic country,” respectively. While both anger and disgust are associated with

less agreement with these statements of democratic support and vigor in the country,

anger is more strongly associated with disagreement. Thus, both of these emotions

detract from perceptions that democracy exists and works in Hungary. It would also

appear that there is some support for Przeworski (1991)’s concern that economic

success is important to sustaining support in nascent democracies; positive economic

evaluations are associated with a rosier view of democracy in Hungary.

The final column of Table 6.1 provides evidence of the unique influence of disgust

upon the key dependent variable of interest: normative commitment to democracy.

This model predicts agreement with the statement “Democracy is the best system

for a country like ours.” The model shows that disgust, but not anger, is significantly

associated with a lack of agreement. This is precisely what H4c predicts: disgust

causes an overall rejection of the political system. Thus, the results support the

suggested disjuncture between angry and disgusted citizens. Angry participants are

moved to action, but the disgusted are left so repulsed by politics that they abandon

the very ideal of democracy. Indeed, this evidence suggests they do not even hold

democracy as an important or desirable value for their country.

6.3 Evidence from Experiments

The lab-in-the-field experiment and the pilot study introduced in Chapter Four

also included measures that explore the relationship between the emotional responses

of disgust and anger and political attitudes and behaviors. In both studies, the short

150

surveys following the manipulations and measures of emotion included a number of

different measures of attitudes towards politics and the democratic system. The

effects of the attacks are evident mainly through the emotions they elicited and in

many cases were not as evident as differences between conditions. This is likely

because the treatments elicited high levels of both anger and disgust among different

participants in each group and these emotions have effects in the opposite direction on

many key variables of interest, drowning out differences among conditions.5 However,

in key areas regarding perceptions of blame, political action, and political efficacy,

anger and disgust exhibit unique and illuminating effects.

Table 6.2 displays models using the lab-in-the-field experiment data to predict

anger and disgust’s effects on attitudes towards the democratic system while control-

ling for other important variables.6 To recall, undergraduate students in social science

classes at a Hungarian university in Budapest took part in this study during the 2014

elections. These models include several controls to isolate the relationship between

emotion and political attitudes. The variable Ideology measures self-placement on

an eleven point left–right scale, with higher values indicating placement to the right.

Hungary is governed by a right-leaning party, and therefore the expectation is that

those on the right — with a higher score on Ideology — will have more positive

evaluations of the government. Likewise, personal economic situation will also exert

a powerful effect on these evaluations, and the variable Economy — an index con-

structed from how severely participants and their families have struggled with three

economic issues7 — captures this with higher levels indicating greater economic prob-

lems. Finally, gender and political knowledge may also be important to understanding

5I return to this topic briefly at the end of this chapter and in detail in the Conclusion Chapter.

6For measure wording, descriptive statistics, and alternative model specifications, see AppendixXX.

7Unemployment, stagnation in earnings, and rising costs

151

political attitudes; the dummy variable Female captures gender and Knowledge rep-

resents a political knowledge scale constructed from six questions asking participants

to identify various prominent politicians.

The dependent variables are three- and four-point ordered Likert scales, so ordered

logit models were estimated. The first column of Table 6.2 analyzes evaluations of the

current government. This is measured by responses from “Poorly” to “Very Well” to

the question “In your opinion, how well has the government performed since taking

office?” As expected, the model shows that those on the right are much more likely

to report that the government has performed well or very well, while those facing

economic hardship are less likely to do so. Anger is associated with lower evaluations

of government performance, which is in line with expectations given the importance

of attributions of blame to anger. There is, however, no relationship between disgust

and this evaluation. The second column shows similar trends, linking anger but not

disgust to an outcome that gets at the intentions of politicians in politics. Here again,

anger but not disgust predicts agreement with the sentiment that “Politicians don’t

care about people like me.” Anger is thus associated with perceptions about who

politicians do or do not attend to, while disgusted participants are unlikely to see

either side, whether the government or the parties of the opposition, as having any

particular intention to favor any group.

The third column reports suggestive evidence that that those who became angry

are more likely to believe that if they are frustrated with politics they can vote to

make a change. This provides support for the hypothesis that political outrages lead

to action where they elicit anger (H4a), and is in line with work in psychology on

anger’s impulse to take action in general (Carver and Harmon-Jones, 2009) and in the

realm of politics specifically (Marcus et al., 2000; Valentino et al., 2011). The finding

that angry voters are more likely to feel empowered to vote and bring about changes in

152

Table 6.2: Ordered Logit Models: Effects of Anger and Disgust on Political Attitudes

Govt Politicians Vote if TooPerformance Care Unhappy Complicated

Anger Scale −1.37∗∗ −1.26∗∗ 0.99α −0.15(0.63) (0.58) (0.60) (0.58)

Disgust Scale 0.81 −0.04 −0.36 −0.92α

(0.63) (0.57) (0.60) (0.58)Ideology 5.84∗∗∗ 0.83 0.95 −0.11

(0.65) (0.51) (0.52) (0.51)Economy −1.80∗∗∗ −1.46∗∗∗ −0.38 −0.12

(0.54) (0.50) (0.49) (0.49)Knowledge 0.27 1.47∗∗∗ 0.88∗∗ 1.84∗∗∗

(0.44) (0.43) (0.41) (0.42)Female 0.27 0.43 1.11∗∗∗ −0.75∗∗∗

(0.28) (0.26) (0.27) (0.26)

Ordered logistic models, standard errors in parentheses.

∗∗∗p < 0.01, ∗∗p < 0.05, ∗p < 0.1, αp < 0.15

the political system also helps understand protest voting — the support of unorthodox

parties in light of discredited mainstream parties — that is a widespread cause of

constant party system instability in the region (Pop-Eleches, 2010). This, in turn,

helps shed light on party system destabilization: anger provides the action impulse

driving protest voting for new upstart parties against the political establishment

(H4b). Finally, the fourth column shows suggestive evidence that disgust, but not

anger, is associated with lower efficacy; disgusted participants are more likely to agree

with the statement that “politics is too complicated for people like me to understand.”

This decrease in efficacy suggests a degree of political disengagement as a result of

disgust (H4c).

The pilot study introduced in Chapter Four also includes several attitudinal mea-

sures that lend further credence to the distinct role of anger in political behavior. As

153

this study only measured anger towards politics in response to populist outrage, the

pilot data can only bring evidence to bear on the effects of anger and not disgust.

While disgust was not measured in this study, however, its effects can often be in-

ferred. Moreover, these data illuminate the role of blame in anger attributions and

highlight the moral nature of political anger. These data focus on attitudes about the

government’s performance and levels of corruption among members of the governing

party, which, to recall the design of the study outlined in Chapter Four, was the tar-

get of an accusation of accepting bribes from multinational corporations in exchange

for work permits for foreigners.

Table 6.3 presents the most pertinent models linking anger at the outrageous

accusations in the treatment conditions to political attitudes. These models include

the same control variables as in the lab-in-the-field models, and like those, are robust

to different specifications. The model in the first column addresses the extent to which

participants feel that the government has performed well since it took office. The

emotional response of moral anger predicts a more negative view of the government’s

performance among the entire study sample. The model in the second column presents

a similar picture concerning the effects of outrage upon perceptions of corruption in

the governing party. The model suggests that those who are angry do indeed view the

party as significantly more corrupt. The next model estimates the effects of outrage

upon feelings towards the governing party, Fidesz. Anger is a statistically significant

predictor of negative feelings towards the governing party. The next two models show

the effects of anger upon sentiment towards Prime Minister Viktor Orban.8 In all four

models, control variables predict as we would expect. Those who attend church more

frequently and who are farther on the right feel warmer towards the right-leaning

8Models using attitudes towards Fidesz parliamentary leader Laszlo Kover as the dependentvariable reveal identical results to those shown towards the Prime Minister

154

Table 6.3: Moral anger and perceptions of government

Govt Perf Party Corrupt Fidesz PM

Anger -.1* .15** -.23* -.29*(.05) (.06) (.18) (.18)

Knowledge -.02 .05 -.08 -.07(.03) (.03) (.09) (.09)

Ideology .13*** -.12*** .54*** .67***(.03) (.03) (.12) (.12)

Age .01** -.01 .02 .06(.005) (.01) (.02) (.02)

Female .17 -.4*** .90 .98(.11) (.14) (.40) (.39)

Economy -.06* .07* -.27** -.27**(.03) (.04) (.11) (.11)

Religiosity .12*** -.07 .37 -.27***(.04) (.06) (.15) (.11)

Constant 1.27 3.4*** 2.14 .8(.4) (.48) (1.36) (1.31)

N 217 217 217 217R-squared .2 .14 .21 .26

Ordered logistic models, standard errors in parentheses.

∗∗∗p < 0.01, ∗∗p < 0.05, ∗p < 0.1, αp < 0.15

government, while those who report having serious economic difficulties in the recent

past feel more negatively.

The pilot data therefore show that moral anger predicts less positive sentiment

towards the target of the attack, which is evidence of anger and blame working hand

in hand to affect political sentiment. However, the pilot data also present a quite

perplexing trend when analyzing affect towards pertinent political actors by condition.

Sentiment towards the main political actors and parties from the manipulation is

shown in Figure 6.2, charted on 10-point feeling thermometers with a 0 representing

the lowest regard and a 10 the highest. The condition means show an interesting

155

0

outraged angry scared sad

0

1

2

3

4

5

6

7

Fidesz Orban Kover Jobbik

control

political

outrage

Figure 6.2: Affect toward pertinent political actors and parties

156

pattern in the data that is repeated in other measures: the political condition had

a more negative effect than the populist outrage condition. That is, the accusation

of mismanagement decreased support towards the governing party, Fidesz, and its

politicians, Prime Minister Viktor Orban and party leader Laszlo Kover, to a greater

extent than the outrage condition’s accusation of taking bribes. Additionally, the

appeal to populist outrage did not increase support for the protest party Jobbik that

made the accusation. Politicians from Jobbik were the key accusers of the government

in the articles, and instead of scoring political points, there was a backlash to the

accusation amounting to a loss of popularity in the political and outrage conditions.

ANOVA confirmed this difference between the control group and the outrage and

political conditions to be statistically significant.

The weakness of the results when comparing experimental conditions to one an-

other versus more robust effects when looking at the direct effect of anger upon views

of the governing party is perplexing. If the outrage condition elicited the greatest

anger, as demonstrated by the analysis in Chapter 4, and anger predicts more neg-

ative feelings towards the governing party, why are mean levels of positive feelings

toward the government higher in outrage condition relative the political condition?

There appears to be something about the outrage condition causing a backlash. The

data indicate that this is the result of pre-existing views respondents have of the

governing party, Fidesz. From work in the Affective Intelligence paradigm (Marcus

et al., 2000; Brader, 2006), we would expect anger to be associated with an increase

in support among partisans as anger is known to increase partisan predispositions.

Given the partisan lens though which Fidesz supporters view politics, they would

likely become very angry at the accusation because they view the attacking party,

Jobbik, as worthy of blame for making a false or malicious claim.

157

Table 6.4: Emotions driving affect towards the governing party

Anger AngerGovt Supporter Others

Angry .28α -.30*(.17) (.18)

Economy -.18 -.22**(.14) (.11)

Knowledge .20 -.15*(.15) (.09)

Ideology .76*** .13(.13) (.12)

Religiosity .05 .23(.16) (.18)

Age .00 .03(.03) (.03)

Female .36 .49(.53) (.41)

Constant 1.83 3.96***(1.96) (1.35)

N 58 159R-squared .46 .09

Ordered logistic models, standard errors in parentheses.

∗∗∗p < 0.01, ∗∗p < 0.05, ∗p < 0.1, αp < 0.15

158

Separating out Fidesz supporters from the rest of the sample and analyzing anger

among each group illuminates these countervailing forces. As the first model of Table

6.4 shows, among Fidsz supporters anger caused higher levels of support. Thus, anger

activates partisanship and causes a rally around the flag effect in favor of Fidesz and

helps explain much of the backlash in the outrage condition when it comes to affect

towards the government.9 However, the next model shows that among all other

participants this emotion caused decreased warmth towards Fidesz. While the effects

of anger were not statistically significant at conventionally accepted levels among

Fidesz supporters, the p-value of .12 is highly suggestive given a sample size of only

58 for this model. These findings help account for some of the surprising non-linearity

demonstrated in overall attitudes towards the governing party between groups. It also

further illuminates the tight linkage between partisan predilections, attributions of

blame, and the dynamics of the anger response in politics.

Finally, though the pilot study did not measure disgust, the implementation and

analysis of this study provided many clues of disgust lurking behind the scenes. First,

though failure to treat and drop outs are problems with all studies and with online

studies in particular, the nearly 50 percent of the study that dropped out upon read-

ing the manipulations is extraordinarily high. What’s more, participants randomly

assigned to the treatment condition or political condition were much more likely to

drop out, suggesting that the mere mention of politics caused many participants to

leave the study. Those that dropped out were less likely than those who finished to

place themselves on the left-right spectrum or to state they identified with any party

during the pre-test. Indeed, this problem affected the design of the lab-in-the-field

experiment, where participants came to the lab and reacted to statements one by

9Similar models using affect towards Prime Minister Orban, party leader Kover, and Jobbikconfirm these findings.

159

one in order to combat failure to treat driven by visceral disgust at reading populist

outrage in the treatment conditions.

6.4 Qualitative Data

Finally, the focus groups and interviews introduced in the preceding chapter also

provide useful data to examine the attitudinal and behavioral ramifications of anger

and disgust on political behavior. Participants who displayed both verbal and non-

verbal signs of anger (slamming fist on table when talking, pointing finger) were likely

to talk about a particular actor who they saw as blameworthy and something that

should be done or that they themselves wanted to do. For instance, in the first group

(young, right-leaning), there was very long conversation about the Simon scandal in

which a leftist MP was allegedly taking bribes from large companies for favorable

policy. Participants were irate, alleging that the entire MSZP leadership probably

knew about it and took part and arguing that they had to be defeated once again

in this election to keep them from harming the country any more. Meanwhile, par-

ticipants showing signs of disgust (making a face as if they smelled something rancid

while describing an experience) instead talked about how hopeless the entire situation

seemed, and displayed obvious disdain at talking about politics at all. Speaking of

this same scandal, numerous members of the fourth group (older, left-leaning) made

the gape face characteristic of disgust (Kelly, 2011),10 and noted that this is just one

of many allegations and there is likely nothing to be done about it. Moreover, angry

participants would often talk in terms of what is or is not democratic, while when

participants appeared disgusted they gave no indication that the type of political

system mattered to them at all.

10This is the face a person would make having smelled something truly revolting — the sides ofthe mouth are pulled back and down into an exaggerated frown and the tongue may protrude, thenose is scrunched.

160

The qualitative data also garnered more specific support for H2b, the hypothesis

that voters angry from populist outrage will be more likely to support protest parties

and form the microfoundations of renewed party system volatility. In one interview,

a young woman explained her plan to support Jobbik in the upcoming elections. “I’ll

vote for Jobbik, even though they scare me a little bit” she declared, before nothing

that “at least they tell the truth about the others [other political parties]; at least

with them in the parliament they will tell the people when the others do more lying.”

Thus, while she found Jobbik’s specific proposals downright scary, she still wanted

them in parliament to hold the discredited mainstream parties’ feet to the fire. This

is precisely the causal sequence Pop-Eleches (2010) identified in his study of protest

voting in ECE: this interviewee did not ascribe to the vast majority of Jobbik’s ultra-

nationalistic, xenophobic, anti-democratic platform, but instead felt that the other

parties were guilty of repeatedly lying to the public and had lost their credibility to

govern. A similar theme came out in the third focus group, in which several of the

younger, left-leaning participants responded that though they knew Fidesz had to

go, they could not bring themselves to ever vote for MSZP after the 2006 scandals.

Instead, they would vote for a smattering breakaway leftist parties or the new Politics

Can Be Different party.

6.5 Discussion

While the previous chapters documented the rise of populist outrage in politics

and showed how these attacks elicit anger and disgust at different times and among

different people, this chapter demonstrated that these emotions lead to different po-

litical outcomes. Anger causes action against the blameworthy transgressing actor,

while disgust causes disengagement from politics and the democratic system. While

the hypothesis that anger increases party system volatility (H4b) is only partially

161

borne out here, there is support for the idea insofar as specific actor evaluations de-

crease as a function of anger and angry participants are moved to take action against

what angers them. Thus, more work needs to be done tying anger specifically to

party system volatility. Importantly, the results presented in this chapter show that

the distinction between disgust and anger helps us understand different dimensions

of support for democracy. Anger can detract from support for democracy as it works.

Disgust has the more pernicious effect of decreasing normative commitment to the

democratic system as a whole. Thus, as disgust spreads throughout the populace,

desire to have a democracy in the first place plummets.

Both anger and disgust did have corrosive effects on underlying political attitudes

important to a healthy democratic political culture. With the exceptions of disgusted

individuals being less likely to see a role for political parties in democratic politics

and the power of anger to increase the perceived stakes of political competition,

immediate anger and disgust are often both predictive of negative attitudes towards

politics, democracy, and groups. However, instances in which these emotions push

the public in opposite directions often illuminate political outcomes. As this chapter

demonstrates, this is the case in Hungary in recent years. The disgusted feel only

passionate aversion from the democratic system. The angry, on the other hand,

support ever more radical anti-establishment politics that dismantles the pluralistic

nature of liberal democracy.

This distinction also informs the broader debate over when political attacks mo-

bilize and when they have the opposite effect. Ever since the landmark work on the

“demobilization effect” (Ansolabehere et al., 1994), a vibrant debate over whether

attack politics turn people off or actually motivate them to take part in politics has

raged on. Anger and disgust help understand the contradictory findings on work in

negative campaigning. Where these attacks causes anger they will mobilize. However,

162

voters who are disgusted by these attacks will be demobilized. The associations driv-

ing the disgust response and the appraisals driving anger shed light on the conditions

under which a political attack is more likely to drive engagement or disengagement,

and many specific contexts and combinations of these emotions remain open for in-

vestigation. Moreover, given the link between having a side in the fight demonstrated

in Chapter 4 and in the final analyses of this chapter, we can also better understand

individual level variables that moderate the relationship between attack politics and

political involvement.

While this chapter, and the dissertation more broadly, is most concerned with the

distinction between anger and disgust, their co-occurrence is an equally important

topic for study. People are both angered and disgusted by politics, and while the

data presented in this chapter demonstrate unique marginal effects of each of these

emotions upon political attitude and behavior, the level of co-occurrence of anger

and disgust is high. This is likely the case when it comes to what Montero et al.

(1997) label political disaffection. This is even more dangerous when anger and

disgust combine towards a group. Indeed, those who are angry at a group may take

action to punish that group, and if they also feel disgusted towards that group, their

dehumanization of group members may facilitate particularly dangerous retaliatory

action. In a study addressing contempt, which for the purposes of this discussion is

functionally equivalent to disgust, Fischer and Roseman argue that the combination

of anger and disgust/contempt is dangerous “as it implies that the inclination to

attack co-occurs with the inclination to derogate and exclude a person who is seen

as inferior or bad, which may result in hostile acts that are not held in check by

affection or social relationships” (Fischer and Roseman, 2007, 113). In the following

concluding chapter I address these extensions fully.

163

Chapter 7: Implications and Future Directions

“The public has become justifiably disgusted with politics” (Kamber, 1997, p. xiii)

7.1 Research Goals, Methods, and Findings

In this dissertation I set out to understand political volatility and disengagement

that have engendered democratic backsliding in Hungary. These events are surpris-

ing in the context of Hungary, which was long seen to be one of the most successful

and stable post-communist democracies. Over the past four years, however, the gov-

ernment has ridden roughshod over the institutions of liberal democracy, passing a

constitution that severely limits the freedom of the press and judiciary and reformat-

ting the electoral system to strongly favor their continued power. Through all of this,

the Hungarian public has remained quiescent.

This research was inspired by a desire to understand this disengagement in the face

of the systematic dismantling of the democracy that citizens had yearned for during

the long and brutal period of state socialist rule. The answer I found lies in the way

people in Hungary feel about their political system, and the roots of these feelings

in the political information environment. In Chapter Two I built off of research on

how increased international constraint, changing information technologies, and the

post-communist legacy affect parties’ political appeals and the political dialogue to

hypothesize that political debates in Hungary and other post-communist European

164

Union member states are increasingly dominated by populist outrage. I defined pop-

ulist outrage as a type of political appeal accusing a transgressor of committing a

morally depraved act against the good, homogeneous people of the country. The re-

search in social, evolutionary, and political psychology that I reviewed in the theory

chapter led me to hypothesize that these appeals may inspire anger at points, and that

this anger leads to political action and protest voting that helps explain persistent

party system volatility decades after the transition to competitive multiparty politics.

However, building on this same research on the psychology of emotional response, I

argued that many citizens will also react to populist outrage by becoming disgusted

with politics. Given disgust’s evolutionary role in expelling harmful pathogens from

the body and dangerous people or ideas from groups, I hypothesized that disgust

towards politics would lead to a visceral rejection of politics that could account for

many among the Hungarian public turning a blind eye to the dismantling of their

democratic institutions, as citizens disgusted by politics would not see democracy as

worth protecting in the first place.

In Chapter Three I turned to testing the first hypothesis of this dissertation, that

populist outrage has increased in Hungary in large part as a function of increased

European Union (EU) constraint on the ability of domestic political parties to distin-

guish themselves from one another with distinct policy appeals (H1 ). In order to test

this hypothesis, I gathered an original data set of hundreds of articles covering each

political campaign from 1994 until 2014 by the Hungarian News Agency. I developed

two dictionaries of populist outrage, the first constructed inductively from paradig-

matic examples of populist outrage in the recent campaigns, and the second adapted

from the Affective Norms for English Words database constructed by Bradley and

Lang (1999). Both dictionaries document an increased proportion of populist outrage

words in political coverage by the Hungarian News Agency, despite this organization

165

working to provide the least emotional and salacious source of news. Moreover, by

both measures, populist outrage increased sharply in the early 2000s, following the

brunt of EU accession negotiations and constraint. Having shown an increase in news

coverage, I then turned to tracking populist outrage in appeals from political parties

themselves. For this I relied upon the Comparative Manifestos Project data (CMP

- Volkens et al. (2014)) on the political appeals that parties make in their published

platforms at each election. I derived several populist sub-scales and a general pop-

ulism scale from the various CMP coding categories. Once again, these data point to

an increase in populism across East Central Europe that coincides largely with EU

accession and membership.

The next two chapters turned to analyzing the ways Hungarian citizens react to

the rising tide of populist outrage in their political information environment. Chapter

Four draws upon two experiments and an original survey of 1,000 Hungarian citizens

to analyze the immediate emotional reactions to populist outrage. Building off of an

extensive body of research on anger and selective partisan perception, I hypothesized

that instances of populist outrage would anger those citizens with partisan predilec-

tions as they would attribute blame when they saw an accusation (H2a). I argued

that those in the middle would be less likely to attribute blame for a given accusation

to any specific actor and would therefore be more likely to feel disgusted by populist

outrage (H2b). The survey and experimental data largely supported these hypothe-

ses. When exposed to instances of populist outrage, partisans were uniquely moved

to anger, while those in the middle were left disgusted but not angry. However, one

major finding deviated from expectations. Partisans were not only angry, they were

also disgusted.

The co-occurrence of anger and disgust was also a surprising finding of Chapter

Five, in which I turned to analyzing the dynamics of reactions to populist outrage over

166

time and repeated exposure. Drawing upon neurological and evolutionary theories

about the nature and origins of disgust, I hypothesized that disgust would supplant

anger as a response to populist outrage given repeated exposure to the same objects of

outrage (H3a) and as a citizen aged and gained greater exposure to populist outrage

over the course of greater experience with politics (H3b). Drawing upon a series

of focus groups, interviews, and the same survey data introduced in Chapter Four,

I found that disgust does indeed increase as both a function of repeated exposure

to the same examples of populist outrage and with greater experience with politics

over the life span. However, these data suggest that disgust does not replace anger

as the dominant emotional response to populist outrage. Instead, more and more

disgust supplements ever-increasing anger, combining into a virulent anti-politics of

passionate rejection of the political.

In Chapter Six I turned to the political ramifications of anger and disgust towards

politics, drawing upon the experimental, survey, and focus group data to test the

hypotheses that anger is associated with a desire to fix the offending situation H4a,

that this impulse can lead to the vote switching and protest voting that drives party

system volatility (H4b), and that disgusted citizens are less likely to support the

very idea of democracy (H4c). The data largely support these claims. Anger is a

positive predictor of an intention to vote as well as wishing to take action to right

a wrong, though the specific link to protest voting received less empirical validation.

The results show strong evidence that disgust leads to disengagement from politics

in general, and disgusted citizens had lower political efficacy and were less likely to

believe that democracy is a good thing for their country in the first place.

Thus, the general findings of this dissertation from multiple methods and perspec-

tives indicate that many Hungarian citizens, disgusted as they are by the parade of

outlandish and outrageous allegations standing in for political dialogue, have turned

167

away, not only from political participation, but also from the idea that democracy

is a worthy end for Hungary. This is exactly the trend we see growing in democ-

racies around the world, where “to label an activity or process as ‘political’ is, it

seems, invariably to deride and distance oneself from it” (Hay, 2007, p. 5). These

are not “critical citizens” (Norris, 1999) or “distrusting democrats” (Moehler, 2008)

empowered by their skepticism to hold their leaders’ feet to the fire. Instead, this

dissertation uncovers disgusted citizens sickened by their political process and so fed

up with politics that they abandon the very idea of democracy, helping to understand

the ability of the current Hungarian leadership to tighten its iron grip on power as

the majority of the populace willfully turns a blind eye. Along the way, the preceding

chapters raised several noteworthy implications for the study of democracy and public

opinion in the twenty-first century. Indeed, the declaration at the top of this page

was made about the United States, and the trends noted in this dissertation are not

contained to Hungary, East Central Europe, or even new democracies.

7.2 Limitations and Future Research

The work in the preceding pages opens up a research agenda on the implications

of the complex interplay between elite appeal making and citizens’ emotional reac-

tions. While it is my great hope that the reader has found the questions posed and

answers provided enlightening, this research suffers from a number of shortcomings.

Namely, the methods brought to bear on these trends often fall short of conclusive

measurement of both internal emotional dynamics and actual behavior in the real

world. Moreover, it largely focuses upon one case, so questions of generalizability are

both warranted and fair. There are also open questions pertaining to the endogenous

relationship between anger and partisan predilections, the co-occurrence of emotions

originally hypothesized to occur and influence political behavior separately, and the

168

causes and consequences of disgust directed towards specific groups instead of gener-

alized towards the political system as a whole. At this point of reflection upon the

research agenda to date, I detail these present shortcomings as well as directions for

future scholarship in the hope that in considering the caveats and failures of the work

as it stands today I will encourage future illuminating study.

7.2.1 Empirical Refinement

Studying emotions is hard. While scholars in both psychology (Zajonc, 1980;

Lazarus, 1991; Damasio, 2005) and political science (Marcus et al., 2000; Lodge and

Taber, 2013) acknowledge their primacy in directing human cognition and behavior,

measuring the origins, intensity, and consequences of emotional states is a fraught

process. Thus, I suspect that the relative newness of emotions as a major area of

study in political science has more to do with the difficulty of providing empirically

rigorous, externally valid measurement than a general sense that emotions are unim-

portant to political behavior. As with almost all other political psychology studies

of emotion, this dissertation relies mostly upon participants’ self-reporting of their

emotions. This is an obviously weak strategy as it is easy to think of ways that par-

ticipants might fail to identify their true emotional state. The repeated appearance

of the same findings reported in Chapters Four through Six pertaining to the corre-

lates of anger and disgust across multiple methods lends credence to the results, and

supplementation by observation of body language and tone in the qualitative data

adds somewhat more validity. However, when considering the finding that people ex-

perience and are influenced by multiple emotions at once (addressed below) and the

further wrinkle that this measurement took place in Hungarian, there are concerns of

both internal and external validity. Future replication and empirical expansion is no

169

doubt required. Fortunately, disgust and anger have somewhat opposite physiologi-

cal markers and are associated with measurably distinct portions of the brain (Kelly,

2011). Thus, future research should leverage fMRI techniques, observation of facial

expression (Ekman and Friesen, 1971), appetite behavior, and physiological markers

such as skin conductivity, heart rate, and blood pressure to confirm what type of

stimuli elicits which emotion in which participant.

While measuring emotions is itself a very difficult business, matching emotional

states at the point of measurement to actual political behavior in the real world is

harder still. Again, as with the vast majority of research on emotions in politics, and in

political behavior more broadly, the current work makes claims about actual behaviors

such as voting from self-reported responses. Given the likelihood of social desirability

bias and the possibility that emotional states at play in experimental work would wear

off before affecting behavior, there is an obvious problem of external validity. While

the recurrence of similar trends across hypothesis tests and methodology once again

helps lend credence to the claims presented above, future research should work on

ways to measure behavioral implications as they occur. My own attempts to do so in

the lab-in-the-field experiment reported in Chapters Four and Six met resistance from

my university administration where I collaborated as their policies explicitly forbid

researchers from including any political behaviors in studies or distributing political

petitions or campaign materials. Thus, matching emotional states to measures of

political behavior in the real world remains an important avenue of future research.

The most difficult hypothesis to provide empirical leverage on concerns the evo-

lution of emotional response over time. The focus group, interview, and survey data

brought to bear on this issue in Chapter Five represent a starting point for this

process, but the over time development of disgust towards politics is the area that

requires the most immediate replication and supplementation. More qualitative data

170

in other cases and contexts would be extremely useful. However, large scale data that

could stand the test of rigorous statistical analysis would be even better. The gold

standard would be panel data tracking the same participants over long periods of time

and repeatedly measuring how specific political events as well as politics in general

make them feel. These costly data were well beyond my resources in conducting this

dissertation research, but given the rapid uptick in studies of emotions in politics,

it is my hope that more ambitious data collection on emotional reactions to politics

will occur in the near future. If this dissertation can push the agenda towards more

careful and widespread measurement of citizens’ emotional experiences and how this

directs political opinion and behavior, I have surpassed my own expectations (and

likely those of the reader).

Finally, the vast majority of the data in this dissertation focus solely upon Hun-

gary, and the extent to which the findings presented here generalize to other cases

remains an open question. While the Comparative Manifesto Data allowed for an

analysis of populist appeals in party platforms across three other post-communist

democracies, the dearth of existing data on emotional reactions to politics necessi-

tated reliance upon original data collection for the other hypothesis tests. As will be

discussed more fully below, the trends hypothesized to fuel an increase in populist

outrage are global, and skimming recent scholarship or just the mainstream news

on any given day provides some surface validity for the claim that populist outrage

and political disgust are on the rise in diverse cases around the world. However, it

is my hope that future data collection will expand the focus of this research to new

cases and scenarios, adding to our understanding of acrimonious politics and their

implications for political behavior and democratic governance.

171

7.2.2 Of Time, Outrage, and Partisan Predilections

One of the key findings of this research is that citizens’ partisan predispositions

moderate their emotional reaction to populist outrage. Partisans see either the ac-

cuser or the accused as blameworthy and thus feel anger, while those in the middle

are more likely to react with disgust and thus become more disengaged with politics.

This finding is in line with previous work on the role partisanship plays in influencing

subsequent political opinion (Campbell et al., 1960; Westen et al., 2006; Lodge and

Taber, 2013). Additionally, the role that partisan predilections play in determining

whether a citizen reacts to populist outrage with anger or disgust in these studies

mirrors a trend observed in democracies old and young: while many citizens are able

to stay permanently enraged by their political opponents, many others in the middle

fall into a vast chasm of disengagement. As Ansolabehere and Iyengar (2005) observe,

voters are increasingly divided into the unaligned who do not participate and the par-

tisan loyalists who do. Those with a dog in the fight become ever more enraged while

those in the middle become more and more disgusted and disillusioned with the very

process of politics, removing exactly the moderates that an angry, polarized political

environment requires most.

The present findings regarding partisanship’s moderating role between anger and

disgust also aligns with work by Ceka (2012) on post-communist public opinion. He

shows that there is a surprising correlation between higher levels of robust political

competition among parties and lower satisfaction with democracy among Eastern Eu-

ropeans. However, this finding only holds for citizens who lack partisan predilections.

I would argue that this is due to these unaligned citizens becoming more and more

disgusted with the very process of robust political competition, as populist outrage

is likely a hallmark of robustness (measured by Ceka as closeness of elections) in the

172

contexts of contemporary ECE. Moreover, while Ceka concludes that as party sys-

tems consolidate and partisanship becomes more widespread the problem of robust

political competition weakening support for democracy will fade away, there is no

evidence that levels of partisanship are on the rise in ECE or that party systems are

consolidating across the region. On the contrary, as populist outrage persists and

pushes more citizens towards political disgust, the problem of democratic support for

democracy may indeed continue to grow in breadth and severity.

The key role partisan predilections play in evoking anger also highlights the im-

portance of building on work analyzing the origins of partisanship in ECE and other

new democracies. This begs a prior question, however: where does partisanship

come from? What is different about the minority of voters who become affiliated

with a party in the early years of party competition, thus spurring them to be more

likely to react with anger and action than disgust and disinterest? How partisan-

ship emerges remains one of the greatest unanswered questions in political behavior,

and new regimes such as those in post-communist Europe offer a rare opportunity to

track the advent of partisanship in the contexts of the earliest stages of multi-party

political competition. An important caveat, however, is that partisanship may sim-

ply not arise for many within post-communist Europe and other new regimes. The

processes of dealignment has been well documented even in the older democracies

in Western Europe (Dalton and Wattenberg, 2000), and we may therefore expect a

lack of alignment in the first place. This still begs the question of what is different

about that minority of citizens that did develop a degree of self-reported partisan-

ship. Unfortunately this dissertation has little to say about this particular issue, but

its findings would suggest that anger will both increase partisanship and come from

it, while disgust will propel people away from party identification. Future research

would do well to take on the daunting task of wrestling with this endogeneity.

173

A related implication concerns the role of parties more generally. Mair (2013)

points to the passing of party democracy as the main cause of the withdrawal of voters

and elites at the root of the problems of decreasing participation on the part of citizens

and increasing populism, polarization, and extremism. Interestingly, Ansolabehere

and Iyengar (2005) come from a completely different angle but also argue for the

importance of party organizations in fixing the problems contemporary democratic

governance. In their study of the United States, they argue that the prominence of

negative advertising is in large part due to the lack of strong party organizations

that mobilize people to participate. Instead of parties working to sell a ticket as a

whole, the weakness of American parties leaves individual candidates on their own

to win races in the best way they see fit. They argue that this may often mean

driving up an opponents’ “negatives,” especially for candidates who benefit from

suppressing turnout in a race. Thus, Ansolabehere and Iyengar’s solution to the

problems of negative campaigning in the US is exactly what Mair argues has gone

missing in Europe: stronger party organizations. My own analysis comports these

findings: the weakness of parties in ECE has acted as a strong incentive to rely

upon negative, outrage-based appeals, and this trend is exacerbated by the media.

The rise of populist outrage purveyor extraordinaire Donald Trump to the top of the

Republican ballot in 2016 to the dismay of the GOP leadership lends further credence

to these claims, and future research stands to contribute a great deal in examining the

link between party organizational strength and the nature of political appeal making.

7.2.3 The Co-occurrence of Disgust and Anger

One recurring unexpected finding throughout this dissertation is that anger and

disgust co-occur with great frequency. This was true in two respects. First, in test-

ing the role of partisan predilections in moderating the relationship between populist

174

outrage and emotional response, I found that partisans were not only angry, they also

registered signs of disgust. Thus, while partisans were unique for registering anger at

populist outrage, they reported levels of disgust equal to and sometimes surpassing

those of the unaligned. Second, while the younger focus groups demonstrated unadul-

terated anger towards politics, older groups displayed disgust combined with an even

greater measure of anger. While at the outset this research was most concerned with

the distinction between anger and disgust, their co-occurrence is an equally important

topic for study. As addressed above, future research would do very well to theorize

how these emotions interact with one another.

In my view, the most pressing hypothesis to investigate regarding the combination

of anger and disgust felt towards the same object at the same time is when this object

is a social group. The disgust reported by partisans in Chapter Four is likely directed

towards the other political side. This is logical given the proposed mechanism for

the development of disgust: repeated association with irksome, unpleasant immoral

actions. Fidesz partisans have had great cause to associate MSZP with disgusting

immoral acts and vice versa, as is the case between Democrats and Republicans in the

United States and between political camps in virtually any political system. That the

other side of politics can elicit disgust is a particularly treacherous possibility. Recent

work by Iyengar et al. (2012) on affective polarization, or intense disliking of the other

side that they show to be higher than any other type of social discrimination, is likely

driven by this same combination of anger and disgust.

That already marginalized social groups may become the objects of both anger

and disgust is even more troubling. Appeals to populist outrage are often directed

towards political opponents or the political establishment in general, but they are

also often targeted towards distinct social groups. For instance, in Hungary stories

175

of “gypsy crime and terror”1 are extremely common in the media and in the polit-

ical dialogue of the right. These stories frame the Roma minority as taking great

resources from society in the form of social welfare programs and human rights legis-

lation while attacking and stealing from hard-working citizens who lack such special

privileges. This narrative resonates with large portions of the Hungarian population.

One recent study shows increasingly prevalent anti-Roma sentiment among the Hun-

garian public, and that these attitudes account for a great deal of the support for

the radical-right, anti-democratic Jobbik party (Karacsony and Rona, 2010). More-

over, widespread anti-Semitic rhetoric in Hungary is likely also associated with rising

anti-Semitic prejudice. A 2012 report by the American Anti-Defamation League an-

alyzing attitudes towards Jews in Europe shows Hungary to have both the highest

absolute level of popular anti-Semitic attitudes in Europe as well as the greatest

negative change over recent years.2 A vast Israeli-American conspiracy against the

wholesome, victimized Hungarian nation is an age-old trope of the political right in

Hungary, and Jobbik politicians have even gone so far as to suggest making lists of

all Jewish Hungarian Members of Parliament for the purposes of national security.3

Disgust may well at the root of these attitudes, and populist outrage is likely

a cause of increasing disgust. This is enormously concerning given studies showing

disgust towards other humans is the first step of the dehumanization central to the

darkest chapters of human history (Allport, 1954; Harris and Fiske, 2006). Disgust’s

combination with anger in this circumstance constitutes a worst case scenario. In-

deed, those who are angry at a group may take action to punish that group, and if

1cigny b?nzs

2http://www.adl.org/Anti_semitism/adl_anti-semitism_presentation_february_2012.

pdf

3http://www.euronews.com/2012/12/03/outrage-in-hungary-over-jewish-register-

remark/

176

they also feel disgusted towards that group, their dehumanization of group members

may facilitate particularly dangerous aggression. As Fischer and Roseman note, the

combination of anger and disgust/contempt is dangerous “as it implies that the in-

clination to attack co-occurs with the inclination to derogate and exclude a person

who is seen as inferior or bad, which may result in hostile acts that are not held in

check by affection or social relationships” (Fischer and Roseman, 2007, 113). My

most immediate future empirical task will therefore be to turn to ascertaining the

sources of disgust and anger towards vulnerable social groups in an effort to draw

attention to this dangerous trend and work to prevent it.

7.3 Implications for Scholars and Policymakers

Despite the shortcomings and extensions noted above, this research points to a

number of important implications for both scholars and policymakers. First, studies

of emotions in politics primarily focus upon the United States, and by examining

these trends in Hungary, this work adds to the growing body of work on anger in

politics and probes how findings from the US travel to different circumstances. As

emotions are based upon deep foundations in the human psyche, we should expect

them to be largely constant across country and culture. Even so, specific patterns of

elicitation and expression will likely be contingent upon unique political and cultural

factors, and the United States is in many ways exceptional. It is an enormous country

with the world’s oldest democratic institutions and no recent history of warfare within

its boundaries. It is a uniquely multicultural country with a uniquely complicated

relationship with race. It has a remarkably stable two-party system and an electorate

that seems both to turn out less at the polls and know less about politics than com-

parable cases. These are just some of the factors that could make our understanding

of the way emotions affect politics somewhat contingent upon the American context.

177

Extending the study of emotion in politics to other countries informs the study of

emotions in politics by providing new contexts in which to replicate and extend the

field’s findings from studies focused on the US, clearing up empirical debates and

shedding light on inconsistencies.

Another key contribution is the focus on the interaction between politics and a

new emotion. Political psychology focuses mainly upon a limited number of emotions,

and while the Affective Intelligence paradigm (Marcus et al., 2000) in particular has

been crucial in encouraging a fascinating research agenda, it is clear that there is

much to explore when it comes to the study of emotions in politics (Brader, 2006).

By examining disgust, an emotion that is rarely implicated in political science,4 this

research provides a fuller account of the way various emotions interact in political

perception and behavior. Part of an ongoing trend in the study of emotions in politics

is bringing more emotions and more nuance into the fold. Just as Huddy et al.

(2005) disaggregated aversion into anger and anxiety and thereby increased our ability

understand political outcomes, further distinguishing between anger and disgust will

shed light on dual trends of political volatility and disaffection plaguing democracies

around the world.

This work also connects to the vast scholarly literature on the effects of political at-

tacks. To date this body of research reveals contradictory findings (Lau et al., 2007),

as in some circumstances they turn into political action (Montada and Schneider,

1989; Lodewijkx et al., 2008; Mark, 2009) whereas in others they turn into with-

drawal and abstention (Vandenbroek, 2012; Kamber, 1997). The distinction between

disgust and anger offers a way forward on this debate: where political attacks anger

they increase participation and where they disgust they decrease it. A related ques-

tion concerns participation in post-communist Europe in particular. While scholars

4As of this writing, Vandenbroek (2012)’s dissertation is the only work in political science ofwhich I am aware

178

explain waning participation with disappointment at corruption (Kobach, 2001) and

a lack of perceived fairness (Mason, 2003), actual empirical evidence has been thin.

In a comprehensive analysis of turnout in the region, Pacek et al. (2009) show that

political disillusionment can either increase or decrease the propensity to vote, de-

pending on how it is measured and modeled. The distinct attitudinal and behavioral

effects of anger and disgust account for these contradictory findings: discontent in-

spires different emotions in different people. Future work improving measurement and

explicitly theorizing about how these emotions come about individually and together

will continue to shed light on political participation across diverse contexts.

Finally, the evidence presented here shows how the structure of supranational gov-

ernance forces elites’ hands to utilize a strategy that leads to pernicious consequences

for the quality of democracy. By fleshing out this important causal mechanism at the

heart of the democratic deficit, this research draws attention to the imperative of de-

signing international institutions such that national political actors are able to differ-

entiate themselves on the policy issues most important to voters. Indeed, scholarship

shows that fiscal restraint drives party system volatility (Nooruddin and Chhibber,

2008), and prominent scholars have pointed to the role of neoliberal global institutions

and technocratic leadership in creating a political void (Hay, 2007; Mair, 2013). The

present work explains what moves in to fill this void: the politics of outrage. These

trends are global, but especially important to understand in the European Union.

The EU has contributed to the rise of populist outrage both in creating a wellspring

of public resentment against a distant, hostile enemy in Brussels (Meny and Surel,

2002), and by binding the hands of domestic policymakers and making them search

for new means of differentiation and competition beyond programmatic distinctions

(Nanou and Dorussen, 2013). The rising tide of popular ire directed at the Euro-

pean Union from Fidesz and Jobbik in Hungary is matched by populist-nationalist

179

parties across Europe, from the France’s Front National to the United Kingdom’s In-

dependence Party. The present work thus constitutes one more voice calling for new

thinking by policymakers on how to either make EU level political competition more

meaningful or how to re-invigorate the power of domestic political policymakers.

Such new thinking is all the more important given how high the stakes are when

it comes to keeping the political system from becoming an object of disgust. Because

the evolutionary purpose of disgust is to guard the body against pathogens in a world

full of potentially fatal contaminants, this emotional response is endowed with a hair-

trigger that is prone to false positives and difficult to undo (Kelly, 2011). In other

words, it is easy for something to become an object of disgust, but nearly impossible

for it to become clean again, an aspect of the emotional response that is just as true for

moral disgust as it is for its biological counterpart (Chapman et al., 2009). Indeed,

disgust leaves a more lasting impression in the memory than even fear (Chapman

et al., 2013). Thus, for those who come to feel disgust towards democratic politics, it

will be difficult to turn them back toward the idea of democracy as a worthy ideal or of

politics as a realm in which to address pressing social concerns. There is evidence that

political scandals cause many citizens to feel that the disgraced politicians accused

of outrageous transgressions are the tip of the iceberg of a system that is rotten to

the core, not exceptions (Sabato, 1991; Patterson, 2009). Keeping specific instances

of outrage from generalizing into a dialogue steeped in outrage is critical, as when

it comes to political disgust, the old adage that an ounce of prevention is worth a

pound of treatment is doubly true.

On a related note, however, anger and even populist outrage have their useful

places in political behavior. Examining the politics of outrage and the transforma-

tion of anger into disgust provides leverage with which to explore the contradictory

180

nature of moral anger when it comes to politics; in some situations it elicits the partic-

ipation so critical to accountability and in others spurs the deep disengagement that

threatens the viability of democratic governance. The angry will turn disappointment

into action, while the disgusted reject the entire system. While in this dissertation,

populist outrage and the associated anger it inspires has been viewed as a harmful

development because of the associated disgust it elicits, many studies actually view

moral anger as a pro-social emotion. Indeed, it is not always the case that negative

campaigning and anger are a bad thing when it comes to politics, as they can inspire

pro-social political action, ranging from voting to getting involved in a protest for a

worthy cause (Marcus et al., 2000; Brader, 2006; Mark, 2009; Valentino et al., 2011).

Some outrageous instances in politics really do require fixing, and anger makes that

possible. Thus, one key consideration is how to reap the participatory virtues of moral

anger without it curdling into disgust. It strikes me that the best way to do this is to

save the outrageous rhetoric for the truly serious transgressions, and to somehow level

serious penalties against actors in the media and political arena who scream outrage

when it is not merited. How to do this in an increasingly fragmented, competitive,

pervasive media environment remains elusive. This constitutes a tragedy of the com-

mons problem in which an individual politician or media source may benefit in the

short-term from riding a tide of outrage and indignation to victory or popularity,

but as this tactic metastasizes throughout the body politic, citizens exit the political

system and turn off the news. The bottom line is that we must always keep disgust

in the back of our mind when studying the effects of anger in politics.

Wrestling with these questions is an important task for scholars and policymakers

investing in deepening and protecting democratic governance. At the beginning of

this dissertation, I quoted the most memorable moment of my fieldwork, when my

interviewee promised to scrawl either the profane word for the male or female anatomy

181

across his ballot. At the end of our conversation, this most quotable of interviewees

wished me luck in my study of the election and warned, “You’re at the circus now” in

an amused but concerned grin. The underlying sentiment was quite common. Many

Hungarians I spoke to felt that as an outsider I could not hope to comprehend how

outlandish Hungarian politics were. I suspect, however, that this sincere contention

is false. On the contrary, events in Hungary are all too common. As outrage-fueled

political movements become increasingly prevalent around the world, disgust is a vital

variable to account for in order to understand the reasons for growing anti-political

sentiment and democratic malaise documented around the world (Mair, 2013). From

the current assault on democracy in Poland to Trump’s rise to GOP nominee to the

election of Duterte as Filipino president, the headlines of the week of this writing, like

almost any week these days, provide ample reason to worry that populist outrage and

a popular disdain for the participatory norms of democracy are global problems.

182

Bibliography

Adriaansen, M. L., Van Praag, P., and De Vreese, C. H. (2010). Substance matters:How news content can reduce political cynicism. International Journal of PublicOpinion Research, pages 1–25.

Allport, G. W. (1954). The Nature of Prejudice. Addison-Wesley, Reading, MA.

Almond, G. A. and Verba, S. (1963). The Civic Culture: Political Attitudes andDemocracy in Five Nations. Princeton University Press.

Ansolabehere, S. and Iyengar, S. (2005). Going Negative. Free Press, New York.

Ansolabehere, S., Iyengar, S., Simon, A., and Valentino, N. (1994). Does attack adver-tising demobilize the electorate? American Political Science Review, 88(4):829–838.

Bakke, E. and Sitter, N. (2005). Patterns of stability party competition and strategyin central europe since 1989. Party Politics, 11(2):243–263.

Batson, C. D., Kennedy, C. L., Nord, L.-A., Stocks, E., Fleming, D. A., Marzette,C. M., Lishner, D. A., Hayes, R. E., Kolchinsky, L. M., and Zerger, T. (2007).Anger at unfairness: is it moral outrage? European Journal of Social Psychology,37(6):1272–1285.

Bermeo, N. (2016). On democratic backsliding. Journal of Democracy, 27(1):5–19.

Berry, J. M. and Sobieraj, S. (2013). The Outrage Industry: Political Opinion Mediaand the New Incivility. Oxford University Press, New York, NY.

Bertoa, C. and Mair, P. (2010). Two decades on: How institutionalized are thepost-communist party systems? EUI Woking Papers, SPS/2010, 3.

Best, S. J. and Krueger, B. S. (2011). Government monitoring and political partic-ipation in the united states: The distinct roles of anger and anxiety. AmericanPolitics Research, 39(1):85–117.

Blom, R., Carpenter, S., Bowe, B. J., and Lange, R. (2014). Frequent contributorswithin us newspaper comment forums an examination of their civility and infor-mation value. American Behavioral Scientist, 58(10):1314–1328.

183

Boukes, M. and Boomgaarden, H. G. (2015). Soft news with hard consequences? in-troducing a nuanced measure of soft versus hard news exposure and its relationshipwith political cynicism. Communication Research, 42(5):701–731.

Brader, T. (2005). Striking a responsive chord: How political ads motivate andpersuade voters by appealing to emotions. American Journal of Political Science,49(2):388–405.

Brader, T. (2006). Campaigning for hearts and minds: How emotional appeals inpolitical ads work. University of Chicago Press, Chicago, IL.

Bradley, M. M. and Lang, P. J. (1999). Affective norms for english words (anew):Instruction manual and affective ratings. Technical report, Citeseer.

Bunce, V. (2010). The political transition. In Wolchik, S. L. and Curry, J. L., editors,Central and East European politics: From communism to democracy, pages 43–66.Rowman & Littlefield, London, UK.

Campbell, A., Converse, P. E., Miller, W. E., and Stokes, D. E. (1960). The Americanvoter. University of Chicago Press, Chicago, IL.

Canovan, M. (2002). Taking politics to the people: populism as the ideology ofdemocracy. Democracies and the populist challenge, pages 25–44.

Cappella, J. N. and Jamieson, K. H. (1997). Spiral of cynicism: The press and thepublic good. Oxford University Press, New York and London.

Carey, J. and Reynolds, A. (2007). Parties and accountable government in newdemocracies. Party Politics, 13(2):255–274.

Carnahan, D. (2012). A motivational approach towards understanding the conse-quences of selective exposure. In APSA 2012 Annual Meeting Paper.

Carver, C. S. and Harmon-Jones, E. (2009). Anger is an approach-related affect:evidence and implications. Psychological Bulletin, 135(2):183–204.

Catterberg, G. and Moreno, A. (2006). The individual bases of political trust: Trendsin new and established democracies. International Journal of Public Opinion Re-search, 18(1):31–48.

Ceka, B. (2012). The perils of political competition: Explaining participationand trust in political parties in eastern europe. Comparative Political Studies,46(12):1610–1635.

Chapman, H. A. and Anderson, A. K. (2011). Varieties of moral emotional experience.Emotion Review, 3(3):255–257.

184

Chapman, H. A., Johannes, K., Poppenk, J. L., Moscovitch, M., and Anderson, A. K.(2013). Evidence for the differential salience of disgust and fear in episodic memory.Journal of Experimental Psychology, 142(4):1100–1112.

Chapman, H. A., Kim, D. A., Susskind, J. M., and Anderson, A. K. (2009). In badtaste: Evidence for the oral origins of moral disgust. Science, 323(5918):1222–1226.

Chhibber, P. and Kollman, K. (2009). The formation of national party systems:Federalism and party competition in Canada, Great Britain, India, and the UnitedStates. Princeton University Press, Princeton, NJ.

Chiru, M. and Gherghina, S. (2012). When voter loyalty fails: party performance andcorruption in bulgaria and romania. European Political Science Review, 4(1):29–49.

Chmiel, A., Sobkowicz, P., Sienkiewicz, J., Paltoglou, G., Buckley, K., Thelwall, M.,and Ho lyst, J. A. (2011). Negative emotions boost user activity at bbc forum.Physica A: Statistical Mechanics and its Applications, 390(16):2936–2944.

Coppedge, M. (2005). Explaining democratic deterioration in venezuela throughnested inference. In Hagopian, F. and Mainwaring, S. P., editors, The third wave ofdemocratization in Latin America: Advances and Setbacks, pages 289–318. Cam-bridge University Press, Cambridege.

Cottrell, C. A. and Neuberg, S. L. (2005). Different emotional reactions to differ-ent groups: a sociofunctional threat-based approach to” prejudice”. Journal ofPersonality and Social Psychology, 88(5):770.

Crabtree, C. and Golder, M. (forthcoming). Party system volatility in post-communisteurope. British Journal of Political Science.

Dalton, R. J. and Wattenberg, M. P. (2000). Parties without partisans: politicalchange in advanced industrial democracies. Oxford University Press, Oxford, UK.

Damasio, A. (2005). Descartes’ error: Emotion, reason, and the human brain. Pen-guin Books.

Dassonneville, R., Blais, A., and Dejaeghere, Y. (2015). Staying with the party,switching or exiting? a comparative analysis of determinants of party switchingand abstaining. Journal of Elections, Public Opinion & Parties, pages 1–19.

Dassonneville, R. and Hooghe, M. (2015). Economic indicators and electoral volatility:economic effects on electoral volatility in western europe, 1950–2013. ComparativeEuropean Politics, 9.

De Tocqueville, A. (2004). Democracy in America, volume 147. Digireads. com.

Diamond, L. J., Linz, J. J., Lipset, S. M., and Samudavanija, C.-A. (1989). Democracyin developing countries, volume 4. Lynne Rienner Boulder, CO.

185

Dunn, E. C. (2004). Privatizing Poland: Baby food, Big Business, and the Remakingof Labor. Cornell University Press, Ithaca, NY.

Duverger, M. (1954). Political Parties: Their Organization and Activities in theModern State. Methuen, London.

Easton, D. (1965). A Framework for Political Analysis, volume 25. Prentice-HallEnglewood Cliffs, NJ.

Ekman, P. and Friesen, W. V. (1971). Constants across cultures in the face andemotion. Journal of personality and social psychology, 17(2):124.

Enyedi, Z. and Toka, G. (2007). The only game in town: Party politics in hungary.In Webb, P. and White, S., editors, Party Politics in New Democracies, pages147–178. Oxford University Press, Oxford, UK.

Evans, G. and Whitefield, S. (1995). The politics and economics of democratic com-mitment: Support for democracy in transition societies. British Journal of PoliticalScience, 25(04):485–514.

Fan, R., Zhao, J., Chen, Y., and Xu, K. (2014). Anger is more influential than joy:Sentiment correlation in weibo. PloS one, 9(10):e110184.

Farah, B. G., Barnes, S. H., and Heunks, F. (1979). Political disaffection. In Barnes,S. H., editor, Political Action: Mass Participation in Five Western Democracies.Sage, Beverly Hills, CA.

Finkel, S. E., Muller, E. N., and Seligson, M. A. (1989). Economic crisis, incumbentperformance and regime support: A comparison of longitudinal data from westgermany and costa rica. British Journal of Political Science, 19(3):329–351.

Fischer, A. H. and Roseman, I. J. (2007). Beat them or ban them: the characteristicsand social functions of anger and contempt. Journal of Personality and SocialPsychology, 93(1):103.

Fischhoff, B., Gonzalez, R. M., Lerner, J. S., and Small, D. A. (2005). Evolving judg-ments of terror risks: foresight, hindsight, and emotion. Journal of ExperimentalPsychology: Applied, 11(2):124.

Follesdal, A. and Hix, S. (2006). Why there is a democratic deficit in the eu: Aresponse to majone and moravcsik. JCMS: Journal of Common Market Studies,44(3):533–562.

Geer, J. G. (2012). The news media and the rise of negativity in presidential cam-paigns. PS: Political Science & Politics, 45(3):422–427.

186

Gerber, A. S., Green, D. P., and Shachar, R. (2003). Voting may be habit-forming:evidence from a randomized field experiment. American Journal of Political Sci-ence, 47(3):540–550.

Gerner, K. (2007). Open wounds? trianon, the holocaust and the hungarian trauma.In Mithander, C., Sundholm, J., and Holmgren, M., editors, Collective Traumas:Memories of War and Conflict in Twentieth-Century Europe, pages 79–109. P.I.E.Peter Lang, Brussels, Belgium.

Giner-Sorolla, R. (2012). Judging passions. Psychology Press, New York, NY.

Glynn, C. J., Herbst, S., O’Keefe, G. J., Shapiro, R. Y., and Lindeman, M. (1999).Public Opinion. Westview Press Boulder, CO.

Graham, J., Haidt, J., Nosek, B. A., et al. (2009). Liberals and conservatives rely ondifferent sets of moral foundations. Journal of personality and social psychology,96(5):1029–1046.

Grzymala-Busse, A. and Innes, A. (2003). Great expectations: The eu and domesticpolitical competition in east central europe. East European Politics and Societies,17(1):64–73.

Gunther, R. and Montero, J. R. (2006). The multidimensionality of political supportfor new democracies: Conceptual redefinition and empirical refinement. In Torcal,M. and Montero, J. R., editors, Political Disaffection in Contemporary Democra-cies: Social Capital, Institutions and Politics, pages 46–78. Routledge, New York.

Gutierrez, R. and Giner-Sorolla, R. (2007). Anger, disgust, and presumption of harmas reactions to taboo-breaking behaviors. Emotion, 7(4):853.

Haggard, S. and Kaufman, R. R. (1995). The political economy of democratic transi-tions. Princeton University Press.

Haidt, J. (2003). The moral emotions. In Scherer, K. and Goldsmith, H. H., editors,Handbook of affective sciences, pages 852–870. Oxford University Press, Oxford,UK.

Haidt, J., Rozin, P., McCauley, C., and Imada, S. (1997). Body, psyche, and culture:The relationship between disgust and morality. Psychology & Developing Societies,9(1):107–131.

Harris, L. T. and Fiske, S. T. (2006). Dehumanizing the lowest of the low: Neu-roimaging responses to extreme out-groups. Psychological Science, 17(10):847–853.

Hay, C. (2007). Why we hate politics. Polity Press, Cambridge, UK.

Hellman, J. (1998). Winners take all. World politics, 50(2):203–234.

187

Herman, L. E. (2015). Re-evaluating the post-communist success story: Party eliteloyalty, citizen mobilization and the erosion of hungarian democracy. EuropeanPolitical Science Review, pages 1–34.

Howard, M. M. (2003). The weakness of civil society in post-communist Europe.Cambridge University Press, Cambridge, UK.

Huddy, L., Feldman, S., Taber, C., and Lahav, G. (2005). Threat, anxiety, andsupport of antiterrorism policies. American Journal of Political Science, 49(3):593–608.

Hume, D. (2007). A treatise of human nature. In Norton, D. F. and Norton, M. J.,editors, A treatise of human nature: A critical edition. Clarendon Press, Oxford,UK.

Huntington, S. P. (1968). Political Order in Changing Societies. Yale UniversityPress.

Hutcherson, C. A. and Gross, J. J. (2011). The moral emotions: A social–functionalistaccount of anger, disgust, and contempt. Journal of personality and social psychol-ogy, 100(4):719.

Imbir, K. K. (2015). Affective norms for 1,586 polish words (anpw): Duality-of-mindapproach. Behavior research methods, 47(3):860–870.

Innes, A. (2002). Party competition in postcommunist europe: The great electorallottery. Comparative Politics, 35(1):85–104.

Ito, T. A., Larsen, J. T., Smith, N. K., and Cacioppo, J. T. (1998). Negative infor-mation weighs more heavily on the brain: the negativity bias in evaluative catego-rizations. Journal of personality and social psychology, 75(4):887.

Iyengar, S., Sood, G., and Lelkes, Y. (2012). Affect, not ideology a social identityperspective on polarization. Public Opinion Quarterly, page nfs038.

Jabbi, M., Bastiaansen, J., and Keysers, C. (2008). A common anterior insula repre-sentation of disgust observation, experience and imagination shows divergent func-tional connectivity pathways. PLoS ONE, 3(8):2939.

Jansen, H. J. and Koop, R. (2006). Pundits, ideologues, and the ranters: The britishcolumbia election online. Canadian Journal of Communication, 30(4).

Jarvis, W. B. G. and Petty, R. E. (1996). The need to evaluate. Journal of Personalityand Social Psychology, 70(1):172–194.

Jowitt, K. (1992). New world disorder: The Leninist extinction. University of Cali-fornia Press, Berekely, CA.

188

Kamber, V. (1997). Poison politics: are negative campaigns destroying democracy?Insight Books, Plenum Press.

Kapstein, E. B. and Converse, N. (2008). The Fate of Young Democracies. CambridgeUniversity Press Cambridge.

Karacsony, G. and Rona, D. (2010). A jobbik titka (jobbik’s secret). A szelsojobb magyarorszagi megerosodesenek lehetseges okairol. Politikatudomanyi Szemle,19(1):31–63.

Katz, R. S. and Mair, P. (2009). The cartel party thesis: A restatement. Perspectiveson politics, 7(04):753–766.

Kelly, D. R. (2011). Yuck!: the nature and moral significance of disgust. The MITPress, Cambridge, MA.

Kenesei, I. (2004). A Nyelv es Nyelvek. Akademiai Kiado, Budapest, Hungary.

Kirchheimer, O. (1966). The transformation of the western european party systems.In LaPalombara, J. and Weiner, M., editors, Political Parties and Political Devel-opment, pages 177–200. Princeton University Press, Princeton, NJ.

Klingemann, H.-D. (1999). Mapping political support in the 1990s: A global analysis.In Norris, P., editor, Critical Citiznes: Global Support for Democratic Governance.Oxford University Press, Oxford.

Kobach, K. (2001). Lessons learned in the participation game. In Auer, A. and Butzer,M., editors, Direct Democracy: The Eastern and Central European Experience,pages 292–309. Ashgate Publishing, Burlington, VT.

Kornberg, A. and Clarke, H. D. (1992). Citizens and Community: Political Supportin a Representative Democracy. Cambridge University Press, Cambridge.

Kostadinova, T. (2012). Political corruption in Eastern Europe: Politics After Com-munism. Lynne Rienner Publishers Boulder, CO, Boulder, CO.

Krosnick, J. A. and McGraw, K. M. (2002). Psychological political science versuspolitical psychology true to its name: A plea for balance. Political psychology,pages 79–94.

Lau, R. R. and Rovner, I. B. (2009). Negative campaigning. Annual Review ofPolitical Science, 12:285–306.

Lau, R. R., Sigelman, L., and Rovner, I. B. (2007). The effects of negative politicalcampaigns: a meta-analytic reassessment. Journal of Politics, 69(4):1176–1209.

Lazarus, R. S. (1991). Progress on a cognitive-motivational-relational theory of emo-tion. American Psychologist, 46(8):819.

189

Lerner, J. S., Gonzalez, R. M., Small, D. A., and Fischhoff, B. (2003). Effects of fearand anger on perceived risks of terrorism a national field experiment. Psychologicalscience, 14(2):144–150.

Lerner, J. S. and Keltner, D. (2000). Beyond valence: Toward a model of emotion-specific influences on judgement and choice. Cognition & Emotion, 14(4):473–493.

Lerner, J. S. and Tiedens, L. Z. (2006). Portrait of the angry decision maker: Howappraisal tendencies shape anger’s influence on cognition. Journal of BehavioralDecision Making, 19(2):115–137.

Lewis-Beck, M. S. (1988). Economics and the american voter: Past, present, future.Political Behavior, 10(1):5–21.

Lodewijkx, H. F., Kersten, G. L., and Van Zomeren, M. (2008). Dual pathways to en-gage in silent marches against violence: moral outrage, moral cleansing and modesof identification. Journal of Community & Applied Social Psychology, 18(3):153–167.

Lodge, M. and Taber, C. S. (2005). The automaticity of affect for political leaders,groups, and issues: An experimental test of the hot cognition hypothesis. PoliticalPsychology, 26(3):455–482.

Lodge, M. and Taber, C. S. (2013). The rationalizing voter. Cambridge UniversityPress, Cambridge, UK.

Lowe, W. (2015). Yoshikoder: Cross-platform multilingual content analysis. javasoftware version 0.6.5.

MacKuen, M., Wolak, J., Keele, L., and Marcus, G. E. (2010). Civic engagements:Resolute partisanship or reflective deliberation? American Journal of PoliticalScience, 54(2):440–458.

Mainwaring, S. (1998). Party systems in the third wave. Journal of Democracy,9(3):67–81.

Mair, P. (1996). What is different about post-communist party systems? In Stud-ies in Public Policy No. 259, University of Strathclyde Centre for Public Policy.University of Strathclyde Press, Glasgow.

Mair, P. (2002). Populist democracy vs party democracy. In Meny, Y. and Surel, Y.,editors, Democracies and the populist challenge, pages 81–98. Palgrave Macmillan,London.

Mair, P. (2013). Ruling the void: The hollowing of Western democracy. Verso Books,London.

190

Marcus, G. E., Neuman, W. R., and MacKuen, M. (2000). Affective intelligence andpolitical judgment. University of Chicago Press, Chicago, IL.

Mark, D. (2009). Going dirty: The art of negative campaigning. Rowman & Littlefield.

Mason, D. S. (2003). Fairness matters: Equity and the transition to democracy.World Policy Journal, 20(4):48–56.

McDermott, R. (2010). Emotional manipulation of political identity. In LeChaminant,W., editor, Manipulating Democracy: Democratic Theory, Political Psychology,and Mass Media, pages 113–134. Routledge, London.

Meny, Y. and Surel, Y. (2002). The constitutive ambiguity of populism. In Meny, Y.and Surel, Y., editors, Democracies and the populist challenge, pages 1–21. PalgraveMacmillan, London.

Mickiewicz, E. (2008). Television, Power, and the Public in Russia. CambridgeUniversity Press.

Miller, W. E. and Shanks, J. M. (1996). The new American voter. Harvard UniversityPress, Cambridge, MA.

Miller, W. I. (1998). The Anatomy of Disgust. Harvard University Press, Cambridge,MA.

Moehler, D. C. (2008). Distrusting democrats: Outcomes of participatory constitutionmaking. University of Michigan Press, Ann Arbor, MI.

Moehler, D. C. (2009). Critical citizens and submissive subjects: Election losers andwinners in africa. British Journal of Political Science, 39(02):345–366.

Moll, J., de Oliveira-Souza, R., Moll, F. T., Ignacio, F. A., Bramati, I. E., Caparelli-Daquer, E. M., and Eslinger, P. J. (2005). The moral affiliations of disgust: Afunctional mri study. Cognitive and behavioral neurology, 18(1):68–78.

Montada, L. and Schneider, A. (1989). Justice and emotional reactions to the disad-vantaged. Social Justice Research, 3(4):313–344.

Montero, J. R., Gunther, R., and Torcal, M. (1997). Democracy in spain: Legitimacy,discontent, and disaffection. Studies in Comparative International Development,32(3):124–160.

Morgan, D. L. (1996). Focus groups. Annual review of sociology, pages 129–152.

Morgan, J. (2011). Bankrupt Representation and Party System Collapse. Penn StatePress.

191

Mudde, C. (2001). In the name of the peasantry, the proletariat, and the people:Populisms in eastern europe. East European Politics and Societies, 15(10):33–53.

Mudde, C. (2007). Populist radical right parties in Europe. Cambridge UniversityPress Cambridge, New York, NY.

Mutz, D. C. (2007). Effects of in-your-face television discourse on perceptions of alegitimate opposition. American Political Science Review, 101(04):621–635.

Mutz, D. C. and Reeves, B. (2005). The new videomalaise: Effects of televisedincivility on political trust. American Political Science Review, 99(01):1–15.

Nanou, K. and Dorussen, H. (2013). European integration and electoral democracy:How the european union constrains party competition in the member states. Eu-ropean Journal of Political Research, 52(1):71–93.

Nooruddin, I. and Chhibber, P. (2008). Unstable politics, fiscal space, and electoralvolatility in the indian states. Comparative Political Studies, 41(8):1069–1091.

Norris, P. (1999). Critical Citizens: Global Support for Democratic Government:Global Support for Democratic Government. Oxford University Press, Oxford, UK.

Nugier, A., Niedenthal, P. M., Brauer, M., and Chekroun, P. (2007). Moral and angryemotions provoked by informal social control. Cognition and Emotion, 21(8):1699–1720.

Olson, D. M. (1998). Party formation and party system consolidation in the newdemocracies of central europe. Political Studies, 46(3):432–464.

Ostrogorski, M. (1902). Democracy and the organization of political parties, volume 2.Macmillan.

Pacek, A. C., Pop-Eleches, G., and Tucker, J. A. (2009). Disenchanted or discerning:voter turnout in post-communist countries. The Journal of Politics, 71(02):473–491.

Patterson, T. E. (2009). The vanishing voter: Public involvement in an age of uncer-tainty. Random House Digital, Inc., New York, NY.

Petersen, M. B. (2010). Distinct emotions, distinct domains: Anger, anxiety andperceptions of intentionality. The Journal of Politics, 72(02):357–365.

Plutzer, E. (2002). Becoming a habitual voter: Inertia, resources, and growth inyoung adulthood. American Political Science Review, 96(01):41–56.

Pop-Eleches, G. (2010). Throwing out the bums: protest voting and unorthodoxparties after communism. World Politics, 62(2):221–260.

192

Powell, E. N. and Tucker, J. A. (2014). Revisiting electoral volatility in post-communist countries: new data, new results and new approaches. British Journalof Political Science, 44(1):123–147.

Prinz, J. (2007). The emotional construction of morals. Oxford University Press,New York.

Prior, M. (2005). News vs. entertainment: How increasing media choice widensgaps in political knowledge and turnout. American Journal of Political Science,49(3):577–592.

Przeworski, A. (1991). Democracy and the market: Political and economic reforms inEastern Europe and Latin America. Cambridge University Press, Cambridge, UK.

Putnam, R. D. (1994). Making democracy work: Civic traditions in modern Italy.Princeton University Press.

Quigley, B. M. and Tedeschi, J. T. (1996). Mediating effects of blame attributions onfeelings of anger. Personality and Social Psychology Bulletin, 22(12):1280–1288.

Redondo, J., Fraga, I., Padron, I., and Comesana, M. (2007). The spanish adaptationof anew (affective norms for english words). Behavior research methods, 39(3):600–605.

Ridout, T. N. and Franz, M. (2008). Evaluating measures of campaign tone. PoliticalCommunication, 25(2):158–179.

Rose, R. (2007). Learning to support new regimes in europe. Journal of Democracy,18(3):111–125.

Rose, R. and Mishler, W. (1998). Negative and positive party identification in post-communist countries. Electoral Studies, 17(2):217–234.

Roseman, I. J., Wiest, C., and Swartz, T. S. (1994). Phenomenology, behaviors, andgoals differentiate discrete emotions. Journal of Personality and Social Psychology,67(2):206–221.

Rozin, P., Haidt, J., and McCauley, C. R. (2000). Disgust. In Lewis, M. and Haviland-Jones, J. M., editors, Handbook of Emotions, pages 637–653. The Guilford Press,New York, NY.

Rozin, P., Lowery, L., Imada, S., and Haidt, J. (1999). The cad triad hypothesis:a mapping between three moral emotions (contempt, anger, disgust) and threemoral codes (community, autonomy, divinity). Journal of Personality and SocialPsychology, 76(4):574.

Rozin, P. and Royzman, E. B. (2001). Negativity bias, negativity dominance, andcontagion. Personality and social psychology review, 5(4):296–320.

193

Rozin, P. and Singh, L. (1999). The moralization of cigarette smoking in the unitedstates. Journal of Consumer Psychology, 8(3):321–337.

Rupnik, J. (2014). From the revolutions of 1989 to democracy fatigue in eastern eu-rope. In Rupnik, J., editor, 1989 as a political world event: democracy, Europe andthe new international system in the age of globalization, pages 56–71. Routledge,New York, NY.

Russell, P. S. and Giner-Sorolla, R. (2011). Moral anger, but not moral disgust,responds to intentionality. Emotion, 11(2):233.

Ryan, T. J. (2012). What makes us click? demonstrating incentives for angry dis-course with digital–age field experiments. Journal of Politics, 74(4):1138–1152.

Sabato, L. (1991). Feeding frenzy: How attack journalism has transformed Americanpolitics. Free Press, New York, NY.

Santana, A. D. (2014). Virtuous or vitriolic: The effect of anonymity on civility inonline newspaper reader comment boards. Journalism Practice, 8(1):18–33.

Scheppele, K. L. (2013). The rule of law and the frankenstate: why governancechecklists do not work. Governance, 26(4):559–562.

Seawright, J. (2012). Party-System Collapse: The Roots of Crisis in Peru andVenezuela. Stanford University Press.

Sikk, A. (2005). How unstable? volatility and the genuinely new parties in easterneurope. European Journal of Political Research, 44(3):391–412.

Taber, C. S. and Lodge, M. (2006). Motivated skepticism in the evaluation of politicalbeliefs. American Journal of Political Science, 50(3):755–769.

Taggart, P. (2002). Populism and the pathology of representative politics. In Meny,Y. and Surel, Y., editors, Democracies and the populist challenge, pages 62–80.Palgrave Macmillan, London.

Tavits, M. (2005). The development of stable party support: Electoral dynamics inpost-communist europe. American Journal of Political Science, 49(2):283–298.

Tavits, M. and Annus, T. (2006). Learning to make votes count: The role of demo-cratic experience. Electoral Studies, 25(1):72–90.

Tavits, M. and Letki, N. (2009). When left is right: party ideology and policy inpost-communist europe. American Political Science Review, 103(4):555–569.

Tooby, J., Cosmides, L., and Price, M. E. (2006). Cognitive adaptations for n-person exchange: the evolutionary roots of organizational behavior. Managerialand Decision Economics, 27(2-3):103–129.

194

Vachudova, M. A. (2005). Europe undivided: democracy, leverage, and integrationafter communism. Oxford University Press, Oxford, UK.

Valentino, N., Gregorowicz, K., and Groenendyk, E. (2009). Efficacy, emotions andthe habit of participation. Political Behavior, 31(3):307–330.

Valentino, N. A., Brader, T., Groenendyk, E. W., Gregorowicz, K., and Hutchings,V. L. (2011). Election nights alright for fighting: The role of emotions in politicalparticipation. Journal of Politics, 73(1):156–170.

Valentino, N. A., Hutchings, V. L., Banks, A. J., and Davis, A. K. (2008). Is a worriedcitizen a good citizen? emotions, political information seeking, and learning via theinternet. Political Psychology, 29(2):247–273.

Vandenbroek, L. M. (2012). Principled abstention: a theory of emotions and non-voting in us presidential elections. Doctoral Dissertation in Political Science at theUniverstiy of Texas at Austin.

Verba, S. and Nie, N. H. (1972). Participation in America. University of ChicagoPress, Chicago, IL.

Volkens, A., Lehmann, P., Merz, N., Regel, S., and Werner, A. (2014). The ManifestoData Collection. Manifesto Project (MRG / CMP / MARPOR). Version 2014b.Wissenschaftszentrum Berlin fr Sozialforschung, Berlin. with Schultze, Henrike.

Wessels, B. and Klingemann, H.-D. (2006). Parties and votersrepresentative consoli-dation in central and eastern europe? International Journal of Sociology, 36(2):11–44.

Westen, D., Blagov, P. S., Harenski, K., Kilts, C., and Hamann, S. (2006). Neuralbases of motivated reasoning: An fmri study of emotional constraints on parti-san political judgment in the 2004 us presidential election. Journal of cognitiveneuroscience, 18(11):1947–1958.

Zajonc, R. B. (1980). Feeling and thinking: Preferences need no inferences. Americanpsychologist, 35(2):151.

Zaller, J. (1992). The nature and origins of mass opinion. Cambridge universitypress.

Zielinski, J. (2002). Translating social cleavages into party systems. World Politics,54(1):184–211.

Zielinski, J., Slomczynski, K. M., and Shabad, G. (2005). Electoral control innew democracies: the perverse incentives of fluid party systems. World Politics,57(03):365–395.

195

Appendix A: Content Analysis Supplementary Materials

A.1 The Seven Sins of the Gyurcsany Coalition Pamphlet

Figure A.1: The first page of the “7 Sins of the Gyurcsany Coalition”

196

English Translation Original HungarianThe Gyurcsany coalition’s seven sins A Gyurcsny-koalci ht fbneThey don’t deserve another chance! Nem rdemelnek tbb eslyt!Because they had eight years to proveit

Mert vot 8 vk, hogy bizonytsanak

1. In 2009 they took a months retire-ment from the elderly

1. 2009-ben elvettek egyhavi nyugdjataz idsektl

2. Between 2002 and 2010 they raisedutilities 15 times; they tripled the priceof gas and doubled the price of electric-ity

2. 2002 s 2010 kztt 15-szr emeltk arezsit, hromszorosra emeltk a gz s du-pljra a villany rt

3. They took a year of assistance fromnew mothers and got rid of tax creditsfor having children

3.Elvettek egyvnyi gyest a kismamktl seltrltk a gyermekek utn jr adkedvesm-nyt

4. They made more than a quarter mil-lion unemployed, and unemploymentrose from 5.8 to 11.2 percent between2002 and 2010

4.Tbb mint negyed-milli tettekmunkanlkliv, ezzel a munkanlklisg2002 s 2010 5,8 szzalkrl 11,2-re ntt.

5. They introduced copayments, givingfees for doctor examinations to the sickand elderly

5. Bevezettk a vizitdjat, fizetss tettkaz orvosi vizsglatokat a betegeknek s azidseknek

6. They stole BKV (Budapest PublicTransit) money in Nokia boxes, deplet-ing the common wealth by millions.

6. Nokis-dobozokban loptk el a BKVpnzt, millikkal krostottk meg a kzvagy-ont.

7. They took a month of wages. Theytook the 13th month payment fromclose to a million people.

7. Elvettek egzhavi munkabrt. Kzelegymilli embertl vettk el a 13. havi jut-tatst.

197

A.2 Manifestos Data Scales

Below are the Manifestos Data indicators that go in to each of the three subscalesof populism. The overall populism measure found in the body of the text is anadditive index of all of these subscales.

Subscale 1: Nationalistic Populism

1. per109: Negative references to internationalism

2. per110: Negative references to the European Union and/or European Com-munity

3. per601: Positive references to the country’s traditional national way of life

4. per605: Positive references to law and order

5. per606: Positive references to civic mindedness

6. per608: Negative references to multiculturalism

Subscale 2: Anti-Communist/Corruption

1. per304: Mentions of Political Corruption

2. per305: References to the party’s political authority and ability to deliverclean, transparent, efficient governance

3. per3053: Anti-communist rhetoric

Subscale 3: Economic populism

1. per406: Positive references to protectionism

2. per408: Mentions of broad positive economic goals

3. per410: Broad mentions of economic growth

4. per411: Advocacy of building infrastructure and investing in technology toincrease economic prosperity

198

A.3 Economic Populism Results

The overall populism measure provides an interesting first look and is a powerful

bit of evidence that populism is on the increase. This, however, is a very broad scale

and obscures a great deal of potentially meaningful variation. Thus, the subscales and

individual components of the subscales were also charted in order to uncover trends

at a lower level of aggregation. Economic populism presents the clearest trend of

any of the subscales. Therefore, it appears to be the case that the over-time gradual

increases in populism documented in the aggregate populism measure above is most

often a function of a steady increase in economic populism over time.

Why should economic populism in particular increase? This finding makes some

sense in light of the tight control that the European Union and other powerful in-

ternational institutions of the global neo-liberal order exerted upon these cases over

time. Indeed, the EU exerts a particularly strong grip upon economic issues. The

finding that economic populism has increased in ECE is in line with Mudde (2001)’s

analysis of populist appeals in Europe. At his writing he expressed great surprised

that economic populism had not erupted in ECE and argued it held great potential

(Mudde, 2001). These data suggests that after Mudde’s analysis his prediction did in

fact occur. Whereas he pointed to the post-communist legacy in isolation as a root

cause, these data suggest that this legacy in addition to the constraint of European

Union accession and membership work together to drive economic populism in ECE.

199

(a) Fidesz (b) Socialists

(c) Democratic Forum (d) Liberal Democrats

Figure A.2: Economic Populism in Hungary

200

(a) Civic Democratic Party (b) Christian Democratic Union

(c) Social Democratic Party (d) Communists

Figure A.3: Economic Populism in the Czech Republic

201

(a) Civic Platform (b) Democratic Left Alliance

(c) Law and Justice (d) Polish Peasants Party

Figure A.4: Economic Populism in Poland

202

(a) Party of Democratic Left (-2002)Smer (2002-)

(b) Movement for a Democratic Slo-vakia

(c) Democratic and Christian Union (d) Slovak Nationalist Party

(e) Hungarian Coalition / Bridge

Figure A.5: Economic Populism in the Slovak Republic

203

A.4 Nationalistic Populism

(a) Civic Democratic Party (b) Christian Democratic Union

(c) Social Democratic Party (d) Communists

Figure A.6: Nationalistic / Anti-EU Populism in the Czech Republic

204

(a) Civic Platform (b) Democratic Left Alliance

(c) Law and Justice (d) Polish Peasants Party

Figure A.7: Nationalistic / Anti-EU Populism in Poland

205

(a) Party of Democratic Left (-2002)Smer (2002-)

(b) Movement for a Democratic Slo-vakia

(c) Democratic and Christian Union (d) Slovak Nationalist Party

(e) Hungarian Coalition / Bridge

Figure A.8: Nationalistic / Anti-EU Populism in the Slovak Republic

206

(a) Fidesz (b) Socialists

(c) Democratic Forum (d) Liberal Democrats

Figure A.9: Anti-Communist/Corruption Populism in Hungary

207

(a) Civic Democratic Party (b) Christian Democratic Union

(c) Social Democratic Party (d) Communists

Figure A.10: Anti-Communist/Corruption Populism in the Czech Republic

208

(a) Civic Platform (b) Democratic Left Alliance

(c) Law and Justice (d) Polish Peasants Party

Figure A.11: Anti-Communist/Corruption Populism in Poland

209

(a) Party of Democratic Left (-2002)Smer (2002-)

(b) Movement for a Democratic Slo-vakia

(c) Democratic and Christian Union (d) Slovak Antiionalist Party

(e) Hungarian Coalition / Bridge

Figure A.12: Anti-Communist/Corruption Populism in the Slovak Republic

210

A.5 Inductive Outrage Dictionary

English Hungarianbad, get bad/worse rosszabuse visszael(es)anger, angry duh(os), merges/meregarrogant onteltassualt tamad(as)bankrupt vagyonbuk(ott), csod(be)betray elrul, hutlenulbetrayal elarulasbizarre bizarr, groteszkblackmail zsarol(as)bluff bloffbrutal llati, brutalischeat, cheater megcsal, csal(o)coercion kenyszer, korlatozcollapse osszeomlik, osszeomlascommunism, communist kommunista, kommunizmuscorrupt korruptcorruption korrupciocoward(ly) gyavacriminal bun, buntetocrisis valsag, krıziscynical cinikuscynicism cinizmusdebauchery, debauch zulles, dozsol(es)

erkolcsront(as), dorbezol(as)destroy lerombol, pusztıtdestructive rombolas, pusztıtasdevour zabal, elnyel, felfaldisgrace kegyvesztett(seg)disorganized, undone szetzullesztdreadful felelmetesembarrassed zavartendanger veszelyeztetenemy, hostile ellenseg(es)err csalodevil gonoszexaggerate tulz; tulozexaggeration tulzasexcuse (n) kifogasfailure bukas

211

English Hungarianfailure, fail megbuk, hibazikfalse hamisfasist fasisztafavoritism, nepotism kivetelezesfiasco, screw up kudarcforbid, object tiltakozfraud(ulent) fondorlat(os), svindligreed moho, nagyetkugreed penzvagyhate gyulol(et)hatred utal(at)hazard, danger, dangerous veszely, veszelyeshell pokolhypocrisy alszenteskedes, kemutatas, hipokrızisillegal illegalisimpair, deplete/depletion karosıt(as)impose rarakincompetent illetekteleninjure; batter serulinsane, insanity orult(seg)insecure megbızhatatlaninsensitive erzeketlen, kozonyosintolerable elviselhetetlenintolerant turelmetlenintruder tolakodoinvader merszallojail bortonlie hazugsaglie(v) hazudmistake hibamoney grubber penzsovarMultinational firms Multinacionalisneglect elhanyagol(as)obscene tragarobsolete osdi, idejetmultoffend vetkezopposition ellenzekiopprobrium, ignominy gyalazatoutrage felhaborıtpathetic szanalmas, patetikus

212

English Hungarianpatronage partfogas, vednok(seg)prostitue prostitualpunish, punishment buntet(es)radical radikalisravenous, rapacious kapzsi, falankrevolution, revolutionary, revolutionize forradal(om)(mar)ridiculous nevetsegesriot lazadasrisk, gamble kockaztat(as)ruin tonkreruin, rot romlik, romol, romloscandal botranysecret titkos, titoksecret, undercover titkolseduce, inveigle, debauch elcsabıt(as)seduction elcsabıtasselfish onos, onzoselfishness onzessensationalist szenzaciosshame szegyen(kezes)sick(ness) beteg(seg)slushfund kenopenzspending koltessquander elpazarolstagnation, stagnate stagnal(as)stressful nyomatekosstupid, idiot hulye, butasubterfuge, pretext urugysuspicion; suspicious gyanutake away elvesztemptation csabıtasthreaten, threat fenyeget(es)throw over, betray, cserbenhagyleave high and drytraitors arulotreasonous hitszegotruthless szoszegougency, urgent surgos, surgetougly csunyaunfaithful hutlenunsubstantiated bizonytalanupset felfordıt

213

English Hungarianvice, excess kicsapong(as), erkolcstelen(seg)victimize/victim raszed(es)war haboruwaste pocsekol, elpusztıtweaken, weakened, weakening gyengul

214

A.6 ANEW Outrage dictionary

Hungarian English Affect Arousalveteles abortion 3.5 5.39visszaeles abuse 1.8 6.83baleset accident 2.04 6.26felo afraid 2 6.67gyotrelem agony 2.43 6.06mentoauto ambulance 2.47 7.33duh anger 2.34 7.63merges angry 2.85 7.17elkınzott anguished 2.12 5.33bosszant annoy 2.74 6.49ontelt arrogant 3.69 5.65vago assassin 3.09 6.28tamadas assualt 2.02 7.51lavina avalanche 3.29 5.54vagyonbukott bankrupt 2 6.21fattyu bastard 3.36 6.07meh bees 3.2 6.51elarul betray 1.68 7.24zsarol blackmail 2.95 6.03istenkaromlas blasphemy 3.75 4.93veres bloody 2.9 6.41bomba bomb 2.1 7.15torott broken 3.05 5.43allati brutal 2.8 6.6golyo bullet 3.29 5.33teher burdened 2.5 5.63temetes burial 2.04 5.08eg burn 2.73 6.22rak cancer 1.5 6.42svabbogar cockroach 2.81 6.11korporso coffin 2.56 5.03zavaros confused 3.21 6.03ellenorzes controlling 3.8 6.1

215

Hungarian English Affect Arousalkorrupt corrupt 3.32 4.67gyava coward 2.74 4.07csattanas crash 2.31 6.95buntett crime 2.89 5.41bunos criminal 2.93 4.79vlsag crisis 2.74 5.44gyotor crucify 2.23 6.47durva crude 3.12 5.07kegyetlen cruel 1.97 5.68osszepresel crush 2.21 5.52vag cut 3.64 5ciklon cyclone 3.6 6.36leszur dagger 3.38 6.14holt dead 1.94 5.73halal death 1.61 4.59adossag debt 2.22 5.68csalas deceit 2.9 5.68elvesztett defeated 2.34 5.09keses delayed 3.07 5.62demon demon 2.11 6.76szokeveny deserter 2.45 5.5ketsegbeesett despairing 2.43 5.68lenez despise 2.02 6.28lerombol destroy 2.64 6.83rombolas destruction 3.16 5.82megvet detest 2.17 6.06ordog devil 2.21 6.07saros dirty 3.08 4.88meghiusıt disappoint 2.39 4.92katasztrofa disaster 1.73 6.33megveto disdainful 3.68 5.04undor disgusted 2.45 5.42hutlen disloyal 1.93 6.56elegedetlen displeased 2.79 5.64kimerult distressed 1.94 6.4haborgat disturb 3.66 5.8elvalas divorce 2.22 6.33felelmetes dreadful 2.25 5.84elont drown 1.92 6.57zavart embarrassed 3.03 5.87felboszult enraged 2.46 7.97gonosz evil 3.23 6.39

216

Hungarian English Affect Arousalkivegzes execution 2.37 5.71bukas failure 1.7 4.95felelem fear 2.76 6.96rettegett fearful 2.25 6.33piszok filth 2.47 5.12tuz fire 3.22 7.17arvız flood 3.19 6fondorlat fraud 2.67 5.75nyugtalan frustrated 2.48 5.61uszok gangrene 2.27 5.7szemet garbage 2.98 5.04pletyka gossip 3.48 5.74penzvagy greed 3.51 4.71zugolodas gripe 3.14 5nyaktilo guillotine 2.48 6.56vetkes guilty 2.63 6.04fegyver gun 3.47 7.02gyulolet hate 2.12 6.95utalat hatred 1.98 6.66fejfajas headache 2.02 5.07pokol hell 2.24 5.38tehetetlen helpless 2.20 5.34remulet horror 2.76 7.21tusz hostage 2.20 6.76ellenseges hostile 2.73 6.44megalaz humiliate 2.24 6.14ehes hungry 3.58 5.13orkan hurricane 3.34 6.83megserul hurt 1.9 5.85hulye idiot 3.16 4.21tudatlansag ignorance 3.07 4.39kor illness 2.48 4.71eretlen immature 3.39 4.15erkolcstelen immoral 3.5 4.98elront impair 3.18 4.04gyenge impotent 2.81 4.57fertozes infection 1.66 5.03kar injury 2.49 5.69orult insane 2.85 5.83bizonytalan insecure 2.36 5.56sertes insult 2.29 6

217

Hungarian English Affect Arousaltolakodo intruder 2.77 6.86merszallo invader 3.05 5.5ingerel irritate 3.11 5.76borton jail 1.95 5.49feltekenyseg jealousy 2.51 6.36gyilkos killer 1.89 7.86kes knife 3.62 5.8per lawsuit 3.37 4.93lepra leprosy 2.09 6.29engedely lice 2.31 5hazugsag lie 2.79 5.96elhagyatottsag loneliness 1.61 4.56elhagyott lonely 2.17 4.51vesztes loser 2.25 4.95elkallodott lost 2.82 5.82tetu louse 2.81 4.98bolond mad 2.44 6.76kukac maggot 2.06 5.28malaria malaria 2.4 4.40rosszindulat malice 2.69 5.86mangorlo mangle 3.9 5.44duhongo maniac 3.76 5.39tragya manure 3.1 4.17meszarlas massacre 2.27 5.33kanyaro measles 2.74 5.05fenyegetes menace 2.88 5.52szenvedes misery 1.93 5.17hiba mistake 2.86 5.18beteges morbid 2.87 5.05tetemnezu morgue 1.92 4.84szunyog mosquito 2.8 4.78gyilkos murderer 1.53 7.47megcsonkıt mutilate 1.82 6.41undorıto nasty 3.58 4.88varrotu needle 3.82 5.36elhanyagolas neglect 2.63 4.83ideges nervous 3.29 6.59remkep nightmare 1.91 7.59felakaszt noose 3.76 4.39serelem nuisance 3.27 4.49visszataszıto obnoxious/vile 3.5 4.74tragar obscene 4.23 5.04vetkez offend 2.76 5.56

218

Hungarian English Affect Arousalfelhaborıt outrage 3.52 6.83fajdalom pain 2.13 6.5vakremulet panic 3.12 7.02benulas paralysis 1.98 4.73buntetes penalty 2.83 5.09hitehagyott pervert 2.79 6.26kartevo pest 3.13 5.62veszekedes quarrel 2.93 6.29veszettseg rabies 1.77 6.1tombol rage 2.41 8.17torkoly rape 1.25 6.81patkany rat 3.02 4.95sajnalkozo regretful 2.27 5.74visszataszıtott rejected 1.5 6.37nevetseg ridicule 3.13 5.83rideg rigid 3.66 4.66lazadas riot 2.96 6.39csotany roach 2.35 6.64rablo robber 2.61 5.62romlott rotten 2.25 4.53udvariatlan rude 2.5 6.31bus sad 1.61 4.13forralas scalding 2.82 5.95botrany scandal 3.32 5.12bunbak scapegoat 3.67 4.53heg scar 3.38 4.79remult scared 2.78 6.82perzseles scorching 3.76 5megvetes scorn 2.84 5.48gunyos scornful 3.02 5.04skorpio scorpion 3.69 5.38sopredek scum 2.43 4.88skorbut scurvy 3.19 4.71tengeribeteg seasick 2.04 5.8onzu selfish 2.42 5.5szigoru severe 3.2 5.26gyalazat shamed 2.5 4.88capa shark 3.65 7.16sikoltas shriek 3.93 5.36beteg sick 1.9 4.29betegseg sickness 2.25 5.61vetek sin 2.8 5.78bun sinful 2.93 6.29

219

Hungarian English Affect Arousalpuhany sissy 3.14 5.17utes slap 2.95 6.46leoles slaughter 1.64 6.77rabszolga slave 1.84 6.21iszap slime 2.68 5.36nyomornegyed slum 2.39 4.78himlo smallpox 2.52 5.58kıgyo snake 3.31 6.82sznob snob 3.36 5.65keseru sour 3.93 1.95pok spider 3.33 5.71ehezo starving 2.39 5.61buz stench 2.19 4.36budosseg stink 3 4.26nyomas stress 2.09 7.45buta stupid 2.31 4.72megfojt suffocate 1.56 6.03ongyilkossag suicide 1.25 5.73sebeszet surgery 2.86 6.35gyanakve suspicious 3.76 6.25szifilisz syphilis 1.68 5.69feszes tense 3.56 6.53termesz termite 3.58 5.39borzaszto terrible 1.93 6.27megfelemlıtett terrified 1.72 7.86terrorista terrorist 1.69 7.27tolvaj thief 2.13 5.89tuske thorn 3.64 5.14felenk timid 3.86 4.11dohany tobacco 3.28 4.83sır tomb 2.94 4.74fogfojas toothache 1.98 5.55tornado tornado 2.55 6.83kınzas torture 1.56 6.1mergezo toxic 2.1 6.4tragedia tragedy 1.78 6.24arulo traitor 2.22 5.78limlom trash 2.67 4.16serules trauma 2.1 6.33baj trouble 3.06 6.85zaklatott troubled 2.17 5.94tumor tumor 2.36 6.51

220

Hungarian English Affect Arousalcsunya ugly 2.43 5.38fekely ulcer 1.78 6.12htlenseg unfaithful 2.05 6.2szerencsetlen unhappy 1.57 4.18felfordıt upset 2 5.86vizelet urine 3.25 4.2hasznalhatatlan useless 2.13 4.87rombolo vandal 2.71 6.4mereg venom 2.68 6.08aldozat victim 2.1 6.06heves violent 2.29 6.89kihıny vomit 2.06 5.75haboru war 2.08 7.49elpusztıt waste 2.93 4.14fegyvertar weapon 3.97 6.03szajha whore 2.30 5.85rossz wicked 2.96 6.09seb wound 2.50 5.8

221

Appendix B: Experiment and Survey SupplementaryMaterials

Experimental Conditions

Control:In light of the many kinds of diets available today, there is a large debate about

the best diets. The Paleo diet has received a lot of attention recently. Please readthis opinion piece – The 5 sins of the Paleo diet — and let us know how you feelabout it. In order to best understand your emotional reactions to each point, we aregoing to present them to you one at a time. After you see each, please indicate theemotion that each statement makes you feel the most, as well as the degree to whichyou feel the statement is credible.

The 5 sins of the Paleo diet:

1. The diet steals many important nutrients from your body by denying it ofnumerous types of food, making it harder for you to stay energetic and healthy.

2. The diet condemns both dairy and fortified cereals, meaning calcium intake canfall to dangerously low levels. This increases the risk of bone related injuriesand illnesses.

3. As with other diets that are extremely low in carbohydrates, as soon as peopleon the diet eat carbohydrates again they often gain any weight they have lostback, plus some.

4. The diet advocates far too much red meat and therefore makes people take intoo much saturated fat. This can cause dangerous increases in cholesterol andall the heart and cardiovascular problems related to cholesterol.

5. The diet industry pushing the Paleo diet is filled with the same people who havepresented numerous other fad diets in the past. Just because cave men ate acertain way does not mean it is the best way for humans to live.

222

Political control :In light of the importance of the agricultural sector to the economy, there is a

large debate about the best agricultural policies. Potential agricultural reforms haveattracted a lot of attention recently. Please read this opinion piece — The 5 problemswith agricultural policy — and let us know how you feel about it. In order to bestunderstand your feelings about each point, we are going to present them to you oneat a time. After you see each, please indicate the emotion that each statement makesyou feel the most, as well as the degree to which you feel the statement is credible.

The 5 problems of agricultural policy:

1. Farm subsidies steal from many of the farmers who need help the most most,and instead benefit larger farms that would be fine without them anyway, arguesAttila Meszterhazy, of MSZP

2. According to Lajos Kosa of Fidesz, most parts of agricultural policy are highlyout of date and need to be re-considered for new farming techniques in thetwenty-first century

3. As noted by the spokeswoman for LMP in a recent press release, agriculturalpolicies as they are now contribute to farm practices that are bad for the nation’senvironment

4. According to Hedegs Lorantn, a candidate for Jobbik, agricultural policy inHungary today does not take into consideration the preservation of the socialstructure in the small towns and villages where Hungarian agriculture mainlyhappens.

5. Many political and economic experts argue that agricultural policy as it standsnow is partly responsible for rising food prices in our supermarkets.

223

Outrage: Left attacks the right In light of the importance of the events inparliament to the country, there is always a debate about those in parliament. Pleaseread this opinion piece — The 5 sins of the Orban government — and let us knowhow you feel about the article. In order to best understand your emotional reactionsto each point, we are going to present them to you one at a time. After you see each,please indicate the emotion that each statement makes you feel the most, as wellas the degree to which you feel the statement is credible. The 5 sins of the Orbangovernment

1. The Fidesz government stole benefits from those who our society depends onthe most, like fireman, soldiers, and pensioners. Of course, they lied about thisall along.

2. The corruption of the Fidesz henchmen knows no ends. They have always usedany political power they could get for their own benefit, lining their own pocketsand getting themselves and their connections rich while real Hungarians suffer.Just think of the shady trafik scandal: they stole the livelihoods of severalthousand families who had those stores for 20 years and divvied up the rewardsamong themselves. They’re not fooling anyone with these nasty tricks.

3. The past years of their government have been the most undemocratic timesince the system change. Their biased redistricting and their media law guttedHungarian democracy and has almost left the country a one-party dictatorshipunder Fidesz. Because of all their treachery and damage to Hungarian democ-racy, they deserve a room in the House of Terror, NOT a role in Hungarianpolitical life.

4. The Orban government is completely out of touch with real Hungarians. Theywaste state money on stadiums when so many Hungarians are unemployed andneed the dignity and security of a job. We can’t pay our bills with stadiums!

5. The Fidesz ultra-nationalists have turned Hungary into an international laugh-ing stock with their phony constitution and their all out assault on Hungariandemocracy. The party is full of liars from Antl with his hidden luxury apart-ments to Pal Schmitt with his plagiarized PhD. They have lied, cheated, andstolen at every turn and naturally they’ll do it again!

224

Outrage: Right attacks the leftIn light of the importance of the events in parliament to the country, there is

always a debate about those in parliament. Please read this opinion piece — The 5sins of the Gyurcsany coalition — and let us know how you feel about the article.In order to best understand your emotional reactions to each point, we are going topresent them to you one at a time. After you see each, please indicate the emotionthat each statement makes you feel the most, as well as the degree to which you feelthe statement is credible.

The 5 sins of the Gyurcsany coalition:

1. Gyurcsany and Bajnai stole benefits from pensioners, students, and pregnantmothers the people our society relies upon most. Of course, they lied aboutthis all along.

2. The corruption of MSZP knows no ends. They have always used any politicalpower they could get for their own benefit, lining their own pockets and gettingthemselves and their connections rich while real Hungarians suffer. Just thinkof Gabor Simon: his foreign bank accounts were filled with money he stole fromHungarian citizens. It’s clear to see that Gyurcsany wasn’t clear of that messeither. They’re not fooling anyone with these nasty tricks.

3. The last MSZP government were the worst years in Hungary since the systemchange. Their biased economic policies gutted the Hungarian economy and leftit in a pathetic state, and they lied about it at szd. Because this treachery, alongwith everything else they did, they deserve a room in the House of Terror, NOTa role in Hungarian politics.

4. The left wing are completely out of touch with real Hungarians. Their policiesled to the credit crisis that crippled so many Hungarian families. We don’t haveSwiss franks to pay MSZP’s friends in foreign banks!

5. The former communists in MSZP turned Hungary into an international laughingstock with their backwards policies and constant stream of scandals. The partyis full of liars, from Mikos Hagyo’s extortion to Gyurcsny’s “lost” diploma. Theyhave lied, cheated, and stolen at every turn and naturally they’ll do it again!

Experiment Measures

225

Table B.1: Summary Statistics

Variable Details Range Mean Standarddeviation

Anger Scale 4–item scale (α = .86); see textnote 7

0–1 .3951 .2632

Disgust Scale 3–item scale (α = .78); see textnote 8

0–1 .3946 .2627

Ideology When people talk about polit-ical themes they often use theexpressions “left” and “right.”Referring to your own views,where would you place yourselfon the scale below? (0 – 11)

0–1 .523 .217

Extreme Computed by folding ideology 0–1 .337 .277Govt Sat How satisfied are you with the

current government’s job in of-fice? (Very unsatisfied — Verysatisfied)

0–1 .401 .260

Vote if un-happy

(Agree/Disagree) If you areunhappy with politics, the bestthing to do is to vote

1–4 2.953 1.012

Too Compli-cated

(Agree/Disagree) Generallypolitics is too complicated forpeople like me to understand

1–4 2.650 .993

Don’t Care (Agree/Disagree) Politiciansdon’t care about people likeme

1–4 2.215 .865

Economy 3–item scale; see appendixnote 1

0–1 .291 .224

Knowledge 6 open-ended questions identi-fying who fills prominent polit-ical offices

0–1 .407 .279

Female Dummy variable 1 = Female 0/1 .749 .434

226

Survey Measures

Emotion measures:Instructions: We would like to ask you about your emotional reaction to a number ofaspects of Hungarian social and political life. We are going to list a number of thingsand ask that you provide your strongest emotion. There are no right or wrong waysto feel about these topics, we are just interested in how normal people like you feel.A lot of the time people have mixed emotions towards things like these, and if that’sthe case, please try to indicate your strongest, first, reaction.Respondents selected oneof the following options from a drop-down menu: happy/content; proud/enthusiastic;angry/outraged; disgusted/repulsed; anxious/afraid; apathetic/nothing; I don’t know/ not familiar with this.

1. Hungarian politics

2. The European Union

3. The “trafik” (cigarette store) scandal

4. Former Prime Minister Gyurcsany’s speech at Oszod

5. The credit crisis

6. The Fidesz constitution

7. Simon-Welsz foreign bank account (money laundering scandal)

8. Paks nuclear plant deal

227

Table B.2: Summary Statistics

Variable Details Range Mean Standarddeviation

Anger Times participant rated an-gry/outraged as dominantresponse

0–8 2.01 2.08

Disgust Times participant rated dis-gusted/repulsed as domi-nant response

0–8 1.27 1.65

Age How old are you? 18–86 47.81 17.01Extreme Computed by folding ideol-

ogy0–5 1.96 1.60

Female Dummy variable 1 = Fe-male

0/1 .52 .50

All following measures in a single battery in which participantswere asked to agree/disagree with the statements on a7–point Likert scale (1 = Completely disagree to 7 = Completely agree))Need to Eval-uate

I think it is better to havedefinite opinions about lotsof things rather than to re-main neutral on most issues(Bizer et al., 2000)

1–7 5.047 1.45

Economy Over recent years, the Hun-garian economy has signifi-cantly improved

1–7 3.56 1.70

Vote Sunday If there were to be electionsthis Sunday, I’d definitelyvote

1–7 5.049 2.13

Satisfactionwith Dem

Generally speaking, I’m sat-isfied with how democracyworks in Hungary

1–7 3.60 1.76

Hungary aCompleteDem

Hungary is a completelydemocratic country

1–7 3.71 1.68

Dem the Best Democracy is the best po-litical system for a countrylike ours

1–7 4.755 1.63

No Party toSupport

There’s no party in parlia-ment, that I could feel goodvoting for.

1–7 3.93 2.00

228

Survey Results (Weighted)

Table B.3: Anger, Disgust, and Attitudes towards DemocracyVote Sunday Satisfaction Hungary a Dem the Best

w/ Dem Complete Dem SystemAnger 0.14∗∗∗ −0.06∗∗∗ −0.11∗∗∗ −0.02

(0.03) (0.02) (0.02) (0.03)Disgust 0.02 −0.03 −0.06∗∗ −0.10∗∗∗

(0.04) (0.03) (0.03) (0.03)Age 0.01∗∗ 0.00 0.00 0.00

(0.00) (0.00) (0.00) (0.00)Education 0.28∗∗∗ 0.06 −0.06 0.09∗

(0.06) (0.04) (0.04) (0.05)Female −0.02 0.18∗∗ −0.03 −0.10

(0.14) (0.09) (0.08) (0.10)Economy 0.27∗∗∗ 0.67∗∗∗ 0.60∗∗∗ 0.28∗∗∗

(0.04) (0.03) (0.02) (0.03)(Intercept) 2.65∗∗∗ 0.99∗∗∗ 1.96∗∗∗ 3.54∗∗∗

(0.33) (0.21) (0.20) (0.25)Adj. R2 0.08 0.45 0.43 0.10Num. obs. 950 942 948 924OLS models with survey weights. Standard errors in parentheses.

∗∗∗p < 0.01, ∗∗p < 0.05, ∗p < 0.1

229

Survey Results Alternative Specification

Table B.4: Anger, Disgust, and Attitudes towards Democracy (with Extreme)Vote Sunday Satisfaction Hungary a Dem the Best

w/ Dem Complete Dem SystemAnger 0.10∗∗∗ −0.07∗∗∗ −0.12∗∗∗ −0.01

(0.03) (0.02) (0.02) (0.03)Disgust 0.02 −0.05∗ −0.06∗∗ −0.12∗∗∗

(0.04) (0.03) (0.02) (0.03)Age 0.01∗∗∗ 0.00 −0.00 0.00

(0.00) (0.00) (0.00) (0.00)Education 0.26∗∗∗ 0.06 −0.07 0.05

(0.07) (0.05) (0.04) (0.06)Female 0.09 0.17∗ −0.01 −0.00

(0.12) (0.09) (0.08) (0.10)Economy 0.19∗∗∗ 0.67∗∗∗ 0.62∗∗∗ 0.27∗∗∗

(0.04) (0.03) (0.02) (0.03)Extreme 0.46∗∗∗ 0.03 −0.02 −0.02

(0.03) (0.02) (0.02) (0.03)(Intercept) 2.12∗∗∗ 1.10∗∗∗ 2.06∗∗∗ 3.70∗∗∗

(0.31) (0.22) (0.21) (0.26)Adj. R2 0.22 0.44 0.44 0.09Num. obs. 950 942 948 924∗∗∗p < 0.01, ∗∗p < 0.05, ∗p < 0.1

230