! ! ! ! ! ! ! ! ! ! Ritual Contention in Divided Societies: Participation in Loyalist Parades! in Northern Ireland ! Jonathan !S. Blake ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in the Graduate School! of Arts and Sciences ! ! COLUMBIA UNIVERSITY! ! 201!5 ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! © 2015! Jonathan !S. Blake All Rights Reserved Abstract! Ritual Contention in Divided Societies: Participation in Loyalist Parades! in Northern Ireland Jonathan !S. Blake !

Each year, Protestant organizations in Northern Ireland perform over 2,500 ritual parades to celebrate and commemorate their culture. Many Catholics, however, see parades as triumphalist and hateful. As a result, parades undermine the political peace process and grassroots peace- building by raising interethnic tension and precipitating riots, including significant violence in recent years. This dissertation asks: Why do people participate in these parades?

To answer this question, I consider loyalist parading as an example of contentious ritual

—symbolic action that makes contested political claims. To understand these parades as ritual actions, I build on two central insights from religious studies, sociology, and anthropology. First, as meaningful and shared practices, rituals provide participants with benefits that are intrinsic to participating in the act itself and do not depend on the achievement of some external outcome.

Second, rituals are multi-vocal, meaning that interpretations of the action can vary across actors.

Participants need not share the interpretation of their actions held by organizers, rivals, or outside observers. Participants, therefore, may not see the ritual as provocative, aggressive, or even contentious. These arguments stand in contrast to traditional explanations for collective action and ethnic conflict that theorize participation in ethnically polarizing events in terms of the achievement of concrete outcomes, such as selective material benefits, provoking the out-group into overreacting, or intimidating them into quiescence. To test my argument, I conducted fieldwork in Belfast, Northern Ireland. I developed and implemented a household survey to measure mass-level opinion, designed and ran an online survey of all Protestant clergy and elected officials in Northern Ireland to measure elite-level opinion, conducted over 80 semi-structured interviews with parade participants and nonparticipants, and observed dozens of hours of parades and related events. I demonstrate that, as expected by my argument, people approach participation in ritual parades as an end in and of itself. The evidence demonstrates that participants do not view parades instrumentally. This means that people make decisions to participate in contentious behavior without consideration of their actions’ profoundly political consequences. The ritual nature of parades severs the expected connection between means (participation) and ends (political consequences), thus creating the environment for sustained conflict. Furthermore, the predictions of influential theories of ethnic conflict—extreme in-group identification or out-group antipathy—and collective action— selective material benefits or sanctions—are not supported by the data. ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! Contents

List of Figures ii

List of Tables iii

Acknowledgements iv

Dedication viii

1 Introduction 1

2 Identity on the March in Northern Ireland 16

3 The Politics of Provocation: Politicians, Pastors, Paramilitaries, and Parades 42

4 A Theory of Ritual Participation 76

5 Parading For Mainly Fun and Process 120

6 Culture, Politics, and the Paradox of Anti-Politics in Loyalist Parading 168

7 For God and Ulster or Private Payoff? Assessing the Role of Collective and Selective Incentives 209 8 Conclusion 260

Bibliography 272

Appendix 305 ! ! ! ! ! ! ! ! ! ! ! ! ! ! !

i List of Figures

2.1 Loyalist and Nationalist Parades, 1985-2013 26

2.2 Contentious Parades, 1999-2013 29

3.1 Parades with Disorder and Elections, 1990-2009 58

3.2 Histogram of Elite Opinions on Parades 70

3.3 Kernel Density Plot of Opinions on Parades by Elected Officials and Clergy 70

7.1 Example Small Area 218

7.2 Screenshot from the Land and Property Service’s Spatial NI Map 218

7.3 Histogram of the Age Participants Began Parading 221

7.4 Percentage of Respondents Marching, by Age 221

7.5 Percentage of Participants in Each Organization 224

7.6 Summary of Respondents’ Parading Behavior, 2012-2013 226 !

ii List of Tables

3.1 The Determinants of Elite Parade Attendance, Logit Models 68

3.2 Elite Opinions on Parade Routes (%) 69

3.3 Parade-Related Activities of Elected Officials and Clergy (%) 72

3.4 Parade Organization Membership among Elites (%) 73

5.1 Types of Purposes Attributed to Loyalist Parades 128

5.2 Purposes of Loyalist Parades Reported by Participants: Intrinsic vs. Instrumental 129

7.1 Information on the Neighborhoods Included in the Survey 215

7.2 Determinants of Current Parade Participation, Logit Models 229

7.3 Robustness Check: Rare Events Logit 254

7.4 Robustness Check: Multiple Imputations 255

A1 Robustness Check: Alternative Measures of Protestant Identification 305

A2 Robustness Check: Alternative Measures of Anti-Catholicism and Family Ties 306

A3 Interviewees Quoted in Dissertation 307 !

iii Acknowledgments

It is my great pleasure to thank all the people who helped me research and write this dissertation. Jack Snyder has guided my thinking since day one. I thank him for trusting me to research what I thought was important and providing criticism to make it sharper. Al Stepan pushed me to think big and speak to debates that matter. Michael Doyle always provided moral support and encouraged my exploration—and directed me to financial support to make it possible. Jim Jasper provided detailed and thoughtful comments on my work that really took it to the next level. And Tim Frye ensured that I was careful with my argument and didn’t step too far.

Two people not on my committee deserve special recognition: Ron Hassner, who has been a teacher, mentor, and cheerleader for a decade; and Lucy Goodhart, whose dedicated guidance set me on the right track as I launched this project. I thank all for them for their continued support.

In Northern Ireland, many people graciously gave me their time and views, not to mention their tea. First and foremost, I want to thank all of the people who invited me into their homes and offices or met me at restaurants, cafes, and pubs for interviews. I truly appreciate their willingness to share their stories. Thanks also to the many people who provided much-needed counsel on my research: John Barry, Jonny Byrne, Paula Devine, John Garry, Neil Jarman, Dave

Magee, Kieran McEvoy, Dirk Schubotz, Peter Shirlow, and Ben Walker. Gladys Ganiel kindly let me use the database of clergy contact information that she and Therese Cullen compiled for the

‘Visioning 21st Century Ecumenism’ research project at the Irish School of Ecumenics. I appreciate the hard work of my survey enumerators, especially Rachel, Brenda, Tracey, Allison, and Julie. Finally, the good folks at Common Grounds Cafe and especially Black Bear Café kept me well caffeinated and let me set up shop.

iv A few people in Northern Ireland went above and beyond. In particular, I want to thank

Dominic Bryan, who taught me much of what I know about parades and who sponsored me as a visitor at the School of History and Anthropology at Queen’s University Belfast; Mark

Hammond, who kindly lent me a bicycle on three of my trips (once before he had even met me!); and Jon Evershed, who provided lots of good craic and was an keen companion in fieldwork.

My research in Northern Ireland was generously supported by a Doctoral Dissertation

Improvement Grant from the National Science Foundation (SES-1263772); the Endeavor

Foundation (formerly the Christian Johnson Endeavor Foundation); Columbia University’s

Department of Political Science and Advanced Consortium on Cooperation, Conflict, and

Complexity (at the Earth Institute); and APSA’s British Politics Group. Many thanks to Joe

Chartier for helping me deal with all the Endeavor grants. Earlier fieldwork in Jerusalem was funded by Columbia’s Center for the Study of Democracy, Toleration, and Religion.

A number of people read drafts of chapters and provided instructive feedback. Kate

Cronin-Furman deserves some sort of medal for doing it more than anyone else. A heartfelt thanks to her and to Hadas Aron, Jonathan Evershed, Lee Ann Fujii, Jeff Goodwin, John Krinsky,

Michele Margolis, Aidan McGarry, Tonya Putnam, Robert Shapiro, Nick Smith, Jon Tonge,

David Weinberg, Lauren Young, and Adam Ziegfeld for valuable comments and critiques. Along the way, I had helpful conversations with Séverine Autesserre, Courtney Bender, David Buckley,

Al Fang, Nils Gilman, Kimuli Kasara, Isabela Mares, Yotam Margalit, Jeremy Menchik, Tonya

Putnam, Bob Scott, Lee Smithey, Paul Staniland, Alissa Stollwerk, and Dorian Warren. Finally, I would like to thank the discussants, audiences, and organizers of the Politics and Protests

Workshop at CUNY Graduate Center, the Cooperative on Working Class Politics in Northern

v Ireland at Queen’s University Belfast, the Columbia University International Politics Seminar, the Political Science Graduate Student Conference at the University of Pennsylvania, and the annual meetings of the American Political Science Association (2014), Association for the Study of Nationalities (2014), International Studies Association (2014 and 2015), and Midwest Political

Science Association (2014).

I would not have made it through grad school without my friends in the trenches. Mike

Smith, Steven White, and I were forever bonded in the cauldron of our first few years. The coffees, beers, pizzas, and outer borough ethnic food quests made grad school a lot more fun.

They, along with Erica Borghard, Kate Cronin-Furman, Simon Collard-Wexler, Jordan Kyle,

Sara Moller, Liz Sperber, and the rest of the Columbia crew, made grad school not so bad.

My in-laws-to-be, Arnie Eisen and Ace Leveen, provided me with food, shelter, and lots of encouragement. In particular, Arnie’s reading of my theory chapter convinced me that I wasn’t desecrating a century’s worth of thinking about ritual. Nathaniel Eisen bravely edited as much of the manuscript as he could in the final hours before I sent it out. I thank them all for welcoming me with open arms over the last years.

The love from my family has kept me going for so long. My parents, Mitch and Judy

Blake, have always encouraged my learning and exploring. Though I’m sure they would have preferred I did my fieldwork around Los Angeles rather Belfast (not to mention my time in

Liberia and Sierra Leone), they provided nothing but enthusiastic support. This dissertation is dedicated to them. Thanks also to my brothers, Aaron and Josh, for their benign neglect when it comes to my work.

vi Last but certainly not least is Shulie. She read much of this dissertation, listened graciously to me yap about all of it, and compiled the bibliography for me—a task I’ll never be finished thanking her for. She cheered me on when things went right and cheered me up when things went wrong. I would not have finished this without her—though that is the least of the reasons for my love. ! ! ! ! ! ! ! ! ! ! ! ! ! !

vii ! ! !

For my parents ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !

viii ! Chapter One ! Introduction

Visit a Protestant neighborhood, town, or village in Northern Ireland during the spring or summer and there is a good chance you will encounter crowds of people packed onto the sidewalks, clutching cans of beer, bottles of Buckfast Tonic Wine, and Union Jacks—some might even be wearing the flag as a cape. There will be food trucks grilling cheap hamburger patties and deep-frying potatoes and battered fish. And there will be outdoor vendors selling posters, t- shirts, hats, and other souvenirs, all bedecked with the icons of Ulster Protestants: the British flag, the Northern Irish flag, King Billy—William III, the Prince of Orange—riding a white horse into battle. It’s the marching season and everyone is gathered for a parade.

Each year, Protestant organizations perform over 2,500 parades across Northern Ireland.

In these parades, all-male Protestant loyal orders, most famously the Orange Order, join with

Protestant marching bands to march through the streets to display and celebrate their shared culture, history, faith, and politics. Through bodies, banners, flags, and music, loyalist parades represent the Protestant nation and its aspirations.

Loyalist parades are an aural and visual riot. Even from a distance, you can hear the sharp cracks of snare drums, high-pitched blasts of flutes, and thunderous thwacks of bass drums. Once the parade appears, it is visually just as loud. First in line is a marching band’s color party, carrying flags (British, Northern Irish, Scottish, Orange Order, 1912 Ulster Volunteer Force,

Royal Irish Rifles, and more) with military precision. Then come the drummers, with side drums slung over their right shoulders and resting on their left hips. Behind marches the bass drummer,

1 slamming on the large drum attached to his chest with great force. Finally come the rows and rows of fluters, instruments raised high. The band, all in matching uniforms, march in unison, playing crowd favorites: “The Sash My Father Wore,” “Derry Walls,” “No Pope in Rome.”

Following closely behind the band is the loyal order lodge. First comes the lodge’s banner, hanging from long poles held by two members. The large banner presents the lodge’s name along with a rich array of iconography: depictions of pivotal scenes in Irish Protestant history, biblical scenes, or portraits of past kings, queens, and unionist political heroes. The lodge members march behind their banner, dressed in dark suits topped by a collarette that denotes their organization. Behind the lodge marches another band, and then another lodge, and so on.

Loyalist parades share two features with other similar public events around the world.

First, they display the nation. By gathering national emblems and bodies in a single space, loyalist parades, like other nationalist rituals, make visible the “imagined community.”1 In so doing, such performances lubricate the imagination, helping each member feel the “deep, horizontal comradeship” of nationalism.2 As a result, rituals of the nation play a prominent role in the culture and politics of modern life. Across varied forms—parade, rally, wreath laying, anthem singing, flag raising, state funeral—nationalist rituals are ways for the nation and its members to display itself and honor itself; mourn collective tragedies and celebrate collective triumphs; declare its present and imagine its future. They are moments when “societies worship themselves brazenly and openly.”3

1 Benedict Anderson, Imagined Communities: Reflections on the Origin and Spread of Nationalism, rev. ed. (London: Verso, 1991).

2 Ibid., p. 7.

3 Ernest Gellner, Nations and Nationalism (Ithaca: Cornell University Press, 1983), p. 56.

2 Consequently, nationalist rituals can, at times, become hotly contested. For the history of nationalism is a history of rival claims over territory, people, and power. Thus, in the absence of politico-cultural homogeneity, any representation of the nation and its ambitions is a potential object of dispute. This contestation is the second shared feature of loyalist parades. Loyalist parades are not only national rituals, but contentious rituals, symbolic actions that make contested political claims. The Protestant past, present, and future envisioned by parades conflicts with the memory, experience, and aspirations of many Catholics. At particular times and in particular places, loyalist parades trigger a cycle of angry protests and counter-protests, further distancing Catholic from Protestant. On occasion, violence erupts and bricks crisscross through skies darkened by the smoke of shattered petrol-bombs.

This feature places loyalist parades in the ignominious company of other divisive symbolic performances: provocative Hindu processions in India, Confederate flag-flying in the

American south, pilgrimages to certain Shinto shrines in Japan, prayer in the disputed holy places of Israel and Palestine, racialized commemorative ceremonies in South Africa, and others.

While the enactment of most contentious rituals does not produce violence or even protest most of the time, it can still exacerbate group tensions, intensify an “us versus them” mentality, and make conflicts more difficult to resolve.

National rituals, both in their banal and contentious forms, are typically thought of and studied as elite phenomena. Indeed, a rich body of scholarship demonstrates the political power of national rituals (and political rituals, more generally) for activists, movements, and rulers.4 In part, this intuition is correct—typically, it is elites who erect monuments, compose national

4 For example, David I. Kertzer, Rituals, Politics, and Power (New Haven: Yale University Press, 1988).

3 anthems, and proclaim holidays. But it is, at best, a partial account. The full story must also include how and why ordinary people interact with monuments, sing national anthems, and celebrate holidays—or not. Many national rites require mass participation, that in democracies, at least, is voluntary.5 And, as Hobsbawm reminds us, while nationalism is “constructed essentially from above, … [it] cannot be understood unless also analyzed from below, that is in terms of the assumptions, hopes, needs, longings and interests of ordinary people, which are not necessarily national and still less nationalist.”6 We cannot, therefore, impose theories of elite motivations onto ordinary participants.

If the elite account is insufficient, how should we explain popular participation?

Magnifying the puzzle, nationalist rituals produce social and political outcomes that benefit the nation as a whole, such as increased solidarity or intimidation of a rival ethnic group. Since these benefits do not discriminate between participants and free-riders, there is no direct incentive to participate.7 Peel away the colorful costumes, lively music, boisterous crowds, and other elements of elaborate spectacle, and contentious rituals, at their most basic, are acts of collective claim-making. In divided societies, they drive wedges between groups, fan the flames of suspicion and hostility, and occasionally spark violence—all while facing the collective action problem. Given all this, I ask: Why do people choose to take part? !

5 On non-democracies, see, for instance, Lisa Wedeen, Ambiguities of Domination: Politics, Rhetoric, and Symbols in Contemporary Syria (Chicago: University of Chicago Press, 1999).

6 E. J. Hobsbawm, Nations and Nationalism Since 1780: Programme, Myth, Reality (New York: Cambridge University Press, 1990), p. 10.

7 Mancur Olson, The Logic of Collective Action: Public Goods and the Theory of Groups (Cambridge: Harvard University Press, 1965). For more general analyses of nationalism as a collective action problem, see Michael Hechter, Principles of Group Solidarity (Berkeley: University of California Press, 1987); and Russell Hardin, One for All: The Logic of Group Conflict (Princeton: Princeton University Press, 1995).

4 The Argument and Its Alternatives

The central claim of this dissertation is that to understand loyalist parades and other contentious rituals, we must approach them as rituals—“symbolic behavior that is socially standardized and repetitive.”8 Though describing an act as “ritual” is often a way to ignore it or denigrate it as “merely symbolic,” I will argue that it is exactly the ritual character of these contentious events that holds the key to their explanation. To better understand the features and dynamics of ritual, I draw on research from sociology, anthropology, and religious studies. I integrate their findings with research on ethnic conflict and contentious politics to provide an explanation that accounts for the anomalous features of contentious rituals. Two central arguments emerge. First, rituals present “process-oriented motivations”9 to participants in the form of benefits that are intrinsic to participation itself. This insight explains how participants overcome the dilemma of collective action. Second, rituals are multi-vocal and their meaning is ambiguous. Symbolic ambiguity suggests that interpretations of a ritual can vary across actors so participants need not share the interpretation of their action held by organizers, rivals, or outside observers. Participants, therefore, may not see the ritual as provocative, aggressive, or even contentious—consequences that may turn them off from taking part. This insight explains how participants overcome the dilemma of divisiveness.

Three prominent existing arguments also provide plausible explanations for participation in contentious rituals. First, theories of elite manipulation suggest that people participate because it is in the interest of group elites. But, as mentioned above, this is an insufficient account for

8 Kertzer, Rituals, Politics, and Power, p. 9.

9 Jon Elster, The Cement of Society: A Study of Social Order (New York: Cambridge University Press, 1989), pp. 44-46.

5 voluntary mass action. We cannot assume that just because a contentious ritual is in the interests of elites, mass action will simply appear.10 Second, the logic of ethnic rivalry theories of conflict suggest that contentious national rituals are ways for a group to assert its dominance and denigrate the ethnic other.11 The implication is that individuals are motivated to participate by ethnic pride and solidarity or grievances against the out-group. But, as I will show in Chapter 7, loyalist parade participants are not distinguishable from nonparticipants by their ethnic attitudes.

I find that participants are not ethnic extremists either in terms of pro-Protestant patriotism or anti-Catholic bigotry. Third, the logic of collective action suggests that people participate in contentious rituals when the private benefits outweigh the private costs.12 Factors found to increase benefits and lower costs include selective incentives, social sanctions for noncooperation, and social ties to participants. Once again, the evidence does not support the prominent theory. Chapter 7 shows that participants do not receive selective material rewards and do not have more preexisting social ties than comparable nonparticipants. The evidence on social sanctioning is less clear: some data suggest that participants do experience mild social sanctioning, but other evidence suggests that they do not.

Thus, my argument and evidence challenge influential theories of ethnic conflict that are based on the instrumental logics of ethnic rivalry and collective action. The ethnic rivalry explanation of conflict assumes that individuals act to achieve an outcome oriented at a rival ethnic group, such as provocation or humiliation. The collective action explanation of conflict

10 For an example of this assumption, see Steven I. Wilkinson, Votes and Violence: Electoral Competition and Ethnic Riots in India (New York: Cambridge University Press, 2004), esp. p. 24.

11 See, for instance, Donald L. Horowitz, Ethnic Groups in Conflict, 2nd. ed. (Berkeley: University of California Press, 2000); and Roger D. Petersen, Understanding Ethnic Violence: Fear, Hatred, and Resentment in Twentieth- Century Eastern Europe (New York: Cambridge University Press, 2002).

12 Olson, The Logic of Collective Action.

6 assumes that individuals act to maximize personal material gain. Both theories treat participation in parades as means toward external ends. I, by contrast, show that people approach participation in ritual parades as an end in and of itself. The evidence demonstrates that participants do not view parades instrumentally. Though the ethnic rivalry and collective action approaches to ethnic conflict have proven to be powerful explanatory tools in a wide range of contexts, by attributing externally-oriented, instrumental motives to participants, they do not satisfactorily explain variation among individuals in Northern Ireland.

The empirical claims I just previewed come from an abundance of original data on individual-level participation in Protestant loyalist parades that I collected during eight months of fieldwork in Northern Ireland. These data have three primary sources. First, I conducted semi- structured interviews with 82 participants and nonparticipants. Second, I designed and implemented a randomized household survey with 228 respondents in nine Protestant neighborhoods in Belfast. Third, I recorded ethnographic observations at many parades, protests, public meetings, marching band practices, and other related events. By collecting data from both participants and nonparticipants, my research addresses shortcomings in existing studies of loyalist parades13 and other “cultural forms of political expression,”14 which only sample participants.

My analysis of Protestant elites in Chapter 3 relies on additional data sources. To measure elite attitudes and behavior regarding parades, I conducted two online surveys: of all Protestant elected officials (57 respondents) and of all Protestant clergy members (212 respondents).

13 See James W. McAuley, Jonathan Tonge, and Andrew Mycock, Loyal to the Core?: Orangeism and Britishness in Northern Ireland (Dublin: Irish Academic Press, 2011).

14 Verta Taylor, Leila J. Rupp, and Joshua Gamson, “Performing Protest: Drag Shows as Tactical Repertoires of the Gay and Lesbian Movement,” Research in Social Movements, Conflicts, and Change, Vol. 25 (2004), p. 106.

7 Contributions to Understanding Ethnic Conflict

Studying participation in contentious rituals touches on fundamental questions about ethnic conflict and collective action. First is the question of the origins of ethnic violence.

Despite the recent micro-level turn in the study of violence,15 only scant scholarly attention has been paid to violence’s triggers.16 They are generally left untheorized and are presumed to appear when conditions are ripe. But if a trigger requires the mobilization of participants, we cannot assume that it will just appear. Mass participation in triggering events is especially puzzling given that if the trigger succeeds, ordinary people are most likely to suffer in the ensuing violence. So a full description of the dynamics of intergroup conflict must include an explanation of the events that ignite violence.17 This dissertation provides a bottom-up account of a major trigger of communal violence in post-conflict Northern Ireland.

A second contribution of this dissertation is asking the question, why do people participate in ethnic conflict short of violence? The literature on ethnic conflict tends to conflate conflict and violence, even though they are conceptually distinct. When “pursued within the institutionalized channels of the polity,” ethnic conflict is a “regular feature of pluralistic

15 See Stathis N. Kalyvas, “The Ontology of ‘Political Violence’: Action and Identity in Civil Wars,” Perspectives on Politics, Vol. 1, No. 3 (September 2003), pp. 475-494.

16 Political science exceptions include Donald L. Horowitz, The Deadly Ethnic Riot (Berkeley: University of California Press, 2001), ch. 8, “The Occasions for Violence”; and Marc Howard Ross, Cultural Contestation in Ethnic Conflict (New York: Cambridge University Press, 2007).

17 For a theoretical approach to historical events, see William H. Sewell, “Political Events as Structural Transformations: Inventing Revolution at the Bastille,” Theory and Society, Vol. 25, No. 6 (December 1996), pp. 841-881. See also, Rogers Brubaker, Nationalism Reframed: Nationhood and the National Question in the New Europe (New York: Cambridge University Press, 1996), pp. 18-20; Mark R. Beissinger, Nationalist Mobilization and the Collapse of the Soviet State (New York: Cambridge University Press, 2002), pp. 11, 14-27; and Geneviève Zubrzycki, The Crosses of Auschwitz: Nationalism and Religion in Post-Communist Poland (Chicago: University of Chicago Press, 2006). Though I do not answer Brubaker’s call for “eventful analyses of nationness and nationalism” (p. 19), this dissertation highlights the importance of specific triggering events in understanding intergroup violence.

8 democracies” that reflects differing political interests and preferences among ethnic groups.18

Even when ethnic disagreements take “the form of strikes and non-violent demonstrations on the streets, it is an expression of conflict to be sure, but it is not a form of ethnic violence.”19

Political mobilization along ethnic lines, therefore, need not end in violence. As Horowitz reminds us: “Even in the most severely divided society, ties of blood do not lead ineluctably to rivers of blood.”20

Studies of participation in ethnic conflict have focused almost exclusively on episodes of violence, but participating in a rebel group,21 riot,22 or genocide23 is not the same as participating an act of nonviolent conflict. Contentious rituals and other episodes of conflict may cause offense, raise tensions, or even precipitate instability and violence, but participation does not carry the high risks to the individual of an act of violence. Neither does participation in a contentious ritual violate widely-held norms against committing harm to other people or their property. On the benefits side, contentious rituals do not provide the opportunities for looting that many have found to incentivize violence.24 Given these dissimilar incentive structures, we should expect people choose to participate in conflict and violence for different reasons, and that the

18 Ashutosh Varshney, “Ethnicity and Ethnic Conflict,” in Carles Boix and Susan Stokes, eds., The Oxford Handbook of Comparative Politics (New York: Oxford University Press, 2007), pp. 279, 278.

19 Ibid., p. 279.

20 Horowitz, Ethnic Groups in Conflict, p. 684. See also James D. Fearon and David D. Laitin, “Explaining Interethnic Cooperation,” American Political Science Review, Vol. 90, No. 4 (December 1996), pp. 715-35.

21 Macartan Humphrey and Jeremy M. Weinstein, “Who Fights? The Determinants of Participation in Civil War,” American Journal of Political Science, Vol. 52, No. 2 (April 2008), pp. 436-455.

22 Alexandra Scacco, “Who Riots? Explaining Individual Participation in Ethnic Violence,” Ph.D. dissertation, Columbia University, 2010.

23 Scott Straus, The Order of Genocide: Race, Power, and War in Rwanda (Ithaca: Cornell University Press, 2006).

24 For example, Edward C. Banfield, The Unheavenly City Revisited (Boston: Little, Brown and Company, 1974); and Paul Collier and Anke Hoeffler, “Greed and Grievance in Civil War,” Oxford Economic Papers, Vol. 56 (2004), pp. 563-595.

9 findings from the study of violence may not necessarily apply to conflict in general.

The third contribution, then, is to offer an explanation for participation in acts of nonviolent ethnic conflict and mobilization. In particular, by approaching participation through the lens of ritual, I provide a proactive perspective on expressive motivation and behavior, which are generally used as a residual explanation when instrumental accounts fail.25 One of the circumstances where instrumental accounts have proven inadequate and arguments about expressive benefits have been developed are violent situations.26 But the instability of such circumstances is not conducive to the assumptions of rational choice theory. In contrast, the stability and predictability of parades provide an easy test for instrumental rational choice theories. Thus the failure of the instrumental accounts to explain participation in parades presents a challenge to these theories’ universal applicability. !

Case Selection: Loyalist Parades in Northern Ireland

The empirical focus of this dissertation is loyalist parading in Northern Ireland. Why the focus on something so seemingly narrow? First, parades are a major political problem and source of disruption in Northern Ireland. Though conflicts over parades hopefully reached their nadir in the mid-1990s, particularly around the Drumcree Church parade in Portadown, old disputes linger and new ones have appeared in recent years. As I write, in April 2015, Protestants continue to protest nightly in north Belfast over the decision to restrict the final leg of the Twelfth of July parade in 2013. At the elite level, political leaders failed to reach an agreement over parades,

25 Alexander A. Schuessler, A Logic of Expressive Choice (Princeton: Princeton University Press, 2000).

26 For instance, Elisabeth Jean Wood, Insurgent Collective Action and Civil War in El Salvador (New York: Cambridge University Press, 2003); and Rachel L. Einwohner, “Opportunity, Honor, and Action in the Warsaw Ghetto Uprising of 1943,” American Journal of Sociology, Vol. 109, No. 3 (November 2003), pp. 650-75.

10 flags, and dealing with the legacies of the Troubles during negotiations in late 2013, and though a new round of talks began in late 2014, they appear slow-going and the outcome is uncertain.

Such disputes over parades regularly test the robustness of the Northern Irish peace agreements, widely considered a model case of negotiated settlement to civil war.27 Parades remain a key unresolved issue and point to the limitations of the peace process, particularly its mismanagement of the politics of culture. As such, the study of parading in Northern Ireland is a study of the continuation of conflict after war. In this dissertation, I study how and why ordinary people choose to contest ethnic relations in the aftermath of ethnic violence and challenge the resilience of peace. This is fundamental to understanding continuing conflict in Northern Ireland and other “post-violence societies.”28 Correctly or not, Northern Ireland’s precarious peace often appears at the mercy of a minority of men in dark suits and orange collarettes, marching up the road to the sound of piercing flutes and booming drums.

Second, loyalist parades are a clear example of a contentious ritual. Contentious rituals— symbolic actions that make contested political claims—are theoretically and empirically compelling. Theoretically, their ritual aspects—such as repetition and goal demotion—present difficulties for prominent existing theories of collective action and conflict. To address these anomalous features, I offer a novel theory of participation that builds on findings from the multi- disciplinary study of rituals. Further, studying contentious rituals enriches our conception of political participation and political activism. Political scientists tend to focus on a narrow range of political activities, so we only see a slice of the contemporary repertoire of contention. This

27 See John McGarry and Brendan O’Leary, The Northern Ireland Conflict: Consociational Engagements (New York: Oxford University Press, 2004); and Timothy J. White, ed., Lessons from the Northern Ireland Peace Process (Madison: University of Wisconsin Press, 2013).

28 John D. Brewer, Peace Processes: A Sociological Approach (Malden, Mass.: Polity, 2010), pp. 16-28.

11 limited focus means that the bulk of research on participation focuses on certain tactics.29

This limitation in the literature means that a vast array of political actions have gone understudied. In particular, political scientists’ ritual-blindness has left us little systematic research on political and politicized rituals. (To my knowledge, this dissertation is the first to systematically study voluntary individual participation in a contentious ritual.) Across a range of identity practices imbued with political meaning and engaging in political claim-making—from

Hindu religious processions through Muslim neighborhoods in India to open-air Papal masses in communist Poland to Shias marking Ashura in Iraq and Lebanon—we have limited empirical knowledge about who participates, why they participate, or how they view the political aspects versus the cultural aspects of their action.

Through the in depth study of parading in Northern Ireland, I am able to answer these questions for one case of contentious ritual. Comparative studies of contentious rituals in multiple countries would be very welcome, but my interest in the individual participants necessitated a single case. Whether my findings apply to other contexts is an open question. Even within the case of Northern Ireland, I limit my research to the Protestant side and Catholic voices are absent throughout the dissertation.30 If this research aimed to explain the dynamics of parades and protests in Northern Ireland, the absence would be inexcusable. But my aim is to explain participation in a contentious ritual, so a focus on the community that performs is justified. I explore variation within one community rather than between the two. Catholics remain an

29 See similar critiques by Fredrick C. Harris and Daniel Gillion, “Expanding the Possibilities: Reconceptualizing Political Participation as a Toolbox,” in Jan Leighley, ed. The Oxford Handbook of American Elections and Political Behavior (New York: Oxford University Press, 2010), pp. 144-61.

30 See the similar discussion in Lee A. Smithey, Unionists, Loyalists, and Conflict Transformation in Northern Ireland (New York: Oxford University Press, 2011), p. 53.

12 important part of the story, but their role is largely filtered through the perceptions of Protestants.

The Route of the Dissertation

In the chapters that follow, I explain why people choose to take part in contentious, ritual parades in Northern Ireland. I will draw on a range of social science literatures to craft my argument and employ diverse types of data and social science methods to test and defend it. But first, I provide the necessary background to the case. Chapter 2 introduces the historical and political contexts of Protestant parading in Northern Ireland. I describe the relevant political situation since the 1998 Belfast/Good Friday Agreement. Then I describe the current parading scene, introducing the major players and the source of conflicts between Protestants and

Catholics. Finally, I trace the history of parades in the north of Ireland since the eighteenth century. I highlight the long relationship between parades, power, and sectarian conflict.

In Chapter 3, I use loyalist parades to rethink the relationship between ethnic conflict and ethnic elites. The leading explanation for contentious rituals is that they are the result of elite planning and manipulation. I argue that this model rest on an inadequate conception of elites that ignores many actors who hold considerable authority in an ethnic community. I present a more inclusive model of elites and show that in the Northern Irish Protestant community it includes political leaders, religious leaders, and ex-paramilitary combatants. Using original survey data, I show that Protestant elites hold diverse opinions about parades, with politicians generally taking more hardline positions and clergy taking more conciliatory positions. Given this split opinion among elites, we need a more satisfying model of mass participation in parades.

I provide this explanation in Chapter 4. As summarized above, I draw on the multi-

13 disciplinary study of ritual to argue, first, that rituals provide process-oriented, intrinsic benefits to participants, and, second, that rituals are multi-vocal and are characterized by ambiguity of meaning. Process benefits help explain why people participate at all; ritual ambiguity helps explain why people participate in rituals with divisive consequences. This chapter also reviews the role of rituals in politics generally and provides a sustained discussion of contentious rituals.

The next two chapters evaluate my argument empirically. Chapter 5 argues that parade participants approach their actions non-instrumentally. In particular, I find evidence for five process-oriented reasons for action: expressing collective identity, commemoration, tradition, defiance, and the pleasures of participation. The benefits motivating these reasons are intrinsic to the very process of participating collectively in parades. Using survey, ethnographic, and, especially, interview data, I explore each reason in detail. I also find one significant instrumental reason for parading: to send a message to in-group and out-group audiences. Some participants do aim to communicate solidarity to fellow Protestants and opposition to Catholic nationalists, but I show that they are secondary reasons to parade. The preponderance of the evidence suggests that participants are most interested in the act of participation, not its outcomes.

Chapter 6 carries this argument forward by exploring how participants seem to ignore the divisive consequences of their parades. I show that despite the widely recognized political causes and consequences of parades, participants understand parades as anti-political—that they transcend politics and exist outside of it. I call this position the paradox of anti-politics. Using interview data, I first sketch the extent of the paradox among participants. I demonstrate their belief that parades are about culture and therefore cannot be about politics. Second, I reveal the political power of the claim of anti-politics. Being outside of politics is useful for participants as

14 it lets them try to shift debates away from criticism and compromise. Third, I argue that the paradox is sustained by the ritual character of parades. Intrinsic benefits mean that participants need not have political motivations and symbolic ambiguity allows participants to maintain their own interpretations of parades. If the parades lacked these features, the paradox might collapse.

Chapter 7 tests two prominent alternative explanations for parade participation: that participation follows the instrumental logics of ethnic rivalry or collective action. I use survey and interview data from participants and comparable nonparticipants to show that the differences between the two groups predicted by the alternative explanations are not present. Where the ethnic rivalry approach predicts participants to be more pro-Protestant and more anti-Catholic than their non-parading neighbors, I find no systematic differences. Where the collective action approach predicts participants to receive selective material benefits and have more pre-existing social ties than nonparticipants, I find that participants actually pay to parade and that the two groups have similar social networks. The evidence is mixed on collective action’s prediction of social sanctioning, however. The survey data show some support it, but the qualitative data suggest that people are not sanctioned to participate.

Chapter 8 concludes by summarizing the dissertation’s argument and main findings. I then briefly consider the similar cases of contentious processions in India and Israel. Harnessing these comparative cases, I end by offering some tentative thoughts about the general nature of contentious rituals.

15

Chapter Two Identity on the March in Northern Ireland ! And I shall have some peace there, for peace comes dropping slow, Dropping from the veils of the morning to where the cricket sings. — W.B. Yeats1 ! In this chapter, I introduce the empirical setting of the dissertation. I first describe the political context of Northern Ireland since the end of the Troubles, highlighting the ways in which the conflict between Protestants and Catholics persists. I show that the years since the

1998 Belfast/Good Friday Agreement (“the Agreement”) have confirmed Yeats’s observation that

“peace comes dropping slow.”2 Second, I detail the landscape of contemporary parading. I demonstrate the prominence of parading in the Protestant community, explain the sources of sectarian parading disputes, and describe the significant players. Third, I review the history of parading in Ireland from the eighteenth century to the present. This history shows how Protestant parades have long been intertwined with sectarian and class politics in the north of Ireland. As a result, sectarian conflict and violence has been a consistent feature of parades since their birth. It is into this turbulent history and political dynamic that participants choose to parade.

Northern Ireland since the Agreement: “Peace without Reconciliation”

The Troubles, as Northern Ireland’s civil war is known, came to a formal end with the

1 W.B. Yeats, “The Lake Isle of Innisfree,” in Richard J. Finneran, ed., The Collected Poems of W.B. Yeats, 2nd ed. (New York: Scribner, 1996), p. 39.

2 Even the name of the peace agreement is disputed: unionists tend toward Belfast, nationalists tend toward Good Friday. Following Brendan O’Leary, “The Nature of the Agreement,” Fordham International Law Journal, Vol. 22, No. 4 (1998), p. 1629, I will call it simply the Agreement.

16 signing of the Agreement in April 1998. The thirty years of violence between and among republican paramilitaries seeking a united Ireland, loyalist paramilitaries seeking to remain in the

United Kingdom, and British security forces killed over 3,700 people and injured at least

40,000.3 In the years since the Agreement, violence has declined dramatically, but conflict between the two communities continues. “Post-conflict” Northern Ireland has been characterized by “peace without reconciliation.”4 The vast majority of students are segregated in Protestant or

Catholics schools, with only 6.5 percent attending formally integrated schools.5 Housing segregation remains high.6 Low-level violence and fear are endemic in interface communities, the areas where Protestant and Catholic neighborhoods meet.7 As a result, the number of “peace walls”—physical barriers separating neighborhoods—has increased since the Agreement.8

Unsurprisingly, there is little contact across the walls. In a survey of residents of interface communities in Belfast, Shirlow finds that under 10 percent work in the other community’s area;

78 percent avoid public facilities simply because they are in the other community’s territory; 88 percent would not go to the other community at night; and 48 percent would not go during the

3 Marie Smyth and Jennifer Hamilton, “The Human Costs of the Troubles,” in Owen Hargie and David Dickson, eds., Researching the Troubles: Social Science Perspectives on the Northern Ireland Conflict (Edinburgh: Mainstream, 2003), pp. 18-19.

4 Paul Nolan, Northern Ireland Peace Monitoring Report, Number 3 (Belfast: Community Relations Council, 2014), p. 11.

5 Ibid., p. 120.

6 Brendan Murtagh, The Politics of Territory: Policy and Segregation in Northern Ireland (London: Palgrave Macmillan, 2002); Brendan Murtagh, “Ethno-Religious Segregation in Post-Conflict Belfast,” Built Environment, Vol. 37, No. 2 (June 2011), pp. 213-25; and Christopher D. Lloyd and Ian Shuttleworth, “Residential Segregation in Northern Ireland in 2001: Assessing the Value of Exploring Spatial Variations,” Environment and Planning A, Vol. 44 (2012), pp. 52-67.

7 Peter Shirlow and Brendan Murtagh, Belfast: Segregation, Violence, and City (London: Pluto, 2006).

8 Nolan, Northern Ireland Peace Monitoring Report, Number 3, pp. 67-70; and Belfast Interface Project, Belfast Interfaces: Security Barriers and Defensive Use of Space (Belfast: Belfast Interface Project, 2012).

17 day. This segregation is driven primarily by fear of the other community and is magnified among young people.9

Among working class Protestants, this reigning sense of fear, along with little opportunity for economic advancement, has led to disenchantment with the peace process. The promised peace dividend never appeared for many people and in fact many Protestants perceive that on many dimensions, conditions have deteriorated since the violence ended. There is a general feeling among Protestants that the peace process has benefited Catholics more than them—a belief that has steadily increased since 1998.10 Through a zero-sum, sectarian prism, there is truth to this. The peace process transformed Northern Ireland from a social, economic, and political system designed to privilege Protestants over Catholics to one premised on equal opportunity.

On top of political, economic, and security reforms to redress anti-Catholic discrimination, Protestants fear losing their demographic advantage. The borders of the province were carved out of northeast Ireland in order to create a territory with a large Protestant majority.

However, the original two-thirds Protestant majority has now declined to near parity: 48.4 percent Protestant and 45.1 percent Catholic, according the 2011 census.11 Northern Ireland has always been characterized by a “double-minority” situation—Catholics are a minority in

Northern Ireland, Protestant are a minority on the island of Ireland—but that is set to change in

9 Peter Shirlow, “Ethno-Sectarianism and the Reproduction of Fear in Belfast,” Capital & Class, Vol. 27, No. 2 (Summer 2003), pp. 85-89.

10 Lee A. Smithey, Unionists, Loyalists, and Conflict Transformation in Northern Ireland (New York: Oxford University Press, 2011), p. 68; and Henry Patterson, “Unionism after Good Friday and St Andrews,” The Political Quarterly, Vol. 83, No. 2 (April-June 2012), p. 249.

11 Eric Kaufmann, “Demographic Change and Conflict in Northern Ireland: Reconciling Qualitative and Quantitative Evidence,” p. 371; and Nolan, Northern Ireland Peace Monitoring Report, Number 3, p. 21.

18 the coming years.12

The shifting demographics matter because the ultimate answer to the sovereignty question—the keystone political issue in the country—lies in the “principle of consent.” Under the terms of the Agreement, a united Ireland “is subject to the… consent of a majority of the people of Northern Ireland.”13 Historically, majority rule meant a Protestant veto over unification, but with a coming Catholic majority this may change. Nevertheless, the union with the United Kingdom currently appears firm—in large part because of growing Catholic acceptance of the constitutional status quo. A 2013 survey shows that 52 percent of Catholics in

Northern Ireland wish to remain in the United Kingdom and only 28 percent desire a united

Ireland. Furthermore, a slim majority of Catholics now view themselves as having at least some

British identity, leaving only 48 percent identifying as exclusively Irish.14 Perhaps most important to the recent stability is “the enthusiastic administration” of Northern Ireland by Sinn

Féin and others “who spent thirty years trying to destroy it.”15

Despite this newfound security, many Protestants fear for their future. At present, these anxieties are focused on a perceived “culture war” against Protestantism. Many believe that

Protestant culture and way of life is under threat from politically assertive republicanism. As a recent report summarizes, “The focus of concern is no longer about Northern Ireland being taken

12 John Whyte, Interpreting Northern Ireland (Oxford: Clarendon, 1990), pp. 100-101.

13 Agreement Reached in the Multi-Party Negotiations, April 10, 1998, Section 2, “Constitutional Issues,” I.ii. Unionists only agreed to ratify the Agreement because of it enshrined majority consent. John McGarry and Brendan O’Leary, “Consociational Theory, Northern Ireland’s Conflict, and its Agreement. Part 1: What Consociationalists Can Learn from Northern Ireland,” Government and Opposition, Vol. 41, No. 1 (Winter 2006), p. 57.

14 ARK, Northern Ireland Life and Times Survey 2013 (Belfast: ARK, 2014). Distributed by ARK. Available at: www.ark.ac.uk/nilt/2013. Accessed August 15, 2014.

15 Patterson, “Unionism After Good Friday and St Andrews,” p. 254.

19 out of Britain, but of ‘Britishness’ being taken out of Northern Ireland.”16 For example, in

December 2012, the Belfast City Council voted to stop flying the Union Flag on the city hall every day of the year and instead fly it on 18 designated days, in line with the policy for government buildings in Britain. This move, supported by the nationalist parties and the nonsectarian Alliance Party, was seen by many Protestants as an assault on their rights and values and another example of Catholic concerns being privileged over theirs. The change in flag policy sparked four months of protests, riots, and violent attacks by Protestants—including burning down an Alliance Party office, attempts to burn others, and death threats to several Alliance elected officials—which led to a “marked deterioration in community relations.”17

There is also a long-standing historical dimension to the Protestant narrative of marginalization and loss. Protestants have long seen their position in Ireland as precarious. Since the sixteenth century plantation, Protestants have been a religious minority in an overwhelmingly

Catholic society. Sporadic episodes of sectarian violence, such as the massacres of Protestants by

Catholic rebels in 1641, confirmed the sense of Catholic threat and promoted a siege mentality that has been held by many Protestants up until this day.18 An independent or autonomous

Ireland, Protestants feared, would curtail religious liberties and subject them to religious persecution. Preventing such an outcome was paramount. The community’s security rested on local Protestant political dominance and the union with the United Kingdom. Any gain by

16 Nolan, Northern Ireland Peace Monitoring Report, Number 3, p. 12. Also, Patterson, “Unionism After Good Friday and St Andrews,” p. 254.

17 Paul Nolan, Dominic Bryan, Clare Dwyer, Katy Hayward, Katy Radford, and Peter Shirlow, The Flag Dispute: Anatomy of a Protest (Belfast: Queen’s University Belfast, 2014), p. 11.

18 Smithey, Unionists, Loyalists, and Conflict Transformation, p. 54. Historians debate the historical validity of the massacres. Jonathan Bardon, A History of Ulster (Belfast: Blackstaff, 1992), pp. 137-139; and Sean Farrell, Rituals and Riots: Sectarian Violence and Political Culture in Ulster, 1784–1886 (Lexington: University Press of Kentucky, 2000), p. 195, note 26.

20 Catholics that could pose a threat to either position was therefore interpreted as a threat to the community’s survival. As a result, Protestant politics, especially since the nineteenth century,

“has been primarily defensive and conservative.”19 The general tenor of modern Protestant politics is summarized in oft-repeated slogans: “no surrender!” “not one inch!” “Ulster says no!” !

A Note on Terminology: Protestants and Catholics

As the previous discussion suggests, Northern Irish society is divided into two ethnic groups: Protestants and Catholics.20 Though there are many divisions within each group, “it is hard to deny the reality of two readily discernible blocs.”21 These two categories contain various dimensions of difference which largely overlap and reinforce each other. Ruane and Todd argue that dimensions of difference premised on religion (Protestant/Catholic), ethnicity (British/Irish), and colonialism (settler/native) have produced two “ideological articulations”: an older distinction between civility and barbarism and a modern distinction between unionism and nationalism.22 The overlap of these cleavages is the root of the division.

Within each ethnic group today, there are two dominant political and ideological strands: unionism and loyalism among Protestants and nationalism and republicanism among Catholics.

Unionism refers to the belief, held mainly by Protestants, that Northern Ireland should remain

19 Smithey, Unionists, Loyalists, and Conflict Transformation, p. 60.

20 I follow Horowitz’s “inclusive conception of ethnicity” where “groups are defined by ascriptive differences, whether the indicium of group identity is color, appearance, language, religion, some other indicator of common origin, or some combination thereof.” Donald L. Horowitz, Ethnic Groups in Conflict, 2nd. ed. (Berkeley: University of California Press, 2000), pp. 17-18.

21 Smithey, Unionists, Loyalists, and Conflict Transformation, p. 56.

22 Joseph Ruane and Jennifer Todd, The Dynamics of Conflict in Northern Ireland: Power, Conflict, and Emancipation (New York: Cambridge University Press, 1996), p. 11.

21 part of the United Kingdom, while nationalism refers to the belief, held mainly by Catholics, that

Northern Ireland should unite with the Republic of Ireland. The Democratic Unionist Party

(DUP) and Ulster Unionist Party (UUP) are the primary political manifestations of unionism; the

Social Democratic and Labour Party is the primary political manifestation of nationalism.

Loyalism is the more hardline version of unionism. It is held mainly by the working class, emphasizes Protestant culture and identity, and is often associated with support for paramilitary violence. Organizationally, loyalism is represented by the Progressive Unionist Party and the

Ulster Defence Association (UDA) and Ulster Volunteer Force (UVF) paramilitaries. Its nationalist counterpart is republicanism, which is represented in Sinn Féin and the Provisional

Irish Republican Army (PIRA) paramilitary.23

Throughout the dissertation, I use “Protestant” and “Catholic” to refer to the two main ethnic communities in Northern Ireland. The terms do not imply religious beliefs or practices, unless otherwise specified. Since I am writing primarily about ethnic relations, I use those terms rather than unionist/loyalist or nationalist/republican, which are political designations, not ascriptive identity groups. !

Parades in Northern Ireland

Each year in Northern Ireland, Protestant organizations perform over 2,500 parades to display their allegiance to the Protestant faith, the Protestant people of Ulster, and the constitutional union between Northern Ireland and Great Britain. During parades, participants

23 Ibid., esp. pp. 84-115; John McGarry and Brendan O’Leary, Explaining Northern Ireland: Broken Images (Oxford: Basil Blackwell, 1995), esp. pp. 13-61, 92-137; and Smithey, Unionists, Loyalists, and Conflict Transformation, pp. 23-24.

22 use their bodies, uniforms, flags, banners, and music to demonstrate their culture and their views to the world. The parades are often festive events along routes lined with cheering fans waving flags and happily singing along to the tunes. Most parades take place in the spring and summer, with the pinnacle of the parading season on the Twelfth of July. On that day, tens of thousands of members of the Orange Order and marching bands parade past throngs of supporters in cities and towns around Northern Ireland to celebrate and commemorate the military victory of the

Protestant King William III (of Orange) over the Catholic King James II at the Battle of the

Boyne in 1690.

However, in Northern Ireland’s divided social landscape, not all citizens view the parades favorably. Many, if not most Catholics, see parades as anti-Catholic triumphalism and provocative “carnival[s] of hate.”24 They associate the loyal orders with Protestant domination and the marching bands, particularly self-styled “blood and thunder,” or “fuck the Pope,” bands, with loyalist paramilitaries. The Twelfth of July, to take the most well-known example, marks a great victory for Protestants, but for Catholics, the battle marked the start of a long era of subjugation to Protestant supremacy. The Twelfth’s content and form symbolize the subsequent centuries of Protestant hegemony in Ireland.

Consequently, groups of Catholic residents often protest parades, causing the police to occasionally block the marchers from entering certain streets. Disputes over parades increase communal tension, harm the political peace process, and undermine grassroots peace-building on a regular basis. These disputes occasionally precipitate violence, including significant riots in

2012 and 2013. The seemingly endless cycles of parades, protests, and violence embody what

24 “A Carnival of Hate,” An Phoblacht/Republican News (Dublin), July 4, 1996, quoted in Martyn Frampton, The Long March: The Political Strategy of Sinn Féin, 1981-2007 (London: Palgrave Macmillan, 2000), p. 126.

23 O’Leary and McGarry call Northern Ireland’s “politics of antagonism.”25

Though very few parades each year are protested—and even fewer turn violent—all of them are intimately political. Every parade makes a claim about the central question of politics: who should rule? Their answer—the United Kingdom—clashes with the aspirations of many

Catholics who seek a united Ireland, free from British dominion. Though the claim is often made obliquely through the use of flags, music, and other symbols, it touches on the question that has dominated politics in Northern Ireland—and before that, all of Ireland—for over a century.26 !

Loyalist Parading Today

Parading remains a major communal activity among Protestants. From April 1, 2013 to

March 31, 2014, there were 4,665 parades in Northern Ireland, 59 percent of which were organized by the “broad Unionist tradition” (2,766 parades). By contrast, only 3 percent were organized by the nationalist community (119 parades).27 As Figure 2.1 demonstrates, the number of Protestant parades, which have always been the vast majority, has steadily increased since the mid-1980s, when systematic counting began.

Estimates of the number of participants are less exact, but it is a significant number of people. There are 35,758 men in the Orange Order, 9,000 men in the Apprentice Boys of Derry,

25 Brendan O’Leary and John McGarry, The Politics of Antagonism (London: Athlone, 1993).

26 See, for instance, Richard Bourke, Peace in Ireland: The War of Ideas (London: Pimlico, 2003); and Alvin Jackson, Home Rule: An Irish History, 1800-2000 (New York: Oxford University Press, 2003).

27 The remaining 38 percent were classified as “other,” which includes “charity, civic, rural and sporting events, as well as church parades.” Parades Commission for Northern Ireland, Annual Report and Financial Statements for the Year Ended 31 March 2014 (Norwich, UK: The Stationary Office, 2014), pp. 7-8.

24 and 29,000 people in marching bands.28 Assuming no overlapping membership (which is not true, and therefore these estimates are the upper bound), there are 73,758 parade participants. Out of a population in Northern Ireland of 1,810,863, paraders represent 4 percent.29 Among

Protestants (875,717 total), the figure is 8 percent. And among Protestant men (424,768 total), who make up the vast majority of parade participants, the figure is 17 percent.30 Even if we assume that one-quarter of participants are members of two or more organizations (and were therefore just double-counted), parade participants would still amount to 6 percent of all

Protestants and 13 percent of all Protestant men.31

Compared with other forms of collective action in Northern Ireland, these participation rates are notable. According to a 2005 survey, in the previous two or three years only 2 percent of respondents took part in a strike, 5 percent attended a political meeting, 5 percent took part in a

“demonstration, picket or march,” and 32 percent signed a petition.32 In the Northern Irish

28 James W. McAuley, Jonathan Tonge, and Andrew Mycock, Loyal to the Core?: Orangeism and Britishness in Northern Ireland (Dublin: Irish Academic Press, 2011), p. 126; Interview with Apprentice Boys senior leader, Derry/ Londonderry; Department of Culture, Arts and Leisure, Marching Bands in Northern Ireland (Belfast: Department of Culture, Arts and Leisure, 2012), p. 6. According to estimates in Northern Ireland Youth Forum, Sons of Ulster: Exploring Loyalist Band Members Attitudes Towards Culture, Identity, and Heritage (Belfast: NIYF, 2013), pp. 4-5, there are 30,000 band members. These figures may include international membership, but the vast majority are in Northern Ireland. Also there is some overlap in membership in the different groups.

29 All population statistics are from the 2011 Northern Ireland Census, Table DC2115NI: Religion or Religion Brought Up in by Age by Sex, available at http://www.ninis2.nisra.gov.uk/Download/Census%202011_Excel/2011/ DC2117NI.xls. Accessed October 29, 2013.

30 Women do participate in parades, but participants are predominantly male. On the role of women in Northern Irish parades, see Linda Racioppi and Katherine O’Sullivan See, “Ulstermen and Loyalist Ladies on Parade: Gendering Unionism in Northern Ireland,” International Feminist Journal of Politics, Vol. 2, No. 1 (Spring 2000), pp. 1-29; and Katy Radford, “Drum Rolls and Gender Roles in Protestant Marching Bands in Belfast,” British Journal of Ethnomusicology, Vol. 10, No. 2 (2001), pp. 37-59.

31 In my survey, detailed in Chapter 7, nine percent of participants have been in both loyal orders and bands, but not necessarily at the same time. Dominic Bryan, Orange Parades: The Politics of Ritual, Tradition, and Control (London: Pluto, 2000), p. 115, estimates that over half of Apprentice Boys members are also in the Orange Order. Assuming one-quarter overlapping membership is a reasonable compromise.

32 ARK, Northern Ireland Life and Times Survey 2005 (Belfast: ARK, 2006). Distributed by ARK. Available at: www.ark.ac.uk/nilt/2005/Democratic_Participation. Accessed August 15, 2014.

25 context, therefore, parading is a significant public activity that involves a sizable portion of the population. These figures resonate with Tonge, McAuley, and Mycock’s finding that despite a loss in members in past decades, the Orange Order “still more than quadruples the combined memberships of all Northern Ireland’s political parties.”33 ! Figure 2.1: Loyalist and Nationalist Parades, 1985-2013 Figure 2.1: Loyalist and Nationalist Parades, 1985-2013

3000 Loyalist Parades Nationalist Parades 2500 2000 1500 Number of parades Number 1000 500 0

1985 1990 1995 2000 2005 2010 2015

Year Sources: Royal Ulster Constabulary and Police Service of Northern Ireland (1985-2000), Parades Commission (2001-2013) !

Contentious Parades

As already noted, disputes over parades are a major arena in which the conflict between

Protestants and Catholics persists. The annual parading season, which runs from March to

August, reignites debates over a host of parade-related topics. The most controversial parades are known as “contentious parades” and are designated as such by the Parades Commission, the

33 McAuley, Tonge, and Mycock, Loyal to the Core?, p. 1.

26 independent statutory body charged with regulating parades.34 Thus the term is used in Northern

Ireland to describe a subset of parades that are disputed. Note that throughout the dissertation, I use contentious to mean non-institutionalized claim-making that bears on the rights and interests of others.35 By this usage, all parades, whether contested or not, are contentious performances.36

To avoid confusion, what the Parades Commission calls a “contentious parade,” I will call a

“disputed parade,” “contested parade,” or “controversial parade.”

The most common dispute over a specific parade is the route it takes, since some pass by or through Catholic neighborhoods or towns. The paraders insist that marching on their traditional routes is a civil right, but many of the Catholic residents see parading by their homes and churches as triumphalist, hateful, and transgressive. As anthropologist Allen Feldman argues, it “transforms the adjacent community into an involuntary audience and an object of defilement through the aggressive display of political symbols and music.”37 This sectarian divide is clearly evident in public opinion: in a 2010 survey, 72 percent of Catholics and 8 percent of Protestants stated that parades should not be allowed in Catholic neighborhoods, while .3 percent of

Catholics and 48 percent of Protestants stated that the parades should be allowed to march

34 The Parades Commission defines contentious parades as “those that are considered as having the potential of raising concerns and community tensions, and which consequently are considered in more detail by the Parades Commission.” Parades Commission for Northern Ireland, Annual Report and Financial Statements for the Year Ended 31 March 2012 (Norwich, UK: The Stationary Office, 2012), pp. 9. In the 2014 Annual Report, the Parades Commission changed the term to “sensitive parades” (p. 8).

35 Doug McAdam, Sidney Tarrow, and Charles Tilly, Dynamics of Contention (New York: Cambridge University Press, 2001), p. 5.

36 Charles Tilly, Contentious Performances (New York: Cambridge University Press, 2008).

37 Allen Feldman, The Formations of Violence: The Narrative of the Body and Political Terror in Northern Ireland (Chicago: University of Chicago Press, 1991), p. 29.

27 anywhere they like.38

The most well-known parading dispute occurred in the mid-1990s over a route in the small city of Portadown. Starting in 1995, Catholic residents organized to stop the Orange

Order’s annual parade to and from Drumcree Church, which passed through their neighborhood.

In 1996, the police sided with the residents and announced that the parade had to be re-routed elsewhere. But after Protestants across the province responded with protests and violence, the police reversed their decision. The Catholic residents finally won in 1998 and ever since the police have blocked parades from entering the area.

A similar dynamic is ongoing in north Belfast. For years, Catholics from the Ardoyne neighborhood protested and rioted in response to parades on a road adjacent their community.

Against the backdrop of increasingly severe riots following parades in recent years, the failure to reach an agreement between the Orange Order and Ardoyne residents groups led the Parades

Commission to prohibit part of the route on July 12, 2013. The decision was followed by several nights of rioting by Protestants.

Parades can also become disputed because of the behavior of paraders or supporters. For example, parades on Belfast’s lower Ormeau Road became hotly disputed in the mid-1990s after several people at a parade mocked the murder of five Catholics by the UDA as they passed the site of their massacre. And parades passing St. Patrick’s Catholic Church in Belfast became contested after a band was videotaped marching in circles and playing a (plausibly) sectarian song in front of the church during the 2012 Twelfth parade.

38 My calculation from Jonathan Tonge, Bernadette C. Hayes, and Paul Mitchell, Northern Ireland General Election Attitudes Survey, 2010 [computer file]. Colchester, Essex: UK Data Archive [distributor], August 2010. SN: 6553, http://dx.doi.org/10.5255/UKDA-SN-6553-1.

28 In 2013-14, the Parades Commission designated 491 parades as disputed (11 percent of the total).39 Almost all of these were organized by the unionist community (96 percent), and nearly half were organized by two specific groups: the Orange Order in Portadown, who have made a perfunctory effort to march each Sunday since 1998, and the nightly parades in north

Belfast to protest the 2013 decision.40 Though disputed parades are a minority of the total, they dominate media coverage, public discussions, and, in many sectors, public perceptions.

Figure 2.2. Contentious Parades, 1999-2013 Figure 2.2: Contentious Parades, 1999-2013 500 Total Parades Loyalist Parades Nationalist Parades 400 300 200 Contentious parades Contentious 100 0

2000 2002 2004 2006 2008 2010 2012

Year Source: Parades Commission

The most obvious negative consequence of disputed parades is the occurrence of sectarian violence. Although only a small number of parades turn violent, there have been

39 The Parades Commission placed restrictions on 88 percent of them, such changing the route or restrictions on timing, music, size, or the number of supporters. Parades Commission, Annual Report 2014, p. 8.

40 These nightly parades began in 2013, which explains the large spike in Figure 2.2.

29 instances of public disorder each year since the Agreement,41 including significant riots in 2005,

2010, 2011, 2012, and 2013. Between July 2012 and September 2013, for example, 682 police officers were injured at parades and related protests and disorder.42 The general dynamic is that

Catholics riot against police after the parade has passed by with heavy police (and until 2006, military) protection. But Protestants have rioted too, particularly in response to parade routes being blocked by the police.

Even without outbreaks of street violence, there are serious social and political repercussions to disputed parades. While these disputes might seem like a local affair, the symbolism involved makes them a national political issue that impedes reconciliation between

Catholics and Protestants. In the halls of Stormont, the devolved provincial parliament, parties are deadlocked over the issue. On the streets, communal tensions rise during the marching season, with survey evidence showing that fear of the other community and avoidance of their neighborhoods increase during that period.43 Paramilitary groups on both sides exploit the marching season to mobilize new members and assert their strength.44 Further disruptions to normal life are caused by the heavy security at disputed parades. Contested parades often include the presence of police, many in full riot gear, and armored police Land Rovers barricading streets. For residents, parades can mean their neighborhood is seemingly transformed into a

41 There was an annual average of 18 instances of public disorder reported by the police from 1998 to 2009. Police Service of Northern Ireland, “Parades with Disorder, 1990-2009,” spreadsheet in possession of author.

42 Nolan, Northern Ireland Peace Monitoring Report, Number 3, p. 159.

43 Shirlow and Murtagh, Belfast.

44 Paul Nolan, Northern Ireland Peace Monitoring Report, Number 1 (Belfast: Community Relations Council, 2012), pp. 45, 47, 73, 78; and Paul Nolan, Northern Ireland Peace Monitoring Report, Number 2 (Belfast: Community Relations Council, 2013), pp. 62, 168.

30 temporary battle zone. !

Loyal Orders and Marching Bands

Broadly, there are two types of parading organizations: the loyal orders and marching bands. The loyal orders are all-male fraternal orders dedicated to the promotion of Protestant culture and the maintenance of Northern Ireland’s place in the United Kingdom. They are seen by many Protestants as the embodiment of the community’s politics and values. The three main orders are the Orange Order, the Royal Black Institution, and the Apprentice Boys of Derry.45

The Orange Order is the largest and most prominent loyal order. Though its membership and political power have declined since the mid-twentieth century, it still holds outsized influence within unionist politics. A recent survey of members of the DUP, Northern Ireland’s largest political party and the senior party in the power-sharing executive, shows that 35 percent are

Orangemen (the term for members of the Order). The proportion increases to 54 percent among

DUP elected officials.46

The Orange Order is a hierarchical organization with four layers: Grand Orange Lodge of

Ireland, County Lodges, District Lodges, and private lodges.47 Most important to the lives of members are the 1,400 private lodges. Each has its own name and number—for example, Pride of Ballymacarrett Loyal Orange Lodge (LOL) 1075 or Ulster Defenders of the Realm LOL 710

45 There are also smaller orders including the Royal Arch Purple, the Independent Orange Order, and the Orange Order’s all-female sister organization, the Association of Loyal Orangewomen of Ireland.

46 Jonathan Tonge, Máire Braniff, Thomas Hennessey, James W. McAuley, and Sophie A. Whiting, The Democratic Unionist Party: From Protest to Power (Oxford: Oxford University Press, 2014), pp. 139, 149-152.

47 Though Grand Lodge is at the top, it only has limited power over the organization and its members. Bryan, Orange Parades, pp. 97, 101-103. The organization as a whole has only a handful of paid employees and functions almost entirely on the basis of volunteers.

31 —bylaws, and budget. They also almost all have a banner that they carry on parade that represents something of importance to the lodge. Popular images include King William III at the

Battle of the Boyne, the Crown and Bible, biblical scenes, churches and other places, and portraits of deceased members.48 When Orangemen parade, they generally parade with their lodge, unless they hold a leadership role at another level.

The Royal Black Institution is a higher “degree” of the Orange Order. Although it is a separate organization, one must first be a member of the Orange Order to join. It is seen as more focused on religion than politics. This is reflected in its banners, which focus on scenes and images from the Bible.49 The Apprentice Boys of Derry is a separate organization that is dedicated to commemorating the 1688-1689 siege of Derry.50 Thus their main parades take place in the city to mark the beginning and end of the siege. The organization’s respect for the city of

Derry/Londonderry and the reality that the city is majority Catholic has lead the Apprentice Boys toward a more pragmatic strategy dealing with opposition to their parades than the Orange Order.

The second type of parading organization is the marching band. While some of the 600 bands51 are quite musically talented, the majority are known as blood and thunder bands because the bass drummer’s hands often bleed after hours of smashing the drum with all his might.

Needless to say, blood and thunder bands are noted more for their volume than skill. The general trend in loyalist parading since the 1970s has been the rising prominence of bands, particularly

48 Neil Jarman, Material Conflicts: Parades and Visual Displays in Northern Ireland (Oxford: Berg, 1997), pp. 172-184.

49 Ibid., pp. 184-187.

50 Bryan, Orange Parades, pp. 114-115.

51 Department of Culture, Arts and Leisure, Marching Bands in Northern Ireland, p. 12.

32 blood and thunder bands.52 Many people associate bands with the loyalist paramilitaries, and during the Troubles they were often closely linked. The extent of their current ties, however, is debated. Today, although both the loyal orders and bands have a reputation for sectarianism in many quarters, bands are seen as particularly hateful. Band members are viewed by many as

“thugs” out to offend Catholics by flying loyalist paramilitary flags, carrying banners commemorating paramilitary members, and performing paramilitary or anti-Catholic tunes.53

There are important differences between and among the loyal orders and bands, but they share core features, values, and interests,54 so for the purposes of this dissertation I consider them together. Throughout, I refer to parade participation as parading with either a loyal order or a band, without distinction. !

The History of Parading in Ireland

The origins of political parading in Ireland date to the eighteenth century, when Anglican landowners began commemorating the seventeenth century events that secured their ascendancy.

In 1688, King James II, a Catholic, was overthrown in the Glorious Revolution by his daughter,

Mary, and her husband King William III, a Protestant Dutch prince. After seeking refuge in

France, James tried to regain the throne, landing first in Ireland where his Jacobite supporters were already fighting Williamites. Jacobite forces, for instance, besieged the city of Derry, a

52 Bryan, Orange Parades, p. 145.

53 Popular loyalist band tunes include “The Billy Boys,” with the lyrics “we’re up to our necks in Fenian blood / surrender or you’ll die,” and “No Pope of Rome,” with the lyrics “Oh give me a home / Where there’s no Pope of Rome / Where there’s nothing but Protestants stay. … No chapels to sadden my eyes / No nuns and no priests and no Rosary beads / Every day is the Twelfth of July.” Available at: http://rangerspedia.org/index.php/ No_Pope_Of_Rome. Accessed July 7, 2014.

54 Bryan, Orange Parades, p. 116.

33 Williamite stronghold, from April to July 1689, when the Royal Navy arrived.55 William landed in Ireland with a multinational European army in June 1690 and met James by the River Boyne on the east coast of Ireland on July 1, 1690 (July 11 in today’s Gregorian calendar). At the Battle of the Boyne, William defeated James, who retreated to France and abandoned his campaign for the throne. William’s victory (along with his more decisive victory over remaining Jacobite forces at Aughim one year later) assured Protestant dominance in Ireland.56

For the eighteenth century Anglican gentry, major benefactors of the Williamite triumph, the events of William’s victorious military campaign (Derry, Boyne, and Aughrim) needed to be commemorated and celebrated. They created commemorative societies that marked the

Williamite victories with banquets, bonfires, fireworks, and parades throughout the eighteenth century.57 But commemorations really rose to prominence in the century’s last years with the formation of the Orange Order, or Orange Institution, in County Armagh in 1795.58 Formed by members of the Protestant Peep O’Day Boys militia after a skirmish with their Catholic rivals, the Defenders, the Order appeared as a Protestant self-defense movement in a period of severe sectarian clashes in Armagh.59 However, the Orange Order sought to appear respectable—unlike the rowdy Peep O’Day Boys—and took great interest in regalia, icons, and symbols. Inspired by

55 Bardon, History of Ulster, pp. 154-158.

56 Ibid., pp. 161-165.

57 Jacqueline R. Hill, “National Festivals, the State and ‘Protestant Ascendancy’ in Ireland, 1790-1829,” Irish Historical Studies, Vol. 24, No. 93 (May 1984), pp. 30-51; Bryan, Orange Parades, p. 31; and Martyn J. Powell, “Political Toasting in Eighteenth-Century Ireland,” History, Vol. 91, No. 304 (October 2006), p. 512.

58 As late as 1789 and 1790, the centenaries of Derry and the Boyne, respectively, were celebrated by Protestants and Catholics alike as victories for liberty, not for Protestantism. Brian Walker, “1641, 1689, 1690 And All That: The Unionist Sense of History,” The Irish Review, No. 12 (Spring-Summer 1992), p. 57.

59 Peter Gibbon, The Origins of Ulster Unionism: The Formation of Popular Protestant Politics and Ideology in Nineteenth-Century Ireland (Manchester: Manchester University Press, 1975), pp. 22-42; Farrell, Rituals and Riots, pp. 10-46; and Bardon, History of Ulster, pp. 223-227.

34 the reformist Volunteer movement of the late 1770s and 1780s, the Orange Order quickly adopted the public commemorative parade as its principal form of mobilization and political expression.60 Thus the Orange Order held its first Boyne commemorative parade in July 1796.

But even at this first parade, a sectarian clash erupted and one person was killed.61

In their early years of existence, the combustable nature of Orange parades made the state weary of them. But with the 1798 United Irishmen Rebellion, the government and the upper classes warmed to the Orange Order, which was increasingly vocally pro-state and whose members participated in the forces mustered to defeat the rebels.62 “The failed Rising,” comments Dominic Bryan, “completed the process of turning the Orange Institution from a small, geographically and political limited group into a large, nationwide influential organization.”63 This influence would only increase as the Order entered the nineteenth century.

The years after the 1800 Acts of Union which united the Kingdoms of Great Britain and

Ireland were marked by continued sectarian unrest at parades, as the relationship between parades and Protestant-Catholic violence grew.64 The rise in disorderly parades, however, was not linear. Rather, violence and “the strength of the Orange Order throughout the nineteenth

60 Bryan, Orange Parades, pp. 31-3.

61 Ibid., p. 33.

62 In a great irony of history, the United Irishmen, the first major Irish republican movement, was largely Presbyterian. All of the founders of the Society of United Irishmen save one were Presbyterian. A.R. Holmes, “Covenanter Politics: Evangelicalism, Political Liberalism and Ulster Presbyterians, 1798-1914,” English Historical Review, Vol. 125, No. 513 (April 2010), p. 340. See also Ian R. McBride, Scripture Politics: Ulster Presbyterians and Irish Radicalism in Late Eighteenth-Century Ireland (Oxford: Clarendon Press, 1998), pp. 165-206.

63 Bryan, Orange Parades, p. 35. See also J.G. Simms, “Remembering 1690,” Studies: An Irish Quarterly Review, Vol. 63, No. 251 (Autumn 1974), pp. 241-2.

64 Bryan, Orange Parades, pp. 35-36.

35 century waxed and waned in direct proportion to the perceived urgency of the Catholic threat.”65

In the first half of the century, this is exemplified by conflicts of the mid-1820s. As Daniel

O’Connell’s Catholic Association pushed for Catholic emancipation, parades increasingly became “riotous assemblies.”66 In response to Catholic political mobilization, the Orange Order paraded more often and parades were “particularly contentious and a significant marker of local dominance.”67 For example, in 1829, just months after the passage of the emancipation law,

Twelfth parades in Ulster were larger and more widespread than ever before. This aggressive parading sparked riots that killed at least forty people.68 The parades communicated Protestant strength and continued resistance to Catholic rights to fellow Protestants, Catholics, and the

British Parliament. “The message was clear,” argues historian Sean Farrell, “ignorant politicians

[in Westminster] might have changed the law on the statute books, but things would not change on the ground in Ulster.”69

In what was to become a pattern in times of serious unrest, the upper class leadership withdrew from parades as the rowdiness challenged their interest in an orderly society. The lower class mass membership, however, continued parading, even after the Orange Order was outlawed from 1825 to 1828 and parades in Ireland were banned from 1832 to 1844 and 1850 to 1872.70

Orange elites continued to try to maintain control over parades in order to use them to

65 David Hempton and Myrtle Hull, Evangelical Protestantism in Ulster Society 1740-1890 (London: Routledge, 1992), p. 21

66 Jarman, Material Conflicts, p. 53.

67 Ibid.

68 Farrell, Rituals and Riots, pp. 96-99.

69 Ibid., 108

70 Bryan, Orange Parades, p. 37.

36 promote cross-class Protestant unity and loyalty to the state. But this proved a struggle, as the landed gentry abandoned the Orange Order due to the mid-century violence and membership grew among the rising urban working class.71 Urban industrialization led to new levels of sectarian conflict, and Belfast parades were marked by unrest in 1852, 1853, and 1855, eventually exploding in the massive riots of July 1857.72

The flip side of the urban violence was that parades became central sites of democratic party politics.73 The huge crowds became an attractive resource for politicians starting in the

1860s. In the late 1860s, parades proliferated—despite still being proscribed by the Party

Processions Act of 1850 that was not repealed until 1872—and began to be considered

“respectable.”74 Parades became bigger events and were frequented by unionist politicians, who used them as political rallies. The Orange Order’s political power grew.75

This power was used in full to oppose the 1886 Home Rule Bill pushed by Irish nationalists and supported by William Gladstone’s Liberal government.76 Senior unionist politicians now flocked to the Orange Order and regularly appeared at parades. “The Twelfth,” argues Bryan, “started to be used as a symbol of Protestant unity in the cause” of unionism.77 By

71 Ibid., p. 38.

72 Belfast riots in this era were so common and predictable, Gibbon, Origins of Ulster Unionism, p. 86, argues they were “integrated into the local social order.” See his analysis in pp. 67-86; Bryan, Orange Parades, p. 40; Farrell, Rituals and Riots, pp. 143-150; and Frank Wright, Two Lands on One Soil: Ulster Politics before Home Rule (New York: St. Martin’s, 1996), pp. 250-254.

73 Gibbon, Origins of Ulster Unionism, pp. 94-104.

74 Bryan, Orange Parades, p. 44.

75 See the extensive discussion of this period in Wright, Two Lands on One Soil, pp. 284-382.

76 Ibid., p. 479.

77 Bryan, Orange Parades, p. 50.

37 the century’s close, support for the Orange Order and its parades reached new heights among the landed and industrial elites, whose economic interests were closely tied to the union and its guaranteed access to the British Empire’s global markets.78

The Orange Order again flexed its political muscle as it mobilized militant opposition to the third Home Rule Bill in 1912. Many Orangemen were among 237,000 unionist men who signed the Ulster’s Solemn League and Covenant, committing themselves to “all means which may be found necessary to defeat the present conspiracy.”79 This commitment was a serious threat, as several months earlier unionist leaders and prominent Orangemen Edward Carson and

James Craig formed the Ulster Volunteer Force paramilitary. Members of the Orange Order enlisted en masse.80 Tensions mounted and the crisis only ended because of the outbreak of

World War I.

Irish nationalists took advantage of British occupation with the war to push for independence in the failed 1916 Easter Rising and the establishment of a parliament, the Dáil

Eirean, in 1919. Protestants again responded to Catholic mobilization by parading more often and with more vigor. When Ireland finally gained independence in 1922, the six northeastern- most counties of Ulster, where Protestants were a majority, remained part of the United Kingdom as Northern Ireland.81

In the new Northern Ireland, unionist rule was hegemonic and the Orange Order was

78 Ibid., p. 51.

79 Bardon, History of Ulster, p. 437.

80 Bryan, Orange Parades, p. 55.

81 Bardon, History of Ulster, pp. 476-479. The traditional province of Ulster contains nine counties. Northern Ireland is made up of counties Antrim, Armagh, Down, Fermanagh, Londonderry, and Tyrone. The other three Ulster counties (Cavan, Donegal, and Monaghan) had significant Protestant minorities, but remained in Ireland.

38 closely tied to the ruling party. As a result, parades became “rituals of state” that “allowed the full expression of a Protestant state.”82 Throughout the half-century of Ulster Unionist Party rule

(1921-1972), all but three government ministers were members of the Orange Order.83 In this era, membership in the Order was sine qua non for advancement in unionist politics as well as a major route to employment and political patronage. Unionist politicians used parades to legitimize their rule, connect with voters, and call for Protestant unity. They understood their economic, political, cultural, and religious interests to be protected by membership in the United

Kingdom. Maintaining the union, in turn, rested on the cross-class alliance of Protestants.84 Thus throughout the history of Northern Ireland, parades have been used to demonstrate as well as urge unity.

After World War II, parading entered a period of relative calm. Sectarian conflict was largely dormant and the parades were populated by the middle class, who brought a sense of respectability. Calls for Protestant unity continued as the powerful unionist government at

Stormont ruled Northern Ireland for the benefit of the Protestant majority—to the exclusion of meaningful representation of the Catholic minority. The discriminatory regime came under increasing pressure in the late 1960s as Catholics, inspired by the African-American movement, started mobilizing for civil rights. Once again, Protestants used parades as a form of countermobilization: disrupting civil rights marches, provoking sectarian violence, and

82 Bryan, Orange Parades, pp. 60, 66.

83 Ibid., p. 60.

84 Paul Bew, Peter Gibbon, and Henry Patterson, Northern Ireland, 1921-1996: Political Forces and Social Classes (London: Serif, 1996); and Mark McGovern and Peter Shirlow, “Counter-Insurgency, Deindustrialisation and the Political Economy of Ulster Loyalism,” in Peter Shirlow and Mark McGovern, eds., Who Are ‘The People?’: Unionism, Protestantism and Loyalism in Late Twentieth Century Ireland (London: Pluto, 1997), pp. 176-198.

39 preventing unionist leaders from making concessions to Catholics protesters.85 This rise of naked sectarianism and disorder again led the middle and upper classes to abandon parading—a trend that has never reversed.86 Through 1968 and 1969, protests, counter-protests, and sectarian tension increased across Northern Ireland. Then, on August 12, 1969, rioting broke out at an

Apprentice Boys parade in Derry.87 The Battle of the Bogside, as the riot is known, escalated, triggering violence around the country, and is widely considered the start of the Troubles.

The Troubles had three important effects on parading. The imposition of direct rule from

Westminster in 1972 meant that the Orange Order lost its connections to the levers of powers in

Northern Ireland.88 Parades were no longer a “ritual of state” and instead came to represent communal defense. An indication of this shift is the rise of blood and thunder marching bands associated with loyalist paramilitaries in working class neighborhoods.89 Parades became rowdier and lost any remaining “respectability.” Finally, as Catholic families in Protestant majority neighborhoods were intimidated from their homes, and vice-versa, the Troubles created more firmly defined ethnic neighborhoods with recognizable boundaries. Parades thus came to be seen as violations of those boundaries by Catholic residents, and opposition to them

85 Bryan, Orange Parades, pp. 78-87; and Jarman, Material Conflicts, pp. 76-79.

86 Bryan, Orange Parades, pp. 78, 93. For an analysis of recent trends in the Orange Order’s class composition, see Eric P. Kaufmann, The Orange Order: A Contemporary Northern Irish History (New York: Oxford University Press, 2007), pp. 269-274.

87 Bourke, Peace in Ireland, pp. 99-102. Jarman, Material Conflicts, p. 78, argues that though “the cycle of parades in 1968-70 did not cause the Troubles… they proved critical in opening up the fracture zones in Northern Irish life that had been obscured and ignored for so long.”

88 Bryan, Orange Parades, p. 78.

89 Ibid., p. 92.

40 increased.90 Catholic opposition to parades to parades through their neighborhoods really took off in the 1980s and 1990s, leading to the current conflicts described above. !

Conclusion

This chapter has provided an overview of parading politics in Northern Ireland. It laid out social, historical, and political contexts through which parades pass. I showed that parading is a source of long-standing conflict between Protestants and Catholics, which today takes place in a country with a largely stable, but not yet consolidated, ethnic peace. Parades, in fact, are one of the major strains on the peace process. By representing the pride and power of Protestants while denigrating Catholics, parades frequently test the limits of compromise and reconciliation among political leaders and in local communities.

Given these well-known consequences of parades, the rest of the dissertation seeks to explain why people participate in them. The next chapter assess the most prominent explanation in the literature on ethnic conflict: elites plan and promote contentious rituals to enhance their own power. I will show that Protestant elites, when accurately categorized, are split over divisive parades. Accordingly, the question of participation remains. In the chapters that follow, I develop an argument for why people participate in contentious rituals and test it on original individual- level data from paraders and non-participants.

90 Ibid., p. 95.

41 !

Chapter Three The Politics of Provocation: Politicians, Pastors, Paramilitaries, and Parades ! ! This Parades Commission determination creates a serious situation for Northern Ireland. We know, having seen republican threats of violence being rewarded, the conclusion is swiftly drawn that violence pays. We have, for some time, been aware that such an absurd parades determination would bring with it a very real risk of widespread violence and disorder. ! — Joint statement by unionist political leadership1 We would encourage all who participate in public parades and protests to obey the law and avoid any behaviour which is not for the overall good of society and respectfully urge those parading organisations which espouse Christian values to uphold those values through behaviour in the public square which honours the Lord Jesus Christ and reflects the teaching of the Scripture. — Statement by Dr. Michael Barry, Moderator of the General Assembly of the Presbyterian Church of Ireland2! The Twelfth really is the worst day of the year for me — because of all the trouble that happens on the way back from the field. ! — Jackie McDonald, UDA South Belfast Brigadier3 The afternoon of June 21, 2013, was, like many in Belfast, Northern Ireland, gray and wet. But the intermittent rain did not stop the crowds who were out to watch the parade. It was the annual Tour of the North parade, when the Orange Order and fifteen marching bands marched from a Protestant neighborhood in north Belfast to a Protestant neighborhood in west

Belfast. A large crowd of cheering Protestant supporters lined the Crumlin Road as they marched by. But there was a problem. Due to Belfast’s segregated ethnic geography, the parade also

1 Joint statement by unionist political leaders announcing their response to the Parades Commission’s decision to restrict the final leg of the 2014 Twelfth of July parade in north Belfast. Many people at the time read it as a veiled threat of violence. It was signed by the Democratic Unionist Party, Ulster Unionist Party, Traditional Unionist Voice, Progressive Unionist Party, and Ulster Political Research Group (West Belfast). “Unionist Parties Oppose ‘Unjust’ Twelfth Parade Decision: Leaders’ Joint Statement in Full,” Belfast Telegraph, July 3, 2014.

2 “Statement from the Presbyterian Moderator Re Bonfires and Parades,” Presbyterian Church in Ireland, July 7, 2014. Available at: https://www.presbyterianireland.org/News/Article/July-2014/Statement-from-the-Presbyterian- Moderator-re-bonfi. Accessed March 16, 2015.

3 Ivan Little, “Twelfth Is the Worst Day of My Year, Says Top UDA Boss,” Belfast Telegraph, July 9, 2012.

42 passed by a Catholic neighborhood, where residents view the parade as hateful bravado by

Protestant supremacist organizations. The Tour of the North was seen as especially problematic because its route passed by St. Patrick’s Catholic Church, the newest parading flashpoint in

Northern Ireland. Parading past the church had been contested since the previous summer when a marching band was caught on video playing a controversial song, known as “The Famine Song,” outside the church. “The Famine Song” is the same tune as the Beach Boys’ classic “Sloop John

B,” but with modified lyrics, such as “the famine’s over, why don’t you go home?”4 The band, the Young Conway Volunteers, maintained that they were playing the Beach Boys’ version, but the Catholic community did not buy it. So Catholics also gathered on the street, but to protest the parade as it passed by. Between the two crowds stood a heavy presence of police in riot gear, using the sunken Westlink highway as a buffer zone.5

There was a lot of build-up to the 2013 Tour of the North. The last time a major

Protestant parade had marched past St. Patrick’s Church it ended in a riot.6 So the police and others were anxious about a repeat performance. Secondly, the Tour of the North marks the beginning of the crescendo of the Protestant parading season which peaks each year several weeks later on Twelfth of July. As a result, observers were concerned that trouble at this parade

4 A Scottish court has ruled that the song, associated with the Glasgow Rangers Football Club supporters, is racist. For more details on the controversy, see “Q&A: How St Patrick’s Became a Flashpoint,” BBC News Online, August 30, 2012. Available at: http://www.bbc.co.uk/news/uk-northern-ireland-19423910. Accessed February 21, 2014.

5 The Westlink was designed to serve such a purpose. Dominic Bryan, “Titanic Town: Living in a Landscape of Conflict,” in S.J. Connolly, ed., Belfast 400: People, Place, and History (Liverpool: Liverpool University Press, 2012), p. 336. For a general discussion and list of interfaces, including such “natural” barriers, see Belfast Interface Project, Belfast Interfaces: Security Barriers and Defensive Use of Space (Belfast: Belfast Interface Project, 2012).

6 “Police Attacked Amid Parade Trouble,” Belfast News Letter, August 25, 2012.

43 might set a violent tone for the 2013 parading season.7 And on top of all of that, Ireland was entering its “decade of centenaries” (2012-2022), when contentious events from the era of Irish partition would be commemorated by one community or the other. Increased tension was expected as each community celebrated centenaries “pregnant with menace.”8

A predominant explanation for aggressive and confrontational acts such as this parade points to the role of ethnic elites. Ethnic elites, a number of scholars argue, use provocative events to spark conflict in order to enhance their political power. At first glance, the Tour of the

North supports this argument. At the head of the parade marched two senior unionist politicians,

William Humphrey and Nelson McCausland. Both are Democratic Unionist Party members of the Northern Ireland Assembly (MLA), the province’s devolved legislature, from North Belfast, and McCausland was also the Minister for Social Development. Along with rest of the paraders, these two politicians marched past their cheering supporters, through the line of police officers and land rovers, in front of the protesters, and then by the church. The protesters had their political representatives as well. In the crowd stood two Sinn Féin MLAs from North Belfast,

Gerry Kelly and Carál Ní Chuilín, the Minster for Culture, Arts, and Leisure. Both Kelly and Ní

Chuilín are also former Provisional IRA prisoners. Thus, the unionist and nationalist politicians both got to go home to their voters as defenders of their respective communities.

A second glance, however, reveals complications to the story. On the Protestant side, the

7 Violence last erupted the Tour of the North parade only two years before, in 2011. Deborah McAleese, “Marchers and Police Clash at Tour of North Parade in Belfast,” Belfast Telegraph, June 20, 2011.

8 Colin Kidd, “On the Window Ledge of the Union,” London Review of Books, February 7, 2014, p. 16. See Jonathan Evershed, “Beyond What Actually Happened: Loyalist Spectral Politics and the Problematic Privileging of ‘History’ during Northern Ireland’s Decade of Centenaries,” Irish Journal of Anthropology, forthcoming; and Dominic Bryan, Mike Cronin, Tina O’Toole, and Catriona Pennell, “Ireland’s Decade of Commemorations: A Roundtable,” New Hibernia Review, Vol. 17, No. 3 (Autumn 2013), pp. 63-86.

44 focus here, the two MLAs were not the only ethnic elites present that day. If we return to the cheering crowd of supporters, we would also find a highly regarded minister from a Presbyterian church in north Belfast. I had met him before, so I immediately noticed that, unlike our previous meeting, he was dressed like a pastor. With his baby blue shirt and white clerical collar, he could be instantly identified as a man of God. His choice of dress was deliberate: he was at the parade to use his ecclesiastical authority to help deescalate the situation if it turned violent. Not too far away, at the very front of the crowd, stood several ex-loyalist paramilitary members, including senior leaders of the Ulster Defence Association (UDA) and the Ulster Volunteer Force (UVF).

They were also there to help with crowd control and also to serve as community representatives, at one point parleying with police officers in the middle of the street to discuss something out of earshot. Like the pastor, these local bosses were at the parade to keep things calm.

The presence of clergymen and ex-paramilitaries was no coincidence—they always attend contentious parades as marshals, negotiators, representatives, and peacekeepers.

Whenever there is a possibility for conflict, they gather nearest to the potential flashpoint so that they are in position to stop it. The Reverend Jack, a Methodist minister I interviewed, describes it as “trying to be a presence.” Tommy, a former UDA commander, calls it “managing” the parade.

In fact, many clergy members and ex-paramilitaries view deescalatory work as part of their job.

The presence of these other Protestant elites with missions of conflict prevention, however, raises a question for a dominant model of ethnic conflict. According to this elite-driven model, these elites should not be trying to discourage conflict—indeed, in some versions of the argument these types of elites would not even exist. In this chapter, I use the case of Northern

Ireland to rethink the relationship between ethnic elites and ethnic conflict. Relying on micro-

45 data collect during extensive fieldwork among Protestant communities in Northern Ireland, I show that the dominant model ignores important dynamics and is insufficient to fully explain ethnic conflict and ethnic peace at the local level.

I begin by assessing dominant approaches to ethnic elites and conflict as well as “rituals of provocation,” symbolic events designed to trigger conflict. I argue that existing models cannot explain the local dynamics of peace and conflict because they rest on a insufficient conception of elites. To correct this deficiency, I present a more expansive and intuitive definition of ethnic elites premised on the amount of authority a person holds within a ethnic community. In the fourth section, I identify the relevant elites in the Protestant community: politicians, clergy, and ex-paramilitary prisoners. The fifth section introduces data drawn from original surveys sent to all Protestant clergy and elected official in Northern Ireland. The sixth section demonstrates the diversity in elite opinion on parades. I show that politicians take a consistently more aggressive position while clergy take a more conciliatory position. In the final section, I highlight the failure of elite-based models to explain mass participation, setting the stage for the rest of the dissertation. !

Ethnic Elites, Provocation, and Ethnic Conflict

One of the most prominent explanations for ethnic conflict looks to the role of ethnic elites. In part a response to the popularity of “ancient hatred” explanations proffered by observers and participants of the civil wars that followed the Cold War, many scholars pointed to the ways in which violence was caused by rational and self-interested leaders rather than fanatical masses.

The general conclusion from this research is that “ethnic violence is provoked by elites seeking

46 to gain, maintain, or increase their hold on political power.”9

There are numerous ways elites could go about sparking conflict, but in this chapter I focus on one: rituals of provocation.10 A form of contentious ritual, rituals of provocation are deliberately offensive and transgressive public, symbolic actions designed to increase tension, and potentially instigate violence, between groups. Anthropologist Marc Gaborieau, who coined the term in a study of Hindu-Muslim violence in South Asia, views them as “well defined” and

“codified procedures to start… hostilities.”11 Such rituals have “two main ingredients”: “the selection of key symbols representing each community” and then “the selection of the means by which such symbols may be most effectively desecrated.”12 “Purposefully desecrating one of those symbols, or threatening to do so, or spreading rumours that it has been done,” he finds, “is enough to spark off violence.”13

Rituals of provocation often appear insignificant, trivial, or even pass unnoticed to outsiders and therefore “the popular response frequently appears out of proportion to the

9 James D. Fearon and David D. Laitin, “Violence and the Social Construction of Ethnic Identity,” International Organization, Vol. 54, No. 4 (Autumn 2000), p. 846. Examples include: Jack Snyder and Karen Ballentine, “Nationalism and the Marketplace of Ideas,” International Security, Vol. 21, No. 2 (Fall 1996), pp. 5-40; Rui J.P. de Figueiredo, Jr., and Barry R. Weingast, “The Rationality of Fear: Political Opportunism and Ethnic Conflict,” in Barbara F. Walter and Jack Snyder, eds., Civil Wars, Insecurity, and Intervention (New York: Columbia University Press, 1999), pp. 261-302; Jack Snyder, From Voting to Violence: Democratization and Nationalist Conflict (New York: Norton, 2000); V.P. Gagnon, The Myth of Ethnic War: Serbia and Croatia in the 1990’s (Ithaca: Cornell University Press, 2004); and Benjamin A. Valentino, Final Solutions: Mass Killing and Genocide in the 20th Century (Ithaca: Cornell University Press, 2004).

10 For a general discussion of the precipitants of ethnic violence, including rituals of provocation, see Donald L. Horowitz, The Deadly Ethnic Riot (Berkeley: University of California Press, 2001), chap. 8, “The Occasions for Violence.” He argues that to successfully trigger violence an event “must be anger producing. It must be threatening or transgressive. By transgressive, I mean that the precipitant constitutes a blatant display of what ethnic strangers are not to be allowed to do with impunity” (p. 268).

11 Marc Gaborieau, “From Al-Beruni to Jinnah: Idiom, Ritual and Ideology of the Hindu-Muslim Confrontation in South Asia,” Anthropology Today, Vol. 1, No. 3 (June 1985), p. 9

12 Ibid.

13 Ibid., p. 10.

47 seemingly meaningless stakes of the triggering events. However,” writes Mark Beissinger, “these events obviously have great meaning for those swept up by them.”14 This is because conflict over cultural expressions in divided societies cut to the core of group identity and belonging.15 As

Edward Linenthal notes, “Desecration registers as a threat to cherished identities.”16

Processions, the specific form of ritual that I focus on, are a particularly potent way to provoke a violent reaction because their “content frequently goes to the heart of the conflict and because [they are] already on the edge of violence.”17 Donald Horowitz estimates that

“processions, demonstrations, and mass meetings precipitate violence in perhaps one-third to one-half of all ethnic riots.”18 This deadly pattern is found in divided societies across time and space, with examples spanning from Protestant-Catholic violence in sixteenth-century France to violence among Hindus, Muslims, Sikhs, Buddhists, and others in contemporary South Asia.19

The literature on rituals of provocation tends to follow the logic of elite-driven conflict.

Though the rituals themselves are generally mass events teeming with frenzied crowds, most scholars argue that they are best explained by ethnic elites plotting from the shadows. Divisive

14 Mark R. Beissinger, “Nationalist Violence and the State: Political Authority and Contentious Repertoires in the Former USSR,” Comparative Politics, Vol. 30, No. 4 (July 1998), p. 402; also, Horowitz, Deadly Ethnic Riot, p. 270; and Marc Howard Ross, “Preface,” in Ross, ed., Culture and Belonging in Divided Societies: Contestation and Symbolic Landscapes (Philadelphia: University of Pennsylvania Press, 2009), p. ix.

15 Marc Howard Ross, Cultural Contestation in Ethnic Conflict (New York: Cambridge University Press, 2007).

16 Edward T. Linenthal, “Epilogue,” in Ross, ed., Culture and Belonging in Divided Societies, p. 283.

17 Horowitz, Deadly Ethnic Riot, p. 275.

18 Ibid., p. 272.

19 On France, see Natalie Zemon Davis, “Rites of Violence: Religious Riot in Sixteenth-Century France,” Past & Present, Vol. 59 (May 1973), pp. 72-74; and Barbara B. Diefendorf, Beneath the Cross: Catholics and Huguenots in Sixteenth-Century Paris (New York: Oxford University Press, 1991), pp. 59, 63. On Sri Lanka, see Stanley J. Tambiah, Leveling Crowds: Ethnonationalist Conflicts and Collective Violence in South Asia (Berkeley: University of California Press, 1996).

48 events, such as rituals of provocation, are useful to ethnic elites in several ways. They can be used to assert group status and dominance and thereby intimidate a rival group into submission.20

Alternatively, they can be used to provoke reaction (or better yet, overreaction) from the other group. A violent response from the other group can polarize society; reinforce ethnic boundaries; promote fear, distrust, and suspicion; and generate a negative image of the other group in domestic and international courts of opinion. The result of these processes can include in-group mobilization, rallying around a leader, strengthening a political organization, and discrediting moderate leaders and policies.21

This dynamic has been studied most extensively in India. In particular, scholars of Indian politics show that Hindu nationalist politicians use religious processions through Muslim neighborhoods to spark polarizing ethnic riots which are electorally advantageous.22 Steven

Wilkinson explains the logic:

Parties that represent elites within ethnic groups will often… use polarizing antiminority events in an effort to encourage members of their wider ethnic

20 Donald L. Horowitz, Ethnic Groups in Conflict, 2nd. ed. (Berkeley: University of California Press, 2000); Roger D. Petersen, Understanding Ethnic Violence: Fear, Hatred, and Resentment in Twentieth-Century Eastern Europe (New York: Cambridge University Press, 2002); and Roger D. Petersen, Western Intervention in the Balkans: The Strategic Use of Emotions in Conflict (New York: Cambridge University Press, 2011).

21 Rogers Brubaker and David D. Laitin, “Ethnic and Nationalist Violence,” Annual Review of Sociology, Vol. 24 (1998), p. 433; Ron Aminzade and Doug McAdam, “Emotions and Contentious Politics,” in Ronald R. Aminzade, Jack A. Goldstone, Doug McAdam, Elizabeth J. Perry, William H. Sewell, Sidney Tarrow, and Charles Tilly, eds., Silence and Voice in the Study of Contentious Politics (New York: Cambridge University Press, 2001), pp. 43-4; Steven I. Wilkinson, Votes and Violence: Electoral Competition and Ethnic Riots in India (New York: Cambridge University Press, 2004); and Adrienne LeBas, “Polarization as Craft: Party Formation and State Violence in Zimbabwe,” Comparative Politics, Vol. 38, No. 4 (July 2006), pp. 419-38. Scholars of terrorism have argued that acts of terror can be used for similar purposes. See Andrew Kydd and Barbara F. Walter, “Sabotaging the Peace: The Politics of Extremist Violence,” International Organization, Vol. 56, No. 2 (Spring 2002), pp. 263-296; Andrew Kydd and Barbara F. Walter, “The Strategies of Terrorism.” International Security, Vol. 31, No. 1 (Summer 2006), pp. 49-80; and Ethan Bueno de Mesquita and Eric S. Dickson, “The Propaganda of the Deed: Terrorism, Counterterrorism, and Mobilization,” American Journal of Political Science, Vol. 51 (April 2007), pp. 364-81.

22 Christophe Jaffrelot, The Hindu Nationalist Movement in India (New York: Columbia University Press, 1996); Paul R. Brass, Theft of an Idol: Text and Context in the Representation of Collective Violence (Princeton: Princeton University Press, 1997); Wilkinson, Votes and Violence; also, Tambiah, Leveling Crowds, p. 241.

49 category to identify with their party and the ‘majority’ identity rather than a party that is identified with economic redistribution or some ideological agenda. These antiminority events… are designed to spark a minority countermobilization (preferably a violent countermobilization that can be portrayed as threatening to the majority) that will polarize the majority ethnic group behind the political party 23 !that has the strongest antiminority identity. In India, this means that upper-caste Hindu politicians and parties must increase the salience of a

Hindu political identity in the minds’ of middle- and lower-caste voters before elections. “They meet this challenge by highlighting the threat posed by Muslims,” Wilkinson writes:

A favorite strategy of Hindu party leaders who calculate that they will gain electorally from polarization around a Hindu identity is to organize unusually large religious processions that take new routes through minority neighborhoods, to hoist the national flag over a disputed site, or to sponsor processions to 24 !celebrate national anniversary. In a general review of the literature on ethnic conflict, Brubaker and Laitin conclude that

“instigative and provocative actions are ordinarily undertaken by vulnerable incumbents seeking to deflect within-group challenges to their position by redefining the fundamental lines of conflict as inter- rather than (as challengers would have it) intragroup; but they may also be undertaken by challengers seeking to discredit incumbents.”25 The hypothesis is clear: elites are behind rituals of provocation. !

A New View of Elites

The problem with political elitist theories is that their view of elites is at once too macro

23 Wilkinson, Votes and Violence, p. 4.

24 Ibid., pp. 23-24.

25 Brubaker and Laitin, “Ethnic and Nationalist Violence,” p. 433.

50 and too narrow to study rituals of provocation.26 They are too macro in that they focus exclusively on national-level elites.27 This is not necessarily a problem for their own purposes which is to explain large-scale war, violence, or mobilization. But when applied to rituals of provocation, which are generally enacted on a very local stage, the argument can be misleading in at least two ways. First, the emphasis on national leaders may be misplaced since, as recent micro-level conflict research has demonstrated, national-level politics are often less important on the ground than local elites, rivalries, networks, and events.28 Second, once we are investigating the local level, the focus on political elites seems overly narrow and restrictive. In local ethnic affairs, the landscape of ethnic elites is far more diverse. Some models of elite-led conflict factor in mild diversity by distinguishing between moderates and hardliners or “old elites” and “rising counter-elites,”29 but the reality is much messier.30 At the local level, there can be numerous leaders with varying interests and preferences that range from convergent to contradictory—to say nothing about varying influence and resources. Additionally, unlike large-scale collective actions, such as war, rituals of provocation do not require many people or resources. As a result, they do not need to be organized by the state or institutions with state-like capabilities (such as

26 For a general critique of elite theories of ethnic conflict, see Petersen, Understanding Ethnic Violence, pp. 34-6.

27 Sherrill Stroschein, “Politics is Local: Ethnoreligious Dynamics under the Microscope,” Ethnopolitics, Vol. 6, No. 2 (June 2007), p. 174.

28 Stathis N. Kalyvas, “The Ontology of ‘Political Violence’: Action and Identity in Civil Wars,” Perspectives on Politics, Vol. 1, No. 3 (September 2003), pp. 475-94; Stathis N. Kalyvas, The Logic of Violence in Civil War (New York: Cambridge University Press, 2006); and Scott Straus, The Order of Genocide: Race, Power, and War in Rwanda (Ithaca: Cornell University Press, 2006).

29 Snyder and Ballentine, “Nationalism and the Marketplace of Ideas,” p. 16.

30 Related is the bigger problem that many studies of ethnic politics and conflict model each group as a unitary actor; for example, Barry R. Posen, “The Security Dilemma and Ethnic Conflict,” Survival, Vol. 35, No. 1 (Spring 1993), pp. 27-47. For a critique, see Kanchan Chandra, “Introduction,” in Kanchan Chandra, ed., Constructivist Theories of Ethnic Politics (New York: Oxford University Press, 2012), pp. 1-47. For a broader critique of studies that rely on “groupism,” see Rogers Brubaker, Ethnicity without Groups (Cambridge: Harvard University Press, 2006).

51 militias). The most necessary resource is some knowledge about what the other group holds sacred and how best to defile them.31 So while they may be mass events, rituals of provocation can also be executed by a small cadre. Our existing theories leave us ill-equipped to make sense of such a situation. They lead us to expect that a single and essentially unified “Elite” plans, promotes, and executes rituals of provocation in order to polarize society for political gain.

In a recent study of ethnic mobilization by minority Hungarian communities in Romania and Slovakia, Sherrill Stroschein modifies existing elite-led theories of conflict in two important ways.32 First, she demonstrates that elites in her cases actually moderated nationalist mobilization, rather than promoted it. Second, she explicitly defines ethnic elites and carefully theorizes their relations with elites of other groups and with their co-ethnic masses—a task most prior works do vaguely at best. Building on the logic of the elite-led mobilization argument, she defines elites as “individuals who have incentives to pursue political goals, i.e. officeholders and party leaders.”33 She then identifies a third group of actors who sit between elites and masses:

“mid-range elites such as intellectuals, journalists and religious leaders.”34 She carves out a new category for them because “these ‘opinion-making persons’ are not formally elites, but neither are they merely ordinary people.”35 For Stroschein, it is inappropriate to call this class of people

31 Gaborieau, “From Al-Beruni to Jinnah,” p. 9.

32 Sherrill Stroschein, “Microdynamics of Bilateral Ethnic Mobilization,” Ethnopolitics, Vol. 10, No. 1 (March 2011), pp. 1-34; and Sherrill Stroschein, Ethnic Struggle, Coexistence, and Democratization in Eastern Europe (New York: Cambridge University Press, 2012).

33 Stroschein, “Microdynamics,” p. 5; also, Stroschein, Ethnic Struggle, p. 20.

34 Stroschein, “Microdynamics,” p. 8; also, Stroschein, Ethnic Struggle, p. 20.

35 Stroschein, Ethnic Struggle, p. 20. See also Arend Lijphart, Democracy in Plural Societies: A Comparative Exploration (New Haven: Yale University Press, 1977), p. 53, who refers to “the middle-level group that can be described as subelite political activists.”

52 “elites” since they sit outside of formal political institutions, arguing that this “is consistent with the elite manipulation view, which posits elites as having political incentives to engage in manipulation.”36

This is an inadequate conception of ethnic elites. By inserting “political incentives to engage in manipulation” into the definition of elite actors, we veer towards tautology. Of course elites cause conflict when elites are defined as actors who are incentivized to start conflicts.

Second, this stacks the deck in favor of political elites over other potential elites before the research has begun. How can we expect the preferences of “mid-range elites” to ever triumph when they are competing against full-blown elites? Third, individuals who are not officeholders and party leaders still have “incentives to pursue political goals” since they have preferences over and interests in political outcomes defined both narrowly (the processes of government) and broadly (power generally).37 Fourth, why should political motives take pride of place over economic, ideological, or moral ones? Again we are stacking the deck in favor of a particular argument.

So why create the new category “mid-range elites” just to put all the non-governmental elites in? A better approach, I argue, is to use a more inclusive and intuitive definition of ethnic elites. Therefore, I define ethnic elites as individuals who hold authority, either institutionalized or non-institutionalized, within an ethnic community. This eliminates the need to create additional categories—now we can conceptualize all elites on a continuum with a single variable: authority. Categorizing elites becomes an empirical matter, rather than definitional.

36 Stroschein, “Microdynamics,” p. 8.

37 See Andrew Mason, “Politics and the State,” Political Studies, Vol. 38 (1990), pp. 575-87.

53 Now it becomes clear that political elites are simply one type of the broader class of ethnic elites.

Authority is measured as being seen by at least a significant segment of the ethnic group as a legitimate representative for the community and/or someone who people in the community turn to for help (for example, to solve disputes, to pay for school fees, or to help find a job). As noted, authority can be institutionalized in a formal hierarchical position, such as a regional governor or bishop, or non-institutionalized and reside with the person by means of wealth, birth into a prominent family, or personal charisma.38 Many elites have both. Additionally, authority and, therefore, elite status can exist on the national, regional, or local level. For example, a tribal chief might have strong authority in a region but little authority at the national level. Conversely, a national leader may have very little authority in a particular region where power resides with a local tribal chief. Using authority as the measure also allows us see that in fact some officeholders may have very little power at any level, where as a pastor with a large and devoted following has a lot. Thus, it makes little sense to call this officeholder a full elite and the pastor a mid-range elite. When debating the role of elites in ethnic mobilization and conflict, we are better off studying the powerful cleric than the pitiful public official. Other definitions would send us looking in wrong place.

This definition is both more inclusive and more intuitive than most existing approaches

(many of which do not even define who elites are). It is more inclusive because it covers

38 The general relationship between formal and informal political power, rules, and institutions is theorized in Gretchen Helmke and Steven Levitsky, “Informal Institutions and Comparative Politics: A Research Agenda,” Perspectives on Politics, Vol. 2, No. 4 (December 2004), pp. 725-39; and Gretchen Helmke and Steven Levitsky, eds., Informal Institutions and Democracy: Lessons from Latin America (Baltimore: Johns Hopkins University Press, 2006). Max Weber famously described three types of authority. My institutionalized authority aligns with Weber’s “legal” authority, while my non-institutionalized authority aligns with his “traditional” and “charismatic” authorities. He introduces the concepts in Max Weber, “Politics as a Vocation,” in H.H. Gerth and C. Wright Mills, trans. and eds., From Max Weber: Essays in Sociology (New York: Oxford University Press, 1972), pp. 78-9, and elaborates them in “The Sociology of Charismatic Authority,” pp. 245-52 and “The Social Psychology of the World Religions,” pp. 295-301.

54 politicians, who are generally what is meant by elites, but also individuals whose authority rests outside of political institutions. It is more intuitive because it looks at who people actually turn to to speak for the community (including defining the community’s interests) and when they need something. In these matters, the government is often quite peripheral to people’s daily life. The head of an ethnic political party would likely be counted as an ethnic elite, but she may not be the person that ordinary people turn to when they are in need. The people they turn to may only hold authority on the local level, but in the daily lives of many people in the ethnic group these local authorities are more important and should therefore be counted as ethnic elites. They are the ones that people might turn to in times of crisis for instructions on how to proceed, and not proceed. The real logic behind the elite manipulation argument is the logic of power. It is about elites getting people to do something that they would not otherwise do. So elites should be defined as individuals with this power in an ethnic group.

With a more expansive understanding of ethnic elites, we can now begin to explore how elites deal with rituals of provocation. I will do so in the case of Northern Ireland, where parades by Protestant groups are an archetypical ritual of provocation.39 In the following sections, I use my definition of elites to identify the relevant actors in Northern Ireland. !

Parades, Conflict, and Protestant Elites in Northern Ireland

Given the divisive consequences of parades catalogued in previous chapter, the elite model of ethnic conflict leads us to expect that loyalist parades are planned and promoted by

39 For example, Brubaker and Laitin, “Ethnic and Nationalist Violence,” p. 445; Aminzade and McAdam, “Emotions and Contentious Politics,” p. 44; and Andreas Wimmer and Conrad Schetter, “Ethnic Violence,” in Wilhelm Heitmeyer and John Hagan, eds., International Handbook of Violence Research (Dordecht, Netherlands: Kluwer Academic Publishers, 2003), p. 256.

55 Protestant ethnic elites looking to gain political advantage. And in fact, many observers of parades explain them in these terms. A number of scholars argue that unionist political elites have used parades as a political tool to generate cross-class and cross-denomination Protestant unity. Indeed, Paul Arthur argues that Orangeism is specifically “designed to overcome interdenominational conflict and the social [class] tensions prevalent among Protestants.”40

Steven Wilkinson argues that in nineteenth-century Ulster, unionist politicians used provocative parades before elections in order solidify the Protestant vote. He uses this brief case study as evidence that his electoral theory of ethnic violence theory works outside of India.41

In the twentieth century, parades were a particularly important tool at moments when sectarian peace opened the door for labor agitation and class-based political organizing. For example, the early years of Northern Ireland’s existence, the ruling Ulster Unionist Party (UUP) was concerned that if the union appeared solid, working class Protestants would vote on class interests rather than ethnic interests. Party leaders specifically worried that if working class

Protestants viewed the constitutional question as resolved, they would vote for the Labour Party and break the cross-class coalition which kept the UUP in power. Therefore, the UUP’s priority was “to keep the Union as the one burning issue.”42 Parades are an effective way to do so, because they “offer occasions where political and cultural differences are emphasized and the

40 Paul Arthur, Government and Politics of Northern Ireland, 2nd ed. (London: Longman, 1984), p. 40.

41 Wilkinson, Votes and Violence, pp. 212-18.

42 Marc Mulholland, The Longest War: Northern Ireland’s Troubled History (New York: Oxford University Press, 2002), pp. 28-45; the quote is from p. 45. This dynamic was also a concern to UUP leadership in the decade after World War II. Henry Patterson and Eric Kaufmann, Unionism and Orangeism in Northern Ireland since 1945: The Decline of the Loyal Family (Manchester: Manchester University Press, 2007), p. 29; and Henry Patterson, Ireland Since 1939 (New York: Oxford University Press, 2002), p. 144.

56 tensions and anger produced mobilize loyalties along sectarian lines.”43 Unionist political elites also used parades to grab and hold the attention of Westminster.44

Nationalist and republican elites have also used parades as a tool to enhance their own power. In fact, as I will demonstrate in Chapter 6, it is widely believed among Protestants that the entire controversy around parades was engineered by republicans to further their political agenda. For example, the Rev. Mervyn Gibson, a prominent Belfast Presbyterian minister and senior Orange Order chaplain, articulated this sentiment when he said that parades “are only contentious when republicans want them to be so.”45 Sinn Féin president Gerry Adams helped to fuel this belief at a party meeting in 1997 when he asked rhetorically, “Do you think Drumcree happened by accident?”46 Despite the conspiratorial view pushed by Gibson and the self- aggrandizing view pushed by Adams, there is evidence that the republican movement has used parading disputes as a political tactic.47 Historian Martyn Frampton identifies “clear political and strategic benefits that the republican movement could seek to gain by its actions” opposing parades. For example, “the issue could easily be used to show Sinn Féin as the stridden voice of the northern nationalist community, standing in the face of apparent Unionist aggression.”48

43 Marc Howard Ross, “Psychocultural Interpretations and Dramas: Identity Dynamics in Ethnic Conflict,” Political Psychology, Vol. 22, No. 1 (2001), p. 158.

44 Smithey, Unionists, Loyalists, and Conflict Transformation, p. 123.

45 Philip Bradfield, “Parade Tensions Are Created, Claims Orange Order Cleric,” Belfast News Letter, July 2, 2013.

46 Malachi O’Doherty, The Trouble with Guns: Republican Strategy and the Provisional IRA (Belfast: Blackstaff, 1998), p. 176.

47 Ibid., 173-76 and 183; and James Dingley, “Marching Down the Garvaghy Road: Republican Tactics and State Response to the Orangemen’s Claim to March their Traditional Route Home after the Drumcree Church Service,” Terrorism and Political Violence, Vol. 14, No. 3 (Autumn 2002), pp. 42-79.

48 Martyn Frampton, The Long March: The Political Strategy of Sinn Féin, 1981-2007 (London: Palgrave Macmillan, 2000), p. 126.

57 Loyalist parading seems very much like an elite-driven ritual of provocation. Certainly the logic of political elites using parades to promote conflict is an important element of the dynamics of parades and parading disputes. This chapter does not aim to overturn this logic.

Rather, I hope to show that it is not the only elite strategy. As a first cut of evidence that a political elite electoral logic does not drive everything take Figure 3.1. If the argument made by

Wilkinson—that political elites use rituals of provocation to spark conflict before elections in order to motivate voters to vote on ethnic lines—holds true in contemporary Northern Ireland, we should expect to see parade-related violence concentrated before elections. But as we see in the graph, that is not the case. Figure 3.1 plots the number of parades with disorder (as reported by the Police Service of Northern Ireland49) each month from 1990 through 2009 along with every election held in Northern Ireland (the lighter vertical dashed lines in 1995 and 2000 indicate by-elections held in just one constituency; all the rest are nationwide elections). Rather than clustering before elections, if anything there is a stronger pattern of disorderly parades clustering in the months after elections. This does not mean that political elites are not using parades strategically, but it suggests that we need to look further to fully understand the situation. !

Figure 3.1. Parades with Disorder and Elections, 1990-2009 Figure 3.1: Parades with Disorder and Elections, by Month 12 8 4 0

# of parades with disorder with # of parades 1990 1995 2000 2005 2010

Black line indicates parades, dashed lines indicate elections (Source: PSNI)

49 Police Service of Northern Ireland, “Parades with Disorder, 1990-2009,” spreadsheet in possession of author.

58 My argument is that looking at parades just from the perspective of political elites alone provides a limited view. It ignores other relevant ethnic elites and therefore circumscribes important questions. Indeed by defining elites in this restricted way, we select on the dependent variable. We are more likely to see a connection between elites and conflict when we only investigate political leaders, the elites most likely to find conflict advantageous. Pulling the lens back to reveal the full scope of elites forces us to ask why some elites seem to instigate conflict while others do not and why some elite voices are listened to over others. !

Ulster’s Protestant Elite

So who are the elites of the Protestant ethnic community?50 Based on extensive fieldwork in Northern Ireland in 2012-2014, I identify three primary classes of Protestant ethnic elites: unionist politicians, Protestant clergy, and loyalist paramilitary ex-prisoners. These three groups hold authority in the community, are seen as legitimate representatives of the community, and members of the community turn to them for help.51 In this section, I briefly describe the characteristics of each group of elites. !

UNIONIST POLITICIANS

Political leaders are widely understood as ethnic elites. This holds true in Northern

50 Following most scholars, I believe that the Protestant community is best describe as an ethnic group, and not a religious group, though religion plays an important role, as I will discuss. See, for example, John McGarry and Brendan O’Leary, Explaining Northern Ireland: Broken Images (Oxford: Basil Blackwell, 1995).

51 This does not mean that any individual, or even an entire class of elites, are seen as legitimate representatives by the whole community. The reality of politics at any level anywhere makes this nearly impossible. As long as some significant portion of a community sees them as legitimate is enough to make them legitimate. Contestation, of course, is part of politics.

59 Ireland, where the Protestant community is represented by two major political parties and several smaller ones. The political stance of the Protestant community is broadly understood as unionism, meaning support for the continued position of Northern Ireland in the United

Kingdom. The two main unionist parties are the Democratic Unionist Party (DUP) and the Ulster

Unionist Party (UUP). The UUP was the ruling party of Northern Ireland from its founding in

1921 until direct rule from Westminster in 1972, and remained powerful through the transition to peace. In recent years, however, the DUP’s power has risen dramatically and it has nearly driven the UUP into obscurity. Smaller parties include the hardline Traditional Unionist Party (TUV) and the Progressive Unionist Party (PUP), a left-leaning party affiliated with the Ulster Volunteer

Force paramilitary. There are unionists elected to all four levels of government in Northern

Ireland. Currently, unionist parties hold two of three Northern Irish seats in the European

Parliament; eight of eighteen seats in the British Parliament (44 percent); fifty-four of 108 seats in the devolved Northern Ireland Assembly (50 percent); and 282 of 582 seats of the twenty-six local government districts (48 percent).52

As elected officials, they clearly have authority vested in their institutional positions.

Though that authority, and therefore influence, varies from national level to local level. MPs,

MEPs, and prominent MLAs are accepted to speak for the Protestant community at large. Less prominent MLAs and local councilors tend to speak for a narrower segment of the community, namely the constituency they represent. Likewise people might seek an MP for help with a nation-wide project, but seek a councilor for help navigating the public housing system. In interviews I conducted with ordinary Protestant citizens, they often spoke highly of their local

52 In 2015, the twenty-six council districts will be consolidated into eleven.

60 councilors and suggested that they were grateful for the work they do. !

PROTESTANT CLERGY

As noted above, the Protestants of Northern Ireland are best considered an ethnic or national community, rather than a religious one, but, members of the clergy are ethnic elites of the community.53 There are at least two reasons why this appropriate. One, the boundaries of ethnicity and the boundaries of religion are largely the same in Northern Ireland. That is to say,

Protestant as an ethnic group and Protestant as a religious group are largely coterminous. Two,

Northern Ireland remains a religiously active society. Though secularization has affected the province, it is much more religious than other advanced industrial democracies, in terms of church attendance, identification, and beliefs.54 Clergy, then, have the (literal) power of the pulpit. And they let their political and social views be known, both from the pulpit and in conversations with congregants.

The clergy also meet my criteria of being people of authority in the ethnic community.

Clergy members often represent the community and its views in the media and official government commissions. For example, the Protestant community’s witness to the Provisional

IRA’s weapons decommissioning and its representative to the Independent Review on Parades

53 Clergy have played an important role in ethnic mobilization in other cases as well. For example, see Aldon D. Morris, The Origins of the Civil Rights Movement (New York: Free Press, 1984) and Doug McAdam, Political Process and the Development of Black Insurgency, 1930-1970, 2nd ed. (Chicago: University of Chicago Press, 1999), pp. 132-36, on African American ministers and the civil rights movement; and Guillermo Trejo, Popular Movements in Autocracies: Religion, Repression, and Indigenous Collective Action in Mexico (New York: Cambridge University Press, 2012) on Roman Catholic bishops and priests and indigenous mobilization in Mexico. Of course, clergy can oppose and even suppress mobilization as well. On the conditions that favor a socially quiescent religion versus a socially oppositional religion, see Dwight B. Billings, “Religion as Opposition: A Gramscian Analysis,” American Journal of Sociology, Vol. 96, No. 1 (July 1990), pp. 1-31.

54 Claire Mitchell, Religion, Identity and Politics in Northern Ireland: Boundaries of Belonging and Belief (Aldershot: Ashgate, 2006), esp. pp. 22-28.

61 and Marches in Northern Ireland were both clergymen.55 Some have even directly entered electoral politics, most famously Rev. Ian Paisley, the founder and long-time leader of the DUP and First Minister in 2007 and 2008.56 Though Paisley is generally associated with zealous opposition to compromise, recent research highlights the pivotal role that the Protestant and

Catholic clergies played in the peace process.57 Additionally, people come to clergy for advice and guidance on matters spiritual and otherwise. Like elected officials, their authority varies in scope. Some clergy have a national audience due to personal popularity or syndication or due to their institutional position in a church; others are a voice of authority only in their congregation. !

LOYALIST PARAMILITARIES AND EX-PARAMILITARY PRISONERS

The third category is local “big men,” primarily ex-loyalist paramilitary prisoners.

Approximately 8,000 loyalist paramilitary members served time in prison during the Troubles, and nearly all were released as part of the peace process.58 Many ex-prisoners have had trouble readjusting to society: rates of unemployment and substance abuse are high. But some have managed to become forces for positive change within their communities.59

55 The Methodist minister Rev. Harold Good and the Presbyterian minister Dr. Rev. John Dunlop, respectively. Their Catholic counterparts were both priests.

56 According to Steve Bruce, Paisley: Religion and Politics in Northern Ireland (New York: Oxford University Press, 2007), p. 1, Paisley is the only person in modern Europe who founded both a church and a political party.

57 John D. Brewer, Gareth I. Higgins, and Francis Teeney, Religion, Civil Society, and Peace in Northern Ireland (New York: Oxford University Press, 2011).

58 James W. McAuley, Jonathan Tonge, and Peter Shirlow, “Conflict, Transformation, and Former Paramilitary Prisoners in Northern Ireland,” Terrorism and Political Violence, Vol. 22, No. 1 (2009), p. 25.

59 See Peter Shirlow, The End of Ulster Loyalism? (Manchester: Manchester Unviersty Press, 2012); Kieran McEvoy and Peter Shirlow, “Re-imaging DDR: Ex-Combatants, Leadership and Moral Agency in Conflict Transformation,” Theoretical Criminology, Vol. 13, No. 1 (2009), pp. 31-59; McAuley, Tonge, and Shirlow, “Conflict, Transformation, and Former Paramilitary Prisoners in Northern Ireland”; and Claire Mitchell, “The Limits of Legitimacy: Former Loyalist Combatants and Peace-Building in Northern Ireland,” Irish Political Studies, Vol. 23, No. 1 (February 2008), pp. 1-19.

62 There are two primary ways in which they promote positive change. First, many are formally employed at peace and reconciliation organizations, often working with at-risk youth to steer them away from crime and violence.60 Second, they act as informal local security providers in their communities: many of the men who fought in the Troubles now work to actively suppress violence. Residents alert ex-combatants when local youths are throwing bricks and bottles at Catholic houses and they head to the scene and break it up. As one told me, if a pastor or peace worker showed up to a riot, the rioters would tell them to leave (in less polite language); but when he arrives, they scatter immediately. As part of this work, they coordinate with former republican paramilitary members via cell phone; warning them, for instance, that a gang of

Protestant teens is heading towards a Catholic neighborhood, so the ex-IRA man should keep the

Catholic youth inside.61 Former belligerents now collaborate to keep their neighborhoods calm.

In both their work as reconciliation advocates and security providers, ex-prisoners draw on the legitimacy they hold for having served time for the cause. This legitimacy and influence, however, is almost entirely confined to working class neighborhoods where they live and where they fought. Middle-class Protestants were never comfortable with the loyalist paramilitaries, and still do not want to be associated with them. !

Survey Data

To study the preferences of ethnic elites in the Protestant community of Northern Ireland,

60 Organizations include Northern Ireland Alternatives, EPIC, Prisoners in Partnership, Intercomm, and many others.

61 Mitchell, “The Limits of Legitimacy,” p. 8; Michael Hamilton, Working Relationships: An Evaluation of the Community Mobile Phone Networks in Northern Ireland (Belfast: Community Relations Council, 2001); and Michael Hall, ed., ‘It’s Good To Talk’: The Experiences of the Springfield Mobile Phone Network (Newtownabbey, UK: Island Publications, 2003).

63 I conducted two online surveys: one of members of the clergy and one of elected officials. Since generating a systematic sample of ex-paramilitary members would be next to impossible, they were not surveyed. This section describes the data collection process. The survey was administered using Qualtrics, a survey website. The email invitations to take part in the survey were sent out on March 6, 2013, with reminders on March 12 and April 5. To increase response rates, invitations were personalized and each email was sent on a different day of the week.62

The choice of an online survey was based on time and budget constraints. Three problems with internet surveys are that people without known email addresses are excluded from the sample; there is a potential for low response rates; and, as with all self-administered surveys, the sample is likely to be biased toward people interested in the topic.63 Administering a postal survey, or a postal component for people without email addresses, could have helped with the first issue, but as I explain below, I actually have an email address for nearly all members of the study population. Using mailed questionnaires could have helped raised the response rate, but I deemed it not worth the tradeoff. And regarding the sample’s representativeness, recent research suggests that Internet surveys of political elites produce samples nearly as representative as postal surveys.64 I will now explain each sample in more depth. ! !

62 Dirk Heerwegh and Geert Loosveldt, “Personalizing E-mail Contacts: Its Influence on Web Survey Response Rate and Social Desirability Response Bias,” International Journal of Public Opinion Research, Vol. 19, No. 2 (Summer 2007), pp. 258-68.

63 Floyd J. Fowler, Jr., Survey Research Methods, 4th ed. (Thousand Oaks, CA: Sage, 2009), p. 72.

64 Samuel H. Fisher III and Rebekah Herrick, “Old versus New: The Comparative Efficiency of Mail and Internet Surveys of State Legislators,” State Politics and Policy Quarterly, Vol. 13, No. 2 (June 2013), pp. 147-63. See also Michael D. Kaplowitz, Timothy D. Hadlock, and Ralph Levine, “A Comparison of Web and Mail Survey Response Rates,” Public Opinion Quarterly, Vol. 68, No. 1 (2004), pp. 94-101.

64 Protestant Clergy Sample

Email invitations with links to the survey were sent to all members of the clergy of the

Protestant denominations with a presence in Northern Ireland.65 The starting point for the list of clergy members comes from a comprehensive database created in 2009 by Gladys Ganiel and

Therese Cullen as part of the “Visioning 21st Century Ecumenism” research project at the Irish

School of Ecumenics of Trinity College, Dublin.66 I updated the database with the most recent information available, adding new clergy members; removing members who are deceased, retired, or currently living outside of Northern Ireland; and updating email addresses. For the three main Protestant denominations (Presbyterianism, Church of Ireland, and Methodism), I used church publications to check the list of clergy members. Between the three of them, I have email addresses for 759 of 819 clergy (92.7 percent).67 These three denominations jointly comprise 86 percent of non-Roman Catholic Christians in Northern Ireland.68 For the smaller denominations, I primarily relied on the Ganiel and Cullen database and supplemented from

65 However, since the focus of the study is on ethnic elites, rather religious elites, clergy working at congregations that cater to immigrant communities were excluded. Exclusion was based on the 2009 “Directory of Migrant-Led Churches” published by the All-Ireland Churches Consultative Meeting on Racism, or else having a distinctive nationally-based church name, for example, the Belfast Chinese Christian Church. Available at http:// ireland.anglican.org/archive/hardgospel/cmsfiles/files/migrantledchurches_09.pdf. Accessed 11 February 2014.

66 “Cullen compiled a database of email and postal addresses of clergy, pastors, ministers and faith leaders, gathering this information from Denominational Directories, websites, and telephone directories.” Gladys Ganiel, 21st Century Faith – Results of the Survey of Clergy, Pastors, Ministers and Faith Leaders (Belfast: Irish School of Ecumenics, n.d.), p. 9, available at: http://www.ecumenics.ie/wp-content/uploads/Clergy-Survey-Report.pdf. Accessed February 14, 2013. See also Gladys Ganiel, “Surveying Religion’s Public Role: Perspectives on Reconciliation, Diversity and Ecumenicism in Northern Ireland,” Shared Space, Vol. 9 (March 2010), pp. 53-68.

67 For the Presbyterian church, see “Ministers Email Addresses,” available at: http://www.presbyterianireland.org/ ministers/mina.html. Accessed February 6, 2013; and “Presbyterian Church Online Directory,” available at: http:// presbyterianireland.org/congregations/index.html. Accessed February 6, 2013. For the Church of Ireland: Susan Hood, ed., The Church of Ireland Directory 2013 (Belfast: DCG Publications Ireland, 2012), pp. 27-97. For the Methodist church: Methodist Church in Ireland, Minutes of the Conference: Enniskillen 2012 (Belfast: Edenderry, 2012), “Appendix 6: Minister and Probationers,” pp. 133-52.

68 According to the 2011 Census, there are 752,555 non-Roman Catholic Christians, of which 345,101 are Presbyterian, 248,821 are Church of Ireland, and 54,253 are Methodist.

65 online sources when I could.69 In all, I sent invitations to take the survey to 844 email address and received 212 valid responses (25.2 percent response rate).70 !

Elected Officials Sample

The list of elected officials comes from the websites of the Northern Ireland Assembly and the 26 Local Government Councils; I did not sample members of the British or European

Parliaments. Usually, each member’s email address was provided. When it was not, I deduced an email address from the pattern of available email addresses. The list includes all members of the unionist political parties (DUP, UUP, PUP, and TUV), as well as members of the Alliance Party and independents. If I could find online that these people were not Protestant, I excluded them from the list. I sent the invitation to the survey to all 397 elected officials who met those criteria

(65 members of the Northern Ireland Assembly and 332 members of Local Government

Councils), and received 72 valid responses (18.1 percent response rate).71 For the analysis, I exclude several Catholic respondents and representatives of the Alliance party since the study is interested in the views of Protestant ethnic elites and Alliance is a non-sectarian party. This leaves 57 elected officials in the sample.

69 There are available online directories of the Free Presbyterian Church, Elim Pentecostal Church, Reformed Presbyterian Church, Association of Baptist Churches in Ireland, and Christian Brethren, but many of these websites are missing most information or state that they are incomplete. For instance, I only have email address for 22 of 62 Free Presbyterian ministers. For all other denominations, I relied entirely on the database. See, respectively, “Free Presbyterian Church Information Page,” www.freepres.org/churchlist.asp?loc=ni; “The Elim Churches in Northern Ireland,” http://elimireland.org/component/option,com_magazine/func,show_edition/id,4/Itemid,96/; “Reformed Presbyterian Church of Ireland – Congregations,” http://rpc.org/directory; “Association of Baptist Churches in Ireland – Churches,” www.baptistsinireland.org/churches.php; “Gospel Halls – Northern Ireland,” www.gospelhall.org.uk/lists/nireland.html. All accessed February 14, 2013.

70 Two people responded saying that they no longer live in Northern Ireland and 8 people started the survey but did not complete it. In her survey, Ganiel had a 16.9 percent response rate in Northern Ireland. Ganiel, 21st Century Faith, p. 9.

71 Nine others started the survey but did not complete it.

66 The Diversity of Elite Opinion

In this section, I use original surveys of Protestant elected officials and clergy in Northern

Ireland to show that the two groups hold different opinions regarding parades. In particular, members of the clergy care less about parades and hold more conciliatory opinions about them.

First, clergy are far less likely to express that parades are personally important to them.72

Among clergy, 58 percent of respondents said that parades are not at all important, while 7 percent said that parades were very or extremely important. In contrast, only 14 percent of elected officials said that parades were not at important personally, while 53 percent said they were very or extremely important. The mean responses for each group are significantly different from each other (p=0.00).

This personal interest is reflected in their parade attendance. Respondents were asked how many parades they have attended as a spectator or supporter on average over the past five years. Only 30 percent of clergy members attended even one parade per year, whereas 75 percent of elected officials had. This could be a reflection of appealing to voters, but, as Table 3.1 shows, how important elites think parades are to their constituents/congregants is not associated with attendance.73 This logit regression examines the determinants of attending any parades versus no parades. Both the measure of the personal importance of parades and a dummy indicator of being an elected official (as opposed to clergy) are positively and significantly associated to parade attendance, while the measure of importance to their constituents/congregants is not.

72 “Are loyalist parades important to you personally?” Response options ranged on a five point scale from extremely important to not at all important.

73 “Are loyalist parades important to your [constituents/congregants]?” Response options ranged on a five point scale from extremely important to not at all important.

67 Table 3.1. The Determinants of Elite Parade Attendance, Logit Models Personal importance of parades 0.95*** (0.22) Importance of parades to constituents/congregants 0.13 (0.20) Elected official 1.08* (0.53) Constant -1.64*** (0.41) Observations 196 Standard errors in parentheses * p<0.05, ** p<0.01, *** p<0.001

Beyond finding parades less important than politicians, the clergy hold more conciliatory and more critical views on parades. One question asked whether the respondent believes that the loyal orders should be able to march wherever they want, without restrictions; be able to march through mainly Catholic areas only if there is prior agreement with local residents; or should not be allowed to march through mainly Catholic areas. The responses range from most to least hardline. Table 3.2 shows that the hardline view is far more prevalent among elected officials than the clergy: 41 percent of elected officials compared to only 7 percent of clergy members. In contrast, two-thirds of clerics take the position that parades should only take place after negotiations and compromise with the local residents. One-quarter of the clergy also take the position that parades simply should not pass through Nationalist communities, while only 17 percent of elected officials feel the same.74

74 This same question was asked in the Northern Ireland 2010 General Election Survey. Among Protestants in the poll 48 percent said they should march wherever they want, 44 percent said there should be prior agreement, and 8 percent said parades should not march in Nationalist areas.

68 Table 3.2. Elite Opinions on Parade Routes (%) Elected Clergy P-value Total % Officials (Two- (N) Tailed) March wherever they want 41 7 0.00*** 14% (33) March only with prior agreement 41 68 0.00*** 63% (148) Should not march in Nationalist areas 17 25 0.29 23% (55)

Observations 46 190 * p<0.05, ** p<0.01, *** p<0.001

What is more, clergy are more likely to believe that parades are provocative.75 Only 4 percent of clergy took the most hardline position that no parades are provocative, while 36 percent of elected officials expressed the same opinion. The majority of each group said that a few parades are provocative (64 percent of clergy and 56 percent of politicians), but far more clergy than elected officials said that more than a few are provocative (32 percent versus 8 percent; p=0.00).

The difference in opinions between the clergy and politicians is perhaps most evident using an index of polarizing questions on parades. Respondents were asked to mark on a five- point scale if they agreed or disagreed with 14 statements about parades. Examples include: “I am proud of loyalist parades in Northern Ireland”; “the loyal orders and marching bands have done everything that they can to prevent conflict”; “a purpose of loyalist parades is to intimidate the communities they pass”; and “as a whole, loyalist parades are harmful to Northern Irish society.” I transformed each question so the responses increased in pro-parade sentiment, and then summed them to create a 57-point index where 0 is least supportive of parades and 56 is

75 “In your opinion, how many parades are provocative?” The responses were none, a few, many, most, or all parades are provocative.

69 most supportive (Cronbach’s alpha = 0.9481). Looking at all responses together, the index has a fairly normal, mean-centered distribution, giving me confidence that it effectively measured opinions on parades (Figure 3.2). But when we break the histogram into elected officials and clergy, we see that the bell curve masks significant difference in groups (Figure 3.3). In line with my other findings, clergy on average hold opinions that are far more critical of parades, while elected officials tend to be much more supportive. The median score for elected officials is 24, while for the clergy it is 40.5. The different in means is statistically significant (p=0.00).

Figure 3.2. Histogram of Elite Opinions on Parades

Figure 3.3. Kernel Density Plot of Opinions on Parades by Elected Official and Clergy

70 The differences in personal importance of parades and opinions about parades manifest in the public actions that elites take regarding parades. Elected officials and clergy were asked if they had ever done a series of parade-related actions during their time in office or the ministry.

The results are displayed in Table 3.3, where we see wide differences in the actions taken by each group in nearly every category. Nearly half of elected officials (48 percent) have given a public address about parades, either for or against. But only one-fifth of clergy members (22 percent) have given a similar sermon (p=0.00). The differences are starker for giving a speech or sermon that takes a position favorable to parades or opposed to anti-parade protesters. Forty- three percent of elected official have done so, while only 11 percent of clergy have (p=0.00).

Notably, the most common speech given by a politician is in support of parades, while the most common sermon given by clergy is critical of parades.

Interestingly, despite the clergy’s general avoidance of parades and the issue of parading, nearly three-quarters have delivered a sermon or led a prayer at a loyal order church service. This suggests that parades and parading organizations are so embedded in Protestant communities that it is hard for elites to avoid entirely. However, there is also evidence that they may not be entirely happy with the arrangement. Forty-three percent of clergy say that a loyal order has used their church for a ceremony or function in the past year. But of those clerics, over one-quarter say that they are not comfortable with the fact that their church was used in this way. We can speculate that the figure might be even higher among clergy whose churches were not used by a loyal order in the last year. So while 72 percent of clergy have participated in a loyal order religious service, a significant portion may have been uncomfortable with the situation.

One reason why elected officials may care more about parades and hold views that are

71 Table 3.3. Parade-Related Activities of Elected Officials and Clergy (%) ! Elected ! P-score Action Officials Clergy (Two-Tailed) Speech/Sermon supporting parades 37 5 0.00*** Speech/Sermon critical of parades 15 17 0.74 Speech/Sermon supporting protesters 11 1 0.00*** Speech/Sermon critical of protesters 6 5 0.81 Any Parade Related Speech/Sermon 48 21 0.00*** Participated in Parade 52 11 0.00*** Lobbied Parades Commission 54 13 0.00*** Met with Nationalists to discuss parades 26 9 0.00*** Attended community meeting about parades 65 15 0.00*** Helped lodge or band with grant application 43 3 0.00*** Introduced legislation about parades 2 - - Mentioned parades in campaign material 22 (N=51) - - Gave sermon/prayer at loyal order church service - 72 - Observations 46 179 * p<0.05, ** p<0.01, *** p<0.001 more supportive of parades compared to the clergy is that elected officials are more likely to be a member of a parading organization (Table 3.4). Among elected officials, 37 percent are currently members and 46 percent have been a member at some point. Among clergy, in contrast, only 7 percent are currently members and only 18 percent have ever been members. This is a striking given that historically membership in the Orange Order was a practical requirement for election to political office, and high rates of clergy were members as well.76

76 The number of clergy in the Orange Order has been declining since the mid-twentieth century. A 1988 survey of clergy in Northern Ireland found that 12 percent were Orangemen. Eric P. Kaufmann, The Orange Order: A Contemporary Northern Irish History (New York: Oxford University Press, 2007), p. 285.

72 Table 3.4. Parade Organization Membership among Elites (%) Elected ! P-Value Officials Clergy (Two-Tailed) Never member 54 82 0.00*** Past member 10 12 0.69 Current member 37 7 0.00*** Observations 52 198 * p<0.05, ** p<0.01, *** p<0.001

This section has shown that Protestant elites are not unified when it comes to parades, and therefore we should not characterize them as one unitary Elite. I found that unionist politicians do strongly support parades and take a hardline position on contentious parades, but clergy have a different view. The clergy are a more diverse group, but as a whole they do not support parades as vigorously and many are outright opposed to the controversial parading.77

These findings present a more nuanced picture of the ethnic elites of the Protestant community.

Conclusion

The argument that elites manipulate ethnic conflict to further their political agenda is compelling and often useful. The problem with it is that by assuming that elites are a unitary class (or at best an elite divided into moderates and hardliners along a single issue dimension), the theory of elite-led conflict obscures important diversity among ethnic elites. And by homogenizing the various elite groups and interests, this approach buries important questions about ethnic elites and their relationship to conflict. For example, why are political elites often more successful at mobilizing people than other types of elites? Why do elite voices favoring

77 Even among Orange Order chaplains, there are important hardliners, but the majority are moderates. Ibid., 285-6.

73 conflict drown out elite voices opposing it? As it stands now, the theory of elite-driven conflict cannot explain why certain elite voices are heard over others. Future research should interrogate why this is the case. To do so requires careful micro-level research that identifies the relevant elites; their sources of power; and their interests, ideologies, and preferences. In this chapter, I provided a preliminary analysis along these lines.

A further problem with the elite-led model is that it does not answer the “insistent question of why the followers follow.”78 We cannot explain individual level participation solely with reference to the interests of elites.79 And mass participation matters, since without it, most rituals of provocation would not occur. My findings in this chapter make this general problem even more pronounced. Even if elite interests were a sufficient explanation, Northern Irish

Protestants are divided over parades. There is not a single “elite preference” to guide mass behavior. Moreover, the elites who are more supportive of parades—politicians—are strongly disliked in Northern Ireland. Protestants are especially disenchanted with their elected representatives. In a 2009 poll, 87 percent of Protestants stated that they did not trust politicians.80 In a 2003 poll, 75 percent agreed that “those we elect lose touch with people pretty quickly” and 77 percent agreed that “parties are only interested in people’s votes, not in their

78 Horowitz, Ethnic Groups in Conflict, p. 140.

79 See Kalyvas, “The Ontology of ‘Political Violence.’”

80 ARK, Northern Ireland Life and Times Survey, 2009, distributed by ARK, 2009, available at: http:// www.ark.ac.uk/nilt/2009/Political_Attitudes/POLTRUST.html. Accessed March 6, 2015. The question is: “How much would you say you trust politicians generally? A great deal, a fair amount, not very much, or not at all.”

74 opinions.”81 And most of the people I interviewed were thoroughly upset with unionist politicians, particular at the national level (local officials were often rated more highly). A question about whether there were political leaders they admired frequently prompted sighs or laughter.

So elite-based accounts, at least on their own, do not account for the 2,500 annual loyalist parades which feature tens of thousands of voluntary participants. How, then, do we explain participation? The remainder of this dissertation presents an answer. In the next chapter, I provide an argument that explains participation in loyalist parades by exploring these actions as rituals. Chapters 5 and 6 use interviews along with survey and ethnographic data to support the ritual approach to participation. Then Chapter 7 demonstrates that plausible alternative explanations—ethnic rivalry and rational collective action—not do fit the data.

81 ARK, Northern Ireland Life and Times Survey, 2003, distributed by ARK, 2003, available at: http:// www.ark.ac.uk/nilt/2003/Political_Attitudes/LOSETCH.html and http://www.ark.ac.uk/nilt/2003/ Political_Attitudes/VOTEINTR.html. Accessed March 6, 2015. The questions are: “Generally speaking those we elect lose touch with people pretty quickly” and “Parties are only interested in people’s votes, not in their opinions.” Agree strongly, agree, neither agree nor disagree, disagree, or disagree strongly.

75 Chapter Four A Theory of Ritual Participation! The man who complained, “They’re just going through a ritual” had the analysis right, but he should not have regretted it. Rituals matter. 1 ! — James M. Jasper ! This chapter introduces a theory to explain participation in contentious rituals: symbolic actions that make contested political claims. It seeks to understand why people participate in events which provide limited private rewards and have socially harmful consequences. I suggest that two insights from the study of ritual help us arrive at an explanation. First, rituals provide intrinsic benefits to participants. This insight explains why people voluntarily participate despite the option to free-ride. And, second, they are multi-vocal and their meaning is ambiguous. The ambiguity of ritual helps us to understand that participants may not share the interpretation of their action held by opponents or other outsiders. In fact, the multi-vocality allows participants to dispute the legitimacy of opponents’ interpretation. I integrate these findings with research on collective action and contentious politics to provide an explanation that accounts for the anomalous features of contentious rituals. From this general argument, I produce specific implications that I will test on data from parades in Northern Ireland in future chapters.

This chapter proceeds as follows. First, I define ritual as a distinct class of human action.

Second, I review the role of rituals in politics, introducing a specific type of political ritual that I call contentious rituals. In contrast to existing models of ritual and conflict, I suggest that contentious rituals are episodes of political conflict, not merely substitutes or descriptions of the conflict’s repetitive nature. Third, I argue that rituals are a form of collective action and, then

1 James M. Jasper, The Art of Moral Protest: Culture, Biography, and Creativity in Social Movements (Chicago: University of Chicago Press, 1997), p. 207.

76 fourth, build on Mancur Olson’s framework to show that participation in ritual is a puzzle.

Specifically, given that contentious rituals produce a public good, why does anybody voluntarily participate? Fifth, I present my explanation: people participate because of the process benefits intrinsic to taking part in a collective ritual. Further, the symbolic ambiguity of ritual explains why people are not dissuaded from participating by the divisive consequences of their actions.

Sixth, I outline four alternative explanations from the literatures on rationalist explanations for ritual, elite-led conflict, ethnic rivalry, and collective action. Finally, I conclude with a preview of the coming chapters. !

Defining Ritual

Edward Leach noted that already nearly fifty years ago “[there was] the widest possible disagreement as to how the word ritual should be understood.”2 Since then, the number of definitions has only increased.3 Some scholars define ritual as an exclusively religious phenomenon. By this definition, acts with secular concerns as their object do not qualify.4 But as critics of the restrictive definition illustrate, the concept becomes overly narrow and obscures a

2 Edmund R. Leach, “Ritual,” in David L. Sills, ed., The International Encyclopedia of the Social Sciences, vol. 13, (New York: Macmillan, 1968), p. 526.

3 The most comprehensive analyses of the concept of ritual are Catherine Bell, Ritual Theory, Ritual Practice (New York: Oxford University Press, 1992); and Catherine Bell, Ritual: Perspectives and Dimensions, rev. ed. (New York: Oxford University Press, 2009).

4 See, e.g., Jack Goody, “Religion and Ritual: The Definitional Problem,” British Journal of Sociology, Vol. 12, No. 2 (June 1961), p. 159; Max Gluckman, “Les Rites de Passage,” in Max Gluckman, ed., Essays on the Ritual of Social Relations (Manchester: Manchester University Press, 1962), pp. 22-23; and Max Gluckman, Politics, Law and Ritual in Tribal Society (Oxford: Basil Blackwell, 1965), p. 251; Victor W. Turner, The Forest of Symbols: Aspects of Ndembu Ritual (Ithaca: Cornell University Press, 1967), p. 19; and Evan M. Zuesse, “Ritual,” in Lindsay Jones, ed., Encyclopedia of Religion, 2nd ed. (Detroit: Macmillan Reference, 2005), p. 7834.

77 more general understanding of ritualized behavior.5 Other scholars define ritual very broadly, identifying it as an aspect of all social action or calling any repeated action a ritual.6 But as critics of the expansive definition point out, the term loses its analytic bite when stretched so thin.7 Following anthropologist David Kertzer, I define rituals as “symbolic behavior that is socially standardized and repetitive.”8 This definition is broad enough to include both religious and secular symbolic acts, but narrow enough to exclude certain actions and make the concept theoretically useful. Additionally, I restrict my analysis to rituals that are public and collective since these are most politically salient. Rituals that are conducted individually or in private, while important, are beyond the scope of my research.

A further point that follows from this definition is that rituals are not only a feature of pre-modern societies and polities. They did not fade away with rationalization, economic development, or the “disenchantment of the world”: modern societies and states continue to practice rituals.9 In fact, many contemporary rituals which appear ancient—royal coronations, for

5 “These theological residues in the notion of ‘ritual’ get in the way of understanding—and especially of evaluating —political rituals,” writes Robert E. Goodin, “Rites of Rulers,” British Journal of Sociology, Vol. 29, No. 3 (September 1978), p. 281. See also the critiques in Sally F. Moore and Barbara G. Myerhoff, eds., Secular Ritual (Assen: Van Gorcum, 1977).

6 See, e.g., Edmund Leach, Political Systems of Highland Burma: A Study of Kachin Social Structure, 2nd. ed. (London: Bell and Sons, 1964), p. 13. In a somewhat different sense, see Erving Goffman, Interaction Rituals: Essays on Face-to-Face Behavior (New York: Pantheon, 1967). See the useful discussion in Andrew L. Roth, “‘Men Wearing Masks’: Issues of Description in the Analysis of Ritual,” Sociological Theory, Vol. 13, No. 3 (November 1995), pp. 315-20.

7 I agree with philosopher John Skorupski’s view: “What use does a term have which brings together a man shaking hands, a man praying to his god, a man refusing to walk under a ladder, a man clapping at the end of a concert, a man placing medicine on his crops? None at all.” John Skorupski, Symbol and Theory: A Philosophical Study of the Theories of Religion in Social Anthropology (Cambridge: Cambridge University Press, 1976), p. 171, quoted in Caroline Humphrey and James Laidlaw, The Archetypal Actions of Ritual: A Theory of Ritual Illustrated by Jain Rite of Worship (Oxford: Claredon Press, 1994), p. 66. See also, Robert Bocock, Ritual in Industrial Society: A Sociological Analysis of Ritualism in Modern England (London: George Allen & Unwin, 1974), p. 15.

8 David I. Kertzer, Ritual, Politics, and Power (New Haven: Yale University Press, 1988), p. 9.

9 See Bocock, Ritual in Industrial Society; Kertzer, Ritual, Politics, and Power.

78 instance—originated in the modern era.10 !

Rituals and Politics

A central reason why rituals remain prominent in modern politics is that they can be valuable resources for political actors. They can legitimate the status quo (or challenges to it), build solidarity, define a political reality, generate strong emotions, gather a crowd of people, represent a group, create common knowledge, or make political claims.11 Rituals can thus mobilize political action or constitute political action.12 All the while, rituals have the advantages of being seen as legitimate forms of action, which makes them difficult for states to regulate, and of being symbolically ambiguous, which provides cover for performing otherwise unacceptable

10 Eric Hobsbawm and Terence Ranger, eds., The Invention of Tradition (Cambridge: Cambridge University Press, 1983).

11 Goodin, “Rites of Rulers”; Clifford Geertz, Negara: The Theatre State in 19th Century Bali (Princeton: Princeton University Press, 1980); Hobsbawm and Ranger, eds., The Invention of Tradition; Kertzer, Ritual, Politics, and Power; Mona Ozouf, Festivals and the French Revolution (Cambridge: Harvard University Press, 1988); Mabel Berezin, Making the Fascist Self: The Political Culture of Interwar Italy (Ithaca: Cornell University Press, 1997); Lisa Wedeen, Ambiguities of Domination: Politics, Rhetoric, and Symbols in Contemporary Syria (Chicago: University of Chicago Press, 1999); Sean Wilentz, ed., Rites of Power: Symbolism, Ritual, and Politics since the Middle Ages (Philadelphia: University of Pennsylvania Press, 1999); Amitai Etzioni, “Toward a Theory of Public Ritual,” Sociological Theory, Vol. 18, No. 1 (March 2000), pp. 44-59; Michael Suk-Young Chwe, Rational Ritual: Culture, Coordination, and Common Knowledge (Princeton: Princeton University Press, 2001); Steven Pfaff and Guobin Yang, “Double-Edged Rituals and the Symbolic Resources of Collective Action: Political Commemorations and the Mobilization of Protest in 1989,” Theory and Society, Vol. 30, No. 4 (August 2001), pp. 539-589; Amitai Etzioni and Jared Bloom, eds., We Are What We Celebrate: Understanding Holidays and Rituals (New York: NYU Press, 2004); and Ziad W. Munson, The Making of Pro-Life Activists: How Social Movement Mobilization Works (Chicago: University of Chicago Press, 2008), esp. pp. 171-177.

12 For examples of the causal model of political ritual, see Francesca Polletta, “Can You Celebrate Dissent?” in Etzioni and Bloom, eds., We Are What We Celebrate, pp. 151-177; and Pfaff and Yang, “Double-Edged Rituals and the Symbolic Resources of Collective Action”; of the constitutive model of political ritual, see Geertz, Negara. I elaborate this typology further in Jonathan S. Blake, “The Variety of Political Rituals: A Typological Analysis,” paper presented at the Association for the Study of Nationalities World Convention, New York, April 24-26, 2014.

79 actions.13 These advantages make them attractive to political actors—and simultaneously troublesome for their political rivals. Therefore, rituals often become central objects of political contestation. I call these contentious rituals.

The sources of a dispute over a ritual are varied.14 Here I propose three. First, actors and groups might contest the insertion of political rhetoric and claim-making into a ritual that is typically understood as apolitical. For example, when partisan political speeches were delivered at the memorial for US Senator Paul Wellstone, the mourning ritual became “perhaps the most politically contentious event” of the year.15 People might also dispute specific symbols used in the ritual that give it a particular interpretation.16 Second, opponents may dispute where the ritual takes place. Muslims in Jerusalem, for instance, dispute Jewish prayer atop the Temple Mount/

Haram al-Sharif.17 The third source of contestation is the ritual’s very performance. In this case, opponents dispute any and all performances of the ritual. For example, many Chinese and

Korean citizens protest any visits by Japanese officials to the Yasukuni Shrine, which commemorates war criminals. In all of these cases, the ritual’s contestation can be intended by its

13 Charles Tilly, Contention and Democracy in Europe, 1650-2000 (New York: Cambridge University Press, 2004), p. 31; Steven I. Wilkinson, Votes and Violence: Electoral Competition and Ethnic Riots in India (New York: Cambridge University Press, 2004), pp. 23-24; Pfaff and Yang, “Double-Edged Rituals and the Symbolic Resources of Collective Action.” For a more general argument on the power of “ritualization,” see Bell, Ritual Theory, Ritual Practice.

14 I ignore disputes over ritual performance within communities that do not have explicitly political ramifications, for example, changing the text of a prayer. Of course, the line between what is political and what is not is fine and shifting.

15 Alyssa Samek and Karrin Vasby Anderson, “The Day the Campaign Died: The Wellstone Memorial, Civic Piety, and Political Propriety,” Communication Quarterly, Vol. 59, No. 2 (April-June 2011), p. 156.

16 Peter Stamatov, “Interpretive Activism and the Political Uses of Verdi’s Operas in the 1840s,” American Sociological Review, Vol. 67, No. 3 (June 2002), pp. 345-366, argues that the political meaning of cultural objects is the result of deliberate work by “interpretive activists,” rather than any inherent property of the object.

17 As do many, if not most, rabbinic authorities. See Ron E. Hassner, War on Sacred Grounds (Ithaca: Cornell University Press, 2009), pp. 113-34; and Gershom Gorenberg, The End of Days: Fundamentalism and the Struggle for the Temple Mount (New York: Free Press, 2000).

80 performers and/or organizers or can be a surprise to them.

Once a ritual becomes disputed, the effects can be anything but symbolic. The contestation and performance of a contentious ritual can cause significant political consequences, including increased polarization, heightened tension, and violence. As Marc Howard Ross argues, “Cultural expressions are not just surface phenomena. They are reflectors of groups’ worldviews and on-going conflicts… [They] play a causal role in conflict… [And they] serve as exacerbaters or inhibitors of conflict.”18 At their most dangerous, contentious rituals can trigger the onset of violent ethnic clashes. Anthropologist Stanley Tambiah finds that the calendar of communal festivals and religious rites in India “can at sensitive times actually channel and direct the shape, expression, timing, and spatial location of ethnic violence.”19 Although this relationship appears to be particularly strong in modern India,20 it can be found in divided

18 Marc Howard Ross, Cultural Contestation in Ethnic Conflict (New York: Cambridge University Press, 2007), p. 3. Emphasis in the original.

19 Stanley Tambiah, Leveling Crowds: Ethnonationalist Conflicts and Collective Violence in South Asia (Berkeley: University of California Press, 1996), p. 240.

20 On India, see Wilkinson, Votes and Violence; Christophe Jaffrelot, The Hindu Nationalist Movement in India (New York: Columbia University Press, 1996); Sudhir Kakar, The Colors of Violence: Cultural Identities, Religion, and Conflict (Chicago: University of Chicago Press, 1996), pp. 44-51; Peter van der Veer, “Riots and Rituals: The Construction of Violence and Public Space in Hindu Nationalism,” in Paul R. Brass, ed., Riots and Pogroms (New York: New York University Press, 1996), pp. 154-76; Paul R. Brass, Theft of an Idol: Text and Context in the Representation of Collective Violence (Princeton: Princeton University Press, 1997); Christophe Jaffrelot, “The Politics of Processions and Hindu-Muslim Riots,” in Amrita Basu and Atul Kohli, eds., Community Conflict and the State in India (Delhi: Oxford University Press, 1998), pp. 58-92; Paul R. Brass, The Production of Hindu-Muslim Violence in Contemporary India (Seattle: University of Washington Press, 2005); and Arafaat A. Valiani, “Processions as Publics: Religious Ceremonials and Modes of Public Sphere Intervention in Western India,” unpublished manuscript, Williams College, 2011.

81 societies around the globe, historically and today.21 In fact, in his comparative study of ethnic riots, Donald Horowitz concludes that contentious rituals are one of the most significant precipitants of ethnic riots around the world.22 But, contentious rituals can also have profound political effects short of violence. Even when they do not spark violent clashes, cultural practices such as flying flags, ritual animal slaughter, celebrating nationalist holidays, erecting monuments, pilgrimages to sacred sites, visits to cemeteries, and mass worship can exacerbate tensions between groups and make conflicts more difficult to resolve.

My concept of contentious rituals, and the relationship between rituals and conflict that it implies, differs from two prominent views of rituals and conflict: the channeling of conflict thesis and the ritualization of conflict thesis. The “channeling of conflict” thesis suggests that in conflictual environments, social tension can be channeled into rituals so that the rituals replace violent conflict. The ritual represents the conflict and allows participants to blow off steam, thus managing tension and preventing violence. Anthropologists call these events “rites of rebellion” or “rituals of reversal” and argue that they “provide an important safety valve for political

21 For general and comparative work, see Donald L. Horowitz, The Deadly Ethnic Riot (Berkeley: University of California Press, 2001), esp. pp. 272-277; Ross, Cultural Contestation in Ethnic Conflict; and Ron E. Hassner, “Sacred Time and Conflict Initiation,” Security Studies, Vol. 20, No. 4 (November 2011), pp. 491-520. For other cases, see, for example, Natalie Zemon Davis, “Rites of Violence: Religious Riot in Sixteenth-Century France,” Past & Present, Vol. 59 (May 1973), pp. 51-91; Bernard Wasserstein, “Patterns of Communal Conflict in Palestine,” in Ada Rapoport-Albert and Steven J. Zipperstein, eds., Jewish History: Essays in Honour of Chimen Abramsky (London: Peter Halban, 1988), pp. 611-628; Barbara B. Diefendorf, Beneath the Cross: Catholics and Huguenots in Sixteenth-Century Paris (New York: Oxford University Press, 1991), pp. 59, 63; Lori A. Allen, “The Polyvalent Politics of Martyr Commemorations in the Palestinian Intifada,” History & Memory Vol. 18, No. 2 (Fall/Winter 2006), pp. 107-38; Vali Nasr, The Shia Revival: How Conflicts Within Islam will Shape the Future (New York: W.W. Norton, 2006); Elliott Horowitz, Reckless Rites: Purim and the Legacy of Jewish Violence (Princeton: Princeton University Press, 2008), pp. 5-8, 10-12.

22 Horowitz, The Deadly Ethnic Riot, pp. 272-277.

82 tensions.”23 For French theorist René Girard, such rituals are the keystone of society. By presenting a scapegoat for sacrifice, rituals prevent cycles of reciprocal violence, and, thereby, society maintains order.24 Historian David Nirenberg offers a rich example of the “channeling of conflict” thesis at work in medieval Spain. In his assessment, the anti-Jewish violence in Girona during Holy Week 1331 “was not as violent as at first appeared”:

All the participants showed their intention to act violently: weapons were displayed, insults were shouted, people took up aggressive stances. These were action sequences that if carried through seriously would have resulted in severe injury to the officials. In fact, the actions were restrained, punches were pulled, and only minor injuries or humiliations resulted. The participants seem to have been following informal protocols, or rules of engagement, that prevented 25 ! excessively brutal violence. Nirenberg concludes that by reading the Holy Week violence as a “ritual sacrifice,” we see that the events actually displaced serious brutality. “By alluding to and containing the original act of vengeance at the foundation of Christian-Jewish relations in the diaspora,” he notes, “Holy Week attacks flirted with but ultimately avoided the repetition of that violence in contemporary society.”26

Whereas the “channeling of conflict” thesis views rituals as conflict, the “ritualization of

23 Kertzer, Ritual, Politics, and Power, p. 131. Central here is Max Gluckman, Rituals of Rebellion in South-East Africa (Manchester: Manchester University Press, 1954); but see also Robert Dirks, “Annual Rituals of Conflict,” American Anthropologist, Vol. 90, No. 4 (December 1988), pp. 856-70; Bell, Ritual Theory, pp. 172-3; and Etzioni, “Toward a Theory of Public Ritual.” For a lively contemporary example, view videos of the daily ceremony of the closing the India-Pakistan border at Wagah widely available on YouTube.com. For a contrasting view, see James Scott, The Art of Not Being Governed: An Anarchist History of Upland Southeast Asia (New Haven: Yale University Press, 2010), p. 298, who argues of “rituals of reversal”—“a brief period [when] the normal social order is suspended or reversed”: “Far from being mere safety valves serving harmlessly to release tension, the better to impose hierarchy the rest of the year, these ritual sites have always been zones of struggle, threatening to spill over into actual revolt.”

24 René Girard, Violence and the Sacred, trans. Patrick Gregory (Baltimore: Johns Hopkins University Press, 1977).

25 David Nirenberg, Communities of Violence: Persecution of Minorities in the Middle Ages (Princeton, 1996), p. 210.

26 Ibid. p. 218.

83 conflict” thesis views conflict as ritual. The “ritualization of conflict” thesis is interested in explaining the recurrent features and patterns in violence. Tambiah calls this the “routinization and ritualization of collective violence,” arguing that these concepts “may help us to perceive some of the organized, anticipated, programmed, and recurring features and phases of seemingly spontaneous, chaotic, and orgiastic actions of the mob.”27 Violence, in this view, is understood as a performance with actors following a pre-written script. For example, anthropologist Marc

Gaborieau argues that “hostile relations [between Indian Hindus and Muslims]… are… acted out in stereotyped behaviors with an intense religious content, which can only be described as rituals.”28 These “rituals of provocation,” are “performed according to well defined procedures,” such as desecrating sacred symbols and sites belonging to the other group. The violence is so scripted that Gaborieau describes it as a “monotonous scenario… acted upon again and again.”29

In some societies, warfare is so highly standardized and symbolic that there is even a “special permanent site for the hostilities,” i.e., a stage. “In many of these cases,” notes Kertzer, “as soon as an individual is seriously wounded, hostilities cease and a round of post-battle ritual begins.”30

These two perspectives offer important insights on the causes and nature of violence, but they are distinct from the approach taken here.31 Rituals, in the the first view, are a substitute for

27 Tambiah, Leveling Crowds, p. 230. Also Davis, “Rites of Violence”; and Rogers Brubaker and David D. Laitin, “Ethnic and Nationalist Violence,” Annual Review of Sociology, Vol. 24 (1998), pp. 444-6.

28 Marc Gaborieau, “From Al-Beruni to Jinnah: Idiom, Ritual and Ideology of the Hindu-Muslim Confrontation in South Asia,” Anthropology Today, Vol. 1, No. 3 (June 1985), p. 9

29 Ibid., p. 10. For a recent political science approach to the performative aspects of violence, see Lee Ann Fujii, “The Puzzle of Extra-Lethal Violence,” Perspectives on Politics, Vol. 11, No. 2 (June 2013), pp. 410-26.

30 Kertzer, Ritual, Politics, and Power, p. 130. See also the discussion of dueling and other forms of limited, staged violence in Randall Collins, Violence: A Micro-Sociological Theory (Princeton: Princeton University Press, 2008), chap. 5, “Staging Fair Fights.”

31 This is not to argue that my view is incompatible with the existing views.

84 conflict and, in the second view, conflict becomes a standardized and repetitive (i.e., ritual-like), but in my view, contentious rituals are a form of conflict. They are a way in which groups express their rights and interests and make claims that bear on the rights and interests of others.

The modifier in contentious ritual thus has a double meaning. First is the plain meaning of

“controversial” or “likely to cause an argument.” Second is the meaning given by the contentious politics research program associated with McAdam, Tarrow, and Tilly, where contentious politics is defined as “interactions in which actors make claims bearing on someone else’s interest.”32

The performance, or attempted performance, of a contentious ritual thus in and of itself is an episode of political conflict, not a substitute for conflict or a description of the form that conflict takes.

Contentious rituals can exacerbate conflict or trigger violence, but the the ritual itself is not an act of violence. Maintaining this distinction requires that we distinguish between precipitants of violence and violence—or “sparks and fires”33—and between conflict and violence. This is important for our coming task of explaining participation, because contentious rituals do not carry the high risks to the individual of violence nor do they violate widely-held

32 Charles Tilly, Contentious Performances (New York: Cambridge University Press, 2008), p. 5; see also Doug McAdam, Sidney Tarrow, and Charles Tilly, Dynamics of Contention (New York: Cambridge University Press, 2001), p. 5. The DOC approach limits contentious politics to claim-making where the “governments appear either as targets, initiators of claims, or third parties” (Ibid.), but Snow and others have usefully criticized this as overly narrow. David A. Snow, “Social Movements as Challenges to Authority: Resistance to an Emerging Conceptual Hegemony,” Research in Social Movements, Conflicts, and Change, Vol. 25 (2004), pp. 3-25.

33 The metaphor is elaborated in Ashutosh Varshney and Joshua Gubler, “The State and Civil Society in Communal Violence: Sparks and Fires,” in Atul Kohli and Prerna Singh, eds., Routledge Handbook of Indian Politics (London: Routledge, 2013), pp. 155-166.

85 norms against committing property damage or bodily harm.34 As a result, contentious rituals, along with some other precipitants of violence, are instances of conflict without being instances of violence. In general, the literature overlooks ethnic conflict short of violence, but, as Varshney points out, distinguishing between ethnic violence and ethnic conflict is necessary since they are conceptually different and may require different explanations.35 Beyond emphasizing the conceptual distinction, focusing on non-violent forms of ethnic conflict, such as contentious rituals, is important to our understanding of the dynamics of ethnic relations in times of peace as in times of war. Given the immense ethnic diversity in the world, ethnic violence is remarkably low, and therefore explaining events that result in tension, but not necessarily violence, is crucial to our general understanding of conflict.36 !

Ritual as Collective Action

As the previous section on ritual and politics made clear, rituals can have a significant impact on the material world. Their impact is not limited to the symbolic or supernatural realms, but spans a diverse range of social and political outcomes: from individuals’ emotions and allegiances to relations between groups to the power of states. A characteristic common to many

34 On high-risk collective action, see Doug McAdam, “Recruitment to High-Risk Activism: The Case of Freedom Summer,” American Journal of Sociology, Vol. 92, No. 1 (July 1986), pp. 64-90; Mara Loveman, “High-Risk Collective Action: Defending Human Rights in Chile, Uruguay, and Argentina,” American Journal of Sociology, Vol. 104 No. 2 (September 1998), pp. 477-525; Jeff Goodwin and Steven Pfaff, “Emotion Work in High-Risk Social Movements: Managing Fear in the U.S. and East German Civil Rights Movements,” in Jeff Goodwin, James M. Jasper, and Francesca Polletta, eds., Passionate Politics: Emotions and Social Movements (Chicago: University of Chicago Press, 2001), pp. 282-302.

35 Ashutosh Varshney, “Ethnicity and Ethnic Conflict,” in Carles Boix and Susan Stokes, eds., The Oxford Handbook of Comparative Politics (New York: Oxford University Press, 2007), pp. 278-9; see also, Charles King, “The Micropolitics of Social Violence,” World Politics, Vol. 56, No. 3 (April 2004), p. 448; and Brubaker and Laitin, “Ethnic and Nationalist Violence.”

36 James D. Fearon and David D. Laitin, “Explaining Interethnic Cooperation,” American Political Science Review, Vol. 90, No. 4 (December 1996), pp. 715-35.

86 of these outcomes is that they produce a public benefit. A ritual that produces such a benefit, or good, is thus a collective action.

Rituals are not often discussed in the language of political economy, but in some ways many of them are usefully classed as collective action according to Olson’s definition.37 This is because many rituals involve multiple people working together to produce a non-excludable public good. Indeed, rituals tend to require participants being physically together,38 acting in physical synchrony—what historian William McNeill calls “muscular bonding.”39 The product of this action can be non-excludable collective, or public, goods. I find four collective goods that a ritual can provide: instrumental efficacy, an integrative social function, group representation, and inherent value.

First, many rituals are performed instrumentally in order to achieve an outcome desired by the community. Anthropologist Mary Douglas calls this “instrumental efficacy,” writing: “Of course Dinka hope that their rites will suspend the natural course of events. Of course they hope that rain rituals will cause rain, healing rituals avert death, harvest rituals produce crops.”40

Instrumental efficacy can thus provide a non-excludable good. To take one of Douglas’s

37 Mancur Olson, The Logic of Collective Action: Public Goods and the Theory of Groups (Cambridge: Harvard University Press, 1965). For a political-economic analysis of ritual, see Chwe, Rational Ritual.

38 See Randall Collins, Interaction Ritual Chains (Princeton: Princeton University Press, 2004), esp. pp. 53-64, where he provides an affirmative answer to the question, “Is bodily presence necessary?” Émile Durkheim, The Elementary Forms of Religious Life, trans. Karen E. Fields (New York: Free Press, 1995), p. 216, for example, describes that during the ritual phase of Australian life “the population comes together, concentrating itself at specified places for a period.”

39 William H. McNeill, Keeping Together in Time: Dance and Drill in Human History (Cambridge: Harvard University Press, 1995), chap. 1, “Muscular Bonding.” Recent psychology research studies ritual in the laboratory by enforcing physical synchrony for their subjects. See, e.g., Scott S. Wiltermuth and Chip Heath, “Synchrony and Cooperation,” Psychological Science, Vol. 20, No. 1 (January 2009), pp. 1-5; and Piercarlo Valdesolo and David DeSteno, “Synchrony and the Social Tuning of Compassion,” Emotion, Vol. 11, No. 2 (April 2011), pp. 262-6.

40 Mary Douglas, Purity and Danger: An Analysis of the Concepts of Pollution and Taboo (London: Routledge, 1996), p. 69.

87 examples, rain is a public good. It falls on the entire community, not just those members who participated in the rain ritual.41 Since rain is non-excludable, rituals performed to provide it are archetypical collective actions.

Second, many rituals provide a social function for the community that enacts it. The most prominent function attributed to rituals is social solidarity. The solidarity-building function of ritual is associated with Émile Durkheim and his followers. Durkheim argued that rituals counteract the push toward individualism created by participation in profane, economic activities. By bringing members of society together to take part in shared practices oriented at shared beliefs and symbols, rituals recreate society and generate strong bonds of solidarity among members. Only with such shared solidarity can social order be maintained.42 Amitai

Etzioni summarizes this position, arguing that rituals “serve to socialize members of a society as well as to reaffirm their commitments to values, and as such serve to sustain the integration of society.” Even rituals which appear to upend social norms in fact “contribute to reinforcement of shared beliefs and institutions indirectly, by releasing tension that results from conformity to societal beliefs and the behavioral prescriptions they entail.”43

Alfred Radcliffe-Brown provides a well-known functional account in his analysis of funeral rites among the Andaman Islanders. “For the society,” notes the anthropologist, “a death

41 That rain rituals may not in fact cause rain does not matter if the participants act with the belief and intention that they do.

42 Durkheim, Elementary Forms of Religious Life; as well as important elaborations, interpretations, and critiques, such as Robert N. Bellah, “Civil Religion in America,” Dædalus, Vol. 96, No. 1 (Winter 1967), pp. 1-21; Steven Lukes, Emile Durkheim: His Life and Work: A Historical and Critical Study (Stanford: Stanford University Press, 1973); Steven Lukes, “Political Ritual and Social Integration,” Sociology, Vol. 9 (1975), pp. 289-308; Kertzer, Ritual, Politics, and Power, esp. pp. 61-67; Roth “‘Men with Masks’”; Etzioni, “Toward a Theory of Public Ritual”; Robert N. Bellah, “The Ritual Roots of Society and Culture,” in Michelle Dillon, ed., Handbook of the Sociology of Religion (New York: Cambridge University Press, 2003), pp. 31-44; and Collins, Interaction Ritual Chains.

43 Etzioni, “Toward a Theory of Public Ritual,” p. 47.

88 is the loss of one of its members, one of its constituent parts…. [A person’s] death constitutes a partial destruction of the social cohesion, the normal social life is disorganized, the social equilibrium is disturbed.” The funeral is the way in which Andaman society “organise[s] itself anew and reach[es] a new condition of equilibrium.” The elaborate symbols and practices of the burial process are not for the dead; they are for the living community to reintegrate itself after suffering a collective trauma.44 In this way, the ritual provides a communally valued social function.

Third, rituals can represent a community to the outside world. As public and purposeful behavior, rituals, deliberately or not, often come to represent the community that performs them.

This is because rituals—like flags, monuments, and other tangible, visible symbols—are realized manifestations of the “imagined community.”45 Like these other totems, rituals portray the community as it wishes to be seen. One effect of this group representation is that rituals can help define the boundaries of the group.46 By representing who the group is, rituals also clarify who the group is not. Thus rituals help police the edges of the community, declaring who is in and who is out. The flip side of representing itself to the outside world is that ritual also represent the community to itself. As Douglas notes, “rituals enact the form of social relations and in giving these relations visible expression they enable people to know their own society.”47 Or as Lukes

44 Alfred R. Radcliffe-Brown, The Andaman Islanders: A Study in Social Anthropology (Cambridge: Cambridge University Press, 1922), p. 285.

45 Benedict Anderson, Imagined Communities: Reflections on the Origin and Spread of Nationalism, rev. ed. (London: Verso, 1991). Compare Hassner’s (War on Sacred Grounds, pp. 62-4) discussion of “sacred space as a social symbol,” where he notes that since sacred sites represent “a religious movement at its most splendorous,” they often come to serve “as the visual representation of a community” and are therefore often targeted for attack.

46 Anthony P. Cohen, The Symbolic Construction of Community (London: Routledge, 1985), pp. 50-63.

47 Douglas, Purity and Danger, p. 129.

89 argues, rituals “define as authoritative certain ways of seeing society.”48 By directing attention of both in-group and out-group members toward a particular vision of the group, rituals are powerful ways for representing a community.

Fourth, rituals can be communally valued not for what they do or provide, but simply for what they are. That is, rituals are often valued for their own sake, as meaningful moments in the life of the community and its members.49 People value the preservation of tradition, which has the capacity to bind each to their own personal past and the past, present, and future they share with the collective.50 What is more, rituals are often valued for being public celebrations, which are often fun and exciting. Rituals are community gatherings and a break from mundane and profane daily life.51 People enjoy the colors, music, song, dance, and other festive elements. The ritual, as a carnivalesque experience, is enjoyable and enriching in its own right. The inherent meaning and pleasure of rituals are thus a fourth collective good they provide. !

The Puzzle of Ritual Participation

If we conceptualize ritual as a form of collective action, we cannot take participation in ritual for granted. As Olson famously argued, rational individuals will not contribute to the

48 Lukes, “Political Rituals and Social Integration,” p. 301.

49 Paul Connerton, How Societies Remember (New York: Cambridge University Press, 1989), p. 45.

50 For a strong versions of this claim, see Mircea Eliade, Patterns in Comparative Religion, trans. Rosemary Sheed (New York: Sheed & Ward, 1958), pp. 392-393; and Claude Lévi-Strauss, The Savage Mind (Chicago: University of Chicago Press, 1966), pp. 236-237.

51 For a vivid example, see Durkheim, Elementary Forms, pp. 216-17; see also the evocative descriptions of crowd gatherings in Elias Canetti, Crowds and Power, trans. Carol Stewart (New York: Farrar, Straus, and Giroux, 1984), e.g., p. 18.

90 provision of public goods; they will free-ride.52 Framing rituals in this way highlights the dilemma of individual participation in a way left unquestioned by sociological, anthropological, and popular theories. It forces us to account for individual choice.53

In his foundational account of religious rituals, Durkheim disregards choice, describing rituals as if they somehow compel participation. For example, in The Elementary Forms of

Religious Life, he claims that “when a native is asked why he follows his rites, he replies that ancestors have always done so and that he must follow their example.”54 Building on Durkheim, cultural sociologist Jeffrey Alexander writes that in early societies “participation in ritual performance is not contingent, either for the actors or the observers. Participation is determined by the established and accepted hierarchies of gender and age, not by individual choices that respond to the sanctions and rewards of social powers or segmented social groups. Every relevant party in the band or tribe must attend to ritual performances.”55 And cognitive and evolutionary anthropologists Pierre Liénard and Pascal Boyer argue that ritual is characterized by feelings of “compulsion”: “given certain circumstances,” they claim, “people just feel that they

52 Olson, Logic of Collective Action.

53 Not all rituals face Olson’s collective action problem. For some rituals, even if they occur in public, the primary purpose is the production of a private good. For example, rites of passage are rituals designed to transition individuals from one stage of life or social role to another (Arnold van Gennep, The Rites of Passage, trans. Monika B. Vizedom and Gabrielle L. Caffee [Chicago: University of Chicago Press, 1960]). These rituals may involve a number of people acting together, but the central benefit is excludable and reserved for the participants. While such a ritual may provide public enjoyment, demarcate symbolic boundaries, or social integration—as the Radcliffe- Brown showed for Andaman funerals—the primary actors act out their roles for a private, selective benefit and the public benefit is secondary. Thus the dilemma of collective action is not a concern.

54 Durkheim, Elementary Forms, p. 192. Emphasis added.

55 Jeffrey C. Alexander, “Cultural Pragmatics: Social Performance Between Ritual and Strategy,” Sociological Theory, Vol. 22, No. 4 (December 2004), p. 535. Emphases added.

91 must perform a specific ritual, that it would be dangerous, unsafe, or improper not to do it.”56

These arguments mirror popular perceptions of rituals and their participants. The dominant image is that of mindless automatons worshipping their gods or saluting their leader—picture, for instance, the sea of Nazis at Nuremberg in Leni Riefenstahl’s 1935 film Triumph of the Will.57

But, rituals do not necessarily compel participation or brainwash. As religion scholar

Catherine Bell argues, “Ritual is never simply or solely a matter of routine, habit, or ‘the dead weight of tradition.’”58 Even when faced with a ritual, humans retain agency and individuals still confront the choice to participate or not. We, therefore, cannot treat participation in rituals as unthinking compliance. Olson’s free-rider problem means that participation in ritual is an outcome that must be explained.

The puzzle of ritual participation is compounded by characteristics of ritual that distinguish it from other collective action, and, in the process, complicate existing theories.

Rituals differ from canonical collective actions, such as labor strikes and peasant rebellions, in at least three critical ways: repetition, goal demotion, and participant perception.

Rituals are characterized by repetition. This alone is not unique: many collective actions are repeated. Unlike repeated contentious performances, however, rituals are repeated out of adherence to rules, not to achieve some extrinsic goal. Workers strike until wages rise and anti-

56 Pierre Liénard and Pascal Boyer, “Whence Collective Rituals? A Cultural Selection Model of Ritualized Behavior,” American Anthropologist, Vol. 108, No. 4 (December 2006), p. 816. Emphasis added.

57 As Mary Douglas, Natural Symbols: Explorations in Cosmology (London: Routledge, 1970), p. 1, bemoans, “Ritual is become [sic] a bad word signifying empty conformity…. Many sociologists, following [Robert K.] Merton, use the term ritualistic for one who performs external gestures without inner commitment to the ideas and values being expressed.” Religion scholar Jonathan Z. Smith, To Take Place: Toward Theory in Ritual (Chicago: University of Chicago Press, 1987), p. 102, attributes the general denigration of ritual in Western scholarship to the “Protestant insistence on the ‘emptiness’ of ritual.”

58 Bell, Ritual Theory, p. 92.

92 nuclear activists will protest until nuclear weapons are banned—i.e., it is an element of strategy.

For rituals, repetition is not strategic, it is the rule.59

This leads to the second distinction, “goal demotion,” meaning that “the actions are divorced from their usual goals.”60 As Liénard and Boyer explain, “Frequent repetition bolsters this intuition that [ritual] actions are disconnected from their ordinary goals.”61 Summarizing anthropologist Roy Rappaport, they place goal demotion first in their list of the “‘obvious’ (i.e., obvious to all anthropologists) aspects of ritual.”62 Elsewhere, they write that “the description of ritual action in terms of goals is either not available or in any case irrelevant.”63 Conversely, scholars normally understand collective action as goal-oriented. The goal may be internal or external, short-term or long-term, strategic or misguided, central or peripheral, but achieving it is a purpose of the action.64 In ritual, this is not necessarily the case. Philosopher Frits Staal famously argues this point in its extreme: “Ritual is pure activity, without meaning or goal,” he writes. “Things are either for their own sake, or for the sake of something else.… [M]y view is that ritual is for its own sake.”65

And third, ritual actors and observers view rituals as a distinct type of action. This

59 For an overview of rituals as rule-governed activity, see Bell, Ritual, pp. 153-5.

60 Liénard and Boyer, “Whence Collective Rituals?,” p. 815. Also Humprey and Laidlaw, The Archetypal Actions of Ritual.

61 Liénard and Boyer, “Whence Collective Rituals?,” p. 816.

62 Ibid., referencing Roy A. Rappaport, Ecology, Meaning and Religion (Berkeley: North Atlantic Books, 1979)

63 Pascal Boyer and Pierre Liénard, “Why Ritualized Behavior? Precaution Systems and Action Parsing in Developmental, Pathological and Cultural Rituals,” Behavioral and Brain Sciences, Vol. 29, No. 6 (2006), p. 11.

64 “Social movement scholars generally agree that protest is undertaken for the purpose of achieving some goal,” writes Rachel L. Einwohner, “Identity Work and Collective Action in a Repressive Context: Jewish Resistance on the ‘Aryan Side’ of the Warsaw Ghetto,” Social Problems, Vol. 53, No. 1 (February 2006), p. 40.

65 Frits Staal, “The Meaninglessness of Ritual,” Numen, Vol. 26, No. 1 (1979), p. 9.

93 element of perceived distinction is central to many theories of ritual. For instance, Durkheim distinguishes between the “two different phases” of Australian life not only in terms of their purposes—the economic phase and the religious phase—but in terms of their mood: the former is

“monotonous, slack, and humdrum,” the latter is a period of wild, unbridled excitement.66 Victor

Turner argues that rituals are characterized by liminality and the production of communitas, an intense sense of community not felt during normal life.67 The deliberate repetition and goal demotion already discussed—along with other elements such as formalism, invariance, and symbolism—play a part in facilitating the experience of this distinction. As Humphrey and

Laidlaw remark, “the participants [of an ordinary meal and a ritual meal]… think that something distinguishes the two events; and… the observable differences often include rules and practices.”68 Catherine Bell refers to this technique of emphasizing difference as “ritualization,”

“the way in which certain social actions strategically distinguish themselves in relation to other actions.” It is “a way of acting that is designed and orchestrated to distinguish and privilege what is being done in comparison to other, usually more quotidian, activities.”69 Thus the perception of difference may be a deliberate construct, but it is a powerful one nonetheless.

The power of successful ritualization is evident in the views of ordinary people, who often recognize rituals as distinct. The beliefs of the people who organize, participate in, observe,

66 Durkheim, Elementary Forms, pp. 216-17.

67 Victor Turner, The Ritual Process: Structure and Anti-Structure (Ithaca: Cornell University Press, 1969).

68 Humphrey and Laidlaw, The Archetypal Actions of Ritual, p. 73; also, p. 5. Also J. David Knottnerus, “Religion, Ritual, and Collective Emotion,” in Christian von Scheve and Mikko Salmela, eds., Collective Emotions (New York: Oxford University Press, 2014), p. 313.

69 Bell, Ritual Theory, p. 74. Therefore, “formality, fixity, and repetition are not intrinsic qualities of ritual so much as they are frequent, but not universal strategies for producing ritualized acts.” Ibid., p. 92. See also Smith, To Take Place, p. 109: “Ritual is, above all, an assertion of difference.”

94 and ignore rituals matter because these views may affect the decisions that people make regarding rituals. And many people see rituals as a time apart, in some way different from normal life. We see this when, for example, a pastor calls a worship service “different from every other gathering” or when Jews ask annually during the Passover Seder, “Why is this night different from all other nights?”70 Participants view ritual practices as distinct from how they live the rest of their lives, including participating in non-ritual collective action. Onlookers, too, often see rituals as unique. As Bell comments, “even a particularly dense foreigner dropped into the middle of things would not mistake a Shī‘ī procession or a Korean kut for just another routine event in the daily life in those communities.”71 For these three reasons—repetition, goal demotion, and perceived difference—it is valuable to understand ritual as a distinct and

“particular type of social action” with “properties that distinguish [it] from other types of action.”72

Consequently, rituals cannot be fully explained with theories of participation developed with other behaviors in mind. Prominent theoretical approaches in political science understand most human actions through a narrowly rationalist and strategic lens that requires an analytic distinction between desired ends (benefits) and the means used to achieve them (costs).73 But, as the previous paragraph adumbrated, in ritual, means and ends are not always neatly separable. As

70 The pastor’s quote is from Timothy J. Nelson, Every Time I Feel the Spirit: Religious Experience and Ritual in an African-American Church (New York: NYU Press, 2005), p. 59.

71 Bell, Ritual, p. 138.

72 Roth, “‘Men Wearing Masks,’” p. 320. Though note that there are other collective actions which approach ritualization. Social and political actions which are often considered expressive, for example protest or voting, take on, to greater and lesser degrees, these three features of ritual.

73 Albert O. Hirschman, Shifting Involvements: Private Interest and Public Action (Princeton: Princeton University Press, 1982), pp. 82-91, esp. pp. 85-6; and Jasper, The Art of Moral Protest, pp. 23-24, 26.

95 Liénard and Boyer suggest, “the standard connections between means and ends seem broken.”74

Further, an implicit assumption of rational choice theory is that people approach all actions in the same way.75 But, as noted above, people explicitly approach rituals differently. To fully explain participation in rituals, we require a theory that can account for these peculiar features. In the next section, I sketch such a theory. !

A Theory of Participation in Contentious Rituals

To generate a theory of participation in contentious ritual, I begin with two fundamental insights from the sociology and anthropology of ritual and religious studies. The first insight is that rituals are not “merely symbolic” reflections of reality; for participants, rituals do things.

Specifically, rituals provide benefits for participants such as collective effervescence, emotional energy, a sense of belonging, and meaningful interpretations of the world.76 This helps us explain ritual participation generally. The second insight is that rituals are multi-vocal and ambiguous.

They do not have a fixed meaning across time or individuals. This helps us explain participation in contentious rituals specifically. Together these two observations suggest that rituals provide valued experiences for participants and that these experiences are not necessarily the same for every participant.

74 Liénard and Boyer, “Whence Collective Rituals?,” p. 816.

75 For example, Anthony Downs, An Economic Theory of Democracy (New York: Harper, 1957), pp. 7-8, states that people approach “every situation with one eye on the gains to be had, the other eye on costs, a delicate ability to balance them, and a strong desire to follow wherever rationality leads” (emphasis added; quoted in Donald P. Green and Ian Shapiro, Pathologies of Rational Choice: A Critique of Applications in Political Science [New Haven: Yale University Press, 1994], p. 35).

76 Additionally, some rituals, particularly religious ones, provide an array of empirically unverifiable outcomes for participants: they cause rain, achieve healing, and so on.

96 The Benefits of Ritual

The insight that rituals provide benefits for participants originates in the earliest days of the social scientific study of ritual. Since then, scholars have identified a range of effects that rituals have. Some rituals have a social effect for the participant. For example, Arnold van

Gennep showed that rituals permit people to transition to a new social identity, such as from childhood to adulthood. Rites of passage both signal this change and, more importantly, cause it, thereby granting the person a new social role.77 Rituals can also have a cognitive effect. They shape how participants understand the world. For instance, Kertzer states that rituals have a

“cognitive effect on people’s definition of political reality.”78 Steven Lukes elaborates this position, arguing: “political ritual should be seen as reinforcing, recreating, and organizing représentations collectives… In this sense, such ritual plays… a cognitive role, rendering intelligible society and social relationships, serving to organize people’s knowledge of the past and present and their capacity to imagine the future.”79

Rituals also have an emotional effect on participants. Durkheim argues that collective rituals produce the experience of “collective effervescence” in participants. Rituals cause

“passions… so torrential that nothing can hold them,” “an intense hyperexcitement of physical and mental life,” and “a state of exaltation” that transport the participant to “a special world inhabited by exceptionally intense forces that invade and transform him” and “excite him to the

77 van Gennep, Rites of Passage.

78 Kertzer, Ritual, Politics, and Power, p. 14.

79 Lukes, “Political Rituals and Social Integration,” p. 301.

97 point of frenzy.”80 Building on Durkheim, Randall Collins elaborates the micro-situational foundations of rituals’ effects. He writes:

The central mechanism of interaction ritual theory is that occasions that combine a high degree of mutual focus of attention… together with a high entrainment— through bodily synchronization, mutual stimulation/arousal of participants’ nervous systems—result in feelings of membership that are attached to cognitive symbols; and result also in the emotional energy of individual participants, giving them feelings of confidence, enthusiasm, and desire for action in what they consider a morally proper path. These moments of high degree of ritual intensity 81 !are high points of experience. These emotions, as Collins argues, come from being together, moving together, and acting together—all elements that rituals combine. Emotions are further enhanced by music, which is a common feature in ritual. As Jasper comments, “music has a strong emotional impact on participants who sing, dance, and move together.”82

Recent psychological research supports these sociological claims. Experiments that simulate rituals through bodily synchrony have demonstrated a number of relevant effects. For example, Valdesolo and DeSteno find that synchronized movement fosters feelings of social connection and compassion.83 Studies by Valdesolo, Ouyang, and DeSteno and Wiltermuth and

80 Durkheim, Elementary Forms, pp. 218 and 220.

81 Collins, Interaction Ritual Chains, p. 42; also, Randall Collins, “Social Movements and the Focus of Emotional Attention,” in Goodwin, Jasper, and Polletta, eds., Passionate Politics, pp. 27-43; and Randall Collins, “Interaction Ritual Chains and Collective Effervescence,” in von Scheve and Salmela, eds., Collective Emotions, pp. 299-311. See also Knottnerus’s “structural ritualization theory,” which builds on Collins, in Knottnerus, “Religion, Ritual, and Collective Emotion”; and J. David Knottnerus, “Collective Events, Rituals, and Emotions,” in Shane R. Thye and Edward J. Lawler, eds., Advances in Group Processes, Vol. 27 (Bingley, UK: Emerald Group Publishing Limited, 2010), pp. 39-61.

82 James M. Jasper, “Emotions and Social Movements: Twenty Years of Theory and Research,” Annual Review of Sociology, Vol. 37 (2011), p. 294. On music and collective action more generally, see Ron Eyerman and Andrew Jamison, Music and Social Movements: Mobilizing Traditions in the Twentieth Century (New York: Cambridge University Press, 1998); and William G. Roy, “How Social Movements Do Culture,” International Journal of Politics, Culture, and Society, Vol. 23, Nos. 2-3 (September 2010), pp. 85-98.

83 Valdesolo and DeSteno, “Synchrony and the Social Tuning of Compassion.”

98 Heath show that synchrony increases in-group cooperation.84 And Konvalinka et al. find that high-risk rituals can even cause physiological arousal in some spectators.85

What is more, rituals are meaningful experiences for participants. In particular, rituals allow participants unmediated access to a cherished, often mythic, past. This can happen in two ways. First, rituals, because of their repetition, imply “continuity with the past.”86 As a result, the participant feels him or herself following the path of their ancestors, walking in their footsteps and reproducing their actions. This is especially potent when the ritual has been passed down within families or in commemorative rituals which “do not simply imply continuity with the past but explicitly claim such continuity.”87 Second, rituals can make participants feel as if they are walking alongside ancestors and participating with them. Scholar of religion Mircea Eliade writes, “Every religious festival… represents the reactualization of a sacred event that took place in a mythical past, ‘in the beginning.’ Religious participation in a festival implies emerging from ordinary temporal duration and reintegration of the mythical time reactualized by the festival itself.”88 Though Eliade is referring specifically to religion, the same can apply to secular rituals as well. Ethnic, national, and political rituals also often celebrate a mythic past that is

84 Piercarlo Valdesolo, Jennifer Ouyang, and David DeSteno, “The Rhythm of Joint Action: Synchrony Promotes Cooperative Ability,” Journal of Experimental Social Psychology, Vol. 46, No. 4 (July 2010), pp. 693-5; and Wiltermuth and Heath, “Synchrony and Cooperation.” This includes cooperation in aggressive and destructive behavior: Scott S. Wiltermuth, “Synchronous Activity Boosts Compliance with Requests to Aggress,” Journal of Experimental Social Psychology, Vol. 48, No. 1 (January 2012), pp. 453-6; and Scott Wiltermuth, “Synchrony and Destructive Obedience,” Social Influence, Vol. 7, No. 2 (2012), pp. 78-89.

85 Ivana Konvalinka, Dimitris Xygalatas, Joseph Bulbulia, Uffe Schjødt, Else-Marie Jegindø, Sebastian Wallot, Guy Van Orden, and Andreas Roepstorff, “Synchronized Arousal Between Performers and Related Spectators in a Fire- Walking Ritual,” Proceedings of the National Academy of Science, Vol. 108, No. 20 (May 17, 2011), pp. 8514-9.

86 Connerton, How Societies Remember, p. 45.

87 Ibid.

88 Mircea Eliade, The Sacred and the Profane: The Nature of Religion, trans. Willard R. Trask (Orlando: Harcourt, 1959), pp. 68-9.

99 “reactualized” through the ritual’s performance.89

The result of all these effects of ritual is that participation is often meaningful and pleasurable. Connerton, for instance, argues, “Rites have capacity to give value and meaning to the life of those who perform them.”90 And Kertzer, in his wide-ranging study of rituals and politics, notes, “People derive a great deal of satisfaction from their participation in ritual.”91 The crowds, sounds, movement, purpose, symbols, solidarity, attention, effervescence, and so on come together in ritual, creating moments of great experience for participants.

All of the effects of rituals just identified share an important feature: they are “intrinsic to the process of participation itself” and not reliant on the successful achievement of a particular outcome.92 Therefore, participation is not motivated by expectations of outcome benefits—the benefits to the individual that result from achieving the goal of the collective action—but by expectations of process-regarding, or intrinsic, benefits which are internal and inherent to the process of acting collectively.93 Whereas outcome benefits, whether collective or selective, are determined by the outcome of the collective action, process benefits are the satisfactions gained from participation in collective action itself and not in the action’s consequences. Process

89 Connerton, How Societies Remember; Hobsbawm and Ranger, The Invention of Tradition; John R. Gillis, ed. Commemorations: The Politics of National Identity (Princeton: Princeton University Press, 1996); Yael Zerubavel, Recovered Roots: Collective Memory and the Making of Israeli National Tradition (Chicago: University of Chicago Press, 1995); Anthony D. Smith, The Ethnic Origins of Nations (Oxford: Blackwell, 1986).

90 Connerton, How Societies Remember, p. 45.

91 Kertzer, Ritual, Politics, and Power, p. 14.

92 Elisabeth Jean Wood, Insurgent Collective Action and Civil War in El Salvador (New York: Cambridge University Press, 2003), p. 240.

93 See Jon Elster, The Cement of Society: A Study of Social Order (New York: Cambridge University Press, 1989), pp. 34-46; Kay Lehman Schlozman, Sidney Verba, and Henry E. Brady, “Participation’s Not a Paradox: The View from American Activists,” British Journal of Political Science, Vol. 25, No. 1 (January 1995), pp. 1-36; Jasper, Art of Moral Protest; Wood, Insurgent Collective Action, pp. 231-56.

100 benefits are still a type of instrumental motive and fit the future-oriented calculation of doing X in order to achieve Y. However, the Y of process-regarding behavior is intrinsic to the action. So to act on process-regarding motives is to act to obtain the benefits intrinsic in the collective action, such as the excitement of collective effervescence and emotional energy or the satisfaction of “express[ing] allegiance to moral visions.”94 My argument, therefore, is not that participants are irrational; they are simply acting to receive a different type of benefit. The benefit is participation, not its external consequences.

Importantly, process benefits are exclusively available to participants. Only participants receive the social, cognitive, emotional, and other effects that rituals produce. Process-oriented benefits thus share a characteristic of selective incentives. Yet they are not the same as nor are they reducible to selective incentives. Selective incentives are a consequence of action. In other words, participation is the means and selective incentives are the ends. Further, action can be analytically distinguished from the incentives: action and reward are separate. Process benefits, in contrast, emerge from the acting, they are of the action. With process benefits we cannot distinguish means and ends: the action is the reward. And as Elisabeth Wood argues, “Reasons that refer irreducibly to the process itself or to a non instrumental value and not to the outcomes of action are problematic because they challenge the consequentialist framework of rational choice.”95 Describing process benefits as a type of selective benefits (as classically conceived) is not useful.

Intrinsic benefits are not unique to rituals. Scholars have found them to motivate a wide

94 Jasper, Art of Moral Protest, p. 14.

95 Wood, Insurgent Collective Action, p. 253.

101 range of collective behavior, especially when the actions or their contexts lack the usual conditions conducive to collective contention. For example, Wood finds that during the civil war in El Salvador, both nonparticipants and participants received access to liberated land, but acts of rebellion provided the pride and pleasure in agency only to campesinos who participated.96

Rachel Einwohner finds that the Warsaw Ghetto Uprising took place under conditions of reduced political opportunity, but resistance gave Warsaw Jews the occasion to assert their dignity and honor in the face of certain death.97 A number of scholars have attributed the choice to vote to

“expressive motives,” since it is otherwise irrational to go to the polls.98 But in all of these cases, there is no ex ante reason to suspect that process-oriented motives would trump outcome- oriented, instrumental ones. Generally, scholars find expressive motives only when all else fails.

By pointing to specific features of ritual, my argument suggests an a priori justification of when to expect intrinsic motives to matter.

Though process benefits can motivate many types of behaviors, they are especially associated with rituals because rituals are a non-instrumental form of action.99 It is not the case that rituals cannot have instrumental uses or be instrumentally effective. Rather, what

96 Ibid., pp. 231-56; and Elisabeth Jean Wood, “The Emotional Benefits of Insurgency in El Salvador,” in Goodwin, Jasper, and Polletta, eds., Passionate Politics, pp. 267-80.

97 Rachel L. Einwohner, “Opportunity, Honor, and Action in the Warsaw Ghetto Uprising of 1943,” American Journal of Sociology, Vol. 109, No. 3 (November 2003), pp. 650-75.

98 The claim originates in Downs, Economic Theory of Democracy, p. 48. A large literature has followed, e.g., Geoffrey Brennan and Loren Lomasky, Democracy and Decision: The Pure Theory of Electoral Preference (New York: Cambridge University Press, 1993); Alexander A. Schuessler, A Logic of Expressive Choice (Princeton: Princeton University Press, 2000); Alan Hamlin and Colin Jennings, “Expressive Political Behaviour: Foundations, Scope and Implications,” British Journal of Political Science, Vol. 41, No. 3 (July 2011), pp. 645-70. But see Amit Ahuja and Pradeep Chhibber, “Why the Poor Vote in India: ‘If I Don’t Vote, I Am Dead to the State,” Studies in Comparative International Development, Vol. 47, No. 4 (December 2012), pp. 389-410.

99 Some scholars view ritual not as a type of action, but as an aspect of all action. But even in this schema, ritual is generally seen as the noninstrumental aspect of action and is contrasted with the instrumental aspects of action. See, for instance, Leach, Political Systems of Highland Burma, p. 13.

102 distinguishes ritual from “ordinary functional action” or “everyday action” is that ritual departs from the “normal intentional character” of the latter.100 I draw here on the important theoretical work of anthropologists Caroline Humphrey and James Laidlaw, developed in their study of the

Jain puja ritual in India. Though I differ with their argument that ritual is a quality of action, rather than a class of action, I find their general theory illuminating.

Humphrey and Laidlaw “contrast ritualized action… with action which is not ritualized” and conclude that “ritualization severs the link, present in everyday action, between the

‘intentional meaning’ of the agent and the identity of the act which he or she performs.”101 To explain: non-ritualized action is defined by the actor’s intention. Thus, to use an example of theirs, what makes my arm movement a beckon is that I intend to beckon my friend (i.e., beckoning is my intentional meaning when I move my arm). But for ritualized action, this is not the case. For ritualized action, the normal relationship between the intention and the act is deliberately transformed by the “adoption of a ritual stance.”102 The result of this stance is that

“intentions no longer play the immediate role which they normally do in determining the identity of the acts performed.”103 This is because ritual is prescribed action. When someone performs a ritual, they are following a set of prescribed rules; thus what matters is that the rules are followed, not the intentions behind the performance. Note that this does not mean that the ritual is unintentional or accidental. The actor does have the intention to carry out the ritual, but that

100 Humphrey and Laidlaw, Archetypal Actions of Ritual, pp. 12, 13, 89.

101 Ibid., p. 2.

102 Ibid., p. 94.

103 Ibid.

103 entails the dismissal of their own wants and desires. As a result, the acts which compose a ritual

“are constituted not by the intentions which the actor has in performing them, but by prior stipulation. We thus have,” Humphrey and Laidlaw conclude, “a class of acts in which the intentions which normally serve to identify acts… are discounted.”104

The overall insight from Humphrey and Laidlaw is that ritual lacks the “normal intentional character” of day-to-day action.105 An implication of this understanding of ritual is that the rewards of taking part in a ritual are internal and intrinsic—the consequences of ritual are inconsequential. As Staal argues, in “the applied activities of our ordinary, everyday life,” the results, not the rules, are what matter. In ritual, conversely, “the rules count, but not the results.”

Thus, he states, “whatever value [ritual] has is intrinsic value.”106

When an actor chooses to partake in ritual, she therefore does so on account of the process-oriented benefits inherent in acting ritually. The appeals of ritual are not in the consequences of ritual action; they are the ritual action. !

Ritual Ambiguity

Thus far, our theory addresses rituals in general; with insight is that rituals are multi- vocal and therefore their meaning is ambiguous, we narrow our focus to contentious rituals. By

104 Ibid., p. 97.

105 Ibid., p. 89.

106 Staal, “The Meaninglessness of Ritual,” p. 9.

104 multi-vocality, I mean that “a single symbol may stand for many things.”107 Consider, for instance, all the things that the American flag stands for: a state, a territory, a people, patriotism, militarism, liberty, oppression, opportunity, inequality, and so on. This is particularly the case because of the capacity of symbols to condense many meanings and represent many things in a single form.108 The result is that even a “commonly accepted symbol” holds a “range of meanings”109 and, therefore, different people can understand the exact same symbol in different ways.

Since a ritual can represent so much, the meaning of ritual is ambiguous. This ambiguity is, in fact, a key to their utility. “Symbols are effective because they are imprecise,” as Anthony

Cohen argues.110 The ambiguous character of symbols can also be politically useful, as shown by

Murray Edelman.111 It allows different groups and constituencies to see different things in the symbol or ritual.

Contentious rituals, like all rituals, do not have a fixed or definitive meaning. This is perhaps more obvious in contentious rituals than non-contentious ones, since differing

107 Turner, Forest of Symbols, p. 50. He adds, “This property of individual symbols is true of ritual as a whole.” Political scientists have recently gained interest in the multi-vocality of religion, and especially religious doctrine. See Alfred Stepan, “Religion, Democracy, and the ‘Twin Tolerations,’” Journal of Democracy, Vol. 11, No. 4 (October 2000), pp. 37-57; Alfred Stepan, Arguing Comparative Politics (New York: Oxford University Press, 2001), ch. 11, “The World’s Religious Systems and Democracy: Crafting the ‘Twin Tolerations,’” pp. 213-53; Daniel Philpott, “Explaining the Political Ambivalence of Religion,” American Political Science Review, Vol. 101, No. 3 (2007), pp. 505-25; Daniel Corstange, “Religion, Pluralism, and Iconography in the Public Sphere: Theory and Evidence From Lebanon,” World Politics, Vol. 64, No. 1 (January 2012), pp. 116-60.

108 Turner, Forest of Symbols, p. 28; Murray Edelman, The Symbolic Uses of Politics, rev. ed. (Champaign: University of Illinois Press, 1985), p. 6.

109 Cohen, Symbolic Construction of Community, p. 15. Original emphasis.

110 Ibid., p. 21.

111 Edelman, Symbolic Uses of Politics, esp. pp. 206-8. Kertzer, Ritual, Politics, and Power, pp. 69-75, explains the “virtues of ambiguity.”

105 interpretations of a ritual is often the source of the dispute. Consider, for instance, the many disputed commemorations of figures considered heroes by some and villains by others.

Yet this basic point about multi-vocality and ambiguity is often lost on observers trying to explain the behavior of participants. Analysts often begin their analysis with an argument about what the ritual means or is really about.112 They thus collapse the multiple meanings of the ritual into a single interpretation—often one that the participants themselves would not recognize.

Contentious rituals, in such an analysis, are often viewed as intimidating demonstrations of power or deliberately inflammatory acts. Since we further tend to assume that action is ends- oriented, the conclusion is that participants act in order to intimidate or provoke. This double assumption is problematic on both levels. As I argued earlier, we cannot conflate outcomes and intentions. I now wish to highlight that we cannot project our interpretations of a ritual—no matter how obvious it may appear—onto the participants. The meanings of a ritual are manifold, and the one held by the analyst is just one of many possibilities.

Two people can observe the same ritual and believe completely different things about it.

For example, person A, a participant, may view ritual R as a celebration of national liberation while person B, from a different ethnic group, may view R as a mean-spirited display of chauvinism and a symbol of national oppression. This is particularly true in deeply divided societies, where “value consensus is manifestly absent or minimal.”113 It is entirely logical that people on different sides of an ethnic divide have different understandings of a symbolically

112 Ziad W. Munson, “When a Funeral Isn't Just a Funeral: The Layered Meaning of Everyday Action,” in Nancy T. Ammerman, ed., Everyday Religion: Observing Modern Religious Lives (New York: Oxford University Press, 2007), pp. 122.

113 Lukes, “Political Ritual and Social Integration,” p. 300.

106 ambiguous act.

There are three implications of multi-vocality and ambiguity for a theory of participation.

First, even if opposition to the contentious ritual is well known, it does not necessarily impact the meaning it holds for participants or their intentions in acting. Ambiguity means that the opponents’ interpretation is not universal or definitive. This is bolstered by the fact that in divided societies, members of the different groups tend not to have cross-cutting social networks.

Thus, there is no risk in accidentally offending a friend or neighbor.114

Second, amongst the participants themselves, individuals may hold differing interpretations of the ritual and therefore act for different reasons.115 The ritual’s multi-vocality means that participants can find their own meaning from the range of possibilities. There is no one hegemonic interpretation of the ritual held by all participants.

Third, the ritual’s participants and it organizers or elites may not share interpretations or intentions. It is plausible that organizers and political elites use contentious rituals to polarize society as argued by Wilkinson and others.116 But that does not mean that ordinary participants see it in the same way. Participants, thus, can have their own understanding that is independent of in-group elites, out-group opponents, or even fellow participants. ! !

114 For the opposite social situation, see Theodore Caplow, “Rule Enforcement Without Visible Means: Christmas Gift Giving in Middletown,” American Journal of Sociology, Vol. 89, No. 6 (May 1984), p. 1321, and the discussion of Caplow in Ann Swidler, “Cultural Power and Social Movements,” in Hank Johnston and Bert Klandermans, eds., Social Movements and Culture (Minneapolis: University of Minnesota Press, 1995), pp. 32-33.

115 Cohen, Symbolic Construction of Community, p. 55.

116 Wilkinson, Votes and Violence.

107 Theorizing Loyalist Parades as Contentious Rituals

The theory I have just presented provides a general explanation for participation in contentious rituals, the set of symbolic acts that make contested political claims. The empirical portion of this dissertation focuses on a specific contentious ritual: loyalist parading in Northern

Ireland. In this section I will establish that loyalist parades are contentious rituals and collective actions, highlighting why participation in them is puzzling.

The first crucial point is that loyalist parades are rituals. In fact, they are the most important political or commemorative ritual in the Protestant community of Northern Ireland.117

Recall that in Kertzer’s definition, rituals have three characteristics: they are “action wrapped in a web of symbolism,” they “follow highly structured, standardized sequences,” and they are repetitive.118 Parades meet all three criteria. They are highly symbolic behavior, with everything from the clothing, regalia, and banners to the music, names of the organizations, and the parade routes full of meaning and significance. Parades are socially standardized, following rules of behavior that are well established in society. And parades, are repetitive: they occur year after year, often on dates of historical significance.

Loyalist parades are also at the center of great political contestation in Northern Ireland, making them quintessential contentious rituals.119 Parades are disputed by Catholic nationalists who see them as provocative displays of Protestant chauvinism. In particular, nationalists protest

117 Neil Jarman, Material Conflicts: Parades and Visual Displays in Northern Ireland (Oxford: Berg, 1997); and Dominic Bryan, Orange Parades: The Politics of Ritual, Tradition, and Control (London: Pluto, 2000).

118 Kertzer, Rituals, Politics, and Power, p. 9.

119 See Ross, Cultural Contestation in Ethnic Conflict; Brubaker and Laitin, “Ethnic and Nationalist Violence,” p. 445; and Andreas Wimmer and Conrad Schetter, “Ethnic Violence,” in Wilhelm Heitmeyer and John Hagan, eds., International Handbook of Violence Research (Dordecht, Netherlands: Kluwer Academic Publishers, 2003), p. 256.

108 parades that pass by their neighborhoods or cultural and religious institutions. This cycle of parade and protest raises tensions and fear between groups and has an overall negative impact on community relations and the consolidation of peace. Moreover, the confrontations between protesters, parades, and the police occasionally become violent.

These political outcomes of parades suggest that they provide collective goods for the

Protestant community at large, not just participants. Therefore, they are collective actions. I count at least eleven collective goods that parades are thought to produce. First, parades publicly represent the Protestant community and its values. Second, parades are a tradition that is valued by much of the Protestant community, whether they participate or not. Third, parades project

Protestant power. Fourth, parades mark territory as Protestant. Fifth, parades strengthen and maintain Protestant culture. Sixth, parades perform a socially integrative function. Seventh, parades maintain social memory and ethnic myths. Eighth, parades police the symbolic boundaries of the group. Ninth, parades bolster Protestant unity. Tenth, parades are a festival for anyone to enjoy. And eleventh, parades confront and antagonize Irish nationalists. Not all of these are desired by the Protestant community or even necessarily by the participants, but, once produced by parades, they can affect all members of the community.

As a result, the choice to participate in loyalist parades must be explained.120 Olson’s theoretical challenge means that we cannot simply assume that anyone will participate in

120 By “participate,” I mean marching in a parade as a member of a loyal order or marching band. Parades are large heterogenous events composed of multiple actors and groups, all of which take part in their own way (including supporters, protestors, police, journalists, and passersby). But I focus on the people who actually parade through the streets. On the many actors involved in demonstrations, see Olivier Fillieule, “The Independent Psychological Effects of Participation in Demonstrations,” Mobilization, Vol. 17, No. 3 (September 2012), p. 236; also, Olivier Fillieule and Danielle Tartakowsky, Demonstrations, Phyllis Aronoff and Howard Scott, trans. (Halifax: Fernwood, 2013), pp. 16-17. For parades, it is especially important to clarify this distinction, since some scholars of ritual include audience members as participants. Berezin, Making the Fascist Self, p. 250; and Bocock, Ritual in Industrial Society, p. 59.

109 parades. Further, the fact that the action is contentious and holds the potential for socially harmful consequences makes the puzzle more vexing. If conflict does break out, the ordinary participants are among the most likely to be harmed. Why would anyone want to act in a way that risks such conflict? In the previous section, I provided an explanation. In the next section, I will draw out the observable implications of that explanation for Northern Irish parades. !

Observable Implications

Our argument provides several testable implications for parades in Northern Ireland. The primary implication of the “ritual benefit” argument is that loyalist parade participants should understand their behavior as non-instrumental. When participants discuss parades they should not view them instrumentally as a means to an end—as expected by the ethnic rivalry and collective action approaches I will discuss shortly. Instead, participants’ discussions should center on the value of participation itself. Their descriptions of parades, therefore, should minimize or even neglect the outcomes of parades (exciting the Protestant community, intimidating the Catholic community, receiving a material benefit, etc.). Rather, they should highlight the experience of participation: what it is like to march through the streets. Relatedly, they should not view participation as a cost or burden, but as a benefit in and of itself—despite the fact that it requires significant time and resources. If participants view their actions in instrumental terms, the “ritual benefit” argument seems less plausible.

The implications of the “ritual ambiguity” argument are attitudinal. The multiple available meanings imply that Protestants and Catholics can maintain firm and opposing interpretations of parades, each valid. If, in fact, Protestants do not view parades as hostile

110 attempts to offend and provoke Catholics, then there is no reason to expect that participants hold any more ill will toward Catholics than nonparticipants. Therefore, an implication of ritual’s ambiguity is that parade participants should express the same level of sectarian attitudes as nonparticipants. Due to ritual ambiguity, the fact that Catholics view parades as hateful does not mean that Protestants do too. !

Alternative Explanations

There are four bodies of scholarship which might explain participation in contentious rituals. In this section, I describe them and suggest problems with each approach. I begin with a review rationalist approaches to rituals, and then I address three explanations for ethnic conflict: elite manipulation, ethnic rivalry, and collective action. I conclude that only the latter two provide a plausible explanation for participation and then delineate the hypotheses each argument generates. !

Rationalist Approaches to Ritual

Though ritual was long defined in opposition to rationality, several political scientists have recently worked to explain rituals with a rationalist framework. In Rational Rituals,

Michael Chwe translates rituals into terms more familiar to the discipline. He argues that the source of the political and social importance of rituals is their public format. This publicness allows rituals to create common knowledge––knowledge of the form “we all know that we all know”––which is essential to solving coordination problems. Applying these simple ideas from game theory to an extensive range of examples, Chwe provides a convincing argument that the

111 form of rituals is a key to their power.121

Building on Chwe’s model, David Patel and Alfred Stepan explain the role of rituals in the politics of Iraq and Senegal, respectively. Patel argues that the political importance of Islam stems from the capacity of Islamic rituals to generate common knowledge and coordinate the expectations of individual Muslims.122 In particular, he shows that Friday worship at mosque played two coordinating functions: it allowed localities with only one mosque to maintain social order and it facilitated the creation of a nation-wide Shiite identity and political agenda. Stepan argues that public rituals of respect in Senegal have helped create a democratic and tolerant polity.123 He first shows that horizontal rituals of respect between different religions foster mutual tolerance between groups. Second, he shows that vertical rituals of respect between the state and society have allowed for the creation and maintenance of the “twin tolerations”–– religion’s tolerance of the state and the state’s tolerance of religion––necessary for robust democracy.124

This research makes valuable contributions to our understandings of both rituals and politics. Chwe, Patel, and Stepan present convincing arguments that the public form of rituals affect political outcomes. But they do not address why people participate in the rituals they study.

121 Chwe, Rational Ritual. For a strong critique of Chwe and the idea of culture as common knowledge, see Lisa Wedeen, “Conceptualizing Culture: Possibilities for Political Science,” American Political Science Review, Vol. 96, No. 4. (December 2002), esp. pp. 718-19.

122 David Siddhartha Patel, “Islam, Information, and Social Order: The Strategic Role of Religion in Muslim Societies,” Ph.D. dissertation, Stanford University, 2007.

123 Alfred Stepan, “Rituals of Respect: Sufis and Secularists in Senegal in Comparative Perspective,” Comparative Politics (July 2012), pp. 379-401. For an earlier analysis of the role of rituals in Senegalese politics, see Leonardo A. Villalón, “Sufi Rituals as Rallies: Religious Ceremonies in the Politics of Senegalese State-Society Relations,” Comparative Politics, Vol. 26, No. 4 (July 1994), pp. 415-37.

124 Stepan, “Religion, Democracy, and the ‘Twin Tolerations’”; and Stepan, Arguing Comparative Politics, pp. 213-53.

112 This is problematic because they demonstrate that rituals provide a communally valued function, i.e., a public good. As I demonstrated earlier, rituals that produce a public good are collective actions open to free-riding. Specifying the collective benefits of rituals is insufficient to explain individual choices. Thus the rationalist framework these scholars use to explain macro-outcomes lacks a plausible micro-foundation.

A second approach which fits seamlessly into political science thinking on rituals (though thus far not adopted by political scientists) is the costly-signaling theory of rituals. This approach directly addresses participation in rituals, particularly risky, costly, or time-consuming rituals.

These include rituals involving self-inflicted pain (e.g., self-flagellation on Ashura) or willful destruction of personal property (e.g, the potlatch practiced by communities indigenous to the

Pacific Northwest). The argument is that partaking in a dangerous or painful ritual is a costly signal of trustworthiness and commitment to the group which screens out free-riders. There are two variants of the mechanism. Economists argue that this increases the payoffs provided by the

“club,” while evolutionary anthropologists argue that these mechanisms were necessary for the survival of early humans.125 Both variations, however, rely on a functionalist logic that fails to explain why any particular individual would participate. The economic argument places us back in the realm of collective goods and free riders, while the evolutionary argument leaves no room

125 For the economic argument, see, e.g., Laurence R. Iannacone, “Sacrifice and Stigma: Reducing Free-riding in Cults, Communes, and Other Collectives,” Journal of Political Economy, Vol. 100, No. 2 (April 1992), pp. 271-91; Laurence R. Iannacone, “Why Strict Churches Are Strong,” American Journal of Sociology, Vol. 99, No. 5 (March 1994), pp. 1180-1211; and Eli Berman, “Sect, Subsidy, and Sacrifice: An Economist’s View of Ultra-Orthodox Jews,” Quarterly Journal of Economics, Vol. 115, No. 3 (August 2000), pp. 905-53. For the anthropological approach, see e.g., William Irons, “Religion as a Hard-to-Fake Sign of Commitment,” in Randolph Nesse, ed., The Evolution of Commitment (New York: Russell Sage Foundation, 2001), pp. 292-309; Richard Sosis and Candace Alcorta, “Signaling, Solidarity, and the Sacred: The Evolution of Religious Behavior,” Evolutionary Anthropology, Vol. 12, No. 6 (2003), pp. 264-274; and Richard Sosis and Bradley J. Ruffle, “Religious Ritual and Cooperation: Testing for a Relationship on Israeli Religious and Secular Kibbutzim,” Current Anthropology, Vol. 44, No. 5 (December 2003), pp. 713-22.

113 to explain differences in behavior across individuals. !

Theories of Ethnic Conflict

In this section, I review three approaches to explaining ethnic conflict. The elite-led conflict approach, discussed in the previous chapter, does not address mass participation, so I only describe it briefly. The ethnic rivalry approach views contentious rituals as opportunities for groups and individuals within the group to assert in-group pride and out-group animosity. Thus we expect that participation is motivated by attitudes towards the in-group and out-group. The collective action approach sees contentious rituals as subject to the free-rider problem. Thus we expect that participation is motivated by variables which increase the private benefits or decrease the private costs. !

ELITE-LED CONFLICT APPROACH

The political science research which addresses contentious rituals most directly focuses on the role of strategic elites. As detailed in the previous chapter, the conclusion of this research is that inflammatory acts, such as contentious rituals, are used strategically to polarize a society by provoking the out-group into overreacting in order to promote distrust between communities, create a negative image of the out-group in local or international courts of opinion, or discredit in-group moderates.126 This approach provides an explanation for the incentives of elites to design and promote contentious rituals, but it fails to account for why the masses would participate in them. For instance, Wilkinson writes that Hindu party leaders in India “organize

126 Wilkinson, Votes and Violence; and Brubaker and Laitin, “Ethnic and Nationalist Violence,” p. 433.

114 unusually large religious processions” in order to provoke minorities.127 But who makes up these

“unusually large” crowds? And why did they turn out? Neither Wilkinson nor the other scholars from this approach provide answers. Since we cannot explain mass participation with elite interests, readers are left guessing why people participate in the (often risky) elite-promoted contentious rituals. As a result, this argument does not suggest hypotheses about individual-level participation. And, as I previously demonstrated, Protestant elites in Northern Ireland are divided over loyalist parades. Some support parades, but others oppose them. So even without its silence on the matter of participation, the elite-led approach does not explain the Northern Irish case. !

ETHNIC RIVALRY APPROACH

The ethnic rivalry explanation focuses on the role of ethnic difference and rivalry in motivating conflict. This approach comes in primordial and constructivist flavors, but both camps argue that conflict and contentious rituals stem from the existence of ethnic difference.128

Through this lens, contentious rituals are seen as symbolic assertions of group dominance129 or status130 as well as the out-group’s subordination. As mass public events, contentious rituals

127 Wilkinson, Votes and Violence, p. 24. Emphasis is mine.

128 On the social psychology of ethnic difference, see Henri Tajfel and John C. Turner, “An Integrative Theory of Intergroup Conflict,” in W. G. Austin and S. Worchel, eds., The Social Psychology of Intergroup Relations (Monterey, CA: Brooks/Cole, 1979); and Henri Tajfel and John C. Turner,“The Social Identity Theory of Intergroup Behaviour,” in S. Worchel & W. G. Austin, eds., Psychology of Intergroup Relations (Chicago: Nelson-Hall, 1986).

129 Donald L. Horowitz, Ethnic Groups in Conflict (Berkeley: University of California Press, 1985); and Horowitz, The Deadly Ethnic Riot.

130 Roger D. Petersen, Understanding Ethnic Violence: Fear, Hatred, and Resentment in Twentieth-Century Eastern Europe (New York: Cambridge University Press, 2002); and Roger D. Petersen, Western Intervention in the Balkans: The Strategic Use of Emotions in Conflict (New York: Cambridge University Press, 2011).

115 allow individuals to articulate their ethnic pride or ethnic grievances.131 Although grievances have recently been dismissed by much of the empirical literature on conflict,132 there is a long tradition of grievance-based explanations for ethnic conflict.133 Additionally, many studies that find that grievances have no effect are looking at the emergence of conflict and contentious political movements. Studies that look at individual participation, our interest here, are generally much more favorable toward grievances.134 The logic of these arguments suggests that individual-level participation is best explained by the attachments that people hold to the group.

Participation is a way for people to further the interests of their ethnic group and thus it is expected that the more people identify with their ethnic group, the more likely they are to participate.135

However, these events do not just celebrate the ethnic in-group and help it achieve its goals: they simultaneously denigrate and offend the ethnic out-group. Contentious rituals are a

131 Roger Petersen, Western Intervention in the Balkans, p. 100, specifically predicts that one particular grievance, resentment of ethnic status reversal, leads to “parades and demonstrations that highlight grievances about political status.”

132 Paul Collier and Anke Hoeffler, “Greed and Grievance in Civil War,” Oxford Economic Papers, Vol. 56 (2004), pp. 563-595; James D. Fearon and David D. Laitin, “Ethnicity, Insurgency, and Civil War,” American Political Science Review, Vol. 97, No. 1 (February 2003), pp. 75-90.

133 Most prominently, Ted Robert Gurr, Why Men Rebel (Princeton: Princeton University Press, 1970). See also the recent “meaning-laden approach to grievances” in Erica Simmons, “Grievances Do Matter in Mobilization,” Theory and Society, Vol. 43, No. 5 (September 2014), pp. 513-546.

134 Maurice Pinard, Motivational Dimensions in Social Movement and Contentious Collective Action (Montreal and Kingston: McGill-Queen’s University Press, 2011), esp. pp. 40-44.

135 See Ashutosh Varshney, “Nationalism, Ethnic Conflict, and Rationality,” Perspectives on Politics, Vol. 1, No. 1 (March 2003), p. 93; and Amilcar Antonio Barretto, “Nationalism, Collective Action, and Rationality,” Nationalism and Ethnic Politics, Vol. 18, No. 3 (2012), pp. 327-28. This hypothesis is also supported by studies of social movement mobilization such as Bert Klandermans, Jojanneke van der Toorn, and Jacquelien van Stekelenburg, “Embeddedness and Identity: How Immigrants Turn Grievances into Action,” American Sociological Review, Vol. 73 (December 2008), pp. 992-1012; Bernd Simon, Michael Loewy, Stefan Stürmer, Ulrike Weber, Peter Freytag, Corinna Habig, Claudia Kampmeier, and Peter Spahlinger, “Collective Identification and Social Movement Participation,” Journal of Personality and Social Psychology, Vol. 74, No. 3 (March 1998), pp. 646-58.

116 way to express grievances against or hatred of the out-group. Therefore, it is expected that people are more likely to participate the more they dislike the out-group. Overall, the ethnic rivalry approach suggests that participants are distinguished from nonparticipants by their attitudes towards the in-group and out-group. !

COLLECTIVE ACTION APPROACH

This approach to ethnic conflict builds on Olson’s logic of collective action.136 Given that the private costs of ethnic conflict outweigh the private benefits, thereby creating incentives to free-ride on the actions of others, collective action theorists ask: Why would anyone voluntarily participate? One answer is that participants are provided selective material incentives.137 Thus, it is expected that people are more likely to participate if they receive selective material benefits.

Some theorists of collective action have expanded their understanding of selective rewards to include non-material benefits such as “fun”138 and “reputation,”139 but, as critics have shown, these attempts to rescue the rationalist approach fail to provide a coherent explanation of participation.140 So to have a clear test between the collective action theory and the ritual

136 Olson, Logic of Collective Action.

137 Ibid., p. 51; Samuel L. Popkin, The Rational Peasant: The Political Economy of Rural Society in Vietnam (Berkeley: University of California Press, 1979); Mark I. Lichbach, “What Makes Rational Peasants Revolutionary?: Dilemma, Paradox, and Irony in Peasant Collective Action,” World Politics, Vol. 46, No. 3 (April 1994), pp. 383-418; Mark I. Lichbach, The Rebel’s Dilemma (Ann Arbor: University of Michigan Press, 1995).

138 Edward N. Muller and Karl-Dieter Opp, “Rational Choice and Rebellious Collective Action,” American Political Science Review, Vol. 80, No. 2 (June 1986), pp. 471-87.

139 Dennis Chong, Collective Action and the Civil Rights Movement (Chicago: University of Chicago Press, 1991), esp. chap. 3.

140 Green and Shapiro, Pathologies of Rational Choice; Jasper, Art of Moral Protest, pp. 23-29; Petersen, Understanding Ethnic Violence, pp. 32-33; and Wood, Insurgent Collective Action, pp. 253-54.

117 argument, we must clearly limit the selective benefits of collective action to material benefits.

A second answer to the collective action problem is that rational individuals will participate to avoid social sanctions targeted at nonparticipants.141 Consequently, it is hypothesized that people are more likely to participate if they expect social sanctions for not taking part. Overall, the rationalist collective action approach suggests that participants are distinguished from nonparticipants by their private rewards or punishments.

Finally, research on collective action finds that pre-existing social ties to other participants are an important predictor of participation.142 Social ties can increase the likelihood of mobilization by providing information, nurturing an activist identity and solidarity, or giving social approval and encouragement. Therefore, it is expected that people are more likely to participate if they have social ties to participants. !

Conclusion

This chapter introduced the concept of contentious ritual and proposed a theory of individual participation in them. The theory builds off of two important conclusions from the

141 Michael Taylor, “Rationality and Revolutionary Collective Action,” in Michael Taylor, ed., Rationality and Revolution (New York: Cambridge University Press, 1988), pp. 63-97. See Pamela Oliver, “Rewards and Punishments as Selective Incentives for Collective Action: Theoretical Investigations,” American Journal of Sociology, Vol. 85, No. 6 (May 1980), esp. pp. 1368-71.

142 For reviews of this large literature, see James A. Kitts, “Mobilizing in Black Boxes: Social Networks and Participation in Social Movement Organizations,” Mobilization, Vol. 5, No. 2 (October 2000), pp. 241-57; John Krinsky and Nick Crossley, “Social Movements and Social Networks: Introduction,” Social Movement Studies, Vol. 13, No. 1 (2014), pp. 1-21. Some selected research includes, Doug McAdam, “Recruitment to High-Risk Activism: The Case of Freedom Summer,” American Journal of Sociology, Vol. 92, No. 1 (July 1986), pp. 64-90; Doug McAdam, Freedom Summer (New York: Oxford University Press, 1988); Sharon Erickson Nepstad and Christian Smith, “Rethinking Recruitment to High-Risk/High-Cost Activism: The Case of the Nicaragua Exchange,” Mobilization, Vol. 4, No. 1 (April 1999), pp. 25-40; Alan Schussman and Sarah A. Soule, “Process and Protest: Accounting for Individual Protest Participation,” Social Forces Vol. 84, No. 2 (December 2005), pp. 1083-1108; Alexandra Scacco, “Who Riots? Explaining Individual Participation in Ethnic Violence,” Ph.D. dissertation, Columbia University, 2010; the articles in the recent special issue of Social Movement Studies (“Social Networks and Social Movements,” Vol. 13, No. 1 [2014]) introduced by the Krinsky and Crossley essay cited above.

118 multidisciplinary study of ritual: that rituals provide benefits for participants and that their meaning is ambiguous. This suggests that participation in contentious rituals is fueled by process-oriented motives. In addition to this theory of intrinsic motives, I proposed two plausible alternative explanations for ritual participations premised on different motives. One alternative— ethnic rivalry—is that people are motivated by ideational or ideological concerns about in-group/ out-group relations. The second alternative—collective action—suggests that people are motivated by the external outcomes of their behavior.

In the following three chapters, I present evidence from participants and nonparticipants in loyalist parades in Northern Ireland to assess each argument. Chapters 5 and 6 present evidence to support the theory of ritual participation. Using primarily semi-structured interviews,

I show that participants understand their action non-instrumentally and are most interested in the pleasures and meanings inherent in participation itself. I then demonstrate that participants do not view their actions as political and explore the anti-political rhetoric of participation. In Chapter 7,

I provide evidence from the interviews and a survey of 228 randomly-selected Protestants in

Belfast that the alternative explanations do not help explain who participates or why.

119 Chapter Five Parading Mainly for Fun and Process! ! Outside of [parading culture] you look out and you think, this doesn’t make sense… What possible, rational, sensible reason could there be for these people to be in this? Well, the answer lies under rationality. It lies in something else. ! — Rich, Presbyterian minister and former Orangeman !

The last chapter proposed an explanation for why people participate in contentious rituals. This chapter is the first of three that takes an empirical look at the question. In it, I present interview, survey, and ethnographic data that support my argument of ritual participation.

After a brief discussion of the data and my analytic approach, I show that, as expected by the argument of the last chapter, participants generally view parades non-instrumentally. I demonstrate that their primary reasons for action and understanding of their actions and are not outcome-oriented, but focused on the process of participation itself. Not all reasons to parade, however, are process-oriented, so in the second part I explore instrumental reasons, such as sending a message to Catholics. But I show that the appeal of these instrumental reasons are secondary. !

Data and Methods

The primary data I use in this chapter come from semi-structured interviews I conducted in Northern Ireland during eight months of fieldwork in 2012 (July-August and November-

December), 2013 (April-August), and 2014 (June). Interview subjects were selected both purposely and using snowball sampling. In total, I interviewed 82 individuals: 49 current or former parade participants, and 33 nonparticipants. This chapter focuses on the participant

120 interviews. The majority of interviewees lived in greater Belfast, though I also sought the views of some people from the rest of the province. Reflecting the gender of paraders, the majority of my interviewees were male. Nearly all interviews were digitally recorded and transcribed in full; all were analyzed using NVivo, a qualitative data analysis software.

Semi-structured interviews provide two specific advantages for my research.1 First, the variation in many of the variables I am interested in, such as the meanings and perceived purposes of parades, is too diverse and/or subtle to measure accurately and reliably with close- ended survey questions. Using what Michèle Lamont and Ann Swidler call an “open-ended and pragmatic approach to interviewing,” I am able “to collect data not only, or primarily, about behavior, but also about representations, classification systems, boundary work, identity, imagined realities and cultural ideals, as well as emotional states.”2 The open-ended questions of a semi-structured interview allow me to capture importance nuance in these variables while remaining focused and ensuring that I ask all the necessary questions. Further, open-ended questions provide the opportunity to capture the details of the process of mobilization.

Second, the broad questions I ask allow respondents to help direct the course of the interview toward the topics they find important and meaningful. What they say and the way they say it can reveal a lot about what is on someone’s mind. Thus, according to Blee and Taylor,

1 On the benefits—and limitations—of semi-structured interviews, see Beth L. Leech, ed. “Symposium on Interview Methods in Political Science,” PS: Political Science & Politics, Vol. 23, No. 3 (December 2002), pp. 663-88; Kathleen M. Blee and Verta Taylor, “Semi-Structured Interviewing in Social Movement Research,” in Bert Klandermans and Suzanne Staggenborg, eds., Methods of Social Movement Research (Minneapolis: University of Minnesota Press, 2002), pp. 92-117; H. Russell Bernard, Research Methods in Anthropology, 5th ed. (Lanham, MD: AltaMira, 2011); Layna Mosley, ed., Interview Research in Political Science (Ithaca: Cornell University Press, 2013); Michèle Lamont and Ann Swidler, “Methodological Pluralism and the Possibilities and Limits of Interviewing,” Qualitative Sociology, Vol. 37, No. 2 (June 2014), pp. 153-171.

2 Lamont and Swidler, “Methodological Pluralism,” p. 157.

121 allowing participants and nonparticipants to speak for themselves helps us “understand[] social movement mobilization from the perspective of movement actors or audiences.”3 They continue by arguing that semi-structured interviews “provide greater breadth and depth of information, the opportunity to discover the respondent’s experience and interpretation or reality, and access to people’s ideas, thoughts, and memories in their own words rather than in the words of the research.” The downside that they note is the “reduced ability to make systematic comparisons between interview responses,” but analyzing my interviews alongside my other two data sources helps ameliorate the problem.4

The second form of data I use in this chapter comes from a randomized household survey

I conducted in Belfast in 2013. I only use this quantitative source sparingly throughout the chapter and only in the form of univariate statistics. As a result, I will leave a full discussion of the survey and its sample for Chapter 7, when I rely more heavily on quantitative analyses. For now it is sufficient to say that 228 randomly-selected people were surveyed in nine Protestant neighborhoods about their parading behavior and opinions as well as demographic characteristics. Of the respondents, 28 are current parade participants (12 percent) and 70 participated in some point in their adult life (31 percent).

The third form of data are my fieldnotes from the observation of many parades, protests, public meetings, marching band practices, and other related events. The fieldnotes were written as soon after the event as possible from memory with the aid of written jottings, voice

3 Blee and Taylor, “Semi-Structured Interviewing in Social Movement Research,” p. 92.

4 Ibid., pp. 92-3.

122 recordings, and photographs made during the event.5 Observing the events is an appropriate method because my theory claims the primacy of process benefits, some of which are best measured as they happen in order to avoid the distorting lens of memory. Through informed and guided observation, I looked for specific variables that might impact these intrinsic benefits.

Features which I expected to increase the available process benefits include the crowd, social interactions, symbols, and ceremonies. Therefore, when I attended parades I took particular note of the size and enthusiasm of the crowd; how paraders interact with supporters, protesters, and each other; how physical symbols, such as flags and banners, are used; and behavior during religious and memorial services.

I analyzed the three data sources together and I am more confident in my conclusions because, in general, all the data point in the same direction. When they do not (for example, in the discussion of the impact of social pressure on participation in Chapter 7), I present the results of all the relevant data and carefully explain my interpretation.

In this chapter, in particular, I approach the interview data with care. While collecting the data and analyzing it, I was fully aware that loyalist parades are a heavily politicized topic in

Northern Ireland and that participants and supporters have good reason to present a particular side. I thus conducted the interviews and analyzed the transcripts with skepticism, so that statements that are outside the dominant Protestant narrative are given special value. That said, in my analysis, I take seriously what people said about how they view the world, their own actions, and their own motivations. There are problems with this stance for reasons such as imperfect

5 See Robert M. Emerson, Rachel I. Fretz, and Linda L. Shaw, Writing Ethnographic Fieldnotes (Chicago: University of Chicago Press, 1995).

123 memory and social desirability bias,6 but I believe that when analyzed carefully, these data can provide useful information.

Three dynamics can further increase our confidence in the data. One, in the interviews, I asked about specific, recent events. For example, when asking interviewees to describe parades and their experience in marching, I always asked them to describe the most recent parade they walked it. For some, it was as recent as the previous day, but it was never more than several months. But even when asking about events from years ago, such as their childhood and their initial mobilization into parading, I, two, focused on specific events, choices, and relationships in that period.7 Three, Schlozman, Verba, and Brady note that “there is general agreement that answers about reasons gain significance to the extent that the matters at stake are important ones.”8 Parades are generally extremely important to participants and the controversy surrounding parades only increases the importance to them. !

Process-Oriented Reasons for Participating

How do parade participants understand their own actions? Do they think of what they do as a means toward an end or as an end in and of itself? In this part, I present evidence that participants primarily understanding parading as non-instrumental action. I show that in interviews and the survey, participants do not focus on the outcomes of parades—which would

6 On the problems with self-reported motivations, see Richard E. Nisbett and Timothy DeCamp Wilson, “Telling More Than We Can Know: Verbal Reports on Mental Processes,” Psychological Review, Vol. 84, No. 3 (May 1977), pp. 231-59.

7 Ziad W. Munson, The Making of Pro-Life Activists: How Social Movement Mobilization Works (Chicago: University of Chicago Press, 2009), p. 198.

8 Kay Lehman Schlozman, Sidney Verba, and Henry E. Brady, “Participation's Not a Paradox: The View from American Activists,” British Journal of Political Science, Vol. 25, No. 1 (January 1995), p. 10.

124 be evidence of instrumental thinking—but on the experience of participation, which suggests the appeal of process-oriented motives, the core of the ritual approach presented in Chapter 4.

In particular, I find that participants have are six primary “reasons for acting,” the

“values, norms, commitments, emotions, material interests, and aversions” that motivate action.9

These reasons are expressing their collective identity, commemoration, tradition, defiance, the pleasures of participation, and external communication. The first five, which are more prevalent in the data, have the qualities of process-regarding benefits. That is, their benefits are rooted in the process of participation, not in successfully achieving the goal of the action. The final reason, sending a message, is more outcome-oriented. The benefits derive from successfully sending a message to the intended recipient; simply trying is insufficient.

By dividing participants’ reasons for participating into process-regarding and outcome- regarding, I am not reviving old distinctions between expressive movements and instrumental movements. Many studies have shown that likely all movements use both instrumental and expressive action and are oriented both internally and externally.10 Rather, I am describing the reasons that bring people to participate.11 In fact, I will argue people can have expressive reasons for participating in collective action with external outcomes.

9 Elisabeth Jean Wood, Insurgent Collective Action and Civil War in El Salvador (New York: Cambridge University Press, 2003), p. 231.

10 For example, James M. Jasper, The Art of Moral Protest: Culture, Biography, and Creativity in Social Movements (Chicago: University of Chicago Press, 1997); Mary Bernstein, “Celebration and Suppression: The Strategic Uses of Identity by the Lesbian and Gay Movement,” American Journal of Sociology, Vol. 103, No. 3 (November 1997), pp. 531-565; and Verta Taylor and Nella Van Dyke, “‘Get up, Stand up’: Tactical Repertoires of Social Movements,” in David A. Snow, Sarah A. Soule, and Hanspeter Kriesi, eds., The Blackwell Companion to Social Movements (Malden, Mass.: Blackwell, 2004), pp. 262-293.

11 See Bert Klandermans, “Motivation and Types of Motives (Instrumental, Identity, Ideological Motives),” in David A. Snow, Donatella della Porta, Bert Klandermans, and Doug McAdam, eds., The Wiley-Blackwell Encyclopedia of Social and Political Movements (Malden, Mass.: Wiley-Blackwell, 2013).

125 The expressive or intrinsic character of participants’ reason is well illustrated by Mark, an

Orangeman from Carrickfergus, just outside of Belfast. At the very beginning of the interview, he told me that he first paraded at the age of four, when his uncle, “a very strong member of the

Orange Order,” brought him to the Twelfth of the July to hold the strings that hung from the lodge’s banner. “I just knew this is where I was meant to be,” he recalled. “It was almost as if it was my whole reason for being, to a certain degree—and that may sound very strange from a young child.” He has “been a member of the Orange Order ever since.” As Mark grew older and learned more about what the Orange Order stands for, he saw that it is “exactly what I’m about.

That is who I am.” Speaking rapidly, he continued: “It’s like asking a Jewish person, why do you go to the synagogue? Asking a Catholic person, why do they go to Mass? Why am I in the

Orange? Because that is exactly what I am and it’s— I don’t do it to offend Catholics, to offend anybody. I am in the Orange simply because that’s my identity and that’s my belief.” For Mark, participation is related to his sense of self and part of how he understands his life and his world.

It is a reflection of his deepest held principles and priorities.

Participation, for Mark, is not a means toward an external social or political end. Later on in the interview, he explained to me his view of what it is that parades celebrate: “It’s about that we are a distinct cultural group,” he said. “And we’re still here and we still adhere to old- fashioned British values of honesty and decency.”

“So is that the primary message then?” I asked.

“Eh, the message? I don’t know.” He hesitated before proceeding haltingly: “When I go out parading do I go out to send a message? Hmm. Maybe it could be viewed upon that I am sending a message, but probably it’s more of a statement of what I am.” His normal confidence

126 was tripped up by my question about parading’s message. Despite just stating a plausible message, he was unsure how to answer because he does not view his actions in these instrumental terms. Parades are about his “identity” and are an “outward expression” of his beliefs, as he put it a moments later. As a result, he generally does not think about them in terms of their primary message and he did not have a ready answer for the question. His parsing of

“message” further reveals his non-instrumental reasoning. Although “message” and “statement” are near synonyms, they have different connotations. Sending a message is an attempt to influence an intended recipient, whereas making a statement does not presume a recipient and thus does not attempt to influence them. Sending a message, therefore, is more instrumental while making a statement is more expressive. Mark places parading firmly with the latter.

Intrinsic reasons thus dominate Mark’s thinking about parades, though instrumental reasons appear as well. For instance, he acknowledged that parades have a political edge because they are performed by people who support the union of Northern Ireland and Great Britain. “It is a political statement in that regard,” he stated. “Although, we don’t set out— We simply set out to celebrate, ‘this is what we are and this is what we stand for.’” The statement broadcast from parades is not what he sets out to do; it is an unintentional byproduct. Instrumental reasons are present but they are largely incidental, not a motivating factor.

Mark’s privileging of intrinsic reasons and his ambivalence toward intrinsic reasons is a pattern found throughout the data. As a first cut, I use a question from the survey—“In your opinion, what is the purpose of parades?”—to test whether participants are more likely to

127 Table 5.1. Types of Purposes Attributed to Loyalist Parades Intrinsic Purposes Tradition It is a tradition to parade; to continue that tradition. Culture It is part of our culture to parade; or, parading maintains our culture. Celebration To celebrate Protestant culture and people. Commemoration To commemorate, celebrate, or mark the Protestant past. Social To bring people together to enjoy each others’ company. Fun/Carnival To create a fun environment for people to enjoy; also to compete musically. Instrumental Purposes To Promote To promote a particular agenda, such as Protestantism or Protestant unity. Display Loyalty To display loyalty to the Protestant group, Northern Ireland, or the UK. Take a Stand To show others what one believes in (culture, politics, etc.). Negative To cause trouble, be provocative or send a message of triumphalism. attribute intrinsic, rather than instrumental, purposes to parades.12 Each response was coded with at least one of ten purposes, with some responses receiving multiple codes, using the definitions in Table 5.1. The codes were developed inductively to accurately capture the concepts conveyed by respondents. I then grouped the purposes into two general categories based on their orientation: Intrinsic and Instrumental. Intrinsic purposes (culture, tradition, celebration, commemoration, social, and carnival) are achieved simply by doing the act. For example, successfully continuing a tradition is accomplished by doing the traditional act. Instrumental purposes (taking a stand, displaying loyalty, promoting, and causing a negative outcome) are only achieved if they accomplish something external to the act itself. For example, successfully promoting Protestant culture requires a response from someone else.

Simple two-tailed t-tests show that participants are significantly more likely to mention

12 For a similar empirical approach, see Abby Peterson, Mattias Wahlström, Magnus Wennerhag, Camilo Christancho, and José-Manuel Sabucedo, “May Day Demonstrations in Five European Countries,” Mobilization, Vol. 17, No. 3 (September 2012), pp. 281-300.

128 intrinsic purposes than instrumental purposes. The first column in Table 5.2 displays the results for current participants (92 percent intrinsic vs. 13 percent instrumental; p = 0.00) and the second column shows the results for respondents who have ever paraded since age 16 (88 percent intrinsic vs. 14 percent instrumental; p = 0.00). Participants were thus much more likely to state that the purpose of parading is to “celebrate Protestant culture and tradition” or “to commemorate the victory of the Battle of the Boyne” than “to make a statement” or “marking territory.” Interestingly, current paraders are also significantly more likely than nonparticipants to cite intrinsic purposes (92 percent vs. 73 percent; p = 0.04). ! Table 5.2. Purposes of Loyalist Parades Reported by Participants: Intrinsic vs. Instrumental (%) Current Participants Ever Participated Intrinsic Purpose 92% 88% Culture 46% 47% Tradition 38% 34% Celebration 29% 25% Commemoration 17% 16% Social 4% 8% Fun/Carnival 4% 3% Instrumental Purpose 13% 14% Take a Stand 8% 6% Display Loyalty 4% 2% To Promote 0% 3% Negative 0% 2% Intrinsic-Instrumental Difference 79% 73% P-Value (Two-Tailed) 0.00*** 0.00*** Observations 24 64 Each response could take multiple codes, so columns do not sum to 100%. *p<0.05, ** p<0.01, *** p<0.001

129 The semi-structured interviews confirm that participants are far more interested in the process than the outcome of participation. In analyzing participants’ stated purposes of their actions, the same two categories emerged: intrinsic and instrumental. But intrinsic, process- oriented purposes were cited more often and with the more vigor. Participants’ understanding of parades focus on expressing a collective identity, commemoration, tradition, defiance, and the social and emotional pleasures of participation. The following sections explore each in turn. !

Expressing Collective Identity

For most, if not all, paraders, the central purpose of a parade is the expression and celebration of a multifaceted Protestant identity. In interview after interview, participants described parades as opportunities to articulate “what I am,” to “show your identity,” and to

“express Protestant culture.”13 Among participants, expressing collective identity is a significant reason for action. By collective identity, I mean: “a shared sense of ‘one-ness’ or ‘we-ness’ anchored in real or imagined shared attributes and experiences among those who comprise the collectivity and in relation or contrast to one or more actual or imagined sets of ‘others.’”14

Scholars have found that collective identity is an important cause and consequence of

13 Interview with Mark, July 11, 2013; interview with Albert, August 20, 2012; and interview with Rachel, August 8, 2013.

14 David Snow, “Collective Identity and Expressive Forms,” Center for the Study of Democracy Working Paper, University of California, Irvine, 2001, p. 2, available at: https://escholarship.org/uc/item/2zn1t7bj. For discussions of the use of the concept in the social movement literature, see Francesca Polletta and James M. Jasper, “Collective Identity and Social Movements,” Annual Review of Sociology, Vol. 27 (2001), pp. 283-305; Scott A. Hunt and Robert D. Benford, “Collective Identity, Solidarity, and Commitment,” in Snow, Soule, and Kriesi, eds., The Blackwell Companion to Social Movements, pp. 433-457; and Cristina Flesher Fominaya, “Collective Identity in Social Movements: Central Concepts and Debates,” Sociology Compass, Vol. 4, No. 6 (2010), pp. 393-404.

130 mobilization.15

In the case of parading, the relevant identities are Protestant, unionist, and/or loyalist and to a lesser extent an organizational identity of the parading group that one belongs to. George, for instance, states: “The overarching purpose is to say to the world, ‘Here we are.… Here we are as members of the Protestant, reformed, evangelical faith. This is our cause and we want the world to know.’ I think that’s the purpose.” George understands his actions as a declaration of his religious identity. Parades are a public witness to his membership in a religiously defined group

(and its contrast with Roman Catholicism, evinced in the word “reformed”). As he says later in the interview, Protestantism as a religion is less about specific teachings or values, but about “the pride and the understanding of being a member of it.” I pushed him to discuss what religious teachings were important to him, but by his own admission, he was “stumbling” and “hedging around the question,” giving little more that the vague statement, “I believe in the teachings of the Protestant church.” Far clearer was that he is not Catholic: “I would be vehemently against the teachings of the Roman faith, where they are claiming to bring down God on Mass and his body and blood are being given in bread and wine. If that’s not an abomination, I don’t know what is.” Parades are how George articulates his collective identity, allowing him to declare simultaneous that he is Protestant and he is not Catholic.

Walter also expresses multiple aspects of identity in parades. For him, parades are “about going out and showing our cultural identity—not through violence, but through music and

15 For example, Verta Taylor and Nancy E. Whittier, “Collective Identity in Social Movement Communities: Lesbian Feminist Mobilization,” in Aldon D. Morris and Carol M. Mueller, eds., Frontiers in Social Movement Theory (New Haven: Yale University Press, 1992), pp. 104-129; Bernstein, “Celebration and Suppression”; and Bernd Simon, Michael Loewy, Stefan Stürmer, Ulrike Weber, Peter Freytag, Corinna Habig, Claudia Kampmeier, and Peter Spahlinger, “Collective Identification and Social Movement Participation,” Journal of Personality and Social Psychology, Vol. 74, No. 3 (March 1998), pp. 646-58.

131 through the pageantry of it all. And to show how well and respectful we can be.” Parading is a way to express his ethnic Protestant collective identity and demonstrate its worthy qualities. In addition to the broader Protestant identity, Walter has a specific local identity which he also relishes showing off. “[Our neighborhood] is known throughout the land because of [our band],” he boasts. “It takes [the neighborhood’s] identity everywhere. It’s [written] on our bass drum.”

Most band members are from the same small, tight-knit interface neighborhood, so the identities of the band and the community are intertwined. Walter even displays his attachment to his band on his body, with a tattoo on his leg. Half the band, he estimates, has done the same: “We have our band tattooed on us,” he says. Local and organizational identities, such as Walter and his band, reinforce the broader collective identity as they tie individuals to a knowable segment of the larger identity. Particularly in the case of ethnic movements, the ties to an organization generate a real tie to the imagined community.

Though expressing a collective identity can be used instrumentally to effect an external outcome,16 parade participants, for the most part, do not act with this intention. For example, when I asked Albert what the goal of parading is, he replied: “Just to show your identity. … We are a parading organization and we parade as and when necessary.” When I asked why, he said,

“That’s the thing to do. You show it to your supporters and if your non-supporters object, it’s up to them.” Responding to the same question about the purpose or goal of parading, Robert said,

“To show that we are members of the Protestant community.” Neither Albert nor Robert evince any indication of a goal outside of the parade itself. Their reason for acting is intrinsic of parading; it is not dependent on achieving a particular result.

16 For example, Bernstein, “Celebration and Suppression.”

132 Implicit in their comments is an idea that Ben makes explicit: “Identity isn’t a private thing, it’s a public thing.”17 He emphasizes the Orange Order’s relationship to Christian evangelism and then continues: “Orangeism isn’t something that we are just going to have in our wee halls and have our wee meetings and then go home again. Public expression of it is intrinsic to it, it’s a value of it. It has to be public. It has to be a public manifestation of it.” Parading has been part of Irish Protestant identity for so long that he cannot imagine the identity without its public displays.

But participants do not just express their collective identity because they believe the identity demands it. Expressing a deeply-held identity publicly is a source of process-oriented benefits that participants seek. As Friedman and McAdam argue, “One of the most powerful motivators of individual action is the desire to confirm through behavior a cherished identity.”18

Parading is a way for participants to live their Protestant identity and beliefs. It is a way to publicly enact deep connections to the broader community. Many of the process-benefits of expressing collective identity in a parade are emotional, such as the strong feeling of pride. I will return to these emotional rewards below when I discuss the pleasures of participation. !

Commemoration

A second major purpose for parades cited by participants is commemoration of the

17 Compare to Sherry B. Ortner’s (“Theory in Anthropology since the Sixties,” Comparative Studies in Society and History, Vol. 26, No. 1 [January 1984], p. 129) summary of Geertz: “culture is not something locked inside people’s heads, but rather is embodied in public symbols.” See also Ann Swidler, “Culture in Action: Symbols and Strategies,” American Sociological Review, Vol. 51, No. 2 (April 1986), pp. 273-286; and Ann Swidler, “Cultural Power and Social Movements,” in Hank Johnston and Bert Klandermans, eds., Social Movements and Culture (Minneapolis: University of Minnesota Press, 1995), esp. pp. 26-27.

18 Debra Friedman and Doug McAdam, “Collective Identity and Activism: Networks, Choices, and the Life of a Social Movement,” in Morris and Mueller, eds., Frontiers in Social Movement Theory, p. 169.

133 Protestant past, particularly historic battles. Through the parades, participants honor the great moments in Protestant history and the individuals who gave their lives for the cause. Certain parades commemorate specific events, the most important being the Twelfth of July, which commemorates the Battle of the Boyne (1690); but others include the First of July, which commemorates the Battle of the Somme (1916); and the two main Derry parades, which commemorate the Siege of Derry (1689). But through repetition, which “automatically implies continuity with the past,” even non-commemorative parades call the past to mind.19 For this reason, anthropologist Neil Jarman describes Northern Irish parading as “the performance of memory.”20

Commemoration follows a process-oriented logic: the goal of a commemorative act—that the past be appropriately marked—is inseparable from the means of achieving it. Participants see no extrinsic purpose in commemoration; they are driven to commemorate by a drive to commemorate. They do not see their commemorations as a means for an external end.21 For example, Albert told me that, “The First of July [parade], it commemorates the Battle of the

Somme [and] the 36th (Ulster) Division [in World War I], the Twelfth of July [parade] commemorates the Battle of the Boyne. They’re all battles we remember.”

“You do that to continue the memory?” I asked, probing his motives.

“Yes, to continue the tradition,” he replied.

19 Paul Connerton, How Societies Remember (New York: Cambridge University Press, 1989), p. 45.

20 Neil Jarman, Material Conflicts: Parades and Visual Displays in Northern Ireland (Oxford: Berg, 1997), chap. 1.

21 A vast literature from the past several decades has documented beyond question that elites use commemorations for political purposes. See, for example, the seminal collections of essays in John R. Gillis, ed., Commemorations: The Politics of National Identity (Princeton: Princeton University Press, 1996). Far less is known about the perspectives and motivations of the ordinary people who participate in these events.

134 For Albert, the purpose of marching is to maintain the memory of these great battles and then men who died in them. He understands his actions as part of a long line of war commemoration where the goal is wrapped up in the performance of tradition.

Kenny suggests a similar understanding: “You’ve got commemoration parades, like the

Twelfth. That to me is all part of my history. People fought and died for that and I want to be part of it to keep it going.” His interest is in maintaining the honor of those who gave their lives for the nation. Thus both Albert and Kenny are motivated to honor the memory of the victories and sacrifices of Ulster Protestant history. They are driven to preserve the past. Their comments attest to this intrinsic reasoning: the benefit they receive is the satisfaction of joining with friends “to continue the tradition” and “keep it going.” The motive and reward cannot be separated from the act itself.

It is important to note, however, that despite the intentions of most participants, these acts of remembrance have effects that reverberate across society. Primarily, the commemorated past is a sectarian, exclusively Protestant past. Catholics are almost always excluded—unless they appear as the villain. In their commemorations, participants largely remember, celebrate, and mourn the Protestant dead and those who died for the Protestant nation and the Union. This is common of national collective memory and commemorations, which seek to produce and reproduce myths, and therefore solidarity, of the nation. But in Northern Ireland and other plural societies, “memory tend[s] to divide rather than unite.”22 Divisive commemorations thus reinforce ethnic boundaries. Nevertheless, effect does not imply intent and across all my varied sources of data I found no systematic evidence that the mass membership of parading

22 John R. Gillis, “Memory and Identity: The History of a Relationship,” in Gillis, ed., Commemorations, p. 7.

135 organizations are motivated to perpetuate divisive memory qua divisive. As the participants cited here attest, most genuinely seek to honor their nation’s dead, not to promote a bifurcated past.23 !

Tradition

As the discussions of expressing collective identity and commemoration hint, participants are concerned with the role of tradition. They regularly account for both the overall existence of parades and their own reasons for parading by invoking “tradition.” Table 5.2 quantifies this pattern, showing that one-third of men who ever participated cite tradition as a purpose of parades. The survey also shows that 30 percent of men who have ever participated say that tradition is what attracted them to join their specific lodge or band. In interviews too, tradition is given as a major reason to participate. While some cite the concept without the term (“It’s what we do; we’ve done it for years,” says Albert), references to the term and its derivatives are common in my interviews. Billy, for instance, joined because of “tradition”; Robert began parading because “it was part of the family tradition”; and Kyle got involved because “that was just a family tradition.”

But what are we to make of the assertion of “tradition” as an explanation for action? How should we explain their reasoning? Communication theorist James W. Carey elaborates the dilemma: “Actions motivated by tradition, values, and affections pretty much escape our understanding and end up as the human interest exotica that fill the space between the self-

23 Some even went out of their way to mention their respect for the Catholics of the 16th (Irish) Division of the British army in World War I who died alongside the 36th (Ulster) Division at the Somme in the service of King and Country. “What do you do? Not remember them because they were Catholic?” Kenny asks rhetorically. “They were still soldiers, so that’s the way I look at it.”

136 interest stories and provide the features for the National Enquirer and Charles Kuralt.”24

There are two additional problems when trying to understand tradition as a reason for action. One, participants could be merely mimicking the official language of parading organizations. That is, interviewees might be giving me “little more than organizational slogans repeated as personal beliefs.”25 And “tradition” certainly is an important slogan for parading organizations: Bryan calls it “the dominant discourse used to legitimize the parades.”26 If this is the case, “tradition” is not a reason so much as a justification, and a memetic one at that.

Building on this is a second problem: participants could be using the language of tradition to mask their true motives. If true, tradition is not engrained in participants, but cleverly used by them to deceive observers and interlocutors (American graduate students very much included).

We know from Hobsbawm, Ranger, and their contributors that tradition and traditions are important ways to confer legitimacy on actions and institutions in the present.27 They find that in the modern era, traditions are not “genuine,” but “invented” and therefore are not a valid reason for action. From their perspective (as well as the many who follow them), the invocation of tradition is a red flag and must be exposed. But this story of elite invention ignores non-elite

24 James Carey, “The Dark Continent of American Journalism,” in Eve Stryker Munson and Catherine A. Warren, eds., James Carey: A Critical Reader (Minneapolis: University of Minnesota Press, 1997), p. 181.

25 Kathleen Blee, Inside Organized Racism: Women in the Hate Movement (Berkeley: University of California Press, 2002), p. 201.

26 Dominic Bryan, Orange Parades: The Politics of Ritual, Tradition, and Control (London: Pluto, 2000), esp. pp. 155-172. See, for example, Graham G.W. Montgomery and J. Richard Whitten, The Order on Parade (Belfast: Grand Orange Lodge of Ireland Education Committee, 1995), pp. 7-8. Thinking beyond parades, Jarman, Material Conflicts, p. 25, notes, “Tradition is one of the most over-used words in contemporary Northern Ireland.”

27 Eric Hobsbawm and Terence Ranger, eds., The Invention of Tradition (Cambridge: Cambridge University Press, 1983).

137 participants.28 The very reason why Hobsbawm and Ranger’s critique is so powerful is that traditions do not appear to be invented to most people most of the time; they appear traditional.

Thus when analyzing the motivations of the men and women who take part in a tradition, the feel and appeal of tradition can be strong and authentic. We need not be so suspicious and immediately pounce on the invocation of tradition for fear that it is invented. Rather, we should investigate what participants mean by it and how it operates.

When participants speak of parading as a tradition there are three layers. From most narrow to most general, they are parading as a personal tradition, as a family tradition, and as a communal tradition. Each layer adds depth of meaning to participating in parades by building a sense of continuing the past. This desire for continuity at each level is a powerful motive for participation.

As a personal tradition, parading is something that people have done for many years.

Starting to parade at age five or six is not uncommon, and even infants are brought by their parents to parades. For nearly all paraders, their engagement with parading culture began young.

The survey shows that 99 percent of participants attended parades in their childhood, and 60 percent belonged to the Junior Orange Order or a marching band before they turned 16. As Ben recalls, “it was always just a part of life. It’s what we did.” In many ways these past experiences add up cumulatively. Part of what makes current experiences so valuable is that they align with

28 See a related critique in Ian McBride, “Memory and National Identity in Modern Ireland,” in Ian McBride, ed. History and Memory in Modern Ireland (Cambridge: Cambridge University Press, 2001), p. 9.

138 years of accumulated memories. These memories accompany participants as they march today.29

At the family level, parading can link participants to father, grandfathers, and beyond. It gives participants the ability to take part in an action that was cherished by ancestors. As Joseph remarks: “when I parade with the very same band that my great-grandfather was in, that my father paraded in, there is a sense of continuity. There’s a sense of continuing something which is something to be proud of.” Parading allows Joseph to walk in his ancestors’ footsteps and sustain something that was dear to them. Billy feels similarly: “Carrying on the tradition is important,” he explains. “My tradition is going back generations in my family. So I think that it’s important that I keep that going. I think it’s important that doesn’t fall by the wayside because somebody couldn’t be bothered to take an interest in it.” A family history of parading also contributes to a personal history, since it shapes the environment in which one is raised—providing memories of watching one’s father on parade, for instance.30 This commitment to family can be an important motive for some. It provides a meaning and a purpose that is filled with love and loyalty.

Finally, at the community level, participating is a way to connect with the broader collectivity, past, present, and future. Participants recognize parades as “something my

29 On biographical continuity, see Doug McAdam, “The Biographical Consequences of Activism,” American Sociological Review, Vol. 54, No. 5 (October 1989), pp. 744-760; Silke Roth, “Developing Working-Class Feminism: A Biographical Approach to Social Movement Participation,” Pp. 300–23 in Sheldon Stryker, Timothy J. Owens, and Robert W. White, eds., Self, Identity, and Social Movements (Minneapolis: University of Minnesota Press, 2000); and Robert W. White, “Structural Identity Theory and the Post-Recruitment Activism of Irish Republicans: Persistence, Disengagement, Splits, and Dissidents in Social Movement Organizations,” Social Problems, Vol. 57, No. 3 (August 2010), pp. 341-370.

30 On family background influencing activism, see Donatella della Porta, “Recruitment Processes in Clandestine Political Organizations: Italian Left-Wing Terrorism,” in Bert Klandermans, Hanspeter Kriesi, and Sidney Tarrow, eds., International Social Movement Research, Vol. 1 (Greenwich, CT: JAI Press, 1988), p. 158; Hank Johnston, “New Social Movements and Old Regional Nationalisms” in Enrique Laraña, Hank Johnston, and Joseph R. Gusfield, eds. New Social Movements: From Ideology to Identity (Philadelphia: Temple University Press, 1994), p. 271; Sidney Verba, Kay Lehman Schlozman, and Henry Brady, Voice and Equality: Civic Voluntarism in American Politics (Cambridge, Mass.: Harvard University Press, 1995), p. 437; and White, “Structural Identity Theory and the Post-Recruitment Activism of Irish Republicans,” p. 354.

139 community has always done” and this leads to reverence and a desire to continue.31 Thus Craig says, “There’s tradition involved in it… and it’s a way of life,” and Frankie states, “[we’ve] been doing this for hundreds of years… It’s part of our make-up, it’s part of what we do.” In this way, the authority of tradition, as Weber argued, is legitimate because its “mores [are] sanctified.”32

The communal tradition of parading is continued not only because it is old, though this matters very much. Tradition motivates participation because people have “piety for what actually, allegedly, or presumably has always existed.”33 Because it is old and because it is sanctified, parading becomes “an inviolable norm of conduct” in the eyes of participants.34

The desire to follow tradition at all three levels is a reason for action. Weber suggests as much by placing traditional action among his four types of meaningful social action (along with instrumentally rational, value rational, and emotional action). He is, however, “respectful but hostile” to traditional action and authority, viewing it as non-rational.35 As Craig Calhoun explains, Weber “opposed traditionalism as mere unconscious reflex or unexamined inheritance to rationality as conscious and sensible action.”36

Parade participants seem to share Weber’s separation of instrumental rationality and

31 Interview with Kyle, June 11, 2014.

32 Max Weber, “Politics as a Vocation,” in H.H. Gerth and C. Wright Mills, trans. and eds., From Max Weber: Essays in Sociology (New York: Oxford University Press, 1972), p. 78.

33 Max Weber, “The Social Psychology of World Religions,” in Gerth and Mills, eds., From Max Weber, p. 296.

34 Ibid.

35 Arnold M. Eisen, “Constructing the Usable Past: The Idea of Tradition in Twentieth-Century American Judaism,” in Jack Wertheimer, ed., The Uses of Tradition: Jewish Continuity in the Modern Era (New York: Jewish Theological Seminary of America, 1992), p. 451 n. 7.

36 Craig Jackson Calhoun, “The Radicalism of Tradition: Community Strength or Venerable Disguise and Borrowed Language?,” American Journal of Sociology, Vol. 88, No. 5 (1983), p. 895.

140 tradition, but without the negative judgment of the latter. Indeed, they often treat tradition as a superior mode of and reason for action. Both Tom and Kyle demonstrate the contrast between instrumental action and tradition. I asked Tom what the overall goal of parading is and he replied: “The overall, uh— It’s part of our tradition, it’s part of what we’ve always done, what we do, who we are.” The question about goals seemed to throw him off and he stumbled for an answer. His response suggests that he sees no goal outside the action itself. A goal implies an instrumental orientation, while Tom’s understanding of parades lacks such an orientation.

Parades are a core part of how he understands himself and his place in the world, so thinking about them in terms of goals is nonsensical.

Kyle’s comment exhibits a similar distinction between tradition and instrumentalism. In response a question of whether parades send a message, Kyle said, “I wouldn’t say so. I’d say it’s a celebration of culture, it’s heritage, it’s something we’ve always done.” Sending a message is instrumentally rational, parading is something else: tradition. For both Tom and Kyle, “tradition” means action that is not goal oriented. It is action whose rewards are found in the pleasures and meanings of collective participation in a symbol-laden action that has been passed from generation to generation.

Yet, as mentioned, tradition carries negative connotations for Weber and many other social scientists as the forerunner to rational modernity. Weber’s definition of traditional social action is action “determined by ingrained habituation.” “Strictly traditional behavior,” he continues, “lies very close to the borderline of what can be called meaningfully oriented action…

141 it is very often a matter of almost automatic reaction to habitual stimuli.”37 My interviews demonstrate, however, that paraders do not understand their action as “ingrained habituation,” but pleasurable action full of meaning, much of which is rooted in the long personal, family, and communal histories of parade participation.

Interviewees repeatedly emphasize that the choices to begin parading and to remain active were their own. Frankie, for instance, recalls that though membership in Orange Order goes back several generations in his family, his father had quit at some point. Joining the organization was therefore “completely a choice on my part” and not simply continuing down a path determined by his family. Tom also sees deliberate decisions, not habit, as why he has remained active for so many decades. “My own choice,” he says, “was to stay in the Order. I could have left when I [got a new job] and became a Christian, but at each step I made a conscious decision to stay.” And though these may not be entirely accurate recollections of their mobilization (Chapter 7, for example, discusses structural factors involved in mobilization that are ignored in these tellings), they provide a glimpse into how participants understand their current actions.38 The element of active and informed decision-making that participants note suggests that they do not perceive parading as a habit. Further evidence comes from the fact that participants often quit parading or switch organizations for mundane reasons, not the types of dramatic reasons we might expect are needed to shock people out of a habit.

Thus, tradition as a source of action does not mean habit or something done without

37 Max Weber, Economy and Society: An Outline of Interpretive Sociology, Guenther Roth and Claus Wittich, trans. and eds. (Berkeley: University of California Press, 1978), p. 25. For a discussion of habit in political science, see Ted Hopf, “The Logic of Habit in International Relations,” European Journal of International Relations, Vol. 16, No. 4 (June 2010), pp. 539-61.

38 Munson, The Making of Pro-Life Activists, p. 198.

142 thought. Tradition, contra Marx, does not weigh “like a nightmare on the brain of the living”39 or compel action by eliminating conscious choice. In some circumstances, to be sure, tradition constrains behavior; but in others, such as loyalist parading, tradition motivates behavior. By providing meaning to action, it encourages mobilization.40 Framing an action as traditional connects participants to their past at the level of biography, genealogy, and community. It gives them the opportunity to connect with and continue history. Tradition thus motivates by providing an appealing sense of continuity. Through parading, participants can enter the stream of history to swim alongside forefathers and fading memories of childhood. These layers of history and the accompanying feeling of continuity ensure that parading is not merely a habit, but action

“freighted with culture, memory, and experience.”41 Achieving the experience of living with history and carrying the past into the present is what participants mean when they say they parade because of tradition. !

Defiance

Parading provides the opportunity for some participants to express defiance and moral outrage towards the perceived multifaceted attacks on their way of life by republicanism.42

39 Karl Marx, “The Eighteenth Brumaire of Louis Napoleon,” in Robert C. Tucker, ed., The Marx-Engels Reader (New York: Norton, 1978), p. 437.

40 See also Blee, Inside Organized Racism, p. 169; and Calhoun, “The Radicalism of Tradition.” As Jarman, Material Conflicts, p. 10, writes about loyalist parades: “The apparent historical continuity of ritual is an important feature of its power, the unchanging form is itself a major attraction, to join in and carry on a tradition, to follow in one’s father’s footsteps, or to wear ‘the sash my father wore.’” See also his discussion in pp. 25-28.

41 Courtney Bender, Heaven’s Kitchen: Living Religion at God’s Love We Deliver (Chicago: University of Chicago Press, 2003), p. 51.

42 As Joep Leerssen, “Monument and Trauma: Varieties of Remembrance,” in Ian McBride, ed., History and Memory in Modern Ireland (Cambridge: Cambridge University Press, 2001), p. 215, notes, “Even in Orange marches, the brash assertiveness is an act of defensive defiance rather than mere proclamations of supremacy.”

143 Unlike other oppositional messages that I will discuss below, defiance has the characteristics of process-oriented action. Defiance is “a refusal to acquiesce,” writes Wood. “Its value [is] not contingent on success or even on one’s contributing to the likelihood of success.”43 Its value, rather, is in the act of expression.

Since the end of unionist hegemony in 1972, and especially since the peace process of the mid-1990s, many Protestants have felt humiliated by and resentful of the changes in Northern

Irish society.44 According to Smithey, they “have tended to frame their experience of change… in terms of loss or the perpetual potential for loss.”45 The transition from ethnic dominance to a more level playing field is bound to cause pain in the formerly dominant community, but to make matters worse in the eyes of many Protestants, many of the Catholic politicians now in power were IRA combatants not long ago. It is an attack on their moral vision and their fundamental sense of right and wrong. “Just seeing terrorists sitting in government and now ruling over this country—that to me was I think shocking, to be quite honest,” says Ian, a parade supporter. As

Mark puts it: “Would David Cameron [or] Barack Obama bring… Osama bin Laden in? Give him a place in government? Because that’s what we are expected to do here.” Just having to accept this situation is unbearable enough, but the politicians “who had murdered and maimed and destroyed this country” are also trying to stop parades that they find offensive.46 For many, if

43 Wood, Insurgent Collective Action, p. 233. Also Jasper, The Art of Moral Protest, pp. 37-8.

44 Roger D. Petersen, Understanding Ethnic Violence: Fear, Hatred, and Resentment in Twentieth-Century Eastern Europe (New York: Cambridge University Press, 2002), esp. pp. 40-61, argues that resentment is the natural emotional response to a reordered ethnic status hierarchy.

45 Lee A. Smithey, Unionists, Loyalists, and Conflict Transformation in Northern Ireland (New York: Oxford University Press, 2011), p. 66.

46 Interview with George, August 14, 2012.

144 not all, participants, the irony is sickening. “Why can a convicted bomber tell me that I can’t walk down the road because I’ve got a flute?” Mikey asks with indignation in his voice.

Parades, therefore, are a way to take a stand, voice their disgust, and defy those who want to stop them. George, for instance, says: “Now I don’t mean that to be antagonistic, but [my intention is] to say, ‘We’ve always been here and we’re not going away.’” The result of George’s statement may well antagonizing Catholics, but his motive does not depend on that occurring.

His reason for participating is the “inner moral obligation” to vent his opposition against the moral outrages he sees in his society.47

Defiance, then, is a noninstrumental reason to participate. The reward is in the process, not the outcome of the action. At a large parade I attended with Sammy, he repeatedly pointed out bands to tell me that the town or village they are from used to have a Protestant majority, but now only a few are left. Yet, he would proudly say, they still have a marching band. Why? The band does not defend the Protestant community against the violence of the republican paramilitaries, nor is it designed to counteract demographic changes. The Protestant minority still supports a band out of defiance to what he views as the ethnic cleansing of those towns. !

The Pleasures of Participation

More generally, discussions about parades and what they mean to participants center on the pleasures of participation. Most participants I interviewed happily talked on and on about how much they loved to parade. Their enthusiasm was palpable. The pleasures they discussed were varied, but they emerge from the very act of parading, not the consequences. They take the

47 Bert Klandermans, “Motivations to Action,” in Donatella Della Porta and Mario Diani, eds., The Oxford Handbook of Social Movements (New York: Oxford University Press, forthcoming).

145 form of process-oriented benefits. The two most prominent sources of such benefits are the other people who participate and the emotions experienced while parading.

The social benefits of parading come from the opportunity to spend meaningful time with dear friends and articulate shared values together.48 For instance, what made Michael’s most recent parade “fantastic” was “just being part [of it] and just spending the day with people of like-minded views and showing your cultural identity. It makes me very proud. It excites me.”

He enjoys the company of his friends and acquaintances and the connection he feels with them.

But being with so many other Protestants also concretizes his collective identity and makes him feel part of something bigger. As Sewell notes, “Big demonstrations or mass meetings not only persuade the political authorities that the insurgents are, in Tilly’s words, ‘Worthy, United,

Numerous, and Committed,’ but also help persuade the insurgents of the same things.”49 Being part of such a large, collective event, where the people feel bound together by ideology, faith, and tradition, is exhilarating for Michael.

Rich also expresses positive experience of togetherness. “When you’re walking up the street in an Orange parade,” Rich reminisces, “… waving at people you know and being part of something, there’s a sense of belonging and being owned and owning something. It’s so hard to

48 James M. Jasper, “The Emotions of Protest: Affective and Reactive Emotions In and Around Social Movements,” Sociological Forum, Vol. 13, No. 3 (1998), p. 418.

49 William H. Sewell, “Space in Contentious Politics,” in Ronald R. Aminzade, Jack A. Goldstone, Doug McAdam, Elizabeth J. Perry, William H. Sewell, Sidney Tarrow, and Charles Tilly, eds., Silence and Voice in the Study of Contentious Politics (New York: Cambridge University Press, 2001), p. 58. He is quoting Charles Tilly, “Social Movements and (All Sorts of) Other Political Interactions—Local, National, and International—Including Identities,” Theory and Society, Vol. 27, No. 4, (August 1998), pp. 453-480, but Tilly used the WUNC concept in a number of his works.

146 define.… It’s an undefinable quality of belonging.”50 This feeling of collective belonging is “part of why it’s so hard to leave” a parading organization, yet Rich did just that. He quit the Orange

Order fourteen years earlier, after the signing of the Agreement, because he disagreed with the confrontational and sectarian stances taken by the Order during the peace process and Drumcree crises of the mid-1990s. Rich remains critical of parades and parading organizations—but the power of the experience is still unmistakable to him. Years later, the powerful social benefits that come from taking part in a parade continue to resonate with him. “Once you’re in it, you feel it,” he tells me.

As the interviewees’ discussions of the social experience of participation suggest, a significant part of that experience is emotional.51 Indeed, many of the social pleasures are rooted in the “reciprocal emotions” of activism, “the emotions generated in a social movement… [that] concern participants’ ongoing feeling toward each other. These are the close, affective ties of friendship, love, solidarity, and loyalty, and the more specific emotions they give rise to.”52

In addition to the affective ties between paraders, acting collectively to express deeply held beliefs produces an emotional experience. Even collective identity itself, as Jasper argues,

“is an emotion, a positive affect toward other group members on the grounds of that common

50 Psychology research shows that humans have a deep psychological need to belong to groups. See Roy F. Baumeister and Mark R. Leary, “The Need to Belong: Desire for Interpersonal Attachments as a Fundamental Human Motivation,” Psychological Bulletin, Vol. 117, No. 3 (1995), pp. 497-529.

51 On the role of emotions in activism and conflict, see Jasper, “The Emotions of Protest”; Jeff Goodwin, James M. Jasper, and Francesca Polletta, eds., Passionate Politics: Emotions and Social Movements (Chicago: University of Chicago Press, 2001); Petersen, Understanding Ethnic Violence; Helena Flam and Debra King, eds., Emotions and Social Movements (London: Routledge, 2005); Deborah B. Gould, Moving Politics: Emotion and ACT UP’s Fight against AIDS (Chicago: University of Chicago Press, 2009); James M. Jasper, “Emotions and Social Movements: Twenty Years of Theory and Research,” Annual Review of Sociology, Vol. 37 (2011), pp. 285-303; and Wendy Pearlman, “Emotions and the Microfoundations of the Arab Uprisings,” Perspectives on Politics, Vol. 11, No. 2 (June 2013), pp. 387-409.

52 Jasper, “The Emotions of Protest,” p. 417.

147 membership.”53 This affect manifests in multiple ways and can encourage both mobilization and commitment. Walter, for instance, illustrates the role of reciprocal emotions in maintaining commitment and persistence. “The band has been my main love all the years,” he says. It is “an extended family” to whom he is loyal. As a result, “It’s not only something that you do for a year or two years. It’s something that’ll stay with you all your life.” For Walter, this has meant thirty years of participation.

Of all the emotions experienced by participants, pride is among the most common and reports of it abound in the interviews.54 Pride, along with its opposite emotion, shame, are strong motives for human action.55 Thomas Scheff calls it a “master emotion,” noting that pride

“generates and signals a secure bond” between individuals and between groups.56 I found three primary sources of pride among paraders. First, participants are proud of their cultural heritage and of displaying it to others. They take pride in the opportunity to collectively show their culture to others—both in-group members and members of out-groups. Walter explains:

I love to get out there and show people what I can do and how proudly I can do it; how well I look and how well I sound when I do. It’s a pride thing, probably, for me. And it’s an identity thing, showing my identity as a loyalist to everyone—and I’m not just showing it to loyalist people, I’m showing it to nationalist people. I’m !showing it to people who really don’t care. Showing off who he is as an individual and as a group and how well he performs fills Walter with pride. This “pride thing” is a significant source of participants’ desire to express their

53 Ibid., p. 415.

54 This is not surprising, for as Jasper (Ibid., p. 418) writes: “articulating one’s moral principles is always a source of joy, pride, and fulfillment.”

55 Thomas J. Scheff, Bloody Revenge: Emotions, Nationalism, and War (Boulder, Colo.: Westview, 1994).

56 Ibid., p. 3.

148 collective identity discussed above. It is an emotional aspect of publicly identifying with a cherished identity and community.

Second, participants are proud of the recognition they get for their actions. As Tom describes, “There’s a pride involved in marching. There’s a pride in the crowd cheering and clapping. There’s that pride walking around.” A positive reception such as cheering is a typical source of pride, which “we feel… with achievement, success, and acceptance.”57 Walter explains the pride he feels when attracting the attention of tourists:

We get Japanese people or American people or people from all over the world coming over and asking, ‘Can I have [my] photo taken with you?’ So you’re doing something [or else] why would they want their photo taken with you? … They want their picture taken with somebody who looks well… I mean you’re !standing and you’re smart, your uniform’s clean, your boots are shining. All the positive attention stands in contrast to day-to-day life, when he knows that no one is going to approach him to snap a photo. The positive recognition he gets from visitors makes him proud and confirms his sense that he is doing something worthwhile.

In addition to the anonymous cheers from the crowd or recognition from tourists, pride is enhanced by personal connections between participants and audience members. Along the parade route, participants greet and are greeted by friends, family, co-workers, and acquaintances. “It’s fairly proud and enthralling moment,” says George, “because you can’t walk down the street without someone shouting out at you who knows you.” Participants smile and their faces light up as they shout hello, wave, nod, point, and occasionally jump out of the parade for a handshake or

57 Ibid., p. 39.

149 hug.58 After months of fieldwork, paraders I got to know would even wave to me on occasion.

These exchanges between participants and supporters illustrate the connections between the social and emotional elements of parading. They show that many of the emotions are generated in interaction with other participants and with the audience.

Third, participants are proud of their dignified self-presentation. Parades are an opportunity to dress up in dark suits and collarettes for loyal order members and colorful, elaborate, often military-style, uniforms for band members. “We go out and we’re all in nice suits and nice collarettes—a lot of money spent—and you’re going out and you’re proud… and your friends and family are proud to see you,” states Mikey, further touching on the social and interactive roots of emotions. Billy hones in on the role of fine and ceremonial clothing:

I enjoy the parades. The Twelfth of July, you feel proud. You like to wake up on the Twelfth morning and you get your best suit on, and get your best clothes on. You get your bowler hat out. You get your coat and hat out. You get your white gloves out. You get your cuffs out. You go on parade. You turn out your best to do the organization proud. And it’s just a feeling of [being] so proud of the tradition, ! the culture… And the history of the village, keep the history going. Looking sharp, respectable, and dignified on parade fills Billy, a truck driver by profession, with pride.59 Moreover, wearing clothing with a history connects him to that history.

The pride in wearing a bowler hat and white gloves comes not from being hip to the latest fashions, but in dressing in a manner that consciously invokes the past.60 As Hobsbawm remarks,

58 In fact, many interviewees mentioned that since audience members tend to stand in the same place year after year, they expect to see certain friends and family at specific locations along the parade route. Iddo Tavory, “The Private Life of Public Ritual: Interaction, Sociality and Codification in a Jewish Orthodox Congregation,” Qualitative Sociology, Vol. 36, No. 2 (June 2013), pp. 127, 126, argues that these “predictable forms of interaction” are a common feature of “public worship and ritual.”

59 Bryan, Orange Parades, discusses the role of respectability in parading.

60 See Connerton, How Societies Remember, pp. 45, 48, on how commemorative rituals deliberately refer to the past.

150 “The wigs of lawyers could hardly acquire their modern significance until other people stopped wearing wigs.”61 The same is true for the Orangeman’s démodé dress. Its current significance lies in its obsolescence. Thus, dressing and acting in an explicit and expressive manner produces pride. As Michael summarizes, “It’s always a proud moment to be on parade.”

The pride, excitement, and other positive, satisfying emotions are energizing and encourage sustained participation.62 Craig describes the energizing aspect of parades:

It’s not routine to me. It’s a totally refreshing day. It’s one of those days when after you’ve walked twenty-two miles, you finish up with a load of energy. You’re on a high, so you are. And it’s thoroughly enjoyed, thoroughly enjoyed. If I was to walk five miles there today, I’d have sore feet and blisters. There’s something !magical about being in the band and playing your flute. I never have a blister. Craig ends the day refreshed and full of energy despite being on his feet for many hours and miles. And though he might exaggerate the ability of parades to prevent blisters, he certainly conveys his sentiment. The face-to-face interactions, large cheering crowds, music, and feeling of taking part in something important all contribute to the visceral excitement that Durkheim calls “collective effervescence” and Randall Collins identifies as “emotional energy.” Durkheim noticed the energizing features of collective gatherings, arguing that all groups need to convene occasionally so that members may “renew their common faith by making a public demonstration of it together.” Face-to-face conventions have this effect, he holds, because “in the midst of an

61 Eric Hobsbawm, “Inventing Traditions,” in Hobsbawm and Ranger, eds., The Invention of Tradition, p. 4.

62 Many scholars of collective action have argued the rituals and the emotions generated by them are a central method for maintaining the enthusiasm and commitment of participants. The case of loyalist parading, however, shows that disengagement rates can still be high even when the organization’s primary form of mobilization is an emotionally rich collective ritual. For example, Rosabeth Moss Kanter, Commitment and Community: Communes and Utopias in Sociological Perspective (Cambridge: Harvard University Press, 1972); Jasper, The Art of Moral Protest, esp. chap. 8; Jasper, “The Emotions of Protest,” p. 418; Sharon Erickson Nepstad, “Persistent Resistance: Commitment and Community in the Plowshares Movement,” Social Problems, Vol. 51, No. 1 (February 2004), pp. 52-54.

151 assembly that becomes worked up, we become capable of feelings and conduct of which we are incapable when left to our individual resources.”63 This certainly seems the case in parades.

Building on Durkheim and others, Collins argues that emotional energy is the outcome of any successful interaction ritual. Emotional energy, Collins argues, “makes the individual feel not only good, but exalted, with the sense of doing what is most important and most valuable....

[It] has a powerful motivating effect upon the individual; whoever has experienced this kind of moment wants to repeat it.”64

The energy of parades is enhanced by the music that thunders through them. The music played by loyalist marching bands tends to have a military rhythm that is designed to march to.65

The beat sets a pace for the marchers and reverberates across the city, town, or village. Both marching together and music, Jasper argues, “have unusual capacities to make people melt into a group in feelings of satisfaction, perhaps because so many parts of the brain and body are involved at once.”66 Walter and Rachel illustrate this idea well. Walter explains that “if your band’s playing well and you bass drums sound loud, the hair on the back of your neck actually lifts!” Rachel narrates the experience of a “blow out,” when two bands march past each other and

63 Émile Durkheim, The Elementary Forms of Religious Life, trans. Karen E. Fields (New York: Free Press, 1995), pp. 211-212.

64 Randall Collins, Interaction Ritual Chains (Princeton: Princeton University Press, 2004), pp. 38-39.

65 For analyses of the musical elements of parading, see Gordon Ramsey, Music, Emotion and Identity in Ulster Marching Bands: Flutes, Drums and Loyal Sons (Bern: Peter Lang, 2011); Gordon Ramsey, “Band Practice: Class, Taste and Identity in Ulster Loyalist Flute Bands,” Ethnomusicology Ireland, Vol. 1, No. 1 (February 2011), 1-20; Ray Casserly, “Blood, Thunder, and Drums: Style and Changing Aesthetics of Drumming in Northern Ireland Protestant Bands,” Yearbook for Traditional Music, Vol. 45 (2013), pp. 142-63; and Ray Casserly, “Parading Music and Memory in Northern Ireland,” Música e Cultura, Vol. 9 (2014). On music and movements in general, see, for example, Ron Eyerman and Andrew Jamison, Music and Social Movements: Mobilizing Traditions in the Twentieth Century (New York: Cambridge University Press, 1998); and William G. Roy, “How Social Movements Do Culture,” International Journal of Politics, Culture, and Society, Vol. 23, Nos. 2-3 (September 2010), pp. 85-98.

66 Jasper, “Emotions and Social Movements,” p. 294.

152 play as loud and hard as they can. “It’s the height of being out with the band,” she says. “The adrenaline rush that comes through you when two competing bands playing opposite tunes just start passing each other and blasting it out and the bass drummer is going absolutely mental playing the tune. Yeah, your adrenaline goes up.” In both these cases, the intensity of the musical experience generates a physical reaction among participants, demonstrating the mental and bodily aspects of parading to music.67

These positive emotions and energy of parading also explain why Craig feels that something so often repeated is “not routine.” Conventional outcome-oriented approaches to participation find rituals difficult to explain because they expect that people would get bored or tired of repeating the same action, especially when there is no external goal or material rewards.68 They predict that repeated action would feel “routine” because the sole benefit is received upon its successful completion. But as Craig mentions, it does not feel this way. That is because the allure of parading and other rituals is in the action itself. The appeal is the experience: the pride, excitement, and energy; the quality time with respected friends and comrades; forging a tangible connection to the past. These are what encourage mobilization and sustained activism in a collective ritual.69

67 It is in this context that we should understand Gordon Ramsey’s remark that “It is only in the context of parading to music that loyalism exists as a mass-movement, united in common practice, for in all other practical contexts, political, religious or cultural, loyalists are fragmented and often mutually opposed.” Ramsey, Music, Emotion and Identity in Ulster Marching Bands, p. 147. Original emphasis.

68 Charles Tilly, Regimes and Repertoires (Chicago: University of Chicago Press, 2006), p. 41, remarks that in “contentious claim-making… perfect repetition from one performance to the next breeds boredom and indifference on the part of claimants and objects alike.”

69 This effect is also seen in William McNeil’s well-known recollection of aimless marching during basic training in the US Army during World War II. Though marching around in the blistering Texas sun seemed pointless at the time, decades later his main memory is that he “rather liked strutting around… Words are inadequate to describe the emotion aroused by the prolonged movement in unison that drilling involved.” William H. McNeill, Keeping Together in Time: Dance and Drill in Human History (Cambridge: Harvard University Press, 1995), pp. 1-2.

153 The sense of pride and energy, attained in the company of friends and family, is a prime experience of participation. My findings resonate with Gordon Ramsey’s ethnographic conclusion that marching in loyalist bands is motivated in large part by the “emotional rewards of participation.”70 More generally, they match what Wood calls “emotional in-process benefits,” the “emotion-laden consequences of action experienced only by those participating in that action.”71 Feelings like this make it unsurprising that, despite the lack of personal or collective material gains that I will demonstrate in Chapter 7, participants choose to return time after time.

Overall, the interviews demonstrate that parading is itself a benefit for participants. They look forward to parades, talk about them with their friends, and many, if not most, come to structure their lives around them. Some participants even told me about missing family functions and damaging relationships in order to march in parades. Yet they certainly do not view parading as a cost—on the contrary, as I will show in Chapter 7, they actually pay for the privilege to parade their identity with pride and moral vision alongside dear friends and companions. The paraders I spoke with would reject Olson’s framework out of hand; rather, I believe, they would embrace Hirschman’s view that free-riders “cheat themselves first of all.”72 As Billy says of friends who do not parade, “That’s their loss… You know, they’re missing out.”

* * *

As I mentioned throughout my discussion of collective identity expression, commemoration, tradition, defiance, and the pleasures of participation, they all share a particular

70 Ramsey, Music, Emotion and Identity in Ulster Marching Bands, p. 223.

71 Elisabeth Jean Wood, “The Emotional Benefits of Insurgency in El Salvador,” Goodwin, Jasper, and Polletta, eds., Passionate Politics, p. 268.

72 Albert O. Hirschman, Shifting Involvements: Private Interest and Public Action (Princeton: Princeton University Press, 1982), p. 87. The original is italicized.

154 form. These reasons for parading are intrinsic to marching in the parades. One, they are selective benefits available exclusively to participants. Two, they are rooted in the experience of participation, not the consequences of the parade. Their source is in the sights and sounds, the thoughts and feelings, the memories and people, not the social, political, or economic results, be they personal or collective, of the parade. As Jack, a Methodist minister who was in a band as a young man but quit parading years ago, reflects, “It’s not simply a parade. There’s a multiplicity of issues: … identity, religion, self-worth, self-esteem, nostalgia, family members… there’s so many competing emotions spilling into what looks like a normal day out… music, crowds, the ritual, symbolism, memory, past, pain.” All of these elements, and more, join together to generate the thrilling and meaningful experience of marching down a road to the sound of flutes and drums in Northern Ireland. !

Instrumental Reasons for Participating

Parade participants collectively and individually bring a variety of reasons and motivations with them.73 While the preponderance of the survey, interview, and ethnographic evidence I collected about participants’ thinking is best understood as process-oriented or intrinsic, some outcome-oriented or instrumental reasons exist as well. This is not problem for my argument. Following much of the literature, I have tended to describe rituals as action primarily done “for its own sake.”74 But rituals are also used as a means toward an end: to bring

73 “The simple realization that human motivation is sufficiently complex and is likely to run on more than one motivational ‘fuel’ is sufficiently close to a truism for the statement itself to be trivial,” notes Alexander A. Schuessler, A Logic of Expressive Choice (Princeton: Princeton University Press, 2000), p. 28. Yet much contemporary social science operates without such realization.

74 Frits Staal, “The Meaninglessness of Ritual,” Numen, Vol. 26, No. 1 (1979), p. 9.

155 rain or a good harvest, to increase fertility, to heal the sick. This is the side of ritual that Mary

Douglas calls “instrumental efficacy.”75 Thus, a focus on the ritual elements of human behavior does not preclude outcome-oriented motivations. Likewise, scholars of social movements now generally recognize that the old distinction between instrumental movements and expressive movements is unhelpful. Most movements are now thought to have instrumental and expressive aspects.76 Moreover, my claim that the ritual nature of loyalist parades provide the primary reasons for participation remains strong. The instrumental reasons I will now introduce are at the margins, rather than center, of the average participant’s thinking.

The central outcomes sought by participants in parades are internal and external communication—that is, sending a message to the Protestant in-group and Catholic out-group.77

Many scholars theorize ritual as a communicative device. Tambiah, for instance, defines ritual as

“a culturally constructed system of symbolic communication.”78 And Douglas claims that

“[r]itual is pre-eminently a form of communication.”79 But ritual is an imprecise form of communication. The ambiguity inherent in the meaning of symbols and rituals leaves room for

Protestant and Catholic audiences to receive a different message from the same parade. In the

75 Mary Douglas, Purity and Danger: An Analysis of the Concepts of Pollution and Taboo (London: Routledge, 1996), p. 69.

76 For example, Bernstein, “Celebration and Suppression”; and Taylor and Van Dyke, “‘Get up, Stand up.’”

77 Parades have also been interpreted as sending messages to other paraders, the British government, and unionist political parties and elites, but I rarely, if ever, heard these targets discussed by participants. See Jarman, Material Conflicts; Bryan, Orange Parades; and Lee A. Smithey and Michael P. Young, “Parading Protest: Orange Parades in Northern Ireland and Temperance Parades in Antebellum America,” Social Movement Studies, Vol. 9, No. 4 (November 2010), p. 402.

78 Stanley J. Tambiah, Culture, Thought, and Social Action: An Anthropological Perspective (Cambridge: Harvard University Press, 1985), p. 128.

79 Mary Douglas, Natural Symbols: Explorations in Cosmology (London: Routledge, 1970), p. 20.

156 coming paragraphs, I examine what these competing messages are. !

Unifying Protestants

For members of the Protestant in-group, the primary message is unity and solidarity. The intended effect, then, is social cohesion and communal strength.80 By gathering large crowds and icons of the community, parades not only represent Protestant unity, but cause it as well. Irish

Protestants have long understood communal unity as essential for their strength and survival on an island where they are a minority. But unity has also long been seen as elusive for a group divided by religious denomination, class, and political allegiances.81 Parades have often been seen as a source of and forum for Protestant unity.82

Interviewees share this diagnosis and solution. For example, Robert believes that parades help create cross-denomination and cross-class unity among Protestants by “bring[ing] a lot of people… together” and, crucially, by “show[ing] that we’re together.” By assembling so many people from across the religious and socioeconomic spectrums—“a lot of clergy from different denominations… [and] people from the humblest person, a bus driver, a member of the Loyalists

[paramilitaries] to a doctor and professors… a high court judge… and a lot of our MLAs and

80 Internally-directed messages have often been criticized as merely expressive outbursts of social movements, but as a number of scholars have pointed out, they can lead to highly instrumental outcomes, such as solidarity and renewed commitment. For example, Jasper, The Art of Moral Protest; Bernstein, “Celebration and Suppression”; and Casquete, “The Power of Demonstrations.”

81 Smithey, Unionists, Loyalists, and Conflict Transformation, pp. 208-210. On the Northern Irish state’s strategies for maintaining cross-class Protestant unity, see Mark McGovern and Peter Shirlow, “Counter-Insurgency, Deindustrialisation and the Political Economy of Ulster Loyalism,” in Peter Shirlow and Mark McGovern, eds., Who Are ‘The People?’: Unionism, Protestantism and Loyalism in Late Twentieth Century Ireland (London: Pluto, 1997), pp. 176-198.

82 Bryan, Orange Parades; and Paul Arthur, Government and Politics of Northern Ireland, 2nd ed. (London: Longman, 1984), p. 40.

157 politicians”—parades create a sense of community that might otherwise be lacking among

Protestants. But speaking with Robert, I never felt that seeking this unity was a priority or that he would stop if parading failed to achieve Protestant solidarity. Rather, for Robert and others with whom I spoke, constructing Protestant solidarity is a byproduct of participation—albeit a valued and desired byproduct—not a motive. The production of social cohesion may be a primary social function of parades and other mass rituals, but that does not imply that the individual actors are motivated to achieve the function.83 !

Opposing Irish Nationalism

For members of the Catholic out-group, the primary message is one of opposition.

Parades are a way that participants express opposition to a united Ireland, republican violence, and losses in power and prestige that resulted from the peace process. One manifestation of this opposition is the defiance discussed above with the other process-oriented reasons. The opposition discussed here, in contrast, is outcome-oriented and aimed externally. Unlike defiance, this opposition supposes an out-group audience to receive the message.

Isaac puts opposition squarely as his motivation to mobilize into a parading organization, recalling that “the impetus to join was probably political.” It was the early 1970s, the height of violence, and he did two things “as a way of saying, look, I’m British”: he joined the Orange

Order and he joined the part-time police force. For Isaac, using force and the law and using symbols and rituals are related, even linked, ways of opposing republicanism.

But Isaac is the exception, not the rule. For the majority of participants that I interviewed,

83 Jon Elster, “Marxism, Functionalism, and Game Theory: The Case for Methodological Individualism,” Theory and Society, Vol. 11, No. 4 (July 1982), pp. 453-82.

158 sending a message of opposition is either a minor reason to act or an unintended consequence of participation. George, for instance, articulates one of the most hardline oppositional stances of any of the interviewees. He absolutely intends for his parading to send the message that he wholeheartedly opposes Irish nationalism and its goals. But this developed out of his participation and was not an original reason to join nor a major reason to continue.84 He got involved because all his friends from the army were members and they asked him to join them.

The social element, not communicative role of parading, was his primary reason to act.

Importantly, the oppositional message, though read as aggressive by Catholics and other outsiders, is generally intended by participants as defensive. This defensive message is encapsulated in a phrase repeated by a number of interviewees: “We’re still here.” Mark explains that after the decades of “attacks that have been placed on our community,” parades say, “We’re still here, we’re still living, we’re still breathing.” Joseph elaborates: “‘We’re here, we exist…’

Not necessarily meant to be threatening or intimidating anybody, but ‘we exist’—it’s a simple message, ‘we exist.’ We see ourselves at the heart of our community and heart of the town, heart of the village and we don’t want to be marginalized and excluded.… Certainly there’s been a lot of times in the past where I have paraded… just to say, ‘listen, we’re here, we’re not going to threaten you, but we’re here, we exist.’” Joseph’s message of “we don’t want to be marginalized and excluded” is not the message of a community that feels in command, but one that feels under siege.

Ben expands on this sentiment. The message of parades, Ben states, are “that we exist.

We are a community, we exist, and we have things to celebrate and remember.” But, he believes,

84 See Munson, The Making of Pro-Life Activists.

159 that is not how Catholics view them:

You see, the thing is they are trapped in a mindset [that Protestants are politically dominant]. What have I got to be dominant about? I was born in 1973, a year after Stormont fell. … Unionists have not run Northern Ireland since 1972. What have I got to celebrate? Martin McGuinness [of Sinn Féin] should have been hung for treason against the British state for organizing and developing an attempt to overthrow the power of the United Kingdom… That hasn’t happened. He is now sitting in government as the deputy First Minister in the government of Northern Ireland. He helped murder dozens of people. He’s never been made amenable to justice. What have I got to be dominant about? But I’m still somehow sending this ! superior message that I’m better than them! Ben is exasperated and honestly baffled. Through his eyes, the past decades have been patently awful for Protestants, and he cannot understand how anyone could see otherwise. So as he sees it, paraders are being accused of triumphalism that they do not, even cannot, actually feel.

Some paraders, such as Isaac and Joseph, clearly participate in order to send an defensive message to Catholics. For them, parading, at least some of the time, is an instrumental means toward an external end. But no participant acts solely to send a message; they all also share the expressive, process-oriented motives held by others. Conversely, not all paraders act in order to send a message to Catholics. Several interviewees do not believe that their actions even send a message. We see saw this above in Mark’s hesitation when asked about parades’ message and the distinction he drew between sending a message and making a statement. We see it again in

Steven’s comment about parade terminology. The Orange Order often refers to its Twelfth parades as demonstrations, because they march to a “demonstration field” where there are speeches and benedictions.85 But Steven is not comfortable with this term. “I hate the word

85 See, for example, Grand Orange Lodge of Ireland, “Twelfth 2014,” available at: http:// www.grandorangelodge.co.uk/twelfth14.aspx. Accessed February 21, 2015.

160 ‘demonstration’ being used,” he says. “To me it should be ‘celebration,’ as opposed to

‘demonstration.’ We’re not demonstrating anything, we’re celebrating our culture and heritage.”

The word demonstration, to Steven, connotes protest or deliberately sending a political message.

He therefore sees it as an inappropriate description of parades. The celebration of culture is not a demonstration to Catholics or anyone else. !

Intimidation and Provocation

So far, I have argued that the meaning paraders broadcast is either defensive or process- oriented. What about the reasons for participation that are both aggressive and instrumental, intimidation and provocation? These are the reasons for participation most often attributed to paraders by outsiders. Intimidation is intended as a cheap way to gain compliance from a population by scaring people into submission. It is a way to maintain ethnic dominance.

Provocation is intended to elicit a reaction (and preferably an overreaction) from the out-group in order to influence the in-group. It is a way to mobilize the in-group, reinforce an ethnic boundary, or sideline in-group moderates or dissenters.

Participants never so much as hint that their intention is to intimidate or provoke

Catholics.86 On the contrary, they vigorously distance themselves from the divisive consequences of parades. In interview after interview, paraders emphasized that their actions are not intended to offend or antagonize anyone.87 Isaac, for example, told me earnestly, “I don’t think we’re going out to make an offensive statement. Certainly, anything that ever I have been personally

86 Table A2 in the Appendix shows that participation is not predicted by stating that “sometimes Catholics need to be reminded that they live in the United Kingdom.”

87 Offense and division may be unintended consequences, but they are not unanticipated consequences.

161 involved in, I haven’t seen any particularly offensive actions or anything that anybody could take as offensive.” Kyle recognizes that some Catholics do view parades as offensive, but, he maintains, it is the result of a misinterpretation: “There’s the perception that the bands go out to offend Catholics and to assert dominance over them, but it’s not the case. I mean we don’t go out on parade to dominate anybody or to claim anything is ours or to intimidate anybody.” He continues by pointing out, as several others did, that participants put a lot of time and effort into parades and if the goal was simply to antagonize Catholics, there would be easier ways to do it.

“I mean we practice I would say on average twice a week all year round,” Kyle says about his band. “We don’t do that to offend anybody.”

The time I spent with a marching band in West Belfast provides further support for Kyle’s two points (Kyle was not a member of this band). Participants did spend time and effort in practicing. At the weekly practices, they often practiced a single part of a song repeatedly until the musical director was pleased. And, as best I could tell, most if not all members could read music. These facts are unremarkable from the point of view of a musical group, but seems unnecessary if the goal is to provoke a riot.88

What is more, there were never any discussions of intimidating or provoking Catholics.

In fact, I never heard an overt discussion of politics at all, despite specifically listening for it. I was keenly attuned to any mention of politics, Catholics, or any instrumental goal in the privacy of the band practices, but it never came up.89 Indeed, the two most illuminating episodes in this

88 It also challenges prevailing stereotypes of loyalist marching bands as hoards of “knuckle draggers.”

89 My observations are likely impacted by the fact that the band practices I attended was a melody band, not a blood and thunder band. Melody bands require more musical skill and practice and they do not have the same reputation for aggressive sectarianism. That said, I am near certain I once saw this band in clearly violating a legally-binding Parades Commission decision by loudly playing music as they marched by St. Patrick’s Catholic Church in Belfast. No one would mistake them for goody two-shoes.

162 regard feature the absence of sectarian politics. The first is a recurring episode. Each practice session concluded with the band playing “God Save the Queen,” the United Kingdom’s national anthem. This was generally the only remotely explicit political or nationalist moment during the ninety-minute practice. When played on the streets, where Irish nationalists can hear it, “God

Save the Queen” can take a provocative or defiant flavor. In the privacy of a clubhouse, however, it does not. They generally played it perfunctorily, as it was the final task before they could go home. It seemed to me more “banal nationalism” or “everyday nationhood” than the “contrived occasions for the crystallization of national awareness” that often spill into the streets of

Belfast.90

The second episode involves a t-shirt. On the last practice before the Twelfth of July in

2013, Dan, one of the band leaders, wore a t-shirt with a cartoon of a potato and the words, in all caps, “THE FAMINE IS OVER SO WHY DON’T YOU GO HOME.” These are the lyrics of the expressly sectarian “Famine Song” which caused so much controversy outside St. Patrick’s

Catholic Church in Belfast the previous year. The shirt looked brand new—I noticed that the sleeves were still creased, as if it was just purchased—so I guessed that he had worn it specifically for the band practice. Yet despite the jokes and laughs I imagine Dan was hoping to inspire, I never heard any. Even with this invitation to discuss the intimidation or provocation of

Catholics, I heard no comments by other band members. I was not following Dan the whole night, so I could have missed private conversations, but nothing was said in front of the band as a whole. Furthermore, he wore the shirt in a private, Protestant forum, not in public, where

Catholics could see it. Does this make the act less insensitive? Likely no. But wearing it was not

90 Michael Billig, Banal Nationalism (London: Sage, 1995); and Jon E. Fox and Cynthia Miller-Idriss, “Everyday Nationhood,” Ethnicities, Vol. 8, No. 4 (2008), pp. 536-63, the quote is from p. 545.

163 an attempt to offend Catholics who were not a potential audience for this private act.

Given the prevailing unfavorable stereotypes about loyalist bands, the absence of sectarian conversations around these episodes is striking. But ethnomusicologist Gordon Ramsey reaches a similar conclusion in his long-term ethnographic study of two marching bands:

Participating in a loyalist parade is almost universally seen by outsiders as a display of political and religious allegiance, an act of power or resistance with a political goal. Some within the band may accept such characterization to some degree… Yet… in six years of intensive involvement in the band scene, spanning some of the most politically controversial events of the peace process, I have never heard politics or religion discussed seriously in a bandroom.91

The playing of the national anthem and the non-reaction to the wearing of a potentially offensive t-shirt hint that participants are generally not looking to provoke or intimidate

Catholics. The closest that any participants get to admitting a connection between parades and these outcomes are the few interviewees who mentioned that other participants want to cause trouble. Members of the loyal orders tended to the put the blame on bands, while band members tended to blame other bands or the “blue bag brigade,” young men who follow a parade carrying loads of cheap alcohol. Ramsey similarly describes how the bandsmen he knew referred “to other bands as political bands, by which they meant those bands with paramilitary affiliations.”92

A central reason why participants do not intend to offend Catholics is that they do not believe that Catholics are actually offended. Instead, participants believe that the objections are all disingenuous cant manufactured by the republican movement. Thus, the controversy is invented, Catholics feign the offense, and any violence that may follow a parade is, in fact, unrelated to the parade. “We’re not causing the trouble,” Sammy stresses. “It’s violent

91 Ramsey, Music, Emotion and Identity in Ulster Marching Bands, p. 163.

92 Ibid., p. 226. Original emphasis.

164 republicans who are causing the trouble.” These widely-held beliefs suggest that participants are not consciously trying to offend Catholics or polarize the communities. Since participants do not believe that Catholics get offended, it is unlikely that they act with the goal of offending them.

There is much more to say on this issue and I will return to it in more detail in the next chapter, where I argue that participants’ understanding of culture and cultural action helps them to disassociate parades from their consequences.

Of course, Protestants have strong incentives to say that they do not intend to effect the harmful outcomes of parades, even if it is untrue. Nevertheless, my formal interviews and informal conversations with a broad range of participants and observations at many parades and related events overpowered my strong skepticism and convinced me that most of them are sincere in their beliefs—even if the reader likely has justifiable doubts. More convincing evidence of the demotion of instrumental outcomes is that interviewees also devalued the positive outcomes of parades, such as entertaining the crowd of supporters. For example, when I asked Mikey if parading has a goal, he replied, “It’s not a goal as such.” But he then acknowledged the public service they provide: “There’s hundreds of parades going on in

Northern Ireland and people like to go out and watch them, it’s a spectacle. People like to listen to bands. People like to see their brothers and sisters, their family walking. People like to see it.”

And people who attend parades agree wholeheartedly. Sophie, an enthusiastic attender, describes being at parades this way: “Love it. … You just feel more of your cultural identity. You’re all together and it gives you a wee bit of hope for the future.” And local scholar Michael Hall records a band member recalling: “One woman [at a community meeting in east Belfast] said,

‘We struggle all year to make ends meet, and watching the bands is the only bit of holiday time

165 we get.’”93

So why would participants dismiss these positive outcomes, about which they can be justifiably proud? These beneficial outcomes do happen, and they are a satisfying byproduct of parading, but, they told me, it is not why they act. Both Michael and Howie, for instance, convey this idea. Michael states, “It’s our right to do it. It’s our belief to do it. The crowds and the spectators are an added bonus to us. We will walk the streets whether there is one person watching us or one hundred million people watching us. It’s our right, it’s our identity. It’s our reason we exist.” Howie agrees: “[When there is a large crowd] it’s just nicer. But if there wasn’t as many of a crowd there it wouldn’t bother me… I’ve been to parade where there hasn’t been many people watching it, but it’s still an honor and a privilege for me to walk with the

Apprentice Boys.”

My field observations confirm their claims: on occasion, I was the only person on the street watching the parade, particularly when the parade was in the early morning or the pouring rain. A crowd of cheering supporters certainly makes a livelier atmosphere, but they will parade whether or not there are spectators. As Michael and Howie make clear, entertaining the community is not what drives them. Rather, their motives are internal to the very process of participation: honor, privilege, fulfilling a right, acting on a belief, and living out one’s identity. !

Conclusion

Parades generate significant social and political effects in Northern Ireland. Among other things, they build Protestant solidarity, entertain deprived communities, intensify communal

93 Michael Hall, Towards a Shared Future (5): Ulster’s Marching Bands (Belfast: Island Publications, 2014), p. 9.

166 friction, and occasionally trigger violence. Some of these effects are intended, so the participants’ reasons are ends-oriented and instrumental. But many are not intended, so to understand their participation we must maintain a distinction between the reasons for the action and the consequences of the action. Parades may lead to solidarity, entertainment, tension, provocation, intimidation, or violence, but that does not mean that these outcomes are motivations. To say that they are commits the fallacy of explaining behavior simply by stating its effects. In particular, the divisive consequences of parades are a byproduct of action motivated by other ends and desires.

To attribute them as motives for paraders would deliver “a very distorted picture of their participation.”94 The participants’ reasons for participating are related to effects that are far less visible because they are intrinsic to the process of acting collectively, such as carrying on a tradition and the fulfillment experienced when publicly affirming a cherished identity.

This chapter has demonstrated that, as expected by the argument developed in Chapter 4, participants’ reasons for acting are primarily process-oriented. I used interview, survey, and ethnographic data to show that participants concentrate on the process of participation and largely disregard the outcomes. Yet the outcomes, especially the polarizing outcomes, seem hard to overlook. The next chapter suggests how participants sustain their process-oriented thinking in the face of glaring political consequences of their parades. I will continue to argue that parades produce outcomes that do not necessarily motivate participants and show how they disassociate their means from the ends that are produced. In so doing, I will provide further evidence that participants’ reasons for acting are the intrinsic benefits of publicly parading their beliefs and values with friends and family.

94 Nina Eliasoph and Paul Lichterman, “Making Things Political,” in John R. Hall, Laura Grindstaff, and Ming- Cheng Lo, eds., Handbook of Cultural Sociology (New York: Routledge, 2010), p. 489.

167 Chapter Six !Culture, Politics, and the Paradox of Anti-Politics in Loyalist Parades ! This is the difference in thinking, Jonathan. If there was a parade by a religious group in the United States and they flew the Star Spangled Banner on the parade, would you deem that parade as being political? Probably not. But if you fly the Union Jack in Northern Ireland, it’s political. ! — Mark, Orangeman ! A curious feature of the discourse surrounding loyalist parades in Northern Ireland is the impassioned insistence of many Protestants that parades are apolitical. In this chapter, I will show that at the center of this claim is a paradox: parades have patently and widely recognized political causes and consequences, yet many parade participants insist that they are not just apolitical, but anti-political—that parades transcend politics and exist apart from it. How is it that these activities, which from the outside look to be political activism, are imagined and encountered as non-political by participants? To answer this question, we must examine how participants construct, understand, and experience social categories such as politics and culture.

Through this we can see how they try to remove parades from the realm of politics, making parades “inaccessible to deliberation or contestation.”1

I begin by demonstrating the contours of the paradox in the interviews I conducted with parade participants. The interviews show that many participants fully recognize the politics ingrained in parades, while at the same time maintaining that parades have nothing to do with politics. Parades, they argue, are about culture, a social category they see as mutually exclusive to politics. Second, I explain why this paradox matters. In particular, I argue that there is political power in the claim of anti-politics. It shapes outcomes by shifting the debate away from

1 Nina Eliasoph and Paul Lichterman, “Making Things Political,” in John R. Hall, Laura Grindstaff, and Ming- Cheng Lo, eds., Handbook of Cultural Sociology (New York: Routledge, 2010), p. 483.

168 bargaining and compromise, thereby making the conflict over parades more intractable. Third, I probe how this seemingly untenable paradox is maintained by the ritual nature of parades. As rituals, parades provide participants with apolitical reasons to participate and the symbolic ambiguity of ritual allows for alternative interpretations of the events. Together, these two features create the behavioral and attitudinal environment that sustains the idea of anti-politics, thereby perpetuating this aspect of the Northern Irish conflict. !

The Paradox of Anti-Politics

Though they have served varying political projects over their two century history,

Protestant parades in the north of Ireland have been thoroughly political since their earliest days.2

Their political character remains to this day—that is, parades “striv[e] to influence the distribution of power… among groups within a state.”3 By calling parades political, I refer to two specific characteristics: they make political claims and they have political consequences.

The primary claim made by parades is that the six counties of Northern Ireland should remain part of the United Kingdom. This claim is an answer to the paramount question in

Northern Irish politics: should the territory be part of the Republic of Ireland or the United

Kingdom? The two main answers to this question form the most significant political and social cleavage in the province by far. All politics is debated around the unionist-nationalist cleavage

2 Neil Jarman, Material Conflicts: Parades and Visual Displays in Northern Ireland (Oxford: Berg, 1997), pp. 25-79; Dominic Bryan, Orange Parades: The Politics of Ritual, Tradition, and Control (London: Pluto, 2000), pp. 29-96; Charles Tilly, The Politics of Collective Violence (New York: Cambridge University Press, 2001), pp. 120-7.

3 Max Weber, “Politics as a Vocation,” in H.H. Gerth and C. Wright Mills, trans. and eds., From Max Weber: Essays in Sociology (New York: Oxford University Press, 1972), p. 78. This political character is evident to nearly all outside observers—a remark by Marc Howard Ross, Cultural Contestation in Ethnic Conflict (New York: Cambridge University Press, 2007), p. 88, is archetypal: “These cultural enactments are political statements— provocations and challenges, rights claims, assertions of power, and public acts of commitment.”

169 and all policies are interpreted through it. This question was also central to the thirty years of violence in the late twentieth century.4 At times, parades proclaim political loyalty to the U.K. implicitly using symbols, such as flying the Union flag or playing “God Save the Queen,” while at other times they make it explicitly, such as in speeches on the Twelfth of July. Through these means, each loyalist parade makes a claim about the most important political question in

Northern Ireland.

In addition to claims about the constitutional question, parades make several other, but no less significant, claims. Smithey and Young argue, “Orange parades have operated as expressions of loyalty and dissent across at least three relational domains: between Protestants and Catholics, between Protestant unionists and British governments, and within unionist politics.”5 Between

Protestants and Catholics, parades make claims about power, status, and ethnic hierarchy. For example, Ruane and Todd describe loyalist parades as “a symbolic assertion of power over

Catholics.”6 Further, parading is “the most prominent means of… claiming dominance over territory,” ranging from specific streets to whole neighborhoods to public space generally.7

4 For distinctly political interpretations of the Troubles, see, for example., Richard Bourke, Peace in Ireland: The War of Ideas (London: Pimlico, 2003).

5 Lee A. Smithey and Michael P. Young, “Parading Protest: Orange Parades in Northern Ireland and Temperance Parades in Antebellum America,” Social Movement Studies, Vol. 9, No. 4 (November 2010), p. 402.

6 Joseph Ruane and Jennifer Todd, The Dynamics of Conflict in Northern Ireland: Power, Conflict, and Emancipation (New York: Cambridge University Press, 1996), p. 109.

7 Jarman, Material Conflicts, p. 79. Allen Feldman, The Formations of Violence: The Narrative of the Body and Political Terror in Northern Ireland (Chicago: University of Chicago Press, 1991), chap. 2, “The Spatial Formations of Violence,” discusses the role of public space in the Troubles, with particular attention to parades on pp. 29-30. The aural intensity of parades allows them to mark territory far beyond the specific streets they march on—the sound of flutes and drums echoes across the city. As Tom Boylston remarks: “Territory is not just about the occupation of land. It is just as much about soundscapes and sightlines, not to mention tastes and smells.” Tom Boylston, “What Kind of Territory? On Public Religion and Space in Ethiopia,” The Immanent Frame (blog), August 26, 2014, available at: http://blogs.ssrc.org/tif/2014/08/26/what-kind-of-territory-on-public-religion-and- space-in-ethiopia. Accessed September 11, 2014.

170 Although most participants object to this characterization of their parades and explicitly argue that they are not making these claims vis-a-vis Catholics, they are most certainly read this way by most Catholics, who, whether or not participants intend it, are a relevant audience.8

Between Northern Irish unionists and British governments, parades make claims about the interests of the unionist community, primarily ensuring that London does not abandon them to drift toward Dublin. For example, Smithey notes that “historically, Protestant elites used the latent threat of a large popular loyalist community, publicly manifested in Orange parades, to ensure due attention to the union by British governments.”9 Finally, within unionist politics, parades make hardline claims on unionist political parties, often serving to narrow their bargaining position.10 The Orange Order, for instance, campaigned for a “no” vote on the 1998 referendum on the Agreement. Parades are also used by rival loyalist paramilitary organizations to assert power and claim territory.

Importantly, all of these claims change over time in response to shifting social and political conditions. Despite their self-promoted appearance as a timeless and unchanging traditional ritual, parades and their claims respond to changing currents in all of the relations just

8 Compare Jack Santino, “Public Protest and Popular Style: Resistance from the Right in Northern Ireland and South Boston,” American Anthropologist, Vol. 101, No. 3 (September 1999), p. 518: “Certainly not all parade participants, nor all members of the Orange Order, march to flaunt their power in a territorial or triumphalist way, but many do. Not every Catholic reads the parades in this manner, but many do.” On how social movements construct their audiences, see Kathleen Blee and Amy McDowell, “Social Movement Audiences,” Sociological Forum, Vol. 27, No. 1 (March 2012), pp. 1-20; also James M. Jasper, “A Strategic Approach to Collective Action: Looking for Agency in Social-Movement Choices,” Mobilization, Vol. 9, No. 1 (February 2004), p. 6.

9 Lee A. Smithey, Unionists, Loyalists, and Conflict Transformation in Northern Ireland (New York: Oxford University Press, 2011), p. 123.

10 See, e.g., Henry Patterson, Ireland since 1939 (Oxford: Oxford University Press, 2002), p. 8: The Orange Order’s “political importance resided not simply in the fact that it had representation as of right in the [Ulster Unionist] party at all levels, or that the overwhelming bulk of Unionist MPs and cabinet members were members, but also in its ability to criticize the party and the government if they were seen to deviate from the defence of Protestant interests.”

171 discussed: unionist-nationalist, Protestant-Catholic, unionist-British, and unionist-unionist. For example, Jarman describes how the Orange Order reacted to Daniel O’Connell’s campaign for

Catholic emancipation in the 1820s by “marching more often, in more places and with more people.”11 And Bryan documents how as working-class Protestants became more alienated from the British state over the course of the Trouble, parades lost their position as “an expression of the state.”12 This was reflected in the rise of blood and thunder bands (and the concomitant decline of more “respectable” marching bands), the display of loyalist paramilitary iconography, and the reduced display of the British flag. Paraders’ reactions to political events naturally also engendered responses by other actors. For example, the Royal Ulster Constabulary reported in

1981 that: “Against the emotional background of the Republican hunger strikes and various

Loyalist activities, the traditional parades required a higher level of policing than usual.”13 More serious reactions to periods of severe parade-related violence included the banning of all parades in Ireland by the British government from 1832 to 1845 and 1850 to 1872.14

Political claim-making by parades is not unique to Northern Irish Protestants. Rather, parades are “common features of contentious politics.”15 Historian Susan Davis claims that “as public representations, parades and public ceremonies are political acts: They have pragmatic

11 Jarman, Material Conflicts, pp. 53-56, quote is from p. 56.

12 Bryan, Orange Parades, p. 157.

13 Royal Ulster Constabulary, Chief Constable’s Report, 1981 (Belfast: Royal Ulster Constabulary, 1981), p. 10, Northern Ireland Political Collection, Linen Hall Library, Belfast.

14 Jarman, Material Conflicts, pp. 54-61. Though the prohibitions were not always enforced when it came to Orange parades. Just one striking example: an estimated 100,000 people marched in the Belfast Twelfth in 1870. Ibid., p. 65.

15 Smithey and Young, “Parading Protest,” p. 394.

172 objectives, and concrete, often material results.”16 Indeed parades are often arenas for contesting authority, generating legitimacy, challenging policies, forging new political identities, confronting rival groups, asserting personhood, and claiming citizenship.17 They are thus an important part of the modern repertoire of contention.18

The second reason parades are political is that they have political consequences. They are hotly debated by politicians in the Northern Ireland Assembly and local government councils, as well as in the media. Unionist politicians have used parades to campaign and connect with voters en masse since at least the mid-nineteenth century,19 and the tradition continues today. For example, David Trimble’s role in the Drumcree parade dispute in 1995 helped propel him to the leadership of the Ulster Unionist Party,20 which placed him at the center of the peace process— and, three years later, in Olso to share the 1998 Nobel Peace Prize. More recently, politicians regularly give speeches from the platform on the Twelfth of July and many elected officials have spoken at the ongoing weekly protests over the current parade dispute in north Belfast.

The history of politicking at parades is long, but the history of Protestant-Catholic

16 Susan G. Davis, Parades and Power: Street Theater in Nineteenth-Century Philadelphia (Philadelphia: Temple University Press, 1986), p. 5.

17 There is a rich literature on parades and politics. Notable examples include: Ibid.; Lynn Hunt, Politics, Culture and Class in the French Revolution (Berkeley: University of California Press, 1986); Mona Ozouf, Festivals and the French Revolution, Alan Sheridan, trans. (Cambridge: Harvard University Press, 1988); Mary Ryan, “The American Parade: Representations of the Nineteenth-Century Social Order,” in Lynn Hunt, ed., The New Cultural History (Berkeley: University of California Press, 1989), pp. 131-53; Mabel Berezin, Making the Fascist Self: The Political Culture of Interwar Italy (Ithaca: Cornell University Press, 1997); and Geneviève Zubrzycki, “Aesthetic Revolt and the Remaking of National Identity in Québec, 1960-1969,” Theory and Society, Vol. 42, No. 5 (2013), pp. 423-75. David I. Kertzer, Ritual, Politics, and Power (New Haven: Yale University Press, 1988), provides a general and comparative synthesis.

18 See Charles Tilly, Contentious Performances (New York: Cambridge University Press, 2008).

19 Bryan, Orange Parades, pp. 44-6; Jarman, Material Conflicts, pp. 61-2

20 Bryan, Orange Parades, p. 1.

173 violence is longer. At the Orange Order’s very first Battle of the Boyne commemoration parade, in 1796, a Mr. McMurdie exchanged “some words” with a local militia member, the two “came to blows,” and “Mr. McMurdie received a stab of which he died.”21 Thus began a sustained relationship between parades, disorder, and violence that has continued to this day. Parades have proved particularly likely to stir unrest during periods of Catholic political assertiveness and mobilization, such as O’Connell’s campaign for emancipation in the 1820s and the civil rights movement of the late 1960s. As a result of this combustibility, historically and today, the performance of parades often requires the coercive apparatus of the state—police, riot police, and, until 2006, the British military. Parades also create opportunities for both loyalist and republican paramilitaries to mobilize and assert their continued presence.22 Even in the absence of overt physical violence, parades polarize the two communities and heighten tension between them.23 They thus have a detrimental effect on the political peace process and grassroots peace- building efforts. In this manner, sustained violence has been replaced by largely non-violent conflict over parades.24 Echoing Clausewitz, Kertzer argues that “the Protestant-Catholic struggle in Northern Ireland, like so many other political battles, continues to be waged as much through ritual as any other means.”25

21 Belfast News Letter, 15 July 1796, quoted in Bryan, Orange Parades, p. 33; and Jarman, Material Conflicts, p. 47.

22 Paul Nolan, Northern Ireland Peace Monitoring Report, Number 2 (Belfast: Community Relations Council, 2013), pp. 62, 168.

23 Smithey, Unionists, Loyalists, and Conflict Transformation, p. 122, argues that “although both loyalist and nationalist parading is in some technical sense nonviolent, it is also tied to coercion and intimidation.”

24 See Neil Jarman and Dominic Bryan, Parade and Protest: A Discussion of Parading Disputes in Northern Ireland (Colraine, UK: Centre for the Study of Conflict, University of Ulster, 1996), p. 41; and Ross, Cultural Contestation, p. 5. Compare René Girard, Violence and the Sacred (Baltimore: The Johns Hopkins University Press, 1977).

25 Kertzer, Ritual, Politics, and Power, p.124.

174 The parade participants themselves often recognize one or both of these political characteristics of parades—though, as I will soon demonstrate, they resolutely refuse to recognize parades as political. In particular, three themes emerged in my interviews with participants which betray a political interpretation of parades. First, some participants readily acknowledge that parades make a political claim about the constitutional question. For example,

Jesse, an Orangeman and DUP elected official and party strategist, articulates this view forcefully. He mentioned to me that he got involved in the Orange Order through his political activism, so I asked him to explain the relationship between parades and politics:

Well there is when it comes to identity. There is when you consider— And, so it’s not party political. So you’ll not find the Orange Order advocating for political position of the Ulster Unionist Party, the Democratic Unionist Party, the Progressive Unionist Party, or whatever. But on those big ticket items, on Northern Ireland being a part of the United Kingdom, on reverence for the !monarchical system that we have and our head of state, it is, it is political. Here Jesse makes clear that in his view parades are political because of their message about the

“big ticket items,” namely remaining in the United Kingdom. Though, observe how he begins by qualifying his remarks by stating that parades are not political in the sense of party politics.

Jesse also illustrates a second theme: parades help the political prospects of unionism by unifying the Protestant community. Parades are politically useful “not only [in] energizing people, [and] getting the people on the streets, but [in] showing that there’s such level of support for political aspiration, for a political goal.” As a politician, he understands the political utility of large, public demonstrations. They mobilize citizens and excite them about the unionist cause, which is valuable since it re-energizes the unionist base. They also symbolize, and thus help engender, Protestant unity and communal solidarity. Protestants have long understood unity as

175 necessary for the community’s survival, but difficult to achieve due to internal divisions based on religious denomination, class, and political party allegiances. Parades are seen as a source of and forum for unity across those divisions.26 This even affects people who do not attend parades. By

“showing that there’s such level of support,” parades can influence non-attenders by demonstrating to them that the ideology has huge support as evinced by large crowds. This is important to communicate to supporters as well as opponents of unionism. Protestants see that unionism is alive and well—which might forestall electoral defection—while Catholics see that it remains a formidable political force.

Others speak more explicitly about sending a message to the Catholics, a third political theme. For them, parades are a ritualized means of communication with nationalists and republicans. That parades are ritualistic matters for how the communication is sent, and how it is received.27 In particular, the message of parades is made obliquely through the display of symbols—flags, collarettes, uniforms, music, banners—and through the physical presence of the marchers. So the two necessary ingredients which jointly constitute a parade—symbols and bodies—are simultaneously the means for communication.28 That means that parades generally

26 E.g., Bryan, Orange Parades; Paul Arthur, Government and Politics of Northern Ireland, 2nd ed. (London: Longman, 1984), p. 40. See, for instance, Michael Hall, Towards a Shared Future (5): Ulster’s Marching Bands (Belfast: Island Publications, 2014), p. 30: “The unionist community is more and more fragmented, and the bands are the only thing which is holding the Protestant working-class community together at the present moment.” Like all of the volumes in Hall’s excellent “Island Pamphlets” series, Towards a Shared Future (5) is an edited transcript of a conversation between people from Northern Ireland on a topic of pressing concern. Throughout the chapter, I supplement what I heard in the field with footnoted quotes from anonymous band members who took part in this conversation. Also Gordon Ramsey, Music, Emotion and Identity in Ulster Marching Bands: Flutes, Drums and Loyal Sons (Bern: Peter Lang, 2011), p. 147.

27 Writing on loyalist parades, Lee A. Smithey, “Strategic Collective Action and Collective Identity Reconstruction: Parading Disputes and Two Northern Ireland Towns,” Ph.D. dissertation, University of Texas, Austin, 2002, p. 96, notes: “Ritualistic collective action… facilitates communication with opponents and third parties.”

28 For an anthropological analysis of the relationship between Orange Order icons and the body, see Feldman, Formations of Violence, pp. 57-8.

176 eschew the use of explicit signs, flyers, speeches, chants, or other canonical elements of the contemporary contentious politics repertoire, and instead rely on the core elements of its very performance to communicate a message.29

At the core of the communication to Irish nationalists is the message, “We are here.” For instance, George says that “the intent” is “to say we’ve always been here and we’re not going away.” Republican dreams be damned, George suggests—Protestants are on Ireland and are not going anywhere. Kenny agrees: Parades are “about making the stand that we can’t just be forced out of our own country,” he says. “It shows we’re still here,” states Walter.

Some specifically see their parading as an act of opposition to nationalism and republicanism. For example, John sees parades’ message as one of negating republicanism rather than just affirming unionism. “I look at it as not simply about a parade, but it has to do with civil liberty. It has to do with tyranny. It has to do with [not] giving into the violence, giving in to the people who want to take away your civil rights, which I would look upon as republicanism.”

Through parading, this clergyman sees himself taking a stand against Irish republicanism.30 In all of these examples, we see that communication with nationalists is intentional. Whether they are trying to maintain the Protestant political position or erode that of nationalists, parades are used as a means of political communication.

While participants acknowledge these claims and consequences of parades, they nevertheless generally refuse to accept that they are political. We can begin to see the tension

29 On the modern demonstration repertoire, see Jesus Casquete, “The Power of Demonstrations,” Social Movement Studies, Vol. 5, No. 1 (May 2006), 45-60; Olivier Fillieule and Danielle Tartakowsky, Demonstrations, Phyllis Aronoff and Howard Scott, trans. (Halifax: Fernwood, 2013); and Tilly, Contentious Performances, esp. pp. 71-87.

30 See, also, Hall, Towards a Shared Future (5), p. 27: “Our band culture is the last obstacle standing in the way of militant Irish Republicanism.”

177 caused by this paradox in Mark. On the one hand, Mark makes statements that suggest he believes that parades are inherently political and are used politically. On the other hand, he expresses a reluctance to characterize parades as political actions, identifying the Northern Irish context as the reason why parades are considered political. Regarding the former point, he suggests that parades have a political component due the fact that they commemorate political events. “Obviously there is a political part,” Mark says, “I mean the Battle of the Boyne was won in 1690 by the king of England, which in a way heralded a change in almost the constitution of the United Kingdom that guaranteed the rights of everybody.” Secondly, he argues that parades today make an explicit political claim:

You know, the parades are almost a statement of what we stand for, and we stand for the union between Northern Ireland and the United Kingdom.… It is a political statement in that regard. Although, we don’t set out— We simply set out to celebrate, “This is what we are and this is what we stand for,” and what we !stand for is the union of Great Britain and Northern Ireland. As I noted earlier, “stand[ing] for… the union of Great Britain and Northern Ireland” is one of two predominant positions in all Northern Irish politics. But then Mark’s view shifts as he argues that rather than being fundamentally political, parades are made political by the political context.

He provides two comparisons to show how this is the case. First, he says that while parades in

Northern Ireland are Protestants saying, “We’re still here,” in the Republic of Ireland, parades mean something entirely different since “they have a different set of problems to deal with.” In other words, there is nothing about parades that gives them the political claim of “we’re still here;” it results from the context. And second, he argues that parades have “political connotations… because… you are carrying the Union [Flag].” “This is the difference in thinking,

Jonathan,” he tells me. “If there was a parade by a religious group in the United States and they

178 flew the Star Spangled Banner on the parade, would you deem that parade as being political?

Probably not. But if you fly the Union Jack in Northern Ireland, it’s political.” Again, it is the vexed politics of Northern Ireland that renders parades political, not parades’ nature. So Mark is torn between his understandings of the sources of parades’ politics. He cannot quite seem to decide whether parades are political by nature or whether Northern Ireland’s social, political, and historical context made them so. His reluctance to accept the politics of parades is not unfounded. In fact, it is deeply rooted in the Protestant public discourse around parades. For most participants, as I will now show, there is not even a question.

Many, if not most, participants strongly believe that parades are in no way political.31 In fact, for many participants, parades are anti-political, in that they exist outside the realm of politics. Rather, parades have been strategically politicized by the Republican movement.32 Thus parades, which are not political, have been made political by the enemies of Protestantism.

What do I mean by anti-politics? By identifying paraders’ claims and rhetoric as anti- political, I intend to highlight their view that parades are not merely not political, but above and beyond politics.33 Anti-political is a stronger, more extreme stance than just apolitical. Something that is apolitical is passively not political, whereas being anti-political is an active and antagonistic stance against politics and the political. Things that are anti-political are not on the

31 See also Ramsey, Music, Emotion and Identity, pp. 226-7; and Ray Casserly, “The Fyfe and My Family: Flute Bands in Rathcoole Estate,” Irish Journal of Anthropology, Vol. 13 (2010), p. 11.

32 Eliasoph and Lichterman, “Making Things Political,” p. 483, usefully define politicizing as “action, collective or individual, that makes issues or identities into topics of public deliberation or contestation.” Conversely, depoliticizing is “making once-salient issues or identities inaccessibly to deliberation or contestation.”

33 Hannah Arendt, The Human Condition, 2nd ed. (Chicago: University of Chicago Press, 1998), p. 242, makes a similar distinction when she argues that “love… is not only apolitical but antipolitical.” For a case study of the anti- political effects of love, see Jeff Goodwin, “The Libidinal Constitution of a High-Risk Social Movement: Affectual Ties and Solidarity in the Huk Rebellion, 1946 to 1954,” American Sociological Review, Vol. 62, No. 1 (February 1997), pp. 53-69. Jim Jasper pointed out this connection to me.

179 same spectrum as things that are political; there is an absolute difference in kind.

To understand how parades are anti-political, we must first know how participants view politics. Politics, for most participants, is the domain of elite and electoral competition. In their eyes, politics is about politicians and political parties jockeying for advantage, challenging each other, and competing for votes. This is a narrow, but not incorrect, conception of political life.34

By this definition, contemporary parades are not especially political, particularly when compared to parades in earlier decades when the Orange Order was tightly linked to the ruling Ulster

Unionist Party and Orange parades were practically party rallies.35 In contrast, the parading organizations today do not advocate for a particular party—which is not to say that unionist politicians do not try to use parades, as events and as an issue, to their personal and party’s advantage: they most certainly do. In large part this is because the membership of parading organizations reflects the panoply of unionist political parties.36 As Rachel puts it, “We avoid bringing up politics in band because it’s a very, very, very sensitive issue.… It just takes one

34 In fact, many political scientists might agree. For a discussion of varying approaches to “politics” and “the political,” see Andrew Mason, “Politics and the State,” Political Studies, Vol. 38 (1990), pp. 575-87. The “elite competition as politics” approach I find in the interviews is even narrower than what Mason calls the “narrow approach,” which is anything having to do with government. For an important discussion of how observers should approach “the political” and navigate the differences between their own views and the “native” view, see Nina Eliasoph, Avoiding Politics: How Americans Produce Apathy in Everyday Life (New York: Cambridge University Press, 1998), pp. 14-15.

35 From 1905 to 2005, the Orange Order sent 15 percent of delegates to the Ulster Unionist Council, the governing body of the Ulster Unionist Party. Eric P. Kaufmann, The Orange Order: A Contemporary Northern Irish History (New York: Oxford University Press, 2007), p. 216. The two organizations broke their formal tie in 2005. The Order’s real power in the UUP is reflected in the fact that from the formation of Northern Ireland until the imposition of direct rule, all but three Stormont cabinet ministers were Orangemen. Bryan, Orange Parades, p. 60.

36 See Jocelyn A.J. Evans and Jonathan Tonge, “Unionist Party Competition and the Orange Order Vote in Northern Ireland,” Electoral Studies, Vol. 26 (2007), pp. 156-167; Jon Tonge, Jocelyn Evans, Robert Jeffery, and James W. McAuley, “New Order: Political Change and the Protestant Orange Tradition in Northern Ireland,” British Journal of Politics and International Relations, Vol. 13, No. 3 (2011), pp. 400-419; and Kaufmann, The Orange Order, esp. chap. 8, “Breaking the Link: Orange-UUP Relations after the Good Friday Agreement.”

180 snide comment and a whole band can be torn apart.”37 So at least in this confined sense, parades are not political.

But that is not the main reasoning that drives participants’ belief. Rather, participants believe that parades are not political because they are cultural, which they see as mutually exclusive categories. Culture and politics are often juxtaposed by participants and defined in contrast to each other.38 For example, Craig states that, “[Parades] wouldn’t be seen as political from anybody coming from a parading side. It’s a cultural thing.” We can see the contrasting concepts employed by Robby, a very active parader, as he compares himself to his brother. His brother is very interested in politics: “He can’t not watch the news every night and he was a member of the DUP party and stuff. He was quite politically minded.” Robby, however, has “no real interest” in politics. Instead, he is interested in parades: “As [for] what I do as in the

Orangeism and the band scene and stuff like that… I’m more culturally-minded to politically- minded.” Culture and politics, then, are seen as distinct, even opposing, points of view. For

Robby, culture replaces politics: while his brother lives and breathes politics, Robby lives and breathes culture.

The contrast between the cultural and the political is made explicit by Rachel:

Any time it’s said that the Twelfth of July is making a political statement, no, it’s not. It’s a cultural statement. The only time that it would be making a political statement [is] if it’s a protest against [something], like say flag protest, or if it’s a civil rights march, then that’s political. But if it is cultural, it’s not.

For Rachel, there is no overlap between culture and politics. Politics is protests and “trying to score political points against other Unionist parties,” as she says several moments later; culture is

37 See also Ramsey, Music, Emotion and Identity, p. 145.

38 See also Ibid., p. 227.

181 parades, tradition, music, history, and the like. And if it is one, it cannot be the other.

The belief that parades are cultural and not political is built on more than just definitional and conceptual distinctions. Parades’ non-political nature rests on a comparative and empirical foundation as well. One piece of evidence that several informants cited to me is that parades are not political because they are a widely enjoyed cultural practice. How can parades in Northern

Ireland be political, they ask, when they are simply one example of a global means of self- expression and celebration that elsewhere is apolitical? Edward is certainly correct when he says,

“People all around the world celebrate with parades. It’s not unique to Orangemen, to Northern

Ireland, or to Britain.” So is Mikey, who provides more details to this argument: “The army has parades, schools in America have parades, they have parades on St. Patrick’s Day. People want to parade.” And in all of those cases, he argued, there are no political challenges. Therefore, the logic insists, parades in Northern Ireland are not political either.

Secondly, parades are a tradition, in that they have been practiced for a long time. Again, this is true: parades commemorating the Battle of the Boyne date to the 1740s, the Orange Order held its first parade in 1796, and the Apprentice Boys was founded in 1814.39 The conclusion that participants draw is that since parades have been going on for a long time, they cannot have relevance to contemporary politics. Traditions, by definition, do not change over time: their

“object and characteristic… is invariance.”40 Thus, a tradition that began in the eighteenth century and has continued unchanged to the present cannot speak to today’s political debates.

39 James Kelly, “The Emergence of Political Parading, 1660-1800,” in T.G. Fraser, ed., The Irish Parading Tradition: Following the Drum (London: Palgrave Macmillan, 2000), pp. 15-16; Bryan, Orange Parades, p. 33; T.G. Fraser, “The Apprentice Boys and the Relief of Derry Parades,” in Fraser, ed., Irish Parading Tradition, p. 174.

40 Eric Hobsbawm, “Inventing Traditions,” in Eric Hobsbawm and Terence Ranger, eds., The Invention of Tradition (Cambridge: Cambridge University Press, 1983), p. 2. As Jarman, Material Conflicts, p. 9, notes, repetition in and of rituals “enhances the feeling that they never change.”

182 “How can there be a political parade,” Dan asks, “when it’s been walking that road for one hundred years?”

Relatedly, parades are not only from the past, they are about the past: they are commemorations. Since the subject of parades is the past, not the present, they do not have relevance to contemporary politics. If anything, the era that parades commemorate was a time when Catholics and Protestants were united against a common enemy. As a number of interviewees smugly pointed out to me, William of Orange and Pope Alexander VIII were allies in 1690. How, they ask, can parades which celebrate a cause supported by the Pope himself be considered political by Catholics?

Yet even the staunchest defender of parades’ anti-politics recognizes that parades today are disputed. But if parades are not political by their origins, their subject of focus, or their history, why are they so contested today? How did they become the subject of such intense political controversy? The nearly universally-held answer is that that parades were politicized by the republican movement during the Troubles. Sinn Féin, the Provisional IRA (PIRA), and others deliberately transformed parades into subjects of politics. By doing so, republicans inserted politics where it did not belong and had not been before. And their campaign of politicization succeeded: parades today remain “not political, but they’re perceived as political.”41 This explains why events which, from participants’ perspective, are and always have been cultural are so political today.

41 Interview with Craig, December 12, 2012.

183 The history interviewees told me again and again goes like this.42 For centuries, parades passed uncontested. In fact, they were attended and enjoyed by Protestants and Catholics alike.

Many participants tell stories of childhood Catholic friends who wanted to join the parade with them, or of a Catholic neighbor coming over on the morning of the Twelfth to pin a flower on their father’s lapel.43 This all changed around the 1980s and 1990s. In those years, Sinn Féin and the PIRA developed a strategy to turn the Catholic population vehemently against parades. The intensity of this campaign increased during the peace process as the PIRA began demobilizing troops and Republican prisoners were released. Suddenly, the republican leadership needed something for these men to do beyond armed conflict. So, they placed operatives in Catholic communities to manufacture anti-parade emotions and orchestrate local protests against parades.

A number of informants referenced a well-known rhetorical question that Sinn Féin president

Gerry Adams posed to a party meeting in Athboy, Ireland in 1997: “Do you think Drumcree happened by accident?”44 As protests mounted and succeeded in a particular place, republicans moved their attention to new parade routes, with the intention of one day banning all parades

42 The history is also recounted by parading’s senior leaders in public addresses. See, for example, Grand Secretary of the Grand Orange Lodge of Ireland Drew Nelson’s speech of January 17, 2014, excerpted in “Is the Long War on the Orange Order Over?,” available at: http://www.grandorangelodge.co.uk/news. Accessed 9 February 2014.

43 See also Jack Santino, Signs of War and Peace: Social Conflict and the Uses of Symbols in Public in Northern Ireland (New York: Palgrave, 2001), p. 34. Bryan, Orange Parades, pp. 69-70, points out that the postwar era (1945 to mid-1960s), when parades were relatively harmonious, was the formative period for the generation of who became loyal order leaders during and after the Troubles. Friendlier times were within their living memories which shaped their reaction to growing Catholic opposition.

44 Drumcree Church parade was the most contested parade in the mid-1990s. Dominic Bryan point out to me the irony that this is the only thing Gerry Adams ever said that these men believe.

184 from Northern Ireland.45

This narrative, in varying degrees of detail, is ubiquitous among my interviews. For example, Dan remembers:

Twenty years ago on the Twelfth day, Catholics used to come out. Now it’s all been IRA orchestrated to try to stop bands going where they want to go. The bands used to go by Catholic areas and they would turn the radio up: live and let live, it’ll be by in a minute. But now they’re out in the streets trying to stop !parades. It has changed. Isaac also has rosy memories of a bygone era: “Neighbors lived at peace with each other.

What has happened is that we’ve all become contentious now. You know, you didn’t need a

Parades Commission before the Troubles.” Craig told me that “the reality is people have marched down the same road there for umpteen years… without causing offense until Sinn Féin/

45 Parts of this account are accurate. There was a republican strategy to target parades that began in those years, and Adams was reported to have said that. See Malachi O’Doherty, The Trouble with Guns: Republican Strategy and the Provisional IRA (Belfast: Blackstaff, 1998), pp. 173-76 and 183; Martyn Frampton, The Long March: The Political Strategy of Sinn Féin, 1981-2007 (London: Palgrave Macmillan, 2000), pp. 126-7; Harvey Cox, “Keeping Going: Beyond Good Friday,” in Marianne Elliott, ed., The Long Road to Peace in Northern Ireland, rev. 2nd ed. (Liverpool: University of Liverpool Press, 2007), pp. 162-3; and James Dingley, “Marching Down the Garvaghy Road: Republican Tactics and State Response to the Orangemen’s Claim to March their Traditional Route Home after the Drumcree Church Service,” Terrorism and Political Violence, Vol. 14, No. 3 (Autumn 2002), pp. 42-79. Nevertheless, Bryan cautions that we should not give too much credit to the strategic view which “underestimates the local community dynamics that have long existed in places such as the Garvaghy Road and the political space the peace process created for public opposition to parades.” Dominic Bryan, “Parade Disputes and the Peace Process,” Peace Review, Vol. 13, No. 1 (2001), p. 45. However, my interviewees’ idealized interpretation of the years before 1980s is the result of selective memory: there is a long history of sectarian violence at loyalist parades and their predecessors. See Bryan, Orange Parades; Jarman, Material Conflicts; Sean Farrell, Rituals and Riots: Sectarian Violence and Political Culture in Ulster, 1784–1886 (Lexington: The University Press of Kentucky, 2000); Mark Doyle, Fighting Like the Devil for the Sake of God: Protestants, Catholics and the Origins of Violence in Victorian Belfast (Manchester: Manchester University Press, 2009). This selective memory seems to have a history of its own: for example, just days before July 12, 1969, the Northern Irish Prime Minister James Chichester-Clark and the Inspector-General of the RUC “were assuring the General Officer Commanding in the Province… that Orange marches had never been a source of communal strife.” Bourke, Peace in Ireland, p. 99. Yet my point is not to dismiss my interviewees’ understanding of history; it is to investigate how their understanding of the past works in the present. What kind of power does this narrative have and how does it affect behavior? For a general account of cultural polarization following political polarization, see Ann Swidler, Talk of Love: How Culture Matters (Chicago: University of Chicago Press, 2001), pp. 169-75. She specifically addresses ethnic conflict on p. 259 n. 10. For a similar analysis to mine also based on interviews with Protestants, see Neil Southern, “Territoriality, Alienation, and Loyalist Decommissioning: The Case of the Shankill in Protestant West Belfast,” Terrorism and Political Violence, Vol. 20, No. 1 (2007), pp. 79-80 (though Southern only interviews elites). On recent dynamics, especially the irony that the residents groups got too powerful and now Sinn Féin cannot control them, see Henry Patterson, “Beyond the ‘Micro Group’: The Dissident Republican Challenge,” in P.M. Currie and Max Taylor, eds., Dissident Irish Republicanism (New York: Continuum, 2011), pp. 75-6.

185 IRA politicized it.” Or, in Mark’s telling:

During the thirty years [of the Troubles] the republican community weren’t interested in parades, because they had other things on their mind: how they were going to plan their next atrocity, who they were going to murder next. They seem to have changed tactics and come away from that. ‘Ok, let’s pick a soft target now !during the peace process,’ which is a parade. The story they tell is of a passive Protestant community being acted on by republican agents.46 Paraders are not agents in their own telling of history; politicization happened to them.

Participants speak of parades being “pulled into politics” by “a very serious negative propaganda attack from Sinn Féin.”47 “The situation has been politicized,” says Isaac. As these examples illustrate, when speaking of parades, participants tend to use the passive voice; but when talking about Republicans, their verbs are active, and their language is the language of conspiracy:

“There was a deliberate policy formulated by Sinn Féin/IRA… when they were in prison”; parades become contentious when “republicans deem them to be contentious”; republicans

“sectarianize the unionist tunes”; “they’re going out of the way to cause problems.”48 Billy summarizes the sentiment: “It’s all orchestrated. It’s created, deliberately created conflict where there was no conflict.”49

46 It is common for groups to view their rivals as far more organized and unified than themselves, or than the rival actually is. See Robert Jervis, Perception and Misperception in International Politics (Princeton: Princeton University Press, 1976), chap. 8, “Perceptions of Centralization.”

47 Interview with Craig, December 12, 2012.

48 Interview with Craig, December 12, 2012; Interview with Mark, July 11, 2013; Interview with Sammy, July 9, 2013; Interview with Howie, August 13, 2012. For interpretations of the conspiratorial elements of loyalist ideology, see Ronnie Moore and Andrew Sanders, “Formations of Culture: Nationalism and Conspiracy Ideology in Ulster Loyalism,” Anthropology Today, Vol. 18, No. 6 (December 2002), pp. 14-15.

49 Note that interviewees are now using “politics” differently than they did before. Recall that when discussing unionist politics, they use the term narrowly to mean intra-unionist party politics. Now, when discussing nationalist politics, they use the term much more broadly to mean contestation.

186 A conclusion drawn from this narrative is that all anti-parade sentiment is artificial.50 The only people who are offended by parades are those who choose to be offended by parades. For many Protestants, parades are a synecdoche for their community, so to be offended by parades is to be offended by the the Protestant community—and to challenge parades is to challenge the community’s very existence.51 So those who choose to be offended by parades do so because they hate Protestants and want to drive them off the island of Ireland. Any and all opposition to parades is, consequently, illegitimate.52

For all of these reasons, participants understand parades not just as not political, but in contradistinction to politics, despite their widely known political claims and consequences. The paradox of anti-politics comes through most clearly in individuals who explained to me the patently political functions and effects of parades, but ardently refused to call them political. We

50 Or perhaps even coerced. Billy told me the following: “I had a very good friends in there who are Catholics and talking to them and chatting like the way we are, they would say, ‘The parade’s been going down here for a lifetime. It didn’t annoy me, because I didn’t go to see it. But I didn’t tell you that. I can’t openly say that. If I openly say that, I would be told to get out.’ So they’d be under pressure for that.” And Steven said: “I have been told by Catholic friends that they were visited by Sinn Féin activists and told that if they got anywhere near Orange parades, the Twelfth of July, well, the repercussions would be fair… so then they stopped coming.” Whether or not these stories are true, they are certainly part of the dominant narrative.

51 Marc Howard Ross, Cultural Contestation, esp. pp. 22-6, provides a rich cultural theory of this dynamic, with a specific analysis of loyalist parading in Chapter 4, “Loyalist Parades in Northern Ireland as Recurring Psychocultural Dramas,” and Marc Howard Ross, “Psychocultural Interpretations and Dramas: Identity Dynamics in Ethnic Conflict,” Political Psychology, Vol. 22, No. 1 (2001), pp. 157-78.

52 For an alternative interpretation of this dynamic of ethnic conflict, see Sherill Stroschein, Ethnic Struggle, Coexistence, and Democratization in Eastern Europe (New York: Cambridge University Press, 2012), p. 238. She finds that in Romania and Slovakia, people see also tend to blame ethnic mobilization by another group on manipulation by the other group’s elites. From this she draws the interesting conclusion that this is a mechanism to reduce tensions in the daily interactions with members of the other ethnicity. Rather than blame your neighbor for contentious mobilization, you can blame their elites for manipulating them. This way you can continue shopping at their stores and saying hello on the street. I interpret this tendency to blame rival elites in Northern Ireland very differently. Rather than facilitate good relations, I see it as stripping “the enemy” of their own autonomy and agency. It transforms the other group into one undifferentiated whole who are slaves to their masters. In the context of Protestant-Catholic relations this builds from and feeds into the old idea that Catholics just follow orders. It is at least partially rooted in theological differences between the two faiths, where Protestantism is seen as rooted in personal choice, while Catholicism is rooted in submission (see John Bell, For God, Ulster or Ireland? Religion, Society and Security in Northern Ireland [Belfast: Institute for Conflict Research, 2013], pp. 5, 7, 61-71, 106). Placing the blame on republican elites reinforces the old idea that “Home Rule equals Rome Rule.”

187 see this, for example, with Robert, a retired factory worker very active in the Orange Order in

West Belfast: !JB: Now is parading at all political? Robert: Well I don’t think so. Now there was a time it would have been termed that [because of the connection to the UUP], but I’m thinking over the last 30 years it’s not, it used to be everything had to be a politician on the field, but you find a lot of places that wouldn’t be. Yes, there’s members of the Orange Order that are politicians and they’d maybe speak, but it’s normally they’ve done away !with the political, excuse me, agenda so. !JB: So does parading help to maintain the union? Robert: Yes, it does let people see we’re still here and we haven’t gone away and I !think that’s one of the main things. !JB: What’s one of the main things? Robert: That we see that we maintain the union with Great Britain. [It] lets people !see we’re still very strong in numbers. … !JB: So if parading relates to unionism, is that not political? Or is that— Robert: Not really. I would say most of the Orangemen would—Alliance [Party] people not many and some, well, quite a lot of independents, but most of the Orangemen would be of a unionist family background. No, I wouldn’t, I wouldn’t !say it was political. Robert had just told me that helping to maintain the constitutional union between Northern

Ireland and Great Britain is “one of the main things” about parades. It is also, of course, the main thing in Northern Irish politics. But then when I pushed him to acknowledge the politics involved in that action he demurred and returned to a discussion of party politics.

We see the same dynamic with Walter, the bandsman from a West Belfast interface community. He first argues that parades are not political on the grounds that they are unrelated to political parties:

188 JB: So some people would argue that parading in general is political. I’m !assuming you would disagree. !Walter: Well we don’t have any political agenda. ![…] Walter: I don’t really know any bands that are [political]. The DUP probably !would have had— JB: Well maybe not party political, but political in the most basic sense of the !sovereignty issue or the constitutional issue. Does that play— !Walter: No. !JB: That plays no role in it? Walter: No, not to a great extent. A lot of people believe that they want to be British, but it wouldn’t be played out in a role within a band. The band wouldn’t !play a role of being—of wanting to keep the union, if you know what I mean. Political agendas, according to Walter, are things that politicians and political parties have.

Cultural organizations, such as his band, do not have political agendas. In fact, individual members suppress their political beliefs in the context of the band. But then later in the interview,

I asked Walter if parading has a goal, and he said:

I think it has. It shows that we’re still here. It shows that the people still want to be part of Ulster. They want to remain loyalist, they don’t want to be part of anything else, just our own—This is our wee country. This is our band’s walk on the streets to show that we can do, how many of us there is, and that’s the support !there is for our cause, which is loyalism. So somehow, though Walter wants to “be part of Ulster,” opposes a united Ireland, and supports the (political) cause of loyalism, his is not a political agenda. The reason why he, and many others, construct this contradiction is the subject of the coming section. !

189 The Power of the Paradox of Anti-Politics

Thus far I have demonstrated that the anti-political discourse of parades promoted by participants is paradoxical. But, to paraphrase Lori Beaman, the interesting and important question is not whether or not parades are political. Rather, what is interesting and important is

“what is achieved, which power relations [are] shifted and preserved, by recasting” parades “as

‘cultural’ rather than” political.53 What is achieved by the anti-politics paradox, I argue, is the creation of a powerful political tool to be wielded by parade participants and their allies.

Specifically, the paradox provides them with two advantages. First, the anti-politics paradox protects participants’ self-concept as good people. Participants’ approach to politics is shaped by an understanding of the world where culture is inherently good and politics is inherently bad. By forcefully separating culture from politics, the language of anti-politics places parades and paraders on the moral high ground. Second, I argue that there is political power in the discourse of anti-politics. This discourse shapes debates and political outcomes because if parades are not political, they are immune from critique and they are protected from compromise. Participants use the logic of anti-politics to silence democratic opposition. Participants, therefore, want parades to be outside of politics because, ironically, there is political power in the claim of transcending politics.54 So parades are more than just passively apolitical, they are actively anti- political. These two features of the anti-politics discourse explain why there is such a strong and sustained resistance to thinking of parades as political.

53 Lori G. Beaman, “Battles over Symbols: The ‘Religion’ of the Minority versus the ‘Culture’ of the Majority,” Journal of Law & Religion, Vol. 28, No. 1 (April 2012), p. 79.

54 See also Bryan, Orange Parades, esp. chap. 10, “‘Tradition,’ Control, and Resistance.” Ramsey, Music, Emotion and Identity, p. 228, suggests a third reason why bandsmen in particular resist a political label, “music defined as ‘political’ is devalued as music” (emphasis in the original).

190 Are these not just post hoc rationalizations of their behavior? I believe they are not. The elements of the anti-politics paradox are what sociologists Gresham Sykes and David Matza call

“techniques of neutralization.” Sykes and Matza argue that techniques of neutralization are commonly “viewed as [justifications] following deviant behavior and as protecting the individual from self-blame and the blame of others after the fact. But there is also reason to believe that they precede deviant behavior and make deviant behavior possible.”55 Thus the discourse of anti- politics more than just rationalizes controversial parading behavior—it enables it.

For most participants, as previously discussed, politics and culture are separate spheres of social life. Politics is the sphere of politicians, their parties, and their competition for power.

Culture, meanwhile, is the sphere of faith, tradition, and heritage. Layered on top of this distinction is an implicit normative ranking: culture is good, politics is bad.56 Culture is pure, politics is corrupt. Culture is about communal identity, politics is about individual greed. This negative view of politics emerges in interviewees’ language and argumentation. Parades are

“tarnished as being political,” Rachel told me. Politics is dirty and soils parades. For one thing, politics is about naked self-interest. As George puts it, parades “crept into the political realm because various politicians have used the political network to assist their own ends.” He puts the blame on Sinn Féin, but then also criticizes “our political people, who have used it for their own ends when needed.”

Part of this conceptual division is the idea that “we” do culture, but “they” do politics.

55 Gresham M. Sykes and David Matza, “Techniques of Neutralization: A Theory of Delinquency,” American Sociological Review, Vol. 22, No. 6 (December 1957), p. 666.

56 For a general analysis, see Colin Hay, Why We Hate Politics (London: Polity, 2007); also Eliasoph, Avoiding Politics.

191 Interviewees argue that nationalists have even pushed politics on their own culture. Irish nationalists politicized, and thus excluded Protestants from, elements of Irish culture such as the

Irish language and St. Patrick’s Day.57 Likewise, republicans made parades political to serve their political agenda. Dan sums up the general thinking. He was telling me about protests orchestrated by the IRA, so I asked him, “Why is the IRA interested in this?”

“They want to break the loyalists down. They want a united Ireland,” he replied.

All of the protests, all of the anger, all of the restrictions, then, are part of the republican agenda for a united Ireland. And they are able to pull it off because republicans are brilliant strategists and excellent manipulators: “The republican media/PR machine has been so good,” as

Mark puts it. Protestants, on the other hand, are just plain-spoken, honest folk who were outgunned from every angle, so to speak. Tom explains the problem:

They’re better at manipulating the press, they’re better at spin.… We’re blunt and straight to the point… We call a spade a spade, that’s what it is. We don’t flower it up or dress it up, and sometimes that straight truth needs dressing up. Republicans !dress it up, so people tend to believe what republicans are saying. So “they,” with their sophisticated “propaganda machine,”58 play politics, while “we” do what

57 Interview with Frankie, December 4, 2012; Interview with Steven, December 5, 2012. In a speech at the weekly loyalist protest in North Belfast, Grand Master of the Belfast County Orange Lodge, George Chittick, warned Protestants against learning the Irish language because “it’s part of the republican agenda.” According to the BBC, “He said the Irish language had not been ‘political’ in the past, but this had been changed in recent times by republicans.” Mark Simpson, “Orangeman Says Protestants Should Not Learn Irish Language,” BBC News Online, February 1, 2014, available at: http://www.bbc.co.uk/news/uk-northern-ireland-26000146. Accessed February 5, 2014. And Gregory Campbell, the DUP MP for East Londonderry, stated, “I will use whatever device I have to use to expose [Sinn Féin’s] duplicity, and their politicising of the Irish language.” Jennifer O’Leary, “Why Is Irish Language Divisive Issue in Northern Ireland?,” BBC News Online, December 17, 2014, available at: http:// www.bbc.com/news/uk-northern-ireland-30517834. Accessed December 17, 2014. Again, Gerry Adams has made statements which support the loyalist claim. For example, he wrote: “The revival of the Irish language as the badge of identity, as a component part of our culture and as the filter through which it is exposed, is a central aspect of the reconquest.” Gerry Adams, Free Ireland: Towards a Lasting Peace (New York: Penguin, 1995), p. 139, quoted in Hugh F. Kearney, Ireland: Contested Ideas of Nationalism and History (New York: NYU Press, 2007), p. 128. Ironically, during the nineteenth century, the Irish language was preserved by Presbyterians.

58 Interview with Robert, November 28, 2012. Relatedly, there is a widespread sentiment amongst my interviewees that the media is out to get Protestants, in general, and parades, in particular.

192 we’ve always done, culture.

In bemoaning the politicization of parades, participants also yearn for the lost innocence of the “good old days,” a mythic past of ethnic harmony under unionist rule. The contrast between the friendly parades of the halcyon, pre-political era and today’s anger and hyper- contestation could not be clearer. Scott told me that when unionists controlled the country, they could parade anywhere. During those years, there were no such thing as contentious parades.

Isaac recalls the era when “the RUC [Royal Ulster Constabulary, an all-Protestant police force] policed all those parades and there never was really any trouble.” But these statements inadvertently reveal the truth. The earlier era was not pre-political: parades were always tied up with politics and political power. It’s just that respondents used to support the politics, since it permitted them, and in fact helped them, to parade wherever they liked. Indeed, the very reason that parading is so much more prominent among Protestants than Catholics is that the state promoted Protestant parades while using expansive legislation to suppress Catholic ones.59

Parades “have always been closely associated with Protestant political ascendancy,”60 but the zenith of their power was during the five decades that unionists dominated the Northern

Ireland Parliament in Stormont (1921-1972). During the Stormont years, Protestant were politically supreme, and at the center of their hegemony was the Orange Order. Northern Ireland was “the Orange state” and loyalist parades were “rituals of state.”61 The general mood was

59 Bryan, Orange Parades, p. 61; and Jarman, Material Conflicts, p. 72. One of the laws used to restrict Catholic parades was the Civil Authorities (Special Powers) Act of 1922, about which an Apartheid-era South African Justice Minister said that he would “trade all the coercive powers at his disposal ‘for one clause of the Northern Ireland Special Powers Act.’” Bourke, Peace in Ireland, p. 46.

60 Smithey, Unionists, Loyalists, and Conflict Transformation, p. 116

61 Michael Farrell, Northern Ireland: The Orange State (London: Pluto Press, 1976); and Bryan, Orange Parades, esp. chap. 5, “Rituals of State.”

193 captured in a well-known statement by James Craig, Northern Ireland’s first Prime Minister: “I have always said I am an Orangeman first and a politician and Member of this Parliament afterwards. … All I boast of is that we are a Protestant Parliament and a Protestant State.”62 The

Orange Order, along with the other institutions of unionist hegemony, actively worked to exclude

Catholics from political power and labor markets and to maintain Catholics’ “degraded form of citizenship.”63 A leading Northern Irish human rights organization, points out the bitter irony that while the loyal orders extol and exercise their own rights and liberties, their “political involvement… has often served to deny civil, religious and political liberties to others.”64 And though loyalist parades were generally used to symbolize and display Protestants’ privileged position, at times they were creatively used to maintain it. During the late 1960s, for example, loyal orders would organize “annual parades” for the same time and location as planned civil rights demonstrations, forcing the government to prohibit both marches.65

This history, however, is conveniently ignored. As I discussed previously, participants’ general understanding of the historical shift from Catholic quiescence to protest is that Catholics’ attitudes changed, not that the balance of power in government changed. This is reflected in the heartfelt desire expressed by several participants that one day Catholics would come to love parades as much as Protestants do. As Walter says: “I wish they could feel the passion that we

62 Parliamentary Debates, Northern Ireland, 1933/34, Vol. 16, Column 1091 and 1095, April 24, 1934. The Hansard record is fully digitized and available online at: http://stormontpapers.ahds.ac.uk/.

63 Bourke, Peace in Ireland, p. 11.

64 Pat Finucane Centre, For God and Ulster: An Alternative Guide to the Loyal Orders (Derry: Pat Finucane Centre, 1997), available online at: http://www.patfinucanecentre.org/archive/loyal.html.

65 Jarman, Material Conflicts, 76-7; Tilly, Politics of Collective Violence, p. 125. Also Eric Kaufmann, “Demographic Change and Conflict in Northern Ireland: Reconciling Qualitative and Quantitative Evidence,” Ethnopolitics, Vol. 10, Nos. 3-4 (September-November 2011), pp. 381-2.

194 feel.” Howie states:

I would love the day, and I think it probably does happen, but it’s not being reported—I want to see the day when the Catholic community can come out and enjoy it. Enjoy the spectacle of it and enjoy the music of the bands. That’s the way it used to be whenever I was growing up… We would have traveled into Belfast, whenever we were young teenagers, and them guys [Catholic friends] would !come with us to watch the parade. If only the future would be like the past, everything will return to normal. For many paraders, the solution to current problems surrounding parades is for Catholics to accept, or even embrace, loyalist parades, just like they did in their memories and interpretations of the past.

There is a second aspect to the language of anti-politics and the logic of de-politicization: there is power in anti-politics. This power is premised on the idea that culture transcends politics and is therefore exempt from the practices of democracy, such as critique, debate, and compromise. Culture and cultural practices, as anti-political phenomena, are simply beyond the reach of democratic processes and procedures. The discourses of culture and tradition are thereby a free pass to act in ways that would not normally be acceptable in society. But, as traditional cultural practices, parades simply have to be accepted.

To show how the discourse of anti-politics is used to silence criticism and dismiss compromise, I will focus on two of the most illuminating things said to me in all my interviews.

The first is from Alexander, a senior member of the Orange Order and retired unionist politician.

Sitting in his comfortable living room, sipping milky tea his wife had served us earlier, I asked him, “Are parades at all political?” Alexander replied:

I don’t think that they seem to be political. Some can be. I mean, there have been political processions in Belfast, but the Orange parade is not a political parade as such. The speeches designated by the Grand Lodge, the resolutions to which the speakers are asked to speak as to loyalty, as citizens to the United Kingdom. That

195 could be perceived as political if you don’t like the United Kingdom. [My !emphasis.] In his response, Alexander first distinguishes Orange parades from political parades. Then, he states that that the resolutions proclaimed at parades are not political, rather they are merely about citizenship and loyalty to the state. His last quoted sentence, however, is the most illuminating: “That could be perceived as political if you don’t like the United Kingdom.” Seeing politics where there is not any is a pathology caused by hatred of the union. It has nothing to do with the parades themselves. In other words, we need not take these people seriously, because they are not serious people. By calling parades political, this Orange doyen argues, the opponents of parades show their true cards.

The second remark comes from Rachel, the university-educated, female band member from East Belfast. I already quoted a piece of it earlier, but reproduce the entire segment here:

They’re tarnished as being political because the other side would see that as being political, but when you’re in them with your other friends, they’re not political, they’re cultural. One thing that keeps getting told off is that… this is a political problem. It’s not a political problem, it’s a cultural problem. But it’s ingrained in the minds of people that it’s a political problem because Sinn Féin always puts it down as being over politics. I don’t see— You know, fair enough, people see Ardoyne as being a political tension point because who’s the first people to come out and stand by? It’s politicians. But it’s cultural, they’re not respective of our !culture. [My emphasis.] She begins by contrasting the cultural experience of parades with the negative and false perception of parades as political. Therefore, the problems surrounding parades are cultural problems, not political problems. And then she identifies the real problem: “they’re not respective of our culture.” The crux of the issue is that republicans do not like Protestants, end of story. The implication being that no political solution is possible. Dialogue will not fix the

196 problem, reason will not fix the problem, compromise will not fix the problem. These are political hammers that simply cannot bend this cultural nail. The logic of anti-politics leads to no other conclusion. Opponents cannot have a legitimate problem with parades because as cultural events there is nothing to object to, unless you object to the entire premise of Protestants in your presence. This is made clear in a large banner I saw carried by a group of women in the 2013

July Twelfth parade and displayed by parade supporters in Ardoyne: “END HATRED OF

ORANGE CULTURE.” The message is clear: opposition to parades is caused by and reflects hatred of Protestant culture.

The discourse of tradition and cultural anti-politics does a lot of work for participants. It justifies their actions and delegitimizes opposition. It explains why other people object to them and simultaneously dismisses their objections from the agenda, thereby shaping the political arena. Acknowledging the politics embedded in parades opens the door to a number of distasteful outcomes, including a tarnished self-conception and compromising a cherished tradition.66 The platform of anti-politics, conversely, keeps the door tightly sealed. Therein lies the power of anti- politics. !

The Ritual Foundations of the Paradox of Anti-Politics

Nevertheless, how is it possible that participants maintain the paradox of anti-politics?

Strong personal and political incentives to sustain it notwithstanding, how can they determinedly resist understanding parades as political actions given the political claims and consequences of

66 Further, Ross, Cultural Contestation, p. 14, notes that “each side deeply fears that recognizing the claims of the other invalidates their own.” Jennifer Todd, “Two Traditions in Unionist Political Culture,” Irish Political Studies, Vol. 2, No. 1 (1987), p. 19, argues that this trait is especially acute among unionists.

197 parades that many of them all but outright acknowledge? In this section, I argue that the foundation of the paradox is the ritual nature of parades. As rituals, parades provide participants with alternative (and genuinely apolitical) reasons to participate as well as with symbolic ambiguity that supports multiple and conflicting interpretations of the events. Further, rituals nurture the idea that within them, the normal connection between means and ends is broken.

Finally, the repetitive and formal qualities of rituals suffuse parades with a feeling of invariance, feeding the idea that parades never were and are not now political. This impression of continuity with the past is also a source of legitimacy in contemporary society.67 These four features of ritual sustain the paradox of anti-politics, against evidence to the contrary. And without the claim to anti-politics, parades could not take advantage of the defense mechanisms enumerated above.

Thus, the intractability of the conflict over parades—and all of the hostility and violence that has resulted from it—is supported by parades’ ritual nature.

Parades, like all rituals, provide process-oriented motives to participate. The achievement of process-oriented benefits, such as emotional energy and a sense of belonging, is unrelated to the consequences of the action. Therefore, participants’ own understanding of why they participate has nothing to do with the political effects of their actions, such as ethnic polarization, damage to the peace process, protests, and sometimes violence. For participants, these political outcomes might be an unwelcome, tolerated, or happily welcomed, but in my interviews I found no evidence that they are actively sought. The political consequences are a byproduct of parades, not a motivation to act. Their motivations are the emotions, solidarities, and opportunities for self-expression that are intrinsic in the very act of taking part in a symbol-laden, traditional ritual

67 Hobsbawm and Ranger, eds., The Invention of Tradition; and Paul Connerton, How Societies Remember (New York: Cambridge University Press, 1989).

198 celebration of the nation along with family, friends, and members of the imagined community.

Thus there is a disconnect between the individual-level causes and the social functions and consequences of parades.

Rituals, secondly, provide participants with politically useful ambiguity. As Anthony

Cohen’s discussion of symbols makes clear, even “commonly accepted symbol[s]” hold a “range of meanings.”68 This ambiguity is, in fact, the key to their utility: “Symbols are effective because they are imprecise.”69 The symbolic ambiguity of parades means that Protestants and Catholics can hold different, even contradictory meanings of parades. Paraders and protesters can read parades in opposite ways, and each can still believe themselves entirely correct—and indeed each can be correct.70 This view of ritual stands in contrast to that of Durkheim and the neo-

Durkheimians.71 They view society as coherent and rituals as a major source of social integration. In this model, there is little room for incompatible interpretations of a ritual. But as

Lukes points out in his critique of the neo-Durkheimians, in societies such as Northern Ireland

“value consensus is manifestly absent or minimal.”72 In such divided societies, Lukes—writing

68 Anthony P. Cohen, The Symbolic Construction of Community (London: Routledge, 1985), p. 15. Original emphasis.

69 Ibid., p. 21.

70 This multivocality and ambiguity of rituals can be exploited by elites to use them for controversial purposes while maintaining that they are simply supporting a legitimate cultural practice with no negative connotations. Beyond giving them political cover, it makes it hard for the government “to ban the event, for who could possibly object to the performance of a religious obligation, the raising of the national flag, or the celebration of a national day?” As Wilkinson notes, most states “institutionally privilege[] some forms of mobilization—and in particular, ‘traditional’ religious ceremonies and processions—over others.” Steven I. Wilkinson, Votes and Violence: Electoral Competition and Ethnic Riots in India (New York: Cambridge University Press, 2004), pp. 23-24.

71 Émile Durkheim, The Elementary Forms of Religious Life, Karen E. Fields, trans. (New York: Free Press, 1995 [1912]); and, for example, Edward Shils and Michael Young, “The Meaning of the Coronation,” Sociological Review, Vol. 1, No. 2 (December 1953), pp. 63-81; Robert N. Bellah, “Civil Religion in America,” Daedalus, Vol. 96, No. 1 (Winter 1967), pp. 1-21.

72 Steven Lukes, “Political Ritual and Social Integration,” Sociology, Vol. 9 (May 1975), p. 300.

199 specifically about loyalist parades—argues, “Ritual here exacerbates social conflict and works against (some aspects of) social integration.”73 Thus, given that Cohen argues that even participants in a ritual need not share its meaning,74 it is entirely logical that people on opposite sides of Northern Ireland’s sectarian division do not share parades’ meaning. Ambiguity means that just because many Catholics believe that parades are sectarian and hateful does not mean that Protestants see them the same way.75

The ritual nature of parades also supports participants’ understanding of their politicization. Recall the narrative that paraders tell to explain current controversies: parades had been peaceful for centuries until the republican movement decided to whip up opposition during the Troubles. This story has the effect of exculpating parades and deflecting blame for the unrest of the last two decades. But it is buttressed by certain features of ritual. Since parades and other rituals project a sense of continuity with the past,76 they imply that any change could not have been internal; rather, it must have come from outside the parading community.

The paradox of anti-politics is further made possible by the way that ritual suggests a hazy connection between means and ends. Rituals are non-instrumental acts, they entail “a qualitative departure from the normal intentional character” of action.77 As Connerton writes, rituals are “expressive acts rather than instrumental acts, in the sense that they are either not

73 Ibid.

74 Cohen, Symbolic Construction of Community, p. 55.

75 Ross, Cultural Contestation, p. 63, emphasizes the role of “divergent group psychocultural narratives” in producing such contradictory interpretations of events.

76 Connerton, How Societies Remember, p. 45, argues that commemorative rituals, such as many loyalist parades, “do not simply imply continuity with the past but explicitly claim such continuity.”

77 Caroline Humphrey and James Laidlaw, The Archetypal Actions of Ritual: A Theory of Ritual Illustrated by Jain Rite of Worship (Oxford: Claredon Press, 1994), p. 89.

200 directed to a strategic end, or if they are so directed, as with fertility rites, they fail to achieve their strategic aim.”78 Rituals have “no obvious empirical goals”:

There is no clear point in walking the ox round the dancers—or, for that matter, the dancers around the scene. More generally, in rituals one typically washes instruments that are already clean, one enters rooms to exit them straightaway, one talks to interlocutors that are manifestly absent, and so forth. … True, a given ritual generally has a specific purpose (e.g., healing a particular person) but the set of sequences that compose the ritual are not connected to this goal in the same 79 !way as subactions connect to subgoals in ordinary behavior. We can see the traces of this ruptured means and ends in the interviews. For example, when

Walter says, “People in the band, yes, they want to keep the union, but they wouldn’t use the band as a means towards an end.”

Politics, in contrast, is fundamentally instrumental. Politics is the pursuit of power, and political action is toward that end. Pursuing “‘means’ for the attainment of the actor’s own rationally pursued and calculated ends” is Weber’s very definition of instrumentally rational social action.80 By portraying itself as non-instrumental, ritual thus propels the appearance of being anti-political. Ritual stands against politics, each operating with a different logic. So because parades are rituals, they cannot be political.

Together all of these features make possible and validate participants’ experience of parades as apolitical. As Rachel makes clear, “When you’re in them with your other friends, they’re not political, they’re cultural.” Parades do not feel political, they feel cultural, traditional,

78 Connerton, How Societies Remember, p. 44.

79 Pierre Lienard and Pascal Boyer, “Whence Collective Rituals? A Cultural Selection Model of Ritualized Behavior,” American Anthropologist, Vol. 108, No 4 (December 2006), p. 816.

80 Max Weber, Economy and Society: An Outline of Interpretive Sociology, Guenther Roth and Claus Wittich, trans. and eds. (Berkeley: University of California Press, 1978), p. 24.

201 and ritualistic—the opposite of political. Therefore the complaints by others that parades are sectarian, intimidating, triumphalist, or provocative—i.e., political—do not resonate with participants’ own experience.

Paraders, for example, understand that Catholics think that parades are about confronting

Catholics, but their own experience tells them that this is not true. As Ben says: “The Catholic community has got it in its head [that] it’s all about them: this is us stating our dominance over them.… [But, it’s not.] It’s just basically us being us. It’s as simple as that.” Ben’s language inadvertently echoes a distinction that Schuessler makes between instrumental and non- instrumental behavior. Instrumental behavior, he argues, is characterized by “Doing—individuals perform X in order to do Y”—whereas non-instrumental, expressive behavior is characterized by

“identification, attachment, or Being—individuals perform X as this is how they become X- performers.”81 Ben thus unwittingly illustrates the distinction by contrasting “us stating our dominance over them”—quintessential Doing—with “us being us.”

One element of the parading experience emphasized by several interviewees is that when they are marching, Catholics do not even cross their mind—except for the minutes it takes to walk past any protesters. Mikey, for instance, describes how he sees it:

You’re out with your friends… It’s a social thing. It’s not us going, “Yea, we’re going to annoy the Catholics! Yo! We’re all Protestants! Yo!” It’s not like that. … The parade is like a gathering of friends. … There’s this perception [among] nationalists and republicans—I don’t know whether it’s Catholics in general—… that I’m getting up in the morning and I got to bed on a Thursday night and wake up Saturday morning a sectarian bigot, and I’m going down to annoy Catholics. You don’t care about that. Nobody cares about their religion. I don’t care about them, I’m not thinking of anything. I’m not even thinking of the Protestant religion. I’m getting up and I know I’m going out… with my [Orange] lodge, with

81 Alexander A. Schuessler, A Logic of Expressive Choice (Princeton: Princeton University Press, 2000), p. 30.

202 !all my friends. For Mikey, the contrast between the perception of parades and the actual experience of parades is stark. Like Ben, he is fully aware of the Catholic interpretation of parades, but has never observed it in his own involvement. Parades are an enjoyable time spent with friends; Catholics do not enter the equation. Thus, to paraders like Ben and Mikey, parades do not feel in anyway political.82

Of course, ritual is not the only way to create or maintain a politics of anti-politics. The human rights movement, for instance, has generated an ideological (and perhaps pragmatic) anti- politics premised on a “moral discourse centered on pain and suffering.”83 Political theorist

Wendy Brown writes that human rights:

generally presents itself as something of an antipolitics—a pure defense of the innocent and the powerless against power, a pure defense of the individual against immense and potentially cruel or despotic machineries of culture, state, war, ethnic conflict, tribalism, patriarchy, and other mobilizations or instantiations of 84 !collective power against individuals. It is human rights’ very anti-politics, she argues, that allows it to pursue political projects (such as “liberal imperialism”) unquestioned. Historian Samuel Moyn also finds that the human rights movement and its predecessors mobilized the language of anti-politics to its advantage. For example, in discussing Warsaw Pact dissidents of the 1970s, he writes: “In reality, of course, the movement ‘was political in the sense that it threatened the foundations of Soviet power.’ But it

82 See also Ramsey, Music, Emotion and Identity, p. 163.

83 Wendy Brown, “‘The Most We Can Hope For…’: Human Rights and the Politics of Fatalism,” South Atlantic Quarterly, Vol. 103, No. 2/3 (Spring/Summer 2004), p. 453. On the anti-politics of human rights’ cousin, humanitarianism, see Michael Barnett, Empire of Humanity: A History of Humanitarianism (Ithaca: Cornell University Press, 2011), esp. pp. 212-219.

84 Ibid.

203 was based on a politics that worked precisely by claiming to transcend politics.”85

The moralism of human rights that Brown and Moyn identify as anti-political has been similarly observed in other social movements. Deborah Gould identifies and theorizes the role of moralism in the fracturing of the direct-action AIDS group ACT UP. She finds that moralism can supplant real political dialogue, since moralizing claims question the very essence of your opponent. Therefore, covering an issue with moralizing rhetoric removes it from the possibility of political debate, persuasion, bargains, or compromise.86

In The Anti-Politics Machine, James Ferguson argues that economic development officials in Lesotho cultivated an anti-politics based on scientific authority. He finds that the anti- political, technocratic rhetoric of economic development is a source of power for the development industry. He concludes that “the ‘development’ apparatus” is an “anti-politics machine” whose trick is “the suspension of politics from even the most sensitive political operations.” By using its variation on Midas’s Touch, “a ‘development’ project can end up performing extremely sensitive political operations involving the entrenchment and expansion of institutional state power almost invisibly, under the cover of a neutral, technical mission to which no one can object.”87

85 Samuel Moyn, The Last Utopia: Human Rights in History (Cambridge: Harvard University Press, 2010), pp. 136-7. He is quoting Philip Boobbyer, Conscience, Dissent and Reform in Soviet Russia (New York: Routledge, 2005), p. 89. Looking specifically at the Polish opposition, David Ost, Solidarity and the Politics of Anti-Politics: Opposition and Reform in Poland since 1968 (Philadelphia: Temple University Press, 1990), pp. 16-17, argues that not only did they need to reject state-centered politics, they wanted to abandon state-centered politics in favor of civil society. “Anti-politics… is not a negation of politics,” he writes, “but a relocation of the political public from state to society. It is this ‘anti-political’ project that is crucial to understanding Solidarity’s practice.”

86 Deborah B. Gould, Moving Politics: Emotion and ACT UP’s Fight against AIDS (Chicago: University of Chicago Press, 2009), pp. 378-92.

87 James Ferguson, The Anti-Politics Machine: “Development,” Depoliticization, and Bureaucratic Power in Lesotho (Minneapolis: University of Minnesota Press, 1990), p. 256.

204 Presenting an issue as anti-political operates in all three approaches to power identified by scholars.88 Within Dahl’s one-dimensional approach to power, the discourse of anti-politics can be used to influence decision-makers.89 It is a resource to be wielded by A in her bargaining with B to resolve a conflict over issue J. Within the two-dimensional approach to power elaborated by Bachrach and Baratz, claiming an issue as anti-political is a way to get that issue off the agenda in the first place.90 The second face of power is where this discourse’s influence really takes shape. There may still be a conflict between A and B over issue J, but J is never put up for debate since it is not a political issue and therefore B is prevented from raising the issue in political fora. Within the three-dimensional approach to power theorized by Lukes, the discourse of anti-politics affects B so profoundly that she does not even consider J, which is objectively detrimental to her, to be a problem.91 Here too the power of anti-politics can be great. B cannot even get to the point of conceiving that J is detrimental because she is convinced that J exists outside the realm of political conflict.

This source of power is not lost on many groups and causes who try to use it to their advantage in the defense of culture and traditional practices. The rhetoric of anti-politics is commonly used in an attempt to influence political outcomes related to cultural issues. For example, historian David Hollinger argues that Americans tend to “give religious ideas a pass,” meaning that they follow the “convention of protecting religious ideas from the same kind of

88 My understanding was greatly improved by the discussion in John Gaventa, Power and Powerlessness: Quiescence and Rebellion in an Appalachian Valley (Champaign: University of Illinois Press, 1980), pp. 5-20.

89 Robert A. Dahl, “The Concept of Power,” Behavioral Science, Vol. 2, No. 3 (July 1957), pp. 201-15.

90 Peter Bachrach and Morton S. Baratz, “Two Faces of Power,” American Political Science Review, Vol. 56, No. 4 (December 1962), pp. 947-52.

91 Steven Lukes, Power: A Radical View (London: Macmillan, 1974).

205 critical scrutiny to which we commonly subject ideas about almost everything else.” This convention is not only founded on the “virtues of decency and humility,” but also a

“constitutional tradition that does indeed treat religious ideas as a distinct category…. [And] a history of religious diversity that renders silence a good way to keep the peace.”92 The result of this convention is that American politicians frequently justify their positions on religious grounds and their critics cannot question their reasoning. Religion is a “conversation-stopper,” as Richard

Rorty suggests.93 Proponents of controversial practices such as female genital cutting and flying the Confederate flag in the American south appeal to culture and tradition to defend their actions.

By shifting the debate to culture and tradition, these actors as arguing that society needs to give them a pass. By appealing to culture and tradition, they are moving the debate away from politics.

We see the same logic at work with loyalist parades. If parades are political, then they are open to normal democratic politics and processes. Compromise becomes a necessary part of doing business. Accepting restraints and restrictions on parades becomes inevitable. And, possibly worst of all, to be political is to acknowledge the legitimacy of nationalist opposition. It means taking their objections seriously, which in turn requires a hard look in the mirror and the possibility of seeing something unexpected and undesirable. The distinction between “good, cultural us” and “bad, political them” disappears, and the moral landscape is suddenly level.

This, then, is the power provided by the anti-politics of culture and tradition. It protects parades

92 David A. Hollinger, “Religious Ideas: Should They Be Critically Engaged or Given a Pass?” Representations, Vol. 101, No. 1 (Winter 2008), pp. 145-6.

93 Richard Rorty, “Religion as Conversation-Stopper,” Common Knowledge, Vol. 3, No. 1 (Spring 1994), pp. 1-6; cited in Hollinger, “Religious Ideas,” p. 145.

206 from the inevitable, distasteful compromises inherent in democratic politics; and it protects participants’ moral vision of the world. The source of that power is the ritual nature of parades. !

Conclusion

“When,” ask Taylor, Rupp, and Gamson, “is cultural performance a form of protest?”94

Their answer is that three factors “distinguish between events staged purely for entertainment and those staged for political ends”: contestation, intentionality, and collective identity.95 That is, the event must contest the dominant order, the actors must intend this contestation, and the actors must hold a collective identity.96 This model is a useful corrective that helps us expand the boundaries of what we consider political action. It shows that events that look “merely cultural” from the outside, can be political on the inside. But this chapter has shown that in loyalist parades, the opposite is the case. From the outside, parades look deeply political, but on the inside, participants see them as “merely cultural.” Participants lack intentional political contestation.

Political rituals, Pfaff and Yang argue, are double-edged because they can be used to

94 Verta Taylor, Leila J. Rupp, and Joshua Gamson, “Performing Protest: Drag Shows as Tactical Repertoires of the Gay and Lesbian Movement,” Research in Social Movements, Conflicts, and Change, Vol. 25 (2004), p. 107.

95 Ibid., p. 108.

96 Also Leila J. Rupp and Verta Taylor, Drag Queens at the 801 Cabaret (Chicago: University of Chicago Press, 2003); Verta Taylor and Nella Van Dyke, “‘Get up, Stand up’: Tactical Repertoires of Social Movements,” in David A. Snow, Sarah A. Soule, and Hanspeter Kriesi, eds., The Blackwell Companion to Social Movements (Malden, MA: Blackwell, 2004), pp. 262-293; and Verta Taylor, Katrina Kimport, Nella Van Dyke, and Ellen Ann Andersen, “Culture and Mobilization: Tactical Repertoires, Same-Sex Weddings, and the Impact on Gay Activism,” American Sociological Review, Vol. 74, No. 6 (December 2009), pp. 865-890.

207 support the regime or the opposition.97 This has long been true of loyalist parades, but this chapter highlights a second way in which political rituals have a double-edged character. One on hand, as Taylor and coauthors demonstrate, rituals and other cultural performances can make political claims. But on the other hand, as I showed in this chapter, rituals and other cultural performances can mask political claims. The ritual character of parades allows participants to disavow their political claims and consequences. It allows them to make a political action apolitical, even anti-political. As rituals, parades provide participants with process-oriented reasons to act and symbolic ambiguity that muddles their meaning. This does not diminish the political quality of loyalist parades. Events can be unintentionally political,98 particularly when political opponents interpret them so—as is clearly the case here. But it does push us to clarify how culture is used as a political strategy. By suggesting that cultural performances are can be useful to promote political claims, Taylor and colleagues reveal one edge of the sword. By suggesting that cultural performances can be useful to hide political claims, I uncover the other.

97 Steven Pfaff and Guobin Yang, “Double-Edged Rituals and the Symbolic Resources of Collective Action: Political Commemorations and the Mobilization of Protest in 1989,” Theory and Society, Vol. 30, No. 4 (August 2001), pp. 539-589.

98 Olivier Fillieule, “The Independent Psychological Effects of Participation in Demonstrations,” Mobilization, Vol. 17, No. 3 (September 2012), p. 236; and Rupp and Taylor, Drag Queens at the 801 Cabaret, p. 218.

208 Chapter Seven For God and Ulster or Private Payoff? ! Assessing the Role of Collective and Selective Incentives

This chapter assesses the evidence for plausible alternative explanations for loyalist parade participation: that participation follows the instrumental logics of ethnic rivalry or collective action. In order to rigorously test these arguments, I rely on more evidence that I have in the previous two chapters. In particular, I draw on data from parade participants and comparable nonparticipants so that I can make claims about systematic differences between the two groups. Sampling negative cases is a standard method in modern social science that ensures variation in the dependent variable,1 but existing studies of loyalist parades and other “cultural forms of political expression”2 have only collected data from participants, the positive cases.

This methodological shortcoming limits the claims they can make. In research on loyalist parades, both ethnographic3 and quantitative4 scholars have failed to sample comparable

1 Gary King, Robert O. Keohane, and Sidney Verba, Designing Social Inquiry: Scientific Inference in Qualitative Research (Princeton: Princeton University Press, 1994), pp. 129-37.

2 Verta Taylor, Leila J. Rupp, and Joshua Gamson, “Performing Protest: Drag Shows as Tactical Repertoires of the Gay and Lesbian Movement,” Research in Social Movements, Conflicts, and Change, Vol. 25 (2004), p. 106.

3 Neil Jarman, Material Conflicts: Parades and Visual Displays in Northern Ireland (Oxford: Berg, 1997); Dominic Bryan, Orange Parades: The Politics of Ritual, Tradition, and Control (London: Pluto, 2000); and Gordon Ramsey, Music, Emotion and Identity in Ulster Marching Bands: Flutes, Drums and Loyal Sons (Bern: Peter Lang, 2011).

4 The most systematic quantitative study is James W. McAuley, Jonathan Tonge, and Andrew Mycock, Loyal to the Core?: Orangeism and Britishness in Northern Ireland (Dublin: Irish Academic Press, 2011), but even they only sample members of the Orange Order.

209 nonparticipants. The same is true for research on contentious cultural events in general.5 Even one the most systematic and methodologically sophisticated studies of participants in cultural contentious politics, Verta Taylor et al.’s research on same-sex weddings in San Francisco, is limited by this sampling issue. They “conducted a random survey of all participants in the San

Francisco weddings,” but acknowledge that a “sample of nonparticipant gays and lesbians would be virtually impossible to obtain.”6 I address this by exploiting Belfast’s sectarian housing segregation to construct a random sample of participants and comparable nonparticipants.

Thus, the rest of this chapter overcomes previous methodological limitations by using quantitative and qualitative data collected from participants and nonparticipants. In the next section, I explain how I constructed a random sample to survey. In short, I assume that every person from a Protestant background is a potential parade participant and therefore take a random sample of households in Protestant neighborhoods. Belfast’s severe sectarian segregation makes it possible to generate a pool of potential paraders of whom some choose to parade. I can

5 Taylor, Rupp, and Gamson, “Performing Protest”; Leila J. Rupp and Verta Taylor, Drag Queens at the 801 Cabaret (Chicago: University of Chicago Press, 2003); and Suzanne Staggenborg and Amy Lang, “Culture and Ritual in the Montreal Women’s Movement,” Social Movement Studies, Vol. 6, No. 2 (September 2007), pp. 177-194. Despite increasing methodological sophistication, selecting on the dependent variable is a general problem in research on participation in contentious politics. For important exceptions, see Doug McAdam, “Recruitment to High-Risk Activism: The Case of Freedom Summer,” American Journal of Sociology, Vol. 92, No. 1 (July 1986), pp. 64-90 Bert Klandermans and Dirk Oegema, “Potentials, Networks, Motivations, and Barriers: Steps Towards Participation in Social Movements,” American Sociological Review, Vol. 52, No. 4 (August 1987), pp. 519-531; Doug McAdam and Ronnelle Paulsen, “Specifying the Relationship between Social Ties and Activism,” American Journal of Sociology, Vol. 99, No. 3 (November 1993), pp. 640-667; Sharon Erickson Nepstad and Christian Smith, “Rethinking Recruitment to High-Risk/High-Cost Activism: The Case of the Nicaragua Exchange,” Mobilization, Vol. 4, No. 1 (April 1999): 25-40; and Alan Schussman and Sarah A. Soule, “Process and Protest: Accounting for Individual Protest Participation,” Social Forces, Vol. 84, No. 2 (December 2005), 1083-1108. Macro-level studies that focus on the event, rather than the individual, also fail to sample comparable non-events. See Steven I. Wilkinson, Votes and Violence: Electoral Competition and Ethnic Riots in India (New York: Cambridge University Press, 2004), pp. 38-39; though Wilkinson tries to address the issue in Steven I. Wilkinson, “Which Group Identities Lead to Most Violence? Evidence from India,” in Stathis N. Kalyvas, Ian Shapiro, and Tarek Masoud, eds., Order, Conflict, and Violence (New York: Cambridge University Press, 2008), pp. 271-300.

6 Verta Taylor, Katrina Kimport, Nella Van Dyke, and Ellen Ann Andersen, “Culture and Mobilization: Tactical Repertoires, Same-Sex Weddings, and the Impact on Gay Activism,” American Sociological Review, Vol. 74, No. 6 (December 2009), pp. 872, 873 n. 6.

210 therefore compare those who parade to their neighbors who do not. I then present the texture of the quantitative data, reporting descriptive statistics about paraders, parade attenders, and parade non-attenders. Third, I test two prominent existing theories of participation, ethnic rivalry and collective action. Using the survey data, supplemented by qualitative data, I show that, contrary to the predictions of the ethnic rivalry approach, participants are neither more pro-Protestant nor more anti-Catholic than nonparticipants, and, contrary to the predictions of collective action approach, participants do not receive material benefits nor do pre-existing social ties increase the likelihood of participation. The quantitative data do provide some support for the collective action approach’s expectation of social sanctioning, but the qualitative data are not consistent. !

The Survey

To collect individual-level data, I conducted a randomized survey in nine Protestant neighborhoods in Belfast. The interviews were conducted face-to-face by local interviewers from late May to mid-August 2013. In total, 228 valid surveys were collected. In this section, I describe the survey methodology in detail, explaining the choices I made and constraints I faced. !

Selecting Neighborhoods

The purpose of the survey was to determine who participates in parades and who does not participate in parades. Since loyalist parades are done by the Protestant community, I needed neighborhoods that are predominantly Protestant. A nationally representative sample would include many Catholics who are not potential participants and therefore not a comparable group of nonparticipants. Due to high levels of sectarian housing segregation, finding Protestant

211 neighborhoods is easy.7 I used the 2011 Northern Irish census to restrict my search to census wards with a Protestant majority.8 Secondly, I wanted to maximize the likelihood of gaining parade participants in the sample, so I needed neighborhoods with high concentrations of participants. I based these decisions on conversations with academic experts and leaders of parading organizations, as well as my own observations. Given the working class dominance of

Table 7.1. Information on the Neighborhoods Included in the Survey Comparing: Comparing: % Employed Census information % British full-time Usually Sample resident House % (Source: population -holds Protestant Survey) Survey Census Survey Census East Belfast 1 1596 652 82.1% 39 65.8% 75.9% 53.9% 48.9% East Belfast 2 1538 625 87.1% 34 67.7% 80.8% 44.1% 55.2% East Belfast 3 1401 655 67.5% 13 61.5% 65.5% 16.7% 38.0% South Belfast 1 933 485 70.4% 13 77.8% 70.1% 16.7% 33.6% South Belfast 2 3425 1770 62.0% 34 93.9% 60.0% 29.4% 40.9% South Belfast 3 809 394 78.9% 50 89.8% 73.8% 44.0% 37.6% West Belfast 1 2574 1126 85.0% 26 96.2% 80.9% 34.8% 40.0% West Belfast 2 1127 594 89.7% 10 90.0% 85.0% 11.1% 26.7% West Belfast 3 472 244 87.3% 9 100.0% 77.1% 33.3% 33.9% Note: The census data are the sums of the Small Areas which I sampled in a neighborhood, and not necessarily the entire geographic area which locals consider their neighborhood. Also note that the percentages of British and full-time employment from the census are for the entire population of the area, while my survey only sampled Protestants. We would expect, therefore, that the percent identifying as British should be higher in the survey than the census, which it is in six of the nine areas. Sources: 2011 Northern Ireland Census. Usually resident population and Households: Households (Statistical Geographies); % Protestant: Religion or Religion Brought Up In: KS212NI (statistical geographies); % British National Identity (Classification 2): KS203NI (statistical geographies); % full-time: Economic Activity - Males: KS602NI (statistical geographies).

7 Housing segregation means that almost every neighborhood in Belfast is considered either Protestant or Catholic. See Peter Shirlow and Brendan Murtagh, Belfast: Segregation, Violence, and the City (London: Pluto, 2006); Brendan Murtagh, “Ethno-Religious Segregation in Post-Conflict Belfast,” Built Environment, Vol. 37, No. 2 (2011), pp. 213-225; and Christopher D. Lloyd and Ian Shuttleworth, “Residential Segregation in Northen Ireland 2001: Assessing the Value of Exploring Spatial Variation,” Environment and Planning A, Vol. 44, No. 1 (2012), pp. 52-67.

8 Of the 40 Small Areas where the survey took place, only one had less than 50 percent Protestant population (it had 48.7 percent). Thirty-one were over 75 percent Protestant. All of the neighborhoods would be considered Protestant, Unionist, or Loyalist by locals.

212 contemporary parading, the survey mainly took place in the Victorian- and Edwardian-era red- brick terrace houses and the mid-twentieth century public housing estates. However, in order to gain a more representative sample, I also included two middle-class Protestant neighborhoods, neither of which was known to be a hotbed of parading.

The reality of working-class neighborhoods in Belfast meant that I could not simply pick areas that matched the criteria I was looking for and begin interviewing. The vast housing estates of Northern Ireland are generally controlled by the UDA or UVF paramilitaries and working in these areas meant gaining the consent from the local gatekeepers. In some neighborhoods, I first approached the gatekeepers and then assembled a team of interviewers, in others it was reversed.

Working with the gatekeepers caused both ethical and methodological dilemmas. On the one hand, I needed their approval to conduct the survey in their area. But this raised the ethical concern that my research teams could be viewed as coercive. I could not accept that my interviewers might knock on a door and have the resident feel that they had to answer “or else.”

This also raised a methodological concern in that respondents might feel that they need to give the answer that the paramilitaries want to hear. For both these reasons I worked hard to ensure that I employed interviewers were who not be seen as affiliated with a paramilitary. One effect of this was that a majority of interviewers were female.

The third issue I faced in choosing a neighborhood to sample was finding interviewers.

The neighborhoods in which I was interested in working tend to be highly insular. Locals often described Belfast to me as a collection of villages. Like in a village, within each neighborhood, everyone knows each other and their business, and is suspicious of outsiders. This limited my choice of interviewers. I consulted with many local academics, community activists, and

213 residents, and almost uniformly they told me that I needed to use interviewers that were from each neighborhood.

There were two main reasons why I could not just hire and train university students and send them out into the neighborhoods to conduct the survey. First, there is the issue of getting the door open. I was told repeatedly that respondents would be very suspicious of interviewers they did not recognize and unlikely to complete the questionnaire.9 Second, several people independently mentioned that they thought it could put the interviewers at risk. Residents suspicious of outsiders traipsing the neighborhood with clipboards could warn the interviewers to leave or even call the local paramilitary. Not everyone agreed with this assessment, but enough people I trusted thought it was a concern, so I decided it could not hire a single team to use everywhere. Rather, I hired and trained people to conduct the survey in the neighborhood in which they lived or worked. Therefore, in order to work in a neighborhood, I needed to find a team of interviewers who could conduct the survey there without problem. Of course, there is a trade-off in this decision. While using local interviewers meant that selected respondents were more likely to participate, they also might not to divulge certain things about themselves to people they see on a regular basis. Ultimately, I decided that it was a compromise I had to make; given the possible risk to outside interviewers, I could not ethically employ them.

The issue then became finding suitable interviewers in neighborhoods that fit my criteria of mainly Protestant and strong parading scene. To do this I used every connection I had, made cold calls, and followed every lead. I established five teams (four with three members, one with

9 People still live by what the late Nobel Laureate Seamus Heaney called “The famous / Northern reticence, the tight gag of place” whereby “whatever you say, you say nothing.” Northern Ireland, Heaney continued, is the “land of password, handgrip, wink and nod.” Seamus Heaney, “Whatever You Say, Say Nothing,” Opened Ground: Selected Poems, 1966-1996 (New York: Farrar, Straus, and Giroux, 1998), p. 124.

214 four members) through a local housing association, a restorative justice organization, a women’s center, a youth club, and a personal connection. I trained each team and then monitored their work as they conducted the survey.

In the end, the neighborhoods in which the survey was conducted––three in West Belfast, three in South Belfast, and three in East Belfast––were in part chosen by me based on demographic characteristics, but also chosen for me based on where I where I was able to assemble a team and get local permission to work. (Table 7.1 provides demographic information from the 2011 Census on the nine neighborhoods included in the survey.) The enumerated areas are therefore not representative of any larger population. However, I have no reason to suspect that any additional bias was created based on where I was able to work and where I was not.

Neighborhoods that fell through did so because people not return phone calls, particular individuals were not interested, and so on, not factors about the neighborhood which would likely matter to the results of the survey. In other words, I have no reason to suspect that they neighborhoods where the survey did take place are meaningfully different from similar neighborhoods where it did not take place. That said, since the areas were selected purposefully, rather than randomly, the respondents in the survey are not representative of the Protestant population at large. However, within each neighborhood, respondents were randomly selected, as

I will now explain. !

Selecting Respondents

Once each neighborhood was chosen, the goal was to randomly select individuals to interview. To select individuals, I began by selecting households. First, I used detailed maps

215 created by the Land and Property Services (see Figure 7.1) to generate a list of each address in a

Small Area, the smallest geographic unit in the census.10 Each Small Area generally contains

100-200 households and 200-400 people (see Figure 7.2). Each neighborhood is made up of a number of Small Areas, but their boundaries do not always align with local definitions. I therefore used Small Areas which were entirely within the local understanding of the neighborhood. Using a random number generator, I selected one-quarter of houses in each Small

Area for an interview.

Lists of selected addresses in hand, interviewers went to each house to find target subjects to interview. To be eligible for the survey, a house must have a male who is 18 years old or older and from a Protestant background who usually lives there. Although women do participate in parades, they are a small minority and to include females in the sample would reduce the number of participants surveyed.11 This trade-off reduced costs at the expense of the richness of the data.

If the house had more than one eligible subject, the one with the most recent birthday in the calendar was selected for the sample. If the house had more than one eligible subject, the one with the last (most recent) birthday was selected for the sample. If the household did not have an eligible subject (single-mothers with children and elderly widows were common in these neighborhoods), then the address was replaced using the original randomization. I did the same if the respondent refused to participate. If the respondent was not home, enumerators were to try to

10 Detailed maps that show each address are available at: https://www.spatialni.gov.uk/geoportal/viewer/index.jsp? title=&resource, accessed 7 October 2013. Maps of each Small Area are available at: http://www.ninis2.nisra.gov.uk/ public/StaticMapsAddress.aspx, accessed 7 October 2013.

11 On the role of women, see Linda Racioppi and Katherine O’Sullivan See, “Ulstermen and Loyalist Ladies on Parade: Gendering Unionism in Northern Ireland,” International Feminist Journal of Politics, Vol. 2, No. 1 (Spring 2000), pp. 1-29; and Katy Radford, “Drum Rolls and Gender Roles in Protestant Marching Bands in Belfast,” British Journal of Ethnomusicology, Vol. 10, No. 2 (2001), pp. 37-59.

216 make an appointment, or, if that was not possible, to return to the house until he was home (up to four times). Once the enumerator identified and reached the selected respondent, they conducted the interview in person, recording the responses on paper.12 This process yielded 228 responses.

The last four columns of Table 7.1 compare the census figures to the survey figures on two measures, the percent identifying as British and the percent of men with full-time jobs. As we can see, the survey did a fairly good job matching the general characteristics of each neighborhood.

Further, I believe that this procedure constructed an appropriate sample of nonparticipating Protestants. That is, they can be reasonably considered a pool of potential participants. First, they are drawn from the same neighborhoods, many of which have well established parading organizations. Second, they have considerable social ties to participants.

Among nonparticipants in the sample 67 percent have close friends who parade, 35 percent had a father who paraded, 64 percent had other family members parade (71 percent had any family), and 28 percent have personally been asked to parade. Third, a majority of nonparticipants (64 percent) attend parades, an action which suggests ideological support and mobilization potential.13 And finally, a substantial number of nonparticipants have participated at some point in their life. Thirty percent of them marched before the age of 16 and 21 percent have marched after the age of 16. Altogether, these structural and biographical features of the nonparticipant sample suggest that they are a pool of potential participants comparable to the participants.

12 In one neighborhood, interviewers had trouble gaining cooperation for face-to-face interviews and therefore left the questionnaire to be filled out by the selected respondent. Wherever possible, they sat with the respondent as they filled it out.

13 Further, 81 percent of nonparticipants identify as British and 84 percent identify as unionist.

217 Figure 7.1. Example Small Area14

Figure 7.2. Screenshot from the Land and Property Service’s Spatial NI Map15

The Interview Process

Once the enumerator identified and reached the selected respondent, they conducted the

14 From the Ballysillian neighborhood in North Belfast (not in the survey). Available at: www.ninis2.nisra.gov.uk/ public/StaticMapViewer.aspx?geoCode=N00000910&geoLevel=SA. Accessed September 25, 2013.

15 From a section of the Small Area from Figure 7.1. Available at: www.spatialni.gov.uk/geoportal/viewer/index.jsp? title=&resource. Accessed Accessed September 25, 2013.

218 interview in person, recording the responses on paper.16 On average, the interview took 52 minutes. At the end, respondents were asked if they would agree to participate in a follow up survey as well. Those who did gave their contact information to the enumerator who recorded it on a separate sheet of paper. To incentivize participation in the second survey, respondents were told that if they completed the second interview they would be entered into a lottery to win £100. !

Potential Biases

As already noted, the data are biased due to the deliberate selection of the areas of enumeration. I must additionally note a potential bias stemming from individuals’ choice of responding to the survey. Individuals selected for the survey were told that this was a survey for an American university about “parades and culture in Northern Ireland.” This could have led men with little or no interest in parades not to participate. Though conversely, it could have led who do parade but do not wish to discuss it publicly to not participate. Unfortunately, there is no way to estimate or correct for these biases. !

Descriptive Statistics

The Paraders

In this section, I present descriptive statistics that begin to paint a quantitative picture of parade participants. Twelve percent of the sample are active marchers, but in this section I look at both current participants and former participants, i.e., people who marched in parades at some

16 In one neighborhood, interviewers had trouble gaining cooperation for face-to-face interviews and therefore left the questionnaire to be filled out by the selected respondent. Wherever possible, they sat with the respondent as they filled it out.

219 point after the age of 16. Seventy men (31 percent of the sample) fall into this category.17 I exclude boys who only paraded when they were younger than 16 since I am interested in mature decisions and it is unclear how much of a choice children make about such matter and how much is their parents making the choice for them. However, this does include men who began parading before they were 16 as long as they continued to participate after they were 16. The reason is that while the decision to join at, say, age six may not been made entirely (or at all) by the child, the decision to continue parading at age 18, 30, and 60 is made by an adult. I refer to marching after the age of 16 as marching as an adult.

When do paraders begin participating? On average, participants joined their first parading organization when they were 18.3 years old. The median age, however, is 16, which is reflected in the distribution’s skew toward youth seen in Figure 7.3. Men in their late teens and early twenties are most likely to be active paraders. Figure 7.4 shows the percentage of men in the sample who participated at each age. The percentage rapidly increases from age 3 to age 17, where it peaks, and then gradually declines as men get older. The data get less precise as age increases since the denominator only includes men who have reached each age. So whereas all men in the sample had been 17 years old (since they had to be at least 18 to participate in the survey), only half of the sample had been older than 42, and only one-quarter had been older than 57.

17 There are no robust estimates for the number of paraders in Northern Ireland to which we can compare my findings. According to a 2010 survey of 994 residents of Northern Ireland conducted by the University of Liverpool, 28.6 percent of Protestant males are members of the Orange Order (69 out of 241). The figure rises to 33.9 percent of Protestant males when I restrict the sample to Belfast residents (19 out of 56). Unfortunately, the 2010 survey just asked about membership in the Orange Order and not in other Loyal Orders or marching bands. If we sum the estimates of members in the Orange Order, Apprentice Boys, and marching bands and then adjust for the fact that there is some overlap in membership, I calculate that 13 to 17 percent of Protestant men are current parade participants. In my survey, 12.4 percent of the sample is a current member of a parading organization. Jonathan Tonge, Bernadette C. Hayes, and Paul Mitchell, Northern Ireland General Election Attitudes Survey, 2010, Distributed by UK Data Archive, 2010, doi: 10.5255/UKDA-SN-6553-1.

220 Figure 7.3. Histogram of the Age Participants Began Parading 25 20 15 10 Frequency 5 0

0 10 20 30 40 50

Age ! Figure 7.4. Percentage of Respondents Marching, by Age 20 15 10 Percentage 5 0

0 20 40 60 80

Age

Many participants actually began parading even earlier in their lives. Sixty percent of participants belonged to the Junior Orange Order or marching band before the age of 16 and 59

221 percent marched in parades as a small child even before formally joining a group (e.g., marching alongside their father for part of a parade). Even higher numbers were active in parading culture in other ways. Nearly all participants (99 percent) attended parades as a child and 93 percent collected wood for Eleventh Night bonfires, which are lit to mark the start of the Twelfth of

July.18 These figures are all significantly higher (at the 99 percent significance level) for participants than non-participants: among non-participants, only 8 percent belonged to a parading organization before age 16, 18 percent marched as a child, 87 percent attended parades, and 72 percent collected wood for bonfires.

We have even more information about the 28 current paraders. In the last year, they marched an average of 9.9 times, with a median of 8 times and a range from 1 to 32. In the last year, 82 percent have marched on a route labeled contentious by the Parades Commission, and

89 percent have marched on a route that was protested by Catholics. Thus, a large majority of participants in the sample—drawn from three of Belfast’s four parliamentary constituencies

(South, West, and East)—have participated in a disputed parade in 2012-13, even though only a very small percent of parades during that period were considered contentious by the Parades

18 In Protestant communities across Northern Ireland, large bonfires are lit the night before the Twelfth of July. The bonfires are festive events usually celebrated by the whole neighborhood or town. The bonfires are built by groups of mainly young boys and men who often spend weeks collecting wood and other flammable objects (including, car tires, couches, and mattresses), and then building the bonfires. Some of them are true feats of engineering, with precisely placed wooden pallets stacked well over fifty feet high. Boys will spend all day and night guarding the growing towers from attempts by boys from rival neighborhoods (both Protestant and Catholic) to set it alight early. When they are finally burned at midnight of July Twelfth, the bonfires are often adorned with images of nationalist and republican political leaders and topped with the Irish tricolor flag. In recent years, the flames have also consumed the flags of Poland and Cote d’Ivoire (the former a reaction to the large, and Catholic, Polish immigrant community; the latter a mistake, since the Ivorian flag is mirror image of the Irish flag) and an effigy of a popular local Catholic priest who had recently committed suicide.

222 Commission.19 This is likely because portions of three of the major arterial streets into the city center are disputed by Catholic residents (Donegall Street from the north, Newtownards Road from the east, and Ormeau Road from the south). !

PATTERNS OF PARTICIPATION

Although I label all of these men as paraders or marchers, they in fact were members of diverse organizations and followed differing trajectories of participation. Of the 70 respondents who have marched as adults, 47 percent were members of a loyal order at some point in the life and 59 percent were members of a marching band. Among loyal order members, 88 percent were in the Orange Order, 27 percent were in the Apprentice Boys of Derry, and 18 percent were in the

Royal Black Institution. Among band members, 68 percent have been in blood and thunder bands and 24 percent have been in other types of marching bands.20 There is also some cross- membership, with 9 percent of participants having been in both a band and a loyal order (though not necessarily at the same time). Figure 7.5 displays these graphically, with the general categories in dark gray and the more specific organizations or categories in light gray.

The trend of switching membership between or among bands and loyal orders is common among activists. This reflects the simple fact that people often move in and out of organizations.

19 In 2012-13, 5 percent of parades were deemed contentious (215 of 4,4449). Even though contentious parades are disproportionately Protestant (while 58 percent of all parades are Protestant, 90 percent of contentious parades are Protestant), only about 7.5 percent of Protestant parades nationwide are deemed contentious. The figures are not broken down geographically, so we cannot calculate the proportion of Belfast parades that are contentious, but the figure is low. Parades Commission for Northern Ireland, Annual Report and Financial Statements for the Year Ended 31 March 2013 (London: Stationary Office, 2014), pp. 10-11.

20 These figures do not add to 100 percent due to missing data on the type of band several respondents were in.

223 Corrigal-Brown has analyzed this pattern in her study of American social movements.21 She identifies three ideal-type trajectories of participation: persistence, where an individual maintains their activism over time; disengagement, where an individual quits their activism and never again gets involved; and individual abeyance, where an individual stops participating but rejoins later in life. The same patterns exist among paraders. Twenty-one percent of participants are persistent, meaning they joined one organization and have remained a member ever since. Fifty- nine percent of participants have disengaged, meaning they left their parading organization and have not paraded since. And 19 percent of participants have undergone some form of individual abeyance, meaning they are currently active paraders, but have also left a previous organization.22

Figure 7.5. Percentage of Participants in Each Organization (N=70)

21 Catherine Corrigal-Brown, Patterns of Protest: Trajectories of Participation in Social Movements (Stanford: Stanford University Press, 2011).

22 There are several timelines that could constitute this pattern in the data, such as joining Groups A and B at the same time and continuing membership in A but quitting B; or joining Group A, quitting, and then not parading for several years before joining Group B.

224 !

The Attenders

As large events, parades consist of more than those actually marching. A central feature of performance, whether theatrical or political, is the audience. In fact, as Fillieule and

Tartakowsky point out, public demonstrations have multiple audiences, from the observers present at the event to those who read about it in a newspaper.23 Here, however, I look at just the most immediate audience of parades, then men who attend as supporters or spectators. Among non-paraders (that is, people who do not currently parade, in contrast to the previous section which examined people who ever paraded as an adult), 64 percent attended at least one parade in the last year. This represents 56 percent of the entire sample. On average, attenders attended 5.6 parades, with a median of 4 parades. The mode, however, is 1, meaning that many probably only attended the Twelfth of July. !

The Non-Attenders

This leaves those who did not attend any parades in the last year. They represent 31 percent of the sample. When asked why they did not attend, non-attenders have a range of reasons. A majority (55 percent) said they had no interest, while 20 percent said they had no time. The former reason suggests they had no intention to attend, but the latter suggests they may have liked to if they had the time. However, some non-attenders gave reasons which explicitly suggest aversion toward parades. Ten percent did not attend because they think parades are

23 Olivier Fillieule and Danielle Tartakowsky, Demonstrations, trans. Phyllis Aronoff and Howard Scott (Halifax: Fernwood, 2013), pp. 16-17; also Olivier Fillieule, “The Independent Psychological Effects of Participation in Demonstrations,” Mobilization, Vol. 17, No. 3 (September 2012), p. 236; and Kathleen Blee and Amy McDowell, “Social Movement Audiences,” Sociological Forum, Vol. 27, No. 1 (March 2012), pp. 1-20.

225 sectarian; 4 percent find parades intimidating; 4 percent do not like big crowds or loud music; 3 percent believe parades are associated with paramilitaries; and 1 percent said they had Catholic friends or family they did not wish to offend. Further evidence that the majority of non-attenders do not attend deliberately (rather than because of a health constraints or lack of time) is that only

14 percent of men who did not attend a parade in 2012/13 said that they intend to go watch the

Twelfth of July parade in 2013 (whereas 87 percent of men who attended a parade in 2012/13 intend to go to the 2013 Twelfth).24 Figure 7.6 summarizes the sample’s parading choices in the year between the summer 2012 and summer 2013. ! Figure 7.6. Summary of Respondents’ Parading Behavior, 2012-2013 Respondents' choices about parades, 2012 - 2013 (%) 60

56.2 50 40

30 31.4 Percent 20

12.4 10 0

Did not attend Attended Marched

24 The Twelfth of July is a public holiday in Northern Ireland, so most people do not have work, and many shops are closed, making even running errands inconvenient.

226 Assessing Alternative Arguments

In this section, I use the quantitative survey data as well as the qualitative interview and ethnographic data to test the hypotheses derived from two prominent theories of ethnic conflict: ethnic rivalry and collective action. As I explained earlier, this analysis addresses methodological shortcomings of prior research by using data on participants and nonparticipants. Thus, I provide more sound tests of these important hypotheses than previously available in studies of cultural collective action.

The dependent variable in the quantitative analysis, Parade Participant, takes a value of

1 if he is currently a member of a loyal order or marching band and a value of 0 if he is not.25

Twelve percent of the sample are current parade participants (N=28). I estimate the determinants of participation using logistic regressions with standard errors clustered by neighborhood and interviewer fixed effects. The results are presented in Table 7.2. After the initial analyses, I provide two robustness checks. First, I reestimate the regressions using rare events logit.

Although, King and Zeng state that rare events logit is most useful when less that 5 percent of observations are “events,” it is worth re-running the models since the proportion of participants is still low and the number of observations is small.26 Second, I account for deleted observations due to missing data by reestimating my original models using multiple imputations. !

25 A benefit of this outcome variable is that it measures participation in a specific, discrete action, rather than general support for a cause or vague “movement participation.” Using this type of precisely measurable dependent variable is recommended by McAdam, “Recruitment to High-Risk Activism”; and Gregory L. Wiltfang and Doug McAdam, “The Costs and Risks of Social Activism: A Study of Sanctuary Movement Activism,” Social Forces, Vol. 69, No. 4 (June 1991), pp. 987-1010.

26 Gary King and Langche Zeng, “Logistic Regression in Rare Events Data,” Political Analysis, Vol. 9, No. 2 (Spring 2001), p. 157.

227 Testing the Ethnic Rivalry Hypotheses

The ethnic conflict approach to participation looks to and an individual’s feelings towards their in-group and the out-group. The first hypothesis is that people are more likely to participate the more they identify with their own ethnic group. I measure positive feelings towards the in- group with a measure, Protestant Identification, that captures how strongly the respondent identifies with the Protestant community. Protestant Identification is an additive scale of three survey questions, each with five response categories.27 I find that stronger identification with the

Protestant community is not associated with an increased probability of parade participation.

This result remains in alternative measures of identification (see Appendix Table A1), including entering each component variable individually or substituting with an indicator for self- description as British (as opposed to Irish, Northern Irish, or Ulster-Scots).28

The second ethnic conflict hypothesis is that people are more likely to participate the more hostility they feel towards the rival out-group. I measure negative feelings towards the out- group with a measure, Anti-Catholicism, that captures the degree of anti-Catholic views professed by the respondent. This variable is scaled from 0 to 1 to account for differing numbers

27 Respondents were first asked if they strongly agree, agree, neither agree nor disagree, disagree or strongly disagree with the statements “I feel strong ties with other Protestants in Northern Ireland” and “In many respects, I am like most other Protestants in Northern Ireland” (coded from -2 to 2). Then they were asked “Would you be proud to be called an Ulster Protestant?” on a five-point scale from “not at all” to “very much” (coded 0 to 4). The three responses were summed to produce a scale ranging from -4 to 8. For similar measures, see Bert Klandermans, Jojanneke van der Toorn, and Jacquelien van Stekelenburg, “Embeddedness and Identity: How Immigrants Turn Grievances into Action,” American Sociological Review, Vol. 73 (December 2008), pp. 992-1012; Bernd Simon, Michael Loewy, Stefan Stürmer, Ulrike Weber, Peter Freytag, Corinna Habig, Claudia Kampmeier, and Peter Spahlinger, “Collective Identification and Social Movement Participation,” Journal of Personality and Social Psychology, Vol. 74, No. 3 (March 1998), pp. 646-58.

28 Generalized ordered logit analyses (not shown) demonstrate that non-attenders have lower in-group identification than attenders and paraders,

228 Table 7.2. Determinants of Current Parade Participation, Logit Models Model 1 Model 2 Model 3 (No controls) (Main model) (Disaggregate social pressure) Ethnic Rivalry Protestant Identification 0.04 [0.11] -0.07 [0.13] -0.07 [0.13] Anti-Catholicism 1.88 [1.26] 2.56 [1.92] 3.07 [1.87] Collective Action Social Pressure 0.68 [0.15]*** 0.79 [0.17]*** Family Expected Participation 0.56 [0.32]* Community Thinks Less of Non-Participants 1.12 [0.39]*** Social Ties Family Marched -0.04 [0.83] -0.11 [0.80] -0.04 [0.83] Close Friends at Age 16 0.31 [0.30] 0.31 [0.38] 0.29 [0.38] Been Asked to March 0.43 [0.53] 0.22 [0.92] 0.30 [0.90] Control Variables Marched as Youth 0.93 [0.47]** 0.96 [0.49]** Age -0.05 [0.03] -0.05 [0.03] Education -1.14 [0.94] -1.12 [0.91] Children under 18 -1.48 [0.74]** -1.51 [0.71]** Full-Time Job 2.59 [0.83]*** 2.56 [0.80]*** Church Attendance 0.35 [0.15]** 0.35 [0.16]** Constant -4.93 [0.90]*** -4.50 [2.14]** -4.99 [2.45]** Number of observations 174 160 160 Correctly predicted 90.80% 90.6% 91.25% Reduction in error 23.81% 16.67% 22.22% Standard errors clustered at neighborhood level in brackets. Enumerator fixed effects are not reported. * p<0.10, ** p<0.05, *** p<0.01 of response categories in the questions.29 I find that anti-Catholic prejudice is not a significant predictor of participation in any of the model specifications. The null finding holds if I recreate

29 The first question asked how much the respondent would mind if a close family member married a Roman Catholic. The second asked “Do you think that sometimes Catholics need to be reminded that they live in the United Kingdom?” The third asked “How much of the sectarian tension that exists in Northern Ireland today do you think Catholics are responsible for creating?” The fourth: “Do you strongly agree, agree, disagree, or strongly disagree that over the past few years, Catholics have gotten more economically than they deserve.” Each variable was scaled from 0 to 1, with 1 as the most anti-Catholic view, then added together, and then re-scaled from 0 to 1. The third and fourth questions, as well as the scaling, are adapted from P. J. Henry and David O. Sears, “The Symbolic Racism 2000 Scale,” Political Psychology, Vol. 23, No. 2 (June 2002), pp. 253-283.

229 the scale without the question on whether Catholics deserve their economic gains, with had substantially more non-responses than the other three questions (Appendix Table A2).

Both identification with the in-group and hostility toward the out-group are plausibly endogenous to parade participation. It is easy to imagine that membership in a parading organization and marching in parades increases one’s attachment to the Protestant community and prejudice towards the Catholic community. However, collecting a valid retrospective measure of these attitudes from before a respondent joined would be near impossible. We would believe answers to a question like, “How did you feel about Catholics thirty years ago, when you were eight years old?” We cannot, therefore, attribute any causality and settle only for association. Had I found a positive and significant relationship between in-group and out-group attitudes and participation, we could attribute little to the correlation. The standard model of participation argues that attitudes precede and motivate participation. But we also know that attitudes can be a function of involvement. As a result, distinctive attitudes among participants could be a cause or an effect of their participation. In a cross-sectional study, we have no way of knowing. Finding no difference in attitudes, however, actually gives us more confidence that holding particularly strong attitudes did not motivate participation. Unless we assume that people with less of an attachment to Protestants and prejudice towards Catholics join parading organizations and then increase those attitudes through their participation,30 the data show that, regarding these two positions, parade participants reflect the communities they come from.

By sampling nonparticipants as well as parade participants, the survey reveals that,

30 A highly unlikely, but not impossible, scenario. For example, Ziad W. Munson, The Making of Pro-Life Activists: How Social Movement Mobilization Works (Chicago: University of Chicago Press, 2008), p. 189, finds that 23 percent of his sample of pro-life activists were pro-choice prior to mobilization; another 20 percent expressed ambivalent beliefs about abortion.

230 ceteris paribus, participants do not express notably extreme ethnic attitudes. Their degree of in- group identification and out-group prejudice does not stand out from their non-parading neighbors. Note that this does not mean that paraders do not express high levels of Protestant identification or anti-Catholicism, only that they do not express levels that are notably higher than nonparticipants. This result matches a robust finding across number studies of social movement mobilization: ideological support for the movement does not explain why some people join while others do not. For example, Klandermans and Oegema find that 74 percent of the respondents that they interviewed before a large Dutch peace demonstration in 1983 supported the goals of the rally. But when they reinterviewed respondents after the demonstration, only 4 percent had attended.31 In his study of the 1964 Mississippi Freedom

Summer, Doug McAdam concludes that attitudinal and ideological support does not distinguish participants from applicants who withdrew before the summer. Each group was equally supportive of the goals of the project, thus ideology cannot explain differences in participation.32

Ideological sympathy is clearly not enough. And Ziad Munson’s research on pro-life activists demonstrates that people with a wide-range of attitudes toward abortion become active in the movement, leading him to conclude that “beliefs about abortion do not generally lead people into activism.” Even activists who considered themselves pro-life before they were mobilized often held “thin beliefs” about abortion, as opposed to the “more robust ideological commitments” that we expect to cause activism.33

31 Klandermans and Dirk Oegema, “Potentials, Networks, Motivations, and Barriers,” p. 524.

32 McAdam, “Recruitment to High-Risk Activism”; and Doug McAdam, Freedom Summer (New York: Oxford University Press, 1988).

33 Munson, The Making of Pro-life Activists, pp. 189, 186.

231 !

Testing the Collective Action Hypotheses

Rather than looking at the respondent’s attitudes, the collective action approach looks to the private costs and benefits of participation. In this section, I test three hypotheses: participants receive selective material benefits, participants are socially pressured into parading, and participants have more pre-existing social ties than nonparticipants.

SELECTIVE MATERIAL BENEFITS AND MOBILIZATION

Selective material benefits have long been seen as a way to overcome the free rider problem and produce collective action. In contemporary parading, however, I found almost no evidence that there are any selective material benefits directly available to participants. There is limited evidence that indirect material benefits may be available to participants in the form of loans, but for the overwhelming majority of participants, parading is a net financial loss. In this section, I document that the vast majority of paraders do not have a material incentive to participate; that those who have happened to receive economic gain are a minority, and that these gains are too small and unreliable to realistically incentivize rational participation; and that paraders in fact pay a direct cost to participate and open themselves to the risk of problems at work. From this evidence, I conclude that Olson’s primary solution to the collective action problem does not operate in loyalist parading.

Throughout my fieldwork I sought evidence that parade participants received selective material incentives as predicted by the collective action approach to participation. I found almost none, and that which I did find was unsystematic. Quantitatively, only three out of the seventy paraders in my survey reported that being in a loyal order or band had ever helped him

232 financially.34 The remaining 96 percent of participants responded that they had never received an economic benefit from parading. Additionally, current parade participants were asked “Why did you march last 12th?” and the interviewers noted their open-ended responses. Not one participant mentioned anything even slightly related to material gain.

My interviews reveal the same pattern. Nearly every time I asked a parade participant if there were any personal financial benefits the answer was an emphatic “no.” In fact, without skipping a beat, most respondents would continue by saying that parading actually costs a lot of money, a topic I will elaborate shortly. Time and again I heard from respondents that they have never benefited economically from their participation, neither in the form of direct rewards nor from preferential access to employment, contracts, or promotions. This is a big change from generations past when membership in the Orange Order was a path to employment and promotion in many industries. In some fields, such as unionist politics, membership was a virtual requirement. The lack of tangible benefits, according to one member, even leads a number of people to quit. “You get people joining and it’s not what they expected and then they leave,”

Howie told me.

“What do they expect?” I asked.

“I don’t know. I think they maybe expect financial benefits or other benefits.”

But even non-participants I spoke with know that parading costs participants far more than it materially benefits them. And among non-participants I surveyed at random, only 11 percent reported that they thought that joining a parading organization benefited people

34 The question read: “Did being in your Order/band ever help you financially, such as by helping you to get a small loan, job, or promotion?” This question was only asked of men who had ever been parade participants, so it not included in the regression.

233 financially. So it seems unlikely expectations of financial gain are a major motive for joining a parading organization.

That said, a limited number of participants did mention minor economic benefits available to paraders. In particular, several told me that by joining a wide network, membership can help people find employment.35 For example, Walter and Chris are bandmates who co-own a small business and they hire three other fellow band members for casual work “a couple of days here and a couple of days there.” Walter explains: “if you’re in a band and you become unemployed and there are people in your band that are working, they’re going to be looking for work for you.” But when I then asked if help in finding a job was a reason why people joined bands, he cut me off: “No, no, no,” he made clear. Similarly, Albert suggests that if he needed to hire a plumber or other tradesmen, he would hire one he knows through the Orange Order.

“Not,” he notes, “because we get it cheaper. Just to give them the job.”

In Walter and Albert’s comments, we see a feeling of wanting to help one’s own. This is not specific to parading, but a general phenomenon of finding work through clubs and other social organizations. As Howie says, “It would be the same as… if you were in a golfing society, and you were an accountant and you said, ‘I’m out of work,’ a guy over there’s looking for an accountant.” Though he emphasizes that while “it could happen… in my experience nothing like that has ever happened.” Social connections formed through parading, therefore, occasionally lead to work opportunities. But very few interviewees mentioned it and those that did only discussed opportunities for odd jobs, not sustained employment. Additionally, these social ties are not unique to parade organizations and are just as likely to exist through any potential

35 Ray Casserly, “The Fyfe and My Family: Flute Bands in Rathcoole Estate,” Irish Journal of Anthropology, Vol. 13, No. 1 (2010), pp. 8-12, also finds that marching band members gain skills and job opportunities.

234 connection.

Two interviewees told me about financial benefits that were more directly related to their parading activities. Nigel owns a company that does business all over Ireland and he believes that his membership in the loyal orders actually helped him gain clients in the Republic of

Ireland. “It probably opened some doors for me, believe it or not, in the South of Ireland, not the

North,” he says. People there found his membership intriguing and it started conversations that led to personal bonds with potential clients. Robby also mentions benefits that come specifically from parading. He told me his Orange lodge runs a small private safety net for its members who fall into financial trouble. His lodge collects funds that they give to charity but “if that arose that you were having difficulties, you could apply to the charity for some money to help you through.” More informally, Robby tells me that “if you needed a lend of a couple of hundred quid for a week or two, there’s plenty of boys there that if they had it they’d give you it. At the end of the day because we’re all friends.” Yet despite asking each of my interviewees if there were financial benefits to parading, and often citing loans as an example of a benefit, Robby is the only one who mentioned these formal or informal loan opportunities. This leads me to believe that they are likely uncommon and certainly not on high people’s minds. So there is some evidence that joining a band or loyal order can bring limited financial benefits in the form of occasional employment or a loan. But these benefits are both small, apparently uncommon, and vastly outweighed by the costs of participation.

Personal Economic Costs of Parading. The most common response when I asked if there were personal economic benefits to parading was that it actually cost participants to parade. As Billy

235 quipped, “The financial benefit is that it actually costs you money! There’s absolutely no financial gain in being in the Orange Order or in the band.” Loyal order members have to pay dues, membership fees, and other incidental expenses, but it is band members who really pay dearly. Buying new musical instruments and uniforms can costs members hundreds if not thousands of pounds, and these purchases are largely self-financed. Steven described some of his band’s expenses: £700 for a flute, £600 for a drum, £250 to £400 for a uniform, and £25-30,000 for an upcoming trip for the whole band to the Somme battlefield in France. And for all those expenses, he emphasizes, “We’re subsidized by no one… Everything we do we do ourselves.”

Bands like his can make some money by selling CDs, DVDs, performing at functions, and hosting fundraising parades, but generally this generates a pittance compared to their expenses and members have to pay a significant amount out of pocket. Some bands get an occasional grant from the Arts Council of Northern Ireland and other funding agencies (including the Irish government), but it all goes to the organization, not the members. As Mikey tells me: “Any grants that the bands get, they’re for something. They’re not just handing money over to them, they’re for ten new flutes, new drums, uniforms. That all has to be counted for with receipts.” A recent report estimated that the loyal orders and bands spend £15.4 million per year (roughly

$23.5 million at the time the report was released) on expenses such as uniforms and regalia, catering, transportation, musical instruments, and capital projects.36 !

Problems at the Workplace. Besides these direct costs of membership, there are also the indirect economic costs of parading. Specifically, many participants believe that their parading could hurt

36 RSM McClure Watters, “The Socio-Economic Impact of the Traditional Protestant Parading Sector in Northern Ireland,” May 2013, pp. 5-6.

236 them in finding employment or advancing in their careers. Seventy percent of current paraders in the survey reported that they would not put a parading leadership position on a resume or job application. Since their membership is a potential source of conflict at work, most participants keep quiet about their parading at the workplace. The two general justifications for this silence is that they would not want to offend Catholic colleagues or other co-workers who are upset by parades and/or they believe that their membership might hurt their chances at advancement. Even those who do not think that being a parader would necessarily hurt them at work tend to keep it quiet. For example, Lee recalled a situation at his office where one female colleague—“I assume she’s Catholic, I don’t really know to be honest”—was complaining about parades delaying traffic throughout the city. A male colleague from England then mentioned that Lee was in those parades, leading the first colleague to “shut up” about it. The English man tried to goad the conversation a bit further, but the woman tried to end the conversation. Lee reflected that he

“wasn’t going to get into it in case I offended anybody.”

Avoiding this kind of awkwardness and uncomfortableness leads participants to keep private. Scott, who plays a flute in a band, goes so far as to lie to his colleagues rather than tell them about parades. He tells me that when people at his middle-class job ask him about weekend plans, he says he is not doing anything or makes something up. The exception that proves the rule is that when his band performed a prestigious concert at a major arts’ venue, he was willing to talk about it at work. The acceptability of performing at a reputable concert hall, rather through the raucous and liquored streets of working-class neighborhoods, made him feel comfortable to even invite colleagues from his middle-class workplace to the show. He seemed proud that several came to see him perform. In contrast, he was clear that he would not invite

237 them to a regular band parade because he would not want people from work to make assumptions and form the wrong opinions of him.

Besides not wanting to offend colleagues, participants also discussed that being in a loyal order or marching band could hurt their career. Michael, for example, is a manager in an industry that is “predominantly made up of Catholics.” As a result, he knows that for the sake of his professional life, he needs to keep the fact that he is an Orangeman quiet. He chose his words carefully as he told me the following: “I— I know my— I know my position. I know I can’t go out and say, I’m— ‘Arms up, I’m in the Orange Order here.’ I know it might be detrimental to my progression, so I just keep it to myself. You know, I’m not ashamed of it, but I don’t— I don’t—”

He paused, so I began to ask, “Do you think it could affect your—”

“To answer your question, yes. There is a risk factor in my standing in my organization.”

That said, Michael told me that though their parading activities could be an issue at work, he does not feel that he has ever actually been discriminated against. In fact in none of my formal interviews did anyone feel that they had personally discriminated against. I kept hearing complaints or rumors about people getting fired, but no one seemed to personally know anyone who it actually happened to. The exception is that one very active participant I knew told me that he had been forced out of his job because he was a member of the Orange Order. I was unable to follow up with him about the details, but in my many interactions with this man, he struck me as prone to exaggeration and quick to see anti-Protestant discrimination everywhere. So while his story is quite possibly true, I cannot confirm it.

Both the interviews and the survey suggest that selective material benefits are not a cause

238 of parade participation. Although Olson’s hypothesis is intuitive and others have found evidence of selective incentives at work in a wide-range of collective action,37 including Orange parades in previous decades,38 I find that they are unrelated to parades in Northern Ireland today. My survey interviewers thought that even looking for financial benefits to parading was futile. The survey instrument included the open-ended question “What attracted you to the specific lodge/ band that you joined first?” Interviewers were to mark any of the fourteen listed items that the respondent mentioned. One of them was “financial/employment,” and on several times when going over the survey with the interviewers during training, they laughed or told me that there was no reason to have it as an option. I would insist that though it might be unlikely, I really wanted to know if anyone mentioned it as a reason. In the end, my interviewers were of course correct: in all of the surveys that they administered the box was never ticked. !

Social Sanctioning and Mobilization

The second hypothesis is that people are more likely to participate if they expect to pay a social cost for not participating. I measure this variable, Social Pressure, by summing the level of sanctioning expected from two sources: family and community. Pressure from the family is measured by how much the respondent believes his Family Expected Participation.39 Pressure from the community is measured by whether the respondent believes that his Community Thinks

37 Mancur Olson, The Logic of Collective Action: Public Goods and the Theory of Groups (Cambridge: Harvard University Press, 1965); Samuel L. Popkin, The Rational Peasant: The Political Economy of Rural Society in Vietnam (Berkeley: University of California Press, 1979); Mark I. Lichbach, “What Makes Rational Peasants Revolutionary?: Dilemma, Paradox, and Irony in Peasant Collective Action,” World Politics, Vol. 46, No. 3 (April 1994), pp. 383-418; and Mark I. Lichbach, The Rebel’s Dilemma (Ann Arbor: University of Michigan Press, 1995).

38 Bryan, Orange Parades.

39 “Do you think that your family expected you to join a loyal order or band? Would you say definitely, somewhat, not really, or definitely not?”

239 Less of Nonparticipants.40 Each component part is measured from 0 indicating no pressure to 3 indicating most pressure, and the sum, Social Pressure, ranges from 0 to 6. I find that my measure of social sanctions is positively and significantly associated with participation in loyalist parades. Substantively, if all other variables in Model 2 in Table 7.2 are held at their median value, changing Social Pressure from its minimum to maximum increases the probability of participating 10 percent.41 In Model 3, I disaggregate social pressure into its two components.

The results show that the correlation is substantively and statistically stronger with the measure of community-based social pressure. Men who believe that their community thinks less of parade nonparticipants are more likely to be paraders.

The interviews tell a somewhat different story about social sanctioning. Bar several exceptions, the men I interviewed recall experiencing no social pressure to participate in parades and they believe that their experience is the norm. There are two potential forms of sanctioning from family, friends, or the community: direct and indirect. In the direct form, paraders pressure an individual into joining, say by teasing them, constantly asking them to join, or even threatening to stop socializing with them. In the indirect path, an individual feels that they need to join a parading organization in order to maintain their status or reputation, specifically their reputation as a “good Prod”—a loyal defender of the Protestant faith and stalwart of the Union.

Throughout my interviews I find occasional examples of both types of pressure, but overall my informants agree that there is minimal to no social pressure on men to participate in parades.

Evidence of direct pressure in action comes most clearly from Samuel. Samuel is not a

40 “Do people in this community think less of people who choose not to join loyal orders or bands? Would you say definitely, somewhat, not really, or definitely not?”

41 Calculated using Clarify for Stata 10. The standard errors could not be clustered in the simulated model.

240 member of any parading organization, and has no real interest in parades or parading culture. He is “happy enough once a year, twice a year watch a parade, go to a bonfire and then I would be happy enough not to mention it for another twelve months.” But, he is involved in unionist politics and most of his close friends are members of the loyal orders. His friends tease him about not joining and ask him join constantly: “I mean I have been asked, I don’t know how many times to join the Orange Order, every single week for about the past three years,” he says.

“Do you feel pressure[d]?” I asked.

“Yeah, yeah.”

“Because they are friends or because—?”

“Yeah. Not because of anything else, just because they are my friends. Yeah, I do, no doubt about it.”

One strategy that his friends use is to appeal to his sense of Protestantism and unionism.

When Samuel tells his friends that he simply has no interest in joining the loyal orders, he

“instantly get[s] a lecture, ‘Well you should have an interest, you know. You call yourself a unionist, you call yourself a member of the unionist community, you’ll do anything you can within peaceful means to maintain the union, well this is one way that we mark our territory.’”

And though it is just friendly teasing—a “bit of banter” in his words—it is directed at his self- image and identity as a Protestant and unionist. “If I go on holiday [over the Twelfth of July],

[my friend] would send me a message, ‘Are you going again then, Seamus?’” By calling him by a stereotypically Irish/Catholic name, his friend suggests that Samuel’s loyalty is in question.

Although these comments are delivered and received in jest they demonstrate that this is usable material for a joke. If there was nothing wrong with skipping town during the Twelfth, calling

241 Samuel “Seamus” would make no sense. But instead the quip makes complete sense both to

Samuel and his friend.

Despite the pressure that he has felt personally, Samuel does not believe that it is a common experience. He explains that it is merely a product of who he socializes with:

“Obviously I associate with circles of people who are active in politics and who are in [loyal orders], so obviously I am going to get that because that is who I choose to be friends with.” It is clear from talking with him that he finds his friends’ incessant nagging annoying, but it is just part of their friendships, as gentle teasing is common among many friends. So Samuel does experience direct pressure, but it is specific to his particular social situation and more of an irritation than anything else—and he has easily been able to resist it.

More common is the absence of direct pressure or social sanctioning. Even between fathers and sons, the relationship where we would most expect it, I found little evidence. Many men were asked by their fathers to follow in their footsteps and join them in a parading organization, but felt no pressure from it. Several informants recall that was, in fact, their father not asking them or pressuring them to join that inspired their decision. Alexander states that

“what kind of impressed me was I asked my dad should I join, and he says, ‘It’s your decision, you decide.’ That encouraged me to join more, because I didn’t feel like I was being forced to join, and I wasn’t.… And it impressed me more that he didn’t try and encourage me. He just said,

‘I’d like you to join, I’d like you to follow in my footsteps, but clearly it’s a decision for you.’

That made me really want to join.” The fact that his father made it clear that the decision was one that he had to make for himself left a lasting impression on the informant and encouraged his membership.

242 Indirect social pressure is just as rare. This form of pressure is more general and not directed at any particular individual, but present in the social atmosphere. It is about maintaining one’s reputation in the community as loyal to the cause. Despite several exceptions, the general experience of Protestant men does not appear to include diffuse pressure to parade aimed at one’s reputation as a loyal member of the Protestant community. Mark, for instance, grew up in a “very staunch Protestant area,” the exact place where we would expect young men to feel that they had to join parades. But in his recollection, “There was no real drive that you had to do something

[join a lodge or band].”

Among participants, I did not even find agreement about whether paraders are respected, high-status members of the community. Many interviewees believe that paraders do gain recognition and respect through their membership: the community admires them for the commitment they have made to represent and defend Protestant culture. But many others reported that they do not feel that parading brings them any additional respect among their peers and neighbors. Billy claims that Orangemen are “not put on a pedestal against somebody in the community that’s not in the Orange.” Tom agrees, saying that “there’s no particular kudos [for parading]…. You’re not elevated” by the community. “Like everything,” he tells me, “some are

[respected], some aren’t.”

Nonparticipants I spoke to agree. Sophie, for example, told me that paraders are “just ordinary fellows. Just [guys we] went to school with, worked with, grew up with. We all know each other.” Even Sophie, a massive parade supporter, sees participants as completely ordinary because they are so embedded in the community. It is hard to see someone as particularly special or esteemed when you have known them your whole life through school, work, and the

243 neighborhood.

Some of the clearest instances of social pressure that I heard in my interviews were people who felt pressured not to participate in parades. Most commonly, this is sons whose parents did not want them to parade. Some parents wanted their children to avoid bands altogether. As a youth, Michael liked the bands and followed the bands, but his “mom would never have allowed [him] to join a band” because, he believes, “people would associate bands with alcohol.” For a similar reason, Rachel was “a wee bit apprehensive about telling [her] parents that [she] was in a band.” Her family had a “very strong Christian background” where

“sometimes bands are seen as being alcoholic monsters.” Other parents were concerned about violence associated with parades. For example, Matt recalls that when he was younger “there was a lot more trouble… around parades,” so his mother did not want him to join a band.

Similarly, Jamie states that “back then bands were very, very associated with paramilitaries” and so his father “wouldn’t let me join until I was a bit older.”

There are also cases where the parents are not opposed to bands in general, but object to their son joining a specific band. Scott, for instance, saw a particular band play well and really wanted to join. His father could see his son’s interest, and told him that he could not join because of the band’s purported association with a paramilitary. As a result, Scott had to wait until he was

18 to join the band. Lee’s band does not have any paramilitary affiliations, but is a blood and thunder style band, which his father objects to. For his father, who has been a member and instructor of high quality marching bands for many years, blood and thunder bands are a sonic disgrace. “If he had his way, I wouldn’t have joined my band,” Lee says between laughs.

A final point that suggests that social sanctioning is not a significant cause of

244 participation is that parading organizations are not structured to coerce people into them. The loyal orders, despite having large nationwide memberships, have very small professional staffs.

They do not have the resources to pressure individuals into joining. They don’t even succeed in maintaining the members they have. Of the 70 respondents who have paraded as adults, only 40 percent are currently members of a parading organization. Fully 60 percent have withdrawn from parading. In fact, several interviewees explained that men who join without being fully committed to parading are likely to quit anyway, thus invalidating the logic of pressuring people to join.

Regarding the role of social sanctioning, then, the quantitative and qualitative evidence diverge. The statistical analysis suggests that men who report social pressure are more likely to participate. The semi-structured interviews, conversely, suggest that social pressure plays little to no role in motivating participation. Both forms of evidence have their problems. For the quantitative data, participants could be more likely to believe retrospective that their family expected them to participate, since they see participation as a good thing. Similarly, they could more likely to report that their community thinks less of nonparticipants, because it reflects their thinking or desires. For the qualitative evidence, people telling the narrative of how they joined a beloved organization that is central to their self-image may wish to downplay the possibility that joining was not entirely voluntary. I do not believe they are misrepresenting it to me, but rather that the story they tell themselves emphasizes free choice over outside pressure.42 When explaining positive personal developments, agency is more appealing than structure.

Overall, the data on social sanctioning is contradictory and inconclusive. Pressure can

42 The discourse of volunteerism is a common theme in mobilization narratives throughout the interviews I conducted. Perhaps it is related to the virtue of voluntary faith in some Protestant theology.

245 work through multiple channels and in multiple directions, and the data do not provide a clear answer. Rather than simply accept the data which is more conducive to my ritual argument—the qualitative—I believe it better to recognize the complexity and give the benefit of the doubt to the more troubling evidence. Thus I conclude that there is tepid support for the collective action approach’s hypothesis that social sanctioning causes mobilization. !

Social Ties and Mobilization

A second major approach to participation in collective action looks to social ties. The general hypothesis is that people are more likely to participate if they have members of their social network who participate. In the survey, I measure two specific social ties for each respondent: Family Marched measures whether or not family members were paraders and Close

Friends (Age 16) measures how many of their close friends at age 16 were paraders.43 This variable is retrospective since current friends are clearly endogenous to parade participation.

Third, I measure whether or not a respondent has Been Asked personally to join a parading organization.44 None of these measures are statistically associated with loyalist parade participation. This is a striking finding, given the robustness of the result across a range of

43 Family Marched takes the value of 0 if no family marched, 1 if either the father or other family members marched, and 2 if both the father and other family members marched. Close Friends (Age 16) ranges from 0 for none or almost none to 4 for all or almost all.

44 See David A. Snow, Louis A. Zurcher, Jr. and Sheldon Ekland-Olson, “Social Networks and Social Movements: A Microstructural Approach to Differential Recruitment,” American Sociological Review, Vol. 45, No. 5 (October 1980), p. 795; Sidney Verba, Kay Lehman Schlozman, and Henry E. Brady, Voice and Equality: Civic Voluntarism in American Politics (Cambridge: Harvard University Press, 1995); Schussman and Soule, “Process and Protest”; and Stefaan Walgrave and Ruud Wouters, “The Missing Link in the Diffusion of Protest: Asking Others,” American Journal of Sociology, Vol. 119, No. 6 (May 2014), pp. 1670-1709.

246 studies on mobilization,45 including loyalist parading.46 A problem with the research on parades, in particular, is that they fail to sample nonparticipants thus they cannot compare the social networks of participants to those of nonparticipants. Consequently, they fail to realize that while participants did know many participants prior to joining, nonparticipants also know many participants. This is a reflection of how deeply parading organizations are embedded in

Protestant communities. It is hard not to know a member of a loyal order or band. Among men who have not participated as an adult, 69 percent have family who march, 62 percent have current friends who march, and 22 percent have been asked to march. These figures are even higher in working class neighborhoods, where parading organizations are stronger (77, 78, and

28 percent, respectively).

Although the statistical evidence demonstrates that having social ties to participants does not increase one’s likelihood of participation, those who do choose to join generally do so through social ties. Pre-existing social ties, therefore, remain a central pathway to mobilization.47

We can see this both quantitative and qualitatively. Among men who have ever paraded, 77 percent cited social ties as what attracted them to the specific parading organization that they joined (59 percent said friends, 17 percent said their father, and 20 mentioned other family; they could list more than one attraction). It is, therefore, unsurprising that 93 percent of them already

45 For reviews, see James A. Kitts, “Mobilizing in Black Boxes: Social Networks and Participation in Social Movement Organizations,” Mobilization, Vol. 5, No. 2 (October 2000), pp. 241-57; John Krinsky and Nick Crossley, “Social Movements and Social Networks: Introduction,” Social Movement Studies, Vol. 13, No. 1 (2014), pp. 1-21.

46 James W. McAuley and Jonathan Tonge, “‘For God and for the Crown’: Contemporary Political and Social Attitudes among Orange Order Members in Northern Ireland,” Political Psychology, Vol. 28, No. 1 (February 2007), p. 37; Northern Ireland Youth Forum, Sons of Ulster: Exploring Loyalist Band Members Attitudes towards Culture, Identity and Heritage (Belfast: Northern Ireland Youth Forum, 2013), p. 14.

47 My survey and interviews show that there are rare cases of men joining with no pre-existing social ties to other parades, but they are exceptional.

247 knew a few (37 percent) or many (55 percent) members in the group at the time of joining.

The interviews confirm these trends. Jamie is typical when he relates that he chose the band to join because he “wanted to be in a band that was more home. And the [band] has quite a few of my family: one, two, three, four family members in it, cousins and stuff like that. And then all the other guys in it, there’s quite a few I used to go to school with.” The band is the only band from the neighborhood that he grew up in, so it draws it members from the community of his childhood. Even though he now live in a different part of East Belfast, Jamie says that joining the this band was a “no-brainer.” George, too, is typical, if perhaps a bit eager. When he retired from the British army, buddies from his unit approached him about joining the loyal orders that they belonged to:

I was asked to join the Apprentice Boys. I had an idea what it was all about, but I said certainly. So I joined that. And within three weeks I was asked if I wanted to join the [Orange Order] lodge that the lads were in. I said, I would. So I joined that. And within three months, I was asked if I wanted to join the [Royal] Black !that they belonged to. I said, I would. As George illustrates, just knowing participants is often insufficient to spur mobilization.

It requires being asked. In George’s case, parading was not even on his radar when his friends asked him; for others, they might have been thinking about joining for some time and then finally decide to do it when they are approached. For example, Robert was a big supporter of parades—“I would have went out to every parade”—but had never taken the next step to actually parade himself. Then “some of the men said to me, ‘Boy, it’s about time you joined.’ And I joined.” The importance of being asked is a common finding in studies of participation in collective action. Schussman and Soule, in their study of protests in the United States, find that being asked to participate is such an important factor in explaining participation that they posit a

248 two-stage model where first a person is asked to participate or not and then they decide to participate.48 In loyalist parading, participants are more likely to have been asked to march than nonparticipants (54 percent versus 22 percent), but among men who have been asked to march, only 51 percent actually did.

Social ties, despite not affecting the likelihood of initial mobilization, clearly still matter greatly for those who are mobilized. This becomes especially apparent when we focus on the role of family. Family ties are often mentioned as a big motive for joining. It is common for participants to explain their own participation by stating simply that it is a “family tradition.”

Participants often place themselves in a lineage that traces back several generations of marchers.

For example, Billy states proudly, “My grandfather was an Orangeman, my father was an

Orangeman, and I’m an Orangeman.” And though he has no sons, he hopes that his grandsons will be Orangemen one day.

The most lauded family tie is between father and son. It is celebrated in perhaps the most popular Orange tune, “The Sash.” The song can be heard many times in a single parade, often with spectators singing along to the familiar refrain, “The sash my father wore.” Parading can be a way to literally follow in one’s father’s footsteps, and many sons do. But many sons do not. In the survey, 86 respondents had a father who marched. Of them, 50 percent have marched themselves and 50 percent have not. For example, Lee relates how he and one brother chose to follow in their father’s footsteps and join the Orange Order, while his other two brothers did not.

One of his brothers will occasionally go to watch a parade, but otherwise does not care; the other brother could not be bothered to leave his Xbox.

48 Schussman and Soule, “Process and Protest.”

249 Thus, even men with strong social ties to participants often do not participate. In fact, as noted earlier, a majority of nonparticipants have friends and family that march. Gary’s story illustrates one possible path away from parading. Gary grew up in working-class Protestant East

Belfast, where Orangeism and band culture were the way of life. His family, too, was heavily involved: “My family were up to their neck and beyond in the Orange [Order] and the [Royal]

Black [Institution],” he recalls. His father was the past Master of his lodge and a lay chaplain; his uncle was a chaplain to the Grand Lodge. And it rubbed off on him as well. He says, “As a kid, the two biggest days in the year for me were Christmas and the Twelfth… It was a big, big part of my year. I mean, my birthday was two days before it, and I looked forward to the Twelfth more than I ever looked forward to my birthday. Massive, absolutely massive in my young life.”

But then two things changed in his life. He was admitted to a grammar school and had an evangelical conversion, and suddenly his social world was flipped upside-down. His new friends and peers at “church and school, were not part of that Orange culture. They were part of a middle-class culture,” where parading less acceptable. So he never joined.

We must also recognize that social ties do not always push people toward participation.

As McAdam and Paulsen correctly note, “Individuals are invariably embedded in many organizational or associational networks or individual relationships that may expose the individual to conflicting behavioral pressures.”49 People live in complex social worlds with lots of social ties to all sorts of people. So even those who have many ties to parade participants likely also have ties to those who do not participate in parades. As I demonstrated earlier, some interviewees felt pressured to not participate. There is no ex ante reason to believe that the ties to

49 McAdam and Paulsen, “Specifying the Relationship between Social Ties and Activism,” p. 641.

250 participants should matter more than the ties to non-participants. Or that pressure from friends and family to participate should outweigh pressure from other friends and family to not participate.

Contrary to the findings of many prominent and diverse studies,50 I find that social ties are not a correlate of mobilization in collective action. The three quantitative measures of ties are statistically insignificant in the regression analysis and interview data show that people with ties to participants choose both to participate and to not. But the data do suggest that social ties remain an important part of the story, just not a causal one. Rather, social ties are the major pathway through which people choose to participate. When people do participate it is generally because of family or friends and their connections influence which specific organizations they join. Also, when participants decide to move to a new band or lodge (because their old one disbanded, personal disagreements, looking for something new, etc.), they move to one where they have friends or family. !

Control Variables

The models in Table 7.2 also all include six control variables. Marched as Youth is an ordinal variable which captures prior participation. It takes a value of 0 if the respondent never marched in a parade before he turned 16, a value of 1 if he marched as a young boy but not as a formal member of a parading organization, and a value of 2 if he marched as a member of a

50 For example, McAdam, “Recruitment to High-Risk Activism”; McAdam, Freedom Summer; Nepstad and Smith, “Rethinking Recruitment to High-Risk/High-Cost Activism”; Schussman and Soule, “Process and Protest”; and Alexandra Scacco, “Who Riots? Explaining Individual Participation in Ethnic Violence,” Ph.D. dissertation, Columbia University, 2010.

251 junior loyal order or band. Consistent with previous studies,51 I find that prior participation increase the likelihood of current participation. Education takes 0 for no qualifications, 1 for qualifications less than degree level, and 2 for degree level. I do not find any relationship between education level and participation. The coefficient is negative, as expected, but it is not significant. Church Attendance ranges from 0 for never attending church to 6 for attending more than once a week. This proxy for religiosity is positively related to participation: paraders attend church more frequently than non-paraders.

Three variables measure biographical availability. The biographical availability hypothesis claims that people are more likely to participate in contentious politics at points in their lives when they have fewer competing claims to their time, such as families or careers.

Research on the effect of biographical availability is inconclusive with studies pointing in opposite directions.52 In my survey, Children under 18 and Full-Time Job are dummy indicators and Age is recorded in years. These variables deliver mixed results about biographical availability. I find that men with young children are less likely to be participants, but men with full time jobs are more likely. And though the young are generally thought to be more available, I find that age is not significantly associated with the likelihood of participation.

Finally, though no measure of income is included in the analysis, we can still say something about it.53 Income seems to be negatively correlated with likelihood of participation.

Other model specifications were run with a dummy indicator that takes the value of 1 if the

51 Corrigal-Brown, Patterns of Protest; and Wolfgang Rüdig and Georgios Karyotis, “Who Protests in Greece: Opposition to Mass Austerity,” British Journal of Political Science, Vol. 44, No. 3 (July 2014), pp. 487-513.

52 McAdam, “Recruitment to High-Risk Activism”; McAdam, Freedom Summer; and Nepstad and Smith, “Rethinking Recruitment to High-Risk/High-Cost Activism.”

53 Only 65 percent of respondents answered the income question.

252 respondent lives in a middle-class neighborhood and 0 for a working-class neighborhood (not shown). The variable had to be dropped, however, because living in a middle-class area perfectly predicts not participating in parades.54 Sixteen percent of working-class residents are paraders, while only 4 percent of middle-class residents are. Finally, education is a reasonable proxy for income (Pearson’s r = 0.51; p = 0.000), and, as previously mentioned, education is negatively associated with propensity to participate, although not significantly so. !

Robustness Checks

In this section, I test the robustness of the quantitative results of the previous section by accounting for potential problems with the data. First, I account for the low proportion of

“positive cases” (participants) in the dependent variable using a rare events logit. Second, I account for missing data in the dataset using multiple imputations. !

RARE EVENTS LOGIT

The first issue is that only 12.4 percent of the dataset are current participants. Given the small number of observations, this low proportion of “positive cases” or “events” in the binary dependent variable can cause standard logistic regressions to malfunction. Specifically, logistic regressions can underestimate the likelihood of the rare event.55 Twelve percent is not considered considered “rare” (King and Zeng suggest under 5 percent), but given the small number of observations, it is worthwhile to compare the original results to results that correct for the rarity.

54 In fact, three middle class residents are participants, but they are dropped from the analysis for missing variables.

55 King and Zeng, “Logistic Regression in Rare Events Data.”

253 The results are robust to the correction in the rare events logistic regression. There are no significant differences between the original results in and the results in Table 7.3. As before, the measures of the ethnic rivalry and social ties are not significantly related to the probability of participation. Social pressure remains significant, though the coefficient is reduced. !

Table 7.3. Robustness Check: Rare Events Logit Model 1 Model 2 Model 3 (No controls) (Main model) (Disaggregate social pressure) Ethnic Rivalry Protestant Identification 0.02 [0.10] -0.09 [0.12] -0.09 [0.12] Anti-Catholicism 1.64 [1.20] 1.80 [1.76] 2.07 [1.71] Collective Action Social Pressure 0.61 [0.15]*** 0.58 [0.16]*** Family Expected Participation 0.44 [0.29] Community Thinks Less of Nonpartic. 0.76 [0.36]** Social Ties Family Marched 0.24 [0.88] -0.07 [0.74] -0.01 [0.76] Close Friends at Age 16 0.29 [0.29] 0.22 [0.35] 0.19 [0.35] Been Asked to March 0.42 [0.51] 0.18 [0.85] 0.25 [0.82] Control Variables Marched as Youth 0.73 [0.43]* 0.71 [0.45] Age -0.04 [0.03] -0.03 [0.03] Education -0.84 [0.86] -0.80 [0.83] Children under 18 -1.06 [0.68] -1.05 [0.65] Full-Time Job 1.84 [0.76]** 1.75 [0.73]** Church Attendance 0.28 [0.14]** 0.26 [0.14]* Constant -4.34 [0.86]*** -3.09 [1.97] -3.34 [2.24] Number of observations 174 160 160 Standard errors clustered at neighborhood level in brackets. Enumerator fixed effects are not reported. * p<0.10, ** p<0.05, *** p<0.01 !

254 MULTIPLE IMPUTATIONS

To account for the loss of observations due to missing data, I reestimate the main model using the AMELIA II program.56 The program imputed values for missing data in the

Table 7.4. Robustness Check: Multiple Imputations Model 1 Model 2 Model 3 Model 4 (No controls) (Main model) (Disaggregate (With income) social pressure) Ethnic Rivalry Protestant Identification 0.19 [0.12] 0.11 [0.15] 0.13 [0.16] 0.13 [0.14] Anti-Catholicism 1.27 [1.14] 2.14 [1.40] 2.53 [1.38]* 2.38 [1.80] Collective Action Social Pressure 0.84 [0.19]*** 0.76 [0.22]*** 0.76 [0.24]*** Family Expected Participation 0.57 [0.20]** Community Thinks Less of Nonpartic. 0.75 [0.34]** Social Ties Family Marched 0.25 [0.82] 0.13 [0.70] 0.29 [0.61] 0.09 [0.74] Close Friends at Age 16 0.21 [0.32] 0.10 [0.33] 0.01 [0.33] 0.11 [0.34] Been Asked to March 0.63 [0.52] 0.49 [0.55] 0.49 [0.49] 0.46 [0.57] Control Variables Marched as Youth 0.71 [0.33]** 0.70 [0.28]** 0.74 [0.35]** Age -0.02 [0.02] -0.02 [0.02] -0.02 [0.02] Education -0.26 [0.66] -0.36 [0.68] -0.33 [0.66] Children under 18 -1.01 [0.52]* -0.88 [0.50]* -1.07 [0.46]** Full-Time Job 0.96 [0.66] 0.80 [0.55] 0.81 [0.98] Church Attendance 0.25 [0.09]*** 0.27 [0.10]** 0.25 [0.08]*** Income 0.10 [0.27] Constant -5.71 [0.83]*** -5.42 [1.27]*** -5.55 [1.25]*** -5.88 [1.87]** Number of observations 227 227 227 227 Standard errors clustered at neighborhood level in brackets. Enumerator fixed effects are not reported. * p<0.10, ** p<0.05, *** p<0.01

56 James Honaker and Gary King, “What to Do about Missing Values in Time-Series Cross-Section Data,” American Journal of Political Science, Vol. 54, No. 3 (April 2010), pp. 561-81; and James Honaker, Gary King, and Matthew Blackwell, “AMELIA II: A Program for Missing Data,” Harvard University, 2012.

255 independent variables, but I deleted the one observation that was missing the dependent variable.

I then used the new imputed datasets to estimate the logistic regressions. As we can see in Table

7.4, the results with the five new datasets are largely similar to the original results, increasing the confidence in the initial estimation. Participants remain undistinguished by their ethnic attitudes and social ties, but do report greater social pressure.

There are two points to notice in Table 7.4. The first is that Anti-Catholicism is statistically significant at the 90 percent level for the first time in Model 3. This clearly conflicts with my argument and the evidence in the rest of the chapter. I acknowledge this discrepancy, but believe that the preponderance of the evidence favors the argument that sectarian attitudes do not differentiate participants from nonparticipants.

In several other of this chapter’s regressions, however, Anti-Catholicism is close to statistically significant at the 90 percent level. To further examine what is driving this near- significance, I disaggregate the variable in the Appendix (Table A2, Model A4). Breaking the measure into its four component parts is revealing. The results show that parading is perhaps about expressing grievances, but not about expressing deep-seated hatred. Paraders are not more likely than nonparticipants to express concern about a close family member marrying a Catholic, express the belief that Catholics sometimes need to be reminded that they live in the United

Kingdom, nor the belief that Catholics are primarily responsible for the sectarian tension in

Northern Ireland. Rather, they more likely to state that they agree that “over the past few years,

Catholics have gotten more economically than they deserve.” All four of these variables capture a dimension of bias against Catholics, but the former two focus on bias against Catholics as people (i.e., I wouldn’t want them in my family and they are inherently traitorous) while the

256 latter two focus on bias against Catholic behavior or outcomes (i.e., raising the level of sectarian tension and succeeding economically). The one measure that is statistically significant captures the latter form of bias: a negative view of Catholic economic outcomes. So paraders remain undifferentiated by bias directed at Catholics qua Catholics. Rather, they express high levels of economic-based grievances against perceived Catholic success.

The second point is that Model 4 includes a measure of Income. The variable is not significantly related to parade participation, a finding that substantiates my earlier suspicion.

Thus, participants are no wealthier or poorer than nonparticipating neighbors. !

Conclusion

This chapter used original survey and interview data to assess hypotheses derived from prominent explanations for participation in ethnic conflict. Care was taken to collect data from both participants and comparable nonparticipants in order to properly test these arguments. And though addressing these methodological concerns proved laborious and costly, the resulting data challenge the scholarly and popular conventional wisdoms. Where the ethnic rivalry approach expects that participants identify with the in-group and are prejudiced against the out-group more than nonparticipants, I find that they are not. Widely-held opinions in Northern Ireland mirror these arguments. Many Protestants hail paraders as “super Prods”—staunchly loyal and dedicated members of the Protestant community—and many Catholics denounce paraders as extreme sectarians, but my data dispute these characterizations. I do not argue that participants are not loyal Protestants nor biased against Catholics. I simply claim that they are no more so than their neighbors who choose not to parade. In this way, paraders are similar to other social

257 movement participants who tend to not be differentiated by ideological sympathy for the movement.57

The collective action approach provides three specific explanations for individual-level participation, of which I find thin support for one. There is robust quantitative evidence that participants are more likely to report social pressure, but the interview data strongly suggest that social pressure play no significant role in mobilization. Thus any conclusion is perforce inconclusive. The data on selective material rewards and the role of social ties, however, are in agreement. Parade participation is not motivated, or even correlated with, the receipt of material rewards. I found no systematic evidence that parade participants gain economically as a result of their actions. On the contrary, they pay—monetarily and in perceived risk to job security—for the chance to parade. Social ties, too, do not increase the probability of participation. Prior research and the conventional wisdom are partly true: paraders do have many social ties to other participants at the time of joining.58 But I found that nonparticipants also have many social ties to parades. So much so that social ties do not statistically differentiate participants from nonparticipants.

The variation is large part explain by variables outside the parameters of the dominant theories of ethnic conflict. Participants do not follow a logic of extreme sectarianism or of precise cost-benefit analysis, rather the regression analysis shows that the factors associated with participation are participating as a youth, church attendance, having full-time employment, and not having children under 18 years old. The latter two variables provided mixed support for the

57 See, for example, Klandermans and Oegema, “Potentials, Networks, Motivations, and Barriers”; McAdam, “Recruitment to High-Risk Activism”; McAdam, Freedom Summer; and Munson, Making of Pro-Life Activists.

58 McAuley, Tonge, and Mycock, Loyal to the Core?

258 biographical availability explanation, with absence of young children supporting the explanation and working full-time opposing it.

259 Chapter Eight ! Conclusion

In a speech less than three years after the Agreement was signed, British diplomat David

Goodall stated: “Peace, in short, is within reach of becoming a habit which no one wants to break.”1 Though largely true, many people in Northern Ireland appear willing to test the habit’s breaking point on a regular basis. This dissertation has examined why that is. Rather than simply assume that large-scale challenges are the product of elite spoilers to the peace process, I examined the choices made by ordinary people to participate or not in loyalist parades. I argued that people choose to take part in loyalist parades because they are a ritual. I found that participants are most interested in the benefits intrinsic to participation in parades, rather than selective material gains or the chance to intimidate Catholics. As a result, people make decisions to participate in contentious parades without consideration of their actions’ profoundly political consequences. The ritual nature of parades severs the expected connection between participation and the external, often negative, consequences, thus creating the environment for sustained conflict. !

Review of the Argument and Main Findings

Contentious rituals, such as loyalist parades, present several puzzles for scholars of ethnic conflict and collective action. First, since contentious rituals produce collective outcomes, they face the free-rider problem. Second, the outcomes that contentious rituals produce are often

1 David Goodall, “Hillsborough to Belfast: Is It the Final Lap?” in Marinne Elliott, ed., The Long Road to Peace in Northern Ireland: Peace Lectures from the Institute of Irish Studies at Liverpool University, rev. ed. (Liverpool: Liverpool University Press, 2007), p. 132.

260 socially harmful, with participants among the most likely to suffer. And third, contentious rituals are characterized by distinctive aspects of ritual such as repetition and goal demotion. Given these features, why would anyone decide to participate? Dominant existing explanations do not provide compelling answers. Elite manipulation arguments do not account for mass participation.

Though these theories assume that ethnic elites are clever and calculating, they fail to give any agency to ordinary people. What is more, I presented survey data from Northern Ireland showing that there is not one elite opinion for the masses to just follow.

Perhaps participants share the goals of conflict-seeking elites. Theories of ethnic conflict suggest that participants hold extreme ethnic attitudes: participants should be in-group chauvinists and hate the out-group. But my survey of randomly selected participants and comparable nonparticipants suggests that participants are not distinguished by their attachment to other Protestants or their feelings toward Catholics.

Rather than seeking the collective outcomes for the in-group and against the out-group, participants might be most interested in themselves. Theories of collective action suggest that people are more likely to participate when the personal benefits of participation are higher than the personal costs. But again, the evidence refuted the claim. The survey and interviews showed that paraders do not receive material incentives, but pay to be able to take part. The evidence regarding social sanction was more mixed. There was some quantitative data suggesting that participants are more likely to experience pressure to parade; but the interview data suggested no such pressure. Finally, pre-existing social ties to other participants, a robust predictor of participation in collective action, also did not distinguish paraders from non-paraders. Parading organizations are so embedded in Protestant communities that even non-participants have friends

261 and family involved in them.

The data I collected in Northern Ireland do not support these prominent theories premised on instrumental logics. This is because existing theories do not account for contentious rituals as rituals. My argument, in contrast, rests on two fundamental insights from multi-disciplinary research on rituals. I argued that rituals provide participants with process benefits intrinsic to the very act of participation and that rituals are multi-vocal and their meaning is ambiguous.

Together, these claims explain why people participate in contentious rituals that produce collective and divisive outcomes.

To substantiate my argument, I provided qualitative and quantitative evidence that showed that participants primarily understand their behavior non-instrumentally. They are interested in the internal processes, not the external consequences, of participation. Specifically, I identified five process-oriented reasons for acting: collective identity expression, commemoration, tradition, defiance, and the pleasures of participation. External communication with Protestants and Catholics is a key instrumental reason for acting, but did not appear to motivate participation on its own.

To explain how participants seem to ignore the serious external political consequences of their parades, I argued that they understand parades as anti-political. I showed that participants define culture against politics and place parades firmly in the culture category. Their beliefs are sustained by the ritual nature of parades, which provides participants with process-oriented reasons and symbolic ambiguity to maintain their own interpretations.

In sum, the empirical evidence presented in this dissertation supports the view that contentious rituals must be understood as rituals. The importance of loyalist parades in

262 perpetuating sectarian conflict and violence in Northern Ireland shows that rituals have significant material effects on political life. Though I suggested that contentious rituals have similar effects in other divided societies, I did not prove it. In what remains of this dissertation, I will provide short accounts of divisive processions in Jerusalem and India and reflect on the comparisons. !

Expanding the Horizon

The cases of contentious rituals in Jerusalem and India provide fruitful comparisons to

Northern Ireland. In all three places, processions have long histories of fusing nationalism, religion, territory, power, and violence. They also all continue to polarize communities and ignite conflict to this day. !

Israeli Processions in Jerusalem

The most prominent disputed procession in contemporary Jerusalem takes place on

Jerusalem Day (Yom Yerushalayim). On this day, Israeli Jews celebrate the capture of the Old

City and the reunification of Jerusalem during the 1967 Six-Day War. The victory for Israel, however, is understood as a defeat for the city’s Arab residents, who live as second-class citizens.

In recent years, the day’s celebrations include the “Flag Dance” (rikud degalim), a march by tens of thousands of primarily right-wing, religious-nationalist teenagers waving Israeli flags through

263 the city.2 Since 2011, the parades route has included Arab neighborhoods in the Old City and

East Jerusalem.3 Each year since, the parade has been marked by violence between Jews and

Arabs.4 Fights have also broken out between right-wing and left-wing Israeli Jews.5 Issawi Freij, an Israeli Arab Knesset Member, has called the Flag Dance “nothing but a euphemism for a parade of hatred and provocation on the part of thousands of radical right-wing activists in the midst of the Arab neighborhoods,” claiming, “It’s not Jerusalem that celebrants are happy about; it’s belligerence, arrogance and provocation.”6 During the celebration, Palestinian shopkeepers are ordered by the police to close their shops during the march to “prevent friction.”7 Parading past shuttered Arab shops and homes, the celebrants chanted “let your village burn,” “death to

2 Joel Greenberg, “Celebration or Provocation? A Stroll Through the Old City on Jerusalem Day,” Haaretz.com, May 21, 2012. Available at: http://www.haaretz.com/news/features/double-take-celebration-or-provocation-a-stroll- through-the-old-city-on-jerusalem-day-1.431791; and Mitch Ginsburg, “Jerusalem’s Annual Liberation Party Degenerates, Again, From Sweet Fervor to Mini-Rioting,” Times of Israel, May 21, 2012. Available at: http:// www.timesofisrael.com/the-biggest-party-of-the-year-national-religious-teens-march-to-the-western-wall/. Accessed March 28, 2015.

3 Rachel Busbridge, “Frontier Jerusalem: Blurred Separation and Uneasy Coexistence in a Divided City,” Thesis Eleven, Vol. 121, No. 1 (April 2014), p. 77.

4 Omri Efraim, “Violent Clashes Erupt During Jerusalem Day Parade,” Ynetnews.com, June 1, 2011. Available at: http://www.ynetnews.com/articles/0,7340,L-4077124,00.html; Omri Efraim, “Jerusalem: 30,000 Take Part in ‘Flag Dance’ Parade,” Ynetnews.com, May 20, 2012. Available at: http://www.ynetnews.com/articles/ 0,7340,L-4231760,00.html; Gavriel Fiske, “On Jerusalem Day, Clashes and Arrests in Old City,” Times of Israel, May 8, 2013. Available at: http://www.timesofisrael.com/on-jerusalem-day-clashes-and-arrests-in-old-city/ #ixzz3VQhmhXlD; and Jonathan Lis, “Left and Right Play Tug-of-War over Jerusalem Day,” Haaretz.com, May 28, 2014. Available at: http://www.haaretz.com/news/national/.premium-1.595780. Accessed March 28, 2015.

5 Omri Efraim, “Jerusalem: 30,000 Take Part in ‘Flag Dance’ Parade,” Ynetnews.com, May 20, 2012. Available at: http://www.ynetnews.com/articles/0,7340,L-4231760,00.html. Accessed March 28, 2015.

6 Jonathan Lis, “Left and Right Play Tug-of-War over Jerusalem Day,” Haaretz.com, May 28, 2014. Available at: http://www.haaretz.com/news/national/.premium-1.595780. Accessed March 28, 2015.

7 Joel Greenberg, “Celebration or Provocation? A Stroll Through the Old City on Jerusalem Day,” Haaretz.com, May 21, 2012. Available at: http://www.haaretz.com/news/features/double-take-celebration-or-provocation-a-stroll- through-the-old-city-on-jerusalem-day-1.431791. Accessed March 28, 2015.

264 the Arabs,” and “death to all leftists.”8

In addition to the Flag Dance on Jerusalem Day, other contentious rituals reverberate across the city. For instance, to mark the first day of each Hebrew month, hundreds of nationalist religious Israelis circumambulate the Temple Mount/Haram al-Sharif during the “circling the gates” ceremony. During the ritual, which necessitates the closing of streets in the Old City’s

Muslim Quarter, participants dance, sing, and pray for the rebuilding of the Temple.9 There are also the many contested ritual performances in holy places claimed by multiple religious groups.

Attempts at Jewish prayer atop the Temple Mount/Haram al-Sharif have the most consequential political ramifications, but even disputes between Christian denominations over the location of rituals inside the Church of the Holy Sepulcher have erupted in brawls.10

Conflicts over public religious rituals in Jerusalem date to at least the early twentieth century, when significant numbers of Jews began to immigrate. Wasserstein comments that the

“calendar of communal violence in [Mandatory] Palestine was closely bound up with the calendar of religious festivity.”11 It was always especially dangerous when multiple religions celebrated holidays on the same day. The crowds gathered to performed the required holiday

8 Noam Sheizaf, “Watch: Jerusalem Day’s Racist March, Escorted by Police, +972, June 1, 2011. Available at: http://972mag.com/watch-jerusalem-days-racist-march-escorted-by-police/15554/; Nir Hasson, “Right-Wing March to Pass through East Jerusalem, Despite Past Spats,” Haaretz.com, May 15, 2012. Available at: http:// www.haaretz.com/news/national/right-wing-march-to-pass-through-east-jerusalem-despite-past-spats-1.430568. Accessed March 28, 2015.

9 Yizhar Be’er, Dangerous Liaison: The Dynamics of the Rise of the Temple Movements and Their Implications (Jerusalem: Keshev and Ir Amim, 2013), pp. 43-44.

10 Ron E. Hassner, War on Sacred Grounds (Ithaca: Cornell University Press, 2009), p. 76; and Michael Dumper, Jerusalem Unbound: Geography, History, and the Future of the Holy City (New York: Columbia University Press, 2014), p. 112.

11 Bernard Wasserstein, “Patterns of Communal Conflict in Palestine,” in Ada Rapoport-Albert and Steven J. Zipperstein, eds., Jewish History: Essays in Honour of Chimen Abramsky (London: Peter Halban, 1988), p. 611.

265 rituals often met in combat.12 The “clash of dates” continues to cause problems in the city today.13 !

Hindu Processions in India

The role of ritual processions in producing conflict and violence is most pronounced in

India. According to one anthropologist, the “direct connection between ritual performances in public space and riots [in India] seems… obvious.”14 Episodes of violence erupting from religious processions can be found throughout the last three centuries,15 but the connection between ritual and riot deepened in the 1980s. Starting then, Hindu nationalists promoted the use of yatra processions for political mobilization and to unify the Hindu nation.16 In the years that followed, “the religious element almost disappeared from them; they were converted into demonstrations of strength, pure and simple.”17 Jaffrelot argues that since processions build

Hindu solidarity by (momentarily) erasing internal caste boundaries, claim space on behalf of the

Hindu community, and clearly demarcate the in-group in contrast to the out-group, they are “one

12 Ibid.

13 Dumper, Jerusalem Unbound, p. 231.

14 Peter van der Veer, “Riots and Rituals: The Construction of Violence and Public Space in Hindu Nationalism,” in Paul R. Brass, ed., Riots and Pogroms (New York: New York University Press, 1996), p. 155.

15 C. A. Bayly, “The Pre-History of ‘Communalism’? Religious Conflict in India, 1700-1860,” Modern Asian Studies, Vol. 19, No. 2 (1985), pp. 177-203; and Anand A. Yang, “Sacred Symbol and Sacred Space in Rural India: Community Mobilization in the ‘Anti-Cow Killing’ Riot of 1893,” Comparative Studies in Society and History, Vol. 22, No. 4 (October 1980), pp. 576-596.

16 Jackie Assayag, “Ritual Action or Political Reaction? The Invention of Hindu Nationalist Processions in India during the 1980s,” South Asia Research, Vol. 18, No. 2 (1998), pp. 125-146; and Arafaat A. Valiani, “Processions as Publics: Religious Ceremonials and Modes of Public Sphere Intervention in Western India,” unpublished manuscript, Williams College, 2011, p. 5.

17 Christophe Jaffrelot, “The Politics of Processions and Hindu-Muslim Riots,” in Amrita Basu and Atul Kohli, eds., Community Conflict and the State in India (Delhi: Oxford University Press, 1998), p. 71.

266 of the institutions Hindu nationalists are most eager to exploit.”18

Hindu processions are a common occurrence in India, but two are especially notable in the recent history of nationalist politics. The Ekatmata Yatra (“Pilgrimage of One-Soulness”) was invented in 1983 with the explicit goal of uniting India’s Hindus behind “an undiluted version of Hindutva,” or Hindu nationalism.19 Massive processions left different parts of India heading toward the geographic center of the country, carrying water from the sacred Ganges

River that they distributed along the way. The processions were at once displays of religious devotion drawing on elements of traditional Hindu rituals and demonstrations of the political strength of the nationalist movement.20 The Rath Yatra (“Chariot Pilgrimage”) undertaken in

1990 by L.K. Advani, president of the nationalist Bharatiya Janata Party, from Sommath,

Gujarat, to Ayodhya, Uttar Pradesh, 10,000 kilometers away, is probably the most well-known case. The pilgrimage, which was timed to coincide with religious festivals and whose vehicles were adorned with Hindu symbols, was designed to gather support for the campaign to demolish the Babri Mosque in Ayodhya, which was believed to sit on the birthplace of the Hindu god

18 Christophe Jaffrelot, “The Hindu Nationalist Reinterpretation of Pilgrimage in India: The Limits of Yatra Politics,” Nations and Nationalism, Vol. 15, No. 1 (2009), p. 9. Also Steven I. Wilkinson, Votes and Violence: Electoral Competition and Ethnic Riots in India (New York: Cambridge University Press, 2004). On the political implications of the relative egalitarianism of religious rituals in India, see Pradeep K. Chhibber, Religious Practice and Democracy in India (New York: Cambridge University Press, 2014).

19 Jaffrelot, “The Hindu Nationalist Reinterpretation of Pilgrimage,” p. 10.

20 Assayag, “Ritual Action or Political Reaction?” p. 135; Christophe Jaffrelot, The Hindu Nationalist Movement in India (New York: Columbia University Press, 1996), pp. 360-362; and Jaffrelot, “The Hindu Nationalist Reinterpretation of Pilgrimage,” pp. 9-11.

267 Rama.21 Advani’s Rath Yatra triggered many communal riots along its path.22

Beyond these two famous, one-time national processions, there are many local processions across India. This being India, local does not mean small. In the Gujarati city of

Ahmedabad, for example, one annual procession (the Jagganath Rath Yatra or “Chariot

Pilgrimage of Lord Jagganath”) alone has had upwards of 200,000 participants in recent years.23

Each year, the ritual passes through Muslim neighborhoods with heavy security and a curfew on residents.24 According to ethnographic research by Valiani, participants “seemed to take advantage of the thick police cover which surrounded them and taunted the Muslims with catcalls and slogans.”25 This pattern of unwelcome Hindu religious processions entering Muslim neighborhoods is a major trigger of ethnic riots in India.26 !

Concluding Thoughts

A counterintuitive pattern emerges from the comparison of processions by Protestants in

Ulster, Israelis in Jerusalem, and Hindus in India. In all three cases, the provocative rituals are

21 Advani never reached Ayodhya, he was arrested along the way. But 40,000 activists continued without him and stormed the mosque. Thirty people died in ensuing violence. The ashes of the dead were then carried on processions that triggered more riots. Jaffrelot, “The Hindu Nationalist Reinterpretation of Pilgrimage,” p. 13.

22 Jaffrelot, The Hindu Nationalist Movement, pp. 416-419; Assayag, “Ritual Action or Political Reaction?” pp. 137-138; Jaffrelot, “The Politics of Processions,” pp. 81-84; Thomas Blom Hansen, The Saffron Wave: Democracy and Hindu Nationalism in Modern India (Princeton: Princeton University Press, 1999), pp. 164-165; and Jaffrelot, “The Hindu Nationalist Reinterpretation of Pilgrimage,” pp. 11-13

23 Valiani, “Processions as Publics,” p. 29.

24 Ibid., pp. 50-52. At times, the procession has even defied requests by the police and army to avoid Muslim areas. Ornit Shani, Communalism, Caste and Hindu Nationalism: The Violence in Gujarat (Cambridge: Cambridge University Press, 2007), p. 100.

25 Ibid., p. 56.

26 Jaffrelot, “The Politics of Processions”; Wilkinson, Votes and Violence; and Paul R. Brass, The Production of Hindu-Muslim Violence in Contemporary India (Seattle: University of Washington Press, 2005), esp. pp. 364-365.

268 performed by members of the majority ethnic group.27 We might assume that symbolic aggressions are weapons of the weak, something a group does when it lacks material power. But symbolic confrontations are not a substitute for strength; they require strength. As these cases reveal, the ability to carry out mass provocations in public space is a luxury of the powerful. This is because “ritual rights,” like all rights, are not self-enforcing. They “ultimately depend on the authority and coercive capacities of the state.”28 In particular, parading through an hostile area generally requires the permission and protection of the state. In Jerusalem and India, that includes imposing curfews on minority residents; in all three it requires significant policing operations.

Thus one of the ways that provocative processions project dominance is by embodying dominance. They are performed with the state’s sanction and with state-provided security.29

Subordinate groups seeking to provoke are generally granted neither permission nor protection.

As a Palestinian journalists writes: “Imagine for a moment that Palestinians decide to celebrate their heritage in West Jerusalem and march through Jaffa and Ben Yehuda streets! Would they be given full police protection? Would the police dare to ask shops in West Jerusalem to close their doors to reduce tension?”30

This is not to say that minorities in divided societies do not engage in symbolic

27 Though Protestants are no longer the majority in Northern Ireland, they remain the largest group. But parading developed when they demographically and political dominant.

28 Roger Friedland and Richard Hecht, “The Bodies of Nations: A Comparative Study of Religious Violence in Jerusalem and Ayodhya,”History of Religions, Vol. 38, No. 2 (November 1998), p. 112.

29 Dominic Bryan, Orange Parades: The Politics of Ritual, Tradition, and Control (London: Pluto, 2000), p. 95, argues that during the Troubles, opposition to parades mounted not only because of what parades represented to Catholics, but because their performance meant the presence of the reviled police forces.

30 Aziz Abu Sarah, “Palestinians Asked to Close Their Shops for Jerusalem Day,” +972, May 20, 2012. Available at: http://972mag.com/palestinians-asked-to-close-their-shops-for-jerusalem-day/46355/. Accessed March 28, 2015.

269 provocations. They do, but they tend to antagonize through non-collective actions. Prevented from large-scale, mass provocations, minorities use smaller-scale, but no less inflammatory, acts.

For example, in 1982, Sikh militants placed two severed cow heads outside a Hindu temple in

Amritsar, India. The act, performed before dawn, likely took no more than a few people acting under the cover of darkness. But it worked as intended: the desecration triggered rioting.31

Another tentative conclusion we can draw from the comparison is that successful symbolic provocation is the product of “intimate enmity.”32 Effectively antagonizing the other group requires knowing them well: what they hold dear, what is taboo, where is sacred, when they are particularly sensitive to insults. Without this knowledge, attempts to enrage could fall flat—desecrating a profane place will prompt little more than a shrug. But with the right knowledge—gained from the continuous interaction that comes living in close proximity

—“symbolic challenges… can rapidly spawn a spiraling tornado of violence.”33 Like an unhappy married couple, groups in divided societies know how to push each others’ buttons like no one else. So in addition to rioting after discovering the cow heads, Hindus flung cigarettes into Sikh holy places, knowing that tobacco is taboo for Sikhs.34

While these actions are characterized by intimacy, they lack empathy. Despite, or perhaps because of, the proximity of the ethnic communities in divided societies, many people seem

31 Stanley Tambiah, Leveling Crowds: Ethnonationalist Conflicts and Collective Violence in South Asia (Berkeley: University of California Press, 1996), p. 236.

32 I have seen the term in Meron Benvenisti, Intimate Enemies: Jews and Arabs in a Shared Land (Berkeley: University of California Press, 1995), p. 82; and Naveeda Khan, “The Acoustics of Muslim Striving: Loudspeaker Use in Ritual Practice in Pakistan,” Comparative Studies in Society and History, Vol. 53, No. 3 (July 2011), p. 579. Without using the term, Marc Gaborieau, “From Al-Beruni to Jinnah: Idiom, Ritual and Ideology of the Hindu- Muslim Confrontation in South Asia,” Anthropology Today, Vol. 1, No. 3 (June 1985), pp. 9-10, describes the concept in great detail.

33 Tambiah, Leveling Crowds, p. 236.

34 Ibid.

270 unable or unwilling to empathize across the ethnic boundary. There is an unwillingness to recognize that a ritual that is so wonderful for you may be so terrible for someone else. Returning to Northern Ireland, neither Protestant paraders nor Catholic protesters seem prepared to imagine the significant pain they impose on the other. Most paraders are unprepared to accept that marching by Catholic homes and churches causes real hurt for many Catholics, for whom parades are degrading symbols of hate. Most protesters are unprepared to accept that stopping parades causes real hurt for many Protestants, for whom parades are a deeply meaningful ritual inseparable from collective and personal identities, as well as cherished moments in life. A degree of empathy might break the cycle of mutual antagonizing and create the conditions for true and lasting peace.

271 ! Bibliography !Adams, Gerry. Free Ireland: Towards a Lasting Peace. New York: Penguin, 1995. Ahuja, Amit, and Pradeep Chhibber. “Why the Poor Vote in India: ‘If I Don’t Vote, I Am Dead to ! the State.” Studies in Comparative International Development 47, no. 4 (2012): 389-410. Alexander, Jeffrey C. “Cultural Pragmatics: Social Performance Between Ritual and Strategy.” ! Sociological Theory 22, no. 4 (2004): 527–573. Allen, Lori A. “The Polyvalent Politics of Martyr Commemorations in the Palestinian Intifada.” ! History & Memory 18, no. 2 (2006): 107-138. Aminzade, Ron, and Doug McAdam. “Emotions and Contentious Politics.” In Silence and Voice in the Study of Contentious Politics, edited by Ronald R. Aminzade, Jack A. Goldstone, Doug McAdam, Elizabeth J. Perry, William H. Sewell, Sidney Tarrow, and Charles Tilly, 14-50. ! New York: Cambridge University Press, 2001. Anderson, Benedict. Imagined Communities: Reflections on the Origin and Spread of ! Nationalism, rev. ed. London: Verso, 1991. Arendt, Hannah. The Human Condition, 2nd ed. Chicago: University of Chicago Press, 1998.

!ARK. “Northern Ireland Life and Times Survey 2003.” ARK. http://www.ark.ac.uk/nilt/2003. !ARK. “Northern Ireland Life and Times Survey 2005.” ARK. http://www.ark.ac.uk/nilt/2005. !ARK. “Northern Ireland Life and Times Survey 2009.” ARK. http://www.ark.ac.uk/nilt/2009. !ARK. “Northern Ireland Life and Times Survey 2013.” ARK. http://www.ark.ac.uk/nilt/2013. !Arthur, Paul. Government and Politics of Northern Ireland, 2nd ed. London: Longman, 1984. Assayag, Jackie. “Ritual Action or Political Reaction? The Invention of Hindu Nationalist ! Processions in India during the 1980s.” South Asia Research 18, no. 2 (1998): 125-146. Bachrach, Peter, and Morton S. Baratz. “Two Faces of Power.” American Political Science ! Review 56, no. 4 (1962): 947-952. !Banfield, Edward C. The Unheavenly City Revisited. Boston: Little, Brown and Company, 1974. !Bardon, Jonathan. A History of Ulster. Belfast: Blackstaff, 1992.

272 Barnett, Michael. Empire of Humanity: A History of Humanitarianism. Ithaca: Cornell ! University Press, 2011. Barretto, Amilcar Antonio. “Nationalism, Collective Action, and Rationality.” Nationalism and ! Ethnic Politics 18, no. 3 (2012): 316-336. Baumeister, Roy F., and Mark R. Leary. “The Need to Belong: Desire for Interpersonal Attachments as a Fundamental Human Motivation.” Psychological Bulletin 117, no. 3 ! (1995): 497-529. Bayly, C. A. “The Pre-History of ‘Communalism’? Religious Conflict in India, 1700-1860.” ! Modern Asian Studies 19, no. 2 (1985): 177-203. Beaman, Lori G. “Battles over Symbols: The ‘Religion’ of the Minority versus the ‘Culture’ of ! the Majority.” Journal of Law & Religion 28, no. 1 (2012): 67-104. Be’er, Yizhar. Dangerous Liaison: The Dynamics of the Rise of the Temple Movements and Their ! Implications. Jerusalem: Keshev and Ir Amim, 2013. Beissinger, Mark R. “Nationalist Violence and the State: Political Authority and Contentious ! Repertoires in the Former USSR.” Comparative Politics 30, no. 4 (1998): 401-422. Beissinger, Mark R. Nationalist Mobilization and the Collapse of the Soviet State. New York: !Cambridge University Press, 2002. Belfast Interface Project. Belfast Interfaces: Security Barriers and Defensive Use of Space. !Belfast: Belfast Interface Project, 2012. !Bell, Catherine. Ritual Theory, Ritual Practice. New York: Oxford University Press, 1992. Bell, Catherine. Ritual: Perspectives and Dimensions, rev. ed. New York: Oxford University ! Press, 2009. Bell, John. For God, Ulster or Ireland? Religion, Society and Security in Northern Ireland. ! Belfast: Institute for Conflict Research, 2013. Bellah,! Robert N. “Civil Religion in America.” Dædalus 96, no. 1 (1967): 1-21. Bellah, Robert N. “The Ritual Roots of Society and Culture.” In Handbook of the Sociology of ! Religion, edited by Michelle Dillon, 31-44. New York: Cambridge University Press, 2003. Bender, Courtney. Heaven’s Kitchen: Living Religion at God’s Love We Deliver. Chicago: University of Chicago Press, 2003.

273 ! Benvenisti, Meron. Intimate Enemies: Jews and Arabs in a Shared Land. Berkeley: University of ! California Press, 1995. Berezin, Mabel. Making the Fascist Self: The Political Culture of Interwar Italy. Ithaca: Cornell ! University Press, 1997. Berman, Eli. “Sect, Subsidy, and Sacrifice: An Economist’s View of Ultra-Orthodox Jews.” ! Quarterly Journal of Economics 115, no. 3 (2000): 905-953. !Bernard, H. Russell. Research Methods in Anthropology, 5th ed. Lanham, MD: AltaMira, 2011. Bernstein, Mary. “Celebration and Suppression: The Strategic Uses of Identity by the Lesbian ! and Gay Movement,” American Journal of Sociology 103, no. 3 (1997): 531-565. Bew, Paul, Peter Gibbon, and Henry Patterson. Northern Ireland, 1921-1996: Political Forces ! and Social Classes. London: Serif, 1996. !Billig, Michael. Banal Nationalism. London: Sage, 1995. Billings, Dwight B. “Religion as Opposition: A Gramscian Analysis.” American Journal of !Sociology 96, no. 1 (1990): 1-31. Blee, Kathleen. Inside Organized Racism: Women in the Hate Movement. Berkeley: University ! of California Press, 2002. Blee, Kathleen, and Amy McDowell. “Social Movement Audiences.” Sociological Forum 27, no. ! 1 (2012): 1-20. Blee, Kathleen M., and Verta Taylor. “Semi-Structured Interviewing in Social Movement Research.” In Methods of Social Movement Research, edited by Bert Klandermans and !Suzanne Staggenborg, 92-117. Minneapolis: University of Minnesota Press, 2002. Bocock, Robert. Ritual in Industrial Society: A Sociological Analysis of Ritualism in Modern ! England. London: George Allen & Unwin, 1974. !Boobbyer, Philip. Conscience, Dissent and Reform in Soviet Russia. New York: Routledge, 2005. !Bourke, Richard. Peace in Ireland: The War of Ideas. London: Pimlico, 2003. Boyer, Pascal and Pierre Liénard. “Why Ritualized Behavior? Precaution Systems and Action Parsing in Developmental, Pathological and Cultural Rituals.” Behavioral and Brain Sciences 29, no. 6 (2006): 635- 641.

274

Brass, Paul R. Theft of an Idol: Text and Context in the Representation of Collective Violence. ! Princeton: Princeton University Press, 1997. Brass, Paul R. The Production of Hindu-Muslim Violence in Contemporary India. Seattle: ! University of Washington Press, 2005. Brennan, Geoffrey and Loren Lomasky. Democracy and Decision: The Pure Theory of Electoral ! Preference. New York: Cambridge University Press, 1993. !Brewer, John D. Peace Processes: A Sociological Approach. Massachussetts: Polity, 2010. Brewer, John D., Gareth I. Higgins, and Francis Teeney. Religion, Civil Society, and Peace in ! Northern Ireland. New York: Oxford University Press, 2011. Brown, Wendy. “‘The Most We Can Hope For…’: Human Rights and the Politics of Fatalism.” ! South Atlantic Quarterly 103, no. 2/3 (Spring/Summer 2004): 451-463. Brubaker, Roger. Nationalism Reframed: Nationhood and the National Question in the New ! Europe. New York: Cambridge University Press, 1996. !Brubaker, Rogers. Ethnicity without Groups. Cambridge: Harvard University Press, 2006. Brubaker, Rogers, Margit Feischmidt, Jon Fox, and Liana Grancea. Nationalist Politics and ! Everyday Ethnicity in a Transylvanian Town. Princeton: Princeton University Press, 2006. Brubaker, Rogers, and David D. Laitin. “Ethnic and Nationalist Violence.” Annual Review of ! Sociology 24 (1998): 423-452. Bruce, Steve. Paisley: Religion and Politics in Northern Ireland. New York: Oxford University !Press, 2007. Bryan, Dominic. “Titanic Town: Living in a Landscape of Conflict.” In Belfast 400: People, Place, and History, edited by S.J. Connolly, 317-353. Liverpool: Liverpool University Press, !2012. Bryan, Dominic. Orange Parades: The Politics of Ritual, Tradition, and Control. London: Pluto, ! 2000. Bryan, Dominic. “Parade Disputes and the Peace Process.” Peace Review 13, no. 1 (2001): ! 43-49.

275 Bryan, Dominic, Mike Cronin, Tina O’Toole, and Catriona Pennell. “Ireland’s Decade of ! Commemorations: A Roundtable.” New Hibernia Review 17, no. 3 (2013): 63-86. Busbridge, Rachel. “Frontier Jerusalem: Blurred Separation and Uneasy Coexistence in a ! Divided City.” Thesis Eleven 121, no. 1 (2014): 76-100. Calhoun, Craig Jackson. “The Radicalism of Tradition: Community Strength or Venerable Disguise and Borrowed Language?” American Journal of Sociology 88, no. 5 (1983): !886-914. Canetti, Elias. Crowds and Power. Translated by Carol Stewart. New York: Farrar, Straus, and !Giroux, 1984. Caplow, Theodore. “Rule Enforcement Without Visible Means: Christmas Gift Giving in !Middletown.” American Journal of Sociology 89, no. 6 (1984): 1306-1323. Carey, James. “The Dark Continent of American Journalism.” In James Carey: A Critical Reader, edited by Eve Stryker Munson and Catherine A. Warren, 144-190. Minneapolis: !University of Minnesota Press, 1997. Casquete, Jesus. “The Power of Demonstrations.” Social Movement Studies 5, no. 1 (2006): ! 45-60. Casserly, Ray. “The Fyfe and My Family: Flute Bands in Rathcoole Estate.” Irish Journal of !Anthropology 13 (2010): 8-12. Casserly, Ray. “Blood, Thunder, and Drums: Style and Changing Aesthetics of Drumming in ! Northern Ireland Protestant Bands.” Yearbook for Traditional Music 45 (2013): 142-63. !Casserly, Ray. “Parading Music and Memory in Northern Ireland.” Música e Cultura 9 (2014). Chandra, Kanchan. “Introduction.” In Constructivist Theories of Ethnic Politics, edited by !Kanchan Chandra, 1-50. New York: Oxford University Press, 2012. Chhibber, Pradeep K. Religious Practice and Democracy in India. New York: Cambridge ! University Press, 2014. Chong, Dennis. Collective Action and the Civil Rights Movement. Chicago: University of ! Chicago Press, 1991. Chwe, Michael Suk-Young. Rational Ritual: Culture, Coordination, and Common Knowledge. ! Princeton: Princeton University Press, 2001.

276 !Cohen, Anthony P. The Symbolic Construction of Community. London: Routledge, 1985. Collier, Paul, and Anke Hoeffler. “Greed and Grievance in Civil War.” Oxford Economic Papers ! 56 (2004): 563-595. Collins, Randall. “Interaction Ritual Chains and Collective Effervescence.” In Collective Emotions, edited by Christian von Scheve and Mikko Salmela, 299-311. New York: Oxford ! University Press, 2014. Collins, Randall. “Social Movements and the Focus of Emotional Attention.” In Passionate Politics: Emotions and Social Movements, edited by Jeff Goodwin, James M. Jasper, and ! Francesca Polletta, 27-43. Chicago: University of Chicago Press, 2001. !Collins, Randall. Interaction Ritual Chains. Princeton: Princeton University Press, 2004. Collins, Randall. Violence: A Micro-Sociological Theory. Princeton: Princeton University Press, ! 2008. !Connerton, Paul. How Societies Remember. New York: Cambridge University Press, 1989. Corrigal-Brown, Catherine. Patterns of Protest: Trajectories of Participation in Social ! Movements. Stanford: Stanford University Press, 2011. Corstange, Daniel. “Religion, Pluralism, and Iconography in the Public Sphere: Theory and !Evidence From Lebanon.” World Politics 64, no. 1 (2012): 116-160. Cox, Harvey. “Keeping Going: Beyond Good Friday.” In The Long Road to Peace in Northern Ireland: Peace Lectures from the Institute of Irish Studies at Liverpool University, rev. ed., ! edited by Marianne Elliott, 157-174. Liverpool: Liverpool University Press, 2007. !Dahl, Robert A. “The Concept of Power.” Behavioral Science 2, no. 3 (1957): 201-215. Davis, Natalie Zemon. “Rites of Violence: Religious Riot in Sixteenth-Century France.” Past & ! Present 59 (1973): 51-91. Davis, Susan G. Parades and Power: Street Theater in Nineteenth-Century Philadelphia. !Philadelphia: Temple University Press, 1986. de Figueiredo, Jr., Rui J.P. and Barry R. Weingast. “The Rationality of Fear: Political Opportunism and Ethnic Conflict.” In Civil Wars, Insecurity, and Intervention, edited by !Barbara F. Walter and Jack Snyder, 261-302. New York: Columbia University Press, 1999.

277 de Mesquita, Ethan Bueno, and Eric S. Dickson. “The Propaganda of the Deed: Terrorism, Counterterrorism, and Mobilization.” American Journal of Political Science 51 (2007): !364-81. della Porta, Donatella. “Recruitment Processes in Clandestine Political Organizations: Italian Left-Wing Terrorism.” In International Social Movement Research, Vol. 1, edited by Bert Klandermans, Hanspeter Kriesi, and Sidney Tarrow, 155-169. Greenwich, CT: JAI Press, !1988. Department of Culture, Arts and Leisure. “Marching Bands in Northern Ireland.” Belfast: ! Department of Culture, Arts and Leisure, 2012. Diefendorf, Barbara B. Beneath the Cross: Catholics and Huguenots in Sixteenth-Century Paris. ! New York: Oxford University Press, 1991. Dingley, James. “Marching Down the Garvaghy Road: Republican Tactics and State Response to the Orangemen’s Claim to March their Traditional Route Home after the Drumcree Church ! Service.” Terrorism and Political Violence 14, no. 3 (2002): 42-79. !Dirks, Robert. “Annual Rituals of Conflict.” American Anthropologist 90, no. 4 (1988): 856-870. !Douglas, Mary. Natural Symbols: Explorations in Cosmology. London: Routledge, 1970. Douglas, Mary. Purity and Danger: An Analysis of the Concepts of Pollution and Taboo. ! London: Routledge, 1996. !Downs, Anthony. An Economic Theory of Democracy. New York: Harper, 1957. Doyle, Mark. Fighting Like the Devil for the Sake of God: Protestants, Catholics and the Origins ! of Violence in Victorian Belfast. Manchester: Manchester University Press, 2009. Dumper, Michael. Jerusalem Unbound: Geography, History, and the Future of the Holy City. ! New York: Columbia University Press, 2014. Durkheim, Émile. The Elementary Forms of Religious Life. Translated by Karen E. Fields. New ! York: Free Press, 1995 [1912]. Edelman, Murray. The Symbolic Uses of Politics, rev. ed. Champaign: University of Illinois ! Press, 1985. Einwohner, Rachel L. “Opportunity, Honor, and Action in the Warsaw Ghetto Uprising of 1943.” ! American Journal of Sociology 109, no. 3 (2003): 650-75.

278 Einwohner, Rachel L. “Identity Work and Collective Action in a Repressive Context: Jewish Resistance on the ‘Aryan Side’ of the Warsaw Ghetto.” Social Problems 53, no. 1 (2006): ! 38-56. Eisen, Arnold M. “Constructing the Usable Past: The Idea of Tradition in Twentieth-Century American Judaism.” In The Uses of Tradition: Jewish Continuity in the Modern Era, edited !by Jack Wertheimer, 429-461. New York: Jewish Theological Seminary of America, 1992. Eliade, Mircea. Patterns in Comparative Religion. Translated by Rosemary Sheed. New York: ! Sheed & Ward, 1958. Eliade, Mircea. The Sacred and the Profane: The Nature of Religion. Translated by Willard R. !Trask. Orlando: Harcourt, 1959. Eliasoph, Nina. Avoiding Politics: How Americans Produce Apathy in Everyday Life. New York: ! Cambridge University Press, 1998. Eliasoph, Nina, and Paul Lichterman. “Making Things Political.” In Handbook of Cultural Sociology, edited by John R. Hall, Laura Grindstaff, and Lo, Ming-Cheng, 483-493. New !York: Routledge, 2010. Elster, Jon. “Marxism, Functionalism, and Game Theory: The Case for Methodological ! Individualism.” Theory and Society 11, no. 4 (1982): 453-540. Elster, Jon. The Cement of Society: A Study of Social Order. New York: Cambridge University ! Press, 1989. Emerson, Robert M., Rachel I. Fretz, and Linda L. Shaw. Writing Ethnographic Fieldnotes. ! Chicago: University of Chicago Press, 1995. Etzioni, Amitai. “Toward a Theory of Public Ritual.” Sociological Theory 18, no. 1 (2000): ! 44-59. Etzioni, Amitai and Jared Bloom, eds. We Are What We Celebrate: Understanding Holidays and ! Rituals. New York: NYU Press, 2004. Evans, Jocelyn A.J. and Jonathan Tonge. “Unionist Party Competition and the Orange Order ! Vote in Northern Ireland.” Electoral Studies 26 (2007): 156-167. Evershed, Jonathan. “Beyond What Actually Happened: Loyalist Spectral Politics and the Problematic Privileging of ‘History’ during Northern Ireland’s Decade of Centenaries.” Irish ! Journal of Anthropology, forthcoming.

279 Eyerman, Ron, and Andrew Jamison. Music and Social Movements: Mobilizing Traditions in the ! Twentieth Century. New York: Cambridge University Press, 1998. !Farrell, Michael. Northern Ireland: The Orange State. London: Pluto Press, 1976. Farrell, Sean. Rituals and Riots: Sectarian Violence and Political Culture in Ulster, 1784–1886. ! Lexington: The University Press of Kentucky, 2000. Fearon, James D. and David D. Laitin.“Explaining Interethnic Cooperation.” American Political ! Science Review 90, no. 4 (1996): 715-735. Fearon, James D., and David D. Laitin. “Ethnicity, Insurgency, and Civil War.” American ! Political Science Review 97, no. 1 (2003): 75-90. Feldman, Allen. The Formations of Violence: The Narrative of the Body and Political Terror in ! Northern Ireland. Chicago: University of Chicago Press, 1991. Ferguson, James. The Anti-Politics Machine:“Development,” Depoliticization, and Bureaucratic !Power in Lesotho. Minneapolis: University of Minnesota Press, 1990. Fillieule, Olivier and Danielle Tartakowsky. Demonstrations. Translated by Phyllis Aronoff and !Howard Scott. Halifax: Fernwood, 2013. Fillieule, Olivier. “The Independent Psychological Effects of Participation in Demonstrations.” ! Mobilization 17, no. 3 (2012): 235-248. Fisher III, Samuel H. and Rebekah Herrick. “Old versus New: The Comparative Efficiency of Mail and Internet Surveys of State Legislators.” State Politics and Policy Quarterly 13, no. 2 ! (2013): 147-163. !Flam, Helena, and Debra King, eds. Emotions and Social Movements. London: Routledge, 2005. Fominaya, Cristina Flesher. “Collective Identity in Social Movements: Central Concepts and ! Debates,” Sociology Compass 4, no. 6 (2010): 393-404. !Fowler Jr., Floyd J. Survey Research Methods, 4th ed. Thousand Oaks, CA: Sage, 2009. Fox, Jon E., and Cynthia Miller-Idriss. “Everyday Nationhood.” Ethnicities 8, no. 4 (2008): ! 536-563. Frampton, Martyn. The Long March: The Political Strategy of Sinn Féin, 1981-2007. London: ! Palgrave Macmillan, 2000.

280 Fraser, T.G. “The Apprentice Boys and the Relief of Derry Parades.” In The Irish Parading Tradition: Following the Drum, edited by T.G. Fraser, 173-90. London: Palgrave Macmillan, ! 2000. Friedland Roger, and Richard Hecht. “The Bodies of Nations: A Comparative Study of Religious ! Violence in Jerusalem and Ayodhya.” History of Religions 38, no. 2 (1998): 101-149. Friedman, Debra, and Doug McAdam. “Collective Identity and Activism: Networks, Choices, and the Life of a Social Movement.” In Frontiers in Social Movement Theory, edited by ! Aldon D. Morris and Carol M. Mueller, 156-173. New Haven: Yale University Press, 1992. Fujii, Lee Ann. “The Puzzle of Extra-Lethal Violence.” Perspectives on Politics 11, no. 2 (2013): ! 410-426. Gaborieau, Marc. “From Al-Beruni to Jinnah: Idiom, Ritual and Ideology of the Hindu-Muslim ! Confrontation in South Asia.” Anthropology Today 1, no. 3 (1985): 7-14. Gagnon, V.P. The Myth of Ethnic War: Serbia and Croatia in the 1990’s. Ithaca: Cornell ! University Press, 2004. Ganiel, Gladys. “Surveying Religion’s Public Role: Perspectives on Reconciliation, Diversity ! and Ecumenicism in Northern Ireland.” Shared Space 9 (2010): 53-68. Ganiel, Gladys. 21st Century Faith – Results of the Survey of Clergy, Pastors, Ministers and ! Faith Leaders. Belfast: Irish School of Ecumenics, n.d. Gaventa, John. Power and Powerlessness: Quiescence and Rebellion in an Appalachian Valley. ! Champaign: University of Illinois Press, 1980. Geertz, Clifford. Negara: The Theatre State in 19th Century Bali. Princeton: Princeton ! University Press, 1980. !Gellner, Ernest. Nations and Nationalism. Ithaca: Cornell University Press, 1983. Gibbon, Peter. The Origins of Ulster Unionism: The Formation of Popular Protestant Politics and Ideology in Nineteenth-Century Ireland. Manchester: Manchester University Press, !1975. Gillis, John R. “Memory and Identity: The History of a Relationship.” In Commemorations: The Politics of National Identity, edited by John Gillis, 3-24. Princeton: Princeton University ! Press, 1996.

281 Gillis, John R., ed. Commemorations: The Politics of National Identity. Princeton: Princeton !University Press, 1996. Girard, René. Violence and the Sacred. Translated by Patrick Gregory. Baltimore: Johns Hopkins ! University Press, 1977. Gluckman, Max. Rituals of Rebellion in South-East Africa. Manchester: Manchester University ! Press, 1954. Gluckman, Max. “Les Rites de Passage.” In Essays on the Ritual of Social Relations, edited by ! Max Gluckman, 1-52. Manchester: Manchester University Press, 1962. !Gluckman, Max. Politics, Law and Ritual in Tribal Society. Oxford: Basil Blackwell, 1965. Goffman, Erving. Interaction Rituals: Essays on Face-to-Face Behavior. New York: Pantheon, ! 1967. Goodall, David. “Hillsborough to Belfast: Is It the Final Lap?” In The Long Road to Peace in Northern Ireland: Peace Lectures from the Institute of Irish Studies at Liverpool University, ! rev. ed., edited by Marianne Elliott, 124-132. Liverpool: Liverpool University Press, 2007. Goodin,! Robert E. “Rites of Rulers.” British Journal of Sociology 29, no. 3 (1978): 281-299. Goodwin, Jeff. “The Libidinal Constitution of a High-Risk Social Movement: Affectual Ties and Solidarity in the Huk Rebellion, 1946 to 1954.” American Sociological Review 62, no. 1 ! (1997): 53-69. Goodwin, Jeff, James M. Jasper, and Francesca Polletta, eds. Passionate Politics: Emotions and ! Social Movements. Chicago: University of Chicago Press, 2001. Goodwin, Jeff and Steven Pfaff. “Emotion Work in High-Risk Social Movements: Managing Fear in the U.S. and East German Civil Rights Movements.” In Passionate Politics: Emotions and Social Movements, edited by Jeff Goodwin, James M. Jasper, and Francesca !Polletta, 282-302. Chicago: University of Chicago Press, 2001. Goody, Jack. “Religion and Ritual: The Definitional Problem.” British Journal of Sociology 12, ! no. 2 (1961): 142-164. Gorenberg, Gershom. The End of Days: Fundamentalism and the Struggle for the Temple Mount. ! New York: Free Press, 2000. Gould, Deborah B. Moving Politics: Emotion and ACT UP’s Fight against AIDS. Chicago: University of Chicago Press, 2009.

282 ! Green, Donald P., and Ian Shapiro. Pathologies of Rational Choice: A Critique of Applications in ! Political Science. New Haven: Yale University Press, 1994. !Gurr, Ted Robert. Why Men Rebel. Princeton: Princeton University Press, 1970. Hall, Michael, ed. ‘It’s Good To Talk’: The Experiences of the Springfield Mobile Phone ! Network. Newtownabbey, Northern Ireland: Island Publications, 2003. Hall, Michael. Towards a Shared Future (5): Ulster’s Marching Bands. Belfast: Island ! Publications, 2014. Hamilton, Michael. Working Relationships: An Evaluation of the Community Mobile Phone ! Networks in Northern Ireland. Belfast: Community Relations Council, 2001. Hamlin, Alan, and Colin Jennings. “Expressive Political Behaviour: Foundations, Scope and ! Implications.” British Journal of Political Science 41, no. 3 (2011): 645-670. Hansen, Thomas Blom. The Saffron Wave: Democracy and Hindu Nationalism in Modern India. ! Princeton: Princeton University Press, 1999. Hardin, Russell. One for All: The Logic of Group Conflict. Princeton: Princeton University Press, ! 1995. Harris, Frederick C., and Daniel Gillion. “Expanding the Possibilities: Reconceptualizing Political Participation as a Toolbox.” In The Oxford Handbook of American Elections and Political Behavior, edited by Jan Leighley, 144-161. New York: Oxford University Press, ! 2010. !Hassner, Ron E. War on Sacred Grounds. Ithaca: Cornell University Press, 2009. Hassner, Ron E. “Sacred Time and Conflict Initiation.” Security Studies 20, no. 4 (2011): ! 491-520. !Hay, Colin. Why We Hate Politics. London: Polity, 2007. Heaney, Seamus. Opened Ground: Selected Poems, 1966-1996. New York: Farrar, Straus, and ! Giroux, 1998. Hechter, Michael. Principles of Group Solidarity. Berkeley: University of California Press, 1987.

283 Heerwegh, Dirk, and Geert Loosveldt. “Personalizing E-mail Contacts: Its Influence on Web Survey Response Rate and Social Desirability Response Bias.” International Journal of ! Public Opinion Research 19, no. 2 (2007): 258-68. Helmke, Gretchen, and Steven Levitsky. “Informal Institutions and Comparative Politics: A ! Research Agenda.” Perspectives on Politics 2, no. 4 (2004): 725-39. Helmke, Gretchen, and Steven Levitsky, eds., Informal Institutions and Democracy: Lessons ! from Latin America. Baltimore: Johns Hopkins University Press, 2006. Hempton, David and Myrtle Hull. Evangelical Protestantism in Ulster Society 1740-1890. ! London: Routledge, 1992. Henry, P. J., and David O. Sears. “The Symbolic Racism 2000 Scale.” Political Psychology 23, ! no. 2 (2002): 253-283. Hill, Jacqueline R. “National Festivals, the State and ‘Protestant Ascendancy’ in Ireland, ! 1790-1829.” Irish Historical Studies 24, no. 93 (1984): 30-51. Hirschman, Albert O. Shifting Involvements: Private Interest and Public Action. Princeton: ! Princeton University Press, 1982. Hobsbawm, Eric. “Inventing Traditions.” In The Invention of Tradition, edited by Eric !Hobsbawm and Terence Ranger, 1-14. Cambridge: Cambridge University Press, 1983. Hobsbawm, E.J. Nations and Nationalism Since 1780: Programme, Myth, Reality. New York: ! Cambridge University Press, 1990. Hobsbawm, Eric, and Terence Ranger, eds., The Invention of Tradition. Cambridge: Cambridge !University Press, 1983. Hollinger, David A. “Religious Ideas: Should They Be Critically Engaged or Given a Pass?” ! Representations 101, no. 1 (2008): 144-154. Holmes, R. “Covenanter Politics: Evangelicalism, Political Liberalism and Ulster Presbyterians, ! 1798-1914.” English Historical Review 125, no. 513 (2010): 340-369. Honaker, James, and Gary King. “What to Do about Missing Values in Time-Series Cross- !Section Data.” American Journal of Political Science 54, no. 3 (2010): 561-581. Honaker, James, Gary King, and Matthew Blackwell. “AMELIA II: A Program for Missing !Data.” Harvard University, 2012.

284 Hopf, Ted. “The Logic of Habit in International Relations.” European Journal of International ! Relations 16, no. 4 (2010): 539-61. Horowitz, Donald L. Ethnic Groups in Conflict, 2nd. ed. Berkeley: University of California ! Press, 2000. !Horowitz, Donald L. The Deadly Ethnic Riot. Berkeley: University of California Press, 2001. Horowitz, Elliott. Reckless Rites: Purim and the Legacy of Jewish Violence. Princeton: Princeton ! University Press, 2008. Humphrey, Caroline, and James Laidlaw. The Archetypal Actions of Ritual: A Theory of Ritual ! Illustrated by Jain Rite of Worship. Oxford: Claredon Press, 1994. Humphrey, Macartan, and Jeremy M. Weinstein. “Who Fights? The Determinants of ! Participation in Civil War.” American Journal of Political Science 52, no. 2 (2008): 436-455. Hunt, Lynn. Politics, Culture and Class in the French Revolution. Berkeley: University of ! California Press, 1986. Hunt, Scott A., and Robert D. Benford. “Collective Identity, Solidarity, and Commitment.” In The Blackwell Companion to Social Movements, edited by David A. Snow, Sarah A. Soule, ! and Hanspeter Kriesi, 433-457. Malden, Mass.: Blackwell, 2004 Iannacone, Laurence R. “Sacrifice and Stigma: Reducing Free-riding in Cults, Communes, and ! Other Collectives.” Journal of Political Economy 100, no. 2 (1992): 271-91. Iannacone, Laurence R. “Why Strict Churches Are Strong.” American Journal of Sociology 99, ! no. 5 (1994): 1180-1211. Irons, William. “Religion as a Hard-to-Fake Sign of Commitment.” In The Evolution of Commitment, edited by Randolph Nesse, 292-309. New York: Russell Sage Foundation, ! 2001. Jackson, Alvin. Home Rule: An Irish History, 1800-2000. New York: Oxford University Press, ! 2003. Jaffrelot, Christophe. The Hindu Nationalist Movement in India. New York: Columbia University ! Press, 1996. Jaffrelot, Christophe. “The Politics of Processions and Hindu-Muslim Riots.” In Community Conflict and the State in India, edited by Amrita Basu and Atul Kohli, 58-92. Delhi: Oxford

285 !University Press, 1998. Jaffrelot, Christophe. “The Hindu Nationalist Reinterpretation of Pilgrimage in India: The Limits ! of Yatra Politics.” Nations and Nationalism 15, no. 1 (2009): 1-19. Jarman, Neil. Material Conflicts: Parades and Visual Displays in Northern Ireland. Oxford: ! Berg, 1997. Jarman, Neil, and Dominic Bryan. Parade and Protest: A Discussion of Parading Disputes in ! Northern Ireland. Colraine, UK: Centre for the Study of Conflict, University of Ulster, 1996. Jasper, James M. The Art of Moral Protest: Culture, Biography, and Creativity in Social ! Movements. Chicago: University of Chicago Press, 1997. Jasper, James M. “The Emotions of Protest: Affective and Reactive Emotions In and Around ! Social Movements.” Sociological Forum 13, no. 3 (1998): 397-424. Jasper, James M. “A Strategic Approach to Collective Action: Looking for Agency in Social- ! Movement Choices.” Mobilization: An International Quarterly 9, no. 1 (2004): 1-16. Jasper, James M. “Emotions and Social Movements: Twenty Years of Theory and Research.” ! Annual Review of Sociology 37 (2011): 285-303. Jervis, Robert. Perception and Misperception in International Politics. Princeton: Princeton ! University Press, 1976. Johnston, Hank. “New Social Movements and Old Regional Nationalisms.” In New Social Movements: From Ideology to Identity, edited by Enrique Laraña, Hank Johnston, and Joseph ! R. Gusfield, 267-286. Philadelphia: Temple University Press, 1994. Kakar, Sudhir. The Colors of Violence: Cultural Identities, Religion, and Conflict. Chicago: ! University of Chicago Press, 1996. Kalyvas, Stathis N. “The Ontology of ‘Political Violence’: Action and Identity in Civil Wars.” ! Perspectives on Politics 1, no. 3 (2003): 475-494. Kalyvas, Stathis N. The Logic of Violence in Civil War. New York: Cambridge University Press, ! 2006. Kanter, Rosabeth Moss. Commitment and Community: Communes and Utopias in Sociological ! Perspective. Cambridge: Harvard University Press, 1972.

286 Kaplowitz, Michael D., Timothy D. Hadlock, and Ralph Levine. “A Comparison of Web and ! Mail Survey Response Rates.” Public Opinion Quarterly 68, no. 1 (2004): 94-101. Kaufmann, Eric P. The Orange Order: A Contemporary Northern Irish History. New York: ! Oxford University Press, 2007. Kaufmann, Eric. “Demographic Change and Conflict in Northern Ireland: Reconciling ! Qualitative and Quantitative Evidence." Ethnopolitics 10, no. 3-4 (2011): 369-389. Kearney, Hugh F. Ireland: Contested Ideas of Nationalism and History. New York: NYU Press, ! 2007. Kelly, James. “The Emergence of Political Parading, 1660-1800.” In The Irish Parading Tradition: Following the Drum, edited by T.G. Fraser, 9-26. London: Palgrave Macmillan, ! 2000. !Kertzer, David I. Ritual, Politics, and Power. New Haven: Yale University Press, 1988. Khan, Naveeda. “The Acoustics of Muslim Striving: Loudspeaker Use in Ritual Practice in ! Pakistan.” Comparative Studies in Society and History 53, no. 3 (2011): 571-594. !King, Charles. “The Micropolitics of Social Violence.” World Politics 56, no. 3 (2004): 431-455. King, Gary, and Langche Zeng. “Logistic Regression in Rare Events Data.” Political Analysis 9, ! no. 2 (2001): 137-163. King, Gary, Robert O. Keohane, and Sidney Verba. Designing Social Inquiry: Scientific ! Inference in Qualitative Research. Princeton: Princeton University Press, 1994. Kitts, James A. “Mobilizing in Black Boxes: Social Networks and Participation in Social ! Movement Organizations.” Mobilization 5, no. 2 (2000): 241-57. Klandermans, Bert. “Motivation and Types of Motives (Instrumental, Identity, Ideological Motives).” In The Wiley-Blackwell Encyclopedia of Social and Political Movements, edited by David A. Snow, Donatella della Porta, Bert Klandermans, and Doug McAdam. Malden, ! Mass.: Wiley-Blackwell, 2013. Klandermans, Bert. “Motivations to Action.” In The Oxford Handbook of Social Movements, edited by Donatella Della Porta and Mario Diani. New York: Oxford University Press, ! forthcoming.

287 Klandermans, Bert, and Dirk Oegema. “Potentials, Networks, Motivations, and Barriers: Steps Towards Participation in Social Movements.” American Sociological Review 52, no. 4 ! (1987): 519-531. Klandermans, Bert, Jojanneke van der Toorn, and Jacquelien van Stekelenburg. “Embeddedness and Identity: How Immigrants Turn Grievances into Action.” American Sociological Review !73 (2008): 992-1012. Knottnerus, J. David. “Collective Events, Rituals, and Emotions.” In Advances in Group Processes 27, edited by Shane R. Thye and Edward J. Lawler. Bingley, UK: Emerald Group ! Publishing Limited, 2010. Knottnerus, J. David. “Religion, Ritual, and Collective Emotion.” In Collective Emotions, edited by Christian von Scheve and Mikko Salmela, 312-325. New York: Oxford University Press, ! 2014. Konvalinka, Ivana, Dimitris Xygalatas, Joseph Bulbulia, Uffe Schjødt, Else-Marie Jegindø, Sebastian Wallot, Guy Van Orden, and Andreas Roepstorff. “Synchronized Arousal Between Performers and Related Spectators in a Fire-Walking Ritual.” Proceedings of the National ! Academy of Science 108, no. 20 (2011): 8514-8519. Krinsky, John, and Nick Crossley. “Social Movements and Social Networks: Introduction.” ! Social Movement Studies 13, no. 1 (2014) : 1-21. Kydd, Andrew, and Barbara F. Walter. “Sabotaging the Peace: The Politics of Extremist ! Violence.” International Organization 56, no. 2 (2002): 263-296. Kydd, Andrew, and Barbara F. Walter. “The Strategies of Terrorism.” International Security 31, ! no. 1 (2006): 49-80. Lamont, Michèle, and Ann Swidler. “Methodological Pluralism and the Possibilities and Limits ! of Interviewing.” Qualitative Sociology 37, no. 2 (2014): 153-171. Leach, Edmund R. Political Systems of Highland Burma: A Study of Kachin Social Structure, ! 2nd. ed. London: Bell and Sons, 1964. Leach, Edmund R. “Ritual.” In The International Encyclopedia of the Social Sciences 13, edited ! by David L. Sills, 520-526. New York: Macmillan, 1968. LeBas, Adrienne. “Polarization as Craft: Party Formation and State Violence in Zimbabwe.” ! Comparative Politics 38, no. 4 (2006): 419-38.

288 Leech, Beth L. ed. “Symposium on Interview Methods in Political Science.” PS: Political ! Science & Politics 23, no. 3 (2002): 663-88. Leerssen, Joep. “Monument and Trauma: Varieties of Remembrance.” In History and Memory in Modern Ireland, edited by Ian McBride, 204-222. Cambridge: Cambridge University Press, ! 2001. !Lévi-Strauss, Claude. The Savage Mind. Chicago: University of Chicago Press, 1966. Lichbach, Mark I. “What Makes Rational Peasants Revolutionary?: Dilemma, Paradox, and ! Irony in Peasant Collective Action.” World Politics 46, no. 3 (1994): 383-418. !Lichbach, Mark I. The Rebel’s Dilemma. Ann Arbor: University of Michigan Press, 1995. Lienard, Pierre, and Pascal Boyer. “Whence Collective Rituals? A Cultural Selection Model of ! Ritualized Behavior.” American Anthropologist 108, no. 4 (2006): 814-827. Lijphart, Arend. Democracy in Plural Societies: A Comparative Exploration. New Haven: Yale ! University Press, 1977. Linenthal, Edward T. “Epilogue.” In Culture and Belonging in Divided Societies: Contestation and Symbolic Landscapes, edited by Marc Howard Ross, 281-286. Philadelphia: University ! of Pennsylvania Press, 2009. Lloyd, Christopher D., and Ian Shuttleworth. “Residential Segregation in Northen Ireland 2001: Assessing the Value of Exploring Spatial Variation.” Environment and Planning A 44, no. 1 ! (2012): 52-67. Loveman, Mara. “High-Risk Collective Action: Defending Human Rights in Chile, Uruguay, and ! Argentina.” American Journal of Sociology 104, no. 2 (1998): 477-525. Lukes, Steven. Emile Durkheim: His Life and Work: A Historical and Critical Study. Stanford: ! Stanford University Press, 1973. !Lukes, Steven. Power: A Radical View. London: Macmillan, 1974. !Lukes, Steven. “Political Ritual and Social Integration.” Sociology 9 (1975): 289-308. Marx, Karl. “The Eighteenth Brumaire of Louis Napoleon.” In The Marx-Engels Reader, edited ! by Robert C. Tucker, 594-617. New York: Norton, 1978. !Mason, Andrew. “Politics and the State.” Political Studies 38 (1990): 575-87.

289 McAdam, Doug. “Recruitment to High-Risk Activism: The Case of Freedom Summer.” ! American Journal of Sociology 92, no. 1 (1986): 64-90. !McAdam, Doug. Freedom Summer. New York: Oxford University Press, 1988. McAdam, Doug. “The Biographical Consequences of Activism.” American Sociological Review ! 54, no. 5 (1989): 744-760. McAdam, Doug. Political Process and the Development of Black Insurgency, 1930-1970, 2nd ! ed. Chicago: University of Chicago Press, 1999. McAdam, Doug, and Ronnelle Paulsen. “Specifying the Relationship between Social Ties and ! Activism.” American Journal of Sociology 99, no. 3 (1993): 640-667. McAdam, Doug, Sidney Tarrow, and Charles Tilly. Dynamics of Contention. New York: ! Cambridge University Press, 2001. McAuley, James W. and Jonathan Tonge. “‘For God and for the Crown’: Contemporary Political and Social Attitudes among Orange Order Members in Northern Ireland.” Political ! Psychology 28, no. 1 (2007): 33-52. McAuley, James W., Jonathan Tonge, and Andrew Mycock. Loyal to the Core?: Orangeism and ! Britishness in Northern Ireland. Dublin: Irish Academic Press, 2011. McAuley, James W., Jonathan Tonge, and Peter Shirlow. “Conflict, Transformation, and Former Paramilitary Prisoners in Northern Ireland.” Terrorism and Political Violence 22, no. 1 (2009): 22-40. ! McBride, Ian. “Memory and National Identity in Modern Ireland.” In History and Memory in Modern Ireland, edited by Ian McBride, 1-42. Cambridge: Cambridge University Press, 2001. ! McBride, Ian R. Scripture Politics: Ulster Presbyterians and Irish Radicalism in Late ! Eighteenth-Century Ireland. Oxford: Clarendon Press, 1998. McEvoy, Kieran, and Peter Shirlow. “Re-imaging DDR: Ex-combatants, Leadership and Moral ! Agency in Conflict Transformation.” Theoretical Criminology 13, no. 1 (2009): 31-59. McGarry, John, and Brendan O’Leary. Explaining Northern Ireland: Broken Images. Oxford: ! Basil Blackwell, 1995.

290 McGarry, John, and Brendan O’Leary. “Consociational Theory, Northern Ireland’s Conflict, and its Agreement. Part 1: What Consociationalists Can Learn from Northern Ireland.” ! Government and Opposition 41, no. 1 (2006): 43-63. McGarry, John, and Brendan O’Leary. The Northern Ireland Conflict: Consociational ! Engagements. New York: Oxford University Press, 2004. McGovern, Mark, and Peter Shirlow. “Counter-Insurgency, Deindustrialisation and the Political Economy of Ulster Loyalism.” In Who Are ‘The People?’: Unionism, Protestantism and Loyalism in Late Twentieth Century Ireland, edited by Peter Shirlow and Mark McGovern, ! 176-198. London: Pluto, 1997. McNeill, William H. Keeping Together in Time: Dance and Drill in Human History. Cambridge: ! Harvard University Press, 1995. Mitchell, Claire. “The Limits of Legitimacy: Former Loyalist Combatants and Peace-Building in ! Northern Ireland.” Irish Political Studies 23, no. 1 (2008): 1-19. Mitchell, Claire. Religion, Identity and Politics in Northern Ireland: Boundaries of Belonging ! and Belief. Aldershot: Ashgate, 2006. Montgomery, Graham G.W., and J. Richard Whitten. The Order on Parade. Belfast: Grand ! Orange Lodge of Ireland Education Committee, 1995. Moore, Ronnie, and Andrew Sanders. “Formations of Culture: Nationalism and Conspiracy ! Ideology in Ulster Loyalism.” Anthropology Today 18, no. 6 (2002): 9-15. !Moore, Sally F., and Barbara G. Myerhoff, eds. Secular Ritual. Assen: Van Gorcum, 1977. !Morris, Aldon D. The Origins of the Civil Rights Movement. New York: Free Press, 1984. Mosley, Layna, ed. Interview Research in Political Science. Ithaca: Cornell University Press, ! 2013. Moyn, Samuel. The Last Utopia: Human Rights in History. Cambridge: Harvard University ! Press, 2010. Mulholland, Marc. The Longest War: Northern Ireland’s Troubled History. New York: Oxford ! University Press, 2002. Muller, Edward N., and Karl-Dieter Opp. “Rational Choice and Rebellious Collective Action.” ! American Political Science Review 80, no. 2 (1986): 471-487.

291 Munson, Ziad W. “When a Funeral Isn't Just a Funeral: The Layered Meaning of Everyday Action.” In Everyday Religion: Observing Modern Religious Lives, edited by Nancy T. ! Ammerman, 121-136. New York: Oxford University Press, 2007. Munson, Ziad W. The Making of Pro-Life Activists: How Social Movement Mobilization Works. ! Chicago: University of Chicago Press, 2008. Murtagh, Brendan. “Ethno-Religious Segregation in Post-Conflict Belfast.” Built Environment ! 37, no. 2 (2011): 213-225. Murtagh, Brendan. The Politics of Territory: Policy and Segregation in Northern Ireland. ! London: Palgrave Macmillan, 2002. Nasr, Vali. The Shia Revival: How Conflicts Within Islam will Shape the Future. New York: ! W.W. Norton, 2006. Nelson, Timothy J. Every Time I Feel the Spirit: Religious Experience and Ritual in an African- ! American Church. New York: NYU Press, 2005. Nepstad, Sharon Erickson. “Persistent Resistance: Commitment and Community in the ! Plowshares Movement.” Social Problems 51, no. 1 (2004): 43-60. Nepstad, Sharon Erickson, and Christian Smith. “Rethinking Recruitment to High-Risk/High- ! Cost Activism: The Case of the Nicaragua Exchange.” Mobilization 4, no. 1 (1999): 25-40. Nirenberg, David. Communities of Violence: Persecution of Minorities in the Middle Ages. ! Princeton: Princeton University Press, 1996. Nisbett, Richard E., and Timothy DeCamp Wilson. “Telling More Than We Can Know: Verbal ! Reports on Mental Processes.” Psychological Review 84, no. 3 (1977): 231-59. Nolan, Paul. Northern Ireland Peace Monitoring Report, Number 1. Belfast: Community ! Relations Council, 2012. Nolan, Paul. Northern Ireland Peace Monitoring Report, Number 2. Belfast: Community ! Relations Council, 2013. Nolan, Paul. Northern Ireland Peace Monitoring Report, Number 3. Belfast: Community ! Relations Council, 2014. Nolan, Paul, Dominic Bryan, Clare Dwyer, Katy Hayward, Katy Radford, and Peter Shirlow. The ! Flag Dispute: Anatomy of a Protest. Belfast: Queen’s University Belfast, 2014.

292 Northern Ireland Youth Forum. Sons of Ulster: Exploring Loyalist Band Members Attitudes ! towards Culture, Identity and Heritage. Belfast: Northern Ireland Youth Forum, 2013. O’Doherty, Malachi. The Trouble with Guns: Republican Strategy and the Provisional IRA. ! Belfast: Blackstaff, 1998. O’Leary, Brendan. “The Nature of the Agreement.” Fordham International Law Journal 22, no. ! 4 (1998): 1628-1667. !O’Leary, Brendan, and John McGarry. The Politics of Antagonism. London: Athlone, 1993. Oliver, Pamela. “Rewards and Punishments as Selective Incentives for Collective Action: ! Theoretical Investigations.” American Journal of Sociology 85, no. 6 (1980): 1356-1375. Olson, Mancur. The Logic of Collective Action: Public Goods and the Theory of Groups. ! Cambridge: Harvard University Press, 1965. Ortner, Sherry B. “Theory in Anthropology since the Sixties.” Comparative Studies in Society ! and History 26, no. 1 (1984): 126-166. Ost, David. Solidarity and the Politics of Anti-Politics: Opposition and Reform in Poland since ! 1968. Philadelphia: Temple University Press, 1990. Ozouf, Mona. Festivals and the French Revolution. Translated by Alan Sheridan. Cambridge: ! Harvard University Press, 1988. Parades Commission for Northern Ireland. Annual Report and Financial Statements for the Year ! Ended 31 March 2012. Norwich, UK: The Stationary Office, 2012. Parades Commission for Northern Ireland. Annual Report and Financial Statements for the Year ! Ended 31 March 2013. London: Stationary Office, 2013. Parades Commission for Northern Ireland. Annual Report and Financial Statements for the Year ! Ended 31 March 2014. Norwich, UK: The Stationary Office, 2014. Pat Finucane Centre. For God and Ulster: An Alternative Guide to the Loyal Orders. Derry: Pat Finucane Centre, 1997.

Patel, David Siddhartha. “Islam, Information, and Social Order: The Strategic Role of Religion ! in Muslim Societies.” Ph.D. dissertation. Stanford University, 2007. !Patterson, Henry. Ireland since 1939. Oxford: Oxford University Press, 2002.

293 Patterson, Henry. “Beyond the ‘Micro Group’: The Dissident Republican Challenge.” In Dissident Irish Republicanism, edited by P.M. Currie and Max Taylor, 65-96. New York: ! Continuum, 2011. Patterson, Henry. “Unionism after Good Friday and St Andrews.” The Political Quarterly 83, no. ! 2 (2012): 247-255. Patterson, Henry, and Eric Kaufmann. Unionism and Orangeism in Northern Ireland since 1945: ! The Decline of the Loyal Family. Manchester: Manchester University Press, 2007. Pearlman, Wendy. “Emotions and the Microfoundations of the Arab Uprisings.” Perspectives on ! Politics 11, no. 2 (2013): 387-409. Petersen, Roger D. Understanding Ethnic Violence: Fear, Hatred, and Resentment in Twentieth- ! Century Eastern Europe. New York: Cambridge University Press, 2002. Petersen, Roger D. Western Intervention in the Balkans: The Strategic Use of Emotions in ! Conflict. New York: Cambridge University Press, 2011. Peterson, Abby, Mattias Wahlström, Magnus Wennerhag, Camilo Christancho, and José-Manuel Sabucedo. “May Day Demonstrations in Five European Countries.” Mobilization 17, no. 3 ! (2012): 281-300. Pfaff, Steven, and Guobin Yang. “Double-Edged Rituals and the Symbolic Resources of Collective Action: Political Commemorations and the Mobilization of Protest in 1989.” ! Theory and Society 30, no. 4 (2001): 539-589. Philpott, Daniel. “Explaining the Political Ambivalence of Religion.” American Political Science ! Review 101, no. 3 (2007): 505-25. Pinard, Maurice. Motivational Dimensions in Social Movement and Contentious Collective ! Action. Montreal and Kingston: McGill-Queen’s University Press, 2011. Polletta, Francesca, and James M. Jasper. “Collective Identity and Social Movements.” Annual ! Review of Sociology 27 (2001): 283-305. Popkin, Samuel L. The Rational Peasant: The Political Economy of Rural Society in Vietnam. ! Berkeley: University of California Press, 1979. !Posen, Barry R. “The Security Dilemma and Ethnic Conflict.” Survival 35, no. 1 (1993): 27-47. Powell, Martyn J. “Political Toasting in Eighteenth-Century Ireland.” History 91, no. 304 (2006): 508-529.

294 ! Racioppi, Linda, and Katherine O’Sullivan See. “Ulstermen and Loyalist Ladies on Parade: Gendering Unionism in Northern Ireland.” International Feminist Journal of Politics 2, no. 1 ! (2000): 1-29. Radcliffe-Brown, Alfred R. The Andaman Islanders: A Study in Social Anthropology. ! Cambridge: Cambridge University Press, 1922. Radford, Katy. “Drum Rolls and Gender Roles in Protestant Marching Bands in Belfast.” British ! Journal of Ethnomusicology 10, no. 2 (2001): 37-59. Ramsey, Gordon. “Band Practice: Class, Taste and Identity in Ulster Loyalist Flute Bands.” ! Ethnomusicology Ireland 1, no. 1 ( 2011): 1-20. Ramsey, Gordon. Music, Emotion and Identity in Ulster Marching Bands: Flutes, Drums and ! Loyal Sons. Bern: Peter Lang, 2011. !Rappaport, Roy A. Ecology, Meaning and Religion. Berkeley: North Atlantic Books, 1979. !Rorty, Richard. “Religion as Conversation-Stopper.” Common Knowledge 3, no. 1 (1994): 1-6. Ross, Marc Howard. “Psychocultural Interpretations and Dramas: Identity Dynamics in Ethnic ! Conflict.” Political Psychology 22, no. 1 (2001): 157-178. Ross, Marc Howard. Cultural Contestation in Ethnic Conflict. New York: Cambridge University ! Press, 2007. Ross, Marc Howard. “Preface.” In Culture and Belonging in Divided Societies: Contestation and Symbolic Landscapes, edited by Marc Howard Ross, ix-xi. Philadelphia: University of ! Pennsylvania Press, 2009. Roth, Andrew L. “‘Men Wearing Masks’: Issues of Description in the Analysis of Ritual.” ! Sociological Theory 13, no. 3 (1995): 301-327. Roth, Silke. “Developing Working-Class Feminism: A Biographical Approach to Social Movement Participation.” In Self, Identity, and Social Movements, edited by Sheldon Stryker, Timothy J. Owens, and Robert W. White, 300-323. Minneapolis: University of Minnesota ! Press, 2000. Roy, William G. “How Social Movements Do Culture.” International Journal of Politics, ! Culture, and Society 23, nos. 2-3 (2010): 85-98.

295 Royal Ulster Constabulary. Chief Constable’s Report, 1981. Belfast: Royal Ulster Constabulary, ! 1981. RSM McClure Watters. “The Socio-Economic Impact of the Traditional Protestant Parading ! Sector in Northern Ireland.” Belfast: RSM McClure Watters, 2013. Ruane, Joseph, and Jennifer Todd. The Dynamics of Conflict in Northern Ireland: Power, Conflict, and Emancipation. New York: Cambridge University Press, 1996. ! Rüdig, Wolfgang, and Georgios Karyotis. “Who Protests in Greece: Opposition to Mass Austerity.” British Journal of Political Science 44, no. 3 (2014): 487-513. ! Rupp, Leila J., and Verta Taylor. Drag Queens at the 801 Cabaret. Chicago: University of ! Chicago Press, 2003. Ryan, Mary. “The American Parade: Representations of the Nineteenth-Century Social Order.” In The New Cultural History, edited by Lynn Hunt, 131-153. Berkeley: University of California ! Press, 1989. Samek, Alyssa and Karrin Vasby Anderson. “The Day the Campaign Died: The Wellstone Memorial, Civic Piety, and Political Propriety.” Communication Quarterly 59, no. 2 (2011): ! 155-178. Santino, Jack. “Public Protest and Popular Style: Resistance from the Right in Northern Ireland ! and South Boston.” American Anthropologist 101, no. 3 (1999): 515-528. Santino, Jack. Signs of War and Peace: Social Conflict and the Uses of Symbols in Public in ! Northern Ireland. New York: Palgrave, 2001. Scacco, Alexandra. “Who Riots? Explaining Individual Participation in Ethnic Violence.” Ph.D. ! dissertation, Columbia University, 2010. Scheff, Thomas J. Bloody Revenge: Emotions, Nationalism, and War. Boulder, Colo.: Westview, ! 1994. Schlozman, Kay Lehman, Sidney Verba, and Henry E. Brady. “Participation's Not a Paradox: The View from American Activists.” British Journal of Political Science 25, no. 1 (1995): ! 1-36. Schuessler, Alexander A. A Logic of Expressive Choice. Princeton: Princeton University Press, ! 2000.

296 Schussman, Alan, and Sarah A. Soule. “Process and Protest: Accounting for Individual Protest ! Participation.” Social Forces 84, no. 2 (2005) : 1083-1108. Scott, James. The Art of Not Being Governed: An Anarchist History of Upland Southeast Asia. ! New Haven: Yale University Press, 2010. Sewell, William H. “Political Events as Structural Transformations: Inventing Revolution at the ! Bastille.” Theory and Society 25, no. 6 (1996): 841-881. Sewell, William H. “Space in Contentious Politics.” In Silence and Voice in the Study of Contentious Politics, edited by Ronald R. Aminzade, Jack A. Goldstone, Doug McAdam, Elizabeth J. Perry, William H. Sewell, Sidney Tarrow, and Charles Tilly, 51-88. New York: ! Cambridge University Press, 2001. Shani, Ornit. Communalism, Caste and Hindu Nationalism: The Violence in Gujarat. ! Cambridge: Cambridge University Press, 2007. Shils, Edward, and Michael Young. “The Meaning of the Coronation.” Sociological Review 1, ! no. 2 (1953): 63-81. Shirlow, Peter. “Ethno-Sectarianism and the Reproduction of Fear in Belfast.” Capital & Class, ! 27, no. 2 (2003): 77-93. Shirlow, Peter. The End of Ulster Loyalism? Manchester: Manchester Unviersty Press, 2012.

Shirlow, Peter and Brendan Murtagh. Belfast: Segregation, Violence, and City. London: Pluto, ! 2006. Simmons, Erica. “Grievances Do Matter in Mobilization.” Theory and Society 43, no. 5 (2014): ! 513-546. Simms, J.G. “Remembering 1690.” Studies: An Irish Quarterly Review 63, no. 251 (1974): ! 231-242. Simon, Bernd, Michael Loewy, Stefan Stürmer, Ulrike Weber, Peter Freytag, Corinna Habig, Claudia Kampmeier, and Peter Spahlinger. “Collective Identification and Social Movement ! Participation.” Journal of Personality and Social Psychology 74, no. 3 (1998): 646-658. Skorupski, John. Symbol and Theory: A Philosophical Study of the Theories of Religion in Social ! Anthropology. Cambridge: Cambridge University Press, 1976. !Smith, Anthony D. The Ethnic Origins of Nations. Oxford: Blackwell, 1986.

297 Smith, Jonathan Z. To Take Place: Toward Theory in Ritual. Chicago: University of Chicago ! Press, 1987. Smithey, Lee A. “Strategic Collective Action and Collective Identity Reconstruction: Parading Disputes and Two Northern Ireland Towns.” Ph.D. dissertation, University of Texas, Austin, ! 2002. Smithey, Lee A. Unionists, Loyalists, and Conflict Transformation in Northern Ireland. New ! York: Oxford University Press, 2011. Smithey, Lee A. and Michael P. Young. “Parading Protest: Orange Parades in Northern Ireland and Temperance Parades in Antebellum America.” Social Movement Studies 9, no. 4 (2010) : ! 393-410. Smyth, Marie, and Jennifer Hamilton. “The Human Costs of the Troubles.” In Researching the Troubles: Social Science Perspectives on the Northern Ireland Conflict, edited by Owen ! Hargie and David Dickson, 15-36. Edinburgh: Mainstream, 2003. Snow, David A. “Social Movements as Challenges to Authority: Resistance to an Emerging Conceptual Hegemony.” Research in Social Movements, Conflicts, and Change 25 (2004): ! 3-25. Snow, David A., Louis A. Zurcher, Jr. and Sheldon Ekland-Olson. “Social Networks and Social Movements: A Microstructural Approach to Differential Recruitment.” American ! Sociological Review 45, no. 5 (1980): 787-801. Snow, David. “Collective Identity and Expressive Forms.” Center for the Study of Democracy ! Working Paper, University of California, Irvine, 2001. Snyder, Jack. From Voting to Violence: Democratization and Nationalist Conflict. New York: ! Norton, 2000. Snyder, Jack, and Karen Ballentine. “Nationalism and the Marketplace of Ideas.” International ! Security 21, no. 2 (1996): 5-40. Sosis, Richard, and Candace Alcorta. “Signaling, Solidarity, and the Sacred: The Evolution of ! Religious Behavior.” Evolutionary Anthropology 12, no. 6 (2003): 264-274. Sosis, Richard, and Bradley J. Ruffle. “Religious Ritual and Cooperation: Testing for a Relationship on Israeli Religious and Secular Kibbutzim.” Current Anthropology 44, no. 5 ! (2003): 713-22.

298 Southern, Neil. “Territoriality, Alienation, and Loyalist Decommissioning: The Case of the Shankill in Protestant West Belfast.” Terrorism and Political Violence 20, no. 1 (2007): ! 66-86. !Staal, Frits. “The Meaninglessness of Ritual.” Numen 26, no. 1 (1979): 2-22. Staggenborg, Suzanne, and Amy Lang. “Culture and Ritual in the Montreal Women’s ! Movement.” Social Movement Studies 6, no. 2 (2007) : 177-194. Stamatov, Peter. “Interpretive Activism and the Political Uses of Verdi’s Operas in the 1840s.” ! American Sociological Review 67, no. 3 (2002): 345-366. Stepan, Alfred. “Religion, Democracy, and the ‘Twin Tolerations.’” Journal of Democracy 11, ! no. 4 (2000): 37-57. !Stepan, Alfred. Arguing Comparative Politics. New York: Oxford University Press, 2001. Stepan, Alfred. “Rituals of Respect: Sufis and Secularists in Senegal in Comparative ! Perspective.” Comparative Politics (2012): 379-401. Straus, Scott. The Order of Genocide: Race, Power, and War in Rwanda. Ithaca: Cornell ! University Press, 2006. Stroschein, Sherrill. “Politics is Local: Ethnoreligious Dynamics under the Microscope.” ! Ethnopolitics 6, no. 2 (2007): 173-185. Stroschein, Sherrill. “Microdynamics of Bilateral Ethnic Mobilization.” Ethnopolitic 10, no. 1 ! (2011): 1-34. Stroschein, Sherill. Ethnic Struggle, Coexistence, and Democratization in Eastern Europe. New ! York: Cambridge University Press, 2012. Swidler, Ann. “Culture in Action: Symbols and Strategies.” American Sociological Review 51, ! no. 2 (1986): 273-286. Swidler, Ann. “Cultural Power and Social Movements.” In Social Movements and Culture, edited by Hank Johnston and Bert Klandermans, 25-40. Minneapolis: University of Minnesota ! Press, 1995. !Swidler, Ann. Talk of Love: How Culture Matters. Chicago: University of Chicago Press, 2001. Sykes, Gresham M., and David Matza, “Techniques of Neutralization: A Theory of Delinquency.” American Sociological Review 22, no. 6 (1957): 664-670.

299

Tajfel, Henri, and John C. Turner. “An Integrative Theory of Intergroup Conflict.” In The Social Psychology of Intergroup Relation, edited by W. G. Austin and S. Worchel, 33-47. Monterey, ! CA: Brooks/Cole, 1979. Tajfel, Henri, and John C. Turner. “The Social Identity Theory of Intergroup Behaviour.” In Psychology of Intergroup Relations, edited by S. Worchel & W. G. Austin, 7-24. Chicago: ! Nelson-Hall, 1986. Tambiah, Stanley J. Culture, Thought, and Social Action: An Anthropological Perspective. ! Cambridge: Harvard University Press, 1985. Tambiah, Stanley. Leveling Crowds: Ethnonationalist Conflicts and Collective Violence in South ! Asia. Berkeley: University of California Press, 1996. Tavory, Iddo. “The Private Life of Public Ritual: Interaction, Sociality and Codification in a ! Jewish Orthodox Congregation.” Qualitative Sociology 36, no. 2 (2013): 125-139. Taylor, Michael. “Rationality and Revolutionary Collective Action.” In Rationality and ! Revolution, edited by Michael Taylor, 63-97. New York: Cambridge University Press, 1988. Taylor, Verta, Katrina Kimport, Nella Van Dyke, and Ellen Ann Andersen. “Culture and Mobilization: Tactical Repertoires, Same-Sex Weddings, and the Impact on Gay Activism.” ! American Sociological Review 74, no. 6 (2009): 865-890. Taylor, Verta, Leila J. Rupp, and Joshua Gamson. “Performing Protest: Drag Shows as Tactical Repertoires of the Gay and Lesbian Movement.” Research in Social Movements, Conflicts, ! and Change 25 (2004): 105-137. Taylor, Verta, and Nella Van Dyke. “‘Get up, Stand up’: Tactical Repertoires of Social Movements.” In The Blackwell Companion to Social Movements, edited by David A. Snow, ! Sarah A. Soule, and Hanspeter Kriesi, 262-293. Malden, Mass.: Blackwell, 2004. Taylor, Verta, and Nancy E. Whittier. “Collective Identity in Social Movement Communities: Lesbian Feminist Mobilization.” In Frontiers in Social Movement Theory, edited by Aldon D. ! Morris and Carol M. Mueller, 104-129. New Haven: Yale University Press, 1992. Tilly, Charles. “Social Movements and (All Sorts of) Other Political Interactions—Local, National, and International—Including Identities.” Theory and Society 27, no. 4 (1998): ! 453-480. !Tilly, Charles. The Politics of Collective Violence. New York: Cambridge University Press, 2001.

300 Tilly, Charles. Contention and Democracy in Europe, 1650-2000. New York: Cambridge ! University Press, 2004. !Tilly, Charles. Regimes and Repertoires. Chicago: University of Chicago Press, 2006. Tilly, Charles. Contentious Performances. New York: Cambridge University Press, 2008.

Todd, Jennifer. “Two Traditions in Unionist Political Culture.” Irish Political Studies 2, no. 1 ! (1987): 1-26. Tonge, Jonathan, Máire Braniff, Thomas Hennessey, James W. McAuley, and Sophie A. Whiting. The Democratic Unionist Party: From Protest to Power. Oxford: Oxford University Press, ! 2014. Tonge, Jon, Jocelyn Evans, Robert Jeffery, and James W. McAuley. “New Order: Political Change and the Protestant Orange Tradition in Northern Ireland.” British Journal of Politics ! and International Relations13, no. 3 (2011): 400-419. Tonge, Jonathan, Bernadette C. Hayes, and Paul Mitchell. Northern Ireland General Election Attitudes Survey, 2010. Distributed by UK Data Archive, 2010, doi: 10.5255/UKDA- ! SN-6553-1. Trejo, Guillermo. Popular Movements in Autocracies: Religion, Repression, and Indigenous ! Collective Action in Mexico. New York: Cambridge University Press, 2012. Turner, Victor W. The Forest of Symbols: Aspects of Ndembu Ritual. Ithaca: Cornell University ! Press, 1967. Turner, Victor W. The Ritual Process: Structure and Anti-Structure. Ithaca: Cornell University ! Press, 1969. Valdesolo, Piercarlo, and David DeSteno. “Synchrony and the Social Tuning of Compassion.” ! Emotion 11, no. 2 (2011): 262-6. Valdesolo, Piercarlo, Jennifer Ouyang, and David DeSteno. “The Rhythm of Joint Action: Synchrony Promotes Cooperative Ability.” Journal of Experimental Social Psychology 46, ! no. 4 (2010): 693-695. Valentino, Benjamin A. Final Solutions: Mass Killing and Genocide in the 20th Century. Ithaca: ! Cornell University Press, 2004. Valiani, Arafaat A. “Processions as Publics: Religious Ceremonials and Modes of Public Sphere Intervention in Western India.” Unpublished manuscript, Williams College, 2011.

301 ! van der Veer, Peter. “Riots and Rituals: The Construction of Violence and Public Space in Hindu Nationalism.” In Riots and Pogroms, edited by Paul R. Brass, 154-176. New York: New York ! University Press, 1996. van Gennep, Arnold. The Rites of Passage. Translated by Monika B. Vizedom and Gabrielle L. ! Caffee. Chicago: University of Chicago Press, 1960. Varshney, Ashutosh. “Nationalism, Ethnic Conflict, and Rationality.” Perspectives on Politics 1, ! no. 1 (2003): 85-99. Varshney, Ashutosh. “Ethnicity and Ethnic Conflict.” In The Oxford Handbook of Comparative Politics, edited by Carles Boix and Susan Stokes, 278-279. New York: Oxford University ! Press, 2007. Varshney, Ashutosh, and Joshua Gubler. “The State and Civil Society in Communal Violence: Sparks and Fires.” In Routledge Handbook of Indian Politics, edited by Atul Kohli and ! Prerna Singh, 155-166. London: Routledge, 2013. Verba, Sidney, Kay Lehman Schlozman, and Henry E. Brady. Voice and Equality: Civic ! Voluntarism in American Politics. Cambridge: Harvard University Press, 1995. Villalón, Leonardo A. “Sufi Rituals as Rallies: Religious Ceremonies in the Politics of ! Senegalese State-Society Relations.” Comparative Politics 26, no. 4 (1994): 415-37. Walgrave, Stefaan, and Ruud Wouters. “The Missing Link in the Diffusion of Protest: Asking ! Others.” American Journal of Sociology 119, no. 6 (2014): 1670-1709. Walker, Brian. “1641, 1689, 1690 And All That: The Unionist Sense of History.” The Irish ! Review (1992): 56-64. Wasserstein, Bernard. “Patterns of Communal Conflict in Palestine.” In Jewish History: Essays in Honour of Chimen Abramsky, edited by Ada Rapoport-Albert and Steven J. Zipperstein, ! 611-628. London: Peter Halban, 1988. Weber, Max. “Politics as a Vocation.” In From Max Weber: Essays in Sociology, edited by H.H. ! Gerth and C. Wright Mills, 77-128. New York: Oxford University Press, 1972. Weber, Max. “The Social Psychology of World Religions.” In From Max Weber: Essays in Sociology, edited by H.H. Gerth and C. Wright Mills, 267-301. New York: Oxford University ! Press, 1972.

302 Weber, Max. “The Sociology of Charismatic Authority.” In From Max Weber: Essays in Sociology, edited by H.H. Gerth and C. Wright Mills, 245-252. New York: Oxford University ! Press, 1972. Weber, Max. Economy and Society: An Outline of Interpretive Sociology. Translated by Guenther ! Roth and edited by Claus Wittich. Berkeley: University of California Press, 1978. Wedeen, Lisa. Ambiguities of Domination: Politics, Rhetoric, and Symbols in Contemporary ! Syria. Chicago: University of Chicago Press, 1999. Wedeen, Lisa. “Conceptualizing Culture: Possibilities for Political Science.” American Political ! Science Review 96, no. 4. (2002): 713-728. White, Robert W. “Structural Identity Theory and the Post-Recruitment Activism of Irish Republicans: Persistence, Disengagement, Splits, and Dissidents in Social Movement ! Organizations.” Social Problems 57, no. 3 (2010): 341-370. White, Timothy J., ed. Lessons from the Northern Ireland Peace Process. Madison: University of ! Wisconsin Press, 2013. !Whyte, John. Interpreting Northern Ireland. Oxford: Clarendon, 1990. Wilentz, Sean, ed. Rites of Power: Symbolism, Ritual, and Politics since the Middle Ages. ! Philadelphia: University of Pennsylvania Press, 1999. Wilkinson, Steven I. Votes and Violence: Electoral Competition and Ethnic Violence in India. ! New York: Cambridge University Press, 2004. Wilkinson, Steven I. “Which Group Identities Lead to Most Violence? Evidence from India.” In Order, Conflict, and Violence, edited by Stathis N. Kalyvas, Ian Shapiro, and Tarek Masoud, ! 271-300. New York: Cambridge University Press, 2008. Wiltermuth, Scott. “Synchrony and Destructive Obedience.” Social Influence 7, no. 2 (2012): ! 78-89. Wiltermuth, Scott S. “Synchronous Activity Boosts Compliance with Requests to Aggress.” ! Journal of Experimental Social Psychology 48, no. 1 (2012): 453-6. Wiltermuth, Scott S., and Chip Heath. “Synchrony and Cooperation.” Psychological Science 20, ! no. 1 (2009): 1-5. Wiltfang, Gregory L., and Doug McAdam. “The Costs and Risks of Social Activism: A Study of Sanctuary Movement Activism.” Social Forces 69, no. 4 (1991) : 987-1010.

303 ! Wimmer, Andreas, and Conrad Schetter. “Ethnic Violence.” In International Handbook of Violence Research, edited by Wilhelm Heitmeyer and John Hagan, 247-260. Dordecht, ! Netherlands: Kluwer Academic Publishers, 2003. Wood, Elisabeth Jean. “The Emotional Benefits of Insurgency in El Salvador.” In Passionate Politics: Emotions and Social Movements, edited by Jeff Goodwin, James M. Jasper, and ! Francesca Polletta, 267-80. Chicago: University of Chicago Press, 2001. Wood, Elisabeth Jean. Insurgent Collective Action and Civil War in El Salvador, New York: ! Cambridge University Press, 2003. Wright, Frank. Two Lands on One Soil: Ulster Politics before Home Rule. New York: St. ! Martin’s, 1996. Yang, Anand A. “Sacred Symbol and Sacred Space in Rural India: Community Mobilization in the ‘Anti-Cow Killing’ Riot of 1893.” Comparative Studies in Society and History 22, no. 4 ! (1980): 576-596. Yeats, W.B., “The Lake Isle of Innisfree.” In The Collected Poems of W.B. Yeats, 2nd ed., edited ! by Richard J. Finneran, 39. New York: Scribner, 1996. Zerubavel, Yael. Recovered Roots: Collective Memory and the Making of Israeli National ! Tradition. Chicago: University of Chicago Press, 1995. Zubrzycki, Geneviève. The Crosses of Auschwitz: Nationalism and Religion in Post-Communist ! Poland. Chicago: University of Chicago Press, 2006. Zubrzycki, Geneviève. “Aesthetic Revolt and the Remaking of National Identity in Québec, ! 1960-1969.” Theory and Society 42, no. 5 (2013): 423-75. Zuesse, Evan M. “Ritual.” In Encyclopedia of Religion, 2nd ed., edited by Lindsay Jones, 7834-7848. Detroit: Macmillan Reference, 2005.

304 Appendix ! Table A1. Robustness Check: Alternative Measures of Protestant Identification Main Model Model A1 Model A2 Model A3 (Table 7.2, (Disaggregate (British) (Unionist) Model 2) Prot ID) Ethnic Rivalry Protestant Identification -0.09 [0.12] Strong Ties to Other Protestants -0.02 [0.56] Feel Like other Protestants 0.25 [0.29] Proud to Be Called Protestants -0.21 [0.32] British -0.82 [1.20] Unionist 1.39 [1.31] Anti-Catholicism 1.80 [1.76] 2.26 [1.95] 1.95 [1.59] 2.14 [1.58] Collective Action Social Pressure 0.58 [0.16]*** 0.78 [0.16]*** 0.72 [0.24]*** 0.83 [0.24]*** Social Ties Family Marched -0.07 [0.74] -0.15 [0.83] -0.11 [0.89] -0.17 [0.86] Close Friends at Age 16 0.22 [0.35] 0.22 [0.29] 0.68 [0.45] 0.45 [0.38] Been Asked to March 0.18 [0.85] 0.25 [0.89] 0.61 [0.73] 0.00 [0.90] Control Variables Marched as Youth 0.73 [0.43]* 0.96 [0.47]** 0.53 [0.38] 0.84 [0.54] Age -0.04 [0.03] -0.05 [0.03] -0.05 [0.04] -0.06 [0.04]* Education -0.84 [0.86] -1.09 [0.90] -0.72 [1.01] -0.83 [0.81] Children under 18 -1.06 [0.68] -1.42 [0.73]* -1.52 [0.91]* -1.23 [0.84] Full-Time Job 1.84 [0.76]** 2.50 [0.82]*** 2.30 [0.78]*** 2.16 [0.70]*** Church Attendance 0.28 [0.14]** 0.37 [0.14]*** 0.33 [0.13]** 0.32 [0.12]*** Constant -3.09 [1.97] -4.21 [2.26]* -3.91 [2.48] -5.32 [2.25]** Number of observations 160 160 160 164 Standard errors clustered at neighborhood level in brackets. Enumerator fixed effects are not reported. * p<0.10, ** p<0.05, *** p<0.01 ! ! ! ! !

305 Table A2. Robustness Check: Alternative Measures of Anti-Catholicism and Family Ties Main Model Model A4 Model A5 Model A6 (Table 7.2, (Disaggregate (Anti-Cath. w/o (Disaggregate Model 2) Anti- Econ Deserve) Family Catholicism) Marched) Ethnic Rivalry Protestant Identification -0.09 [0.12] -0.05 [0.17] -0.05 [0.14] -0.08 [0.13] Anti-Catholicism 1.80 [1.76] 2.58 [1.95] Caths. Cause Sectarian Tension 0.27 [0.64] Oppose Family Marrying Cath. 0.41 [0.45] Caths. Gained More Econ 1.48 [0.61]** Caths. Need Remind. Live in UK -0.91 [0.78] Alternative Anti-Catholicism 1.66 [1.89] Collective Action Social Pressure 0.58 [0.16]*** 0.79 [0.24]*** 0.82 [0.17]*** 0.82 [0.17]*** Social Ties Family Marched -0.07 [0.74] -0.35 [0.79] -0.09 [0.72] Father Marched -0.35 [1.43] Other Family Marched 0.12 [0.59] Close Friends at Age 16 0.22 [0.35] 0.22 [0.50] 0.36 [0.39] 0.31 [0.40] Been Asked to March 0.18 [0.85] -0.16 [0.83] 0.30 [0.92] 0.27 [0.85] Control Variables Marched as Youth 0.73 [0.43]* 0.86 [0.52]* 0.94 [0.47]** 0.94 [0.48]* Age -0.04 [0.03] -0.07 [0.04]* -0.06 [0.03]* -0.05 [0.03] Education -0.84 [0.86] -1.79 [1.27] -1.23 [1.00] -1.08 [0.90] Children under 18 -1.06 [0.68] -2.06 [1.16]* -1.53 [0.70]** -1.56 [0.75]** Full-Time Job 1.84 [0.76]** 3.06 [1.13]*** 2.49 [0.88]*** 2.63 [0.90]*** Church Attendance 0.28 [0.14]** 0.47 [0.20]** 0.35 [0.16]** 0.34 [0.14]** Constant -3.09 [1.97] -3.38 [1.87]* -3.94 [2.33]* -4.77 [2.23]** Number of observations 160 160 169 160 Standard errors clustered at neighborhood level in brackets. Enumerator fixed effects are not reported. * p<0.10, ** p<0.05, *** p<0.01 ! ! ! ! !

306 Table A3. Interviewees Quoted in Dissertation. (OO is Orange Order, RBI is Royal Black Institution, ABOD is Apprentice Boys of Derry; Leadership role means senior leadership within a parading organization) Interview “Name” Parading Afflilation Occupation Date Albert OO, RBI retired factory worker Aug 20, 2012 Alexander OO retired clergy and politician Jul 19, 2013 Ben OO, ABOD politician Aug 13, 2013 Billy OO, band truck driver Aug 6, 2013 Chris Band small business owner Jul 31, 2012 Craig Band politician Dec 12, 2012 Dan Band bricklayer Aug 18, 2012 David Bands, ABOD community safety worker Jul 30, 2013 Edward OO politician Jul 27, 2012 Frankie OO community center management Dec 4, 2012 Gary N/A clergy Dec 4, 2012 George OO, RBI, ABOD; Leadership role works for parading organization Aug 14, 2012 Howie ABOD, OO factory worker Aug 13, 2012 Ian Parade attender security guard Jun 13, 2014 Isaac OO clergy Aug 12, 2013 Jack N/A clergy Jul 30, 2013 Jamie Band unemployed Dec 13, 2012 Jesse OO politician Aug 19, 2013 John OO; Leadership role clergy Jul 25, 2013 Joseph Band bar tender Jun 11, 2014 Kenny Band factory worker Dec 12, 2012 Kyle Band skilled mechanic Jun 11, 2014 Lee Band, OO civil service Dec 13, 2012 Mark OO small business owner Jul 11, 2013 Matt Parade attender maintainence May 3, 2013

307 Interview “Name” Parading Afflilation Occupation Date Michael OO health & safety supervisor Aug 19, 2013 Mikey OO, RBI, ABOD truck driver Aug 27, 2012 Nigel ABOD, OO, RBI retired business owner May 1, 2013 Rachel Band student Aug 8, 2013 Rich ex-OO clergy Nov 20, 2012 Robby OO, RBI, ABOD tattoo artist Aug 1, 2013 Robert OO retired factory worker Nov 28, 2012 Sammy OO, ABOD, Band youth worker Jul 9, 2013 Samuel Parade attender public relations Aug 19, 2013 Scott Band arts and culture Jul 23, 2012 Sophie Parade attender healthcare Aug 20, 2012 Steven Band, OO, RBI retired retail manager Dec 5, 2012 Tom OO; Leadership role clergy Aug 14, 2012 Tommy N/A ex-paramilitary prisoner Nov 26, 2012 Walter Band small business owner Jul 31, 2012

308