INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. UM I films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter free, while others may be from any type o f computer printer.
The quality of this reproduction is dependent upon the quality of the copy subm itted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction.
In the unlikely event that the author did not send TJMI a complete manuscript and there are missing pages, these \rill be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back o f the book.
Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6” x 9” black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order. UMI A Bell & Howell Infbnnation Company 300 North Zed> Road, Ann Arbor MI 48106-1346 USA 313/761-4700 800/521-0600
LEADERSHIP AND NATIONALIST MOBILIZATION: INVESTIGATING THE LINKAGE BETWEEN STRATEGIES AND CONTEXT
DISSERTATION
Presented in Partial Fulfillment of the Requirements for
the Degree of Doctor of Philosophy in the Graduate
School of the Ohio State University
By
Andrea K. Grove, M A.
*****
The Ohio State University 1999
Dissertation Committee: Approved by Professor Margaret G. Hermann, Adviser
Professor Richard K. Herrmann Adviser Professor Donald A Sylvan Politic Graduate Program UMI Number: 9931605
Copyright 1999 by Grove, Andrea Kathleen
All rights reserved.
UMI Microform 9931605 Copyright 1999, by UMI Company. All rights reserved.
This microform edition is protected against unauthorized copying under Title 17, United States Code.
UMI 300 North Zeeb Road Ann Arbor, MI 48103 Copyright by Andrea K. Grove 1999 ABSTRACT
This dissertation is a comparative, longitudinal study of leadership mobilization of identity groups in Northern Ireland and Rhodesia/Zimbabwe. In the past decade, profound changes around the world have focused attention on issues of identity within states. How do societies achieve a sense of identity, what factors affect change in identity? As conflicts over identity become more visible in world politics and are treated increasingly as “international” issues, policy makers should understand if and how actors in the international environment can affect the debate over the definition of identity and
“mission.”
This study asserts that competition among leaders to define group boundaries and mission often emerges in times of uncertainty. In response to the “battle for the hearts and minds” o f potential followers, support often shifts back and forth between more inclusive and more exclusive leaders. Looking at leadership competition over time in Northern
Ireland and Rhodesia/Zimbabwe, this study examines the fi'ames or strategies that are effective as leaders try to mobilize. Several strategies leaders may use are derived from literature on nationalism, international relations, and social psychology: Injustice, Enemy
Image/Ally Image, Identity, Governance, and Storytelling. For each, a single factor is most important as the root cause of mobilization. Across seven cases, content analysis is used to compare systematically the strategies of competing leaders, in terms of individual strategy profiles and in terms of an exclusivity index. The success of a leader who uses more inclusive or more exclusive strategies is expected to be contingent on the context, which is analyzed in terms of political opportunities in the domestic and international arenas. Hypotheses are derived and evaluated concerning how repression, change in stability of elite alignment, international involvement/mediation, and regional integration affect the likelihood that a leader with more inclusive/exclusive strategies will win.
This research makes substantial contributions. First, in a validation of the method, leaders in fact use the strategies derived from the literature. Further, both the use and success of the strategies are affected by the match between the more “objective” context and the leader’s interpretation of that context. This research also demonstrates that external actors can not only push a leader toward more inclusive strategies by broadening the leader’s constituency, but can also alter the context in a way that makes the leader’s domestic audience more or less “susceptible” to the inclusive strategies. Finally, this study shows how leaders manipulate aspects of the international and domestic contexts to mobilize domestic audiences.
m ACKNOWLEDGMENTS
As with any long project, there have been many phases of work on this dissertation; I would like to thank several people who helped me throughout most of them. To my committee chair, Margaret Hermann, I owe thanks for help all along the way. Her vision that my ideas were so important to the study of leadership kept me going whenever I was wilting. Also, she went to bat for me several times as I sought funding for my research; without her help the trips to Belfast, the Hoover Archives, and the Rhodes
House Library would have been much more difficult. Finally, I want to thank Peg for her mentorship as a woman in political science and for just being a supportive ôiend. 1 want to also thank my committee members Richard Herrmann and Donald Sylvan. Professor
Herrmann’s constructive criticism has held my feet to the fire more than once, but has always helped me improve my own thinking. His enthusiasm for research more generally has always been encouraging. Professor Sylvan deserves many thanks for his commitment to me and to all of his graduate students; I have always been amazed by and grateful for his careful, precise comments— especially in the early and final phases of this study. 1 would also like to thank the librarians at the Linen Hall Library and Hoover Archives for their help, and the Mershon Center for travel funding.
IV On a more personal note, my family, John, Gail, and Jay Grove, have kept me going in the darkest days— especially because they have always believed in me during those times when I have not beheved in myself. Neal Carter has been a tremendous friend—and especially patient when I was bouncing my early ideas off of him. Finally, my best friend, Chris Scholl, has kept me from going insane even while I was making him crazy. I thank him for always being willing to get out the map and go climb a mountain with me whenever the terrain in Columbus got too rough. VITA
October 20, 1969...... Bom - Scott Air Force Base, EL
1992 ...... B.A. Political Science, University of Georgia First Honor Graduate, summa ciim lande
1993 ...... M.A., International Relations and Strategic Studies, Lancaster University, UK
1993...... University Fellowstiip, The Ohio State University
1994-present...... Graduate Teaching and Research Associate, The Ohio State University
PUBLICATIONS
1. Andrea K. Grove. Book Review. The Origins o f the Present Troubles in Northern Ireland (by Caroline Kennedy-Pipe). In Nationalism and Ethnic Politics Vol. 3, No. 4 (Winter 1997).
2. Andrea K Grove. Book Review. The Palestinians: The Making o f a People (by Kimmerling and Migdal). In Nationalism and Ethnic Politics Vol. 2, No. 4 (Winter 1996).
FIELDS OF STUDY
Major Field; Political Science
VI TABLE OF CONTENTS
Page Abstract...... ii
Acknowledgments ...... iv
Vita...... vi
List o f Tables...... xi
Chapters:
1. Introduction ...... 1
1.1 Overview ...... 1 1.2 Significance for the study of International Relations ...... 6 1.3 An integrative approach ...... 9 1.4 The leadership focus ...... 11 1.5 Leadership as persuasion ...... 14 1.6 Social movements literature and mobilization ...... 16 1.6.1 Framing processes ...... 17 1.6.2 Leader mobilization strategies ...... 19 1.6.3 Nationalism literature and strategy models ...... 19 1.6.4 The five strategies...... 21 1.7 The influence of context...... 30 1.7.1 Political opportunities ...... 30 1.7.2 Mobilizing resources ...... 34 1.8 Conclusion ...... 35
2. Research design and methodology ...... 36
2.1 Introduction ...... 36 2.2 Scope conditions ...... 38 2.3 Case selection ...... 40 2.4 Defining the dependent variable: Mobilization episodes ...... 45 2.5 Content analysis ...... 46 2.5.1 Issues in content analysis ...... 46 2.5.2 Selection of texts ...... 50
Vll 2.5.3 Unit of analysis ...... 52 2.5.4 Identity strategy ...... 52 2.5.5 Strategies used with Identity ...... 57 2.5.6 Intercoder reliability ...... 63 2.5.7 Analysis ...... 64 2.6 Inclusive/Exclusive; Categorizing the strategies ...... 65 2.7 Contextual analysis: Pohtical opportunities ...... 69 2.7.1 Change in stability of elite alignment ...... 70 2.7.2 State’s capacity and propensity for repression 76 2.7.3 International involvement, regional integration 78 2.8 Context and individual strategies ...... 85 2.9 Conclusion ...... 87
3. Northern Ireland: Cases and contexts ...... 92 3.1 Introduction ...... 92 3.2 Background ...... 93 3.3 First mobilization episode: 1982-1983 ...... 99 3.3.1 The competitors and the campaigns ...... 99 3.3.2 The context ...... 107 3.3.3 Expectations about individual strategies ...... 114 3.4 Second mobilization episode: late 1985-1987 ...... 115 3.4.1 The competitors and the campaigns ...... 115 3.4.2 The context ...... 118 3.4.3 Expectations about individual strategies ...... 124 3.5 Third mobilization episode: 1996 ...... 125 3.5.1 The competitors and the campaigns ...... 125 3.5.2 The context ...... 134 3.5.3 Expectations about individual strategies ...... 140 3.6 Fourth mobilization episode: 1997 ...... 141 3.6.1 The competitors and the campaigns ...... 141 3.6.2 The context ...... 145 3.6.3 Expectations about individual strategies ...... 150 3.7 Conclusion ...... 150
4. Northern Ireland: Analysis of leadership mobilization strategies ...... 153 4.1 Introduction ...... 153 4.2 First mobilization episode: 1982-1983 ...... 153 4.2.1 Expectations ...... 153 4.2.2 Comparing strategies...... 154 4.2.3 Assessing the predictions: The exclusivity index ...... 157 4.3 Second mobilization episode: late 1985-1987 ...... 165 4.3.1 Expectations ...... 165 4.3.2 Comparing strategies...... 166
viii 4.3.3 Assessing the predictions: The exclusivity index...... 168 4.4 Third mobilization episode: 1996 ...... 175 4.4.1 Expectations ...... 175 4.4.2 Comparing strategies...... 176 4.4.3 Assessing the predictions: The exclusivity index ...... 178 4.5 Fourth mobihzation episode: 1997 ...... 188 4.5.1 Expectations ...... 188 4.5.2 Comparing strategies...... 189 4.5.3 Assessing the predictions: The exclusivity index ...... 191 4.6 Looking over time: Important implications ...... 199
5. Rhodesia/Zimbabwe: Cases and contexts ...... 206 5.1 Introduction ...... 206 5.2 Background ...... 207 5.3 First mobilization episode: 1979 ...... 212 5.3.1 The competitors and the campaigns ...... 212 5.3.2 The context ...... 214 5.3.3 Expectations about individual strategies ...... 223 5.4 Second mobilization episode: 1980 ...... 225 5.4.1 The competitors and the campaigns ...... 226 5.4.2 The context ...... 231 5.4.3 Expectations about individual strategies ...... 240 5.5 Third mobilization episode: 1990 ...... 241 5.5.1 The competitors and the campaigns ...... 244 5.5.2 The context ...... 253 5.5.3 Expectations about individual strategies ...... 267 5.6 Conclusion ...... 269
6. Rhodesia/Zimbabwe: Analysis of mobilization strategies ...... 271 6.1 Introduction ...... 271 6.2 First mobilization episode: 1979 ...... 272 6.2.1 Expectations ...... 272 6.2.2 Comparing strategies...... 272 6.2.3 Assessing the predictions: The exclusivity index ...... 275 6.3 Second mobilization episode: 1980 ...... 282 6.3.1 Expectations ...... 282 6.3.2 Comparing strategies...... 283 6.3.3 Assessing the predictions: The exclusivity index ...... 285
IX 6.4 Third mobilization episode: 1990 ...... 293 6.4.1 Expectations ...... 293 6.4.2 Comparing strategies...... 294 6.4.3 Assessing the predictions: The exclusivity index ...... 297 6.5 Looking across cases: Learning from the unexpected ...... 306
7. Conclusions ...... 311 7.1 Introduction ...... 311 7.2 Comparisons and contrasts ...... 314 7.3 Contributions to work on social movements, nationalism, and international relations ...... 323 7.4 Contributions to work on framing ...... 332 7.5 Contributions to leadership studies ...... 335 7.6 The international dimension and policy relevance ...... 342 7.7 Conclusion ...... 345
Appendices:
Appendix A: Handbook for assessing mobilization strategies ...... 351 Appendix B: List of coded documents ...... 376 Appendix C: Comparison of strategy frequencies, without Storytelling ...... 379
B ibliography...... 389 LIST OF TABLES
Table Page
2.1 Relationship of Changing Elite Alignment to Favored Strategy Type, when A and B play role ...... 75
2.2 Relationship of Changing Elite Alignment to Favored Strategy Type, when A and C play role ...... 75
2.3 Relationship of Repression by Target to Favored Strategy Type ...... 78
2.4 Distribution of Values for Political Opportunities Across Cases ...... 84
2.5 Summary of Indicators ...... 89
3.1 Percentage of Electoral Support for SDLP and Sinn Fein, 1982-1997 ...... 98
3.2 Percentage of Nationalist Vote; SDLP and Sinn Fein, 1982-1997 ...... 99
3.3 Summary o f Expectations for Northern Ireland, Shift 1 ...... 108
3.4 Summary of Expectations for Northern Ireland, Shift 2 ...... 119
3.5 Summary of Expectations for Northern Ireland, Shift 3 ...... 135
3.6 Summary of Expectations for Northern Ireland, Shift 4 ...... 145
3.7 Summary of Northern Ireland Cases ...... 152
4.1 Comparison of Strategies for Shift 1...... 156
4.2 Comparison of Exclusive and Inclusive Strategies for Shift 1 ...... 158
4.3 Comparison of Targets (Injustice and Governance Negative Strategies) for Shift 1...... 161
xi Table Page
4.4 Comparison of Targets (Blame, Threat, and Combination of Enemy Image Strategies) for Shift 1...... 162
4.5 Comparison of Targets (Storytelling Strategy) for Shift 1...... 163
4.6 Comparison of Targets (Ally/Goals and Combination of Ally Strategies, Governance Positive Strategies) for Shift 1...... 164
4.7 Summary o f Findings, Shift 1 ...... 165
4.8 Comparison of Strategies, Shift 2...... 166
4.9 Comparison of Exclusive and Inclusive Strategies for Shift 2 ...... 169
4.10 Comparison of Targets (Injustice and Governance Negative Strategies) for Shift 2...... 171
4.11 Comparison of Targets (Blame, Threat, and Combination of Enemy Image Strategies) for Shift 2...... 172
4.12 Comparison of Targets (Storytelling Strategy), Shift 2...... 173
4.13 Comparison of Targets (Ally/Goals and Combination of Ally Strategies, Governance Positive Strategies) for Shift 2 ...... 174
4.14 S ummary of Findings, Shift 2 ...... 175
4.15 Comparison of Strategies, Shift 3...... 178
4.16 Comparison of Exclusive and Inclusive Strategies for Shift 3...... 180
4.17 Comparison of Targets (Injustice and Governance Negative Strategies) for Shift 3...... 182
4.18 Comparison of Targets (Blame, Threat, and Combination of Enemy Image Strategies) for Shift 3...... 183
4.19 Comparison of Targets (Storytelling Strategy), Shift 3...... 184
4.20 Comparison of Targets (Ally/Goals and Combination of Ally Strategies, Governance Positive Strategies) for Shift 3...... 186
xii Table Page
4.21 Summary of Findings, Shift 3 ...... 187
4.22 Comparison of Strategies, Shift 4...... 190
4.23 Comparison of Exclusive and Inclusive Strategies for Shift 4 ...... 192
4.24 Comparison of Targets (Injustice and Governance Negative Strategies) for Shift 4...... 194
4.25 Comparison of Targets (Blame, Threat, and Combination of Enemy Image Strategies) for Shift 4...... 195
4.26 Comparison of Targets (Storytelling Strategy), Shift 4...... 196
4.27 Comparison of Targets (Ally/Goals and Combination of Ally Strategies, Governance Positive Strategies) for Shift 4 ...... 197
4.28 Summary of Findings, Shift 4 ...... 198
4.29 Comparison of Change Between Shifts (Totals of Exclusivity Index) ...... 201
4.30 Summary of Northern Ireland Cases ...... 204
5.1 Summary of Mobilization Shifts in Zimbabwe ...... 207
5.2 Summary of Expectations for Rhodesia 1979 ...... 215
5.3 Summary of Expectations for Zimbabwe 1980 ...... 232
5.4 Summary of Expectations for Zimbabwe 1990 ...... 254
5.5 Summary of Zimbabwe Cases...... 270
6.1 Comparison of Strategies, Shift 1 ...... 273
6.2 Comparison of Exclusive and Inclusive Strategies for Shift 1 ...... 276
6.3 Comparison of Targets (Injustice and Governance Negative Strategies) for Shift 1...... 278
xiii Table Page
6.4 Comparison of Targets (Blame, Threat, and Combination of Enemy Strategies) for Shift 1...... 279
6.5 Comparison of Targets (Storytelling Strategy) for Shift 1...... 280
6.6 Comparison of Targets (Ally/Goals and Combination of Ally Strategies, Governance Positive Strategies) for Shift 1...... 281
6.7 Summary of Findings, Shift 1 ...... 282
6.8 Comparison of Strategies, Shift 2 ...... 284
6.9 Comparison of Exclusive and Inclusive Strategies for Shift 2 ...... 286
6.10 Comparison of Targets (Injustice and Governance Negative Strategies) for Shift 2...... 287
6.11 Comparison of Targets (Blame, Threat, and Combination of Enemy Strategies) for Shift 2...... 289
6.12 Comparison of Targets (Storytelling Strategy) for Shift 2...... 290
6.13 Comparison of Targets (More Inclusive Strategies) for Shift 2 ...... 291
6.14 Summary of Findings, Shift 2 ...... 292
6.15 Comparison of Strategies, Shift 3 ...... 295
6.16 Comparison of Exclusive and Inclusive Strategies for Shift 3 ...... 298
6.17 Comparison of Targets (Injustice and Governance Negative Strategies) for Shift 3...... 300
6.18 Comparison of Targets (Blame, Threat, and Combination of Enemy Strategies) for Shift 3...... 301
6.19 Comparison of Targets (Storytelling Strategy) for Shift 3...... 302
6.20 Comparison of Targets (More Inclusive Strategies) for Shift 3 ...... 303
6.21 Summary of Findings, Shift 3 ...... 304
xiv Table Page
6.22 Comparison of Change Between Shifts, Totals of Exclusivity Index ...... 307
6.23 Individual Assessment of Contextual Variables Across Zimbabwe Cases...... 309
7.1 Summary of Northern Ireland and Zimbabwe Cases ...... 315
7.2 Comparison of Individual Strategies and Contextual Variables Highlighted by Winning Leader ...... 320
8.1 Comparison of Strategies without Storytelling, Northern Ireland Shift 1...... 380
8.2 Comparison of Strategies without Storytelling, Northern Ireland Shift 2 ...... 381
8.3 Comparison of Strategies without Storytelhng, Northern Ireland Shift 3 ...... 382
8.4 Comparison of Strategies without Storytelling, Northern Ireland Shift 4 ...... 383
8.5 Comparison of Strategies without Storytelling, Rhodesia Shift 1...... 385
8.6 Comparison of Strategies without Storytelling, Zimbabwe Shift 2 ...... 386
8.7 Comparison of Strategies without Storytelling, Zimbabwe Shift 3 ...... 387
XV CHAPTER 1
INTRODUCTION
1.1 Overview
In the past decade, profound changes in the international system have focused
scholarly attention on issues of identity. How do societies achieve a sense of identity?
What factors affect change in identity? Is it possible to have a strong sense of national
identity without a strong enemy? Can people develop a new sense of identity?
Given the current range of challenges on the U.S. foreign pohcy agenda, decision
makers urgently need answers to these questions. Several examples convey the
importance of these issues. First, news coverage has documented Russia’s post-Soviet
struggle to develop a new “national idea that can define its essence and inspire its citizens’
(Gordon, 1998). In light of this “struggle,” the United States has devoted substantial
resources to support Yeltsin and send the message that the exclusive rhetoric of a
Zhirinovsky should not be the language that resonates with Russians in the late twentieth century. Second, the South Afiican government still struggles to redefine that country’s national idea in an inclusive fashion after years of “South Afiica” being synonymous with
“exclusivity.” International opinion has sought to bolster the “rainbow” vision with economic and political support. Third, Western policymakers spend precious resources to recreate and redefine a multinational—or inclusive—Bosnia, without evidence based on systematic, empirical research showing whether identities are malleable in a way that can facilitate the nation-building process. Even more importantly, policy makers need evidence that actors in the international environment can afifect the internal debate over the definition of identities and interests.
This study seeks to make a substantial contribution by specifying bow the post colonial and international context affects the debate within nationalist groups. When there is uncertainty within a nation or a nationalist movement over the meaning of group boundaries and the “mission” the nation should pursue, competing leaders enter into a
“battle for the hearts and minds” of the nation. In response to these battles, support for competing leaders often shifts back and forth. This project looks at this kind of leadership competition over time in Northern Ireland and Rhodesia/Zimbabwe, asking, “What are the frames or strategies that are effective as leaders try to mobilize people around their ideas?” A second important question involves the context: Are there particular aspects of the international and domestic contexts which promote the success of either more exclusive or more inclusive strategies? For the international context, the roles of two major variables are explored: (1) international involvement/mediation by states, international organizations, or non-governmental organizations; and (2) regional integration. In the domestic context, the effects of two variables are investigated: (1) State repression; and (2) change in the stability of elite alignment, including the relationship between the controlling power and the former colonial power' and the relationship between the controlling power and the regional power^ (McAdam, McCarthy andZald, 1996).
A structured, focused comparison (George, 1979) of seven cases is employed here to build theory about leaders’ mobilization strategies and the conditions surrounding their success."* Cases include periods in these countries in which leaders competed and one of them benefited fi'om a shift in electoral patterns fi'om the previous period. The study then looks back to investigate the strategies that were used and how particular international and domestic variables affected the contest between the leaders. These countries were chosen because they are characterized by protracted but dynamic social conflicts which involve struggles by political minorities, and are fed to varying degrees by previous experience as
(British) colonies. In addition, for these countries there is a wealth of information both documenting mobilization attempts of competing leaders over time and explaining which leaders were more successful in these attempts. In the seven cases, there is appropriate variation in the international and domestic variables as well (see Table 2.4).
Secondary sources that report and interpret electoral shifts are employed to define the time periods in each country where leaders both succeeded and failed in their efforts to
‘ For Northern Ireland, this includes the unionists/Protestants and Britain; for Rhodesia tliis is the whites/Rhodesians and Britain; for Zimbabwe tliis is the Mugabe govenunent and Britain. The terms "controlling power,” "regional power,” and “colonial power” are explained in Chapter 2.
■ For Northern Ireland, this is the unionists and the Republic of Ireland; for Rhodesia this is the wliites/Rhodesians and South Afiica; for Zimbabwe this is the Mugabe government and South Africa.
^ The way in which the term “success” is used in this study is explained in Chapter 2.
3 win support from a particular group. For these seven cases (1982/83, 1986/87, 1996, and
1997 for Northern Ireland; and 1979, 1980, and 1990 for Zimbabwe), the method o f
content analysis is used to compare systematically the strategies of competing leaders.
Five strategies leaders may use are suggested by theoretical perspectives on ethnic
conflict, mobilization, and social identity theory; Identity, Injustice, Enemy or Ally Image,
Governance, and Storytelling. Each strategy highlights a different aspect of a situation
leaders tend to emphasize (for example, similarities and differences between groups,
relative deprivation, threat, the effects of political institutions). The leader strategies are
compared in two ways. First, the particular strategies they used are investigated (does one
leader emphasize injustices and another emphasize threats from an enemy?). Second, the
strategies are classified on a scale from inclusive to exclusive (with regard to other groups
in the society), and leaders are compared on this dimension (is a particular leader very
exclusive while another is inclusive?).
The success of a leader who uses inclusive or exclusive strategies is expected to be
contingent upon the situational context, which is analyzed in terms of political
opportunities in the domestic and international arenas. Hypotheses about the conditions
under which these different kinds of appeals (the particular strategies, then in terms of the
exclusive/inclusive dimension) are successful in mobilizing people are derived and tested.
A central contribution of this research is a greater specification of the role of the
international dimension of opportunities, that is, how international involvement and
regional integration affect leaders’ mobilization efforts. Indeed, recent work has begun to recognize the significance o f international actors in conflicts over national identity; the research findings here address the question empirically and show that the international community can play a pivotal role. A greater understanding of how the leaders themselves interpret and use the international setting to pursue domestic goals is also achieved.
After comparing the strategies competing leaders used in these cases, this research will provide answers to the following questions; (1) Are the strategies suggested in the extant literature in fact employed by leaders? (2) Are there differences in the combinations of strategies used by effective versus ineffective leaders, and are there patterns in these results that suggest particular “winning strategies?” (3) In the cases where leaders were effective, were the strategies generally exclusive, or were there times when more inclusive strategies can “work”?; and (4) Are there certain aspects of the context (both international and domestic) present when leaders with exclusive strategies won? Are certain aspects of context present when leaders with inclusive strategies won? In approaching these questions, this research integrates work from political psychology on framing, persuasion, and identity; sociology literature on social movements; political science literature on leadership, nationalism, foreign policy, and (in terms of questions asked) constructivist international relations.
The remainder of this chapter elaborates the framework in terms of this integrative approach, discussing what each body of literature offers. Chapter 2 discusses the research design and the methodology applied to the case studies. The “methods” chapter also includes several hypotheses about the relationships among the strategies and the contextual variables (types of opportunities). Chapter 3 describes the mobilization shifts in Northern Ireland and discusses the context for these four cases, stating the hypotheses about whether the political opportunity variables favor more inclusive or more exclusive strategies. Chapter 4 presents the results of the content analysis for the Northern Ireland cases and evaluates the hypotheses stated in Chapter 3. Chapters 5 and 6 focus in the same way on the Rhodesia/Zimbabwe cases. The concluding chapter compares the findings of the case studies and discusses implications for theory and for further research.
1.2 Significance for the Study of International Relations
In recent years. International Relations scholars in several subfields—from security studies to conflict resolution to the domestic sources of foreign policy—have called for more empirical research about political identity construction and manipulation
(for example, see Krause and Williams, 1996; McSweeney, 1996; Mercer, 1995; Risse,
1998; Waever e t a i, 1993; Wendt, 1994). Common questions deal with the sources of identity for states and societies, how and why identity may change, and how identity may affect and is affected by interstate relations.
While scholars are asking these questions, there is little agreement on how they should be answered. This study draws on a subset of literature in international relations to formulate several questions and then turns to other bodies of literature to build a framework for empirical research. Constructivist scholars are especially helpful in asking questions about the importance of identity (for example, Krause and Williams, 1996).
Although the “constructivist” label encompasses a diverse array of perspectives, this project draws on a particular idea of the Copenhagen school, which distinguishes between state security, concerned with sovereignty, and societal security, focused on identity (Buzan, 1991; Waever et al., 1993). They make problematic the assumption that a state
and the society within its purview have a high degree of cohesion, a view that makes the
current state of the global system less puzzling (for a statement of the view that the world
is destined to disintegrate into ethnically defined smaller units, see Fuller, 1997).
This perspective argues that the current era is marked by a “renewed process of
decolonization” that has been changing the participants in the state system. The societal
security school of international relations argues that, despite widespread acceptance that
this process is over, “the principle of self-determination stiU has the power to destroy
weak territorial states v/hose populations see them as empires” (Waever et al., 1993,
p. 13). The perception of substantial segments of a society that the state imposes
unrepresentative rule damages societal security or the “binding idea” that is a necessary
component of a strong state (Buzan, 1991). When it becomes unclear who makes up the
nation, “when the criteria for differentiating the inside of states fi-om the outside become
blurred and ambiguous,” the internal sovereignty of the state and the state’s very existence become threatened (Doty, 1996, p. 122). These are crucial questions, but in fact three central issues remain unexplored: when can analysts expect challenges to this “binding idea,” what strategies used by challengers tend to be successful, and what role can other actors in the international system play in promoting or preventing these challenges?
These remaining questions are addressed in the present study by examining an important piece of the puzzle: the little understood role of political leaders in the dynamics of collective identity mobilization. The view taken here acknowledges the contributions of constructivist perspectives, but aims for a more systematic, empirical approach to research than deemed possible or desirable by constructivists. Leaders are an
undervalued focus in most work which looks at this process, even when it is not addressed from a constructivist viewpoint. In fact, in uncertain or “ambiguous” times, leaders often emerge to capture peoples’ imaginations. How do leaders successfully translate aspirations into a political platform and mobilize potential followers to support it? What role do they play in framing vague societal norms to point to particular political actions?
By bringing into focus the relationship between leaders, followers, and historical contexts to address questions also asked by scholars of security and nationalism, an avenue is opened for a political psychology approach (for a statement of the importance of leadership in post-Cold War international politics in general, and the contributions to the study of leadership by decision making approaches, see Hermann and Hagan, 1998).
Drawing upon the perspective and methods of political psychology, this study addresses these questions about leadership and identity more systematically and empirically than previous approaches/
This project investigates the strategies leaders have used to mobilize populations or segments of a population around a difrerent vision of identity. Further, showing the strategies ofcom peting leaders demonstrates how the construction of the situation that became accepted could have been difrerent if another leader had been successful. This
^ [ realize that not all of these approaches regard systematic empirical work as a value. Thus, the critical theory literature may fall short on these criteria because it is not a goal of their work. As for most empirical work on nationalism, it tends to fall short because it is overly general (with anecdotal evidence) or focused on specific case studies. The latter results in the inability for the scholars to make any kind of generalizations. This project calls for a middle groimd, acknowledging the importance of specific contexts, but calling for the necessity to look at how certain variables may be affected by different contexts.
8 research contributes to international relations in several ways. Scholars have shown that the constitution and reconstitution of states, based on the political use of identity, can have significant effects on international and regional politics (Doty, 1996; Krause and Williams,
1996). Any steps toward a better understanding of this mobilization process adds to our knowledge.
Moreover, we have little knowledge about the role of third parties and more general international norms in the mobilization process, particularly how these factors may alternatively constrain and empower leaders to mobilize individuals into groups with particular missions. Each of the cases—four episodes in Northern Ireland, three in
Zimbabwe— involve active third parties and other external pressures. Finally, the findings of this project speak to literature that seeks to understand the “second image reversed”
(Gourevitch, 1978). Indeed, the end result is a better understanding of the relationship between international factors (specifically, international involvement or mediation and regional integration) and domestic politics.
1.3 An Integrative Approach
This project draws on and contributes to work in several areas in addition to the central concerns of international relations work noted above. From the perspective of the nationalism literature, this research provides a means for understanding how applicable the assumptions of several prominent theories are to recent cases having an ethnonational dimension. This researcher has derived strategies that leaders tend to use from five popular theories that speak to facets of collective behavior. Since each theoretical perspective emphasizes specific structural and situational variables as having the most explanatory value for the phenomena o f identification and/or mobilization, it is argued that each implies different situational aspects emphasized by leaders.
This project especially contributes to the development of concepts in social movements hterature and leadership research. Indeed, since the concept of “frame” was introduced to the study of social movements (for example, see Snow et al., 1986), there has been little progress in developing types of frames or systematically comparing the use of frames in mobilization processes (for an exception, see Gamson, 1992). Moreover, scholars in this area have emphasized the need to understand better the international and domestic contexts within which different frames may be used effectively (Oberschall,
1996; Zald, 1996). This research tackles these tasks directly.
Research on leadership has also been moving toward greater understanding of the kinds of messages that tend to be effective in mobilizing followers, while persuasion research in psychology explores how messages and their sources affect attitude change. In the former area, Shamir and his colleagues (1994) have even posited a theory focused on the rhetoric of the charismatic or transformational leader. The present study should help lend greater specificity to the more general, sometimes vague hypotheses in the work on transformational leadership, while integrating them with the concepts of power and politics from the international relations field. The following sections discuss the ways in which these various literatures inform the current research, beginning with leadership.
1 0 1.4 The Leadership Focus
Leadership can be a powerful factor in constructing identity. As Foucault (cited in
Kertzer, 1996) notes, the abihty to lock people into certain identities is the kernel of the
exercise of power. What gives leaders this kind of power? In addressing this aspect of
the research questions posed here, it is helpful to consider models o f leadership. Literature
about leadership tends to differ in the emphasis given to leader characteristics, audience
characteristics, the relationship between leaders and followers, and context (Hermann,
1986). Because the goal here is to understand the latter (relationship between leaders,
followers, and context), special attention is paid to the interaction o f these elements.
There are several studies that do advocate a more specific focus on the rhetoric of
leaders and how rhetoric, audience, and context interact (Shamir, Arthur and House,
1994; Smith and Smith, 1994). In a study on the rhetoric of the American presidency,
Campbell and Jamieson (1992, p. 5) emphasize that presidents do not merely adapt to
their audiences; they transform those who hear them into the audience the leader desires.
These authors argue that Presidents constitute the people with their speeches, persuading audiences to see themselves in ways compatible with the leaders’ goals and views of government. This requires the leader to skillfully assess the general “tenor of the times.”
In the context of this argument, one scholar (cited in De Waal, 1990, p. 43) said about
African leaders that “an Afiican nationalist leader has to speak courageously and uncompromisingly what the people are whispering...” Another (De Waal, 1990, p. 44) asserts, “Words are deeds; they change and reorder reality.” These ideas inform this
11 study, but still more empirical and comparative work pulling these insights together is
needed. Further, most research on leader rhetoric is limited to the American context, and
within that, to the President.
When the question of the power of language employed by leaders is broached, it is
important to note the theoretical contribution made by literature within the field of
linguistics—especially since much of the work in that field has preceded more recent
pohtical science research that investigates the use of language as a pohtical act. For
example, Bourdieu's (1991) work argues that the political arena is the “site” where agents
engage in the formation and transformation of their visions of the world. In this process,
the world itself is formed and transformed. As summarized by an editor (Thompson in
Bourdieu, 1991, p. 26) of Bourdieu’s collective essays;
Through the production of slogans, programmes, and commentaries o f various kinds, agents in the pohtical field are continuously engaged in a labour of representation by which they seek to construct and impose a particular vision of the social world, while at the same time seeking to mobilize the support of those upon whom their power ultimately depends.
In fine with the view taken here, pohtical language cannot be treated as separate firom the
context out of which it emerges. The science of narratology within the linguistics
hterature also emphasizes the power of language in creating and privileging one
construction or version of reality over others (for example, see Bal, 1991). While these
perspectives point out that and explain why an understanding of rhetoric is crucial to an understanding of pohtical acts, this work does not actuaUy enter into the empirical world of politics to develop and test propositions about the imphcations of their views.
12 As noted by the linguistics view and by some of the leadership literature mentioned
above, the significance of leaders’ messages depends on context. In some contexts, an
individual leader may not be able to make a difference. The research focus here is hmited
to cases where the contextual conditions make the political environment ambiguous and
therefore “ripe for restructuring” (Greenstein, 1987, p. 41; Hermann, 1976). ^ Building on
Hargrove (1989, p. 58), who argues that state and society are “held together in part by
shared meanings, language, symbols, and purposes,” this research shows how leaders play a crucial role in cultivating and seeing to the spread of these meanings during periods of flux. For example. Bums (1978, pp. 117-118) argues that there are times when potential followers are “taught” what is best for them:
Hope (looking forward with desire and with belief in possibility) can readily be escalated into aspiration (eagerly and ambitiously desirous of a higher goal) by leadership and other socializing influences... Leaders can also play on fears of failure. Leaders can, in turn, help convert hopes and aspirations into sanctioned expectations. Expectations carry more psychological and political force than hopes and aspirations. They are more purposeful, focused, and affect-laden; the expectation is directed toward more specific and explicit goals, ones that are valued by the builder of expectations... Leadership plays an even more consequential role in converting economic and social expectation into p o litica l demands.
This perspective rejects a deterministic approach that follower characteristics and/or situational conditions dictate the political platform: instead a number of different options can be construed as meeting more abstract expectations. An effective leader must
’ These conditions are elaborated in tlie next chapter.
13 skillfully persuade the public to accept his “definition of the situation.” The literature on persuasion from the field of psychology sheds some light on the process by which leaders mobilize aspiring publics.
1.5 Leadership as Persuasion
The leadership process described above clearly involves the art of persuasion.
Because of this connection, persuasion literature was consulted. It offers some information about the sources of messages (here, leaders) in situations where group membership plays a role. However, as with the work on leadership, this literature indicates that more research about the kinds of messages used to persuade is necessary.
Petty and Cacioppo (1981; 1986) have posited the Elaboration Likelihood Model of persuasion (ELM). This model builds on the research program concerned with attitudes in general, but more specifically addresses the debate about how people may be influenced to change their attitudes. Petty and Cacioppo (1986, p. 2) wanted to go beyond the stalemate over the issues of “if when, and how the traditional source, message, recipient, and channel variables affected attitude change.”
The ELM posits that conflicting research findings on the effects of these variables can be explained with a general model consisting of two “routes to persuasion.” The central route to persuasion is likely to occur when the individual carefully evaluates the merits of the information presented, while the peripheral route is more likely to occur as a
14 result of a simple contextual cue (such as the source of the argument being attractive).
When the central route of persuasion is employed, the change in attitudes is more enduring.
In terms of this framework, pohtical science hterature assumes that a known leader, as an authority figure, provides a cue to people that would encourage processing by the more peripheral route. However, leaders often influence enduring change in attitudes as weU; it is necessary to push pohtical scientists researching issues o f influence to take a more complex view. Psychologists Petty and Cacioppo (1986) and pohtical scientist John Zaller (1994) both show that people are more persuaded by a source they see as representative of the group of which they are a member; this finding is less helpful when several leaders from the same ingroup are competing, however. The crucial question then becomes, how can we know which person’s “spin” wiU win? Is it something about the message? Is the pubhc using some criterion to judge which of the two competing leaders is more “prototypical” of the group?
In the process of exploring the issue of group membership effects, McGarty et al.
(1994) show that when group memberships are relevant to a judgment individuals are asked to make, they engage in thoughtful processing. This means that they carefully consider the arguments that are presented. Since leaders often use references to group memberships, this may explain how leaders may induce enduring change and avoid simple processing based on the cue.
Therefore, when multiple leaders compete to persuade, it becomes clear that we really need to explore the message itself. Unfortunately, the literature on persuasion is less
15 helpful here; “One of the least researched and understood questions in the psychology of persuasion is ‘what makes an argument persuasive?’” (Petty and Cacioppo, 1986, p. 132).
The same is the case for the p o litics of persuasion. Thus far this literature review has reinforced the point that the work of many scholars ends up at the same unanswered questions. Pulling together some of this work points us in a promising direction. Indeed, some of the basic insights of persuasion research should be integrated with work in social movements literature that has recently begun to address this question of persuasive appeals’ contents.
1.6 Social Movements Literature and Mobilization
Literature on social movements is relevant to this line of research because it too considers the appeals of (movement) leaders, and adds the insight that context is crucial.
First, work in this field has explored the role of “framing processes” in case studies, showing that when social movement leaders compete to mobilize people, the key to success often lies in the way in which one competitor is able to “hit a nerve” with a particular slogan or definition of the situation. Second, social movements literature emphasizes that the success of certain frames in mobilization contests is affected by two contextual variables: political opportunities and mobilizing resources (McAdam,
McCarthy and Zald, 1996). This three-tiered framework is adapted for this dissertation, with “framing processes” operationalized as “leader mobilization strategies.”
Opportunities and mobilizing resources provide a comprehensive way to organize contextual variables, which have implications for the effects of different kinds of frames,
16 or strategies. This section derives the mobilization strategy concept from the work on framing, and the next section explains how the roles of opportunities and mobilizing resources are conceptualized. In the case studies, the mobilizing resources were rarely significant. Though they are described in the backgrounds to each campaign, they are not included as independent variables in this study.
1.6.1 Framing Processes
This dissertation draws on the basic idea that frames are a key factor in mobilizing people (Snow et al., 1986; Zald, 1996). Scholars who study framing processes argue that political opportunities and resources (see below) only offer the potential for mobilization.
A key mediating factor is “the shared meanings and definitions that people bring to their situation.” People have to have a grievance about some aspect of their lives and the optimism that collective action in which they participate can address the problem.
“Conditioning the presence or absence of these perceptions is that complex of social psychological dynamics—collective attribution, social construction— ...”, which have been labeled framing processes by Snow at al. (1986). A more narrow definition describes framing processes as the “conscious strategic efforts by groups of people to fashion shared understandings of the world and of themselves that legitimate and motivate collective action” (McAdam, 1996, pp. 5-6).
Work in political psychology has also shown the importance of framing for how people view particular issues (Iyengar and Kinder, 1991; Nelson, Clawson and Oxley,
1997). Defining framing as a “process by which a communication source, such as a news
17 organization, defines and constructs a political issue or public controversy,” Nelson et al.
(1997) use experimental methods to show the impact of alternative news frames on
tolerance for activities of the Ku Klux Bilan. This project draws on the fact that this
previous research has shown the importance of fi’aming, but then takes a slightly dififerent
angle than most of the extant political psychology literature since the focus here is on
competing leaders’ efforts to win out in the fi-aming process. Indeed, while the social
movements literature points out that leaders get involved in "fi-aming contests” as they
compete to be the voice of the movement, this subfield also falls short of understanding
the competition (McAdam, McCarthy and Zald, 1996, pp. 269-270).
In sum, little systematic work or comparative research has investigated firaming
processes or the relationship among fi-aming, political opportunities, and mobilizing
resources—even though there are many calls for such an enterprise. This dissertation
tackles at least part of this task, and by doing so contributes to theory about the fi-aming
concept.^ Further, integrating the sociology work with ideas fi-om international relations
and psychology may provide additional leverage and allow these disparate approaches to
“speak to each other.” While the social movements literature introduces and applies the
“frames” concept, it may be enriched by a fuller understanding of the politics and power
relationships that go into the construction of fi-ames and the dominance of some fi-ames
over others.
^ In fact, McAdam et al. (1996: 19) state that an interesting project would be to focus on the similarities and differences in framing strategies employed by movement groups in specified coimtries; this is somewhat similar to my interests here (though with different subject matter than social movement groups). 18 1.6.2 Leader Mohïlization. Strategies
Frames facilitate the listeners’ interpretation of which aspects of a situation merit action and what kind of actions can fix the problem. This project uses the concept of
“strategies” to denote specific ways in which leaders engage in the general process of fi-aming. The hterature on nationahsm (fi-om pohtical science and sociology) and intergroup relations (firom social psychology) suggests five strategy models leaders may employ. The bases for such strategies are described in these literatures; relative deprivation theory (Gurr, 1970); image theory and social identity theory (Herrmann, 1988;
Horowitz, 1985; Tajfel and Turner, 1986); imagined communities perspectives (Anderson,
1991; Smith, 1995); optimal distinctiveness theory (Brewer, 1991, 1996); and elite consolidation theories (Gunther and Mughan, 1993; Lijphart, 1977).
The following section derives a strategy fi-om each of these theoretic approaches.
The research then explores how vahd each perspective is to leaders who are engaged in fi-aming the issues so as to gain influence over particular constituencies. Prehminary research shows that, in fact, leaders rarely focus on a single set of factors which each theory would emphasize. Instead, combinations of strategies tend to appeal-, pointing to the need for a more sophisticated integration of differing perspectives on nationalism and intergroup relations. The next section derives models fi-om the five theories.
1.6.3 Nationalism Literature and Stratesv Models
An enormous literature addresses different facets of collective mobilization, nationalism, intergroup behavior, and ethnic conflict. In developing mobilization
19 strategies, this project takes five of the more popular perspectives and asks what each implies about the reasons people become mobilized for group behavior. Each has implications for (1) the dimensions on which groups are delineated and how individuals adopt these group identities; and (2) the reasons people become mobilized for group behavior.
An adaptation of Brewer’s Optimal Distinctiveness Theory (1991, 1996), a social psychological theory, provides a systematic way to understand differences in the way competing leaders portray groups. This approach looks at how leaders suggest that people in particular groups are similar to and different fi-om others. This is elaborated below. Then the four remaining theoretical views noted above (relative deprivation, outgroup image, elite consolidation, and imagined communities) offer models for the ways in which leaders connect portrayals of groups to particular lines of policy and/or action.
Together, the derived strategies cover the aspects of any given situation which may be the focus; Who are we? Who are our fiiends/enemies? (Identity delineation)? What is our situation (Justice)? How do others affect us or relate to us (Outgroup Image, which may be Enemy or Ally)? How should we govern our relationships with others (Governance)?
What binds us together as a nation? What is our shared past and future (Storytelling)?
The next chapter elaborates on the method for detecting these strategies in leaders’ public statements.
20 1.6.4 The Strategies
Identity Strategy
According to social identity theory, the extent to which an individual identifies
with potentially relevant categories (town, state, nation, religion, gender, class) depends
on the context of interaction. Brewer’s Optimal Distinctiveness Theory (GOT) offers a
way to think about a theory of context that focuses on the needs of individuals to be
unique yet belong to groups at the same time. She argues that people judge themselves in
terms of similarity and difference and need to view the groups o f which they are members
as distinct from other groups. However, the similarity of people within a group facilitates
the establishment of bonds among members. This theory posits an optimal point at which
an individual’s needs for similarity and difference are balanced. ODT makes a contribution
to previous social psychological approaches because it asserts that certain needs and not
group size help determine which social identities become sahent to an individual in certain
situations.
Brewer is looking at individuals, using experimental methods (as a social
psychologist), and not investigating the role of leadership. However, this researcher and
other students of pohtical psychology (Carter, Grove and Kille, 1997; Carter,
forthcoming) argue that her ideas offer insight to political scientists studying pohtical
identity and leadership. Because context affects the particular memberships with which we identify at any given time, the tension between the two needs can be exploited by leaders who manipulate the context, framing aspects that may be more and less evocative in terms of making certain identities politically relevant. This is not an argument that
21 leaders know about these needs and consciously choose particular identity levels in the
effort to meet the needs of assimilation and differentiation. However, the theory offers the
general insight that the successful leader will indeed appeal to these needs at the optimal
levels.^
From this perspective, it is important to understand the various dimensions of
group membership that a leader may emphasize at any point in time; indeed, the results of
Gamson's (1992, p. 84) study of media framing in the U.S. show the importance of
"identity" frames in motivating collective action. Leaders can play an important role in
making particular social identities relevant to a political issue. National identity is often the
most salient social identity, or at least is imphcitly assumed to be by most theories looking
at international politics. However, this is an assumption which has been “overtaken by
events” in the current international system. By coding for the ways leaders depict groups
and the exclusivity/inclusivity of the identities leaders evoke, we can trace when more
local or transnational, more civic, ethnic, or religious dimensions of identity are argued to
be the most essential to a pohtical group. Also important is that this coding helps us keep
track of when various ingroups and outgroups are depicted as sharing some characteristics
with others or not. In sum, the way a leader depicts group boundaries and characteristics
is pivotal because a leader focusing on this kind of information can promote an
overarching identity, where group memberships are mutually compatible. On the other
^Several interviews revealed the language that this theory might predict if it were taken to this level. For example, Chrissy McCauley. Belfast City Council member for Sinn Fein agreed that Hume’s SDLP focused a great deal on Europe and a vision for the future, but didn’t focus enough (in contrast to her own party) on the needs of people "on the ground.” In her opinion, this lack of adequate attention to the local level and the daily lives of people was costing the SDLP in terms of support in the poor areas of Catholic Belfast in more recent elections (1996 and 1997). 22 hand, the leadership strategy may involve trying to convince the people that they cannot be loyal to several groups at once (especially ethnic nation and state). The two components of the Identity strategy (group reference and group delineation) and indicators for them are described more in Chapter 2 and in Appendix A.
Justice Strategy
The Justice strategy is derived from ideas prominent in relative deprivation theory.
The general perspective of relative deprivation theory (RDT), a structural approach, is that frustration with inequalities between/among groups is the source of conflict. This approach has been formulated and explored by political scientists (for example, Gurr,
1970), sociologists (for example, Davies, 1963) and social psychologists (for example.
Walker and Pettigrew, 1984). RDT posits that a group perceives injustice and will rebel violently against this situation when its members believe the group is being treated unfairly politically, economically, or culturally by a state. A feeling o f relative deprivation develops when people perceive a discrepancy between their value expectations and value capabilities. The former are what group members think the group deserves, and value capabilities are what the people think they can achieve; the focus is on the situation of the ingroup. When conditions in society raise expectations more than they increase the capability to realize these expectations, discontent is intensified, as is the potential for collective violence (Gurr, 1970, pp. 13-14). Although the specific focus for Gurr was political violence, his ideas have been applied to other forms o f action in defense o f the group, such as group protest that is not necessarily violent.
23 Relative deprivation theory predicts that when individual group members identify with the group, are cohesive, and perceive the group as the object of policies which grant it less status than people in a similar group (based on economic, political, or cultural status), they will be inspired to improve their situation. RDT does not theorize the link between structure, the individual, and the group, however. It is also unclear when disadvantaged group members will engage in individual or collective action to better their situation (Wright, Taylor and Moghaddam, 1990, p. 1995). “Deprivations have to be put into a common frame of reference through the activities of political elites, thus overcoming the isolation of the individual” (Lederer, 1986, p. 365). As Gamson's (1992, pp. 31-32) study on framing found, an “injustice” frame was necessary to encourage people to think of their problems in terms of group membership rather than as individuals .
Although Gurr is not known for incorporating the kinds of appeals leaders make into his model, he actually did emphasize the importance of leaders’ slogans in mobilizing groups around particular grievances (see especially pp. 200-209 in Gurr, 1970). Gurr
(1970, pp. 201-202) even posits that normative justifications for action are more likely to be persuasive when “the symbolic appeals offer plausible explanations of the sources of relative deprivation, identify political targets for violence, and provide symbols of group identification.” Thus, while his empirical focus was different, in this one hypothesis Gurr’s insight provides support for the focus on leaders’ competing appeals, as well as the integrative approach of considering not just appeals about deprivation but about identity symbols and outgroup images as well.
24 This study takes insight from relative deprivation theory and the critiques to develop a second mobilization strategy. A leader who employs a Justice strategy (which may have a negative or positive variant, see below) emphasizes grievances about the current situation of group members. The emphases of this type of appeal are (1) the salience of group membership and the cohesiveness of the group; (2) the ways in which the group ‘as group’ is subject to great injustices; (3) and the fact that the situation is especially grievous because the deprived group is equally deserving compared to the more privileged group(s). The basic theme of a leader’s appeal is, “We deserve more than what we are getting.” The subject of this strategy is not limited to the ingroup; the leader may also discuss the situation of another group in terms of injustice (or justice). A positive valuation of this strategy may also be common. This positive valuation focuses on how the group has made recent gains.
Outgroup Image Strategy (Enemy or Ally)
Research in International Relations has examined the pohtical relevance of outgroups and outgroup images (Coser, 1956; Cottam, 1986; Cottam, 1977; Finlay, Holsti and Fagan, 1967; Flerrmann, 1985, 1988; Mercer, 1995). In addition, work on nationalism and ethnic conflict emphasizes how leaders manipulate threat to the group in order to galvanize group solidarity (for example, see Gagnon, 1994/95). In this work, a bias against outgroups is generally expected, and is (not surprisingly) what marks polarized conflicts (Horowitz, 1985, Chapters 1-4). The minimal group paradigm in social psychology explores this phenomena in a laboratory setting (Tajfel and Turner, 1986), demonstrating how the mere perception of belonging to different groups will trigger
25 discrimination favoring the ingroup. Further, the work shows that the threat of a dififerent group making gains is enough to take losses for one’s own group to prevent the other from actually making these gains. Notwithstanding the power of threat, leaders may also claim that relations with outgroups present opportunities (Herrmann, 1985, 1988).
Whether any outgroup is depicted as an enemy or an ally, presenting a threat or an opportunity, depends on the purpose of the leader.
Leaders’ portrayals of other groups are used to justify their positions about relations with outgroups. Leaders using the Outgroup Image strategy concentrate on the characteristics of the outgroup and the potential effects on the ingroup of interaction with the outgroup. Drawing on work in International Relations that emphasizes other-images as allies or enemies, there are two subcategories for this strategy (one portraying the outgroup in a negative light and one portraying the outgroup in a positive hght).*
For the negative Outgroup Image strategy, the leader will emphasize other salient group(s) and why they must be opposed. The basic line of argument may be that the outgroup threatens the ingroup, is to blame for the problems of the ingroup, or has betrayed the ingroup.
Alternatively, the strategy might take on a more positive emphasis, noting similarities between the ingroup and outgroup and relating these to why they should work together. Emphases involve how goals or values are similar, how the situation of ingroup
* The work on images posits several other images ehtes may have of others. For example, Herrmann's (Herrmann and Fischerkeller, 1995) theory includes (with the Enemy and Ally images) the Imperialist the Degenerate, the Colony, and the Barbarian. For this project 1 have adapted only the Enemy-Ally dimension because it is the basic dichotomv I am interested in theoretically. 26 and outgroup are similar, and how the cultures are similar. Changes in valuation over time as well as changes in the salient outgroups will document the dynamic nature o f national discourse.
Governance Strategy
The fourth strategy draws on elite consolidation theories within the literature on democratization. Often debate among contending leaders is about institutions and governmental aspects of group relations, especially in cases where negotiations are in process or being debated. How leaders frame potential and existing structures of governance— including laws, institutions, and relationships with other elites—should therefore be analyzed.
Consociationalism, which limits the power of the majority group while dispersing power to subgroups, is an important way to moderate cleavages in plural societies
(Lijphart, 1977; Lijphart, Rogowski and Weaver, 1993; Nordlinger, 1972). The four defining characteristics of consociationalism are: a grand coalition of all ethnic groups; a mutual veto in decision making; ethnic proportionality in the allotment of particular opportunities and offices; and ethnic autonomy (Lijphart, 1977). Electoral and territorial arrangements can encourage alignments across ethnic divisions. They also affect competition of leaders within a group (Horowitz, 1985). Leaders must show how the institutional framework affects their group.
27 Whether institutions promote or moderate conflict in the long run depends partially on how leaders portray the function of institutions to their publics. “Successful” attempts to form consociational systems depend on the efforts by elites to pursue common goals, not in the particular institutional arrangements.
In the Governance strategy derived from this general perspective, the leader’s message is focused on (1) institutions of the state and/or (2) relationships with other leaders involved in the “bargaining” over how the people will be governed. He may focus on how institutions function to strengthen commitment to the larger regime by being
“responsive” and giving citizens opportunities for political participation in the given system. Another element of this strategy is emphasis on the need to cooperate with the other subgroup leaders in order to collaborate in strengthening the greater “nation.”
On the other hand, leaders may talk about the necessity of having institutions which emphasize protection of parochial interests against other groups while downplaying membership in the larger structure. Along with this, references to other subgroups’ leaders may be derogatory and emphasize the importance of not collaborating to change or maintain the government institutions. The basic theme will be, “Let’s change [or reinforce] institutions to pursue our goals. Let’s work with [or resist] others to make this government work.”
Storytelling Strategy
“Imagined communities” approaches offer insight for the final strategy hypothesized to be important; these perspectives focus on how leaders draw on aspects of shared history and a shared fate which bind individuals into a community and have
28 particular implications for action in the current situation (Anderson, 1991; Connor, 1994;
Hobsbawm, 1990; Smith, 1995). For example, Anthony Smith (1995) describes nationalist leaders as “archaeologists” who shape interpretations of the ethnic past to relate to an understanding of the present period. He implies that leaders may be successful if they can create a story of past glory and convince the group that it has a common glorious destiny. Note that the emphasis here is not on any story the leader tells but on stories that link the present people and their situation to the past.
Because leaders can and do use history differently, and some may use it less than others, paying attention to their use of the Storytelling strategy is important. The literature from which this strategy is derived implies that use of the past will enhance a leader’s chance of successful mobilization, but this is an empirical question that has not been tested systematically (Stem, 1995).^ The information emphasized by a leader using this strategy includes a characterization of the rich past and how it relates to the future.
There will be many references to a common past and common fate, national heroes, and martyrs. The leader may focus on how the past must be redeemed and/or on how it is important to avoid repeating the past.
This strategy is expected to be a “mode” of using the other strategies, which may
(hypothetically) serve to enhance the Justice or Outgroup Image appeals, for example.
This means that the emphasis may be on the injustice suffered by the ingroup, but the
though I focus on the rhetoric of political leaders, I realize that the imagined communities perspective discusses the wider range of actors who emphasize links with the past (the intellectual class, for example). However, I argue that (by way of the national story being an important part of any individual’s view) leaders pick up on the mythology, some more than others.
29 current situation wül be tied to injustices in the past, for example. Treating this as a mode is justified by work on problem representation within the foreign policy decision making hterature. This work is similar to firaming in some ways, and finds that policy makers tend to be more effective in persuading colleagues when they use the “story” model of reasoning, though again the emphasis here is on a particular kind of story (Sylvan, Ostrom and Gannon, 1994).
1.7 The Influence of Context
As leaders compete to define the identity and mission of potential followers, the context is crucial to this contest. In dififerent situations, the same strategies will have variable success in articulating the feelings of a group of people. Further, the kind of situation a leader thinks he faces may push him to use dififerent strategies from other periods. For this reason leadership scholars have argued that we must study the nexus among leaders, followers, and context, as noted above. Returning to the social movements literature, two sets of variables should be considered when comparing the contexts of mobilization attempts; political opportunities and mobilizing resources.
1.7.1 Political Opportunities
Those scholars who focus on political opportunities look at aspects of the broader political system (especially institutions) which structure the opportunities for collective action, as well as its form and extent (Tarrow, 1983; Tilly, 1978). Several kinds of pohtical opportunities are suggested in this hterature: the relative openness or closure of the institutionalized political system, change in the stability of that broad set o f elite alignments that typically undergird a polity, the presence of elite aUies, and the state’s capacity and propensity for repression. Relative changes in one or more of these dimensions provides an “opening” where a leader can advocate a mission or plan and then rally the pubhc around his/her idea (McAdam, McCarthy and Zald, 1996, p. 10). While the focus of this study (defining communal identity and mission) is less collective action than collective “ideas,” so to speak, political opportunities are Just as important in this realm.
Indeed, most situations of conflict over the identity of states emerged when external and internal shocks opened a period of uncertainty by calhng into question the legitimacy of particular institutions and/or elite ahgnments that had provided stability. We see this with the Northern Ireland and Zimbabwe cases, as well as with others such as Lebanon.
An “opening” in the institutional system is a scope condition for the cases in this study (see Chapter 2), therefore it is not included as a type of opportunity examined separately in each case. Also, since preliminary research for this project revealed little variation in the elite allies, change in stability of elite ahgnment and repression were chosen as the two domestic context variables.
As argued above, it is also crucial to consider the role of the international environment in affecting which leaders’ strategies may win. Indeed, the external context can offer both opportunities to and impose constraints on leaders. The social movements literature has recently begun to address this issue (for example, Oberschall, 1996) but
31 could benefit fi-om work in International Relations/" Further, scholars of internal conflict
have often indicated the importance of an international dimension in afiecting the dynamics
of conflict. Modelski (1964, cited in Ganguly and Taras, 1998, p. 69) noted that
incumbent leaders “always and by definition have international connections, simply
because they are in charge of the legitimate machinery of the state, which includes the
diplomatic and other international networks; the insurgents [or challengers], by virtue of
having to approximate the incumbents as closely as possible in order to supplant them,
must develop the same machinery.”
Several “exogenous factors of the global system which set the structural
parameters in the evolution of [identity] politics” (Ollapally and Cooley, 1996, pp. 480-81)
are suggested by recent work. According to Ollapally and Cooley (1996), both a material
and an ideational dimension may come into play. First, the effort to mobilize around
communal memberships is directed against a state or government that can no longer
provide economic or physical security. This material dimension explains why the status
quo weakens and opens the opportunity for mobilization. Second, an ideational dimension
affects why certain identities take precedence over others. The influence of ideas fi'om the
international arena such as democracy, capitalism, socialism, human rights, anti
colonialism, can validate or delegitimize the discourse of an aspiring leader.
Gurr’s (1993) M inorities a t R isk project also discusses the increasing importance
of the international arena for ethnonational politics. In particular, he discusses four ways
In fact, in a review of this part of the social movements literature, McAdam (1996: 35) notes, "...movement scholars have, to date, grossly undervalued the impact o f global political and economic processes in structuring the domestic possibilities...”
32 that the international dimension afifects communal conflicts, by moving actors on both sides of the conflict toward negotiation. These four factors are: (1) the spread of democratization; (2) pressures on those in power—by organizations such as the United
Nations and the Organization for Security and Cooperation in Europe—to implement international principles of nondiscrimination; (3) cost-benefit assessments by political leaders that strategies of accommodation are low-cost alternatives to strategies of forcible assimilation or repression; and (4) the shift in strategies and tactics of politically mobilized minorities to take advantage of both novel sources of international support and political opportunities provided by more responsive states (Gurr, 1993).
This study adapts these ideas that have begun to emerge in the extant literature but advances their contribution by (1) deriving more specific hypotheses about how both the material and ideational dimensions may be significant, and (2) testing and refining these hypotheses in the Northern Ireland and Rhodesia/Zimbabwe cases. In particular, two international variables are the focus in the seven case studies: involvement/mediation by states, international organizations, and non-state actors; and regional integration.
Chapter 2 derives hypotheses concerning how each of these opportunities favors particular kinds of leader strategies. Also, the next chapter presents the method for analyzing this set of variables including the indicators, sources of indicators for each case, and how effects are recorded. Unfortunately, the extant literature does not offer hypotheses about the effects of combinations of these four kinds of opportunities. For
33 this reason, an inductive look at how combinations of variables are associated with the success of more exclusive or more inclusive strategies is a major contribution of this project.
1.7.2 Mobilizing Resources
A second factor which social movements scholars find to be important is mobilization structures or mobilizing resources, such as “.. .meso-level groups, organizations, and informal networks that comprise the building blocks of social movements and revolutions...” (McAdam, McCarthy and Zald, 1996, p. 3). This grows out of earlier versions of resource mobilization theory, which argued that the presence of grievances was not enough to predict group reactions to injustice. Instead, a key element is the organizational infrastructure (both formal and informal) that enables the leadership to reach a critical mass of people (Jenkins, 1983). Leaders without a recognized party, access to the radio/television/print media, and an estabhshed network of followers will be poorly equipped to address an audience and spread their ideas.
In addition, several aspects important from a pohtical science perspective are not included in the sociology or psychology literature. For example, depending on his position in the government and in the societal subgroup of which he is a part, a politician’s mobilizing resources may include patronage networks or other channels through which to co-opt potential dissenters. In other words, mobilization may be easier if the leader can make credible promises about future material benefits. In addition, the international dimension is important; a leader may increase the legitimacy of his cause and organization
34 in the eyes of the potential domestic audience by setting up international linkages. These
links may contribute money that enables a leader’s organization to expand its contacts with potential followers. “
It is argued here that three categories of resources could be important: funding/patronage, party and campaign organization, and media access/coverage. The research for this project revealed that there was rarely change in these dimensions that was not matched by the other leader. For this reason and to simplify the framework, these variables are not discussed further except in one case where a significant change occurred to give one leader a distinct advantage (Adams in 1996 and 1997).
1.8 Conclusion
This chapter has argued that leader mobihzation strategies are a unique and important focus in the study of societal—and ultimately state—security. Exploring the kinds of strategies leaders employ to construct political identities will integrate and advance research in several areas and contribute to both conceptual and empirical development. Several relevant literatures were reviewed, including constructivist views of identity, foreign policy, leadership, persuasion, social movements, social identity, nationahsm, and intergroup behavior. The result is an integrative framework that takes insight from each of these areas yet seeks to advance that insight to understand leaders’ efforts to mobilize groups and the role of contextual variables in the mobilization process.
The next chapter discusses the research design and methodology.
' ' Here I do not include the effects of international military support for a particular leader and his/her organization. This is the subject of many extant studies, however. 35 CHAPTER 2
RESEARCH DESIGN AND METHODOLOGY
2.1 Introduction
This chapter focuses on several aspects o f the research design, which involves the use of the structured, focused comparison of several cases (George, 1979). This method involves choosing a small number of cases in accord with research designed to answer focused questions and assessing the values of the variables for each case. To summarize, the overarching research questions for which the cases are assessed are, “What combinations of strategies do leaders use to mobilize potential followers around more and less exclusive definitions of national identity, and which of these strategies are associated with successful mobilization attempts?” and “How do important variables in the international and domestic contexts affect the success of leaders employing exclusive versus inclusive strategies?” More specific questions pertaining to each variable are elaborated in this chapter.
36 Four mobilization periods in Northern Ireland and three in Zimbabwe are investigated, for a total of seven cases/ These cases were chosen based on secondary sources which describe electoral gains that shift back and forth between leaders; these were shifts occurring after the “legitimation crisis” (see below). The “success of mobilization efforts” (the dependent variable) is defined as gaining a substantial percentage of the nationalist electorate’s vote compared to the other main competitor. The method for evaluating the electoral results is elaborated later in this chapter.
The strategies were derived from a range of extant literature; this chapter will discuss the coding for the individual strategies and also categorize them in terms of how inclusive/exclusive each is. The context is conceptualized in terms of political opportunities. This chapter will derive hypotheses about four types of political opportunities whether or not each “encourages” the use and dominance of more inclusive versus exclusive framing strategies. The case study chapters describe the predictions (i.e., which strategies are expected to succeed) for each case based on these hypotheses and compares these to the results of the content analysis of the competing leaders’ strategies.
This chapter begins with a discussion of the scope conditions delimiting the study and narrowing the set of possible cases. Second, other case selection criteria are explained; the case chapters discuss the choice of the Northern Ireland and Zimbabwe cases in terms of these criteria. Next, the measurement of the dependent variable is discussed and the selection of the time periods within each country case is explained.
Fourth, the method for analyzing the strategies (content analysis of leaders’ statements) is
' The "breakpoints” in these cases are defined by secondary soiurces, so that there is generally agreement on how the mobilization shifts are perceived. 37 discussed, though the coding handbook in Appendix A describes the method in greater
detail. The subsequent section operationalizes the contextual variables (types of political
opportunities) and derives hypotheses about the relationship between the opportunities
and the degree of exclusivity that each type of opportunity is expected to encourage in w inning strategies. Finally, the method for analyzing the contextual variables is presented.
The table at the end of this chapter summarizes the variables, the indicators, and the
sources of data that were used.
2.2 Scope Conditions
Cases were chosen in which there was an ongoing debate over identity in states holding contested elections; in the conclusions it will be important to keep in mind that the research implications are limited to this type of case. In such periods of debate, it is argued that individual leaders are likely to have an impact on the political process. When there is a lack of consensus about the fiiture, some types of mobilization strategies have a high degree of resonance and bring about shifts in the orientation of the group.^ What are the conditions under which individuals are more likely to have an impact on the course of events? A general propitious condition is when the political environment “admits of restructuring” (Greenstein, 1987, p. 41; also Hermann, 1976). Further, when there is a widely shared sense o f uncertainty about the future of the state and society, issues of identity are often the focus of debate. Uncertainty may be a result of international or
■ This idea is consonant with, and an example of “political opportunities” for mobilization discussed in the social movements literature. 38 domestic changes; the central scope condition for this project is a legitimation crisis
brought about by external influences and the social mobilization of a part of the domestic
society (for example, Deutsch, 1953). These are explained briefly.
Domestic factors contribute to a legitimation crisis during demographic
transformations which change the balance between societal groups. Formerly politically
quiescent and/or dominated populations begin to grow in size, more individuals within
these populations achieve higher levels of education, and claims surface that the political
institutions of the current system are not adequately representative. Of course, these
changes may not be entirely domestic since the “demonstration effect” of political
mobilization in other countries might encourage political minorities to mobilize (for a
similar argument about mobilization for democracy, see Huntington, 1991). In fact, this was the case for both Northern Ireland and Rhodesia; the civil rights movement in
Northern Ireland patterned itself on the American civil rights movement, while the
Zimbabwe nationalist movement looked to other successful African nationalist groups.
Factors in the international arena play a significant role in affecting the balance among internal populations. Refugees from other states, territorial claims by regional neighbors, encouragement by irredentist powers, or the demonstration effect noted above may affect the claims a societal subgroup makes on the central government. Also, other types of involvement from the international arena can and have played an important role, such as structural adjustment programs of the International Monetary Fund and Cold War- linked policies of the superpowers. Both of these examples often “created” support for and against regimes that may not have been sustained in another global context (see
39 Brown, 1996). This research looks at cases where, because of both domestic and
international factors, political minorities began to demand reconstitution of the state; the status quo was no longer working, meaning the government lost the legitimacy necessary to avoid crisis.^
2.3 Case Selection
As described in the previous chapter, the configurations of leadership strategies and the political opportunities are the principal foci as the independent variables. So that inferences may be made about the effectiveness of various strategies, each case must show observable competition between leaders. Because a point of interest is how the leader reacts within and interprets different contexts, the design does not attempt to control for context across time periods within each case. However, the cases do show variation on the two international and two domestic context variables (see below).
The method used to select for variation on the dependent variable is less conventional; it is meant as an exploratory way of addressing the difficulties of measuring how messages resonated with their audience in historical cases. Ideally, an instrument would be created to measure the identity profile of a society in terms of how inclusive and exclusive (defined fi'om a particular perspective) various group boundaries are (especially the national identity boundaries). This approach is intractable, especially for investigating
^ I do not want to imply that the previously quiescent or dominated segments of the population ever saw the government as legitimate. Instead, the government being questioned may have maintained rule by means of coercion. If this was the case, the crisis of legitimacy may have arisen because the use of force was no longer feasible given the extent of protest, external powers may be warning against the consequences of use of coercion, etc.
40 historical cases. There is potential for such an approach on contemporary cases; the
“identity exclusivity” concept may be operationalized with a combination of survey
instruments, thus drawing on the tools of somewhat similar work in social psychology
Instead, for this study, secondary sources are employed to choose cases where
leaders competed and one leader won. It is common for there to be competition between
two or more leaders to lead nationalist movements, and it is equally common that one
particular leader is able to gain the upper hand during any given period. Sources are
elections data and narrative accounts^ which show substantial shifts toward and away from
pohtical parties. Then, the leaders’ strategies employed during these shifts are analyzed.
Given that the scope conditions have narrowed the field to cases in a reconstitution phase,
these shifts are often addressed in case hterature as significant puzzles. Because the
phenomena of nationahsm and identity are dynamic, cases include episodes where people
followed leaders calling for an inclusive, integrative country and concomitant institutions
and cases where they foUowed leaders caUing for an exclusive country, defined in a way that excludes other groups within the borders. So that a larger number of cases can be
compared, four time periods within Northern Ireland and three in Rhodesia/Zimbabwe are
investigated. These cases exhibit this shift between inclusive and exclusive ideas. By
showing the variation within cases across time, this approach rejects the notion that
■" An approach using opinion surveys was recently published (Thaler, 1997); Marilyim. Brewer and her students at Ohio State University’s Department of Psychology have constructed experiments using surveys in order to measure social identity.
^ The indicators vary slightly by availability for the different cases.
41 particular institutions (such as majoritarian or consociational structures) have an independent, constraining effect on the possible strategies leaders may use, a popular claim among scholars o f segmented party systems (Horowitz, 1985; Mitchell, 1995).
Another category of case selection criteria involves the kind of data this analysis requires. As implied above, substantial election and polling data is necessary to identify the dependent variable values. Second, the focus on content analysis of leadership appeals limits the study to cases where documents are available. Specifically, a substantial number of leaders’ speeches and interviews made around the named time periods are required.
Based on these criteria, the cases selected are time periods in Northern Ireland^ and Rhodesia/Zimbabwe. These cases firom different continents and cultures may impart a sense of the generalizahility of certain types and combinations of leadership strategies. In addition, the variation in types o f communal rivalries (religious and arguably national in
Northern Ireland, and racial and ethnic in Rhodesia/Zimbabwe) will allow an indirect exploration of whether these tyqies make an actual difference in the strategies observed.’
Third, these countries share a history of colonial domination but differ on whether the people leaders seek to mobilize are a minority (Northern Ireland) or a majority
(Zimbabwe) in the society; whether this difference affects mobilization strategy use and
^Selection of the Northern Ireland case although it is not a sovereign state is in line with the way the general literature has treated it. For example, Lijphart (1977, pp. 134-135) states, “Although it is not a sovereign state, it [Northern Ireland] may be considered on a par with independent countries because from 1921 until the imposition of direct rule from London in 1972 tlie powers of the Stormont (the Northern Ireland parliament) were so generously interpreted that they far exceeded the usual powers of subnational govenunental units even in federal states.” Lijphart also quotes a descriptor of the province from Budge and O’Leary (1973: 143): “...a self- governing province with some of the trappings of sovereignty.”
For example, one comparative study concludes that the fundamental loyalty mechanisms at work in all these kinds of membership are similar (Smock and Smock, 1975, p. 306).
42 effectiveness is explored. Although it is recognized that this one study cannot make definite conclusions on these differences, hypotheses will be derived and can be explored in future research.
Both Northern Ireland and Zimbabwe are constituted of people identifying with groups in existence before the pohtical entity was created, and both have historic divisions.
Further, people in both cases have a number of identities to which leaders may appeal. In
Northern Ireland the perceived communities are nationalists, who tend to be Catholics and identify with Ireland, and unionists, who tend to be Protestants and identify with Britain.
There are also perceived groups within these larger groups; republicans within the nationahst community and loyahsts within the unionist community. Other divisions are also possible, including class and party divisions. In Zimbabwe, perceived groups are ethnic and racial: blacks, whites, Asians; Shona, Ndebele. Again, many other divisions are possible, including provincial differences and the strong identifications with kinship subgroups within the Shona. This research will explore how these were used in the mobilization process, since in fact any o f the above can be fi-amed as more or less inclusive.
Related to the range of possible identity bases, both sets of cases have very significant international components, including the involvement of foreign powers (for example, Ireland and South Afiica). Throughout all the time periods in both countries, international factors also affected elites’ incentive structures, though they did not
43 determine the mobilizing strategies employed (British, Irish, and American involvement in
Northern Ireland; British, South African, American, and Front Line States involvement in
Rhodesia/Zimbabwe).
Finally, these cases meet the scope conditions, as the case chapters describe.
Briefly, in the 1960s, each experienced demographic transformations whereby the
dominated groups grew rapidly, socioeconomic development proceeded and encouraged
urbanization, and the formerly quiescent groups began to undergo politicization. The
states moved closer toward legitimation crises where the identity of the country (as Irish,
British, or uniquely Northern Irish; as Rhodesian or African or uniquely Zimbabwean) was
the focus of debate. During these time periods, contested elections were held.
Northern Ireland may even be considered a critical case; it is ranked among the
world’s most intractable problems because of the failure to build a unifying sense of
nationhood that the major societal groups consider legitimate. Observers tend to focus on
the inevitability of this conflict, because the party and voting system dictated that extremist
leaders would always succeed in outflanking any moderates (MitcheU, 1995). But,
interestingly, by studying the case over time and comparing the rhetoric o f moderate and
extremist leaders on the nationalist side, the research here may show that moderate/inclusive appeals can be significant even in this case. Despite the fact that
Northern Ireland still lacks a sense of common identity, there have been differences in mobilization shifts. For example, in Northern Ireland there have been significant percentage shifts toward Hume’s more conciliatory SDLP and away from the exclusivist nationalists; at least one period showed a reversal o f this shift. Even though there still is
44 not peace in Northern Ireland, the civil rights movement and the SDLP’s abihty to shape the movement’s aspirations into political dialogue have fundamentally changed the nature of the discourse about the conflict.
2.4 Defining the Dependent Variable: Mobilization Episodes
Since the dependent variable is the popularity of a leader’s message vis-à-vis the people that leader is trying to mobilize, a proxy measure is used; swings toward and away from the leader’s party that are observable in elections data and narrative accounts of country experts. When possible, media reports are used to assess public opinion of these leaders and why segments of the public voted as they did. As the secondary sources were surveyed by this researcher to find out the periods which have been marked by mobilization shifts, several questions served as a guide:
1. When did leadership competition within the political minority become evident?
No cases previous to this intra-group competition are examined.
2. Was the “new” competition the result of a leader/party joining the race against an existing party, or were both competitors new?
2a. If the former (which was often the case), did the new competitor make gains that observers and academic accounts describe as significant? If results of previous elections are available, did the new competitor take votes away fi'om the existing leader/party?
If yes to one or both, then that leader is considered to be the beneficiary of the mobilization shift. If yes to the second question, this may afifect interpretation of results. 2b. If the latter (both competitors new), did one leader receive more electoral support than another? If results of previous elections are available, did the new competitor take votes away from the existing leader/party?
45 If no to first question, then case not selected. If yes to one or both, then the leader that the results, observers, and academic analyses argue won is considered to be the beneficiary of the mobihzation shift. If yes to the second question, this may affect interpretation of results.
Using these guidehnes the following time periods are the seven cases: Northern Ireland in
1982/83, 1986/87, 1996, and 1997*; and Rhodesia in 1979 and Zimbabwe in 1980 and
1990.
2.5 Content Analysis
The method used to observe the strategies a leader employs is content analysis, “a procedure in which a judgment is made whether a specific unit of material contains certain words or ideas assumed to indicate a theme or variable under study” (Hermann, 1983,
1987, p. 1). The research goals have imphcations for the type of content analysis to be used; for this project, a combination of quahtative and quantitative approaches is employed in an attempt to address shortcomings of both. After discussing content analysis as a method, the content analysis procedure for this project is explained.
2.5.1 Issues in Content Analvsis
Types of content analysis are usually categorized as quantitative or qualitative.
Although this distinction is often taken as “systematic” versus “soft,” Alexander George
(1959, p. 9) notes there is a more important distinction between the two, that is they differ
*1996/97 was originally considered a single episode, but important differences were found to exist in the contexts of these periods. It is more fruitful for the purposes of this research to consider these two years as separate cases.
46 in the “aspects o f communicative content from which the analyst draws inferences regarding noncontent variables.” Quantitative content analysis focuses on the frequency of specified content characteristics. The underlying assumption is that the more often a specified theme or variable appears in the units of text being analyzed, the more important and representative that theme or variable is to/of the speaker (Hermann, 1983, 1987, p. 1).
Because we wish to know the dominant type(s) of strategies leaders use, the frequency approach is important for this project. However, it is only a part of the information necessary to address the research questions asked here.
In contrast to a quantitative approach, quahtative analysis (which George calls
“nonfrequency”) is concerned with whether particular content variables are either
“present” or “absent,” and perhaps the structural relationship of content variables. The analyst thoroughly examines textual material to get a sense of the material in view o f the research questions. As Hermann (1977, p. 81) notes, the quahtative content analyst is sensitive to semantic, historical, pohtical, and situational contexts in which communications are embedded. In this way, the analyst overcomes a critique of quantitative analysis that it fails to take account of context and what the counted words mean (De Sola Pool, 1959, p. 202). A quahtative approach aUows the analyst to get at important aspects of text that are missed by coding what is there; George (1959) argues that this is especially important in dealing with propaganda analysis, which is similar to the research proposed here. The concern here is how often certain strategies are employed, but the relationships between and among difterent strategies are also crucial.
47 To accomplish the research goals, a type of contingency analysis is used for this project. This approach is “a method of message study designed to index... associational structures...” (Osgood, 1959, p. 41). It should be noted that the contingency analysis proposed here is an adaptation of Osgood’s use. He was interested in the associations among the cognitions of the speaker. Here the focus is the structure of arguments made to convince the audience, not the structure of the leaders’ perceptions. This adaptation this requires a more qualitative approach than Osgood’s, since judgments must be made about the kinds of units being coded.
“Costs” for content analysis approaches are factors which diminish the ability of the analyst to make strong inferences (Winter and Stewart, 1977, pp. 31-32). Several are relevant here and are addressed as the method used in the current project is explicated in the first part of this chapter. First, it is necessary to establish that the sample of documents is representative of the verbal output of the person studied, which may mean that it is representative of all statements made, of output only on certain topics, or of a certain type of output (for example, campaign speeches). The kind of representativeness sought should be informed by the purpose of the investigation. Second, intercoder reliability is a crucial issue, and requires that “...the categories for analysis of content [be] described or defined in such a way that different people, working independently, will make the same judgments when using the same material (Winter and Stewart, 1977, p. 33).
Third, the variables assessed should be linked to theory. Finally, the content samples
48 should be drawn from standardized or otherwise comparable documents or speeches
(Winter and Stewart, 1977, p. 36). These standards are all addressed in this dissertation, including intercoder reliability.
There are also several potential problems with the qualitative approach. Most qualitative analysts are less systematic in data collection than those who use a quantitative approach. Intercoder reliability is used less frequently, and it is rarely indicated which results are based on frequently occurring pieces of information and which are inferred
(Hermann, 1977, p. 81). The method used here was designed to avoid these shortcomings, by being explicit about inferences and by achieving a respectable degree of intercoder reliability. While this distinction between qualitative and quantitative analysis is often made, in fact the dichotomy exaggerates the degree to which both approaches involve aspects of the other. Quantitative approaches still often rely on interpretations about what kind of words should “count” as representative of a characteristic, and qualitative approaches generally have to assess frequency of appearance. With that mind, the approach developed here attempts to be rigorous in its definition of how texts are interpreted and how various aspects of the content are weighted.
Finally, a more fundamental critique of any content analysis approach involves the interpretation of meaning. For example. Sylvan, Majeski, and Milliken (1991, p. 328) argue that by establishing coding categories and then looking for our categories in the data, we cannot imderstand the constmal within the given culture. It is first necessary to
“ground” the categories we use and their interrelations in the speakers’ understandings of their statements. What this school of thought often underestimates is the degree to which
49 the “grounding” process does in fact reify the categories established post-hoc.® In other words, these scholars argue that they interpret a text without preconceived categories,
letting the text speak to the analyst about the categories in which the text’s authors were thinking or within which they were “operating.” In the end, however, the analyst is still
inferring what those categories are. In this project, the perspective adopted, especially when defining views as “inclusive” and “exclusive,” is set out clearly. This means that arguments about cultural interpretation can begin with the stated assumptions and can be critiqued on the basis of those assumptions. The next sections constitute more detailed discussion of the content analysis procedure.
2.5.2 Selection of Texts
After choosing the type of content analysis, the selection of texts to be coded is the next issue. Public pronouncements, such as speeches, articles, editorials, in which the leader appeals to potential constituents are the appropriate texts for the framework presented here. A common criticism o f using public rhetoric as data is that this rhetoric is instrumental and does not provide a dependable basis to infer characteristics of a leader; speeches are carefully structured and premeditated for the receivers (Holsti, 1976). Instead, interviews are valued for the relative spontaneity of response because these are more representative
(Hermann, 1977). In this project, however, the focus is in fact on the carefully planned, instrumental mobilizing strategies. Because the point is to investigate the mobilizing strategies leaders used when they were the beneficiaries and losers of shifts among voters.
® I am grateful to Don Sylvan for reminding me of this issue.
50 materials of interest are the speeches and addresses these leaders gave during the period
prior to the elections. The ideal study might investigate appeals in a number forms
(interviews, speeches, editorials) and to a range of audiences (domestic, international), and then
compare results according to audience. However, for any case there may be data limitations,
as there were for the cases investigated in this study.
The most comparable set of speeches both over time and between leaders for the
Northern Ireland cases are speeches at the annual party conferences. These were consistently
available for every year and most clearly establish the parties’ views of the current situation and
future agenda. Further, in Northern Ireland the content of these speeches is often reported in
the media so that the central messages from the leader often are heard by a wider audience."
During some periods, other kinds of statements are used also if similar forms existed for both
leaders. For example, in 1996 and 1997, both Hume and Adams made statements about the
peace process that were closer to the elections than their party conferences. This kind of
material is used to increase the data that are analyzed.
For the Rhodesia/Zimbabwe cases, the leaders often used radio addresses and speeches
to rallies to get their points across. Once Mugabe was in power, he had the last say in the
access others had to most forms of media. Because of some variation across cases, the
materials used in each case for this country are described in the case chapters; a full listing is
Occasionally, especially around significant events in the course of the conflict, the leaders delivered addresses which were attended by significant numbers of people and widely reported. There were also several interviews surrounding certain events. These speeches and interviews can also be coded and then compared to the party addresses to see if the kinds of texts and audiences makes a difierence in the strategies used.
' ' These are also ideal sources of rhetoric for future analyses because each party has a wdtsite on which they post these annual addresses as well as other speeches by party leaders.
51 found in Appendix B. Because the focus is on party leaders, the conclusions will be limited to those types of leaders; still, this limitation is minor since most leaders do tend to form parties or attach themselves to parties.
2.5.3 Unit o f Analvsis
The next issue to be addressed in any content analysis approach is the unit of analysis for the coding. The unit here is the topic; this allows us to be more inductive about the groups and strategies a leader lumps together, instead of looking more mechanically by sentence or paragraph. Five general topics were devised (see Appendix A). For each topic delineated by the speaker, the coder records the ingroups, outgroups, and strategies connected with these groups. Speeches had a range of approximately 18-25 topics each; the average length of each topic was 252 words. While some critics of content analysis may argue that the analyst is imposing the topic on the leader by this method, the categories of topics are clearly set out here and the coding o f them can be critiqued on those terms.
2.5.4 Identity Strategy
Because the objective is to compare which people/groups the leaders deem relevant and the labels or bases of identification that dififerent leaders (and the same leaders over time) use to describe potential groups in the society, the coder should underline all the groups the leader mentions within each unit (topic). How the leader describes both ingroups and outgroups (or “self’ and “others”) is of interest. “Outgroup” is used here to mean any group that the leader distinguishes as an “other,” or not part of the “self.” “Ingroup” is used here
52 to mean the group with which the leader identifies. For example, Zimbabwe’s Mugabe
may use the following ingroup descriptions: “AJfricans,” “Zimbabweans.” He may name
outgroups such as, “Europeans,” “British.”
K group reference code is assigned, then, instead of using the particular labels
(“Catholics,” “pohticians,” “working class,” “Zimbabweans,”), for this project a method
more generalizable across time and cases is used to record the group delineation. Lists of
labels that tend to be used to describe people in each country were created inductively (for
the group references), and then the lists were categorized according to degrees of
exclusivity (for the delineation): very exclusive, moderately exclusive, moderately
inclusive, and very inclusive.
Group Reference
Before the groups are coded in terms of labels, it is essential to know which people
the leader is trying to label. For example, when someone says “unionists” it is important
to know if this label is given to what is considered the political majority in Northern
Ireland, or if the speaker is using the island of Ireland as a reference point and therefore
presenting this group as a minority. If the group reference is to people on the island of
Ireland, “Protestant” is a very exclusive way of describing people. On the other hand,
calling the majority in Northern Ireland “Protestant” is moderately inclusive (since more
exclusive labels are commonly used: loyalist, Orangemen, unionists). This is important because at another time a leader may use the labels “Protestant majority” or “Irish of a different tradition.” In these other terms, the leader is potentially including and excluding dififerent people.
53 As this example may indicate, coding decisions for the group reference may not be straightforward and could require using the context of the speech or topic to determine which people are being labeled a certain way. Trial coding and intercoder reliability tests show that this method is effective. For the Northern Ireland cases, possible group references include people living in the borders of Northern Ireland (which may be called British or Irish, may be broken down into Nationalists/Unionists or Catholics/Protestants, or may be labeled working class or establishment, for example), those living in the Republic of Ireland, those in the island of Ireland, those living in mainland Britain, people in both areas, those in Europe, or people in other parts of the world/all people in the world.
For the Rhodesia/Zimbabwe cases, group references include people in the borders of
Rhodesia/Zimbabwe, those in Africa, or global references. For example, the first may be broken down into Rhodesians meaning whites or Zimbabweans meaning blacks; it may be referred to as Zimbabweans meaning all people in the borders. Note that this coding leaves room to record when those in particular borders may be defined as part of something larger.
For the Irish cases, “Europeans” is a potential label; for Rhodesia/Zimbabwe, part of a greater
African nation or developing world may come into play; none of these are predefined as mutually exclusive. It should be apparent that for the Identity strategy, the coding is set up to allow the data to speak for itself so that the analysis is as interpretive as possible in recording the way leaders suggest group boundaries be drawn.
A leader may make important references to ingroups and outgroups in forms other than proper nouns. For example, “Irish republicanism” and “British repression” should be
54 coded as references to groups because it is clear to whom the leader is referring. On the
other hand, words that connote forces and not humans should not be coded as groups. An
example here is “apartheid.” This would not be coded, but “Botha’s apartheid junta”
would. The following list summarizes how to know which references to groups should be
coded:
—Group names: “Irish republicans” —A more specific description of the pronouns “we” and “they” : “We are republicans”; “they are unionists” —Reference to a group that can be made exphcit: “Our enemies in high places”; “the party who supported the legislative reforms”; note that contextual knowledge is necessary here to know who the groups are who may have supported or opposed particular measures. —Indications of ownership, actions, or intentions of groups: “British presence in this country”; “fi-eedom fighters”; “Dublin’s refusal”; “our position” (when it is clear from a previous sentence who the “our” refers to) —Capital cities or pohtical leaders when used to talk about governments: “Thatcher and Reagan support Dubhn’s position.” —Individuals as exemplars o f groups: “John Hume sat down at the negotiating table”
An important issue is that identities are often nested, so that people may be a part of many different ingroups. The idea of nesting and presenting a picture of group as multidimensional and crosscutting may be a key way of encouraging multiple loyalties.
This project is sensitive to this in the analysis of the ingroup references; the different groups a leader “suggests” are ingroups will be compared. The most frequent references may be the “central” identity, while a few references emphasize a higher level such as
“European” or “Aftican.”
55 Group Delineation
As noted, the values for this variable are very exclusive, moderately exclusive, moderately inclusive, and very inclusive. It is not suggested here that there is an objective definition of “inclusive” and “exclusive,” but in order to observe differences between leaders and changes over time, it is necessary to assert a reference point. This project approaches this issue by considering the group reference: who are the people the leader refers to when using a particular label? The possible references for the Ireland case include people in the borders of Northern Ireland, the border of the Republic o f Ireland, the island of Ireland, et cetera, (see coding handbook). The labels are assigned degrees of exclusivity based on the “perspective” of the group reference used with the labels. For example, consider two groups with the same group reference named by a leader:
“Catholics” and the “working class folks in Northern Ireland.” From the perspective of the Northern Ireland group reference, “Catholics” is considered “moderately exclusive”
(because “nationalists” in the Northern Ireland context is perceived by the leaders as and considered here as even more exclusive) and “working class” is considered “moderately inclusive” (since all are not in the working class in this area, it is not coded as “very inclusive”). The “label lists” were created fi'om research on these cases and from trial coding on material that was not coded for the dissertation. For each set of cases, a scale was assigned to the label lists, from very exclusive definitions to more inclusive labels, so that for the delineation column the possible values for both ingroups and outgroups are 0-
4 (0 if it is too ambiguous to code).
56 The exception to using this delineation scale occurs when the group reference is to
“people in the world.” Because there is such a wide range of possibilities, it did not make
sense to try to set up a scale. Instead, a list of labels with numbers for each is given. This
will help determine which international actors are talked about and how. For example,
one party may always talk about international allies who are always called “oppressed
people,” while another may often mention “democrats.” Finally, for the group reference
and delineation, there are codes for ambiguous references. For example, these may be
used when a leader uses “we” or “them” but the coder cannot discern with high
confidence from surrounding sentences to whom the leader refers.
In summary, the Identity variables (both for ingroups and outgroups) are linked,
or have two levels. The first level (group reference) denotes the people to whom a leader
is referring, and the second (group delineation) is the label a leader attaches to those
people.
2.5.5 Strategies Used With Identity
After coding for the group reference and delineation, the general coding method is
to read through the topic in which these groups are discussed and record which
mobilization strategies are used with the ingroups and outgroups. Care must be taken to
record the strategy with the proper group. In general, this means that no outgroup image
strategies will be recorded on ingroup lines, for example. Matching the strategy with the group to which it applies is generally straightforward, but may be slightly more difScult with the Governance strategy (rules for this are elaborated below). A key issue is making
57 coding decisions about the frequencies of strategy use. Since this is related to recording
the ingroups and outgroups as discussed above, this section begins where coding for the
Identity strategy ended.
Frequencies and Multiple Strategies
An important issue concerns what to do when the same group reference/group
label pair is used multiple times, because it is related to strategy frequencies. This is
important in recording the strategies used, because each return to an aforementioned
outgroup may include the use of another strategy, or the second use of the same strategy
used the first time the outgroup was mentioned. If the leader uses the exact same
description for a group (same reference and delineation) one or more times within the
same topic, the coder should enter the codes only once (on one “outgroup” or “ingroup”
line). Multiple references include using “they” (them, their, etc.) when it is clear the leader
is referring to a group previously described; “The British government should persuade the
unionists. They need to work with Ireland too.”*^
Again, it is crucial to keep track of frequencies or uses of additional strategies for
each outgroup. Since the topics may be as long as 500 words (though the average is
about 250), it will not be unusual to have several strategies marked for each outgroup.
For example, in the following sentence, the first outgroup is the British and the second is the Irish government; but the second reference to the British is an additional strategy to be recorded (on the same outgroup line as the first):
'■ None of these examples come from any speeches. I have made them up.
58 “The British are destroying our culture and have no plans of easing up on us. The Irish government should promote talks. The monetary policies of the British have led to massive unemployment, which has only promoted further conflict.”
Here the topic is 2 (discussing situation where emphasis is blame); the two outgroups are the British and the Irish government. The first strategy used for the British is Enemy
Image/threat and the second is Enemy Image/blame.
The general mles for coding fi'equencies and multiple strategies are as follows. If several sentences serve to paint the picture of one strategy, the code is recorded only once.
However, if there is then any interruption in this thought before the same strategy is seen again, such as a turn to another group or use o fanother strategy, the return to any o f the strategies usedpreviously is recorded as another usage. Another case where multiple usage is recorded is when the leader begins a new paragraph using the same strategy with which the previous paragraph ended. This latter case is straightforward but the former (in italics) needs clarification. Take the following example:
“We, the Northern Irish people, suffer because of Britain. The injustice in our situation cannot be tolerated. People are without jobs, without decent housing, without the rights that every human should have. British policy for decades has taken every opportunity away. The British government’s monetary practices have made this a hopeless life. Our people deserve better than this.”
In this sentence, the ingroup is the Northern Irish people. There is only one outgroup, British government. In the first sentence, the speaker uses the Enemy hnage/blame strategy with
Britain and the Justice/negative strategy with the Northern Irish people. The next two sentences are also the Justice/negative strategy, but do not count this as a second usage because it follows on the first. The fourth and fifth sentences go back to blaming Britain, and so a second usage of the Enemy Image/blame strategy is recorded. The final sentence goes
59 back to the first ingroup and thus a second usage of the Justice/negative strategy is recorded.
The subsequent sections take the mobilization strategies individually, describing the general themes that are coded for each. See Appendix A for the detailed coding rules.
Justice Strategy
Key indicators for either of the Justice strategies include discussion of the situation of the group being described. This may be more common with the ingroup but not necessarily: leaders often talk about other groups or subgroups of the larger ingroup as well. The negative variant of the Justice (or Injustice) strategy is being used when the leader dwells on the fact that the group “deserves more than we are getting.” Here the group as a group (not just individuals) is subject to unfairness, is making unreasonable sacrifices, or is not being adequately rewarded for its eftbrts. Also code for this strategy when the leader discusses the struggle of the group and/or when the leader dwells on the poor situation in which a group finds itself.
For the positive variant of this strategy, look for mentions that the group is being treated fairly or more fairly than in the past, and/or of the recent gains that have been or are being made. This strategy is coded when the main theme is achieving success and overcoming particular obstacles. When the leader’s discussion fits into these descriptions, but tlie main emphasis is on institutions or policies that have caused the situation, the codes for the Governance strategy are used.
Outgroup Image Strategy (Enemy Image or Ally Image)
These strategies are coded when the leader discusses any groups other than the ingroup and is not characterizing the situation of the group in a way that is described as
60 the Justice Negative or Justice Positive strategies. The emphasis is on the character of other groups, the groups’ roles in the conflict or in the situation of the ingroup. The leader may emphasize positive or negative aspects.
For the Enemy Image strategy, there are several categories. The “ambiguous” code is used if the leader is talking about a group in a negative way that cannot be characterized as pointing out a threat, blaming an actor, or indicating the betrayal of an actor. The leader is coded as using the Enemy Image/Threat strategy if the emphasis is on the threatening nature of the group. For this, the leader may not use the term “threat” directly, but may be creating a sense of fear or darkness about the group, that its presence is ominous. If the leader is blaming the group for any given situation (this may appear commonly with the Justice/negative strategy), he is said to use the Enemy Image/Blame strategy. Finally, the leader is coded for the Enemy Image/Betrayal strategy if he discusses the outgroup’s role as a betrayer to the ingroup. This may require a contextual reading, but in general the intercoder tests show the coding for this strategy is rehable.
For the Ally Image strategy, there are also four categories. The ambiguous variant is recorded if the other codes do not adequately categorize the leader’s discussion of the outgroup but he seems to be characterizing it in a positive light. If the leader emphasizes that the outgroup and the ingroup have common goals or values (both want to fight poverty, both want to get the British out of one place), the AUy Image/Goals similar strategy is recorded. If the reference goes beyond identification of common goals to a characterization that the two groups are in similar situations (both are poor because of same reason, both have British occupier and want to get that power out), the leader is said
61 to use an Ally Image/Situation similar strategy. If the leader emphasizes that the groups
have a similar culture, the Ally Image/Culture similar strategy is used; this strategy was in
fact used only rarely.
Governance Strategy (Positive and Negative)
The Governance strategy is indicated by discussions of government institutions, laws, alliances with other eûtes or parties, proposals for change, taking part in elections, and/or talks with other parties (including a peace process), individuals, or governments
(“governance factors”)- The negative variant is coded when the emphasis is on the harm that has been done, is being done, or will be done to a group by particular governance factors. The positive variant is coded when the leader focuses on the benefits in the past, present, or future of particular governance factors.
Storytelling Strategy
The Storytelling strategy is treated as a mode of presentation or persuasion whereby the leader “embellishes” the discussion that uses the other strategies (Identity, Justice,
Outgroup Image, and Governance) by drawing on the past or a vision of the group’s destiny.
The leader weaves a “story” around the imagined community, connecting the people of the present with those in the past and/or future. The Storytelling strategy is coded when the leader discusses the sacrifices of the past, which may include mentioning martyrs/those who have died for the cause (these examples would also be the Justice/negative strategy). If the leader discusses martyrs to show the tragedy of the situation this strategy is coded; he may also use
62 martyrs to say they have pushed the nation’s cause forward. Characterizations of how bad the
situation in the past was, how the roots of the current problems are in the past, or that the past
must be avoided are also included.
The Storytelling strategy is also coded when the leader focuses on gaining strength from the past, when the emphasis is on redeeming the past, when he makes calls to realize the ideal nation/the right to self-determination, or when the focus is on future gains, dreams, and/or aspirations. Finally, this strategy is coded when the leader mentions heroes of the past, not as martyrs but as people who have had a positive force in moving the nation forward.
2.5.6 Intercoder Reliabilitv
In order to increase the reliability of the coding scheme, two undergraduate honors students at Ohio State University underwent eight weeks of training, both learning the coding scheme and working through texts that would not be used in the dissertation analysis. At the end of this training, several intercoder tests were conducted and further sessions were held to improve the precision of the instructions. Where there were coding disagreements in these sessions, this trainer explained to the coders why the particular codes should be marked. In the few cases where this trainer agreed with the interpretation of the coder, specific instructions were added to the codebook to clarify the rules.
Not including the eight weeks of training, one coder spent 100 hours on the coding of speeches for both cases; the other coder was only able to code half the material and spent about 50 hours doing so. The coders scored different materials, and coded all the
This work was support by a grant from the Ohio State University Graduate School, the Graduate Student Alumni Research Award (GSARA), Spring 1998. 63 text in each speech they were assigned. In total, all the texts coded for the dissertation
were coded by this researcher and coded a second time by one of the students.
The scores on the intercoder tests began at 56% for the initial coding (after the coders had only read through the instructions on their own). After training, the scores remained between 76% and 82%. Given that this coding scheme calls for far more than a simple word count, this score is respectable. It is also encouraging that the coders improved the scores so dramatically once the training sessions were complete. Finally, the intercoder training sessions helped eliminate several coding categories that simply called for too much Interpretation on the part of the coder; in other words, if there were systematic patterns of “misses” in the coding, the category was eliminated. An example of this was whether the leader referred to outgroups as subgroups of a larger ingroup or not.
Theoretically, this information is important in understanding how the leader sees multiple and/or nested identities. In reality, one could not easily determine this information from the speeches.
2.5.7 Analysis
The coding sheets are entered into SPSS, taking each ingroup or outgroup as a case.
This is necessary in order to determine which strategies appear with the particular references and delineations. The statistical analysis then allows comparisons (over time, between leaders, and by topics) of which groups are deemed relevant, how people are depicted in terms of group identities, and which strategies appear with particular group references and delineations
64 (the targets of the strategies). The data are analyzed using simple cross-tabulations and frequencies (see case chapters). The next section discusses how the data are compiled to determine which leader’s strategies are more and less exclusive.
2.6 Inclusive/Exclusive: Categorizing the Strategies
The analysis of strategies will compare how often leaders use the different strategies, how they combine the strategies, and which groups are the targets of the strategies in the different periods. This analysis offers insight about the applicability of extant Uterature to the aspects of conflicts that leaders really emphasize (is relative deprivation a more common emphasis than threat or governance?). Further, the analysis will provide the means to test the hypotheses presented below. Then, beyond comparison of the exclusivity of leaders’ strategies in the single cases, which allows this hypothesis testing and refinement, changes in the degrees of exclusivity over time will be explored to see how the rhetoric evolved over the years.
The first section in each case analysis looks at the percentage (of all the strategies the leaders used) that each individual strategy was used. As discussed below, predictions are made in Chapters 3 and 5 about the individual strategies that the contextual variables are expected to promote. The frequency comparison allows an evaluation of those expectations. The second and third parts of each case analysis investigates the nature of the strategies (in terms of degree of exclusivity) and the nature of the targets (also in terms of degree of exclusivity).
65 Nature o f Strategies
Of the five strategies, some are explicitly more exclusive than others—Threat is
considered more exclusive than Governance Positive, for example. An exclusivity ratio is
calculated for each leader by dividing the percentage of exclusive strategies used by the
percentage inclusive strategies used in that case. The exclusive strategies are Justice
Negative, Governance Negative, Enemy Image (all variants). Storytelling, and Identity
(“very exclusive” delineations for particular ingroup references; this is explained for each
country). The inclusive strategies are Justice Positive, Governance Positive, Ally Image
(all variants), and Identity (“very inclusive” delineations for particular ingroup references).
The leader with the higher ratio is considered more exclusive on this measure, but the
degree of exclusivity also depends upon the target of the strategy. Indeed, an initial
reaction upon reading this classification may be to say that an Injustice strategy can be
used in a more exclusive or more inclusive way, depending on whether the leader is
arguing that everyone in the society is subject to injustice, or if just a subgroup of society is. Since, as ?*ated before, the reference is the society under examination, the latter is more exclusive than the former. This idea is not disputed here, but in order to study the differences between more exclusive Injustice strategies and more inclusive Injustice strategies, for example, a second measure of target (or subject) exclusivity is included.
The classification of strategies (not targets) is now explained briefly.
The Enemy Image strategy emphasizes enemies or scapegoats, and therefore builds boundaries among the groups that are placed in opposition when this strategy is used. On the other hand, the positive variant of this strategy emphasizes allies and other items two
66 or more groups have in common. Thus, the former is more exclusive than the latter. The
Justice Negative strategy emphasizes the poor situation in which the ingroup finds itself, and is considered exclusive because it emphasizes ingroup membership and plight, without reference to others. The Justice Positive strategy is considered more inclusive because the leader is emphasizing gains and how a situation of injustice is improving.
The use of the Governance Negative strategy can be considered more inclusive than no references to Governance (Positive or Negative). This is because talk of governance factors (see above) signifies that the leader is at least considering the institutional setting and negotiation with others and possibly is advocating working within the system or government. This approach is in contrast to extreme groups whose only modus is violence, such as liberation movements/terrorist groups (or to use the more neutral term, paramilitary organizations). Since all leaders were found to at least some of both the positive and negative variants. Governance Negative is considered with the more exclusive strategies and Governance Positive with the more inclusive strategies.
The Storytelling strategy is considered to add to the exclusivity of leaders’ strategies. As previously discussed. Storytelling is a “mode” of employing the other appeals. Since this approach emphasizes an imagmed community and the bonds among people from centuries past with those of the present and future, it is expected to make people of a group feel more exclusive toward those in other groups. Indeed, when scholars emphasize this view they argue that leaders draw on the past and the imagined community to promote ingroup solidarity. The treatment of the Storytelling strategy as a
67 more exclusive strategy receives support from the empirical data as well; as will be
discussed, it tends to be used overwhelmingly with the Injustice, Enemy Image, and
Governance Negative strategies.
Nature o f Targets
The “target” is the subject of the strategy, and how exclusive a leader is overall changes depending about whom he is talking. For example, coding may reveal that both leaders have similar frequencies for the Justice Negative strategy but the targets are very different: one may argue that the target of deprivation is the entire country while the other claims that the target is only the political minority. The latter is considered a more exclusive strategy. For each strategy, an exclusivity ratio for the targets is calculated as follows: the percentage (of total group references) made to groups considered more exclusive is divided by the percentage of more inclusive references (for each country, the inclusive and exclusive references are specified and justified). Finally, two summary ratios are created by adding the ratios for the targets of the strategies for which the subject tends to be an outgroup (Enemy Image and Ally Image) and an ingroup (the Storytelling,
Justice, and Governance strategies).
Summary o f Findings
For an overall exclusivity score, the three scores described above are summed and weighted by the N (or number of groups) for the case: nature of strategies, nature of ingroup targets, and nature of outgroup targets. In all but the first case, the change from the previous time period is also given. The higher the exclusivity ratio, the more exclusive the leader’s appeals.
68 2.7 Contextual Analysis: Political Opportunities
The framework described in the previous chapter has two parts; (1) leaders’
mobilization strategies; (2) political opportunities (stability of elite alignment, degree of
state repression, international involvement, and regional integration). The political
opportunities group together aspects of the domestic and international contexts which may
be expected to constrain or empower leaders’ efforts. Thus, once an effort is made to
predict the effects of context, that is, whether inclusive or exclusive appeals will be
“promoted” by the situation, the content analysis allows a testing of the predictions by
comparing the exclusivity of the appeals of the winning versus the “losing” leader. Also,
in an inductive approach, the cases are compared to see if particular contexts promote
success of certain combinations of the individual strategies (Injustice, Enemy Image,
Governance, and so on).
The problem with the social movements literature from which the concept of
“political opportunities” is drawn is that it does not provide much insight about how the
kinds of opportunities can be made operational (i.e., indicators of the various
opportunities are not made explicit), nor does it suggest concrete hypotheses about the
relationships between opportunities and framing. Even more problematic, it does not
suggest how the types of opportunities may mediate each other. For example, if the degree of repression and elite alignment are favorable to one leader but international intervention favors the other, whom does the opportunity structure favor?
To begin to tackle these unanswered questions, this study puts forth several hypotheses concerning the relationship between context and the success of inclusive and
69 exclusive appeals. Included in this discussion is the operationalization of these concepts and indicators for them; secondary sources o f various kinds are the sources of data for this set of variables. In the Northern Ireland cases, interviews with party members provide an additional source.
Four kinds of political opportunities are investigated. Two are domestic in nature and are drawn from the social movements literature: change in the stability of elite alignment and the state’s capacity and propensity for repression. A major contribution of this study is the inclusion of two additional political opportunities which are international in nature: international involvement or mediation, and regional integration.
2.7.1 Change in Stabilitv of Elite Alignment
A political opportunity is expected to be perceived when there is a change in the elite alignment which has stood in the way of the political minority meeting its goals.
“Elite alignment” is the broad set of relationships that undergird the establishment
(McAdam, 1996, p. 27). The focus is on the change in relationships among the elites who have maintained the status quo that keeps the political minority out o f the political system.
The sets of elites important here are the colonial power (though this may be a former colonial power, the relationship often remains important); the dominant power in the disputed country (Northern Ireland or Zimbabwe), here called the controlling power, and the primary regional actor that has helped maintain a balance. In Northern Ireland the relevant alignment concerns Britain, the unionists, and the Republic of Ireland. In
Rhodesia, relevant actors are Britain, the Rhodesians (whites), and South AJfrica. In
70 Zimbabwe, this has meant Britain (because of the Lancaster House Agreement), the white
Zimbabweans (pre-independence) and the ruling party ZANU-PF (post-independence), and South Afiica.
In the case study chapters, the hypotheses described below are worded specifically for these various actors. For example, for Northern Ireland, hypotheses concern changes in the relationship between the stabihty of the British-unionist relationship, since this formed the basis of the state, and the British-Republic of Ireland relationship since the
Anglo-Irish Treaty was necessary to create Northern Ireland (and these states both must agree to change the status by international law).
To go beyond the general suggestions in the literature that changes in elite alignment provide opportunities for mobilization efforts, this study examines how the alignment of these three types of actors has “tightened” or “loosened,” and hypothesizes which combinations favor exclusive or inclusive views. The relationships examined are between the colonial power and the controlhng power (alignment A), the colonial power and the regional power (alignment B), and the controlling power and the regional power
(ahgnment C). Alignment A should be present in all cases, but it is possible that only alignment B or C will be present.
The terms “tighter” and “looser” are conceptualized as follows. To say alignment
A is “tighter” means that the colonial power is more likely to back the demands of the controlling power to maintain the status quo. When alignment A is “looser,” the colonial power has moved away from backing the Controlling power’s demands. When alignment
B is “tighter,” this means the regional actor and the colonial power are working together.
71 If they favor the status quo this could help the controlling power, but if they favor
recognizing minority demands, a tightening of the colonial-regional alignment could
undermine the controlling power. The “tightness” or cohesion within the installed power’s
bloc is also important, but shifts in it may be a result of loosening in the relationship with
the colonial power (in a period of uncertainty that such loosening produces). To say
alignment C is tighter, the controlling power and the regional power are working together,
which supports the status quo. When this relationship loosens, it means that the regional
power is no longer supporting the status quo. The importance of alignments B and C will
differ depending on the country’s history. For example, alignment C for Northern Ireland
(the relationship between the Republic of Ireland and Protestants in Northern Ireland) is almost non-existent, but is important for Rhodesia (the relationship between South Afiica and whites in Rhodesia loosened over time). This is a judgement that can be made for any case, meaning that this conceptualization can be generalized. Similarly with alignment B, the relationship between Britain and Dublin is very important, but the relationship between
Britain and South Afiica is relatively insignificant.
What are the expectations for the relationship between elite alignments and promotion of exclusive/inclusive strategies? It is posited here that more exclusive strategies used by leaders of the political minority will dominate when the status quo endures or when it looks as if nothing else is working. This section describes what kind of elite alignments contribute to this perception, then the tables below summarize the hypotheses relating combinations of alignment to the expectations about what kind of strategies will win. First taking cases where A and B are the most important relationships
72 and using the terminology introduced in the previous paragraph, three situations lead to this expectation. First, when both alignments A and B loosen, this means that the colonial, regional, and controlling powers are not talking with each other to address the situation. The loosening between the controlling and the colonial powers creates an opportunity for the minority to mobilize, but the majority is not being pushed to address those grievances. An open battle for control would be expected, with any non-violent or moderate actors being outflanked completely.
Second, when alignments A and B tighten, some kind of accommodation is under negotiation with the regional power but the controlling power is receiving support for the status quo from the colonial power. It is posited here that in this situation, exclusive appeals are again most attractive to the minority because it appears that their grievances are not being addressed. If alignment C is relevant, this situation is intensified if alignment
C tightens because the controlling power is receiving even more support in maintaining the status quo.
A third case of alignment that favors exclusive appeals (i.e. anti-negotiation, anti establishment views) is when the colonial power-controlling power alignment tightens
(alignment A), atid the colonial power-regional actor alignment (alignment B) loosens.
In this scenario, the colonial power may appear completely intransigent so moderate appeals do not offer much hope. Again, this situation is intensified if alignment C tightens, because the controlling power is receiving even more support in maintaining the status quo.
73 The fourth possible combination favors more inclusive rhetoric, and it is posited
that a leader using inclusive rhetoric is more likely to be successful when alignment A
loosens and alignment B tightens. This situation is further enhanced when alignment C
loosens. Therefore, by this logic, there is only one possible alignment that favors leaders
promoting inclusive ideas in cases where A and C are the most important: when the colonial power stands firm against the controlling power’s demands for “protection” AND when the colonial power and the regional actor work together to address the minority’s demands (and when the regional actor is not aligned with the controlling power).
Looking at cases where only alignments A and C are important, when A and C both tighten, exclusive strategies are favored because all are supporting the status quo.
When A tightens and C loosens, the controlling power is still receiving support from the colonial power fo r the status quo and exclusive strategies are favored. When A loosens and C tightens, the colonial power withdraws support fo r the status quo but the regional power still supports it: exclusive strategies are favored. The only case that favors inclusive strategies is when alignments A and C loosen, because support is withdrawn fo r the status quo. Here we see room for the other political opportunities to interact with this variable because while the elite alignment “removes” obstacles to the idea of inclusivity, what else the controlling power is doing (such as repression) and the role of other international actors may be important. Lack of support for the status quo by the regional and colonial power does not necessarily mean the pohtical minority will perceive growing
74 equality and respond to inclusive appeals. Tables 2.1 and 2.2 summarize the preceding discussion and the hypotheses about which combinations of alignments favor more exclusive or more inclusive strategies.
Alignment B (colonial/regional)
Alignment A (colonial/controlling) Tightening Loosening
Tightening Exclusive Exclusive
Loosening Inclusive Exclusive
Table 2.1; Relationship of Changing Ehte Alignment to Favored Strategy Type, when A and B play role
Alignment C (controliing/regional)
Alignment A (colonial/controlling) Tightening Loosening
Tightening Exclusive Exclusive
Loosening Exclusive Inclusive
Table 2.2: Relationship of Changing Elite Alignment to Favored Strategy Type, when A and C play role
With these propositions in mind, the case studies are compared on the change in elite alignment variable. In line with the method of structured, focused comparison.
75 questions asked of each case involve the role of the three groups described here, and whether the relationships are tightening or loosening is assessed through a comparative analysis of historical accounts of the chosen cases. If there is more cooperation between the actors than in the previous period, this is considered tightening. When there is less cooperation than in the previous period, this is considered loosening.
2.7.2 State’s Capacitv and Propensitv for Repression
A large body of work has addressed the relationship between repression and political violence. Although the dependent variable for this study is not violence, the issue of mobilization is related and leads one to expect similar effects. The general hypothesis in much of the Uterature is two-pronged: If a regime is repressive, then it will be difiUcult to mobilize people against the status quo. If this repression increases, mobilization will become even more difficult. Despite all the research, there is no definitive conclusion about the effects of repression. Still, one survey (Hoover and Kowalewski, 1992, cited in
Krain, 1998) of over 100 studies between 1965 and 1990 shows that seventy percent of these studies confirm this negative relationship between repression and mobilization.
Empirical studies have shown that the relationship between repression and mobilization is, not surprisingly, more complicated than this Unear view. There may be a
“window” of opportunity hi which leaders are able to mobilize groups because o f state repression. In fact, Irvin’s (1993) study o f the IRA shows that recruitment of hard-Uners goes up when there is more state repression. On the other hand, once in power, Mugabe’s de facto one-party state maintained the Rhodesian security apparatus and has largely
76 deflected anyone from seriously challenging ZANU-PF rule (Weitzer, 1990)/'*
Monitoring changes in the levels of repression over time provides a way to examine the
relationship more closely, so that a better understanding can be reached on this important issue. A more dynamic approach is especially necessary when the purpose is to understand mobilization shifts between competing groups over time.
Repression is measured here by the system the state uses to control dissent.
Common elements are emergency powers acts and the stringency with which they are employed by policing forces; censorship; harassment; and curfews. Initial research into the Northern Ireland case demonstrated the importance of specifying the targets of repression, and therefore for each case whether the targets of repression were the political minority or only a subset of it (such as members of a single party) must be specified.
Propositions deal with level and target of repression. The questions asked of each case are: Compared to the previous time period, did the state pass more legislation, and/or exert more time and resources to arrest and control people who were classified as the political minority; were there repressive events of state violence (crackdowns on riots, for example)? Compared to the previous time period, did the state pass more legislation, and/or exert more time and resources to arrest and control any of the leaders’ parties? If answers to these questions are yes, then this variable is labeled “higher” for that target
(community or party); if no, then “lower.”
Hypotheses are as follows. I f the community is victim o f repression, then the leader with more exclusive strategies is expected to be more succes^l. The logic is that
" This is why the 1990 elections were considered somewhat of a success for opposition leader Tekere, though he only received a small percentage of the vote. 77 appeals to protest resonate more than calls to negotiate under repressive conditions. The state may crush this opposition as in Tiananmen’s Square, but it is expected to gain followers prior to a response from the state. I f a single party is subject to repression, but things are promised to get better for most o f the community, then the leader with more inclusive strategies is expected to be more succes^l. The following table summarizes these expectations.
Repression: Party as Target Repression: Community Higher Lower as Target Higher Exclusive Exclusive
Lower Inclusive Inclusive
Table 2.3: Relationship of Repression by Target to Favored Strategy Type
2.7.3 International Involvement/Mediation and Regional Integration
Change in the stability of elite alignment and repression are often regarded as the most important “proximate” categories of opportunities in the social movements literature, but it is often argued as well that international involvement (from intervention, to mediation, to more symbolic diplomacy) can be pivotal and may feed into the other opportunities. For example, the spread of democratic norms globally explains partially why the United States has pushed ahead a negotiated settlement in Ireland, and this development in turn may have encouraged changes in elite alignment that are favorable to
78 the winning of inclusive strategies. Likewise, in Rhodesia, the sense in the international community that the rule of the large black majority by the tiny white minority under the internal settlement of 1979 was against democratic norms led to the rejection of that settlement and the pressure on the Rhodesians to end their repressive regime. Before independent Zimbabwe was recognized, all parties had to negotiate a more “democratic” settlement. Further, this international dimension may make a difference if all other dimensions are equally balanced in favor of both competing parties.
Within the broad category of “international factors,” which dimensions should be considered and how should their effects on internal competition be analyzed? The previous chapter reviewed some work in the international relations field which has begun to consider the international dimension of nationalist or ethnic conflicts. This study seeks to build on work by Gurr and others by distilling two major categories of international factors which are generalizable across cases, and then examining the seven cases to see if and how the factors were important (Brown, 1993, 1996; Gurr, 1993; Ollapally and
Cooley, 1996). The variables are (a) intervention, mediation, or other actions by states, international organizations, or non-governmental organizations; and (d) regional integration. Secondary sources which provide analyses of the countries, relevant international organizations, and, if applicable, the progress of regional organizations of which the country is a part are used to evaluate these variables. Recent work (Sisk, 1998, p. 163) has raised the issue that different international actors may work at cross-purposes
(for example, states versus non-govemmental organizations). For this study, a summary
79 assessment is made to see if international factors actually have any influence at all. The
current approach will then tell us if future, more detailed research into the role o f
“competing” international actors is needed.
Extant work offers little insight in terms of hypotheses about the role of the
international arena, instead tending to note the importance of the external dimension and
calling for further study (while still somewhat general. Brown, 1993, 1996 are positive
contributions to this area of research). Indeed, most international relations research does
not say much about (1) the struggle within nations or groups to define identity and
mission, and (2) the effects of international factors on this struggle. This means that the
following hypotheses are suggested by previous work, but the case studies in this project
will be the basis of more specific propositions about international factors, to be tested in
further research. For the international and regional factors, the following subsections state
the kind of questions that were asked, the way in which the variable is expected to relate
to inclusive versus exclusive strategies, and the kinds of data used to evaluate the factors.
International Involvement/Mediation
Do the influential actors to which the political minority appeals (most often the
United Nations, and/or the United States, and/or a regional state or organization) consider the issue an internal issue to be dealt with by a sovereign state, or is the issue defined as an international problem/threat to international security? Relatedly, is there mediation or intervention by the UN, IMF, EU, OAU, United States, or other global or regional actors?
If so, is the involvement neutral or is it on the side of one competing leader? Three related hypotheses stem from these questions. I f the international community regards the
80 situation as an internai conflict (visible in United Nations statements and resolutions, for
example), and there is no action by international actors to address perceived problems,
people in the target community (political minority) are likely to be more open to
exclusive appeals. I f the relevant actors in the international community call for that
community to help resolve the problem but are otherwise “neutral, ” more inclusive
strategies are expected to win. I f the intervention or mediation by actors in the
international community favors one leader over another, then that leader is expected to
win whether his strategies are inclusive or exclusive. Data include U.S. State Department
statements and reports; other governments’ statements of support; United Nations
statements, resolutions, or reports; declarations by IDs or NGOs; news coverage from the
New York Times, the Times of London, and the Economist, and secondary sources
describing this aspect of the cases.
Regional Integration
For this variable, questions asked about each case are; Are there regional
initiatives promoting an overarching system and identity, economic and/or political? Do
these regional organizations have any authority to affect political and/or economic aspects
of the country? If the answer to both or one of these questions is no, the integration level
should be labeled “decreasing” or “unchanged,” depending on the circumstances. If the
answer to either or both questions is yes, an additional question is asked: Have
relationships with others in the region expanded since the last period? If yes, then the
integration level should be labeled “increasing,” if no, then it is considered “unchanged”
(and is not expected to be significant).
81 The expected effects of this variable are derived from the fundamental premise of
functionalist theory. Haas described integration as the process of shifting loyalties toward
a new center (Haas, 1958); in the terms of this project, this means people are likely to
have more inclusive views of their identities as integration proceeds because they perceive
higher-level common bonds with all people in a region. Though this makes sense
logically, Haas argued that extreme “micro-nationalisms” might be promoted by regional
integration. As summarized by Ishiyama and Breuning (1998, p. 10), “the logic behind
this expectation is that regional integration permits previously isolated ethnic groups to
become more visible and.. .interact across national boundaries.” In terms of the present
framework, the implication is that the audience may be more open to exclusive appeals in
an integrating region. Ishiyama and Breuning find that Haas’s expectation is not
supported in a study of European cases (1998, Chapter 7). By exploring the counter
hypothesis, the study of Northern Ireland and Zimbabwe cases should provide additional
evidence.
Propositions are that, if integration proceeds, (1) cooperation among the regional power and installed power will increase and more inclusive strategies are favored; and
(2) cooperation among other states in the region will enhance the needfo r peace and
cooperation in the disputed area, also promoting inclusive strategies. If there is no
regional integration, this variable is not expected to be significant. Sources of data for this
variable are secondary source studies of the region, which include both quantitative and
qualitative descriptions of the progress of integration.
82 Although this section has stated hypotheses about each of four context (“pohtical
opportunity”) variables, it must be remembered that there is little basis on which to
suggest the relationships between the two international variables or among the
international dimensions and the two domestic factors (change in stability of elite
ahgnment and repression). The combinations of the opportunities in support of inclusive and exclusive strategies still have to be considered. Indeed, the perspective of this research is that the interaction of these dimensions is where we need greater understanding, for “all other things” are rarely equal. In this study, when the expectations point in different directions, more attention is paid to change from the previous period.
For example, if all four variables favor exclusive strategies in one case, the expectation is that the leader with the more exclusive strategies wins. In another case, if two variables change to favor inclusive strategies, the prediction would be that inclusive strategies are more favored. Table 2.4 shows that the cases show a range of variation across values for these contextual variables. The case studies in Chapters 4 and 6 evaluate role of the international dimension and the other categories of opportunities in terms of the propositions discussed above, and then the conclusion approaches the question of the relationship among the kinds of opportunities in an inductive fashion.
83 Case 1 Case 2 Case 3 Case 4 Case 5 Case 6 Case 7 NI 82-83 N1 86-87 N1 96 N1 97 Rliod 79 Zim 80 Zim 90
Stability of elite alignment: Colonial power-Controlling pwr Tighter Looser no change Tighter Looser Tighter Looser Colonial power-Regional pwr Looser Tighter no change Looser n/a n/a n/a Controlling power-regional pwr n/a n/a n/a n/a Tighter Looser Looser
Repression: Community Higher Lower no change Higher Higher Lower Higher Subgroup/party Higher no change Lower Higher Higher Lower Higher
International Internal International International International International International Internal ^ Intei*vention/mediation issue issue; opinion issue; US issue; issue; opinion issue issue by states, lOs, or NGOs: favors Hume and others spotlight on against but little recognize Adams Muzorewa involvement Adams
Regional integration: no change increase increase no change no change no change increase
Table 2,4: Distribution of Values for Political Opportunities Across Cases 2.8 Context and Individual Strategies
Now that the four dimensions of context have been discussed, the expectations about which individual strategies (as opposed to exclusive/inclusive overall) tend to be promoted by particular contexts will be stated. Here the context is discussed both in terms of the four political opportunities and more generally. The purpose of probing the validity of these expectations across the cases is to gain a better understanding about when the different theories from which the strategies are drawn may be more applicable, or more relevant in the way leaders construct a given situation.
As noted above, the major bodies of literature give us a “bottom line” about different constructions of reality or bases of explanation successful leaders in mobilization situations tend to use. We know that these theories are not mutually exclusive and that they even address different aspects of ethnopohtical life. For example, social identity theory has been applied to the ways in which people balance different identities according to the size and other characteristics of group membership. But relative deprivation theory is focused on the factors that push people to political action (primarily violence/rebellion).
What we do not know enough about is how varying pohtical contexts affect which ways of seeing the world leaders will emphasize successfully (which strategies appeal to potential audiences). Recall that all the hypotheses stated below are meant to apply to cases that fit the scope conditions for this study; where social mobihzation has occurred
(in terms of increasing education, communications, and economic development, for example) and where a “legitimation crisis” if underway.
85 If the ingroup situation is poor economically, and/or when the group is the subject of political, cultural, or economic discrimination, then the Injustice (or Justice negative) strategy is most likely to be used by the successful leader. Also, if the ingroup is subject to a high degree of repression than others in the society, then it is expected that the successful leader will use appeals to injustice. If the ingroup is subject to violence or under threat by members of other groups in society, then the Enemy Image strategies
(Threat, Blame, and Betrayal) are likely to be used by the successful leader.
If there is willingness on the part of others to talk with the ingroup and/or bring the disputing parties to negotiations, then the Governance strategies are expected to be used by a successful leader. The better the situation, the more likely the Governance strategies used will be the positive variant. If international support for negotiations is present, then the Governance Positive strategies are expected to be more successful in this context. If there is change in stability of elite alignment to favor the ingroup, then Governance
Positive strategies should be favored; if change in the stability of elite alignment favors the major outgroup (such as Protestants in the Northern Ireland cases), then Governance
Negative strategies are expected to appear more in the rhetoric of the successful leader.
If the situation moves toward negotiation and accommodation, then it is expected that Ally Image strategies will be used more frequently by the successful leader. If the leader is trying to take the ingroup in a new direction such as from peace to war or from war to negotiations, and/or when he is calling on group members to make sacrifices
(Stem, 1995), then the Storytelling strategy is expected to be used frequently.
86 Finally, all leaders will use the Identity strategy, but the question here is not when the general Identity strategy is expected but when more exclusive or more inclusive references to Identity are expected to be used by a more successful leader. A simple hypothesis is that if a situation favors a move toward conciliation (such as lessened repression and international support for negotiations), then the more inclusive Identity strategy (including promoting of multiple loyalties) is expected. If regional integration is increasing (see discussion above), then the successful leader is expected to use more inclusive Identity strategies. If the situation is one of great repression and conflict, then the successful leader is expected to use more exclusive Identity strategies. Indeed, in such a situation, group boundaries are expected to be solidified and anyone promoting multiple loyalties is expected to be discredited.
This section has stated the kinds of conditions that are expected to promote individual strategies; the case studies allow an evaluation of these basic suggestions which should help to formulate more specific ideas about when the suggestions of extant theories inform each other.
2.9 Conclusion
This chapter has focused on the research design and methodology. The goal has been to communicate the scope conditions for the study, the case selection strategy, and the justification for choosing the particular areas of Northern Ireland and
Rhodesia/Zimbabwe. In addition, this chapter discussed how the dependent variable was conceptualized and briefly listed the seven cases of mobilization. Now that the method for
87 examining the context and leaders’ appeals has also been described, the next chapter begins with the contextual analysis. Table 2.5 summarizes the variables, indicators, and sources of data.
88 Variable Coding Categories Indicators/ Questions Source of Data
Dependent variable Shift in support Success/Failure Are there (a) observable shifts or patterns in elections data that (b) show significantly increased support for one leader at expense of another? If yes. then the leader benefiting is considered more successful Secondary sources that analyze elections history and party competition
Independent variables
Leaders' Mobilization (1) More exclusive/more (l)Does the leader's strategy' Strategies inclusive (2) Use of specific profile score higher or lower than strategy types the other’s on the e.xclusivit}^ index? If higher, then More exclusive (2) Inductive look at covariation among specific strategies, opportunities, and dependent variable Content analysis of leader’s speeches
Political opportunities: Changes in stability Tightening/Loosening Do the relationships among the colonial power, controlling power, and regional actor grow stronger and more cooperative in maintaining the status quo or in undermining it or do they grow weaker and less cooperative in maintaining or undermining the status quo? If stronger, then tightening; if weaker, then loosening Expert descriptions of country histories and cinrent situation (continued)
Table 2.5: Summary of Indicators
89 Table 2.5: Summary of Indicators (continued)
Variable Coding Categories Indicators/Questions Source of Data
State Repression Combinations of high and Compared to the previous period, low on the community and did the state pass more legislation the individual parties and/or exert more time and resources to arrest and control either the commtinity of the pohtical minoritj’^ or any of the leaders’ parties? Were there repressive events of state violence? If yes, then Higher Secondary source descriptions of policing, justice procedmes
International involvement or Define as internal issue Do the influential actors to which mediation by states, inter and neutral/Define as inter the pohtical minority appeals national organizations, or national issue and neutral/ consider the issue internal and to non-governmental actors Define as international issue be dealt with by sovereign state, or and more sympathetic to one is the issue defined as an leader international problem or threat to international secmity? If internal/neutral, promotes winning of more exclusive strategies; if international/neutral, promotes winning of more inclusive strategies; if more sympathetic to one leader, promotes the winning of that leader UN statements, resolutions, and/or reports; US government statements and reports; secondary sources; news reports from the New York Times and Times o f London (continued)
Table 2,5 : Summary of Indicators
90 Table 2.5; Summary of Indicators (continued)
Variable Coding Categories Indicators/Questions Source of Data
Regional integration No change/ Are there regional initiatives that Increase promote an overarching system and identity (political or economic)? Do these regional organizations have potential authority to affect political and/or economic aspects of the country? If no to either or both, then No change; if yes to either or both, have relationships with others in region expanded since the last period? If yes, then Increase; if no. then No change. Secondary sources' studies of the region, including both qualitative and quantitative descriptions of the progress of integration
Table 2.5: Summary of Indicators
91 CHAPTER 3
NORTHERN IRELAND: CASES AND CONTEXTS
3.1 Introduction
This chapter discusses the four mobilization episodes in Northern Ireland: 1982-
1983, 1986-1987, 1996, and 1997. Each section focuses on a single episode and is organized as follows. First, how Northern Ireland meets the case selection criteria is explained; next the cases are discussed in more detail as evidence is brought to bear to show which leader is considered the winner in each “battle for the hearts and minds” of the perceived nationalist electorate. Third, subsections focus on each category of political opportunities: change in stability of elite alignment; state repression; international involvement/mediation; and regional integration. Applying the propositions about the effects of context on the success of leaders’ mobilization strategies, the case-specific expectations about the winning types of strategies are stated. In this discussion, the evidence from the case study research is summarized. The next chapter presents the results of the content analysis so that the kinds of strategies each leader did employ may be compared and the propositions about the relationship between context and winning strategies may be evaluated.
92 3.2 Background
The Northern Ireland case fits within the scope conditions for this study. In the
1960s, this area experienced demographic transformations whereby the dominated group
(Catholics, or the perceived nationalist community) grew rapidly, socioeconomic development proceeded and encouraged urbanization, and the nationalist community began to undergo rapidly increasing politicization. The state moved closer toward a legitimation crisis where the identity of the country (as Irish, British, or uniquely Northern
Irish) was the focus of debate.
After Ireland’s long history as a British colony, Northern Ireland remained in union with Britain as part of an agreement to end the Irish war of independence. Since the majority Protestants were opposed to union with Ireland, which was desired by the minority Cathohc population, the North remained a province of the United Kingdom.
Notwithstanding its signing of the Anglo-Irish Treaty that established this arrangement, the Irish Free State maintained constitutional claims to all the territory of the island. The facts that London allowed considerable autonomy from the 1920s and that Protestants had political dominance enabled them to maintain a regime that discriminated against Catholics
(for example, in terms of access to voting rights, jobs, housing, and education).
Consumed with economic problems and the need to build the Irish republic, nationalists in the South did not pay much attention to the nationahst cause in Northern Ireland. It was not until the 1960s, in the form of the civil rights movement, that Cathohcs in Northern
Ireland began to protest their situation in large numbers. This movement in turn provided
93 the opportunity for the emergence of several new parties and revamped traditional parties, with different ideas about the meaning that the Irish nationalist identity should have for
Catholics in the province/
As noted, the legitimation crisis surfaced in the late 1960s, largely as a culmination of changes in the demographics of Northern Ireland. The Catholic population was growing more rapidly than the majority Protestant population, rendering it increasingly difficult for the latter to maintain hegemonic control. As a result o f modernization and increasing industrialization, more rural peoples were moving to the cities. In addition, education reforms from earher years were producing a larger and more highly educated
Cathohc population. These changes made this population susceptible to political mobilization demanding greater representation in the majoritarian (Protestant) system.
These demands by Catholics were largely unique, since the more traditional nationalists had centered only on rejecting an “illegitimate” continuing British presence on Irish soil.
The focus of Cathohc politicians had been unification with Ireland and boycott of
Northern Ireland politics as well as Westminster pohtics (participation in the British government).
Although the Irish Republican Army claimed to be the voice of Cathohcs before the civil rights movement, the IRA was winning “no sustained international pubhcity or sympathy, and so httle support from Northern Cathohcs that they felt obliged to abandon armed struggle” by the 1960s. In their 1962 cease-fire announcement, they acknowledged their failure to mobilize the Cathohc population (O'Leary and McGarry,
' It was not until the early 1980s that there was direct electoral competition among nationalist parties; for this reason the first case is 1982. 94 1993, p. 161). In 1967, the civil rights movement dramatically changed the ERA
representation of the problem, which had focused on non-recognition of British control
and the need to assert nationahst claims with physical force. The new view was that
Cathohcs should demand representation and improved hving standards within the society
of Northern Ireland; emerging pohticians appealed to Westminster that these were
Catholic/nationahst rights under British law. Many in the movement viewed this approach as a way to improve the underprivileged situation of Cathohcs until unification with
Ireland could occur peacefully (Purdie, 1990). Indeed, “The Cathohc middle class, while not abandoning their nationahst sentiments, began to seek the reform of Northern Ireland as their first goal; and before long for many it became the overriding goal” (O'Leary and
McGarry, 1993, p. 160). Support grew in the Catholic community for reform in Northern
Ireland, at least as an intermediate solution (O'Leary and McGarry, 1993, p. 168).
Initial concessions by the unionist/Protestant government under Terence O’Neill only served to fan the flames of protest among Cathohcs and set the stage for the new
Social Democratic and Labour Party (the SDLP, of which John Hume eventually became the leader) to transform the sentiment of the civil rights movement into a new pohtical platform. The main plank was that violence was not the answer and unity with Ireland could only be achieved by consent from all communities (O'Leary and McGarry, 1993, p.
170).
Between 1969 and 1971, the British government increased its involvement, trying to balance reform with repression in the Cathohc areas. Eventually, internment (arrest and detention without trial) of suspected IRA members made the Cathohcs susceptible to
95 further political mobilization (O'Leary and McGarry, 1993, p. 176). With Northern
Ireland ungovernable by Stormont, in 1972 the British government abolished the
Protestant-controlled body and assumed direct rule o f the province. Since 1972, the
British have governed Northern Ireland through the Northern Ireland OfiSce (NIG), with occasional attempts to set up power-sharing assembhes. The elections for the fourth such attempt in the fall of 1982 saw the rise of Sinn Fein and changed the landscape o f the conflict. Because this 1982 election was the first in which the SDLP and Sinn Fein competed to be the voice of the nationalist community, it is the first episode examined in this study. The poorest and youngest Catholics (and until 1982 abstentionist) were mobilized by Provisional Sinn Fein with the hunger strikes of 1980-1981. PSF was established as the political arm of the IRA in 1970, but was not legalized by the British until 1974. Because of its policy of “principled abstentionism,” it did not contest elections until 1982. In that year, it was able to take support firom the SDLP, and SF’s success shocked observers at the time.
Despite the deadlock that occurred after Sinn Fein gains in 1982-1983, the SDLP turned the situation to its favor. Hume encouraged constitutional national parties in
Northern Ireland and the Republic to come up with an agreed negotiating strategy with the British. The reasoning was that this would get around the unionist veto and stall Sinn
Fein. The political parties of Ireland, also worried about growing support for Sinn Fein and following an increasingly cooperative dialogue with Britain on the situation, agreed to establish the New Ireland Forum to deliberate in 1983-1984. The proposals that came out
96 of this forum, first dismissed by Britain, eventually became the basis for new talks and the
eventual signing o f the Anglo-Irish Agreement in 1985 (O'Leary and McGarry, 1993, pp.
213-215).
The ALA was important in the representation of the situation in Northern Ireland in
a fundamental way. It distinguished between traditional nationalists who argued that the
unification of Ireland was a basic right o f the majority of Irish people on the whole island,
and the revisionist view that Irish unification should and could only be achieved with the
consent o f the majority within Northern Ireland (O'Leary and McGarry, 1993, p. 228).
This reinforced the way the SDLP was fi'aming the situation. The agreement set the stage
for a second mobilization shift, this time toward Hume and the SDLP.
The third and fourth shifts occurred ten years after the ALA, and perhaps because
of that treaty; indeed, analysts argue that the setbacks of the post-ALA years had forced
Sinn Fein to redouble its efforts. Between 1987 and 1996, many changes in the situation
occurred and are discussed below. In this new context, Adams and Sinn Fein were
successful in increasing their support by fifty percent in 1996, back to the 1983 level.
Significantly, they captured a large percentage of the SDLP transfer votes. Research revealed that events after the 1996 elections brought about even more dramatic changes in the context, thus these two years are broken down into two cases even though the shift was toward Adams in both. Again, the case study in this chapter will describe these years in more detail. An interesting point about 1996 and 1997 is that both scholars of Northern
Ireland and journalists who cover Adams and Sinn Fein tend to highlight the fact that these
97 actors have changed their outlook to become more inclusive than their earher views; it will
be interesting to explore this fact in terms of the theoretical framework of this study and
with the methodological approach developed here.
In summary, the four periods analyzed in this study are 1982-1983, 1986-1987,
1996, and 1997. Tables 3.1 and 3.2 show these electoral shifts. Since this project
investigates the dynamic struggle to pohtically mobilize the nationalist/Catholic population
in the period of modernization, the focus here is the leaders who are seen as the most
prominent nationalist leaders. The individuals focused on then are SDLP leader John
Hume,^ and Sinn Fein President Gerry Adams.
Hume/SDLP Adams/Sinn Fein 1982 Assembly 18.8% 10.1% 1983 Westminster 17.9% 13.4% 1986 Westminster by-elections 12.1% 6.6% 1987 Westminster 21.1% 11.4% 1996 Forum 21.4% 15.5% 1997 Westminster 24.1% 16.1% Numbers are percentage of total vote.
Table 3.1: Percentage of Electoral Support for SDLP and Sinn Fein, 1982-1997'
■ In the initial years, Gerry Fitt was the leader of this party, but John Hume was considered to be the Jess pliant, more nationalistic” visionary. He took over the lead of the SDLP in 1979. 98 Hume/SDLP Adams/Sinn Fein 1982 Assembly 65.1% 34.9% 1983 Westminster 57.2% 42.8% 1986 Westminster by-elections 64.6% 35.4% 1987 Westminster 64.9% 35.1% 1996 Forum 58% 42% 1997 Westminster 60% 40% Numbers are percentage of nationalist vote.
Table 3.2: Percentage of Nationalist Vote: SDLP and Sinn Fein, 1982-1997“
3.3 First Mobilization Episode: 1982-1983
3.3.1 The Competitors and the Campaigns
The mobilization episodes for the first period of Sinn Fein-SDLP competition are
centered on the 1982 Assembly campaign and the 1983 Westminster election, where both
Adams and Hume were elected to Westminster for the first time.“ The SDLP’s successful mobilization of nationalist voters began in 1973; however, to maintain a parallel analysis, this study begins with 1982, the first campaign in which Provisional Sinn Fein challenged the SDLP.-*
In this October election, the Social Democratic and Labour Party won 18.8% of the vote, while Provisional Sinn Fein received 10.1%. Despite the SDLP receiving a higher total (which was also a gain of 1.3% over its win in the 1981 local government elections), analysts of the conflict attribute success to PSF for the fact that, in its first
“ After these elections, Hume but not Adams would take a seat in Westminster, because Sinn Fein members refused to swear allegiance to the Queen.
* PSF had contested the by-elections to first elect Bobby Sands and then a replacement for him but in these elections the SDLP decided not to contest the PSF candidates. The 1982 Assembly elections were the first in which both parties fielded candidates, even though both did so on abstentionist tickets. 99 contested election, it jumped to 10% and captured 34.9% of the nationalist electorate. In
the 1983 Westminster elections, the PSF campaign had an even greater impact, reducing
the SDLP’s 8.7% lead in October to a 4.5% lead only eight months later. Indeed, Sinn
Fein’s percentage of the total vote jumped to 13.4% while the SDLP’s fell to 17.9%. Sinn
Fein’s gains are more striking when the percentage of the nationalist vote is examined;
PSF captured 42.4% of the nationalist vote in 1983. Further, Sinn Fein won a major
symbolic victory when leader Gerry Adams captured the West Belfast seat of the SDLP’s
former leader Gerry Fitt.^
Observers were shocked by the rise of PSF and Gerry Adams. A newspaper
commentary before the 1982 election showed the attitude many took to the PSF bid. The
following are the words of Conor Cruise O’Brien (an Irish politician who opposed Sinn
Fein):
... At the time of the hunger strikes, the Provo s could appear in their most acceptable role: that of martyrs. But at the end of the hunger strike, the Provo leaders denounced a Catholic priest, a fervent republican sympathiser, for helping relatives of hunger strikers to get the prisoners to end their fast, and live. That made a sour end to the H-Block campaign. Memories o f that period are now likely to do Provo candidates more harm than good. But many Catholics are likely to abstain from voting, in an election for an Assembly in which their representatives will abstain from taking their seats...(in The London Observer, 3 October 1982, quoted in Irvin, 1993).
’ Adams was Sinn Fein Vice President in this period, but after the himger strike period and subsequent decision to contest elections in the North he was recognized as the leader in the North. This decision was not fully supported by all of Provisional Sinn Fein, but the Northern faction, led by Adams, went on television and radio in the North to publicize their decision. The President, who gave up power to Adams formally in 1983, was not allowed by the British to travel in Northern Ireland at all (Keimedy-Pipe, 1997). 100 The gains were so unimaginable to policy makers in Britain and Ireland, that Sirm Fein’s unexpected support at the expense of the SDLP inspired those against physical force nationalism to strategize a way to halt the gains of Sinn Fein’s “electoral strategy”
(O'Leary and McGarry, 1993).
In 1982, both parties stood on the ground that they would not take seats in the
Assembly, the latest British approach to restoring governance to Northern Ireland. Still, area specialists argue that this election was very significant in terms of mobilizing the nationalist electorate:
For both the SDLP and Sinn Fein, committed to boycotting the Assembly, the election was an exercise in winning the hearts and minds of the anti unionist voters over the means by which the unification of Ireland should be achieved. SDLP offered a new political dialogue set within the broad context of Anglo-Irish relations. Sinn Fein offered a combination of orthodox political action and armed struggle summed up in the statement by Danny Morrison “with a ballot paper in one hand and an Armalite in the other” (Elliott and Wilford, 1983, p. 33).
A brief description of the Assembly idea sheds light on the reasons for the nationalist opposition to it. The 1973 Sunningdale Agreement’s attempt at power-sharing among parties within Northern Ireland and its institutional link with the Republic of
Ireland were destroyed by a coahtion of unionist groups whose encouragement of a worker’s strike brought the province to a standstill. After this power-sharing executive dissolved in 1974, the British government unsuccessfully tried to broker a multi-party
“convention” (1975) and a “conference” (1980). In April 1982, Prime Minister Margaret
Thatcher’s new Secretary of State for Northern Ireland, James Prior, proposed another consociational arrangement whereby power would be slowly given back to a new
101 assembly as différent policy areas were negotiated by agreement of a 70% majority.
Republic o f Ireland leader Haughey and opposition leader FitzGerald were both critical;
the SDLP objected to the lack of an Irish dimension and feared that power-sharing would
not be insured. As one analysis summarized the SDLP view, “The absence of any even
token recognition of the ‘Irish Dimension’ in the 1982 Act was, for the SDLP, the final
nail in the coffin of the proposals. In their view, the Act threatened to return majority rule
and thus raised the spectre of gross structured discrimination against the minority
population” (Elhott and Wilford, 1983, p. 30). Further, there was no one in the unionist
camp expressing any desire to seek compromise, a fact that made the A.ssembly all the
more unpromising (O'Leary and McGarry, 1993, p. 212).
In this atmosphere, the Assembly elections took place. The PSF and SDLP
platforms and the rhetoric of the two leaders demonstrate very different analyses of the
situation. PSF’s main appeal was, “Break the British Connection; Smash Stormont” {An
Phoblacht/Repiiblican NewsA^ September 1986). They argued that the SDLP would not maintain its abstentionism after the election, and instead would give in to British and unionist pressure. Further, Sinn Fein presented themselves as the new leadership of the nationalist population. John Hume and his cronies were “collaborators” with the British imperialists. To counter the propaganda of the other parties, PSF’s manifesto argued that a vote for their candidates was not a vote for violence but for a “principled stand against the British, a vote for peace with justice. It is a vote against the British presence in
102 reland, and a victory for Sinn Fein will inflict, like Bobby Sands’ election victory,^ a psychological defeat on Thatcher and her military chiefs” (quoted in Elliott and Wilford,
1983, p. 33)
While the PSF campaign was primarily negative, the SDLP supplemented its rejection of the Assembly with an alternative proposal. A “Council for a New Ireland” would provide an Anglo-Irish Framework where members of the new Assembly would work together with members of the Republic of Ireland’s parliament (the Dail) (Elliott and
Wilford, 1983, p. 31). John Hume, leading the SDLP campaign, maintained his belief that any internal solution was impossible, and argued that cross-border cooperation could deal with social and economic issues as well as the constitutional problems. One analysis argues that the salience of the Council proposal for the SDLP campaign actually diminished as the party candidates shifted greater attention to the threat posed by Sinn
Fein. As a result, the SDLP platform focused less on the Council and more on the fact that the violence of the PSF (the “Provos,” short for Provisionals) discredited the latter from earning the label “nationalist.” Towards the end of the campaign, Hume focused more on the differences between his party and Sinn Fein. He framed the vote in terms of violent versus non-violent approaches to conflict resolution, and on the eve of the vote attacked Sinn Fein for its role in causing many of the problems in the Northern Ireland community (Elliott and Wilford, 1983, p. 32).
After the 1982 elections. Prior’s Assembly did convene, but served only as a
“talking shop” for unionists until it was dissolved in the summer o f 1986. In a forum
® Bobby Sands was the political prisoner who was the first to die in the 1981 hunger strikes; he was elected in a by-election to replace a deceased member of Parliament, and then died a prisoner shortly after. 103 without either of the nationalist parties and without the support of the Irish government,
the main unionist party (UUP) promoted reform through greater integration with Britain
and Ian Paisley’s Democratic Unionist Party continued to extol the virtues of majority rule
(O'Leary and McGarry, 1993, p. 213). In this context of virtually no movement toward
solutions acceptable to the nationalist community, Sirm Fein maintained its momentum from the Assembly elections during the 1983 Westminster poll (see Tables 3.1 and 3.2
above).
The campaigns for the 1983 election focused on the conflict for the most part and
again showed the different analyses of SF and the SDLP. The SDLP platform criticized both Thatcher’s monetarist policies and paramilitary violence as the sources of economic problems and joblessness in Northern Ireland, and—as a jab at Sinn Fein’s abstentionism—emphasized that the SDLP was the only nationalist party prepared to offer ameliorative measures. The major theme for the SDLP campaign was its Council of
Ireland idea that had been floated as an alternative during the Assembly vote. By this June election however, Hume had been successful in launching the Fomm for a New Ireland with parties from the Irish Republic just weeks before the election.
Criticism of the Forum idea, especially the fact that it did not involve Sinn Fein or any unionists, was at the core of the SF campaign. In an AP/RN editorial, Sinn Fein described the Forum idea as a “Save the SDLP strategem,” and pointed out that a comparison of the participating parties’ leaders opening addresses at the Fomm showed stark differences in their views of the problems and possible solutions. These differences would render the Fomm useless, in Sinn Fein’s view; “Thus, with FitzGerald almost
104 openly rejecting Irish unity, Haughey declining to consider constitutional change, and
Spring determined to focus his eyes on the twenty-six counties alone, John Hume had little ground beneath his feet when he rose to speak” {AP/RN, 2 June 1983). In this editorial and in its election manifesto, Sinn Fein characterized the SDLP’s approach as “dangerous” to nationalists because it ignored that British occupation is the real source of the problem, and instead focused more on unionist fears that the Protestant people would receive unjust treatment in a united Ireland. Sinn Fein’s campaign also justified its abstentionism fi’om
Westminster, emphasized its work at the local level through advice centers, and argued against the “conspiracy” to isolate Sion Fein using censorship and the Forum (SF manifesto, 1983). Finally, in a significant move, Sinn Fein dropped its longstanding £/>e
Nita idea after its party conference in late 1982. This was its vision for a federal Ireland in which the North would be one unit, giving unionists a strong voice.
For mobilizing resources, which may be considered in terms of the fimdraising/patronage, media access/coverage, and party/campaign organization each leader could bring to bear, the research into this aspect of the case revealed that both leaders had similar “quality” of mobilizing resources, but the kinds of resources they had enabled them to reach different kinds of audiences. In terms of party and campaign organization, neither side was very organized in this first time period. Party headquarters for both were always being attacked and bombed. Further, Sinn Fein candidates out
105 canvassing were harassed by the RUC and the British Army, but at the same time SDLP
candidates were harassed in some of the working class Belfast areas known as ERA
strongholds/
The state’s involvement with television and radio limited the access Sinn Fein had
to most forms of media. SF has developed Republican News as a way to get around the
censorship of the British, and have used this outlet not only as a way to seek increased
support within the nationalist community but also as a way to reach “world liberal
opinion.” In 1979, Republican News combined with .<4/7 Phoblacht and a leading figure
within northern republicanism, Danny Morrison, took over. From 1979 until 1982,
Morrison built the weekly into a paper with an international focus, devoting more and
more space to overseas causes perceived as similar to Sirm Fein’s, such as the African
National Congress in South Afiica, the Palestine Liberation Organization, and the
Nicaraguan resistance (Davis, 1994, pp. 65-66).
In contrast, Hume and his party worked through more “normal” political charmels
such as bringing home works projects fi'om the European Community and campaigning in
the United States for attention for the Northern Ireland problem and the constitutional
nationalist position. Hume’s side also placed campaign advertisements in newspapers read
by more middle-class voters (in contrast to the “underground” An PhoblachùRepublican
News), and Hume himself had a periodic column in the Irish News through which he had been able to disseminate his party’s perspective. Overall, because the channels used by both leaders were so different it is difficult to judge which should be considered the “best
' These same themes appeared repeatedly in this researcher’s interviews with SF and SDLP members (Belfast, September 1997). 106 equipped.” It actually seems to depend on the audience: Hume was best equipped to mobilize people who did not live in the working class areas where Sinn Fein bad gotten the people into the streets—and the only place where An Phoblacht/Republican News was distributed. Hume was better equipped to reach those people who would see his articles in the Irish News and see his face on television—the emerging middle class.
The next section returns to the propositions derived in Chapter 2, summarizing the evidence and weighing whether the selected aspects of context would be expected to favor inclusive or exclusive strategies.
3.3.2 The Context
For this first case, three of the four political opportunity variables indicate that the leader with the more exclusive strategies is expected to win; the regional integration variable is not expected to be significant. Table 3.3 summarizes this assessment and the rest of this section presents the evidence.
107 Change in stability of elite alignment; Colonial-Controlling (A) Tighter Colonial-Regional (B) Looser Strategy' predicted to win Exclusive
Repression: Community Higher Subgroup/party Higher Strategy predicted to win Exclusive
Intervention/Mediation Absent from International Commimity Strategj' predicted to win Exclusive
Regional Integration Strategy predicted to win not signif
Overall prediction for Exclusive Winning strategies (3 of 3)
Individual strategies favored Injustice, Enemy Image, Storytelling
Table 3.3: Summary of Expectations for Northern Ireland, Shift 1
Change in Stability o f Elite Alignment
The evidence gathered for this case shows that the relationships between the unionists and the British and the Republic of Ireland and the British both favored the status quo. In the terms employed here, the British-unionist alignment got tighter and the relationship between the British and the Irish repubhc entered a phase of deterioration. In this scenario, the British appeared intransigent to peaceful demands from the nationalist community, and no options were open on any other front either. Thus, the hypothesis is that this aspect of the context promoted the winning of the leader with more exclusive mobilization strategies.
108 British and unionists were still closely aligned in this period, which was visible with
Thatcher’s now-famous statement that Northern Ireland was as British as Finchley, a
constituency in England. This strong aUiance was also apparent when Prior gave in to the
unionist demand to eliminate the power-sharing guarantee and the Irish dimension from
any plans for an assembly. Indeed, in 1970s, SDL? was offered power-sharing as a right
of the nationalist community, as well as the recognition that an Irish dimension was
necessarj'. In the 1982 White Paper proposing the new Assembly, the Irish dimension was
rejected, and unionists were effectively given a veto over power-sharing since they would
have over 70% of the membership (White, 1984, p. 237).
The cohesion among unionists also favored the status quo, which would have further undermined any British efforts to reach out to nationahsts. Indeed, all of the unionist parties were strongly against any measure of cooperation (a key factor that would change in the 1990s). In March 1982, Hume dehvered a speech in which he appealed to the unionist community to work with nationalists, but no cooperation was forthcoming
(White, 1984, p. 238). Then the unionist parties convened the Assembly (after the
October 1982 elections) without any nationalist parties and proposed only closer integration into Britain or majority voting rules as “reform” measures (O'Leary and
McGarry, 1993, p. 213). Further, when invited to the Forum for a New Ireland, the Ulster
Unionist Party, the Democratic Unionist Party, and the Alliance Party of Northern Ireland
(these are the major Unionist parties) all rejected invitations to join in negotiations with the SDLP and the parties from the Republic of Ireland.
109 On the other hand, the cooperation between the British and Irish governments that was central to Hume’s plan was unstable in this period. Thatcher and Ireland’s FitzGerald had agreed to set up an Anglo-Irish Inter-Govemmental Conference in November 1981, an institution which Hume had hoped would show nationalists that “constitutional politics could still deliver the goods” (White, 1984, p. 228). However, Irish opposition at the
United Nations (under Haughey’s leadership) to Britain’s involvement in the Falklands
War in April 1982 interrupted this cooperation until November 1983. Therefore, during the entire period examined here, there was deadlock (Bew and Gillespie, 1993). Also, the split between Thatcher and Haughey on the Falklands issue was said to boost unionist confidence in Britain’s guarantee to the majority Protestants in Northern Ireland (White,
1984, p. 228).
Further undermining the idea of changing the status quo with help fi"om the
Republic, the cooperative effort among Irish constitutional nationalist parties in the Forum for a New Ireland was unstable because of the weak commitment of the Irish parties at the beginning. Despite FitzGerald’s commitment to Hume’s idea, the economic crisis in the
South made the rest of the government wary about giving unity such high priority (White,
1984, p. 244). Haughey, whose rhetoric was much Uke Sinn Fein’s and who may have been an advocate for a more republican approach, was involved with internal political problems.
Had Britain and Ireland engaged in efforts to address some nationalist complaints, and had Britain been willing to go against most unionist opinion—as happened in the
110 1990s— more inclusive strategies might have seemed plausible. Instead, the alignment of
elite groups maintaining the status quo offered more incentives for exclusive strategies to
succeed.
Repression
The evidence shows that the level of repression against both community and the
SF/IRA group was at the highest levels in 1982 and 1983 than in any o f the other four time periods examined. Based on the discussion in Chapter 2, this situation is expected to promote exclusive strategies. No evidence was found that revealed any easing of repressive measures.
First, looking at repression against the whole community, security legislation was being applied to its fullest extent. For example, the Northern Ireland Emergency
Provisions Act, instituted in 1973, gave powers to the Army and the RUG to stop and question any one “suspicious” and to search houses without warrants. This Act was reinforced by the Prevention of Terrorism Acts of 1974 and 1976. This kind of emergency legislation created an environment of repression (Kennedy-Pipe, 1997, p. 65).
In addition, in this period the RUG, seen as a Protestant police force, took over the security regime from British army; and there was more frequent use of plastic bullets resulting in numerous killings.
Further, the supergrass system was started in late 1981, and mass trials began in
1982. “Supergrasses” was the name given to “converted terrorists” or informants, who gave names of those involved in terrorist activities. Allegedly, they were mistreated and forced into this role by the security forces. From November 1981 to November 1983,
111 over 600 people were charged with terrorist activity. Also in 1982, after Sinn Fein did so well in the Assembly elections, the RUC adopted what has been described as a “shoot to kill” policy (Irvin, 1993, p. 322; see also Kennedy-Pipe, 1997).
Perhaps the most symbolic evidence of high repression in this period was the behavior of the British government that led to hunger strikes by prisoners accused of terrorism. In the 1980-81 period, the British government had refused all concessions to political prisoners, and their reaction was to go on hunger strike to demand the acknowledgment that they were not criminals (but political prisoners). The British intransigence on the hunger strike issue resulted in rioting, and in response to this the security forces used plastic bullets on people in the nationalist areas.
As a result of the new vigor to enforce security legislation, repression was also high against the BRA/Sinn Fein. Indeed, the Sinn Fein/IRA movement was a special target for enforcement of the Prevention of Terrorism Acts. Further, Sinn Fein was subject to media censorship under laws to prevent terrorism. In the Republic, Section 31 banned
Sinn Fein spokespersons, and in Britain there was a ban (lifted after 1983) on certain SF leaders— such as Gerry Adams—from travelling to England, Scotland, or Wales.
In summary, during this period the British treatment of the hunger strikers was fresh in peoples’ minds, and this event had “shown” nationalists that the British were willing to ignore basic human rights to protect the unionist community.
112 International Involvement/Mediation and Regional Integration
For this first time period, evidence supports the idea that the conflict was treated
more as an internal issue; this variable therefore favors exclusive strategies. In this post
hunger strike period, general international opinion seemed to be against what was viewed
as Britain’s repression of the minority. For example, in 1982 the European Commission
on Human Rights condemned the British reaction to the hunger strikes and the British
army’s use of plastic bullets (O'Leary and McGarry, 1993, p. 214). This opinion favored
both the views of the SDLP and SF in that it brought increased attention to the situation.
However, there was a perception that Sinn Fein and the ERA were one and the same—
terrorist organizations that had to be dealt with forcefully.
Despite this support for some nationalist views, this general public opinion did not translate into action. The norms of conflict resolution (and intervention in conflicts) in the
international system were such that the European Community and United States were still reluctant to get involved; also the US was committed to its “special relationship” with
Britain. The United Nations also would not get involved because of a British veto.
Ireland had called for a peacekeeping force to be sent in 1969, but Britain exercised its veto, arguing that intervention would violated the UN Charter since Northern Ireland was a domestic issue (Coogan, 1996, p. 90). Finally, public opinion in the Republic of Ireland was hostile to ideas about unification and the socialist goals put forth by Adams, but they were not supportive of Hume either. Instead the focus was internal, dealing with an
113 economic crisis and also avoiding the issue of the North altogether, behavior of the Irish
government was discussed above. Again, the assessment is that the predominant
international treatment was that the status of Northern Ireland was an internal—British—
issue.
The regional integration variable is not considered significant in this period because
there is no increase or decrease in integration; this case is prior to the Single European Act
of 1987 which truly set off the advance of greater pohtical integration (see Goodman,
1996).
3.3.3 Expectations about Individual Strategies
As shown, the evidence used to evaluate the contextual variables predicts that each
variable favors exclusive strategies. In terms of individual strategies, the context above is
expected to promote the success of the Injustice strategy, the Enemy Image strategy, and
the Storytelling strategy. The reason these three are suggested comes fi-om the kinds of
conditions suggested by the theories on which the strategies are based. Because of the great disparity between Catholics and Protestants in Northern Ireland and the fact that this disparity was receiving increased attention, the kind of situation described by relative deprivation theory seems to have been close to the Catholic’s situation; thus the Justice
Negative strategy should appear fi-equently. Also, the deprived situation as well as the repressive behavior of the British and the violence from Protestant community* are the kinds of factors that make emphasis on threat and blame more likely. The Storytelling
* This is not a statement that Cathohcs were not also violent against Protestants. 114 mode is suggested as encouraged by the context because of the need to mobilize people
around a changing sense of identity (challenging the state for equal rights) and because of
the great uncertainty the perceived nationalist community was facing during the continuing
legitimacy crisis.
3.4 Second Mobilization Episode: late 1985-1987
3.4.1 The Competitors and the Campaigns
After Sinn Fein made remarkable gains in the 1982-1983 period, the SDLP, the
Irish parties, and the British government seemed to perceive a shared the goal of stopping the growth in support for republicanism and physical force nationalism. The Anglo-Irish
Agreement was the way they sought to do this. Indeed, the Agreement was intended to stabilize and increase support for the SDLP at the expense of Sinn Fein, and to encourage divisions among unionists that would make power sharing a more acceptable outcome.
Several elections after the ALA, which was signed in November 1985, do show its success for all o f these three goals. The analysis o f the impact on SF-SDLP competition from one observer reflects a consensus among students of the conflict:
It [the Anglo-Irish Agreement] has halted the growth of the Sinn Fein vote, and shows some signs o f reversing it: the SF vote fell in each if the three [1986, 1987, 1989] post-Hillsborough elections... And the SDLP’s position, while not hegemonic, has been restored: it has stemmed and reversed the SF tide, albeit within a growing Nationalist bloc (O'Leary, 1990, p. 16).
115 Observers point out several elections in this second period that reveal this shift/ The focus of this study will be the January 1986 Westminster by-elections, forced as a “mini referendum” by unionists who resigned their seats when the government rejected their calls for an actual referendum, and the June 1987 Westminster elections.’® From the peak of Sinn Fein's support in 1983, the 1986 elections, where fifteen seats were contested, showed a swing firom SF to the SDLP of 10.8%. Elliott (1986, p. 6) notes that an even better test of the Anglo-Irish Agreement’s effect is a comparison between May 1985 local government elections and the by-election. “In this case there was a swing of 6.5% fi-om
SF to SDLP in eight months compared to 4.3% fi-om 1983 to 1985.” Even though
Adams’ West Belfast seat nor Hume’s Foyle seat were contested in 1986, observers took this as a measure with which to gauge the contest between the two parties (Elliott, 1987, p. 13).
Similarly to the January 1986 elections, the 1987 Westminster elections were widely seen as a vote on the Anglo-Irish Agreement (Elliott, 1987). In this election the
SDLP increased its total vote to 154,000—which is higher than it had been since 1974 when they had no nationalist competition—and its number of Westminster seats to 3 of the 17 allotted for Northern Ireland. In comparison to Sinn Fein’s 11.4% of the total
’ Elections for local government councils were held in May 1985, before the signing of the AI A. These were the first non-abstentionist elections for local councils contested by Siim Fein, and the party won 35% of the vote. This was a victory for them (Irvin, 1993), but in the elections following the AIA (those discussed here), the SDLP was widely regarded as cutting back Siim Fein’s earlier gains from the hunger strike period and perhaps from these local elections as well.
The 1989 local government elections show approximately the same percentages for SF and the SDLP as the 1987 elections (actually there was a shift o f. 1 percent to SDLP from SF). This study focuses on the 1986 and 1987 elections because the issues on which it was fought were broader than in local contests, and because the results in the later elections were similar to 1986 and 1987. 116 vote, the SDLP captured 21.1%. Despite the fact that Adams retained his West Belfast
seat, this election was said to show that the SDLP’s constitutional nationalism was gaining
appeal over the earlier gains of republicanism (Taylor, 1990, p. 10).
The themes of the campaigns in this period show many disagreements in the
parties’ views. In 1986 and 1987, the SDLP emphasized the gains o f the Anglo-Irish
Agreement, including improvements in the justice system and commitment to fair
employment and housing allocation. The SDLP also campaigned on its accomplishments
of the AIA for how the treaty recognized the Irish identity of the Northern Ireland
minority and gave formal recognition that the Republic of Ireland should play an
institutionalized role in resolving the conflict. Finally, it formalized the British and Irish
commitment to “unity by consent.” This phrase meant that if majorities in both the North
and South vote for unity, then Britain would not obstruct that goal (O'Leary and
McGarry, 1993).
For Sinn Fein, the 1986 slogan was “Keep up the Pressure,” based on the idea that
it has been the SF vote as well as the “armed struggle” that has forced British into making
promising concessions {AP/RN, 9 January 1986). In 1987, Sinn Fein’s key emphasis was
the failure of the Anglo-Irish Agreement to deliver, especially on issues of justice, housing,
employment, and cultural recognition. Sinn Fein also criticized the AIA because the
‘\inity by consent” idea formalized and validated the unionist veto.
In terms of mobilizing resources, both leaders were able to take advantage of or create conditions in their favor. In this second period, Hume was bringing even more money into Northern Ireland, both from his work in Westminster and from the United
117 States in the form of the pledged International Fund for Ireland. Sinn Fein moved ahead
of the SDLP in terms of party organization in this period by establishing a number of
advice centers in nationalist working class areas. By making more contacts with potential
followers at a local level, Sinn Fein and Adams boosted their image in these areas.
Further, Sinn Fein continued to hold marches and raUies. The media picture was basically
the same as in the last period, and AP/RN countered every “positive” action Hume made
for the nationalist community by describing it as against republicans and harmful to
“ordinal}' people.” Overall, then, it appears that both sides had similar levels of resources
but they used very different channels
3.4.2 The Context
The research showed that all four political opportunity variables predict that a
leader using more inclusive strategies should be more successful in this period. Table 3.4 summarizes these expectations, which are justified below.
118 Change in stability of elite alignment; Colonial-Controlling (A) Looser Colonial-Regional (B) Tighter Strategy predicted to win Inclusive
Repression: Community Lower Subgroup/party Higher Strategy predicted to win Inclusive
Intervention/Mediation PresenL somewhat favoring Hiune from International Community Strategy predicted to win Inclusive
Regional Integration Increasing Strategy' predicted to win Inclusive
Overall prediction for Inclusive Wiiming strategies (4 of 4)
Individual strategies favored Governance Positive, Injustice, Storjtelling
Table 3.4; Summary of Expectations for Northern Ireland, Shift 2
Change in Stability o f Elite Alignment
It is expected that the loosening of the alignment between the British and the
unionists, at the same time that the alignment between the Republic and Britain tightened,
led to the perception that negotiation and compromise were feasible, making inclusive
appeals more likely to succeed. The British-unionist relationship had “loosened” since the first case, meaning that Britain was no longer as willing to rescue the unionist people.
Gone were the days of 1974 when the security forces would stand by and watch Unionist
119 intransigence destroy any chances for negotiations." The good reasons to seek a way out
of Ireland seemed to outweigh the motivation to remain: Thatcher had narrowly escaped
death at the hands of the IRA in the Brighton bombing (October 1984), public opinion in
Britain was increasingly hostile to Britain's involvement, and the opposition Labour Party
(since 1981) had committed itself to promoting unification o f Ireland by consent (O'Leary
and McGarry, 1993). Since 1983, the Liberal Party in Britain had also been promoting the
idea that British troops should be withdrawn fi-om Ireland (Kennedy-Pipe, 1997, p. 106).
The tightening of the British-Irish relationship is exemplified in Britain’s decision
to negotiate with Hume and ROI representatives of constitutional nationalism (the Forum
parties). This was a major shift. It signified to both observers and participants the
recognition by Britain that its efforts to arbitrate the conflict since it abolished Stormont in
1972 had been a complete failure (O'Leary and McGarry, 1993). It should be noted that
British cooperation with the Republic was undermined somewhat by British statements
assuring unionists that the AIA confirmed British sovereignty. In an effort to play both
sides, the British government said the agreement made unionists better off* and any actions toward union with the Republic would have to be by consent of a majority in Northern
Ireland. While Sinn Fein played this up as entrenching the unionist veto, all other parties emphasized how monumental the AIA was in addressing demands of nationalists.
“ In 1974 a ‘"forced” strike engineered by the Ulster Worker’s Council to protest against a power-sharing arrangement with the Catholic minority resulted in some of the worst violence of the Troubles before the British govenunent finally declared a state of emergency. Even then, the British soldiers and the RUC dragged their feet to dismantle barricades manned by UDA (the Ulster Defense Association) (Coogan, 1996. pp. 170-171; Fisk, 1975). 120 Repression
Pressure for reforms from the international community, especially the European
Community, brought changes to the degree of repression the nationalist community
suffered in the second period. Further, the Anglo-Irish Agreement went far toward
dealing with the repressive measures suffered by the community as a whole. On the other
hand, with the cooperation and consent of the European Community, the repression of the
British and Irish states focused in on the IRA (and by association, Sinn Fein), for whom it
was relentless. Along with harassment of people in known IRA areas, both countries
imposed broadcasting bans and signed the European Convention on the Prevention of
Terrorism. Because the repression on the whole community lessened but repression of the
IRA and Sinn Fein activists increased in this period, the expectation is that more inclusive strategies are more hkely to be successful.
Evidence that repression was easing can be summarized briefly. Increasing international pressure on Britain and from Dublin within the AIA meant that the general situation improved somewhat for most Catholics/nationahsts (see Kennedy-Pipe, 1997).
As promised by the SDLP, the Anglo-Irish Agreement focused on improving the justice system for ordinary people in the nationalist community. In 1986 and 1987, the agreement was seen as dealing with crucial problems such as unemployment, housing and the justice system (Rolston, 1987, p. 78). Indeed, because of the AJA, the conduct of security forces was under greater scrutiny from international and British investigators, RUC ofBcers prosecuted by British for interfering with investigations, in 1989 the RUC was reformed by new chief Annesley to be more responsive to both communities (see Kennedy-Pipe,
121 1997). Further, the British stopped using supergrasses, there was increased cooperation between Britain and the Republic on extradition, and Dublin finally signed the European
Convention on the Suppression of Terrorism Act in 1987.
The latter European agreement was one sign that repression against the IRA/SF group increased in this period. In addition, the security forces stepped up their efforts against the IRA (see Kennedy-Pipe, 1997); Sinn Fein accused the RUC and British army of widespread harassment of SF campaigners. It was in this period that repressive measures in the Repubhc and in the UK in response for the IRA bombing campaign were taking their toll on the Provisionals (Rolston, 1987).
There is some evidence that repression on the community increased—but not repression from the state (the focus of research here). Instead, the IRA stepped up its violence in Northern Ireland and elsewhere after the Anglo-Irish Agreement. Many
Cathohcs died in these operations; a major part of Hume’s campaigns were that the largest threat everyone faces is the ERA violence. The fact that they had increased their operations may have made Hume’s accusation seem true (Kennedy-Pipe, 1997).
International Involvement/Mediation and Regional Integration
Research shows that significant changes occurred on the international front since the hunger strike era; the European Community and actors within the United States increasingly called for international attention to the problem. Although in this period there was not direct international mediation, the fact that the issue was no longer viewed as a
122 strictly internal problem means that this variable is expected to favor the leader with more
inclusive strategies. The research for this case revealed the following evidence to support this claim.
First, the United States got more involved. In May 1984, Congress backed
Hume’s Forum strategy in the first resolution on Ireland since 1918 (White, 1984, p. 232).
After AIA was signed, Reagan stated that the U.S. would pledge economic aid (DeYoung,
1985); this aid came in March 1987—Just a few months before the Westminster vote—and was in the amount of $50 million for the International Fund for Ireland (Bew and
Gillespie, 1993, p. 204). Second, in the Republic of Ireland political parties were clear alhes o f Hume with the decision to maintain participation with Hume in New Ireland
Forum, and to reach a compromise agreement on unified position toward Britain. After the AIA was signed, FitzGerald assured nationalists that AIA provides new opportunities, giving Catholics a greater voice in political institutions and providing protection fi-om political and economic discrimination (DeYoung, 1985).
On the European fi-ont, the Haagerup Report was issued in March 1984. This report presented the results o f a study of Northern Ireland’s conflict and significantly went beyond the previous EC stance that Northern Ireland was an internal problem. It supported Hume’s initiative, arguing that the British and Irish governments should set up a fi-amework to resolve conflict. The EC exhibited this change in its view of the conflict in another way, calling for an agreement between Ireland and Britain to deal with the
123 conflict. Further (through Hume’s initiative), the European Community worked to
address the economic aspects of the conflict with infrastructure projects and other aid.
Northern Ireland became a major recipient of EC funds.
The regional integration variable is also expected to favor more inclusive
strategies. The progression of integration meant that Northern Ireland was receiving
economic aid from the EC, perhaps drawing increased attention to the problem of IRA
terrorism in Europe and against EC projects in Northern Ireland (Goodman, 1996).
3.4.3 Expectations about Individual Strategies
For each of the four contextual variables (change in stability of elite alignment,
repression, international involvement, and regional integration), inclusive strategies are
expected to be more successful. Also, almost all international involvement favored
Hume’s views to some extent and was not neutral (Adams and Sinn Fein still received
support from some Irish American groups such as Noraid). In terms of individual
strategies, the improving situation of the nationalist community, increased international attention, and decreased repression suggest that the three most frequently used strategies by the successful leader should be different from the first case. The Injustice strategy should be less frequent, though still present. The Justice Positive strategy, while not one of the three most frequently used strategies, should be observed more by the more successful leader as recent gains are emphasized. Also, the Enemy Image strategies should be less frequent unless the IRA is the target. The Governance Positive strategy should come into play for the first time, as there are many incentives that would validate a view of the
124 situation which mentions frequently the importance of negotiation. Finally, to build
community consensus around the idea that the AIA provides acceptable governance
structures to address Catholics’ aims, the Storytelling strategy is also expected.
3.5 Third Mobilization Episode: 1996
3.5.1 The Campaigns and the Competitors*^
1. Background
The third Irish case examines 1996 because in this period, Gerry Adams and Sinn
Fein increased their support so that the losses to the SDLP in the post-Anglo-Irish
Agreement period were regained. Indeed, in 1996 Forum elections, Siim Fein captured
42% of the nationalist vote—back to the 1983 levels. Adams gained even more support in
1997 as well, but further research into the 1997 case revealed that it should be considered
separately from 1996. As with the two cases discussed above, the nature of this shift is
not only reflected in electoral results but in the perception of their meaning in evaluations
by observers. In fact, with the two recent periods, a sifting frrough American media
coverage of the April 1998 peace talks and agreement makes it clear that Adams and Sinn
Fein had truly captured the limelight. Before presenting more evidence for the shifts in
support to Adams, a summary of the situation between the former period and the most
recent one is necessary because much happened in those ten years.
After the perceived advances of the SDLP and setbacks or stabilizing of SF documented in the previous period, the situation for SF became even more dismal. For
‘•Because of the number of years between the second (1985-1986) and third (1996) mobilization episodes, this background section is longer than the others and is divided into two subsections. 125 example, the Enniskillen bombing in late 1987, the accidental ERA killings of several
Australians in Belgium, the massive damage in the London Stock Exchange bombing in
1992, a broadcasting ban imposed on political parties connected with paramilitary groups
in both the United Kingdom and the Republic of Ireland, and Gerry Adams’ loss of his
West Belfast parliamentary seat in 1992 are a few of the events that damaged nationalist
(and international) sentiment toward Sinn Fein (Bew and Gillespie, 1993).
In the late 1980s and early 1990s, dramatic changes within republicanism and in
the relationship between republicanism and constitutional nationalism were started by and
are the result of many factors including Sinn Fein’s aforementioned difiSculties. In
addition, by 1996, significant changes in the structure of the international system and the
norms of the system had shifted the context of the struggle between the two nationalist parties. For example, the period between the late 1980s and the first IRA cease-fire of
1994 was marked by; (1) the ending of SF’s abstentionism in the Irish Republic’s Dail; (2) changes in the social and political profile of nationalists in Northern Ireland toward a larger middle class with a greater interest in a capitalist economy; (3) the growing economy of the Republic of Ireland; (4) deepening integration in the EU and the effect of the European Union’s single market and regional programs on the need for North-South integration in Ireland; (5) the end of the Cold War; (6) changes in British leadership to
John Major (in 1997, the impending election of Labour leader Tony Blair became even more influential); (7) changes in the United States’ role and the mediation of neutral
George Mitchell; (8) failed talks between Hume and Adams in 1988 and more successful
126 talks in 1993;^^ (9) shifts within the unionist bloc promoted by Britain’s denial of this
peoples’ Britishness in statements such as the Downing Street Declaration, and by “pan
nationalism,” or increased cooperation among Hume, Adams, and the Dublin parties; and finally, (10) enough IRA-British attacks and counterattacks to leave the region in a political and military stalemate.
A 1987 Sirm Fein policy document, A Scenario fo r Peace, was a forerunner of more fundamental shifts within Sinn Fein in the 1990s. This document stated that the established nationalist parties should launch an international campaign to bring about peace, and for the first time hinted that British withdrawal alone would not solve the problems between the two traditions who share the island. Further transformation was encouraged by (1) talks between John Hume and Gerry Adams in 1988 and again in 1993;
(2) Northern Ireland Secretary Brooke’s 1990 statement that Britain would not stand in the way of Irish unity as long as a majority in Northern Ireland voted for it; and (3) the formalization of this statement in the December 1993 Downing Street Declaration signed by British Prime Minister John Major and Irish Taoiseach Albert Reynolds. In the latter.
The 1988 talks foundered because the focus was on difFerences in policies, strategies, and goals between SF and the SDLP; perhaps the largest obstacle was Adams’s refusal to agree with Hume that Britain was "neutral” and would support unity by consent. In 1993, the talks were more productive because inter- party differences were set aside to focus on the issue of an ERA cease-fire and how to get Siim Fein into inclusive negotiations. Hume’s talks with Adams caused a major rift within the SDLP because SF was at a low ebb; dissenters argued that Hume’s actions brought SF back firom their near-defeat. One SDLP cotmcilor stated, "We helped Sinn Fein and now they are standing on our fingers” (interview with author, BelfasL September 1997). The joint statements issued by Hume and Adams during this period were especially objectionable to some SDLP members, especially because these statements were faxed to newspapers on Sinn Fein paper. Further, the fact that Adams and Hume decided not to publish the details of the agreement reached in 1993 created great debate in all circles (Fay, 1996). 127 the British accepted that the people of Ireland alone by agreement between the two parts should be allowed to exercise the right to self-determination on the basis of consent, freely and concurrently given. North and South.
Towards a Lasting Peace in Ireland (1992) articulates a more complete analysis of
Sinn Fein’s recent perspective and reflects changes in how four actors are depicted;
Britain, the Republic of Ireland, the SDLP, and the unionist people. In this statement, SF recognizes Britain’s role as persuader and emphasizes pluralist concepts, which was a major move away from the original “Brits Out” strategy. In the 1992 document and in many speeches and statements since, republicans “discussed the removal o f the unionist veto in terms of winning unionist consent through both positive reconcihation and British involvement in conflict resolution and persuasion” (Bean, 1996, p. 12).
Towards a Lasting Peace also shows a major shift in Sinn Fein’s view of the
Repubhc of Ireland and the SDLP. The new emphasis is on the need for dialogue between republicans and constitutional nationalists, so they can present a united front to negotiate with Britain for the “self-determination” of Ireland. The Dublin government is called on to meet its responsibihty of providing democratic (not sociahst) answers to bring peace to
Ireland: SF had thus moved away from the idea of sociahst revolution to sweep out the collaborationist establishment (Bean, 1996, p. 16).
FinaUy, a change in the republican view of the unionists is visible, from seeing the
Protestant community in Northern Ireland as part of the conflict and not a product of
British occupation. In this way, Sinn Fein has moved closer to the SDLP view of the situation as an internal conflict. Up until recently, the unionists have been viewed as
128 “political collaborators. ..They drew strength from the reality of British power and from a false consciousness” promoted by their privileged position. The key idea was that when the British withdrew, this people would realize who they really are (Irish, albeit of a
“different tradition”) and would join in sharing a political system with other Irish people
(Bean, 1996, p. 18). Now Sinn Fein leaders emphasize the need for nationalists/repubhcans to understand the fears of unionists so that everyone could work together on building a new political system. The unionist people have been used and manipulated by the British, and nationahsts have to make them feel protected and esteemed within new structures. These changes set the stage for the 1996 period.
2. Campaign Summary
This section discusses the struggle between Hume’s SDLP and Adams’s SF to win the vote of nationalists and thus receive a mandate for their particular approaches to the peace process. The significant elections for this period were held on 30 May 1996, and were intended to select those who would attend an all-party talks forum. The forum idea emerged as a compromise measure when a first round o f talks after the 1994 republican and loyalist cease-fires stalled on the issue of decommissioning. The British government insisted that all paramilitary organizations start surrendering their weapons before all-party peace talks could begin. Several o f the parties, supported by the United States, argued that the decommissioning and the talks should take place in parallel. The Unionists, led by newly-elected David Trimble, indicated that they might agree to this idea if constitutional negotiations took place through an elected assembly.
129 This plan for all-party talks was pushed ahead by the actions of American
ambassadors (especially the Ambassador to Ireland, Jean Kennedy Smith), National
Security Advisor Anthony Lake, and former US Senate majority leader George Mitchell.
In November 1995, the work of US actors with the British and Dublin governments led to the issuing of the Downing Street Joint Communiqué. This announced that both governments would work with an International Commission (chaired by Mitchell) to achieve compromise on the decommissioning issue. In January 1996, Mitchell issued a study that said the paramilitaries would probably never surrender their weapons first, and that to start negotiations all parties should commit to six principles of democracy and non violence. This plan would commit Sinn Fein (and the IRA) to the democratic process for the first time, and provide a way to proceed with decommissioning during the peace process (as opposed to before or after) (Coogan, 1996, p. 409).
The Irish parties expected Mitchell’s plan to clear the way for all-party talks to begin by the end of February. However, Unionist pressure on Major led to the call for elections as a precondition for the all-party talks. Major’s decision was an outrage to
Dublin and the Northern Ireland nationalists, as well as Irish-Americans. Frustrated by what they called the British government’s “stalling tactics,” the IRA broke the cease-fire with the London Canary Wharf bombing on 9 February 1996. In response, the Irish and
British governments broke contact with Sinn Fein negotiators until the ERA reinstated the cessation. Still, the elections would be held in May 1996 and talks excluding Sinn Fein would begin that summer.
130 At first the SDLP and Sinn Fein rejected the Forum elections, but after the SDLP
reconsidered, Sinn Fein decided to contest the elections while expressing finrstration with
the SDLP for not working with SF to achieve a common position against the British and
unionists (AP/RN, 25 April 1996). In the 1996 race, SF argued that their campaign would
“seek support for its republican analysis on the basis that there is widespread support for
real talks and a negotiated settlement. Siim Fein will highhght the legitimate demand for
an end to British rule, for equahty and for the release of political prisoners” (AP/RN, 25
April 1996). Overall, Sinn Fein emphasized that people had to vote SF to show Major
that they had a right to a place at the talks table, and a vote for Sinn Fein was a vote for
peace.
The SDLP campaign emphasized their non-violent approach and continued to argue for institutions within Ireland and between Ireland and Britain that would be modeled on the European Union. Before the vote, the SDLP issued a warning to the electorate that “a vote for Sinn Fein in this election was a vote for the on-again, off-again tactics of the ERA which risks bringing the horror of sectarian murder back to our streets.”
On the other hand, a vote for the SDLP was a vote for the integrity of the peace process to which John Hume had dedicated his life (Fay, 1996).
In the Forum elections, Sinn Fein received 15.5% of the vote (winning 17 seats at the Forum), which has been called the highest vote ever for “militant Irish republicanism”
(though SF campaigned on a non-mihtant platform), increasing the vote by over 50% of their share in the 1992 Westminster election. In numbers of votes, Sinn Fein went fi-om
78, 291 in 1992 to 116, 377. Also in this election, SF took the title of “largest party in
131 Belfast,” especially important because Belfast had always been seen as a unionist city
(interviews with Sinn Fein councilor Chrissy McCauley; see also Fay, 1996). Although interpretations of the election focus on the shift toward Sinn Fein within the nationalist bloc, the SDLP did receive 21.37% of the total vote which won that party 21 seats at the
Forum.
The balance between the leaders in terms of mobilizing resources changed from somewhat equal to tilted in Adams’ favor. What is now seen as a response to their suppression under the Anglo-Irish Agreement, Adams and Sinn Fein have developed a powerful party and campaign structure. In terms of fundraising and patronage, Hume continued to bring investment from Europe and the United States to Northern Ireland but still not much money for the SDLP. Since Gerry Adams had obtained permission to enter the U.S. in 1994, he had collected large sums for the Sinn Fein operation. Adams’s travel to the US, especially widely publicized trips in February 1994, September 1994, and autumn 1995 provided finances for the party, investment for the region, and diplomatic support (Coogan, 1996). Further, the legitimacy gained also helped fundraising efforts in other countries.
Sirm Fein was able to use this money to build up its party and campaign organization. By 1996, the focus on local advice centers which began in the mid-1980s had grown into an efficient and effective base within the nationalist community. Far from its abstentionist policy of early years, local Sinn Fein councilors are now perceived as people who get things done for the nationalist community. Sinn Fein activists moved into
132 the trade unions, tenants’ and residents’ associations, and the women’s movement. In the
effort to make the armed struggle more legitimate, branches of the party
have organized classes in Irish history, language and culture, while the main party publication. An Phoblacht, has sought to identify the IRA with other ‘national liberation’ movements, such as the PLO and the ANC. And the advice centers established by Sinn Fein provide an alternative administrative structure: in some Catholic working-class districts they offer the only effective direct link with government, while assisting prisoners’ dependents and organizing ‘neighborhood policing’ and social activities (Taylor, 1990, p. 11).
In this period, Sinn Fein was also participating in politics in the Republic as well as
in Northern Ireland. In 1986, the party voted to participate in the Dail if elected. In the
first elections in which SF participated there (February 1987), the party won only 2% of
vote. In 1987 ard fheis, self-evaluation by the party concluded that the problem getting
votes in the Repubhc was the focus on the national question at the expense of “everyday
interests of people.” As a result of this as well as the changes within Northern Ireland,
SF’s campaigns began to focus increasingly on social and economic policies (Wilson,
1987, pp. 7-8). Thus, by 1996 the party had a revamped approach that they claimed was
more relevant to peoples’ daily lives in North and South.
In terms of media, Sinn Fein and Adams again turned the balance in their favor.
After the 1988 broadcasting bans were imposed on Sinn Fein spokespeople, the party
launched a new International Publicity and Information Committee. This group produces the “Ireland International News Briefing,” which targets overseas (and especially
American) audiences (Miller, 1994, p. 112). By January 1989, earlier formats of international communications that emphasized ERA war operations had been replaced with
133 reports on social issues and coverage of SF political initiatives with only a brief chronology of IRA operations. Further, the photographs depicting armed militants in action or demolished buildings have been replaced with shots of Sinn Fein councilors in action (Irvin, 1992, pp. 82-83).
Once the censorship laws were lifted, Hume had to share the “official” stage with
Adams, who could now appear undubbed on television, and whose articles and editorials began to appear alongside Hume’s in mainstream newspapers. On top of this, Sinn Fein also maintained An Phoblacht/Republiccm News and had extended its media and fundraising reach abroad with a revamped newsletter sent to several countries.
With this background on the campaigns and the evaluation that Adams was considered to benefit from the mobilization shift, the next section discusses the context in terms of the four variables.
3.5.2 The Context
For this case, three of four political opportunities are expected to be significant
(there is no change in the stability of elite alignment), and all three hypothesize that the leader with the more inclusive strategies is expected to win. Table 3.5 summarizes the predictions.
134 Change in stability of elite alignment; Colonial-Controlling (A) Looser Colonial-Regional (B) Tighter Strategy predicted to win Inclusive/no change so not significant Repression: Community Lower Subgroup/party Higher Strategy predicted to win Inclusive
Intervention/Mediation Present somewhat favorable to Adams from International Community Strategy predicted to win Inclusive
Regional Integration Increasing Strategy predicted to win Inclusive
Overall prediction for Inclusive Winning strategies (3 of 3 )
Individual strategies favored Governance Positive, Ally, Injustice, and Storytelling
Table 3.5: Summary of Expectations for Northern Ireland, Shift 3
Change in Stability o f Elite Alignment
This third case shows significant changes in the ehte ahgnment that had originally
maintained the status quo to which nationalists objected. Because the changes are in the
same direction as the previous case, however, this variable is not considered significant for
the predictions in this case (the evidence will still be described here, however). Indeed, the signing of the Anglo-Irish Agreement is symbolic of the way Britain finally stood up to
demands of the unionists and began to work more closely with the Irish government.
Both the further loosening of the British-unionist relationship and the tightening of the British-Irish ahgnment—and their effects on intra-nationahst politics—are symbolized by the Brooke speech of 1990 and reaction to it. For years Hume argued to his audiences
135 that Britain did not have an objection to Irish unity if a majority in Northern Ireland
approved, and that the job of nationahsts was to persuade unionists. Sinn Fein had always
denounced this idea, but SF began to show signs of change after Secretary of State
Brooke announced that “the British government has no selfish strategic or economic
interest in Northern Ireland; our role is to help, enable, encourage.” The Downing Street
Declaration (1993) grew out of the Brooke initiative and eventually produced the
Framework Documents in February 1995. Resembling Hume’s ideas, this framework
discussed three dimensions of negotiation; internal North, East-West, and North-South.
Repression
By 1996, Britain and Ireland (and the United States) were cooperating to bring all
the parties together to negotiations. The lifting of the broadcasting bans were symbolic of the more open climate. The evidence presented below warrants the assessment that the levels of repression during this period were lower on both the community and the individual parties (in the Northern Ireland cases it is in fact only Sinn Fein and the ERA and not the SDLP that are periodically subject to repressive measures).
Indeed, after the cease-fires of 1994, the security forces scaled back their activities in Northern Ireland. Significantly for the sensitivities of nationalists, the British army ended their patrols of the streets of Derry and Belfast (Kennedy-Pipe, 1997, p. 161). The cease-fires also resulted in a scaling back o f security force raids on IRA strongholds.
After the cease-fires, it was announced by Britain that the Ulster Defense Regiment (the
UDR, perceived by nationalists to be a Protestant regiment consisting of sectarian forces who could not get into the British army or the police/Royal Ulster Constabulary) would
136 merge with the Irish Rangers to form the Royal Irish Regiment. Since the Irish Rangers were about thirty percent Catholic and as an organization has been successful in recruiting people from both the North and South of Ireland, the reform was intended to be more fair toward the nationalist population (Kennedy-Pipe, 1997, p. 162). Finally, as noted above, the Repubhc, and then Britain, lifted their censorship pohcies that had been directed at
Sinn Fein. All of this points to a more open atmosphere than in previous periods, in which repubhcans were treated more like pohtical activists and less like criminals.
International Involvement/Mediation and Regional Integration
Since the late 1980s, several aspects of the international dimension continued to evolve toward favoring more inclusive pohcies, according to the propositions derived in
Chapter 2. It is important to note that the post-Cold War “new world order” does not mean that from now on the international dimensions hypothesized to be important will favor inclusivity and conflict resolution in cases of this type; since the end of the Cold War the many “internal” conflicts over issues of nationahsm and identity have been treated in rather different ways.
This third Northern Ireland case is the first of the mobilization episodes affected by the norms of conflict resolution and intervention which had changed dramatically since the end of the Cold War. Advancing peace processes in South Afiica and the Middle East may have demonstrated to republicans that a global trend is ceasing violence to work with others, and they could in turn point to these other cases to show how negotiation is legitimate and can reap benefits. Also, the change in norms of conflict resolution meant that the US was now more involved in offering mediation and guarantees (O'Leary and
137 McGarry, 1993, p. 332). Clinton’s presidency especially took this stance; a major goal of
his foreign policy program has been to resolve regional conflicts. Perhaps the most
powerful embodiment of change was the decision by the United States to grant a visa to
Adams in February 1994, boosting his legitimacy (though in 1996, Adams was not invited
to the White House for St. Patrick’s Day). Adams’s statement in an interview with Vanity
Fair (January 1997, p. 131) shows the power of this move within the context of Northern
Ireland; “Any conflict resolution process in the absence of the two sides all of a sudden having a Damascus Road conversion requires an international dimension...President
Clinton has radicalized entirely the prospect for peace.”
Along with norms of conflict resolution, ideas about sovereignty and international intervention to deal with internal conflicts changed substantially after the Cold War, as the
UN under Boutros-Ghali and the United States formulated “peacemaking” as an important part of the agenda o f the international community. Whereas sovereignty has been held up as a “negative,” to limit the interference of external actors (for example, Jackson, 1990), in recent years the idea of “responsibility” has become connected to it (for example, Lyons and Mastanduno, 1995). This means that a sovereign state has a duty to provide peace to people with in its borders, and all states have a responsibility to intervene if there are violations of this responsibility.
In late 1995 and January 1996, involvement of the international community was crucial as republicans and unionists reached a stalemate over disarmament; this can be summarized briefly. The US and other parties stepped in to argue that decommissioning and negotiations should take place in parallel. On the eve of Clinton’s trip to Northern
138 Ireland, US diplomats working with Irish and British governments were able to announce in late November 1995 that negotiations and decommissioning could proceed in parallel and that an International Commission chaired by former Senator George Mitchell would be set up to figure out the best way to deal with this (Coogan, 1996, p. 407). In January
1996, Mitchell’s report on this issue and a plan to proceed rescued the peace process.
Then, the British government under John Major basically stopped the talks, as it bent to unionist pressure and announced that elections to a negotiating forum must be held.
The American response to the Major announcement o f an Assembly in 1996 can be classified as supporting both Sinn Fein and the SDLP against unionist/British intransigence. Seventy US Senators and Representatives, under the banner of the Ad Hoc
Committee on Ireland, wrote to Clinton to denounce Major’s move. This condemnation was bipartisan, which lent it even more credibility. US media was predominantly hostile to
Britain as well (Coogan, 1996, p. 411). This was followed by a meeting between Adams,
Lake, and Clinton, after which a statement was issued (1 February 1996) supporting
Mitchell’s approach to all-party peace talks and— between the lines—condemning Major’s move (Coogan, 1996, p. 412). It is argued that this kind of involvement was important in affecting leadership mobilization within Northern Ireland because it favored a more inclusive view; we can work with unionists because we have support only for negotiation, and only successful negotiation will bring increased economic aid.
The regional integration variable favored inclusive strategies. Evidence is that by the 1990s, the Single European Market and the regional programs of the EU created a logic that has made North-South integration in Ireland a necessity. It has also made the
139 treatment of Northern Ireland as part o f Britain “iUogical.” This force has even led unionists in Northern Ireland to agree that an “all-Ireland” framework in the SEM is needed, although disagreement exists on the structures for it (Goodman, 1996, p. 114).
Increased European integration also made possible a governmental action that reinforced the US and United Nations view that the situation in Northern Ireland was no longer an internal matter: in September 1995, the European Court of Human Rights ruled that the killing of three IRA personnel by SAS forces at Gibraltar (1988) was illegal (Kennedy-
Pipe, 1997, p. 168).
3.5.3 Expectations about Individual Strategies
In terms of individual strategies in this period, the need to encourage nationalists to support cooperation with the other nationalist party (SF or SDLP) and with unionists means that the successful leader is expected to use a high percentage of Governance
Positive and Ally strategies. Also, the involvement of the United States in the peace process adds to the expected use of Governance Positive. Injustice and Storytelling are also expected as the winning leader needs to maintain cohesion, would be unrealistic to drop the idea that Catholics are subject to deprivation, and needs to connect current concessions to the past struggle in order not to be “selling out” the struggle.
140 3.6 Fourth Mobilization Shift: 1997
3.6.1 The Competitors and the Campaigns
For this final shift, most of the background information provided for 1996 remains relevant. However, there were several differences between these two years that changed the context in which both nationahst leaders sought the mandate of Catholics in Northern
Ireland. Sinn Fein emphasized that their victory in the 1996 Forum elections awarded them a mandate to negotiate on behalf of a significant part of the nationalist community in
Northern Ireland. However, when SF leader Adams and chief negotiator Martin
McGuirmess showed up at Stormont for the talks on Jime 10, 1996, they were refused entry to the first meeting of the Forum. Events in the remainder of summer 1996 appeared to squander chances for peaceful negotiations even further.
The marching season of 1996 must be considered especially important for the background to the 1997 elections. Indeed, though the Westminster elections were not until June, the Drumcree incident of July 1996 is given credit for galvanizing the nationalist population’s support for Sinn Fein. In summary, the Orange Order march, which celebrates the victory of (Protestant) King William over (Catholic) BCing James in
1690, was blocked from coming into nationalist areas (Garvaghy Road) at Drumcree, in
Derry/Londonderry. Royal Ulster Constabulary Chief Armesley, pressured by the unionists via the weak British government, reversed this decision and riots followed.
The events at Drumcree, and a similar outcome on the Lower Ormeau Road, Belfast, showed continued intransigence of Major’s government and its “bias” toward unionists.
" The government was considered "weak” because (1) Major needed Unionist votes in Parliament and (2) he had to call an election soon and it was a foregone conclusion that he would lose to Labour leader Blair. 141 the brutality of the RUC toward Catholics, and continued British treatment of Catholics as second-class citizens/^ One member of Sinn Fein who works in the West Belfast ofiBce as an election director summed up the significance of these events for the 1997 elections. He said that the British made it easy for SF to win; while other parties blamed the IRA for the breakdown of the talks process, “everyone knew that was wrong” (interview with author,
Belfast, September 1997).
In the June 1997 Westminster elections, the nationalist vote was the highest of the total Northern Ireland vote that it had ever been: 40.2%.^® Because o f this overall growth, both SF and the SDLP increased their percentage of the total vote from the previous (1992) Westminster elections: the SDLP received 24.1% and Sinn Fein received
16.1%. However, looking within the nationalist bloc there were noticeable and interesting shifts toward SF that have been the subject of discussion by observers, though the SDLP plays up its absolute gains (interviews with SDLP officers, Belfast, September 1997).
Indeed, one analyst noted that SF had made a “remarkable gain” on the SDLP since 1992.
First, the break down between the two was more like 60 : 40 (SDLP : SF) instead of 70 :
30. Second, and symbohcally important, Adams won back his West Belfast seat from the
SDLP, which he had lost in 1992 (Elliott, 1997). The third point deals with vote transfers in the single transferable vote proportional representation system. Although SF voters
How serious these events were to nationalists and future prospects for peace is well-articulated in an article 'mAn Phoblacht/Republican News on 18 July 1996, aptly titled, “Orange State Exposed: Anger Must Become New Force for Change.”
‘^Census figures from 1991, published in 1996, revealed the changing population in Northern Ireland. Whereas the assumption had been that the population was about one-third Catholic and two-thirds Protestant, the total population of 1,577,836 was revealed to be 43.1% Catholic and 56.9% Protestant (Coogan, 1996, p. 434).
142 frequently transfer their votes to the SDLP, the rate of transfers in the opposite direction is
much lower (Irvin and Moxon-Browne, 1989). While the SDLP-to-SF transfers had been
generally around 7%, in the June 1997 elections, Sinn Fein received 35% of the vote
transfers from SDLP first preference voters (Feeney, 1997).
The SDLP and SF campaigns for these elections were, not surprisingly, focused on
the peace process. The Sinn Fein campaign showed the finstration with the British
government that was galvanized by the Drumcree incident and the decommissioning
issue.*’ Commentators have said that Sinn Fein's campaign was “reminiscent of the post
hunger strike era. Sinn Fein is saying that a vote for it is a vote against the British
government’s intransigence” (Breen, 1997). Indeed, media coverage m .AP/RN sounded
much like it had in the 1970s and early 1980s, emphasizing the repressive role of the
colonial government. While (according to the SDLP) SF had borrowed Hume’s voters in
the 1996 election, in 1997 Sinn Fein “was aiming at ownership” (Breen, 1997).
In terms of mobilizing resources, Adams and Sinn Fein continued to build their
network. A recent New York Times editorial (by Maureen Dowd, April 1998) sums up the
remarkable job that Sinn Fein’s public relations people have done to turn Adams and other
candidates into legitimate politicians for whom the average, middle class person could
respectably vote:
Tomorrow the revolutionary Socialist will stroll through the New York Stock Exchange with its chairman...[Adams] has meetings with top financiers and fund-raisers with celebrities in New York, before coming to the White House...to see the national security adviser, Sandy Berger— with a probable Clinton drop-by. Anjelica Huston, Donald Trump and Martin Sheen have all partied with Gerry. Adams groupies include young
' This was discussed in the previous section: tlie debate over whether the IRA and the loyalist paramilitaries must disarm before attending peace talks. 143 Clintonites who worked on the Irish issue. “H e’s their Ho Chi Minh,” dryly notes Mike Barnacle, the Boston Globe columnist.
In this fourth mobilization episode, one additional factor tilting the balance even
more in favor of Adams and Sinn Fein does stand out, and again is an element of
campaign/party organization. The continuation of marches and ralhes, long an element of
Sinn Fein’s arsenal, may have been a crucial factor in this period that helped shift some of
the SDLP’s support to Adams. During the events of the summer of 1996, Sinn Fein
sponsored marches to condemn the abuses of the British and the RUC. A central theme
was to demand that nationalists be treated as first class citizens and not subjected to the
humihation and physical abuse sponsored by the state. This move by Sinn Fein appealed
to all nationalists and, at least for a short time, Adams and his party spoke the words that
everyone wanted to shout. The SDLP was working behind the scenes to relieve tensions,
and inadvertently allowed themselves to be overrun by Sinn Fein’s leadership. The extensive cultivation and use of mobilization resources by Sirm Fein compared to the
SDLP may have weighed heavily in the outcome of this final mobilization period.
Despite this evidence outside the model applied in the next section (with the focus on the other four contextual factors), one should not conclude that mobilization resources were the only explanatory factor. The analysis will show that the message Adams was trying to get out with these mobilization resources was also a crucial part of the equation.
Indeed, the results o f the mobilization strategy analysis show us an interesting way that
Sinn Fein was able to capture a portion of SDLP supporters. As one observer argued, in
144 this election campaign Sinn Fein may have succeeded by stealing the language of the
SDLP (Feeney, 1997). The next section discusses expectations based on the contextual
variables.
3.6.2 The Context
This section provides the evidence to justify the prediction that a more exclusive
leader is favored by the context in this period. As shown in Table 3.6, two variables
promote more exclusive strategies and one promotes more inclusive strategies. The latter
is moderated by the fact that international involvement is argued to have favored Adams.
Change in stability of elite alignment: Colonial-Controlling (A) Tighter Colonial-Regional (B) Looser Strategy predicted to win Exclusive
Repression: Community Higher Subgroup/party Lower Strategy predicted to win Exclusive
Intervention/Mediation Present, favorable to Adams from International Community Strategy predicted to win Inclusive (but favoring Adams)
Regional Integration no change Strategy predicted to win not significant
Overall prediction for Exclusive Winning strategies (2 of 3)
Individual strategies favored Enemy Image, Governance Positive, Injustice, Storytelling
Table 3.6: Summary of Expectations for Northern Ireland, Shift 4
145 Change in Stability o f Elite Alignment
The change in the alignment from June 1996 until the 1997 elections is a central
reason for considering 1997 as a separate case. Because of events between the 1996
Forum elections and the 1997 Westminster vote, the case moves back to the setting of the
1982-1983 period. Indeed, because there was a tightening of the British-unionist
relationship and a loosening of the British-Irish relationship, the expectation is that more
exclusive strategies are likely to resonate with the pubhc.
Evidence that the British-unionist alignment tightened is exhibited in the British
stance toward Sinn Fein. Sinn Fein won 17 seats in the Forum election in May 1996.
However, when representatives of SF appeared for the first day of the Forum, June 10,
they were refused entry to the talks until the IRA declared another cease-fire. Many saw this as Prime Minister John Major bending to unionist demands, favoring the status quo,
and applying a double standard; loyalist parties were not banned from talks for violence of
summer 1996, though SF’s links to violence (breaking of IRA cease-fire) were reason for exclusion from talks.
Other events in the summer of 1996 seemed to reverse the previous loosening of the British-unionist alignment and now supported the idea that the British were defenders of the “Orange state.” In July 1996, an Orange Order march was blocked from coming into nationalist areas at Drumcree. RUC Chief Annesley reversed this decision and riots followed. Observers and members of both the SDLP and SF said Drumcree was one of the biggest mobilizers for SF support in the 1997 elections, because events at Drumcree
146 march showed intransigence of Major’s government and bias toward unionists, brutality of the RUC toward Catholics, and British treatment of Catholics as second-class citizens.
The argument that the British-Irish alignment was loosening is based on the fact that after the repression of the nationalist community in Northern Ireland by the British during the summer of 1996, Irish Prime Minister Bruton condemned Britain’s behavior and turned a cold shoulder to further cooperation without concessions to nationalists.
Repression
Next to the stability of elite alignment, degree of repression is the second area where there were significant changes from 1996 to 1997. Whereas repression on the community and parties within the community had eased in 1996, the climate changed in the summer of 1996 and in subsequent months. When the unionists pressured Major to get the RUC to change the marching season regulations, the nationalist community was beaten back with plastic bullets as they tried to protest the Orange order marching through nationalist areas. During that summer, other similar events made the RUC and British government appear as if they were once again treating nationalists like second-class citizens. Despite the overall lessening of repression since the extremes of the 1970s and
1980s, this turn of events during the marching season caused such an outrage among all nationalists that it heightened the perceived sense of repression. Marches, rallies, and newspaper editorials from the time and from the elections which followed are indicators of this perception. Importantly, this sense was widespread throughout the nationalist community and not only among republicans, Sinn Fein supporters, and/or working class areas. Therefore, while it may be argued that the repression on the ERA. as target remained
147 at similar levels, the Drumcree incident warrants the conclusion that repression from the
unionists and the British on the nationalist community as a whole increased dramatically.
International Involvement/Mediation and Regional Integration
For these categories, there is one change from 1996 to the 1997 period; the lack of
change in the degree of regional integration means that variable is not expected to be
significant in this case. In large part, the involvement o f the United States as mediator
(especially the George Mitchell group) still favored inclusive strategies. The change here
is with international opinion about involvement in the resolution of the conflict. It was
hypothesized in Chapter 2 that when international opinion is more general (supporting
resolution of the conflict and an end to violence versus supporting a more inclusive or
more exclusive solution) and the intervention of an actor from the international arena (the
UN, the United States, the European Union) is “neutral,” this dimension favors inclusive
strategies. However, it was also stated that if international public opinion and
international intervention support one leader’s view, that leader is more likely to be
successful in mobilizing people within the nationahst community to support him, whether
his strategies are more inclusive or exclusive. While evidence shows that Hume was
favored in the second case (1986-1987), Adams received more attention in 1997.
During the summer and autumn of 1996, Adams campaigned internationally, in person and through the use of editorials and the Sinn Fein newsletter, to bring attention to the renewed repression and to try to get international actors to condemn Britain. In contrast, Hume worked more behind-the-scenes. For this reason, the outrage in the international community—primarily from human rights groups, the European Union, the
148 United States, and the Labour Party (still in opposition) in Britain—provides some evidence that this variable supported more exclusive strategies, that it was necessary to bring international pressure to bear to “get tough” with the unionist parties and British government under John Major. The evidence for this variable is summarized briefly.
American involvement made it clear that the conflict was still being treated as an international issue, but also that the U.S. did not agree with Major’s treatment of Adams and SF. For example, after a devastating June 1996 bombing in Manchester, England,
Clinton issued a statement condemning the violence but assuring the people of Northern
Ireland that the United States was committed to helping maintain the peace process. Then later in June another White House statement pledged, “We’ll continue to remain in touch with Sinn Fein. We’ll continue to remain in touch with all the parties so long as we deem that to be useful in addressing the peace process itself” (Coogan, 1996, p. 433).
The Irish government supported the American view of the situation. In December
1996, Clinton met with Taoiseach Bruton in the United States and both countries backed entry into talks for SF if the latter would renew the cease-fire. And previously, on July
12, 1996, the Republic’s Bruton blamed the British government for the violence of the state against its citizens. Irish Foreign Minister Dick Spring called for an Anglo-Irish conference to deal with the crisis.
The prediction that international mediation favored Adams is qualified because the support was not necessarily for Adams over Hume. The fact that there was no change about the general idea that this was an international concern but that there was a qualified change in the views about the nature of Britain’s orientation to support Adams’ definition
149 of the situation suggests that the dimensions that changed should be given more weight in
the predictions. Further, it does make sense to argue that leveling the playing field so that
Adams has as equal an opportunity as Hume to participate in talks is in fact tipping the
balance in favor of Adams. The discussion o f the results in Chapter 4 will allow a more
specific statement about how this variable may have affected which leader was more
successful.
3.6.3 Expectations about Individual Strategies
In terms of the individual strategies that are favored by this particular context, the increase in repression by the Protestants and the RUC, and the behavior of the British that was seen to support these actors, are expected to encouraged the predominance of Enemy
Image and Injustice strategies. In addition. Governance Positive strategies should continue to be successful because of the continued commitment to negotiations by the
United States and most of the leaders in Northern Ireland. Finally, the Storytelling strategy is expected to be necessary to win in order to connect the current plight to the past struggle yet change people’s ideas that negotiation (now with the support of the international community) is the solution, not hunger strikes or other means.
3.7 Conclusion
This chapter has discussed each of the four Northern Ireland mobilization episodes, detailing the leadership competition and the particular domestic and international contexts in which the struggle to “win the hearts and minds” of the perceived nationalist
150 community took place. For each case, two sets of predictions were made relating the kinds of strategies that a successful versus unsuccessful leader are expected to use in each context: the individual strategies and then more exclusive versus more inclusive strategies.
Understanding the individual strategies which tend to be used by winning leaders in particular contexts will advance our understanding about the lenses through which leaders want their followers to view their situation. This kind of knowledge is important because how one views a situation—for example, in terms of relative deprivation and/or Enemy
Image—affects what kinds of solutions are deemed acceptable—monetary aid and programs to better one subgroup’s plight, or protection from another group perceived as posing the threat of annihilation.
Understanding how repression, change in elite alignment, international involvement, and/or regional integration affect the individual strategies and the predominance of more or less inclusive strategies (as an overall measure) is important for observers who wish to intervene to promote either greater inclusivity or exclusivity. In past cases, how do these sets of variables interact? The next chapter presents the results of the content analysis so that the propositions stated here can be tested. Table 3.7 summarizes the variables across cases.
151 Change in Repression Intemat’I. Regional Overall Winning Stability of involvement integration prediction leader elite alignment 1982-1983 Exclusive Exclusive Exclusive not signif “ Exclusive Adams 1986-1987 Inclusive Inclusive Inclusive not signif. Inclusive Hume 1996 not signif Inclusive Inclusive Inclusive Inclusive Adams 1997 Exclusive Exclusive Inclusive not signif Exclusive Adams ■' As explained above, "not significant” means that these variables do not come into play in the case. For both the change in stability of elite alignment variable and for the regional integration variable, a change is necessary for the hypotheses to apply (see discussion in Chapter 2).
Table 3.7: Summary of Northern Ireland Cases
152 CHAPTER 4
NORTHERN IRELAND;
ANALYSIS OF LEADERSHIP MOBILIZATION STRATEGIES
4.1 Introduction
This chapter evaluates the expectations described in the previous chapter. Taking each of the four mobilization episodes one at a time, the individual strategies used by each leader and the degree of exclusivity' of all the strategies used by the leaders are compared using the method described in Chapter 2. At the end of the chapter, the cases are compared across time to see if other trends in the fit between strategy and context are observable.
4.2 First Mobilization Episode: 1982-1983
4.2.1 Expectations
For this case, change in stability of elite ahgnment, repression, and the international involvement variable all predicted that more exclusive strategies would be more successful; the regional integration variable was not significant in this period. The context was also expected to promote the following individual strategies: Justice Negative
(Injustice), Enemy Image, and Storytelhng.,
153 4.2.2 Comparing Strategies
Table 4.1 compares the percentage of all strategies coded of each individual
strategy. Hume uses three strategies with roughly equal frequency (20-22%); Injustice
(Justice Negative), Governance Positive, and Storytelling. On the other hand, Adams’
two most frequently used strategies are the Enemy Image (23.9%) and the Injustice
(22.2%) strategies. This distribution fits weU with the two leaders’ individual positions in
the politics of the country and in the position of their potential constituents: both must
emphasize the injustices faced by Catholics in Northern Ireland or would probably never
get any support. However, Hume also notes that the people must pursue positive efforts
directed toward negotiation and changing the political system to fix injustices (Governance
Positive), while Adams is more focused on pointing the finger of blame at the British and
the Protestants, emphasizing the threat posed by the SDLP “sell-outs,” and reminding the
Catholics that the Republic of Ireland has betrayed the Irish nation (recall that the three
types of Enemy Image strategies are Threat, Blame, and Betrayal). Another difference is
that Hume’s Identity strategy does focus on Northern Ireland while Adams defines the
group in terms of the entire island of Ireland: 6.3% o f Hume’s strategies are the
Identity/Northern Ireland while Adams compares at only .3%.
Comparing these results with the predictions, the winning leader, Adams, did
indeed use most frequently the three strategies suggested by the context to be most hkely to appeal to nationalist voters. The probability of getting this result randomly is only .018,
154 showing that the findings are significant with an exact test.* In this case, it appears that
Hume’s relatively heavy use o f the Governance Positive strategy was ill-fitting with the
situation in 1982-1983; people did not seem to think working with others and rebuilding
political institutions in Northern Ireland were good ideas.
' The exact test indicates the exact probability of the leader using a specific number of predicted strategies. It is the appropriate test for this kind of data because it does not assume a normal distribution. For the cases in this study in which three strategies are hypothesized, the probability of getting three correct is .018; the probability of getting at least two correct is .286, and the probability of getting at least one correct is .821. For cases in which four strategies are hypothesized, the probability of getting all four correct is .014, the probability of getting at least three of four is .24, the probability of getting at least two of four is .757, and the probability of getting at least one of four is .986. For all seven cases, significance is either .018, .014, or .24, with 2 instances of .018 probability, 3 instances of .014 probability, and 2 instances of .24 probability. The reader should note that all of the .018 and .014 probabilities (5 of 7 total cases) are stronger results than the usual .05 level standard employed for parametric statistics. 155 Hume Adams (n=652) (n=6L8) Identit}" (Northern Ireland 6.3% .3% references Outgroup Negative (Enemy 11.7% 23.9% Image)*** Justice Negative*** 22.5% 22.2%
Storytelling*** 20.2% 14.6%
Governance Negative 7.8% 14.2%
Governance Positive 22.1% 11.2%
Outgroup Positive (Ally Image) 5.7% 7.8%
Justice Positive 3.5% 6%
Total %‘ 99.8% 100.2%
The ”n" here is the total number of strategies and the number given in each cell is the percentage of this total. The strategies in bold are the strategies hypothesized to be used most frequently by the winning leader, in this case that is Gerry Adams. The asterisks denote the actual strategies used most frequently by the winning leader. The chance of getting the three matches randomly was only .018 or 1.8%, making these results in accord with standard canons of significance.
** ** Because Northern Ireland is the focus of the research and considered to be the society imder exam ination, these references are the focus (Adams tends to focus much more on Ireland as the society). This figme excludes references to the leader's party because this is misleading about the groups in the society the leader is defining.
Due to roimding off of the niunbers, these totals do not equal exactly 100%.
Table 4.1 : Comparison of Strategies for Shift T
156 4.2.3 Assessing the Predictions: The Exclusivity Index
This section examines the leaders’ appeals to assess how inclusive/exclusive they were. In the first case, they are compared to each other; in the other three Northern
Ireland cases, the leaders are compared to each other and to the previous period. The analysis proceeds by comparing the strategies the leaders used and the targets of these strategies (divided into strategies which have outgroups as their targets and strategies which have ingroups as their targets); for each, an exclusivity ratio is calculated. A final measure sums the exclusivity ratios for each of the above; the leader with the higher ratio is considered more exclusive than the other.
Nature o f Strategies
As discussed in Chapter 2, the strategy models are generally more or less exclusive in their nature; for example. Enemy Image and Injustice are exclusive. Ally Image is inclusive. Governance is inclusive if it is used in a positive way and exclusive when used to condemn institutions or negotiations. The following table divides the strategies in this manner. The Identity/exclusive strategy includes the Northern Ireland and island of
Ireland references with the “very exclusive” delineation and the Identity/inclusive strategy includes the Northern Ireland and island of Ireland references with the “very inclusive” delineation.^
' The island of Ireland references are included here and not in the previous table because of the different purposes of the two measures: here we want to look specifically at how inclusive and exclusive the leaders are and including the island references gives a fuller picture. 157 More exclusive strategies Hume Adams (n=671) (n=650) Identiw 1% 1.7% (NT, I "very exclusive" references) Outgroup Negative (Enemy Image) 11.3% 22.8%
Justice Negative 21.9% 21.1%
Storytelling 19.7% 14%
Governance Negative 7.6% 13.4%
Total exclusive strategies 61.5% 73.1%
More inclusive strategies Hume Adams (n=671) (n=650) Identity 7.9% 3.5% (NT, I "very inclusive” references”) Outgroup Positive 5.5% 7.3%
Governance Positive 21.5% 10.6%
Justice Positive 3.5% 5.7%
Total inclusive strategies 38.4% 27.3%
Assessment
Hume Adams Exclusivity ratio “ 1.6 2.7 “This ratio is calculated by dividing the total percentage of exclusive strategies by the total per centage of inclusive strategies. The higher number indicates that the leader used more exclusive strategies than the leader with the lower number.
Table 4.2: Comparison of Exclusive and Inclusive Strategies for Shift 1
The comparison of strategy profiles shows that Adams was more exclusive in this first period. Some of the major differences are in the Enemy Image strategies and in both the Governance Negative and Positive strategies. Adams was more likely to use the
158 Enemy Image and Governance Negative strategies while Hume emphasized the positive variant of Governance. This is no surprise given his campaigning for the Forum and
Adams’ abstentionist stance, but it is encouraging that the coding scheme picked up on these differences. Adams and Hume also used the Injustice and Ally Image strategies similar percentages; however if we consider to whom the leaders directed these and other strategies (or the targets), there are important differences.
Before showing the results about the targets, additional information (not all shown in the table above) about the Identity strategies the two leaders used can be compared.
From Table 4.2, it is apparent that Hume more frequently draws boundaries among groups in very inclusive ways when the group references are Northern Ireland and the island of
Ireland (7.9% compared to 3.5% for Adams). Whether he is talking about people in the borders of Northern Ireland or on the entire island, Adams more often uses labels that tend to exclude others. Rarely does he talk about “our whole community” or “all of our people,” as Hume often does. Instead his ingroups tend to exclude Protestants, loyalists, unionists, and the middle class, for example. There are also interesting differences between the leaders concerning references to Europe and the more global references (this is what is not shown in the table). Adams tend to use more global references such as
“oppressed peoples everywhere” and “freedom fighters” (3.5% of 57 in this period, compared to 0 for Hume). On the other hand, Hume uses more of the European or “both islands” level of identity (Britain and Ireland; 2.4% of 81 compared to 0 for Adams). For
Hume, the analysis in the other cases shows that as integration increases, he uses the
European references even more frequently.
159 Nature o f Targets
The picture of which leader is more exclusive in terms of strategies is incomplete without considering the “targets” of the strategies leaders use, because it completes the picture; who is the subject of injustice (Northern Ireland or all working class people or all developing countries)?, who is the enemy threatening the ingroup? Because there are several strategies, the results for this measure are presented in several tables. For each strategy an exclusivity ratio is calculated (see footnotes) and then the ratios for the strategies with outgroups and then ingroups as targets are summed separately. Strategies with outgroups as targets are all variants of the Enemy and Ally Image strategies; strategies with ingroups as targets are the Justice Negative, Governance Positive and
Negative, and Storytelling. Each of the following tables demonstrates that Adams was more exclusive in all of his targets except the total Enemy Image strategies, where the leaders had similar scores. This result fits with what we know about Adams from his use of the Identity strategy above too: he tends to ignore groups in the society of Northern
Ireland when be makes his appeals. Instead, he defines the problem as one of Ireland versus Britain, oppressed versus ruled, working class versus capitalist, for example. The targets are shown below, and the subsequent section summarizes the scores for the exclusivity ratios.
160 Injustice Governance Negative
(More inclusive) Hume Adams Hume Adams (n=147) (n=I37) (n=42) (n=63) Ambiguous 0 0 0 0
World 1.9% 16% 0 12.7%
Europe 0 0 0 3.2%
Britain and Ireland 0 10% 0 1.6%
Britain 1.9% 2% 33.4% 35%
Republic of Ireland 1% 1% 2.4% 11.1%
Northern Ireland 63.5% 3% 45.2% 12.7%
Island of Ireland 31.7% 68% 19.1% 23.8%
Exclusivity ratio “ .5 3.6 .4 .9
“This ratio is a look at the more exclusive references over the more inclusive references. To calculate this ratio, the island of Ireland references are divided by the sum of the Northern Ireland references and the world or global references.
Numbers are percentage of all Justice Negative strategies and Governance Negative strategies (respectively) for which this reference is the target.
Table 4.3: Comparison of Targets (Injustice and Governance Negative Strategies) for Shift 1"
161 Blame Threat Total of Enemy strategies * (More exclusive) Hume Adams Hume Adams Hume Adams (n=29) (n=32) (n=23) (n=33) (n=76) (n=148) Ambiguous 0 0 0 0 0 1%
World 0 21.4% 0 0 0 6.9%
Europe 0 3.6% 0 0 0 1%
Britain and Ireland 0 0 0 0 0 1%
Britain 36% 50% 0 37.9% 24% 36.6%
Republic of Ireland 0 3.6% 0 3.4% 0 9.9%
Northern Ireland 36% 10.7% 57.1% 51.7% 50% 26.7%
Island of Ireland 28% 10.7% 42.9% 6.9% 26% 16.9%
Exclusivity ratio 1.3 3 1.3 7.5 1.9 2
This percentage includes the threat, blame, betrayal, and outgroup image-ambiguous variants.
^ With the Enemy Image strategies, if the leader is arguing that the outgroup is large, this is more exclusive (the more people who are your enemies, the more cohesion of the ingroup is promoted). For this reason, the scale is reversed here and the exclusivity ratio is calculated as follows: the Northern Ireland and world references are summed and divided by the island of Ireland references.
Table 4.4: Comparison of Targets (Blame, Threat, and Combination of Enemy Image Strategies) for Shift 1
162 (More inclusive) Hume Adams (n=132) (n=91) Ambiguous 0 0
World 1% 2.9%
Europe 0 0
Britain and Ireland 0 10.1%
Britain 6.9% 7.2%
Republic of Ireland 1% 0
Northern Ireland 38.6% 8.7%
Island of Ireland 52.5% 71%
Exclusivity ratio 1.3 6.1
“ The target of the Storjielling strategy indicates the group reference that the leader uses when he talks about the "imagined community/’ For example, he may discuss the group or nation as Europe or as Ireland or as Northern Ireland only.
Table 4.5; Comparison of Targets (Storytelling Strategy)"* for Shift 1
163 Shared Goals Total of Ally Governance strategies* Positive (More inclusive) Hume Adams Hume Adams Hume Adams (n=20) (n=36) (n=37) (n=48) (n=101) (n=53) Ambiguous 5% 0 5.9% 0 2% 0
World 15% 10.7% 20.6% 27.5% 1% 1.9%
Europe 5% 0 2.9% 0 3% 0
Britain and Ireland 0 17.9% 0 12.5% 0 1.9%
Britain 25% 7.1% 14.6% 7.5% 13.9% 7.5%
Republic of Ireland 30% 0 17.6% 0 12.9% 3.8%
Northern Ireland 0 7.1% 8.8% 7.5% 33.7% 1.9%
Island of Ireland 20% 57.1% 29.4% 45% 33.7% 83%
Exclusivity ratio 1 3 .9 1.4 .9 7.8
*Tliis percentage includes the goals similar, situation similar, cultnre similar and outgroup image- ambiguous variants.
An inclusive sum is calculated by adding the percentages in the following cells: world, Europe. Britain, and Northern Ireland. These are considered inclusive targets because it means the leader is looking outside the political society for allies and help in governance, and is also looking within the society. The sum of the remaining targets is then divided by the sum of the inclusive to yield a ratio of exclusive to inclusive targets.
Table 4.6: Compaiisoii of Targets (Ally/Goals and Combination of Aliy Strategies, Governance Positive Strategies) for Shift 1
Summary o f Findings
Table 4.7 presents the results of the analysis by showing the exclusivity ratios for
each category and the totals. In both strategies used and targets of strategies, Adams is
164 much more exclusive than Adams in this first period. Therefore, in this case the
hypotheses about the effects of the four contextual variables are shown to be supported by
this analysis.
Exclusivity Exclusivity Exclusivity Total Leader ratio: Nature ratio: Nature of ratio: Nature exclusivity with more of strategies ingroup of outgroup ratio * exclusive targets targets appeals overall Hume 1.6 3.1 6.4 3105 (n= 671) (n=422) (n=113) Adams 2.7 18.4 16.9 11397 XXX (n=650) (n=344) (n=196) This ratio is calculated by first weighting the ratios in the three previous columns by the n for each and then summing the three columns. For example, there were 671 total strategies used by Hume and 422 ingroup targets.
Table 4.7: Summary of Findings, Shift 1
4.3 Second Mobilization Episode: late 1985-1987
4.3.1 Expectations
In the second case of mobilization, the contextual variables all changed, so the hypothesis for each is that the winning leader would be the one with more inclusive strategies. Further, the altered context of changed elite alignment, less repression, increased international involvement, and deepening regional integration is expected to promote the following individual strategies: Governance Positive, Injustice/Justice
Negative, and Storytelling. The next section evaluates the latter, then the exclusivity of leaders’ appeals are compared.
165 4.3.2 Comparing Strategies
As in the previous section, the following table shows the overall distribution of
strategies used by each leader.
Hume Adams (n=1018) (n=870) Identit}" (Northern Ireland 4.3% 1.6% references'^ Outgroup Negative (Enemy Image) 11.6% 24.5%
Justice Negative*** 18.1% 21%
Storytelling*** 15.7% 18.2%
Governance Negative 7.2% 13.9%
Governance Positive*** 24.8% 8%
Outgroup Positive (Ally Image) 10.7% 7.4%
Justice Positive 7.7% 5.4%
Total % = 100.1% 100%
' The "n" here is the total number of strategies and the number given in each cell is the percentage of this total. The strategies in bold are the strategies hypothesized to be used most frequently by the winning leader in this case that is John Hume. The asterisks denote the actual strategies used most frequently by the winning leader. The chance of getting the three matches randomly was only .018, making these results in accord with standard canons of significance.
Because Northern Ireland is the focus of the research and considered to be the society under exam ination. these references are the focus (Adams tends to focus much more on Ireland as the society). This figure excludes references to the leader’s party because this is misleading about the groups in the society the leader is defining.
Due to roimding off of the niunbers, these totals may not equal exactly 100%.
Table 4.8: Comparison o f Strategies, Shift 2“
166 Comparing the predicted top strategies (in bold) to the actual strategies (denoted by asterisks) used by the winning leader, the expectations are fulfilled. Again, the chance of getting the three matches randomly is .018, making these results significant by general standards (.05 or better). One of the greatest contrasts in the strategy use here is that one- quarter of Hume’s strategies are Governance Positive (24.8%) while one-quarter of
Adams’ strategies are Enemy Image (24.5%). Hume’s arguments for negotiation with other groups and leaders and for the change of legislation to help the problems the minority faces indeed seem to have resonated with potential followers, while Adams continued arguments that others primarily pose threats, are to blame, or have betrayed the group did not “make much sense” given the context of increasing cooperation between the
Republic of Ireland and Britain and decreasing repression (for example). At the same time both continue to argue about the unjust situation to similar degrees, though Adams does use this strategy slightly more (21% compared to Hume’s 18.1%). Storytelling was also expected to be used more frequently, and indeed for both leaders this is the third most frequently used strategy. As for the Identity strategy, Hume still appeals to the Northern
Ireland level of identity more than Adams (4.3% of strategies here compared to 1.6%). If we look at references to ingroup identity as a percentage of all references (data not shown in table), 40% of Hume’s 110 ingroup references (excluding references to his political party) are at the Northern Ireland level, compared to 18% of Adams’ 79 ingroup references. By contrast, 78.5% of Adams’ ingroup references were to people on the island of Ireland.
167 To summarize the findings here, as in the first case the two leaders both use two of
three strategies predicted to appear most often according to context, but the winning
leader uses all three: Governance Positive, Justice Negative, and Storytelling.
4.3.3 Assessing the Predictions: The Exclusivitv Index
Nature o f Strategies
The strategy profiles that compare the use of more exclusive to more inclusive
strategies show again that Adams is more exclusive. Also, compared to the previous
period, Adams has become even more exclusive while Hume was even more inclusive
here than in shift 1. This result supports the hypotheses that the four contextual variables favor a leader using more inclusive strategies.
168 More exclusive strategies Hume Adams (n=1043) (n=896) Identity .5% 1.1% (NX, I "very exclusive” references) Outgroup Negative (Enemy Image) 11.3% 23.8%
Justice Negative 17.6% 20.4%
Storytelling 15.3% 17.6%
Governance Negative 7% 13.5%
Total exclusive strategies 51.7% 76.4%
More inclusive strategies Hume Adams (n=1043) (n=896) Identity 6.1% 3.3% (NI, 1 "very inclusive” references) Outgroup Positive (Ally Image) 10.5% 7.1%
Governance Positive 24.2% 7.8%
Justice Positive 7.5% 5.2%
Total inclusive strategies 48.3% 23.4%
Assessment
Hume Adams Exclusivitv ratio ® 1.1 3.3 Change from Shift l'' -.5 +.6 “ This ratio is calculated by dividing the total percentage of exclusive strategies by the total per centage of inclusive strategies. The higher number indicates that the leader used more exclusive strategies than the leader with the lower number.
A positive number here indicates the degree to which the leader's strategy profile became more exclusive, a negative number shows how much it became more inclusive.
Table 4.9: Comparison of Exclusive and Inclusive Strategies for Shift 2
169 The higher degree of exclusivity for Adams is largely due to the fact that he used almost twice as many (in terms of percentages) Enemy Image and Governance Negative appeals.
Indeed, this analysis fits with the facts of the case, that Sinn Fein abhorred the Anglo-Irish
Agreement, that they were being “cornered” by the other parties, and that at this point they did not engage in constructive proposals for future governance. In contrast, Hume used almost three times as many Governance Positive strategies, largely arguing for the
Anglo-Irish Agreement and future negotiations. Hume also used double percentages of more inclusive Identity strategies. The next section looks at the targets.
N ature o f Targets
What the following tables show is that Hume was more inclusive in terms of the targets for most of his strategies. The exception is the Governance Negative strategy; this is explained by the fact that Hume in this period often talked about what Sinn Fein’s actions were doing to negotiations and the building of the new institutions. The group reference for the party and the republican movement is the island of Ireland, which adds to the exclusive score. Another interesting substantive difference is the targets of the Blame and Threat strategies: Adams focuses a great deal of attention on Britain, as he did in
Shift 1. Since the first case, Hume did increase the percentage of times he referred to people in Britain (usually the government) as a threat or to blame. This is the case in part because in this period Hume had gotten even more involved in Westminster and often included several minutes in each speech about the problems for Northern Ireland caused by Thatcher’s monetarist policies. Another interesting aspect of the data is that both
170 Hume and Adams increased by a large percentage the references to groups in the “world’
and “European” references as sharing similar goals. This is a good example of how we
may see the contextual factors showing up in what the leaders say.
Injustice Governance Negative
(More inclusive) Hume Adams Hume Adams (n=I84) (n=183) (n=51) (n=94) Ambiguous .8% .8% 0 0
World 5.6% 9.8% 3.9% 5.3%
Europe 4% .8% 2% 2.1%
Britain and Ireland 6.5% 1.5% 0 8.5%
Britain 0 1.5% 45.1% 24.5%
Republic of Ireland .8% 8.3% 3.9% 24.5%
Northern Ireland 49.2% 11.4% 15.7% 16%
Island of Ireland 33.1% 65.9% 29.4% 19.1%
Exclusivity ratio® .6 3.1 1.5 .9
'‘This ratio is a look at the more exclusive references over the more inclusive references. To calculate this ratio, the island of Ireland references are divided by the stun of the Northern Ireland references and the world or global references.
Numbers are percentage of all Justice Negative strategies and Governance Negative strategies (respectively) for which this reference is the target.
Table 4.10: Compaiison of Targets (Injustice and Governance Negative Strategies) for Shift 2"
171 Blame Threat Total of Enemy strategies* (More exclusive) Hume Adams Hume Adams Hume Adams (n=49) (n=46) (n=19) (n=42) (n=118) (n=213) Ambiguous 2.3% 0 0 0 2.4% 0
World 7% 10.3% 13.3% 7.7% 7.3% 8.7%
Europe 0 0 0 0 0 0
Britain and Ireland 0 7.7% 0 10.3% 1.2% 6.5%
Britain 30.2% 59% 13.3% 33.3% 24.4% 29.6%
Republic of Ireland 0 12.8% 0 12.8% 0 22.4%
Northern Ireland 20.9% 5.1% 13.3% 33.3% 25.6% 23.1%
Island of Ireland 39.5% 5.1% 60% 2.6% 39% 9.4%
Exclusivity ratio** .7 3 .4 15.8 .8 3.4
“This percentage includes the threat, blame, betrayal, and outgroup image-ambiguous variants.
With the Enemy Image strategies, if the leader is arguing that the outgroup is large, this is more exclusive (the more people who are your enemies, the more cohesion of the ingroup is promoted). For this reason, the scale is reversed here and the exclusivity ratio is calculated as follows: the Northern Ireland and world references are summed and divided by the island of Ireland references.
Table 4. II: Comparison of Targets (Blame, Threat, and Combination of Enemy Image Strategies) for Shift 2
172 (More inclusive) Hume Adams (n=160) (n=158) Ambiguous 0 0
World 5.1% 0
Europe 4.3% 0
Britain and Ireland 3.4% 2.7%
Britain 6% 11.8%
Republic of Ireland 0 11.8%
Northern Ireland 30.8% 10%
Island of Ireland 50.4% 63.6%
Exclusivity ratio 1.4 6.4
*The target of the Storytelling strategy indicates the group reference that the leader uses when he talks about the “imagined community/' For example, he may discuss the group or nation as Europe or as Ireland or as Northern Ireland only.
Table 4.12; Comparison of Targets' (Storytelling Strategy), Shift 2
173 Shared Goals Total of Ally Governance strategies* Positive (More inclusive) Hume Adams Hume Adams Hume Adams (n=64) (n=46) (n=109) (n=64) (n=182) (n=48) Ambiguous 0 4.7% 0 3.5% 0 0
World 28.1% 34.9% 23.9% 38.6% 8.8% 0
Europe 24.6% 2.3% 20.7% 1.8% 5.5% 0
Britain and Ireland 0 4.7% 0 3.5% 9.9% 0
Britain 26.3% 4.7% 24% 3.5% 12.6% 6.3%
Republic of Ireland 14% 9.3% 10.9% 8.8% 6% 8.3%
Northern Ireland 3.5% 4.7% 5.5% 7.1% 26.9% 12.5%
Island of Ireland 3.5% 34.9% 15.3% 33.4% 30.2% 72.9%
Exclusivity ratio'’ .2 1 .4 .8 .9 4.3
“This percentage includes the goals similar, situation similar, culture similar and outgroup image- ambiguous variants.
An inclusive sum is calculated by adding the percentages in the following cells: world, Europe, Britain, and Northern Ireland. These are considered inclusive targets because it means the leader is looking outside the political society for allies and help in governance, and is also looking within the society. The sum of the remaining targets is then divided by the sum of the inclusive to yield a ratio of exclusive to inclusive targets.
Table 4.13: Comparison of Targets (Ally/Goals and Combination of Ally Strategies, Governance Positive Strategies) for Shift 2
Summary o f Findings
When we add the exclusivity ratios for each target above, Adams is much more exclusive than Hume, regardless of if he is using the strategies where the targets tend to be ingroups or outgroups. Based on these results and the results from the previous section.
174 we see that Hume is the leader with the more inclusive appeals overall. Further, in this period when Adams’ party lost support, several of the measures show that his appeals moved away fro m the kind of appeal expected to be more successful: as the context favored inclusive definitions o f the situation, Adams became more exclusive. Based on the analysis, the hypotheses relating the each contextual variable to more inclusive strategies are shown to be valid.
Exclusivity Exclusivity Exclusivity Total Leader with Change ratio: ratio: ratio: exclusivity more since last Nature of Nature of Nature of ratio * exclusive period strategies ingroup outgroup appeals targets targets overall Hume 1.1 4.4 2.5 4226 +1121 (n=I018) (n=577) (n=227) Adams 3.3 14.7 24 16,619 XXX +5222 (n=870) (n=483) (n=277) This ratio is calculated by first weighting the ratios in the three previous columns by the n for each and then summing the three colurrms.
Table 4.14: Summary of Findings, Shifi; 2
4.4 Third Mobilization Episode: 1996
4.4.1 Expectations
The analysis of the third shift shows a more complicated picture than the first two, which both seemed to fit so well with expectations. Indeed, because shifts 3 and 4 do not fit perfectly with expectations, it is hoped that comparing all four cases here and the
Zimbabwe cases will lead to a better understanding o f how different dimensions of context interact. In this third case, the change in stability of elite alignment variable was not
175 expected to be significant since there was no change in the expectation that this variable
favors inclusive strategies. It was hypothesized that the repression variable, the regional
integration variable, and the international involvement variable all promote more inclusive
strategies. The caveat discussed in Chapter 3 was that international involvement may have
favored Adams because of the substantial change in the way the United States, Britain, the
Republic of Ireland, and the European Union treated Adams (for example, the United
States granting him a veto and the UK and Ireland lifting their broadcasting bans). In this case there were four individual strategies expected to be favored by the changed international context; Governance Positive, Ally Image, Injustice, and Storytelling. The next section evaluates the latter.
4.4.2 Comparing Strategies
The change in the distribution of strategies used by both leaders with the changed context is striking. Whereas Adams had consistently used Enemy Image strategies around
20-25% of the total strategies he uses, this drops to 9.7% (though still higher than Hume’s
1.9%). For the first time, the most frequently used strategy for Adams is Governance
Positive at about one-third of all strategies he uses. Hume also increased the number of times he appealed to audiences with the Governance Positive strategy—to 43%. In this era of international talks on the fate of the minority in Northern Ireland, it appears that both leaders are compelled to emphasize negotiations with others and negotiation of political structures even more than their standard injustice appeals, which fall to around
14% and 17%. This may reflect the fact that life has improved for the nationalist
176 community over the years and that the two leaders are (at least to some degree) trying to
demobilize people from the struggle and mobilize them around the idea of talking. In
terms of Identity, Hume still uses a greater percentage of references to people in Northern
Ireland as the ingroup. Still, other important differences from previous periods in the
entire Identity strategy (data not shown here) are apparent. Hume continued to increase
his emphasis of the European and more global identity of the ingroup, while Adams still
does not discuss that his audience shares the European level of identity and he uses no
global references here.
In evaluating the expectations for individual strategy use, the predictions do not fare well in that the winning and losing leaders have the same top three strategies
(Governance Positive, Storytelling, and Justice Negative). These three fit with expectations, but for neither leader does the fourth most frequently used strategy match the prediction—the Ally strategy was not as important as expected here. The results from the exact test show that the significance level here is only .24. On the other hand, this case shows that in some contexts, leaders may use similar strategies.
177 Hume Adams (n=269“) (n=677) Identity (Northern Ireland 4.1% .5% references*’ ) Outgroup Negative (Enemy Image) 1.9% 9.7%
Justice Negative*** 16% 13.9%
Storytelling*** 16% 22.7%
Governance Negative*** 2.6% 10.2%
Governance Positive*** 43.1% 32.9%
Outgroup Positive (Ally Image) 7% 6.8%
Justice Positive 9.3% 3.2%
Total %= 100% 99.9%
The "n" here is the total number of strategies and the number given in each cell is the percentage of this total. The strategies in bold are the strategies hypothesized to be used most frequently by the wiiming leaden in this case that is Geny Adams. The asterisks denote the actual strategies used most frequently by the winning leader. Here only three of the four predicted strategies were used by the winning leader. According to the exact test, the significance level here is .24.
Because Northern Ireland is the focus of the research and considered to be the society under exam ination. these references are the focus (Adams tends to focus much more on Ireland as the society). This figure excludes references to the leader’s party because this is misleading about the groups in the societ)' the leader is defining.
Due to rounding off of the numbers, these totals may not equal exactly 100%.
Table 4.15: Comparison of Strategies, Shift 3
4.4.3 Assessing the Predictions: The Exclusivitv Index
Nature o f Strategies
Though the leaders appear to be more similar in the individual strategies they employ, when the strategies are separated into more exclusive and more inclusive types
178 there are still important differences. The sum of the exclusivity ratios shows that Adams is once again more exclusive than Hume. These results seem to invalidate the hypotheses about the role of context, since all four contextual variables predicted that circumstances favored the more inclusive leader (and Adams is said to be the “winning leader”). In fact, a closer look at the changes over time shows that Adams moved a long way toward the inclusive end of the scale; instead of simply showing the hypotheses to be invalid, the data may indicate an interesting though unexpected relationship. The anomalies in this and any other cases will be discussed in the conclusion.
179 More exclusive strategies Hume Adams (n=278) (n=708) Identity 0 .6% (NI, I "very exclusive” references) Outgroup Negative (Enemy Image) 1.8% 9.3%
Justice Negative 15.5% 13.3%
Storytelling 15.5% 21.8%
Governance Negative 2.5% 9.7%
Total exclusive strategies 35.3% 54.7%
More inclusive strategies Hume Adams (n=278) (n=708) Identity 7.2% 4.2% (NI. I "very inclusive” references) Outgroup Positive (Ally Image) 6.8% 6.5%
Governance Positive 41.7% 31.5%
Justice Positive 9% 3.1%
Total inclusive strategies 64.7% 45.3%
Assessment
Hume Adams Exclusivity ratio® .5 1.2 Change since Shift 2^ -.6 -2.1 "*This ratio is calculated by dividing the total percentage of exclusive strategies by the total per centage of inclusive strategies. The higher number indicates that the leader used more exclusive strategies than the leader with the lower number.
A positive number here indicates the degree to which the leader’s strategy profile became more exclusive, a negative number shows how much it became more inclusive.
Table 4.16: Comparison of Exclusive and Inclusive Strategies for Shift 3
180 Overall, Hume is again more inclusive but both leaders moved toward greater inclusivity; Adams also moved much further in this direction. Looking at a few of the more remarkable changes (some of which were noted in the previous section), Adams used far fewer Enemy Image strategies than in the previous period (9.3% compared to
23.8%), as did Hume (11.3% to 1.8%). Another change is the Governance Positive strategy, as discussed above: Adams went from 7.8% to 31.5% (Hume also moved from
24.2% to 41.7%). There were only small reductions in the Governance Negative strategies, perhaps because both leaders still criticized some aspects o f government and negotiations—past and present—while encouraging and suggesting others. Another major reason Adams is more exclusive is that he uses more Storytelling strategies than Hume.
This makes sense from the perspective of his challenges: he has to convince his followers that negotiations are consistent with the goals for which so many freedom fighters have been martyred and embedding his current actions in this mythology may help to do so.
The continued practice of holding the annual conference in Dublin as a symbol of their past and destiny adds to this strategy.
Finally, important changes in the Identity strategy for the inclusive category are seen in both leaders’ appeals. They use between 4-7% of all strategies to emphasize more inclusive group boundaries (delineations) within the references for Northern Ireland and
Ireland.
Nature o f Targets
The following tables show that Adams is more exclusive than Hume for all targets except for the Enemy Image/Blame strategy. For the latter, Hume does not use any of the
181 targets considered more inclusive, so his exclusivity ratio (exclusive divided by inclusive)
is scored 1. The next section discusses the sum of all the ratios and the overall evaluation.
Injustice Governance Negative
(More inclusive) Hume Adams Hume Adams (n=43) (n=94) (n=5) (n=47) Ambiguous 0 1.4% 0 0
World 2.9% 2.8% 40% 8.5%
Europe 5.9% 0 0 0
Britain and Ireland 0 4.2% 0 4.3%
Britain 0 1.4% 20% 46.8%
Republic of Ireland 0 0 0 8.5%
Northern Ireland 50% 2.8% 20% 10.6%
Island of Ireland 41.2% 87.5% 20% 21.3%
Exclusivity ratio“ .8 15.6 .3 1.1
“This ratio is a look at the more exclusive references over the more inclusive references. To calculate this ratio, the island of Ireland references are divided by the sum of the Northern Ireland references and the world or global references.
^ Numbers are percentage of all Justice Negative strategies and Governance Negative strategies (respectively) for which this reference is the target.
Table 4.17: Comparison of Targets (Injustice and Governance Negative Strategies) for Shifts"
1 8 2 Blame Threat Total of Enemy strategies" (More exclusive) Hume Adams Hume Adams Hume Adams (n=2) (n=27) (n=0) (n=8) (n=5) (n=66) Ambiguous 0 0 0 0 0 0
World 0 0 0 0 0 0
Europe 0 0 0 0 0 0
Britain and Ireland Û 0 0 12.5% 0 8.3%
Britain 50% 44.4% 0 25% 50% 54.2%
Repubhc of Ireland 0 0 0 0 0 10.4%
Northern Ireland 50% 22.2% 0 50% 25% 16.7%
Island of Ireland 0 33.3% 0 12.5% 25% 10.4%
Exclusivity ratio*’ 1 .7 — — I 1.6
"This percentage includes the threat, blame, betrayal, and outgroup image-ambiguous variants.
With the Enemy Image strategies, if the leader is arguing that the outgroup is large, this is more exclusive (the more people who are your enemies, the more cohesion of the ingroup is promoted). For this reason, the scale is reversed here and the exclusivity ratio is calculated as follows: the Northern Ireland and world references are siunmed and divided bv the island of Ireland references.
Table 4.18: Comparison of Targets (Blame, Threat, and Combination of Enemy Image Strategies) for Shift 3
For the Storytelling strategy targets (Table 4.19), there are several differences
from the previous period. Adams uses the strategy more frequently overall, surpassing
Hume for the first time (though they were almost equal in period 2). Also, Hume shows a move toward Adams in the targets of the strategy. In this period, he shifted slightly from
183 Northern Ireland to the island of Ireland. Also, Hume discussed the Irish diaspora and
included them in his references to the Irish nation (these are many of the “world” targets
seen below).
(More inclusive) Hume Adams (n=43) (n=154) Ambiguous 0 1%
World 16.7% 1%
Europe 3.3% 0
Britain and Ireland 0 1%
Britain 0 12%
Republic of Ireland 0 1%
Northern Ireland 36.7% 5%
Island of Ireland 43.3% 79%
Exclusivity ratio .8 13.2
“The target of the Storjlelling strategy indicates the group reference that the leader uses when he talks about the "imagined community." For example, he may discuss the group or nation as Europe or as Ireland or as Northern Ireland only.
Table 4.19: Comparison of Targets' (Storytelling Strategy), Shift 3
For the Ally strategy where the leader emphasizes shared goals with other actors, both leaders put more emphasis on the Northern Ireland and island of Ireland references.
This is probably part of the effort to emphasize the need to join in negotiations with other actors in these shared areas, since the targets were often the unionist community and the other nationalist party. These differences are large: for “Northern Ireland,” Hume goes
184 from 3.5 to 30%, though Adams drops from 4.7 to 0. For the “island of Ireland,” Hume goes from 3.5 to 20% and Adams moves from 34.9 to 71.9%. Thus, we still see that overall, Hume’s inclusion of the Northern Ireland dimension makes his score more inclusive than Adams.
The total o f Ally strategies shows a similar picture though the “world” references are interesting: 47.1% of Hume’s total Ally strategies are at the global level, an increase from 23.9%; Adams moved from 38.6 down to 26.5%. Despite the decrease for Adams, these numbers show that both emphasize alhes in the global arena, connecting their plight with others, calling for help, and thanking for help in the recent past (the United States and Irish America were common targets). For the Governance Positive strategy, while
Adams used this with greater frequency, Hume’s much greater emphasis on groups within
Northern Ireland means that his score shows more inclusive targets.
185 Shared Goals Total of Ally Governance strategies* Positive (More inclusive) Hume Adams Hume Adams Hume Adams (n=ll) (n=34) (n=19) (n=46) (n=61) (n=135)
Ambiguous 0 0 0 0 1 .6 % .7%
World 2 0 % 25% 47.1% 25.6% 16.4% 6.7%
Europe 30% 0 17.6% 0 6 .6 % .7%
Britain and Ireland 0 0 0 0 1 .6 % 7.4%
Britain 0 3.1% 0 2 .6 % 8 .2 % 9.6%
Republic of Ireland 0 0 0 0 9.8% 5.9%
Northern Ireland 30% 0 17.7% 2 .6 % 31.1% 3%
Island of Ireland 2 0 % 71.9% 17.6% 69.2% 24.6% 65.9%
Exclusivity ratio*’ .3 2 . 6 . 2 2 . 2 . 6 3.8
“ This percentage includes the goals similar, situation similar, culture similar and outgroup image- ambiguous variants.
*’ An inclusive sum is calculated by adding the percentages in the following cells: world, Europe, Britain, and Northern Ireland. These are considered inclusive targets because it means the leader is looking outside the political society for allies and help in governance, and is also looking within the society. The sum of the remaining targets is then divided by the sum of the inclusive to yield a ratio of exclusive to inclusive targets.
Table 4.20: Comparison of Targets (Ally/Goals and Combination of Ally Strategies, Governance Positive Strategies) for Shift 3
Summary of Findings
Table 4.21 shows that Adams has a much higher score on the exclusivity index.
The predictions of the contextual variables were therefore not supported in this case. The predictions were that the more inclusive leader would win. Adams was judged to have been the major victor, but his strategies were more exclusive than Hume’s. On most
186 scores, both Adams and Hume became more inclusive since shift 2, however. Another
important point about this case was the fact that the international variable may have favored Adams, even while promoting more inclusive strategies. This finding hints that while the international community’s involvement may not necessarily promote the winning of the more inclusive leader overall, it may have the effect of pushing the more exclusive leader to be more inclusive compared to his own strategies in the previous period. The conclusion of this chapter and the dissertation will return to the unexpected outcome in relation to all four cases.
Exclusivity Exclusmty Exclusivity Total Leader Change ratio: ratio: ratio: exclusivity with since last Nature of Nature of Nature of ratio^ more period strategies ingroup outgroup exclusive targets targets appeals overall Hume .5 2.5 2.5 574.5 -3652 (n=269) (n=152) (n=24)
Adams 1 . 2 33.7 10.9 16,524 XXX -95
(n=677) (n=430) (n=I1 2 ) This ratio is calculated by first weighting the ratios in the three previous columns by the n for each and then summing the three columns.
Table 4.21; Summary of Findings, Shift 3
187 4.5 Fourth Mobilization Episode
4.5.1 Expectations
In the final Northern Ireland case, there were important changes in the context so
that two of the political opportunity variables which had predicted inclusive strategies in
case 3 now point to the expectation that more exclusive strategies will be used by the
winner. The changes expected to promote exclusive appeals were in the stabihty of elite
alignment and in repression. The international involvement variable still is expected to
promote inclusive ideas though is somewhat favorable to Adams; there was no change in
regional integration. As far as individual strategies. Enemy Image, Governance Positive,
Injustice, and Storytelling are expected to be those most frequently used.
Though of course it is not possible to come to any conclusions by looking at this
one case, the fact that the two variables that changed (change in stability of elite alignment
and repression) might have affected the outcome so strongly is an interesting insight that
should be explored in additional cases. In turn, the international climate may have affected
the “strength” of the repression variable. At a time when all eyes and the resources of the
United States were turned on Northern Ireland, the use of plastic bullets and the images of
the Royal Ulster Constabulary and British army bearing the Catholic protestors may have
been more important. Indeed, the repression levels seen in the marching season and the
perception that Britain was so blatantly taking the side of unionists— and the decision of
Adams and Sinn Fein to focus on these factors the way they did —may have been enough
to outweigh the international variable favoring inclusive strategies.
188 Consideration of the mobilizing resources may prove a “tie breaker” in this case,
only because—unlike the earlier years—there did appear to be significant differences
between the leaders in their ability to get their message to the people. Still, the model
considered here does not include the mobilizing resources as an independent variable. The
results of the content analysis for this case, and then comparison across all four cases (see below), are expected to help generate more specific hypotheses on how the multiple dimensions of opportunities mediate each other.
4.5.2 Comparing Strategies
In looking at the strategy usage for this final period, the winning leader’s (Adams) top four strategies are the top four hypothesized; the significance level using the exact test is .014. There were several differences from the previous period and interesting similarities between the leaders in this fourth case. For example, both leaders use more
Enemy Image strategies; for example, Hume’s use o f this strategy jumps fi'om 1.9% to
10.7%. Still, the Enemy Image strategy does not rank in the top four of Hume’s strategies as it does for Adams. Importantly, despite the negative context, both leaders continued to use the Governance Positive strategy about one-third of the time; for Hume this is a drop of about 10%.
The other two strategies expected to appear most frequently in the winning leaders’ appeals were Justice Negative and Storytelling. Because both leaders use these two strategies as the second (Justice Negative) and third (Storytelling) most frequent, the
189 predictions are weakened slightly. However, Adams does use a higher percentage of
Storytelling (20.6%); and if the focus is on which leader has all four predicted strategies,
the results are quite strong here.
Hume Adams (n=542*) (n=504)
Identity (Northern Ireland 4.4% 1 .6 % references*')
Outgroup Negative (Enemy 10.7% 1 1 . 1 % Image)*** Justice Negative*** 13.1% 14.3%
Storytelling*** 13.5% 2 0 .6 %
Governance Negative 1 1 .8 % 9.5%
Governance Positive*** 31.4% 30.4%
Outgroup Positive (Ally Image) 6 . 1 % 5.2%
Justice Positive 9% 7.3%
Total % " 1 0 0 % 1 0 0 %
The "n" here is the total number of strategies and the number given in each cell is the percentage of this total. The strategies in bold are the strategies hypothesized to be used most frequently by the winning leader in this case that is Gerry Adams. The asterisks denote the actual strategies used most frequently by the wirming leader. Using the exact test, the chances of the leader using four of the four predicted strategies randomly is only .014, making these results significant.
Because Northern Ireland is the focus of the research and considered to be the society under exam ination, these references are the focus (Adams tends to focus much more on Ireland as the society). This figure excludes references to the leader's party because this is misleading about the groups in the societj’ the leader is defining.
Due to rounding off of the numbers, these totals may not equal exactly 100%.
Table 4.22: Comparison of Strategies, Shift 4
190 4.5.3 Assessing the Predictions: The Exclusivity Index
Nature o f Strategies
When the strategies are divided into the more and less exclusive types, Adams shows a higher ratio o f exclusive to inclusive strategies. Further, Adams score was similar to shift 3 but Hume became more exclusive. Overall, it appears that Hume’s strategy profile began to look much more like that of Adams in this period. The shift in Adams’ approach toward Hume occurred in the last case, when Adams showed the large jump in the use of the Governance Positive strategies.
191 More exclusive strategies Hume Adams (n=540) (n=519)
Identity 1.3% .8 % (NI. I •'very exclusive” references)
Outgroup Negative (Enemy Image) 10.7% 1 0 .8 %
Justice Negative 13.1% 13.9%
Storytelling 13.5% 2 0 %
Governance Negathe 11.9% 9.2%
Total exclusive strategies 50.5% 54.7%
More inclusive strategies Hume Adams (n=540) (n=519)
Identity 2 .8 % 3.7% (Nl, I "very inclusive” references)
Outgroup Positive (Ally Image) 6 . 1 % 5%
Governance Positive 31.5% 29.5%
Justice Positive 9.1% 7.1%
Total inclusive strategies 49.5% 45.3%
Assessment
Hume Adams
Exclusivitj’ ratio^ 1 . 0 1 . 2
Change from Shift 3*" +.5 0 ® This ratio is calculated by dividing the total percentage of exclusive strategies by the total per centage of inclusive strategies. The higher number indicates that the leader used more exclusive strategies than the leader with the lower number.
A positive number here indicates the degree to which the leader’s strategy profile became more exclusive, a negative number shows how much it became more inclusive.
Table 4.23: Comparison of Exclusive and Inclusive Strategies for Shift 4
192 When the Identity strategies employed by the leaders in this case are considered, the
leaders show more movement toward each other over time (some of this data is not shown
here). For example, when the leaders described or made reference to groups within the
boundaries of Northern Ireland or the island of Ireland, both leaders moved toward the
other’s way of describing the situation in previous periods. Substantively, both leaders
discussed the nationahst community and Catholics in Northern Ireland a great deal, and
with the island reference often referred to the Six Counties, the North, or nationaUsts in
the North. A number of references still were to “the Irish people,” a very inclusive
reference often made with the Storytelling strategy, as they spoke of the right to self-
determination (see targets, below).
Nature o f Targets
The results for this section were almost uniform: Adams was more exclusive than
Hume in terms of the targets of the strategies, except for Governance Negative where the scores were .7 and .8. There were still important changes: Hume and Adams both used fewer “island of Ireland” targets and more “Northern Ireland” targets when they used the
Governance Negative strategy. For the former (island), Hume went fi'om 41.2 to 23.9%, and Adams went from 87.5 to 70.4%. With regard to the Northern Ireland targets,
Adams showed the most important change: from 2.8 to 24.1. A remarkable change for the Governance Negative targets is the increase in Hume and Adam’s targeting of both global actors and the Northern Ireland groups (see Table 4.24). Also, while Hume still emphasized more than Adams (54.3%) that groups in Northern Ireland were targets of
Injustice (Justice Negative), in this period Adams increasingly used this as a target.
193 Usually it was Catholics or nationalists that he described, but he also talked about the unfortunate situation of the unionist people. In reference to the latter, he tended to say they are victims of pohticians. This is visible in the table below.
Injustice Governance Negative
(More inclusive) Hume Adams Hume Adams (n=7I) (n=72) (n=46) (n=37)
Ambiguous 0 0 0 0
World 4.3% 1.9% 2 .2 % 0
Europe 4.3% 0 6.5% 0
Britain and Ireland 8.7% 1.9% 0 0
Britain 2 .2 % 0 21.7% 37.8%
Republic of Ireland 2 .2 % 1.9% 4.3% 5.4%
Northern Ireland 54.3% 24.1% 37% 32.4%
Island of Ireland 23.9% 70.4% 28.3% 24.3%
Exclusivity ratio* .4 2.7 .7 . 8
“This ratio is a look at the more exclusive references over the more inclusive references. To calculate this ratio, the island of Ireland references are divided by the stun of the Northern Ireland references and the world or global references.
Numbers are percentage of all Justice Negative strategies and Governance Negative strategies (respectively) for which this reference is the target.
Table 4.24: Comparison of Targets (Injustice and Governance Negative Strategies) for Shift 4"
194 Blame Threat Total of Enemy strategies" (More exclusive) Hume Adams Hume Adams Hume Adams
(n=2 2 ) (n=2 0 ) (n=l2 ) (n=9) (n=58) (n=56)
Ambiguous 0 0 0 0 0 0
World 0 0 0 0 0 0
Europe 0 0 12.5% 0 2 .6 % 0
Britain and Ireland 0 0 0 0 0 1 0 .8 %
Britain 44.4% 62.5% 0 42.9% 26.3% 40.5%
Republic of Ireland 0 0 0 0 0 5.4%
Northern Ireland 2 2 .2 % 37.5% 12.5% 57.1% 23.7% 40.5%
Island of Ireland 33.3% 0 75% 0 47.4% 2.7%
Exclusivity ratio" .7 1 . 2 1 .5 15
"This percentage includes the threat, blame, betrayal, and outgroup image-ambiguous variants.
With the Enemy image strategies, if the leader is arguing that the outgroup is large, this is more exclusive (the more people who are your enemies, the more cohesion of the ingroup is promoted). For this reason, the scale is reversed here and the exclusivity ratio is calculated as follows: the Northern Ireland and world references are summed and divided by the island of Ireland references.
Table 4.25: Comparison of Targets (Blame, Threat, and Combination of Enemy Image Strategies) for Shift 4
195 (More inclusive) Hume Adams (n=73) (n=I04)
Ambiguous 0 0
World 3.9 2.9
Europe 7.8 0
Britain and Ireland 3.9 2.9
Britain 3.9 8 . 8
Republic of Ireland 5.9 0
Northern Ireland 27.5 2 0 . 6
Island of Ireland 47.1 64.7
Exclusivity ratio 1.5 2 . 8
“The target of the Storytelling strategy indicates the group reference that the leader uses when he talks about the ‘"imagined community.” For example, he may discuss the group or nation as Europe or as Ireland or as Northern Ireland only.
Table 4.26; Comparison of Targets^ (Storytelling Strategy), Shift 4
Finally, looking at the targets of the Ally Image and Governance Positive strategies, several changes appear. Hume is again more inclusive than Adams. As in the previous period, he talks more about allies in the world and Europe than Adams, and increased this target more than Adams did from period three. Indeed, substantively, both continued to call on the United States, but Hume did so more frequently and then also included others in Europe. Another aspect that should be pointed out is that Adams targeted allies in the island o f Ireland much less frequently than in 1996: 31.6% compared to 69.2%. This is probably because in 1996 he was reaching out to the “other tradition in
196 Ireland,” whereas in 1997, they became a target of the more negative strategies (such as
Blame and Governance Negative). In other words, the 1997 campaign showed a setback for the more conciliatory “definition of the situation” that both Adams and Hume put forth a year earlier in terms of both the frequency of more inclusive strategies and the inclusivity of the targets of these strategies.
Shared Goals Total of Ally Governance strategies' Positive (More inclusive) Hume Adams Hume Adams Hume Adams (n=26) (n=14) (n=33) (n=26) (n=107) (n=95)
Ambiguous 0 0 0 0 0 0
World 37.5% 16.7% 35.7% 26.3% 7.5% 6.3%
Europe 29.2% 0 25% 0 13.1% 1 . 1 %
Britain and Ireland 0 8.3% 0 5.3% 0 2 . 1 %
Britain 16.7% 16.7% 14.3% 15.8% 7.5% 10.5%
Republic of Ireland 4.2% 0 3.6% 0 5.6% 5.3%
Northern Ireland 4.2% 16.7% 7.2% 2 1 . 1 % 34.6% 18.9%
Island of Ireland 8.3% 41.7% 14.2% 31.6% 31.8% 55.8%
Exclusivity ratio*’ . 1 1 . 2 . 6 . 6 1.7
" This percentage includes the goals similar, situation similar, culture similar and outgroup image- ambiguous variants.
An inclusive sum is calculated by adding the percentages in the following cells: world, Europe, Britain, and Northern Ireland. These are considered inclusive targets because it means the leader is looking outside the political society for allies and help in governance, and is also looking witliin the society. The stun of the remaining targets is then divided by the stun of the inclusive to yield a ratio of exclusive to inclusive targets.
Table 4.27; Comparison of Targets (Ally/Goals and Combination o f Ally Strategies, Governance Positive Strategies) for Shift 4
197 Summary o f Findings
Table 4.28 shows the weighted totals o f the exclusivity ratios. When we look at the first column, the two leaders appear very close in their degree of exclusivity.
However, when the targets of the strategies are considered, Adams is much more exclusive for both kinds o f targets and his score overall shows greater exclusivity.
Looking at the change since 1996 shows that Hume did become more exclusive; at the same time Adams became dramatically more inclusive.
Exclusivity Exclusivit) Exclusivit) Total Leader Change ratio: ratio: ratio: exclusivity with since last Nature of Nature of Nature of ratio' more period strategies ingroup outgroup exclusive targets targets appeals overall
Hume 1 3.2 1.7 1967 +1392.5 (n=542) (n=297) (n=91)
Adams 1 . 2 8 18.6 4594 XXX -11,930 (n=504) (n=308) (n=82) This ratio is calculated by first weighting the ratios in the three previous columns by the n for each and then summing the three columns.
Table 4.28; Summary of Findings, Shift 4
The overall prediction for the contextual variables was that the setting favored more exclusive strategies, though it was also stated that the international involvement may have favored Adams (see Chapter 3). Adams was indeed ±e winning leader, thus the hypotheses about the effects are supported in this case.
198 A Note on Analysis o f the Storytelling Strategy
As discussed in Chapter 2, the coding for the Storytelling strategy is not mutually
exclusive with regard to the other strategy models. Instead, it is considered a “mode” that
a leader employs when he uses one of the other strategies. For this reason, a potential
objection to the analyses presented here and in the next chapter is that the frequency
results (the first table presented for each case above) for this strategy overstate the usage
of Storytelling. To address this issue, a separate analysis was conducted that omitted the
Storytelling strategies; this is explained in Appendix C. The important point to note about
the findings from this second analysis is that it strengthens the results presented in
Chapters 4 and 6; the actual frequencies for the other expected strategies are increased.
4.6 Looking Over Time: Important Implications
One of the findings is that leaders used a mix of the strategies in all periods, even
though in some cases the research supports the hypotheses that the context would afreet
the relative frequencies of the various types (when repression was high. Threat was used
more often; when negotiations neared, the Governance strategies increased). This is an
important insight because it shows that contextual factors—political opportunities and the
lack of them— do not necessarily afreet leaders independently as most theories imply.
Because the leaders often responded (in terms of strategies used) to similar contextual
circumstances in different ways, it can be argued that it matters how leaders choose to highlight particular situational factors. After looking at the Zimbabwe cases, the conclusion will return to this finding.
199 Looking at the data over time, a trend is that the leaders seemed to converge somewhat in their strategy profiles, and that both move toward the other. As we have seen, there are differences. Specifically, Hume continues to recognize that unionists do feel loyalty to a different identity but that the European Union model can be applied so that the state identity is not the terminal identity—the highest allegiance. Adams, in contrast, tends to speak as though the unionists are deluded by the British connection, and as soon as it is severed the Irish Protestant tradition will become aware o f their common roots with Catholics. In this way, Hume is still more inclusive with regard to the current society. Still, both now speak of national self-determination and argue for negotiation with the Protestant tradition; in terms of strategies it can be argued that Adams moved toward Hume. The table below compares the changes between shifts by leader and prediction; it shows that, in case 3, the leaders move toward each other. Adams becomes more inclusive while Hume becomes dramatically more exclusive.
200 Shift 1 Shift 2 Shifts Shift 4 Hume 3105 4226 574.5 1967
total/% change (36% more (8 6 % more (242% more exclusive) inclusive) exclusive) Adams 11,397 16,619 16,524 4594
total/% change (46% more (.6 % more (72% exclusive) inclusive) more inclusive) Expected winning Exclusive Inclusive Inclusive Exclusive strategy type
Winner/type of Adams Hume Adams Adams strategy/movement more exclusive more inclusive more exclusive more exclusive from previous moved toward moved slightly moved away from period exclusive end of toward inclusive predicted winning continuum end of continuum direction Were expectations Yes Yes No Yes supported?
“ The higher numbers indicate that the leader showed greater exclusivity in his appeals. For the percent change, the old index is subtracted from the old and this number is divided by the old index.
Table 4.29: Compaiison of Change Between Shifts (Totals of Exclusivity Index ')
When the group references are examined over time, Hume does use more of the island references in recent periods, and also within this the “very inclusive” delineation. In this way he moves toward Adams, whereas Adams over time increasingly uses the
“Northern Ireland” references more in line with Hume. Initially, it was expected that in a situation that moved toward negotiations, the more exclusive leader would have had to move toward the more inclusive leader. In the Northern Ireland cases, we see that the inclusive leader had to move as well, but in the end both leaders are positing multiple levels of identity (especially more and less inclusive groups within Northern Ireland, the island of Ireland, and the world).
201 This finding that both leaders change to be more like the other is very important to
literature on ethnic systems. As was noted in Chapter 2, work on the role of electoral
institutions in divided societies (especially Horowitz, 1985; and Mitchell, 1995)
emphasizes that more moderate leaders are always easily outflanked by more extreme
nationalist appeals. Further, Northern Ireland is usually held up as a case exemplary of
this phenomenon. While this outcome may happen in any given case, the study of the
dynamics of Irish nationalism in Northern Ireland shows that a more moderate leader, and
particular international circumstances, can affect the discourse in significant ways over
time. Ultimately, this finding shows the importance of considering the battle as it plays
out in the rhetoric of nationalist leaders and is affected significantly by domestic and
international contexts, something that institutional work does not explore in any detail
(beyond domestic political institutions).
Finally, one other aspect of the group references stands out: over time Hume has
increasingly promoted the idea that the European identity is something all parties
(Catholics, Protestants, British, Irish) share and that thinking at this level (1) makes more
sense in a globalizing world, and (2) is the only way to overcome the conflict, which the
experiment of the European Union as a whole “proves.” By focusing on the importance
of context in affecting how successful more inclusive strategies like this are, this study
suggests that whether or not people “buy” his picture that they have multiple identities
may depend on the orientation the European Union takes toward Hume and the role the
EU plays in the lives o f the people Hume is trying to convince.
202 Next a summary table is presented that includes all four cases. It is necessary to consider all four cases in order to be more specific about how contextual variables may have interacted and also perhaps how some aspects of context may have been more important than others in affecting the “battle for the hearts and minds” o f the perceived nationalist electorate. Indeed, along with the lack of clear indicators for the types of
“political opportunities,” and “framing,” previous research has not begun to study empirically the conditions under which various opportunities and resources may be more influential than others. Further, it seems to make sense to think not only about the
“opportunities” that exist but that leaders may m ake opportunities by highlighting certain aspects of context: who are our allies, what is repression, what should the international community be doing? This section asserts several “conclusions” from the Irish cases in the form of hypotheses to be tested in future research. The insight from these cases will also be considered in the final evaluation o f the Zimbabwe cases, and the two regions will be compared to each other as the conclusion revisits some of the ideas raised below.
203 Case 1: Case 2: Case 3: Cased: Expectations/ Expectations/ Expectations/ Expectations/ Outcome Outcome Outcome Outcome Change in stability Exclusive/ Inclusive/ no change Exclusive/ of elite alignment Exclusive Inclusive Exclusive
Repression Exclusive/ Inclusive/ Inclusive/ Exclusive/ Exclusive Inclusive Exclusive Exclusive
International Exclusive/ Inclusive/ Inclusive, Inclusive, intervention Exclusive Inclusive favoring Adams/ favoring Adams/ Exclusive Exclusive Regional not significant Inclusive/ Inclusive/ not significant integration Inclusive Exclusive
“ Where the font is in bold, it denotes cases where the prediction and the finding matched.
Table 4.30: Summary o f Northern Ireland Cases'
This table shows that the predictions for the first two cases were straightforward, with all the contextual variables predicting the success of the leader with the most
exclusive strategies (case 1) and then the most inclusive strategies (case 2). In cases 1 and 2, the hypotheses were supported. The picture gets more complex with case 3. In case 3, it was expected that inclusive strategies to win. The winner, Adams, was more exclusive. In case 3, the role of mobilizing resources (external to the contextual model used here) may have made the difference. Indeed, in case 3 Adams and Hume were very different in elements of their funding/patronage, party/campaign organization, and media access/coverage. It may be that the leader with the most and better resources to “get the message out” may be more successful when the differences are so pronounced. This variable may also affect the others: with the international opinion, Adams’ international
204 campaign against renewed repression in shift 4 may have led to the perception among the electorate that Adams was truly out there trying to redress the most urgent grievances of the nationalist community. Future research with cases selected specifically for variation on mobilizing resources will have to investigate these issues; in the cases selected for this study it is not possible to come to any definitive conclusions.
The next chapter presents the research on the Rhodesia/Zimbabwe cases, exploring how the hypotheses in Chapter 2 fared, and evaluating the results in hght of the conclusions from the Northern Ireland cases.
205 CHAPTERS
RHODESIA/ZIMBABWE: CASES AND CONTEXTS
5.1 Introduction
This chapter discusses the mobilization episodes in Rhodesia/Zimbabwe and is organized as follows. The cases are described in terms of the case selection criteria, and evidence is presented to show which leader is considered in this study to be the beneficiary of each mobilization shift. Second, evidence is presented about the contextual variables; change in stability of elite alignment; state repression; international involvement/mediation; and regional integration. Finally, based on this evidence, the propositions developed in
Chapter 2 about the effects of context on the success of leaders’ mobihzation strategies are applied to give case-specific expectations about which strategies the winning leader employed (both individual strategies and in terms of exclusivity). The next chapter presents the results of the content analysis and evaluates the validity of the predictions described here.
The three cases investigated for this Afiican case are the 1979 internal settlement episode which was a referendum on whether the goals of the independence war had been achieved; the 1980 independence elections where the identity of the newly independent
206 State was hotly contested; and the 1990 elections in which a new party challenged the
hegemony of the ruling party which had been aiming to establish an official one-party
state. A difference between these cases and the Northern Ireland cases is that for the
latter, the same two leaders are examined across time. For Zimbabwe, the major focus is
on Mugabe and his challengers. For the first two cases, this is Muzorewa, but
Muzorewa’s bid in 1990 (he received .4% of the vote) was basically ignored by observers compared to the candidacy of Edgar Tekere. After a brief discussion of the historical background, each case is discussed in turn. Table 5.1 summarizes the shifts.
1 9 7 9 a 1980*’ 1990 Mugabe more successful: 63% of black vote 76% of total vote turnout was 1.7
million relative to 2 . 6 million in 1980
Muzorewa less successful than 8 % of black vote (.4% of total vote, but Mugabe focus here is Mugabe-Tekere competition) Tekere 17% of total vote: considered successful because emerged as new competitor “ Because of the nature of this contest, the leaders are not compared in the conventional way; see text below. In the first two elections, whites were allowed a particular number of seats and were voted in separately.
Table 5 .1 ; Summary of Mobilization Shifts in Zimbabwe
5.2 Background
The historical experience of Zimbabwe insures that this country meets the case selection criteria described in Chapter 2; the country went through a severe legitimation
207 crisis in the decolonization era and again in 1990 as much of Afiica was under pressure to convert to multiparty democracy; Zimbabwe has a history of colonial domination which created/encouraged a society with potential communal divisions; there are multiple group identities at the subnational and supranational level to which any leader can appeal; and there has often been international involvement of varying degree.
As was the case with Northern Ireland, Zimbabwe was settled as a British colony, and also in the 1920s (1923), settlers were given the right to rule themselves. To the same effect that power granted from London to Stormont had on the Protestant/unionist majority in Northern Ireland, the self-rule granted to Rhodesians (the white home government) meant that the move for decolonization was strongly resisted. Two differences between the experiences of the two countries were that (1) the order the
Northern Ireland unionists sought to maintain could be justified by the fact that it was based on majority rule; and (2) the British government was more dependent on unionists for political support in the Westminster parliament, giving the unionists some leverage with London. As the evidence presented in this chapter demonstrates, this difference leads us to expect that the values of the “change in stability of elite alignment” variable were not the same in Rhodesia.
In Rhodesia, the whites were a tiny minority (4%) who sought to hold political power over the black majority (96%). In the face of decolonization across Afiica,
Rhodesians feared that Britain would exercise its ultimate authority and grant independence to the Zimbabwe nationalist movements. Consequently, in 1965, the white government of Ian Smith’s Rhodesian Front issued a Unilateral Declaration of
208 Independence (UDI)(Kriger, 1992, pp. 2-3). Britain’s reaction was to ask the United
Nations to impose an oil boycott in 1966. In 1968, the United Nations Security Council imposed comprehensive mandatory sanctions (Sylvester, 1991, p. 46). This reaction to the white government’s behavior encouraged the nationalist movements led by Joshua
Nkomo (leader of ZAPU) and Robert Mugabe (leader of ZANU) to continue their armed struggle which had begun in 1966.‘
The Smith regime did not initially suffer unbearably from the sanctions regime, was able to deflect British demands for new constitutional arrangements, and was also able to fend off the initial guerrilla warfare with massive military support from South Africa.
Finally, circumstances changed in the early 1970s. Importantly, South Africa’s Prime
Minister Vorster had a change of heart about the kind of solution in Rhodesia that would be best for South Africa. Since it looked as if Smith would keep fighting until all
Rhodesian blacks had turned into communists, Vorster initiated a first all-party talks effort in 1974 (Sylvester, 1991, p. 51). He was motivated to avoid the path that decolonization had taken in Mozambique that year and would also occur in Angola soon after. In the former case, Mozambique suffered economic devastation when Portugal and the white settlers withdrew rapidly and were replaced by the Marxist FRELIMO (Front for the
' As is common with wars of national liberation, the regime challengers were actually divided into several competing groups: the main ones here were the Zimbabwe African National Union-Patriotic Front (ZANU-PF) headed by Robert Mugabe and the Patriotic Front-Zimbabwe African Peoples Union (PF- ZAPU) headed by Joshua Nkomo. In the Zimbabwe case, several studies have explored the question of whether these groups were ethnically based, with ZANU representing the Shona ethnic groups (almost 80% of the population) and ZAPU representing the Ndebele (about 18%; the remainder of about 2% is made up of the whites who decided to stay after independence). Despite starting out in the 1960s as a multietlmic coalition and then in the mid-1970s having a few party officials from the other ethnic group, most scholars accept that this characterization is accurate; still, there are significant divisions among kinship groups within the Shona as well (see Schütz, 1990; Sithole, 1995). From both 1976-1980 and 1987-present, the two parties unified. 209 Liberation of Mozambique), a ZANU ally (Tamarkin, 1990). Vorster’s first effort and
several other attempts at negotiations failed, until by 1978, Rhodesia was under
unbearable pressure from sanctions, OPEC oil prices, and the continuing guerrilla war.
Further, the physical and diplomatic support of the Front Line States (Mozambique,
Zambia, and Tanzania) for the armed struggle by the armies of Nkomo and Mugabe
(which merged as the Patriotic Front from 1976-1979), made it look as if there was no
end or victory in sight. At the same time that the war was intensifying. South Afiica was
increasingly reluctant to foot the bill for the Rhodesian resistance. Finally, the increasing
pressure from Britain and the United States pushed Smith to propose a solution that
included black rule.
Smith’s plan grew out of earlier proposals from the failed Geneva Conference of
1976, the Anglo-American proposals which were drawn up in large part by Henry
Kissinger. The “internal settlement” of March 1978 proposed a new constitution that granted universal sufifage and provided for an Afiican majority parliament. This constitution was “ostensibly designed to grant blacks pohtical power but in fact ensur[ed] continued white dominance” (Weiss, 1994). Executive and judicial powers remained with the white minority. As a transitional arrangement, this settlement may have been acceptable to Britain, the United States, and others. However, the stumbling block was that Smith excluded Mugabe or Nkomo from having any role. After Smith negotiated the internal settlement with the nationalist leaders, Muzorewa, Sithole, and Chirau, an election
210 for prime minister of the “new nation” (called Zimbabwe-Rhodesia) was held in April
1979 as a “symbolic referendum o f African acceptance of the agreement” (Stedman, 1991, p. 147).
Although Britain was on the verge of accepting the internal settlement, that country eventually took the view of the United States, the United Nations, and the
Commonwealth states that the plan was illegitimate. The new prime minister would not be recognized by the international community because the nationalist movements in exile were excluded from the settlement.^ Further, the UN sanctions would remain in place until a settlement involving all parties was negotiated. South Africa was the only state to recognize the elections and the new Zimbabwe-Rhodesia.
The very fact that these elections were held up as legitimate by Smith and South
Africa led to increased international pressure on the “transitional government” in
Zimbabwe-Rhodesia. On the regional front, British pressure on South Africa to stop supporting the regime and Front Line States pressure on the nationalist movements to end the war^ finally encouraged all parties to attend the Lancaster House Conference in
September 1979. At this conference all parties submitted to British and Commonwealth mediation, and two months later, in December 1979, the Lancaster House Agreement was signed. This agreement—among other requirements— called for all-party elections in early 1980.
■ Mugabe, Nkomo, and most other nationalist leaders lived outside of Zimbabwe because they would be captured and/or arrested if foimd in the country during this period.
^ These factors are discussed fully in the presentation of the 1980 case below.
211 Given the role of)the UN, the US, the Front Line States, South Africa, and Britain in the politics of this country in 1979 and 1980, it is immediately apparent that for
Zimbabwe, the international intervention/mediation variable should play an important role.
The next section discusses the first mobilization episode, the battle between Muzorewa and the Patriotic Front to be seen as the legitimate voice of Zimbabwe.
5.3 First Mobilization Episode: 1979
5.3.1 The Competitors and the Campaigns
The 1978-79 “internal settlement” period is the first case investigated because the two leaders studied here competed to mobilize the black population around opposing ideas of who belongs in Zimbabwe (or Zimbabwe-Rhodesia) and Zimbabwe’s fixture. Abel
Muzorewa (of the United Afiican National Congress, or UANC, party) competed in the elections on the platform that working with the whites in the power-sharing arrangement was the best way to bring about peace and prosperity.'* In his campaign, he emphasized the programs the new government was beginning to put in place which would make the lives of people better. For Muzorewa, the principal issue in the campaign was ending the war so that all of these policies would be allowed to take effect. Because ZANU and
ZAPU were proscribed from operating among the people in Zimbabwe, in the latter years
^ That Muzorewa stepped in to share power with the white regime against the wishes of Mugabe and Nkomo is somewhat ironic given the way Muzorewa became part of the nationalist movement in the first place. In an effort to unite several of Zimbabwe’s nationalist parties, the Front Line States created the Afiican National Congress with Muzorewa as only a figurehead leader because others were in exile or in prison (Kriger, 1992, pp. 86-87). 212 of the war guerrillas were recruited under the name of the UANC. For this reason,
Muzorewa calculated that they would be loyal to him and that he therefore had the power to end the war (Sithole, 1986, p. 80).
Muzorewa had excellent resources with which to get his message to potential followers. Unlike Mugabe and Nkomo, Muzorewa had state-supported access to television and radio and did not have to contend with security force resistance during political rallies and meetings. In support of the UANC vote, and to counter the PE’s, abstentionist campaign, the Rhodesian Front-Muzorewa government ran radio, television, and newspaper advertisements such as the following: “We are all going to vote! That is what the people want!” (also, for excellent coverage of the propaganda campaign, see
Frederikse, 1982; Sithole, 1986, p. 78).
Although the Mugabe-Nkomo alliance (the Patriotic Front) was excluded from competing in this election, it is investigated as Muzorewa's competitor because its leaders did seek to mobilize voters around its view of the situation: essentially that the only true democracy in Zimbabwe is one in which the majority rules and the people, not the foreign settlers, have the power. For example, Mugabe issued statements and broadcast radio addresses denouncing the elections as illegitimate and calling on people to stay away from the polls to continue the nationalist resistance;^ distributed flyers about the elections; launched an international campaign with the aid of the Front Line States; and in sum, tried to convince people that the vision of Zimbabwe they should support must take into account the will of the people and the “freedom fighters” who were being excluded by the
^ For example, one of the radio broadcasts coded for this study was entitled “Ignore the April bogus elections.” 213 internal settlement. Given the voter turnout in these elections compared to the turnout in
1980, and the support for continued resistance on behalf of Mugabe’s goals, the argument in this study is that Mugabe, as leader of the Patriotic Front, was the beneficiary of mobilization in 1979.'' For example, the turnout was 1,786,453 in 1979, and 2,649,529 in
1980 (Campbell, 1980, p. 11; Sithole, 1986).
5.3.2 The Context
For each of the political opportunity variables. Chapter 2 suggested the values that are expected to promote the winning of either more exclusive or more inclusive strategies, as well as the kinds of contexts that are expected to promote the individual strategies. This section investigates the evidence for the 1979 case. First, Table
5.2 summarizes the expectations, then the discussion that follows provides a summary of the evidence to support the predictions.
® For a difiFerent interpretation, see Tamarkin (1990) who says that there was an impressive voter turnout. That interpretation is disputed by a comparison with the 1980 elections and by the primary resources used in this study, i.e. papers of elections observers for both periods found in the Rhodes House Library at Oxford University. 214 Change in stability of elite alignment; Colonial-Controlling (A) Looser Controlling-Regional (C) Tighter Strateg}' predicted to win Exclusive
Repression: Conununit)" Higher Subgroup/party Higher Strategy predicted to win Exclusive
Involvement/Mediation present, sympathetic to Mugabe from International Community Strategy predicted to win Inclusive and favoring Mugabe
Regional Integration not significant Strategy predicted to win
Overall prediction for Exclusive Winning strategies (2 of 3)
Individual strategies favored Injustice, Enemy Image, Storytelling, Governance Negative
Table 5.2; Summary of Expectations for Rhodesia 1979
Change in Stability o f Elite Alignment
The relationships investigated are those of the colonial power and controlling power (alignment A, Britain and Rhodesia) and the controlling power and the regional power (alignment C, Rhodesia and South Afiica). The evidence gathered for this case supports the argument that this case falls in the lower left of the table provided in Chapter
2, and exclusive strategies were favored. The negative relationship fostered between the colonial and controlling powers by the Rhodesian Front’s Unilateral Declaration of
Independence from Britain, and the subsequent defiance of British efforts to promote transition to majority rule indicate that alignment A loosened. Meanwhile, the relationship
215 between the Rhodesians and South Africa remained relatively tight. South Africa was
backing away from its willingness to continue its massive military aid to Rhodesia, but for
exactly this reason was solidly behind the internal settlement. Thus, the dissolution of
relations between Britain and Rhodesia, along with the close relationship between South
Africa and Rhodesia created the situation where there was support for the status quo by
South Africa and no leverage for change on Britain’s part. As discussed in Chapter 2,
when the political minority cannot observe “cracks” in the status quo, they are less likely
to respond to calls to work with the status quo powers to get the latter to compromise and
help the pohtical minority.
The dynamics of both o f these relationships can be observed in accounts of the
ongoing negotiations involving Britain, the United States, South Africa, the Front Line
States, the Smith regime, and the Patriotic Front (see also Stedman, 1991, Chapter 4;
Tamarkin, 1990, Chapter 7). During mid-1977, Britain and the United States pressured
Smith to accept their proposals on transition to majority rule whhe he stood firm. When these parties met with South Afiica’s Vorster and Botha to convince them to pressure
Smith, they replied that they could not ask Smith to accept proposals that they could not support themselves (Tamarkin, 1990, pp. 190, 196). Later, South Afiica gave full support for the internal settlement and called on the British and the Americans to recognize it and remove the economic sanctions (Tamarkin, 1990, p. 224). One observer argues that, with an internal settlement (and a moderate Afiican-RF coalition) as one option, Pretoria feared
216 the Anglo-American alternative was likely to bring Marxists to power. Looking at the
Mozambique and Angolan situations, South Africa’s best option was to stay aligned to
Smith’s Rhodesia (Tamarkin, 1990, pp. 190, 196).
Repression
An examination of security legislation and the degree o f state violence reveals that
repression on the wider community and the nationalist parties increased, meaning more
exclusive strategies are expected to win. Indeed, several observers noted that the
repression of the peasants was counterproductive and only drove them into the arms of the guerrillas (for example, see Weitzer, 1990).
The laws that instituted “petty apartheid” in Rhodesia go back to the early 1900s.^
However, the research for this project focuses on more immediate increases (or decreases) in state repression. The beginnings of the security apparatus can be seen in 1959, when the Rhodesian government declared its first “state of emergency” and arrested 500 members of the African National Congress. This same year the Unlawful Organizations
Act outlawed several African organizations and the Preventive Detention Act allowed detention without trial. These moves embedded in Zimbabwe’s pohtical culture an idea that endures today; that political opposition is inherently subversive (Weitzer, 1990, pp.
63-64, 70).
Since UDI in 1965, the government has annually declared a state of emergency, which essentially increased the powers of the executive and the security sectors relative to other branches of government. The security forces were granted the authority to detain
' For a description of the repressive apparatus built to secure white domination, (see Weitzer, 1990). 217 people indefinitely without trial. Security legislation proliferated and affected all aspects
o f life for the black majority. Numbers of poUce stations increased, for example, there was
a tightened control of urban townships, membership in political organizations was
proscribed, and gatherings on Sundays and pubhc holidays were prohibited (Kriger, 1992,
pp. 110-111). In 1973, the Law and Order (Maintenance) Act was amended to permit
execution of anyone supporting the guerrillas or failing to report the presence of the rebel
units. Other regulations in 1973 permitted the government to charge collective fines on
entire villages if people in them were suspected o f supporting the guerrillas. Finally, martial law was announced in September 1978, between the agreement of the internal settlement and the elections (Weitzer, 1990, pp. 92, 96).
Growth in expenditures for the security system (both internal and external) are additional indicators of increased repression. From 1971-1972 to 1976-1977, the Ministry of Internal Affairs saw its budget go fi'om RS9.7 million to R$42 million; the budget for the Ministry of Law and Order, which including policing, when fi'om RS17.5 million to
R$50 million; and the Ministry of Defense expenditures went from R$20 million to RS98.7 million. None of these figures include the mihtary aid coming firom South Africa (Weitzer,
1990, p. 87). The legislation, its execution, and the expenditures show that the state’s propensity for repression on opposition parties and the wider community increased over time and was very high in 1978. Several specific examples further illustrate the point.
A widespread practice of the Rhodesian regime, begun in 1974 and continued by
Muzorewa, was the herding of peasants into “protected villages” to keep these rural
Africans ‘^afe” from the guerrillas; the villages were heavily guarded by Africans recruited
218 by the security services and were subjected to strict curfews (sometimes lasting from 6:00 p.m. until noon). By the end of the war, it is estimated that 700,000 peasants had been placed in around 200 camps. The food supplies to these areas were also strictly controlled, in order to keep the guerrillas from getting any surplus of food. These conditions gave the appearance of prison camps (Kriger, 1992; Ncube, 1991; Sylvester,
1991, p. 54; Weitzer, 1990, p. 93).
Another way in which repression on the community was high and increasing was that the government mobilized all its military manpower to provide a “security screen” for the elections. This involved the increased presence of the military in the streets
(Tamarkin, 1990, p. 243). Along with this domestic “security screen,” the transitional government (but because of who held actual power, the Rhodesians) also stepped up what we can call here “repression” on the opposition parties. Meetings and demonstrations by any party opposed to the elections were prevented, for example (Freedom House
Observation Group, 1980). The Muzorewa-Smith government also increased pressure on the guerrilla armies by raiding camps in Mozambique and Zambia, trying to assassinate
Nkomo and other leaders, and in April 1978 instituting a campaign of “economic sabotage” against the countries supporting the guerrillas (Tamarkin, 1990, p. 232).
Eventually, the campaign would cost the Rhodesians in terms of lives lost, but in all it is concluded here that state repression on both community and subgroups of the community was very high in this period. Shght concessions on some of the early petty apartheid laws were made one month before the elections, but for the most part the security apparatus remained in place (Weitzer, 1990, p. 105).
219 International Involvement/Mediation and Regional Integration
International involvement favored inclusive strategies, and favored Mugabe.
International actors were sympathetic to the cause of the pohtical minority in general, but
(with the exception of South Afiica) expUcitly rejected Smith and Muzorewa's idea that an
internal settlement was possible. Instead, the necessity of bringing in the Patriotic Front
leaders was emphasized. Britain and the United States even refused to send observers to
the 1979 elections on the basis that the poll was iUegitimate (Tamarkin, 1990, p. 242). To
Britain, the Commonwealth, the United States, the Front Line States, the United Nations,
the Organization of African Unity’s Liberation Committee, and others the ongoing war in
Rhodesia and the legitimate decolonization of the area was seen as an international issue
(see Baumhogger, Diedrichsen and Engel, 1984 for an extensive collection of the primary
documents and declarations from most of these organizations) This was reinforced with
resolutions coming from all of these political bodies that the internal settlement was illegitimate and that sanctions should remain in place until an internationally-recognized government was elected based on a negotiated settlement (see Stedman, 1991).
For example, the Front Line States ended a summit to discuss their position on the internal settlement with the following statement; “...The Front Line States...reafiirm their total and unswerving support to the armed struggle being waged by the people of
Zimbabwe under the leadership of the Patriotic Front for the attainment of complete independence and the establishment of a genuine democratic government” (Tamarkin,
1990, p. 221). Further, the OAU Liberation Committee enhanced the Patriotic Front’s position by endorsing the decision of the Front Line States to provide the PF with full
220 political, material, and diplomatic support. A few months later at the OAU summit conference, members were calling on “all Zimbabweans devoted to the struggle for the liberation of their country to do so within the Patriotic Front” (quoted in Tamarkin, 1990, p. 221).
Some observers point to the international community's attitude toward Rhodesia as a sign that the United States and others were not wholeheartedly behind a majority-rule settlement. For example, Kriger argues that the US supplied aircraft to the Rhodesians from 1975, did not try to stop mercenary recruitment into the Rhodesian forces, and violated the international sanctions on Rhodesia by importing chrome from 1971-1977.
Both Britain and the United States continued to allow the flow of investment funds (1992, p. 113).
Despite this history, what is significant is that the US position changed dramatically after the Carter administration’s policies started to gather momentum—just as the internal settlement was being negotiated. The views of Secretary of State Vance,
National Security Advisor Brzezinski, and United Nations ambassador Andrew Young were that the United States should work to promote peace and development in Africa.
Dealing with the Rhodesian war was a number one priority', reflected in Vance’s actions in early 1977. For example, he armounced in January 1977 the repeal of the Byrd
Amendment, which allowed US purchase of Rhodesian chrome despite UN sanctions, and condemned the illegality of the Rhodesian government. In addition, the United States and
Britain agreed to step up negotiation efforts, with plans for a conference with all parties and for the estabhshment of a development fund to aid the independent Zimbabwe (see
221 also Stedman, 1991; Tamarkin, 1990, p. 185). Throughout the negotiations going on
before, during, and after the internal settlement, the British and the Americans supported
inclusion of the Patriotic Front in any “legitimate” resolution of the conflict.
The United Nations also engaged in mediation efforts (beyond its ongoing
involvement in the sanctions arena) prior to and during the internal settlement period, by
appointing a special representative to work toward a cease-fire settlement (Tamarkin,
1990, pp. 199-200). Although this mission was quite unsuccessful, it should be
recognized as another effort by the international community to include the Patriotic Front
in negotiations. A second UN action is evidence of diplomatic involvement by that
international organization: In March 1978, the Security Council held a vote on an
African-sponsored resolution that supported an ofBcial UN rejection of the internal
settlement (Tamarkin, 1990, p. 235).*
Finally, the Front Line States were deeply involved: they gave the PF diplomatic
and military support in the sense that the Patriotic Front forces were allowed to remain in
their bases in Mozambique and Zambia; Angola and Tanzania still provided training
facilities; and Botswana maintained its policy of allowing free passage of Nkomo's troops
from Zimbabwe to the training camps. The Soviet Union (helping Nkomo's side of the
PF) and China (helping Mugabe’s side of the PF) may also be said to have intervened in
military ways by supplying arms (Tamarkin, 1990, pp. 221-222). Overall, the definition of
the situation in the international community as an “international issue” promotes inclusive
* This action was tempered fay the following, faut the other evidence afaout the US and UK still leans in the direction argued afaove. While some statements made fay Britain and America rejected the Smith- Muzorewa arrangement outright, in March of 1978 those two countries and other Western members abstained from voting (Tamarkin, 1990, p. 235). 222 strategies; because it was not neutral, most international involvement favored Mugabe.
The one exception was South Africa, whose involvement favored Muzorewa.
In this case, there was little regional economic integration. Sanctions on the
Rhodesian government were hurting the economies of the Front Line States (Stedman,
1991). The main source of trade was with South Africa. For political integration, the relationship between the Zimbabwe nationalist movements and the Front Line States was in existence, but had not changed around this particular period. Further integration on both the political and economic fronts would happen after independence. Based on the propositions in Chapter 2, this situation means that this variable is not considered significant because integration is not acting to promote any sense of common regional identity that might lessen the conflicts among actors within the region, and the present state is not a reduction in integration.
5.3.3 Expectations about Individual Strategies
Along with expectations about contexts that promote more or less exclusive strategies, it is posited here that the context in any mobilization episode will affect which individual strategies the leader is using to mobilize people around a particular definition of the situation. Again, part of the puzzle for this project is to understand which strategies appear to be more successful in particular contexts. Extant work offers some insight but does not speak to combinations of strategies. For example, when repression is especially high against the ingroup, a leader who emphasizes the injustice endured by people, generally the ingroup (the Justice Negative strategy) is expected to be more successful,
223 but must he complete the picture by pinpointing “who” is responsible for “our” suffering
(thereby combining the Enemy Image strategy with the Justice Negative strategy)?
For Rhodesia/Zimbabwe 1979, (1) repression against blacks was high and their
situation was only getting worse; (2) most external actors agreed that the regime in power
was racist and illegitimate; and (3) almost all actors (except Muzorewa and South Africa)
were calling for further negotiations. Indeed, most of the major players in the
international community (South Africa is the one exception) were standing against the
internal settlement in favor of the Patriotic Front’s participation in a future settlement. For
these reasons, it is expected that successful strategies would be Justice Negative, because
the characteristics were in place that promote feelings of relative deprivation (see reason 1
above). Also it is expected that the Enemy Image strategy will be used by the more
successful leader because the political minority was under attack. In regard to the latter,
unlike situations where more general causes are responsible, a leader who “reasonably” pinpoints a responsible party and has external validation of this definition of the situation may be more successful. Also, (3) the fact that nearly everyone involved was highlighting the illegitimacy of the transitional government, it is expected that Governance Negative strategies will be common. Finally, because the situation called for action (either to vote and surrender or to keep fighting), use o f the Storytelling strategy is expected to be relatively frequent (it was stated that the imagined communities perspective should apply most in times of great uncertainty and when leaders are trying to mobilize people to act).
224 5.4 Second Mobilization Episode: 1980
During and after the elections of 1979, the effort to bring an end to the civil war continued, spurred on by pressure from the international community and the growing unwillingness of South Africa (on the Zimbabwe-Rhodesia side^) and the Frontline States
(on the Patriotic Front side) to support continued fighting. For example, since 1977 the
Rhodesian forces had taken the war into Zambia, Mozambique, and Botswana.
Mozambique and Zambia were suffering from disrupted economies and threats to political stability to the extent that Machel (Mozambique’s leader) and Kaunda (Zambia’s leader) applied immense pressure on the PF to go to Lancaster House with the mind to compromise (Mandaza, 1986, pp. 38-39).
The Patriotic Front was also “shaken by both the occasion of the ‘internal settlement’ (and the establishment of its Zimbabwe-Rhodesia), and by the threat that
Britain and the US might recognize such a settlement as constituting the fulfilment [sic] of the much-desired compromise between whites and blacks” (Mandaza, 1986, p. 36).
Indeed, within both the British Parliament and the US Congress, votes to lift the sanctions on Rhodesia had only barely failed (Tamarkin, 1990, p. 239). Finally, prior to the 1979 elections (in August 1978), Smith had tried his hand at wooing Nkomo into the internal settlement government, in an effort to divide and conquer the PF. This aspect may also have encouraged Mugabe to be more compromising at the Lancaster House conference which was finally convened in September 1979 (Mandaza, 1986, p. 317). After describing
® South Africa had invested US$300 million in the war against the nationalists (Sylvester, 1991, p. 167). 225 the mobilization shift in more detail, this section will discuss specifically how regional and international conditions set the context within which the nationalist leaders attempted to mobilize people around their visions of a new, fi-ee Zimbabwe/®
5.4.1 The Competitors and the Campaigns
At the Lancaster House Conference, Muzorewa and Smith faced Mugabe and
Nkomo’s Patriotic Front. The results of the rancorous negotiations were as follows;
Rhodesia’s legal status as a colony under a British governor was re-established, meaning
Muzorewa’s Transitional Government was forced to resign; a cease-fire and arrangements for demobilizing the armies were agreed; plans were made for an election under British supervision in February 1980; a constitution was negotiated that would grant whites twenty seats in the 100-seat parliament for seven years; the Patriotic Front agreed that white land ownership would be protected and land sales could occur at market prices if the seller is willing; the armies were to be integrated; the new government would be given the power to “Afiicanize” the civil service; and Britain and the United States would deliver a large amount of economic aid to the new Zimbabwe so that some land could be bought for the resettlement of black peasants (Stedman, 1991). The first elections for leadership of the new country, in February 1980, are the focus of the second mobilization episode.
During the short interim period between the ceasefire and the February elections, the situation in Zimbabwe was volatile. All sides accused each other o f breaking the rules
See Stedman (1991) for an authoritative and detailed description of the actual negotiating process at Lancaster House. 226 of the ceasefire; the British governor firequently accused Mugabe o f intimidating potential voters and threatened to disqualify his party; and Mugabe and the Front Line States accused Britain of sabotaging the peace and rigging the elections. The bias by Britain against the Patriotic Front leaders was thought to be such an injustice by some parties that
Tanzania’s Nyerere wrote a letter to the UN Secretary-General, urging the UN to send a special investigative team (Baumhogger, Diedrichsen and Engel, 1984, pp. 1333-1334).
In addition, Mugabe survived two attempts on his life, allegedly made by people in the
Rhodesian regime trying to sabotage the transition; and South Afiica on one side and the
Front Line States on the other hinted that a loss for their patron (Muzorewa and Mugabe or Nkomo, respectively) would bring war back to Zimbabwe. As discussed below, one of the reasons the elections were carried out despite all of these obstacles was that each of the major candidates thought he would win.
The 1980 election was contested primarily by Smith and the Rhodesia Front,
Mugabe and ZANU-PF, Muzorewa and the UANC, and Nkomo and ZAPU (who decided to run on his own after Mugabe announced he would not run in a coalition with ZAPU).
For the analysis, the foci are Mugabe and Muzorewa, because historical accounts of the elections argue that the contest was seen to be between these two, Nkomo and Mugabe had similar appeals, and Smith did not try to appeal to the black community. Mugabe and
Muzorewa also competed for the attention of similar audiences, while the Mugabe-Nkomo split was largely along the Shona-Ndebele tribal lines (even though these leaders do not make appeals based on arguments about tribal membership) (Sithole, 1995). Further,
227 many observers expected Muzorewa to win because he was now recognized
internationally as a legitimate contender (as opposed to the 1979 elections) and had been
visible as a force for peace.
To the observer, Muzorewa had far superior mobilizing resources vis-a-vis
Mugabe. He was expected to benefit fi'om: the fact that his steadfast promise to bring
peace and stability to Zimbabwe was being fulfilled, the indirect support of the United
States which did not want to see a socialist regime (Mugabe) come to power, and (for
campaigning) the ready-made infrastructure of his United Methodist Church of which
many potential voters were members." Beyond the church infi'astructure, Muzorewa was
ahead of Mugabe in terms of other campaign resources, including access to the state-run
media and fi'eedom fi'om harassment by the security forces. OfiBcially, each party was
permitted three political broadcasts on radio and television, and could buy time up to an
agreed limit for the transmission of slogans. There were no limits placed on the
distribution of posters, placards, and publication of newspaper advertisements (Campbell,
1980). Despite this “level playing field” in terms of opportunities, Muzorewa had the
advantage in terms of purchasing television, radio, and newspaper spots because of the aid
he received fi'om the Rhodesians and South Africa. South Africa alone spent US$2 million in the election campaign against Mugabe, most of which went to Muzorewa
(Sylvester, 1991, p. 167). Muzorewa even used airplanes donated by South Africa to fly over the country playing Afiican music and broadcasting UANC slogans (Sithole, 1986, p.
" The election results were a surprise to many, Nkomo and Muzorewa thought they each would win more seats; the whites thought Muzorewa would dominate Mugabe and Nkomo (Dashwood, 1997). 228 96, fh.l 1). He also had greater resources and access to distribute posters and leaflets
about the evils of socialism (which was the system Mugabe promised to introduce) and the
threat to the lives of people if the “militants” took office.
On the other hand, it may be argued that potential voters did not see Muzorewa as
the one who had brought peace. Instead, he was discredited by his failure to bring the
guerrillas home and achieve peace after the internal settlement. As one observer argued,
the support of Muzorewa from South Africa, the Rhodesian Front, and the West in
general may have had the opposite effect than intended and actually endeared Mugabe to
the African voters (Sithole, 1986, p.84). Violent intimidation may have played a role in
the elections as well, but accusations came from all sides. The UANC argued that
Mugabe and ZANU (PF) guerrillas violated the agreement to demobilize fighting forces.
The continuing presence ofZANLA forces in the rural areas therefore created large
sections of the country where Muzorewa could not campaign.On the other hand, the
multiple attempts on Mugabe’s life during the 1980 campaign have been attributed to the
RF and South African security branches. Despite the widespread accusations of violence
on both sides, the British governor and the contingent of international observers
pronounced the elections free and fair, thereby giving validity to the results. The Freedom
House Observation Group expressed some reservations about intimidation by parties and
As others have noted, it would probably not have mattered had Muzorewa been allowed into these areas: in particular segments of the society the years of support for the “freedom fighters” would not be reversed by several weeks of electoral campaigning (Sithole, 1986, p. 85). 229 “black and white Afiican governments” that if particular parties did not win, war would resume; however, they too noted that this intimidatio was on both sides (Freedom House
Observation Group, 1980).
In terms of campaigns, Muzorewa adopted some of the Patriotic Front’s earher ideas by promising land reform, free compulsory primary education, adequate housing, and firee health services. His central argument was that his government would be the only protection against Marxist leaders who would usher in socialism sure to bring death to the economy and anyone who disagreed with that regime (Muzorewa, 1980; Sylvester, 1991, p. 67). The following excerpt firom Muzorewa’s “Eve-of-Poll” election broadcast illustrates the point:
A vote for ZANU-PF and the Patriotic Front will be the last fi-ee vote. Under their one-party Marxist state, never again will we vote democratically, for all other parties will be banned and the leaders executed. A vote for ZANU-PF or the Patriotic Front means everyone, black, white, or brown, will lose businesses, farms, houses, land, cars, bicycles, livestock, everything (Baumhogger, Diedrichsen and Engel, 1984, p. 1367).
Mugabe’s platform set forth a program based on socialism: the plan was to acquire land for resettlement; establish collective villages, agriculture and state farms; and promote government partnership in mining and other industries. Most significantly, this candidate emphasized that his party alone was responsible for ending the racist domination of Zimbabwe and would be the only legitimate leader of the firee Zimbabwe (Mugabe,
1980; Sylvester, 1991, p. 68). The following excerpt fi’om Mugabe’s “Eve-of-Poll” radio broadcast is illustrative:
As we get close to the elections 1 would like all o f you to look back into the past and then look ahead into the future. We promised we would carry
230 out a national struggle to achieve full independence...We stood firm by our principles and refused to yield or retreat. Finally, we succeeded and became party to the Lancaster House agreement, which offers peace, fireedom, and independence... Now, ask yourselves which party can be trusted to act honestly, consistently, determinedly, and decisively in fulfilling the wishes of the people. Of course you cannot avoid to say, only ZANU-PF (Baumhogger, Diedrichsen and Engel, 1984, p. 1367).
The wide range of sources consulted for this project remarked that the world—and especially the Rhodesians and South Afiica—was shocked by the outcome of the election.
Mugabe won an overwhelming victory, drawing support of the Shona (his ethnic group) who made up 82% of the population (as expected, Nkomo received the support of the
Ndebele population). The total vote for Mugabe was 63%, which gave ZANU 57 seats.
Muzorewa and the UANC, in contrast, won 8% of the vote and 3 seats (Sithole, 1997, p.
131)."
5.4.2 The Context
The predictions for the political opportunity variables are summarized in Table 5.3.
The rest of this section discusses these expectations.
Nkomo received 24% of the vote and 20 seats. 231 Change in stability of elite alignment; Colonial-Controlling (A) Tighter Controlling-Regional (C) Looser Strategy predicted to win Exclusive (no change, not significant)
Repression: Community Lower Subgroup/party Lower Strategy predicted to win Inclusive
Involvement/Mediation present, sympathetic to Mugabe from International Community Strategy predicted to win Inclusive, favoring Mugabe
Regional Integration Strategy predicted to win not significant
Overall prediction for Inclusive Winning strategies (2 of 2)
Individual strategies favored Governance Positive, Storytelling, Enemy Image, Injustice
Table 5.3; Summary of Expectations for Zimbabwe 1980
Change in Stability o f Elite Alignment
Compared to the previous period, the “tightening” o f alignment A (between the
colonial power and the controlling power) and the “loosening” of alignment B (between the regional power and the controlhng power) points to the expectation that the strategies used by the winning leader will be more exclusive. The evidence is summarized below.
Note that this is the same combined prediction as in 1979, so this variable is not expected to be significant in this case.
Although in the previous period the British Labour government exerted a great deal of pressure on Rhodesians to negotiate and the Conservative government elected in
232 June 1979 pushed the whites to go to Lancaster House, the alignment between the
colonial and controlling powers changed in this second period. Once the Rhodesians were
at the talks, represented by Muzorewa and Smith, Britain bargained hard with the PF to
guarantee economic protection for the whites as well as over-representation in the
Parliament for the first seven years of independence (for a discussion of the negotiations,
see Stedman, 1991). Britain also supported several other reassurances to the Rhodesians:
In December 1979, the British governor announced an amnesty so that those involved in
the war could not be prosecuted. This move may be seen as benefiting the PF as well, but
there was not much concern that a black government would prosecute Afiican nationalists
for war crimes. Further, despite calls right before the 1980 elections by Amnesty
International and some Commonwealth nations for the British interim government to
repeal the security laws and also release people detained under the emergency legislation,
Britain refused to tamper with the Rhodesian legislation (Weitzer, 1990, p. 108). Finally, the PF had wanted a United Nations peacekeeping force or at least UN officials to
monitor the transitional arrangements, but Britain instead insisted on a Commonwealth force under British command. Stedman (1991, p. 183) states that Britain took this route because its negotiators knew the Rhodesians would not tolerate the UN involvement.
Britain played a role in the changes in alignment C—in the loosening of the regional power (South Afnca)/controlling power (Rhodesia) alignment, though other forces were at work as well. After it came to power and then after the 1979 internal settlement elections, Britain’s Conservative government changed course away from the
Labour government’s tactic of pressuring South Afiica. Instead, they realized that South
233 Afiica had to be “courted” in order to break the tight alliance between Rhodesia and
South Afiica. Britain succeeded in doing so prior to and during the Lancaster House
conference (Tamarkin, 1990). Although South Afiica was still on the side of the
Rhodesians, that country was no longer willing to defend the RF’s fight and occasionally
pressed Smith to reach an agreement. As stated above, this shift in both alignments points to the expectation that this variable provides the opportunity for a leader with more exclusive strategies to be more successful.
Repression
Even though the Lancaster House conference restructured the constitution of
Rhodesia, the security apparatus remained intact, with the potential to keep levels of repression as high or higher than in the previous period. However, in terms of its relation to the previous period, research shows that the actual repression on the community and on the political opposition—while still somewhat high—was lessened, favoring inclusive strategies.
Indeed, the dissolution of the transitional government, the subsequent take-over by the British governor, the presence of the Commonwealth observer group, and the decreased support of South Afiica all forced the Rhodesian security forces to de-escalate.
The British used these forces for law and order but compared to the previous period, they were less repressive. The new settlement also legalized opposition parties and extended civil and political rights to the majority. During the international negotiations at the
Lancaster House conference and prior to it. Commonwealth countries and the nationalist parties emphasized that reform of the security apparatus must be instituted once the
234 elections were held even though Britain refused to restructure before the elections
(Weitzer, 1990, pp. 104, 107). The fact that several actors in the international community
were serving a “watchdog” function in the interim period also lessened the degree of
repression (Baumhogger, Diedrichsen and Engel, 1984).
International Involvement/Mediation and Regional Integration
There is no doubt by scholars of the Zimbabwean case that the international
community played a crucial role in the conduct of the war and of the move toward peace.
How did that role affect the strategies leaders used to win the independence election?
Clearly, the situation leading up to the elections of 1980 was defined as international, and
the fact that the international community viewed Zimbabwe’s plight as an international
concern worthy of global efforts to resolve points toward the favoring of more inclusive
strategies. In addition, the Lancaster House constitution imposed upon the parties by
Britain (who “persuaded” all sides to sign it) required a “racial bargain” that promotes
more inclusive strategies (Herbst, 1989, p. 46). As noted above, 20 of 100 seats in the
Parliament were reserved for whites for seven years. The former Rhodesians were also
guaranteed that their land would not be confiscated. Despite the violent rhetoric by
Mugabe in previous years that the blacks would oust the whites off of the land, the
international agreement forced the first independence government to promise whites
compensation in foreign currency for their land—which could only be acquired on a willing seller-willing buyer basis. British and American guarantees for generous contributions to a “fund for land” reinforced the “inclusive” agreement.
235 In terms of whether the international involvement was “neutral” of favoring one
leader over another, several points are relevant. A major goal of the British at Lancaster
House was to convince all actors that they were most likely to win the independence
elections, or else those not expecting to win had incentive to walk away from a settlement
(Stedman, 1991, p. 175). Despite the negotiators’ skills in convincing Muzorewa,
Nkomo, and Mugabe that each would get the most electoral support, it is argued here that
international actions may have created a situation where support for Mugabe predominated—even if verbally some countries expressed that they did not want the
“socialist” to win.
On one hand, several aspects of the international dimension may have supported
Muzorewa’s view of the world and/or undermined Mugabe. Perhaps the strongest evidence is the West’s fear of professed socialist Mugabe and, if only by default, its
“rhetorical” leaning toward Muzorewa. Second, one may make the link that Britain’s pressure on the PF to make concessions to the Rhodesians at Lancaster House (see
“change in stability of elite alignment,” above) meant that Mugabe was less favored by the
British. However, it is argued here that the actions of Britain, the Front Line States,
South Africa, and others such as the Commonwealth states and the Non-Aligned
Movement sent a different message to the newly enfranchised people of Zimbabwe.
First, accounts of the negotiations show (see Baumhogger, Diedrichsen and Engel,
1984; Stedman, 1991) that Britain pushed both Muzorewa and Mugabe on various issues; as a result any message Muzorewa would try to send out during an election campaign was undermined to a greater degree than Mugabe’s appeals would be because there was never
236 any pretense that Mugabe was supported by the West. For example, Britain’s decision, after some wavering, not to recognize the Muzorewa-Smith internal settlement maintained the message that Muzorewa may not be viewed as representative of the people. This message was reinforced by the refusal of Muzorewa’s demands that his government stay in power during the transition. Second, the intemational community as a whole agreed not to move toward lifting sanctions on Zimbabwe-Rhodesia, but once the PF had participated in a settlement this facet changed. This circumstance may have contributed to the perception that the PF leaders, especially Mugabe, were the only ones who could end the country’s hardship.
Third, the fact that Rhodesia and South Aftica, the archenemies of the nationalist movement, supported Muzorewa may have also sent a message that he was indeed the
“puppet” and “traitor” that Mugabe accused him of being. Indeed, the report of the
Freedom House Observer Group (1980) recognized that Muzorewa may have had advantages of any incumbent in elections, but concluded that his connection with the whites and the security services probably hurt his campaign more than it helped. Fourth, the role of the regional community also showed leanings toward Mugabe while alleging that the British governor was biased against Mugabe and Nkomo. The clear afShation between Mugabe and other African leaders was in sharp contrast to widespread condemnation of Muzorewa. Beyond the radio broadcasts by the Zimbabwe nationalists affiliated with Mugabe and/or the PF, state radio in Mozambique, Tanzania, and Zambia also condemned a vote for Muzorewa (based on transcripts of these broadcasts printed in
237 Baumhogger, Diedrichsen and Engel, 1984). The Front Line States also protested to the
United Nations on behalf of the liberation movement leaders, calling for an investigation of
British repression (Baumhogger, Diedrichsen and Engel, 1984, pp. 1332-1333).
Fifth, the way in which the British governor frequently accused the liberation movements about not getting all their troops to assembly points while South African troops were used as guards and a segment of the Rhodesian forces were maintained as a security regime, combined with Rhodesian and South African support for Muzorewa, provided Mugabe with grist for the propaganda mill Broadcasts of the Revolutionary
Voice o f Zimbabwe out of Addis Ababa are exemplary: “Tonight we give a warning to the people of Zimbabwe not to be intimidated into submission by the British-favored
Rhodesian terrorist forces, who are rampaging the entire country...” (BBC/Lexis-Nexis,
1980).
Finally, the guarantees of the British and the United States that they would provide funds for the purchase of lands to be resettled (only if and when the white owners were willing to sell) was a crucial factor enabling Mugabe to demobilize the fighting forces.
One major reason Muzorewa could not bring the guerrillas home in the previous period was that Mugabe and Nkomo continued to mobUize fighters around the goal of winning
Zimbabwe’s land back from the “foreign enemy.” For example, the following passage in
Mugabe’s 1979 New Year Message to the People exemplifies his appeals in the last phase of the war: “Let the People’s firry break into a revolutionary storm that will engulf and sweep the enemy completely from our land.” If Mugabe could not have pointed to the certainty that the settlement achieved the goal of reclaiming Zimbabwe’s land— without
238 continuing to fight the whites—it is doubtful that he would have received the support he
did- The role of promised intemational funds was key, as the following statement by a
ZANU member of the talks (Simbi Mubako, quoted as interviewed by Stedman, 1991, p.
183) indicates: “With the United States offer of aid we had something we could sell to the
people. The Front Line States told us, ‘You have a promise from the United States; if you
feel you have something you can sell to the electorate, then take it.’”
Given these factors, it is predicted that the international context initially set the
stage for the leader using more inclusive strategies to be more successful in his
mobilization efforts. When the kind of international involvement that occurred throughout
the period is examined, it appears that behavior of external actors combined to tilt the
balance in favor of Mugabe.
In reference to the regional integration variable, the degree of integration did not
change significantly from the previous period. Growing out o f mutual opposition to South
Afiica, imperialism, and neocolonialism, there was a strong sense of political identity
among the nationalist movements in Zimbabwe and the Front Line States. As described
elsewhere, the FLS were internationally recognized as political partners of the nationalist
movements. Still, at this point in time no more formal political or economic institutions
had evolved. The cease-fire had just begun in this time period and there were expectations that, once Zimbabwe became independent, a Front Line States organization involving
Zimbabwe would develop. In fact, on April 1, 1980, the month after Mugabe was elected, the “Declaration by the Governments of Independent States of Southern Afiica” elaborated development objectives to bring about greater regional economic integration
239 through the Southern Afiica Development Coordination Conference (SADCC). As the preamble states, "...we state our commitment to pursue policies aimed at the economic liberation and integrated development of our national economies and we call on all concerned to assist us in this high endeavor” (Baumhogger, Diedrichsen and Engel, 1984, p. 1433).
Notwithstanding these intentions expressed after the election, the similar degrees of regional integration in 1979 and 1980 point to the expectation that this variable is not significant in the second shift either, based on the discussion in Chapter 2.
5.4.3 Expectations about Individual Strategies
How might this different context from 1979 also affect the lenses through which the competing leaders were depicting the situation, in terms of the individual strategies?
Which "theories” informed their presentation of reality? Mugabe had to consider that his previous goal was to keep people fighting and resisting. His strategy was to argue that the whites were foreigners who must be driven out so that the land could be returned to real
Zimbabweans. In 1980, he had agreed to allow whites to stay. He had to weave their presence into his picture of “who” Zimbabwe is. Muzorewa had similar goals but also had the added dimension of needing to convince people that his previous government did a good job and would continue to do so. Further, the involvement of the intemational
240 community in promoting negotiations and good governance was an important contextual factor. For these reasons it is expected that the leader using more Governance Positive and Storytelling strategies will be favored by the context.
Each leader also had to convince people not to follow the other: now that both leaders had the goal of ruling Zimbabwe (and not opposite goals of fight/accommodate), we might expect the Enemy Image strategies to be used a great deal by the successful leader. The literature on the Enemy image and the minimal group paradigm tells us that differences between groups are actually emphasized more when the groups are similar.
Finally, a fourth strategy expected to appear frequently in the appeals of a successful leader is the Injustice strategy. This is expected because the situation around which people were successfully mobilized to fight in the previous period had not gone away yet.
A leader who emphasizes only negotiation and positive changes to political institutions in the situation of racial inequality that was ongoing would only be viewed as a sell-out. It is suggested here that the successful leader would still need to emphasize what is wrong while holding out the promise that he has the right way to fix it.
5.5 Third Mobilization Episode: 1990
After his victory in 1980, Mugabe embarked on the project of creating a government of national unity in which whites and PF-ZAPU (Nkomo’s party) would play a significant role (Sithole, 1995, p. 144). In fact, the socialist rhetoric and a willingness to de-politicize ethnicity led the Mugabe government “to outlaw the ‘colonial’ practice of identifying people by ethnic groups, in hope that by so doing ethnicity shall ‘wither away”’
241 (Sithole, 1995, p. 158, fh 8). The need for intemational aid and the incentives to abide by Lancaster House overshadowed Mugabe’s socialist rhetoric and leanings toward
Eastera-bloc countries. After the elections, Mugabe (quoted in Sithole, 1995, p. 144) focused on putting the past behind and urged the public “to join me in a new pledge to forget our grim past, forgive others and forget...” Most observers were shocked at the changes in Mugabe’s rhetoric after he was elected—6om wanting to purge all “foreign settlers” from the land to promising that everyone would have a place in the new
Zimbabwe.
The focus on reconciliation was strictly racial reconciliation; divisions between
ZANU and ZAPU would be more difScult to surmount. Indeed, although he offered a ministry post to a shocked Nkomo (who thought he would win), Mugabe was less successful in reconciling the Shona and Ndebele, as a low-intensity civil war concentrated in two regions lasted from 1982 until the Unity Accord of 1987. In this agreement,
Nkomo’s party was brought into the ruling party as Mugabe moved toward his goal of a one-party state (Schütz, 1990).
After the Unity Accord, Mugabe attempted to improve the “national” look of his government. To do so, he established Nkomo as one of two Co-Vice Presidents; he also appointed several other ZAPU members to senior posts and gave many others various government positions (Sithole, 1995, p. 146). Further, although by 1987 he was no longer required by the Lancaster House agreement to give whites representation, he voluntarily appointed eleven whites to parliament (Herbst, 1989, p. 43). This move got rid of the last traces of racial privilege in the electoral arrangements of the country.
242 The final important change in 1987 was the move firom a parliamentary executive system required by the Lancaster House constitution to a presidential system in which the parliament serves mainly an advisory role. All of these measures were justified with the language of creating unity and moving Zimbabwe closer to a one-party system more appropriate for an Afiican state (Sylvester, 1995). For the most part, there was not objection to the idea of unity until a major corruption scandal was seen as symbolic o f the growing abuses of power.*"*
The next period investigated in this study is defined by the rise of opposition to the idea of a one-party state in 1990. In line with case criteria for this study, this episode was indeed a battle over the future direction of Zimbabwe and over Zimbabwe’s very identity.
As one scholar (Sylvester, 1990, p. 387) familiar with the period notes about ZANU
(PF)’s and ZUM’s 1990 election manifestos, they both “indicate immediately that new
Zimbabwean histories are being scripted and contested. For ZANU (PF) the years of liberation struggle can be regarded as one continuous saga of political unity. For ZUM the important history of the country begins after independence and it is a story of political fi-agmentation within ‘unity.’”
Moyo (1992, pp. 1-2) lists several other reasons why the 1990 general and presidential elections were so pivotal and the most important since independence. Three in particular are significant for this study; the ZANU (PF)ZPF-ZAPU merger, finalized in
December 1989, promised to end the electoral trend of dividing the voters along tribal lines; all Lancaster House provisions giving safeguards to whites would expire three
Known as the Willowgate scandal, party ofBcials were found to have made profits after reselling cars they were able to procure at low prices because of their positions (Sylvester, 1991, p. 86). 243 weeks after the election and could be amended by the Parbament, potentially making race
an issue again (Lancaster House bad made race pobtics a non-negotiable issue); and
because another party rose in protest to the momentum toward a one-party state, “the
elections became a matter o f‘voting for democracy.’”
5.5.1 The Competitors and the Campaigns
In response to the acceleration toward a one-party system and growing corruption
among ZANU (PF) officials, Edgar Tekere (who bad been expelled from ZANU (PF) in
October 1988 for bis outspoken criticism) formed the Zimbabwe Unity Movement. The
party was started in April 1989 to contest a July by-election in which it won 30% of the
vote (Sachikonye, 1991, p. 45). The party showed early success in mobilizing people
against ZANU (PF) propaganda that Zimbabwe was an integrated state that does not need the “evils” of a multiparty system. For example, in July 1989, he spoke at the University
of Zimbabwe and inspired student protests so disruptive that the government ordered the closing of the university.
This early success was impressive to observers of Zimbabwe’s politics at that time.
Africa Confidential printed a feature pointing out the sectors of society in which ZUM found support. These same sectors supported ZUM in the 1990 elections also, according to Moyo’s (1992) survey, with the possible exception of the trade unions who were
“bought out” by Mugabe in 1990 (see below). These responsive groups included the business community, who feared the one-party state will “deter foreign investment and water down recent economic liberalization policies;” unions, because they agreed with
244 Tekere’s argument that the government should not repress dissent and because they felt
that recent economic policies were a sell-out to neocolonial capitalism; peasant farmers,
because Tekere condemned the failure of Mugabe’s government to deal with the land
issue, and he was against legislation that interferes with the authority of traditional chiefs;
students and intellectuals, because of Tekere’s stands against corruption, troops in
Mozambique, and high government spending; and the p ovo or the urban poor (who were
hardest hit by price increases on staples), because of Tekere’s promises to improve the
economy and adopt policies to distribute wealth more equitably. It was also expected that
Tekere would draw support from ZAPU members who disapproved of the unity pact
between PF-ZAPU and ZANU-PF {Africa Confidential, 1989, p. 5).
In the 28 March 1990 general and presidential elections, the Zimbabwe Unity
Movement won a relatively large following, at least for a near-one party state (Sylvester,
1990). It appears that—despite all the disincentives put in place by Mugabe’s authoritarian style—Tekere’s ZUM did something right.On a platform which included the claim that the status quo was unrepresentative of the evolving Zimbabwean society in the 1990s, ZUM won 17% of the popular vote and was supported by the student movement, the business world, and especially urban voters (support was 30% among urban voters). In a country where the population is still largely rural, ZUM’s support translated to only 2 seats to 117 for the united ZANU (PF) in a wirmer take all vote, but in
Since 1990, one scholar argues, a strong civil society has continued to emerge and has increasingly challenged Mugabe’s one-party goal (see Sithole, 1997). 245 a proportional representation system ZUM’s vote numbers would have given the party 20 seats (Moyo, 1992; Sithole, 1995, p. 145; see also Sithole, 1997, p. 131, Table 1).'®
Observers argue that further evidence for voters not responding to Mugabe’s appeals is the turnout: only about 54% of the voters went to the polls, which was down
20% from the previous election (for example, Sylvester, 1990). One observer noted the following about this election: “First, almost one voter in two chose to abstain; second, from those who made the trip to the ballot box, a fair percentage did not hesitate to vote for the opposition in spite of a virulent campaign...” (Quantin, 1992, p. 39). Further support for the argument that Tekere’s movement exerted remarkable influence was evident months after the election at the September 1990 ZANU (PF) party congress when the party oflScially dropped plans to legalize a one-party system (although intemational and regional pressures also had a lot to do with a move away from the one-party idea).
Finally, it may be argued that the influence of ZUM was also visible when, in his inaugural speech, Mugabe pledged to pull troops out of Mozambique as soon as possible and step up the efforts to give land to the people (Bennett, 1990).
In the words of the group of prominent Zimbabwean scholars (Moyo, 1992, p. 16) who conducted public opinion surveys during and after the campaign period, the elections marked a “significant shift in voters’ opinions on the important question of the type of system of government preferred by Zimbabweans”; it was “ ...a critical shift of opinion.”
This 1990 case is also included in this study because observers at the time
Interestingly, Sithole reinterprets this data from his 1995 chapter in a 1997 article, where he then says that this 17% swing to Tekere shows that he was "badly defeated.” For our purposes, this 17% was rather significant 246 and Mugabe himself anticipated ZUM’s challenge as very threatening—they found it realistic that ZUM could pull significant numbers of voters away from Mugabe. In describing a circumstance in part created by ZANU (PF)’s worry, Zimbabwean scholars
(Makumbe, 1991, p. 182) have called the 1990 elections the most viciously contested elections in the nation’s history, as well as the “highest level of politicisation [sic] of the people of Zimbabwe.”
In his campaign, Mugabe emphasized the need for unity and the threats posed to
Zimbabwe’s peace by “dissident elements.” One television commercial showed the smoking ruins of a damaged car with the caption, “If you vote for ZUM, this is what you can expect” (Moyo, 1992). In a country where the security forces still acted under the
State of Emergency legislation, that a ZUM vote would mean death to the voter was not very far-fetched.
At the newly unified party congress, held in December 1989 just a month prior to the start of election campaigning, Mugabe continued his rhetoric that Marxism-Leninism must be the ideology to guide the newly unified party and that becoming an ofScial one- party state was the best way to achieve this goal. The ofiScial campaign platform emphasized ten years of development in health, education, transport, and mral development, and called for future unity, peace, and development. Perhaps because of the growing support for ZUM among voters in earlier by-elections, the idea of the one-party state was hinted at but not explicitly mentioned in ZANU (PF)’s election manifesto.
Mugabe and his party also promised to increase the rate of land purchases fi"om whites so that more black Zimbabweans could be resettled, since in 1990 they would not be
247 constrained by the Lancaster House constitution. Failure to do an adequate job on the issue of resettlement from 1980 to the present was blamed on the constraints of the independence agreement (see the manifesto, ZANU (PF), 1990).
Not surprisingly, Mugabe had far superior mobilizing resources and stepped up his efforts to take advantage of them during this campaign. He attended on average more than four rallies a day to deliver his message around the country, much of which involved denunciation of Tekere and promises of new programs to benefit each area in which he was speaking. Still, despite the fact that Mugabe held so many rallies, the signs of plummeting popularity were prominent; for most of the rallies, the attendance was shockingly sparse, or shocking to the party at least. Numbers for the summer 1989 election are exemplary: at one stadium, booked for 40,000, only 8,000 people attended
{Africa Confidential, 1989). In 1990, Mugabe officially launched his campaign at a rally with special guest Nelson Mandela, who had recently been released from prison in South
Africa. Only 10,000 people (mostly schoolchildren) attended, compared to the 200,000 people who greeted Mugabe’s return from exile in 1980 (also on this issue of rally apathy, see Moyo, 1989, p. 9; Sakaike, 1990).
There were other ways Mugabe made use of basic mobilizing resources. He made
“deals” with several interest groups, such as the Confederation of Zimbabwean Industries and the Commercial Farmers Union (see also the Herald, 5 and 15 March 1990; and
Quantin, 1992, p. 30). Mugabe employed state resources to campaign for his party, including the use of state buses to bring masses of people into Harare for some of the rallies. During the rallies, little was said about the one-party state idea or socialism. The
248 focus on Tekere was that he represented the forces of racism, largely because ZUM
forged an alliance with the Conservative Alliance of Zimbabwe, which consisted of former
Rhodesian Front members. Another frequent theme was that Tekere’s “unity movement”
was actually threatening the unity of the nation, a unity that had taken so many years of
struggle to achieve. As noted above, Mugabe also brought Nelson Mandela to campaign
rallies; this move had much symbolic power since Mandela had just been released from
prison and was the intemational symbol of liberation struggles.
Mugabe also had the state-run newspapers, radio, and television stations at his
disposal. Purchase of media (television and radio) time was allocated based on party
representation in the previous parliament, thus the incumbent mling party was allowed up
to thirty minutes per day and opposition parties only four minutes per day. For this
project, much of this media and newsprint was collected and content analyzed. From the
period of January 1990-March 1990, the media contained (1) from 1-4 articles p er day on
Mugabe’s work with regional and international actors, and (2) from 3-5 daily articles
about Mugabe’s campaign and his condemnation of Tekere for trying to divide Zimbabwe
or bring violence to the people. The first categoiy above includes working for peace in
Mozambique and Angola, standing up to US policies in these countries, and engaging in
relations with the Organization of African Unity and Commonwealth. More specifically,
Mugabe was portrayed as a major player in the fight against apartheid; and in 1988, he
also won the Afiica prize for Leadership for the Sustainable End to Hunger (a US-based award) for Zimbabwe’s work within SADCC (the regional organization designed to develop economic infrastructure; see below). At the time of these elections, Mugabe had
249 also just finished a three-year term as Chairman of the Non-Aligned Movement. It was
this kind of “clout” as “Statesman” that Mugabe was able to emphasize during his
campaign.
Tekere, as had Mugabe, emphasized unity with the implication that Mugabe’s
brand of unity was artificial. The manifesto states; “...W e need the fuU participation of
every Zimbabwean to seek solutions to Zimbabwe’s problems. An exclusive policy just
will not do” (quoted in Sylvester, 1990, p. 390). Also on the issue of unity, while
Mugabe assailed the alliance that ZUM formed with the Conservative Alliance of
Zimbabwe (the main white party), Tekere’s response was that he and ZUM were
overlooking the race issue by working with whites directly. The main emphasis in the
manifesto was the one-party state debate; ZUM also argued that the new system of a
strong presidency should be replaced by a more limited prime minister. Tekere argued
that a one-party system would not reflect the diversity of people who constitute Zimbabwe
and that it is a shield for corruption by ruling party officials.*^ He promised to repeal the
repressive State of Emergency also (SAPA 1990).
Even Mugabe’s foreign policy was a target for Tekere’s platform, as he argued that the support for the Mozambique government and retaliation from the MNR rebels was making victims of too many Zimbabweans. He promised that if elected president he would withdraw Zimbabwe’s troops firom that quagmire (Sylvester, 1991, p. 170). The
’ ' One criticism of Tekere’s campaign was that he did not offer many details about his alternative system to a one-party state, tending to emphasize only the need to promote democratic principles and practices (Rukobo, 1991, p. 128). 250 MNR rebels stepped up attacks on Zimbabwe around election time and in February 1990
killed four children and three adults. These circumstances may have made Tekere’s words
seem welcome.
In terms of the economic problems of the country, Tekere’s prescription had an
intemational focus. He argued that the best solution to many of Zimbabwe’s development
problems was foreign investment, as well as commitment to the free enterprise system
(Rukobo, 1991, pp. 127, 129). Although Tekere pointed out Mugabe’s shortcomings on
the issue, answers to the land resettlement problem remained unbroached. It is thought
that the economic aspects of ZUM’s message hit a chord with voters because Tekere
emphasized that these widely shared grievances, such as rising unemployment, rising cost
of living, and continuing land hunger despite Mugabe’s promises that one of the central
goals for which the war was fought, would be addressed by a ZUM government
(Sachikonye, 1991, p. 46).
Tekere was faced by multiple obstacles, many of his own making. As noted above,
the party formed an alliance with the all-white CAZ in what they argued was a gesture of
true racial reconcihation. Given the ties between CAZ and the former Rhodesian Front,
including the influence of Ian Smith, the CAZ-ZUM alliance was vulnerable to Mugabe’s
argument that electing Tekere would bring to power a neocolonial regime and forfeit
everything for which the African majority had fought. Radio and television broadcasts by
the ruling party alluded to Tekere’s link with the CAZ “Satan” and sometimes raised the
point that ZUM was a made-up front for the CAZ—and the return of white rule
(Sylvester, 1990, p. 396).
251 On his side, Tekere was a hero of the independence war, but he also had personal shortcomings. He was known to have killed a white farmer in 1981, though he had been acquitted on a technicality, and he was known as a “hard-drinking malcontent who may have been free from the stain of corruption but did not seem cut from presidential cloth”
(Sylvester, 1991, p. 89). Perhaps this evaluation makes the amount of support his campaign garnered even more remarkable but also partially explains why he did not have the charisma to sweep Mugabe out of power.
Other obstacles to Tekere and ZUM can be conceptualized as mobilizing resources; for the most part, these obstacles were put up by ZANU-PF and its control of the state. While Mugabe’s party was able to hold large numbers o f raUies and use state buses to bring rural dwellers to them, Tekere had great diflnculty acquiring authorization to organize. ZUM was successful with several rallies at the University and with street- comer assemblies (Sylvester, 1990, p. 397). Reports show that ZUM supporters were arrested and prosecuted for distributing posters and writing slogans (Moyo, 1992). In the end, spray-painted mailboxes and bus stops were the most visible signs o f ZUM’s campaign (Sylvester, 1990, p. 394). There was extensive violence against members of the opposition; one candidate who was expected to win against a prominent ZANU-PF member was shot and seriously wounded.
Further, ZUM and Tekere had little media access. The media content analysis for this study revealed that any coverage of ZUM in the H erald or the Chronicle was almost all negative; any statements made by Tekere were recounted only piecemeal (which made finding data for the content analysis very difficult). Most of the articles were about former
252 ZUM candidates who had “seen the light” and withdrawn their support for the opposition party (for example, the Herald, 27 March 1990). There are two television channels, and both were used to promote Mugabe; one debate between Tekere and Mugabe was aired.
Finally, the party had no permanent ofiBces, did not organize in many rural areas, and had only small campaign funds even though they were accused of receiving contributions from
South Africa (Sylvester, 1990, p. 397). The next section analyzes the domestic and intemational context variables.
5.5.2 The Context
The prediction for each pohtical opportunity variable is stated in Table 5.4.
Because two variables point toward the success of the leader using more inclusive strategies and two point the other way, the overall prediction for this case is indeterminate. In other words, while each variable does indicate what kind of strategies should be expected to win, there is no way to judge what the prediction should be if all the variables are taken together.
253 Change in stability of elite alignment; Colonial-Controlling (A) Looser Controlling-Regional (C) Looser Strategy predicted to win Inclusive
Repression: Communitj" Higher Subgroup/party Higher Strategy predicted to win Exclusive
Involvement/Mediation Absent from Intemational Community Strategy predicted to win Exclusive
Regional Integration Increasing Strategy predicted to win Inclusive
Overall prediction for Indeterminate Winning strategies
Individual strategies favored Governance, Injustice, Enemy Image
Table 5.4; Summary of Expectations for Zimbabwe 1990
Chatige in Stability o f Elite Alignment
The relationship between the controlling power (now Mugabe) and the regional power (apartheid South Africa) turned around once independence was achieved and South
Africa became the primary nemesis of Zimbabwe. The alignment between the (former) colonial power and the controlling power loosened even more in this 1990 mobilization episode, creating a political opportunity for the leader with more inclusive strategies.
The change in alignment A is straightforward. The relationship between Britain and Zimbabwe was stronger in the immediate period after independence because of the lasting restrictions on the government imposed by the British in the Lancaster House
254 constitution. Further, Mugabe’s government wanted to have a working relationship with
Britain because of the development fund guarantees. However, in 1990, the Lancaster
House constitution was about to expire, and Britain had already not lived up to its
promises in terms of the extent of economic assistance for land resettlement (Dashwood,
1997). British disengagement from Zimbabwe was exemplary of a more general trend of
its withdrawal from the affairs of the continent (Ravenhill, 1991, pp. 186-188).
Evidence about the relationship between the regional power and the controlling
power was clear: South Africa and the principles it espoused were Mugabe’s worst
enemies. For example, one of Mugabe’s major foreign policy initiatives was support for
the Marxist FRELEMO government in Mozambique, against the South Africa-supported
MNR. The following spending figure shows the extent of Zimbabwe’s effort: In the first
half of the 1980s, Mugabe’s government spent up to US$500,000 per day to support
between ten and twelve thousand Zimbabwean troops whose mission was to get rid of
MNR sanctuaries and protect the vital ports in Mozambique (Sylvester, 1991, p. 170).
Zimbabwe has also given strong support for South Africa’s ANC (African National
Congress), allowing members of the liberation organization to operate in Zimbabwe. In retaliation for supporting FRELIMO and the ANC, South Africa initiated commando attacks on the ANC offices in Zimbabwe and armed former Rhodesians to engage in sabotage in Zimbabwe (Weitzer, 1990, p. 161).
A second major tack the Mugabe government took to confront apartheid South
Africa began a few years before the 1990 elections: bringing together regional leaders,
Mugabe wanted to impose mandatory, wide-ranging sanctions against Pretoria, especially
255 after South Aftica's anti-apartheid crackdown in 1986. Because of lasting dependence on
South Aftica, his plan was never implemented (Sylvester, 1991, p. 172). Nonetheless, his efforts symbolize the rift between these two states. It is also interesting to note that when
De Klerk began to make conciliatory moves toward the Front Line States in late 1989 and early 1990 (especially his February 1990 address; see below), Mugabe’s media campaign and his election speeches emphasized that the anti-apartheid struggle was far from over and South Aftica was still the major enemy of Aftica.
Repression
In 1990, the state’s capacity and propensity for repression on the community as a whole and especially on any political opposition had increased from the brief respite in later 1979-early 1980, favoring more exclusive strategies.
Earlier declarations of the nationalist groups had stated that upon independence, the repressive security regime would be reformed and the enacting legislation would be repealed so that democratic freedoms would be granted to all. In reality, the Mugabe government has maintained the same system (see Weitzer, 1990 for an extensive study of the continuity in the security regime). As Ncube (1991, p. 159) states, “The Rhodesian linchpin of repressive government which has the effect of criminahsing [sic] ordinary politics, the Law and Order (Maintenance) Act^ has been retained intact.” Further, the
State of Emergency and the powers granted to the government when there is a State of
Emergency—basically rule by executive order—have been renewed every six months since
1965 (it was finally lifted in July 1990). Unlike the brief interlude between the
256 Lancaster House settlement and the independence elections, in 1990 there was no international presence to mediate the statutory provisions for the highly repressive security regime.
Under security legislation, the government has detained thousands of citizens without trial. Throughout the years, Mugabe has justified the need for this situation because of the South African threat, manifested in (1) Pretoria’s support for the
Mozambican rebels (MNR) who operated from Zimbabwe’s soil and eventually declared war on Zimbabwe; (2) alleged South African support for the insurrection of ZAPU dissidents from independence until the mid-1980s. More immediately, in the 1990 election period and the year preceding it, the government’s favorite targets were alleged to be political opponents (Ncube, 1991, p. 159).
Repression of political action against segments o f the wider community was visible in particular situations as well as in the general atmosphere promoted by the regime, which was described above. When students at the University of Zimbabwe protested against government corruption, the regime cracked down and criminalized dissent. Their application for permission to demonstrate was turned down by the police, and in response to peaceful marches the police assaulted and tear-gassed the students and detained students and a number of “subversive” professors. After numerous other incidents, the government announced the closure of the university (Ncube, 1991).
For Mugabe government actions against "dissidents” in Matabeleland (ZAPU stronghold) in the early 1980s, Amnesty International, Aôica Watch, and the Lawyers Committee for Human Rights condemned human rights violations perpetrated by the security forces (Ncube, 1991, p. 161). 257 Within this context of continued societal repression, repression on the opposition party was also a strong force. It was noted above that ZUM candidates were consistently denied the opportunity to hold ralhes and meetings while ZANU (PF) launched a rally campaign more ambitious than ever. Widely reported intimidation by members of the
ZANU (PF) Youth League allegedly prevented even less grandiose modes of campaigning in many areas. Finally, violence directed against candidates and against anyone claiming to be a ZUM member was reported. The most notable case was the shooting of ZUM candidate Patrick Kombayi, who was seen as a serious challenge to the seat of ZANU
(PF) Vice President Simon Muzenda (Moyo, 1992, pp. 76-78).
The analysis of media coverage in both the H erald and the Chronicle (Bulawayo) showed that the government and its media emphasized the threats to law and order posed by ZUM supporters, as if to justify the repression that was being used against them. From
January-March 1990, only about three articles were simple “announcements” of ZUM’s candidacies or manifesto, compared to 10-12 articles denouncing the problems ZUM was causing. Even the Mugabe campaign reflected the wish to violently oppose Tekere’s movement. It was noted above that television advertisements showed wrecked cars and displayed the message that death would come to you if you voted for ZUM. At one rally,
Mugabe (quoted in Ncube, 1991, p. 164) said, “We are saddened that there are others who want to divide us. But people must not listen to small, petty little ants which we can crush.” Another observer (quoted in Sylvester, 1990, p. 395) reported him as saying if the whites “rear their ugly terrorist and racist heads by collaborating with ZUM, we will chop
258 that head off.” Coming from someone else, these phrases might seem like figures of speech. With the force of the state behind him and the continuing arrests of ZUM supporters, Mugabe’s words were ominous.
International Involvement^ediation and Regional Integration
The role o f the international community in promoting more or less exclusive appeals was important in several ways in 1990. After the elections o f 1990 (late 1990 and after), observers in the West and in other Afiican states increasingly saw the question of one-party rule versus multiparty democracy as an international issue. However, actual involvement of these states in promoting multiparty democracy was minimal in March
1990, reflecting the growing marginalization of Afiica in international politics (Harbeson and Rothchild, 1991). This section will discuss that events in the international arena led to increased perception that Zimbabwe after the 1990 elections was a situation of more international concern, but for our purposes in the 1990 mobilization shift, the situation was treated as an internal issue. This means that this variable favors exclusive strategies; it also means that the second research question need not be asked. In this case, it is interesting to consider the evidence related to the second question nonetheless: Was international involvement neutral or sympathetic to one side over another? Although there was no significant international involvement in support of either leader, research shows that particular events and general opinion was more sympathetic to Tekere’s calls for multiparty democracy than to Mugabe’s plans for a one-party state.
Several factors in the world arena promoted the idea that Mugabe’s vision was not a good one. As Moyo (1992) notes, the internal opposition in ZANU-PF and the growth
259 of ZUM was “given further impetus by developments in eastern and central Europe.”
What was happening in the world contradicted some of Mugabe’s ideas about governance.
Social movements were bringing down Romania, Hungary, East Germany, and
Czechoslovakia, for example, protesting against the evils of socialism and one-party
rule—exactly the two aspects Mugabe was promoting in Zimbabwe. Tekere (quoted in
Morrison, 1990) emphasized this comparison in his campaign; “Ceausescu and his wife
are gone. Our Mugabe here is still preaching Marxism-Leninism.” As these events
occurred, new norms were taking hold in the international community that competitive
democratic elections were the new yardstick by which potential aid recipients would be measured; most action that exemplified these norms was delayed until after the
February/March 1990 election period however (see Harbeson and Rothchild, 1991;
Herb St, 1991, pp. 172-173). Mugabe tried to shrug off these external events, arguing that the socialists in Eastern Europe had imposed a system on the people and that Zimbabwe’s socialism would be from the people, in line with African values (SAPA, 1990). Along with Mugabe’s statement to this effect in his 1990 New Year’s Message, many articles and editorials in the H erald and the Chronicle publicized statements to this effect or put forth this argument (for example the H erald, 9 December 1989; the C hronicle, 20 December
1989; Mathema, 1990).
A second international factor made Mugabe’s rebuttal seem ineffective:
In other Afiican countries including Zambia, Kenya, Tanzania, Angola, and Benin, the political systems had either moved officially toward stronger multiparty systems and away from the idea of a one-party state, or were engaged in efforts to do so under pressure from
260 new World Bank and IMF conditions promoting better “governance” in exchange for
economic assistance (see below) (Harbeson and Rothchild, 1991, pp. 2-3). As one scholar
(Makumbe, 1991, p. 186; see also Moyo, 1992, p. 32) noted, this fact may have made
“...any reasonable Zimbabwean think carefully before embarking on the worn-out road to one-party rule.” Still, Mugabe argued that multipartyism was dangerous for Zimbabwe and the idea was an example of neocolonialism: “A multiparty system in our circumstances is nothing but a disastrous way to the doom of our nation... Why must we listen to those who only yesterday were our oppressive coloniahsts, as they now seek to be our mentors of democracy...” {Reuters, 6 April 1990).
Despite Mugabe’s efforts to frame the international trends in his favor, both events and policies in other states would eventually work against people believing his way of representing the problem (though it is not clear that the mobilization episode in 1990 was affected). In addition to the regional transitions to multiparty systems, Britain and
Sweden, major aid donors to Zimbabwe, began to indicate that future aid to Africa would be linked to democracy—to be measured on a multiparty scale {Africa Economic Digest,
5 November 1990). The fact that African states not conforming to the multiparty model would be increasingly marginalized was widely recognized and discussed among the
Zimbabwean elite (see Southern Africa Political and Economic Monthly, Volume 3).
Also, because of Mugabe’s repression of dissent and de facto one-party rule, human rights groups condemned Mugabe’s plans to eliminate multipartyism. In December 1989,
Human Rights Watch issued a report condemning the emergency legislation still in place in
Zimbabwe, as well as the harassment of all political opposition. The report specifically
261 pointed out the Mugabe government’s intolerance of ZUM. This campaign by non
governmental organizations was part of the international context in which the leadership
competition occurred {Africa News, December 1989).
Another international factor affecting receptivity o f particular appeals involves
South Africa; changes in this country prior to and during the 1990 elections were both
important. As Moyo (1992, p. 32) notes, Mandela’s release and the National Party and
ANC’s subsequent commitment to multiparty democracy (not to be fully realized for some time though) was “even more damaging” to Mugabe’s vision of Zimbabwe’s future than the move toward multipartyism in other Afiican countries. In his opening address to
South Africa’s Parliament, F.W. de Klerk (who had replaced Botha in August 1989) armounced that the ban on the Afiican National Congress, the South Afiican Communist
Party, and several other anti-apartheid organizations were being lifted; that Mandela and some other political prisoners would be released; and that the State of Emergency would also be lifted (Center for Afiican Studies, 1990, p. 3). At a time when Mugabe’s opponent was emphasizing the necessity of similar measures in Zimbabwe, these much- publicized changes in South Afiica were likely to have an impact on voters’ perceptions.
The changing orientation of South Afiica is predicted here to have been important in another way as well. Mugabe has always played up external threats to convince
Zimbabweans to accept that unity is a life or death situation; since his rise as a political leader he has tried to rally people around the idea that Zimbabwe faced a hostile world of
Northern neocolonialism (represented in part by South Afiica), South Africa, and the
MNR ready to take advantage of any division in the country (Schütz, 1990, p. 445). The
262 regional peace initiatives in which South Africa was engaged (for example, the independence of Namibia had just recently been negotiated) and the inevitable coming to power in South Africa of a partner in the struggle was soon to change everything, especially one of the severest threats (but not until months after the Zimbabwe elections).
Also, despite the rhetoric, Zimbabwe was more in need of US aid; railing against neocolonialism at the same time may have not made sense. And for the threats posed by
Mozambiquan rebels (MNR), many people in Zimbabwe were tired of the fighting and loss of Zimbabwean lives.
Finally, international pressure from donor countries other than Britain and Sweden was increasing, as the IMF and World Bank were in the initial stages o f instituting
“political liberalization” conditionality requirements. The consensus among members of these multilateral organizations was that a one-party state was an unacceptable candidate for economic aid (Callaghy, 1991). Indeed, while economic liberalization had been the major focus of conditionality, in 1989 the World Bank issued a widely publicized report about long-term development in Africa: Sub-Saharan Africa: From Crisis to Sustainable
Growth (World Bank, 1989). Among other goals, the report points out that African governments must move toward liberalization of governance structures if there is to be successful economic development. The following excerpt (quoted in Callaghy, 1991, p.
58) summarizes the new “project” the World Bank was launching:
Efforts to create an enabling environment and to build capacities will be wasted if the political context is not favorable... Ultimately, better governance requires political renewal. This means a concerted attack on corruption from the highest to lowest levels. This can be done by setting a good example, by strengthening accountability, by encour aging public debate, and by nurturing a free press. It also means...
263 fostering grassroots and nongovernmental organizations (NGOs), such as fanners’ associations, cooperatives, and women’s groups.
This pressure for “good governance” (which sounded much like Tekere’s platform) from multilateral organizations and individual donor countries would grow stronger after the elections and during the adoption of the structural adjustment program (Makumbe, 1991, p. 187).^® In the period investigated in this study, international intervention is still at enough of a minor level to be considered “neutral” and favoring exclusive strategies. If we were to consider the second research question, indirect events in the international arena tilted the balance toward Tekere, though there was not support for Tekere in particular.
Regional integration; Since Zimbabwe’s independence, both political and economic regional integration in southern Africa have been on the rise. Compared to the
European Union, Southern Aftica’s situation may be better labeled “cooperation” than
“integration.” Nonetheless, the increase in regional economic and political initiatives since the previous period means, in terms of the framework developed in Chapter 2, that this variable creates political opportunities for a leader using more inclusive strategies. The evidence of the increase in regional integration is summarized here.
Early impetus toward economic cooperation came from the effort to build the economies of the Front Line States and make them less dependent on trade with apartheid
South Africa (see statement of purpose, quoted in discussion of the 1980 case, above).
Notwithstanding this evidence about the ESAP and pressures for a multiparty system, in this period US-Mugabe relations seemed good. For example, relations with the US had been rocky as Mugabe criticized the American stance on apartheid and US involvement in Angola. However, in August 1988, the United States had granted US$17 million to Zimbabwe to stimulate rural development and private business (Sylvester, 1991, p. 174). 264 On a continental level, another important goal has been to create economic self-rehance
and pohtical bargaining power among Afiican states within the greater world economy
(Lancaster, 1991, p. 253). Indeed, the 1980 Lagos Plan of Action called for an Afiican
Common Market to be established by the year 2000, and it was hoped that progress
toward such a goal would occur through the progress made by regional organizations
(within Afiica), such as SADCC, the PTA and ECOWAS^" (Mandaza, 1988, p. 11).
SADCC was not a common market as in Europe, but was an effort at regional
cooperation to develop economic infrastructure. Areas of suggested cooperation included
the creation of a Southern Afiican Transport and Communications Committee;
establishment of an agricultural research station; a regional development fund with
permanent administrative machinery; harmonization o f industrial and energy policies; and
development of measures to control foot and mouth disease (Grundy, 1991, p. 107). As
Sylvester (1991, p. 168) notes, “It [SADCC] would circumvent the trade routes,
technological reliance, and development plans South Afiica sought to sohdify around itself
and therefore preserve the independence and sovereignty of all signatories and reduce
redundancy in separate national development efforts.” Within SADCC, Zimbabwe has
been in charge of the food and agricultural sector, and is largely seen as a good member
even though at times Mugabe’s government has made some internal development
decisions that oppose the spirit of SADCC (Sylvester, 1991, p. 169).
Another sign of increasing economic integration in the region is the Preferential
Trade Area (PTA; formed in 1981), which promotes reciprocity in tariff reductions and
The full names are Southern Africa Development Cooperation Council, Preferential Trade Agreement, and the Economic Community of West African States, respectively. 265 other agreements promoting free trade among its nineteen member countries. Zimbabwe runs the currency clearinghouse, though in 1990 it still had not been utilized to its envisioned capacity. The performance of the PTA in achieving economic integration has been very limited compared to the EU example in large part because of the position of
African states in the global economy (see Nyang'oro and Shaw, 1998), but its creation and endurance shows that the member states do have the goals of greater integration
(Lancaster, 1991, p. 259).
Some observers think that integration in Southern Africa could increase in the future because of global changes, though others disagree because the short-term costs outweigh the benefits that only come long-term (Lancaster, 1991, p. 263). On the positive side, with declining trade and investment relationships with Europe, the Soviet Union, and the United States, Afiican countries such as Zimbabwe may have to act on the growing importance o f the regional economic arena (Nyang'oro and Shaw, 1998; Ravenhill, 1991).
As noted in discussions of the previous cases, political cooperation in the region has existed since the emergence of the nationalist movements in the 1960s. Since
Zimbabwe became independent, political integration on a more formal scale increased with regard to some foreign policy decisions. Through SADCC and the FLS, the southern
African states (except South Africa) cooperated on issues such as the war in Angola,
Namibia, and Mozambique (Grundy, 1991, pp. 106-107). Although incorporating more members than SADCC, the Organization of African Unity has been another political grouping to which independent Zimbabwe has belonged. Though not granted authority to make binding decisions, the GAU was formed as “...an expression of the aspiration of
266 unity among Aj&icans...” which was ...intended to head ofifthe increasing division of
Afiica into ideological blocs...” (Lancaster, 1991, p. 251; see also Nyang'oro and Shaw,
1998, p. 36).
In summary, this significant increase in regional integration means that this variable
may have created a political opportunity for a leader using more inclusive strategies. The
fairly high degree of cooperation promotes and is promoted by a sense of shared identity
in both Southern Afiica and all of Afiica; the fact that economic and to some degree
political fates are tied to the regional organizations further creates the conditions for
success if a leader promotes this more inclusive level of identity. Still, because in this
region integration is not visible to the same (high) degree that it is in Europe, we might
expect this variable to be less significant across cases even though it is high in this case
relative to the other Zimbabwe cases. There is a stark contrast between the common
passports and shared voting rights held by people in European countries to the situation in
southern Afiica.
5.5.3 Expectations about Individual Strategies
Though the overall prediction discussed above is indeterminate, more concrete
expectations can be stated for the individual strategies the winning leader should use most fi-equently. Three major aspects of the situation going into the elections of 1990 are
expected to influence individual strategy use. The first involves the system of political representation in Zimbabwe sketched against global trends. Zimbabwe in 1990 was nearly a one-party state. Through first including Nkomo and other high-ranking ofiBcials of PF-
267 ZAPU in his government, then forging a formal Unity Agreement between ZANU-PF and
PF-ZAPU in 1987, and finally holding the first unity (ZANU PF) party congress just prior to the 1990 elections, Mugabe had advanced far toward his idea of eliminating all forms of political competition. These changes were happening in the regional and global context of democratization and increasing calls for multiparty democracy. In the period leading up to the 1990 elections, Zimbabwe was also on the verge o f being fireed firom all remaining constraints of the Lancaster House constitution. Substantively this meant that the government would be fi^ee to pursue less inclusive policies with regard to the whites; it was no longer required to provide whites with separate representation or to abide by the willing-buyer land guarantee. This set of factors is expected to promote the winning of a leader using Governance strategies—both the Positive and Negative variants.
A second aspect of the context is expected to influence what kind of individual strategies a winning leader would use: repression. Political freedom, especially entering into and during the election period, was remarkably constrained. People were made to feel that they would be attacked if they voted for anyone other than Mugabe and his party.
It was common knowledge that political opponents would not be tolerated. The third important factor is the economic situation in Zimbabwe, especially compared to Mugabe’s promises of years past. Although the situation has worsened since, the Zimbabwean people had not seen their dreams of the revolution fulfilled. Most of the land still remained in the hands of the white, industrialized farms. Little of the land had been resettled by the black peasants, even though Mugabe had based his legitimacy in early years on being able to redistribute farmland. Further, many ZANU PF oflBcials were
268 viewed as corrupt; and whether all were corrupt or not, wealth redistribution was overtaken by the growth o f a wealthy black elite. This situation opened up a need for someone to explain why this sad state affairs had come about. Whether Mugabe (who blamed the Lancaster House constitution for the land issue) or a challenger could come up with the best explanations and vision for the future was a large part of the 1990 campaign.
It is argued here that these circumstances promote a third and fourth strategy; Injustice because of the continuing poor situation and Enemy Image (because of the need for a scapegoat).
5.6 Conclusion
This chapter has described the mobilization sliifts by providing evidence of the shifts, selecting the leaders to be analyzed, presenting the background information about each campaign, and predicting whether each contextual variable provides political opportunities for the leader employing more inclusive or more exclusive strategies. The individual strategies promoted by ± e context are also hypothesized. The following table summarizes the mobilization shifts as described above. The next chapter presents the results of the content analysis.
269 Change in Repression Intemat’l. Regional Overall Winning Stability of involvement integration prediction leader elite alignment 1979 Exclusive Exclusive Inclusive not signif “ Exclusive Mugabe
1980 not signif Inclusive Inclusive not signif Inclusive Mugabe
1990 Inclusive Exclusive Exclusive Inclusive Indeterminate Tekere
^ As explained above, "not significant” means that these variables do not come into play in the case. For both the change in stability of elite alignment variable and for the regional integration variable, a change is necessary for the hypotheses to apply (see discussion in Chapter 2).
Table 5.5; Summary of Zimbabwe Cases
270 CHAPTER 6
RHODESIA/ZIMBABWE:
ANALYSIS OF LEADERSHIP MOBILIZATION STRATEGIES
6.1 Introduction
This chapter presents the content analysis results for the coding of the leaders’ mobilization strategies in the three Zimbabwe cases. Each section in this chapter is focused on a particular case, then a final section compares the strategies across the three cases. The next chapter then discusses the findings firom the Northern Ireland and
Rhodesia/Zimbabwe mobilization episodes comparatively. The structure of the case discussions in this chapter is similar to that of the Northern Ireland chapter: first the predictions are summarized, then the individual strategies used by each leader are evaluated. Finally, each section analyzes and compares the exclusivity of the strategies for each leader so that the predictions from Chapter 5 may be evaluated: did the leader who was the beneficiary of the mobihzation shift employ the more exclusive or more inclusive strategies predicted by the four types of political opportunities?
271 6.2 First Mobilization Episode: Zimbabwe-Rhodesia 1979
6.2.1 Expectations
The previous chapter discussed how two of the three opportunity variables
(change in stability of elite alignment and change in repression) indicated that the winning leader would be the one to use more exclusive strategies. The exception is the international involvement/mediation variable, which favors inclusive strategies and also favors Mugabe in particular. The individual strategies expected to be favored by context are the Injustice, Enemy Image, Storytelling, and Governance Negative strategies.
6.2.2 Comparing Strategies
To evaluate the latter, the percentage of total strategies that each individual strategy was used are provided in Table 6.1.
272 Mugabe Muzorewa (0=851)“ (n=381) Identity (Zimbabwe)*’ 6.5% 10%
Outgroup Negative (Enemy 17.3% 11.8% Image)***
Justice Negative*** 15.3% 16%
Storytelling*** 19% 17.8%
Governance/negative 12.8% 9.7%
Governance/positive*** 13.2% 26.5%
Outgroup Positive (Ally Image) 7.6% 5.8%
Justice Positive 8.3% 2.4%
Total % " 100 100
' The ”n” here is the total number of strategies and the number given in each cell is the percentage of this total. The strategies in bold are the strategies hypothesized to be used most frequently by the winning leaden in this case that is Mugabe. The asterisks denote the actual strategies used most frequently by the winning leader. The chances of the leader using three of the four predicted strategies randomly is .24; according to standards of significance (generally the .05 level), these findings are not significant.
Because Zimbabwe is the focus of the research and considered to be the society under exam ination. this reference is the focus. This figure excludes references to the leader’s party because this is misleading about the groups in the society the leader is defining.
' Due to roimding off of the numbers, these totals do not equal exactly 100%.
Table 6.1: Comparison of Strategies, Shift 1
In this first Rhodesia/Zimbabwe case, Mugabe is considered the more successful leader. The top four most fi-equently used strategies in descending order were as follows:
Storytelling, Enemy Image, and Justice Negative/Injustice, and Governance Positive
(Mugabe); and Governance Positive, Storytelling, and Justice Negative, and Enemy Image
273 (Muzorewa). For Mugabe, his strategy profile in Table 6.1 shows that he and Muzorewa
used the Injustice strategy and the Storytelling strategy at similar fi'equencies. However,
Mugabe did use the Enemy Image strategy more often than Muzorewa while the latter
employed the Governance Positive strategy (nearly one-third of the total). Indeed,
Muzorewa spent a good deal of his campaign speeches emphasizing the gains made by his
transitional government to date and especially the gains in legislating change that would follow his election. In this context, it appears that the audiences were not “ready” or willing to respond to this talk of positive governance when the war and the reasons it was being fought were still ongoing.
Most unexpected was that in this particular case, the winning leader’s (Mugabe) most frequently used strategy was Storytelling. Recall that this strategy is a mode that a leader uses when employing the other strategies, so in fact we find that instead of the emphasis purely on Injustice and Enemy Image, Mugabe uses these strategies in the
Storytelling mode. ' Overall, in this case the results were not particularly good because the leaders used similar strategies (though the ordering was quite different) and only three of the four predicted were used most frequently. In terms of the exact test, the significance level is .24 (as opposed to the .018 or .014 found for most other cases in this study).
' The coding and analysis did not provide for a way to treat the frequency counts of the Storytelling strategy differendy than the other strategies. For this reason, and as noted in Chapter 4, Appendix C presents a second analysis of the strategy frequencies for each case without the Storytelling frequencies included. The effort in this second analysis is to see if this affects the ordering of the other strategies. Also, the question has been raised of whether the Storytelling strategy should be treated as a “more exciirsive” strategy (for the following section). In Chapter 2, this treatment was justified by the way in which the imagined communities Uteratirre discusses the reasons for leaders appealed to the past and future links among members of a “nation.” Further, in all seven cases, a look at the original coding sheets shows that the leaders used the Storytelling strategy almost always with the more exclusive strategies of Justice Negatwe, Governance Negative, and Enemy Image. 274 6.2.3 Assessing the Predictions: The Exclusivity Index
In this section, the content analysis results are evaluated to determine how inclusive or exclusive the strategies of each leader was; in the first case this is examined relative to each other. In the second and third cases, the scores are compared to the previous case, and therefore two questions are asked; which leader is more exclusive and did the leaders become more or less exclusive from the previous episode?
Nature o f Strategies
As described in Chapter 2, the Exclusivity Index scores the leaders on the nature of the strategies they use and on the targets of these strategies (which are split into two scores, outgroup targets and ingroup targets). The first table separates the strategies into those which are more exclusive and those which are more inclusive. For example. Enemy
Image strategies are expected to promote more exclusive ideas and behavior than Ally
Image strategies. For the Identity strategy, the focus is on the ingroup references to
“people in Zimbabwe,” since in all cases this reference makes up 78 to 100% of all ingroup references. The materials were coded for the group delineations the leaders used when they talked about the “us” in Zimbabwe. In other words, within that area, are people grouped together in narrowly defined collectives or do most of the people within the borders share group memberships? Within the Zimbabwe references, the “very exclusive” delineations are included with the exclusive strategies and the “very inclusive” delineations are placed with the inclusive strategies.
275 More exclusive strategies Mugabe Muzorewa (n=817) (n=361) Identity 1.3% .3% (ZimbabweAer^" exclusive) Enemy Image 18% 12.5%
Justice Negative 16% 16.9%
Storytelling 19.8% 18.8%
Governance Negative 13.3% 10.2%
Total exclusive strategies 68.4% 58.7%
More inclusive strategies Mugabe Muzorewa (n=817) (n=361) Identity 1.2% 4.7% (ZimbabweA^ery inclusive) Ally Image 8% 6 .1%%
Governance Positive 13.7% 28%
Justice Positive 8.7% 2.5%
Total inclusive strategies 31.6% 41.3%
Assessment
Mugabe Muzorewa Exclusivity Ratio “ 2.2 1.42 “This ratio is calculated by dividing the total percentage of exclusive strategies by the total per centage of inclusive strategies. The higher number indicates that the leader used more exclusive strategies than the leader with the lower number.
Table 6.2: Compaiisoa of Exclusive and Inclusive Strategies for Shift 1
On this measure, Mugabe is more exclusive than Muzorewa, largely because
Mugabe uses the Enemy Image strategies as much as 5.5% more than Muzorewa. The other remarkable difference is that Muzorewa uses a particular inclusive strategy almost
276 one-third of the time: Governance Positive. In terms of the Identity strategies, Muzorewa
is also more inclusive than Mugabe. Substantively, this is largely because Muzorewa
often referred to the nation of Zimbabwe in ways that clearly included the white
community (a “very inclusive” reference; 44.7%). Meanwhile, Mugabe did this only about
18% of the time; his largest category was “moderately inclusive” (60%; not shown here)
and comes from his references to the “peasants,” “struggling masses,” and others where he
makes it clear he means the black majority. Mugabe also employs “very exclusive” group
delineations in this period, and substantively these were commonly “freedom fighters.”
Unlike Gerry Adams, in 1979 Mugabe clearly used this group as an ingroup.
Nature o f Targets
Comparing the strategies leaders use only gives us part of the picture about how
inclusive or exclusive the leaders’ appeals are overall. We also have to consider the
targets of these strategies—how the depiction of group identities are combined with
descriptions of a group’s plight and mission. Tables 6.3-6.6 show the results of the
analysis of the targets for the major strategies. For some of the strategies, the target (or subject) of the strategy is the ingroup (Justice Negative and Positive, Governance
Negative and Positive, Storytelling) and for some the target is the outgroup (Enemy Image and Ally Image strategies). Two final scores are summarized in Table 6.6: the exclusivity of the ingroup targets and the exclusivity of the outgroup targets. The final table shows the total scores for each leader on the Exclusivity Index.
277 Injustice Governance Negative
(More inclusive) Mugabe Muzorewa Mugabe Muzorewa (n=85) (n=39) (n=71) (n=29) People in world 2.4% 0 31% 13.8%
People in Africa 20% 7.7% 5.6% 10.3%
People in Zimbabwe 77.6% 92.3% 63.4% 75.9%
Exclusivity ratio ® 3.46 12 1.73 3.15
® Because there are only three references, an exclusivity ratio is used as a comparison measure here (Zimbabwe references divided by the sum of world and Africa references); this is slightly different than the measure used for the more complex Ireland cases, which have eight possible references.
'’Numbers are percentage of all Justice Negative strategies and Governance Negative strategies (respectively) for which this reference is the target.
Table 6.3; Comparison of Targets (Injustice and Governance Negative Strategies) for Shift r
The results above show that for both Mugabe’s Justice Negative and Governance
Negative strategies, he is much more inclusive than Muzorewa. Mugabe is much more likely to include groups in the world and Afiica when he talks about Injustice and how institutional structures and negotiations (Governance Negative) hurt the target group.
Using the example of the Injustice strategy, substantively this means that he argues both that other groups beyond Zimbabwe face injustice and also that those in the ingroup are defined in terms of Afiica and global group references.
The next table compares the leaders' targets for the Enemy Image strategies.
Looking at Mugabe’s exclusivity ratios for the Blame and Threat strategies, he is more exclusive than Muzorewa. This is largely because he tends to refer to the enemy most of
278 the time as the “colonial invader,” and the “settler enemy,” for example. These are global references (with the implication that the enemy is the white man from Britain); for the enemy strategy global references are considered more exclusive because when the enemy encompasses more people the ingroup boundaries are drawn in (more exclusive). Still, looking at the targets for the combination of all the Enemy Image strategies, Muzorewa is more exclusive (Mugabe often accuses those defined by “Zimbabwe” group references of
Betrayal).
Blame Threat Total of Enemy strategies “ (More exclusive) Mugabe Muz. Mugabe Muz. Mugabe Muz. (n=7) (n=8) (n=58) (n=8) (n=107) (n=33) People in world 85.7% 0 62.1% 25% 55.1% 15.1%
People in Africa 0 12.5% 0 25% .9% 15.1%
People in Zimbabwe 14.3% 87.5% 37.9% 50% 43.9% 69.7%
Exclusivity ratio ^ 5.99 .14 1.64 1 1.28 .43
" This percentage includes the threat, blame, betrayal, and outgroup image-ambiguous variants.
With the outgroup strategies, the broader references are considered more exclusive because the "enemy” encompasses more people, making the ingroup smaller and more exclusive. This ratio is the sum of the world and Africa references divided by the Zimbabwe references.
Table 6.4; Comparison of Targets (Blame, Threat, and Combination of Enemy Strategies) for Shift 1
When the targets of the Storytelling strategy are examined, Mugabe is more inclusive than Muzorewa. In this first mobilization episode, Mugabe’s appeals very often mention an “imagined community” called Zimbabwe, hearkening back to the days of Great
279 Zimbabwe (the ruins said to be the seat o f the great empire, still used as a symbol on the currency). He encourages people to honor the memory of freedom fighters from the rebellion against whites in 1896, fighting to finally free the nation so that all the “children of the revolution” who have died over time will not have died in vain. Muzorewa also dreams of an imagined community but not as vividly (this of course is not reflected in the measures). Mugabe is more inclusive because he also uses the imagined communities vision with more references to Afiica or the world, such as “the ancient people of Africa” or “people everywhere who have suffered from the weight of Western colonialism.”
(More inclusive) Mugabe Muzorewa (n=89) (n=42) People in world 7.9% 2.4%
People in Africa 2.2% 4.8%
People in Zimbabwe 89.9% 92.9%
E.xclusi\âty ratio*’ 8.9 12.9
“The target of tlie Storytelling strategy indicates the group reference that the leader uses when he talks about the “imagined communit}".”
Because there are only three references, an exclusivity ratio is used as a comparison measure here (Zimbabwe references divided by the sum of world and Africa references); this is slightly different than the measure used for the more complex Ireland cases, which have eight possible references.
Table 6.5; Compaiison of Targets (Storytelling Strategy)' for Shift 1
Finally, Table 6.6 also shows the way in which Mugabe uses global and Afiica references more frequently than Muzorewa, here with the Ally Image and the Governance
Positive strategies. Mugabe, the winning leader, depicts the situation as one in which
280 many groups of people outside of Zimbabwe (in Afiica and beyond) share the nationalist movement’s goals, while Muzorewa is more likely to focus on others within Zimbabwe as allies. Mugabe also looks to external actors when he talks about the need for positive governance and negotiation.
Shared Goals Total of Ally Governance strategies* Positive (More inclusive) Mugabe Muz. Mugabe Muz. Mugabe Muz. (n=31) (n=10) (n=56) (n=19) (n=76) (n=63) People in world 32.3% 30% 23.2% 15.8% 22.4% 9.5%
People in Africa 41.9% 20% 30.4% 15.8% 3.9% 6.3%
People in Zimbabwe 25.8% 50% 46.4% 68.4% 73.7% 84.1%
Exclusivity ratio .348 1 .866 2.16 2.8 5.32
“This percentage includes the goals similar, situation similar, culture similar and outgroup image- ambiguous variants.
'’This is calculated by dividing the sum of the world and Africa references by the Zimbabwe ref erences. The former are considered inclusive targets because it means the leader is looking outside the political society for allies and help in governance, and is also looking within the society. A liigher number here indicates a more exclusive score.
Table 6.6: Comparison of Targets (Ally/Goals and Combination of Ally Strategies, Governance Positive Strategies) for Shift 1
Summary o f Findings
The totals of the exclusivity ratios for the three categories are given in Table 6.7.
When the exclusivity ratios are weighted by the n’s, the total shows that Mugabe is more exclusive than Muzorewa.
281 Exclusivity Exclusivity Exclusivity Total Leader ratio: Nature ratio: Nature ratio: exclusivity with more of strategies of ingroup Nature of ratio * exclusive targets outgroup appeals targets overall Mugabe 2.2 16.9 10.1 8943 XXX (n=85I) (n=321) (n=163) Muzorewa 1.4 33.4 4.7 6556 (n=381) (n=173) (n=52) Tliis ratio is calculated by first weighting the ratios in the three previous columns by the n for each and then summing the three columns.
Table 6.7; Summary of Findings, Shift 1
According to the expectations described in the previous chapter, the more exclusive leader was expected to win and he did (Mugabe). In this first case, th en , the expectations are supported. The next section discusses the second time Mugabe and Muzorewa competed for potential followers.
6.3 Second Mobilization Episode: Zimbabwe 1980
6.3.1 Expectations
By 1980, the context had changed significantly. The discussion in Chapter 5 showed that both of the two significant opportunity variables predicted that more inclusive strategies are favored. The stability of elite alignment variable did not change, so though it predicted more exclusive strategies, it was not expected to be significant in this case.
However, the lessened repression changed the prediction of that variable from the previous period to the expectation that inclusive strategies would be promoted. Finally, while regional integration is not significant in this period, the international involvement
282 variable again is expected to favor the leader with more inclusive strategies. The
individual strategies favored are Governance Positive, Storytelling, Enemy Image, and
Injustice.
6.3.2 Comparing Strategies
By the election time of 1980, the fighting had ended since the previous period and
was replaced by multilateral negotiations as the method of interaction among the major
players: the campaign period followed international talks at Lancaster House in Britain
that had lasted over three months. The parties had agreed to a new constitution involving
a power-sharing agreement with whites and the bargain that firee Zimbabwe would allow
whites to keep their land (unless they wanted to sell it) in exchange for development funds
from the international community. As elaborated in the previous chapter. Governance
Positive, Storytelling, Enemy Image, and Injustice were expected to be used by more by
Mugabe (the wiimer) than Muzorewa. Table 6.8 presents the percentage of total strategies that each strategy was used so that these expectations can be evaluated.
283 Mugabe Muzorewa (n=387)^ (n=10S7) Identity (Zimbabwe)’’ 3.9% 8.6%
Outgroup Negative (Enemy 13.8% 16.5% Image)*** Justice Negative*** 12.9% 14.4%
Storytelling*** 26% 13.6%
Governance/negative 8.8% 9.5%
Governance/positive*** 14.8% 23.4%
Outgroup Positive (Ally Image) 8.4% 5.6%
Justice Positive 9.8% 8.5%
Total % ' 98.4 100.1%
“ The "n” here is the total number of strategies and the number given in each cell is the percentage of this total. The strategies in bold are the strategies hypothesized to be used most frequently by the winning leaden in this case that is Mugabe. The asterisks denote the actual strategies used most frequently by the winning leader. These results are significant at the .014 level (there is a 1.4% chance of getting all four predicted strategies correct randomly).
Because Zimbabwe is the focus of the research and considered to be the society under exam ination, this reference is the focus. This figure excludes references to the leader's party because this is misleading about the groups in the society the leader is defining.
Due to rounding off of the numbers, these totals do not equal exactly 100%.
Table 6.8: Comparison of Strategies, Shift 2
Mugabe was the winning leader in this second mobilization shift, and it is interesting to note that his strategies changed in the direction expected. Instead of Enemy
Image and Injustice as the most frequently used strategies, Mugabe changes to
Governance Positive (14.8%). Still, as surprising as in the first case, his most frequently
284 used strategy was Storytelling (26%); see footnote 1 above. Third was the Enemy Image strategy (13.8%) with Injustice as the fourth (12.9%). According to the exact test, there is only a very slight chance of getting this result randomly; the findings are significant at the .014 level. Though Mugabe’ strategies fit with the expectations above, when we look at Muzorewa’s top four strategies, he does as well. He employed the Governance Positive strategy most frequently (23.4%), then Enemy Image (16.5%), Injustice (14.4%), and
Storytelling (13.6%). Based on this comparison, it is suggested that Mugabe’s use of the
Storytelling strategy mode when he used the Governance Positive and other strategies added persuasive power. The next part of the analysis will look at the leaders in terms of whose appeals were more and less exclusive and compare the winning type to predictions based on the political opportunity variables.
6.3.3 Assessing the Predictions: The Exclusivity Index
Nature o f Strategies
Table 6.9 compares the nature of the strategies used by both leaders.
285 More exclusive strategies Mugabe Muzorewa (n=373) (n=1043) Identity .8% .3% (Zimbabwe/very exclusive) Enemy Image 14.2% 17.2%
Justice Negative 13.4% 15.1%
Storytelling 26.7% 14.2%
Governance Negative 9.1% 9.9%
Total exclusive strategies 64.2% 56.7%
More inclusive strategies Mugabe Muzorewa (n=364) (n=994) Identity 1.6% 4.4% (Zimbabwe/very inclusive) Ally Image 8.6% 5.8%
Governance Positive 15.3% 24.4%
Justice Positive 10.2% 8.8%
Total inclusive strategies 35.7% 43.4%
Assessment
Mugabe Muzorewa Exclusivitv Ratio ® 1.8 1.3 Change from shift 1 -.4 -.1 “This ratio is calculated by dividing the total percentage of exclusive strategies by the total per centage of inclusive strategies. The higher number indicates that the leader used more exclusive strategies than the leader with the lower number.
Table 6.9; Comparison of Exclusive and Inclusive Strategies for Shift 2
This shows that both leaders are slightly more inclusive than they were in the previous period, but the ratio of exclusive to inclusive strategies indicates that Mugabe is still more
286 exclusive on this measure. As discussed above, this difference comes from the large
differences in the use of the Storytelling strategy (an exclusive appeal) and the Governance
Positive strategy (and inclusive appeal). The next measures, which take a more complete
approach to look at the targets of these strategies, shows that Mugabe is nearly always
more inclusive.
Nature o f Targets
Table 6.10 below considers whom each leader talks about when they use the
Injustice and the Governance Negative strategies.
Injustice Governance Negative
(More inclusive) Mugabe Muzorewa Mugabe Muzorewa (n=31) (n=112) (n=21) (n=78) People in world 3.2% 1.8% 42.9% 10.3%
People in Africa 22.6% 14.3% 4.8% 16.7%
People in Zimbabwe 74.2% 83.9% 52.4% 73.1%
Exclusivity ratio ® 2.88 5.2 1.1 2.7
“ Because there are only three references, an exclusivity ratio is used as a comparison measure here (Zimbabwe references divided by the sum of world and Africa references); this is slightly different than the measure used for the more complex Ireland cases, which have eight possible references.
‘’Numbers are percentage of all Justice Negative strategies and Governance Negative strategies (respectively) for which this reference is the target.
Table 6.10; Comparison of Targets (Injustice and Governance Negative Strategies) for Shift 2 ”
287 For Injustice, it is interesting that once again Mugabe proflfers that (1) those subject to injustice are not only defined in terms of their Zimbabwe identities but are either people in
Zimbabwe whose global or Afiican identities are emphasized by Mugabe, and that
(2) people in Zimbabwe have a common situation of living with injustice with others in
Afiica or in the world. Muzorewa does use people in Afiica as targets 14.3% of the time, but overall he is more exclusive. There are also large differences in the Governance
Negative targets. Nearly 43% of Mugabe’s targets are global; this is because he often talks about the institution of colonialism and the past practices of racism imposed on the people of Zimbabwe as well as the “British bad faith” he argued was exhibited by British support for whites in the interim period prior to elections. For Muzorewa, 73% of his references are to groups or elites in Zimbabwe— often to the socialism and other forms of bad governance that would be imposed if Mugabe were elected. He does make about
17% of his references at the Africa level; these are primarily to the Front Line States and the support they gave to Mugabe. Overall, Mugabe is more inclusive.
The next table compares the targets for the Blame and Threat variants of the
Enemy Image strategies, as well as the targets for all four variants of the Enemy Image strategies combined (Betrayal and ambiguous too).
288 Blame Threat Total of Enemy strategies “ (More exclusive) Mugabe Muz. Mugabe Muz. Mugabe Muz. (n=8) (n=9) (n=15) (n=51) (n=40) (n=147) People in world 62.5% 0 73.3% 9.8% 55% 6 .1%
People in Africa 0 0 0 0 2.5% 5.4%
People in Zimbabwe 37.5% 100% 26.7% 90.2% 42.5% 88.4%
E.xclusivit}’ ratio 1.67 0 2.7 .12 1.35 .13
“This percentage includes the threat, blame, betrayal, and outgroup image-ambiguous variants.
With the outgroup strategies, the broader references are considered more exclusive because the "enemy” encompasses more people, making the ingroup smaller and more exclusive. This ratio is the sum of the world and Africa references divided by the Zimbabwe references.
Table 6.11: Comparison of Targets (Blame, Threat, and Combination of Enemy Strategies) for Shift 2
Mugabe is more exclusive than his competitor for all of these targets. Again, this is
because he often includes groups beyond the borders of Zimbabwe, but with the outgroup
measures these types of references are considered more exclusive instead of more
inclusive. In this case, he often blamed and played up the threat posed by the colonials,
settlers, or foreigners—although the comparison of strategy frequencies showed us that he
does this less frequently overall than he had in the first case. For Muzorewa, his primary
scapegoat and threatening groups/leaders are Mugabe, Nkomo, and the Patriotic Front, all
Zimbabwe targets.
The targets of the Storytelling strategies are presented in Table 6.12. The
comparison of exclusive to inclusive targets shows that Muzorewa is much more exclusive than Mugabe on this measure.
289 (More inclusive) Mugabe Muzorewa (n=47) (n=90) People in world 19.1% 3.3%
People in Afiica 4.3% 6.7%
People in Zimbabwe 76.6% 90%
Exclusivity ratio 3.27 9
“ The target of the Storytelling strategy indicates the group reference that the leader uses when he talks about the "imagined community.”
^ Because there are only three references, an exclusivity ratio is used as a comparison measiu^e here (Zimbabwe references divided by the sum of world and A&ica references); this is slightly different than the measure used for the more complex Ireland cases, which have eight possible references.
Table 6.12: Comparison of Targets (Storytelling Strategy)* for Shift 2
Since Mugabe’s Storytelling strategy use was so frequent, the shift in the inclusivity of his targets is very important. In the first period, the global targets were 7.9% of the total, but this increased to 19.1% in the second case. Combining this with the observations about the group references stated above, Mugabe still tends to refer to whites as not wholly part of “Zimbabwe,” but he has begun to include their presence in the imagined community of
Zimbabwe because most of this 19% are indeed references to the white community.
Finally, Table 6.13 summarizes the targets for the more inclusive strategies. When the leaders talk about groups with similar goals, those in similar situations, or those who do or should represent positive governance, who are these targets?
290 Shared Goals Total of Ally Governance strategies' Positive (More inclusive) Mugabe Muz. Mugabe Muz. Mugabe Muz. (n=12) (n=23) (n=30) (n=59) (n=34) (n=133) People in world 58.3% 39.1% 33.3% 23.7% 14.7% 15%
People in Africa 25% 34.8% 33.3% 16.9% 5.9% 2.3%
People in Zimbabwe 16.7% 26.1% 33.3% 59.3% 79.4% 82.7%
Exclusivity ratio *’ .2 .353 .5 1.46 3.85 4.78
“This percentage includes the goals similar, situation similar, culture similar and outgroup image- ambiguous variants.
‘’This is calculated by dividing the sum of the world and Africa references by the Zimbabwe refer ences. The former are considered inclusive targets because it means the leader is looking outside the political society for allies and help in governance, and is also looking within the society. A higher number here indicates a more exclusive score.
Table 6.13: Comparison of Targets (More Inclusive Strategies) for Shift 2
For the three strategies examined, Muzorewa is more exclusive than Mugabe. This result comes again from Mugabe’s greater tendency to either emphasize the universal aspects of
Zimbabweans’ identities or to connect their plight to allies outside of their own country.
For example, the latter applies to the Shared Goals strategy : 58% of the references to groups with shared goals are international actors.
Summccry o f Findings
The weighted exclusivity ratios for the three sets of results are added in Table
6.14, and show that Muzorewa’s score is higher than Mugabe’s, meaning Muzorewa
291 is judged to be the leader with more exclusive appeals in this second case. It should be noted, however, that both leaders scores moved substantially in opposite directions.
Mugabe became more inclusive than in the previous case, while Muzorewa became more exclusive.
Exclusivity Exclusivity Exclusivity Total Leader Change ratio: ratio: ratio: exclusivity with more since Nature of Nature of Nature of ratio * exclusive last strategies ingroup outgroup appeals period *’ targets targets overall Mugabe 1.8 11.1 6.4 4943 -4000 (n=387) (n=133) (n=70) Muzorewa 1.3 21.7 2.1 14,701 XXX +8145 (n=I087) (n=413) (n=206) This ratio is calculated by first weighting the ratios in the three previous columns by the n for each and then summing the three columns.
^ Here a positive number means the leader’s strategies were more exclusive; a negative number means they became more inclusive.
Table 6.14; Summary of Findings, Shift 2
Comparing these results with the predictions developed in the previous chapter, the expectations are supported in this case: two of three variables were expected to favor more inclusive leaders and Mugabe, the victor here, was indeed more inclusive. There was only one contextual variable that predicted toward more exclusive strategies: change in stability of elite alignment. Because of these results, it appears that because (1) the international community’s involvement favored more inclusive strategies and (2) the repression on the community and the competitors’ parties was less than in the previous period, the “definition of the situation” put forth by the more inclusive leader was
292 supported. In this case, the change in elite alignment that aligned the British more with the Rhodesians (from their previous support for Mugabe’s inclusion in talks) and changed the prediction to favor exclusive strategies did not seem to be significant. In fact, it may be that this change, while theoretically favoring exclusive strategies, actually favored
Mugabe, no matter what kind of strategies he was using. The reasoning here is that
Mugabe’s successful propaganda emphasized the links between the “traitor” Muzorewa and the Rhodesians, and played up the fact that the Rhodesians were trying to assassinate him. The British hard-line with Mugabe may have acted to gain him supporters. The final section of this chapter will compare this case to the others in order to investigate further this anomaly in the predictions.
6.4 Third Mobilization Episode: Zimbabwe 1990
6.4.1 Expectations
In the third Zimbabwe case, the context of leadership competition was again very different; the individual strategies hypothesized as most appealing in this case are
Governance (Negative and Positive), Injustice, and Enemy Image. The overall prediction for the third case about whether the combination of political opportunity variables favors more or less exclusive strategies was difficult to assess because two variables favor more exclusive strategies and two favor more inclusive appeals. Stability of elite alignment had changed by 1990 so that there was basically no elite alignment between the colonial and the controlling power (Britain, Mugabe) or the controlling and the regional power
(Mugabe, South Afiica). It is argued that this promotes inclusive strategies (see Chapter
293 2). The greater degree of repression promotes more exclusive strategies as does the lack of international involvement. Finally, increased regional integration should promote more inclusive strategies. The summary prediction here is indeterminate.
6.4.2 Comparing Strategies
We know that Tekere is considered to be the beneficiary of the mobilization shift in 1990 (though Mugabe stayed in power; see Chapter 5). Do we see that his top four strategies are closer to the expected winning strategies. Governance Positive, Governance
Negative, Injustice, and Enemy Image, than Mugabe’s? Table 6.15 provides the data to evaluate this question.
294 Mugabe Tekere (n=1102)* (n=683) Identity (Zimbabwe)*’ 9.5% 6 .1%
Outgroup Negative (Enemy 4.7% 13.3% Image)*** Justice Negative*** 11.6% 12.2%
Storytelling 11.2% 9.4%
Governance/negative*** 3.2% 17.3%
Governance/positive*** 41.3% 32.4%
Outgroup Positive (Ally Image) 8.5% 6.4%
Justice Positive 9.9% 2.9%
Total % 99.9 100%
The "n” here is the total number of strategies and the number given in each cell is the percentage of this total. The strategies in bold are the strategies hypothesized to be used most frequently by the wirming leader; in this case that is Tekere. The asterisks denote the actual strategies used most frequently by the winning leader. These results are significant at the .014 level.
*’ Because Zimbabwe is the focus of the research and considered to be the society under exam ination this reference is the focus. Tliis figure excludes references to the leader's party because this is misleading about the groups in the society the leader is defining.
Due to rounding off of the numbers, these totals do not equal exactly 100%.
Table 6.15: Comparison of Strategies, Shift 3
The expectations are supported, and according to the exact test are significant
(.014). Both leaders use Governance Positive strategies as they propose that they have the best suggestions for the way Zimbabwe should be governed. Mugabe’s score is 41.3% compared to Tekere’s 32.4%; this is most likely because Mugabe also spends a lot of time talking about how much his government and the institutions he has set up have helped
295 Zimbabweans. In this light, Tekere’s Governance Positive strategies serve more of the function that the context suggests wül be successful. Further, it was also suggested that a winning leader should use the Governance Negative strategy because of the way the move to a one-party state goes against global norms. Unsurprisingly, Mugabe’s use of this strategy is only 3.2% compared to Tekere’s 17.3%; most of Mugabe’s strategies here were about the Lancaster House constitution. Although this does not tell us anything conclusive of course, it does suggest that Mugabe should have been more humble when he spoke of his government’s post-independence accomplishments.
Tekere’s third most frequently used strategy is the Enemy Image strategy: 13.3% compared to 4.7%. The time was right to blame someone for Zimbabwe’s problems; also,
Tekere spent much time talking about the threat that Mugabe and his corrupt cohort posed to the average Zimbabwean, and how Mugabe’s crusade in Mozambique endangered many innocent civilians. Meanwhile, Mugabe’s third most frequently used strategy was Storytelling; in the 1990 context using this approach did not seem to help
Mugabe as much as in the previous two cases. Tekere also uses the Injustice strategy though Mugabe and he both used this 11-12% of the total. This may mean that the combination of the others above—the way in which Tekere used Governance Positive and the fact that he also emphasized Governance Negative and Enemy Image more than
Mugabe—were most important. Finally, an important change in this period is both
Mugabe and his competitor’s increased use of the Identity (Zimbabwe) strategy. Mugabe
296 previously used more global and Africa references to describe his audience, for example.
The next section looks at the strategies the leaders use in terms of how inclusive or
exclusive they are.
6.4.3 Assessing the Predictions: The Exclusivitv Index
This section examines the strategies and targets in order to assess which leader is more exclusive in 1990. From what we know about the case, it would seem that Mugabe would be more exclusive. Tekere’s Zimbabwe Unity Movement called for unity among all
Zimbabweans including whites against Mugabe’s move to create a type of political system that would violate the needs of society. However, it turns out that Mugabe is more inclusive in his appeals. Unlike the previous cases, Mugabe’s more inclusive way of framing the situation was not as appealing in 1990. This makes sense from the perspective that when people feel that their life has not improved, they look for a scapegoat and they also are less likely to respond to inclusive ideas about the plight o f the country as a whole.
Tekere was pointing out that not all Zimbabweans were suffering because the leadership was growing rich at the expense of everyone else.
Nature o f Strategies
Table 6.16 shows the frequency distribution for the more exclusive and more inclusive strategies by leader.
297 More exclusive strategies Mugabe Tekere (n=1047) (n=681) Identit}" .1% 0 (Zimbabwe/very exclusive) Enemy Image 5% 13.4%
Justice Negative 12.2% 12.2%
Storytelling 11.8% 9.4%
Governance Negative 3.3% 17.3%
Total exclusive strategies 32.4% 52.3%
More inclusive strategies Mugabe Tekere (n=1047) (n=681) Identity 4.7% 5.9% (Zimbabwe/very inclusive) Ally Image 9% 6.5%
Governance Positive 43.5% 32.5%
Justice Positive 10.4% 2.9%
Total inclusive strategies 67.6% 47.8%
Assessment 1 Mugabe Tekere Exclusivity Ratio “ .5 1.1 Change from shift 2 -1.32 -.21 “This ratio is calculated by dividing the total percentage of exclusive strategies by the total per centage of inclusive strategies. The higher number indicates that the leader used more exclusive strategies than the leader with the lower number.
Table 6.16; Comparison of Exclusive and Inclusive Strategies for Shift 3
Here we see large differences between the two leaders. Whereas about two-thirds of
Mugabe’s strategies are more inclusive, about 48% of Tekere’s fall into this category. As
298 discussed above, the differences are in Mugabe’s greater use of the Governance Positive
and Justice Positive strategies, on one hand, and Tekere’s greater use of the Governance
Negative and the Enemy Image strategies. In terms of the Identity strategies, not reflected
here, it is interesting to note that in this case both leaders use more Africa ingroup
references than previously; Mugabe’s score here was 13.4% of 134 ingroup references
and Tekere’s score was 14.3% of 49 ingroup references. These figures are pointed out
here because they fit with the expectation of the regional integration variable: as
integration proceeds leaders will increasingly define the nation’s identity in more regional terms.
Nature o f Targets
We can now compare how exclusive the leaders are in terms of strategy' targets.
Looking at the different strategies and their targets, the tables below show that Tekere defines the targets of five of eight strategies more exclusively than Mugabe; on the Ally
Image-Goals strategy the leaders use the same degree of exclusivity. To begin with targets of the Injustice and the Governance Negative strategies. Table 6.30 shows that
Tekere is much more exclusive than Mugabe for both. As in the previous cases, this score is explained by Mugabe’s tactic of using more global and Afiica references, connecting
Zimbabwe’s plight with those in the region and around the world in a manner similar to
Gerry Adams.
299 Injustice Governance Negative
(More inclusive) Mugabe Tekere Mugabe Tekere (n=99) (n=51) (n=30) (n=76) People in world 13.1% 5.9% 23.3% 6 .6%
People in Afiica 16.2% 7.8% 16.7% 7.9%
People in Zimbabwe 70.7% 86.3% 60% 85.5%
E.xclusmty ratio “ 2.41 6.3 1.5 5.9
Because there are only three references, an exclusivity ratio is used as a comparison measure here (Zimbabwe references divided by the sum of world and Africa references); this is slightly different than the measure used for the more complex Ireland cases, which have eight possible references.
’’Niunbers are percentage of all Justice Negative strategies and Governance Negative strategies (respectively) for which this reference is the target.
Table 6.17: Comparison of Targets (Injustice and Governance Negative Strategies) for Shift 3”
The next table shows the exclusivity ratios for the targets of the Enemy Image strategies. For all of these, Mugabe is more exclusive; again (as in the previous cases) it is on these strategies that Mugabe’s scores are an anomaly. By including groups from Africa and the international arena in his attributions of Threat and Blame, or by describing people actually in Zimbabwe with universal references, it is argued here that Mugabe creates a more exclusive picture of the situation.
300 Blame Threat Total of Enemy strate gies “ (More exclusive) Mugabe Tekere Mugabe Tekere Mugabe Tekere (n=6) (n=25) (n=22) (n=20) (n=42) (n=75) People in world 16.7% 0 31.8% 5% 23.8% 2.6%
People in Africa 0 4% 54.5% 0 38.1% 2.6%
People in Zimbabwe 83.3% 96% 13.6% 95% 38.1% 94.7%
Exclusivity ratio*’ .2 .04 6.35 .05 1.62 .05
“This percentage includes the threat, blame, betrayal, and outgroup image-ambiguous variants.
With the outgroup strategies, the broader references are considered more exclusive because the "enemy” encompasses more people, making the ingroup smaller and more exclusive. This ratio is the sum of the world and Africa references divided by the Zimbabwe references.
Table 6.18; Comparison of Targets (Blame, Threat, and Combination of Enemy Strategies) for Shift 3
For the Storytelling targets, Tekere is more exclusive. As in case 2, Mugabe more frequently invokes the past myth of Zimbabwe as it involves international actors and others in Africa (and more frequently, those in Zimbabwe described in terms of their
Afiican identity).
301 (More inclusive) Mugabe Tekere (n=77) (n=43) People in world 7.8% 2.3%
People in Africa 6.5% 7%
People in Zimbabwe 85.7% 90.7%
Exclusivity ratio 5.99 9.75
“The target of the Storytelling strategy indicates the group reference that the leader uses when he talks about the "imagined community."
Because there are only three references, an exclusivity ratio is used as a comparison measure here (Zimbabwe references divided by the sum of world and Africa references); this is slightly different than the measure used for the more complex Ireland cases, which have eight possible references.
Table 6.19: Comparison of Targets (Storytelling Strategy)® for Shift 3
Turning to the Ally strategies, when we look at the groups both leaders target as having shared goals, they have very similar exclusivity ratios. An interesting change from previous episodes is not revealed in those scores, though. More than Mugabe’s competitor in the first two cases (Muzorewa), Tekere also uses Afiica and global references as a high percentage of his total targets. Looking at all of the Ally Image strategies combined (shared goals, similar situation, similar culture, and ambiguous),
Tekere is again more exclusive than Mugabe.
302 Shared Goals Total of Ally Governance strategies® Positive (More inclusive) Mugabe Tekere Mugabe Tekere Mugabe Tekere (n=28) (n=16) (n=80) (n=44) (n=249) (n=133) People in world 35.7% 18.8% 32.5% 22.7% 13.7% 7.5%
People in Africa 39.3% 56.3% 21.2% 22.7% 14.1% 12%
People in Zimbabwe 25% 25% 46.3% 54.5% 72.3% 80.5%
Exclusivity ratio .333 .333 .862 1.2 2.6 4.13
* This percentage includes the goals similar, situation similar, culture similar and outgroup image- ambiguous variants.
This is calculated by dividing the sum of the world and Africa references by the Zimbabwe refer ences. The former are considered inclusive targets because it means the leader is looking outside the political society for allies and help in governance, and is also looking within the society. A higher number here indicates a more exclusive score.
Table 6.20; Comparison of Targets (More Inclusive Strategies) for Shift 3
The final columns of Table 6.20 examine the targets of the Governance Positive strategies. Here Tekere is more exclusive, but again more than Muzorewa, he employs global and Afiica targets. This change may be an indicator of the realization that
Zimbabwe needs the help of external actors to work on its own problems, and is also evidence of the way in which both actors emphasize Zimbabwe’s positive role in regional and international politics.
Summary o f Findings
When the weighted exclusivity ratios are examined, it is apparent that Tekere is more exclusive overall than Mugabe. Table 6.21 presents this information.
303 Exclusivity Exclusivity Exclusivity Total Leader Change ratio: ratio: ratio: exclusivity with more since Nature of Nature of Nature of ratio’ exclusive last strategies ingroup outgroup appeals period targets targets overall Mugabe .5 12.5 9.4 7385 +2442 (n=1102) (n=455) (n=122) Tekere 1.1 26.1 1.7 8862 XXX -5839 (n=683) (n=303) (n=119) This ratio is calculated by first weighting the ratios in the three previous coltunns by the n for each and then smnming the three columns.
^ Here a positive number means the leader’s strategies were more exclusive; a negative number means they became more inclusive. This change for Tekere can only be considered as 'change in Mugabe’s competitor.”
Table 6.21: Summary of Findings, Shift 3
In this third case, Mugabe’s appeals became more exclusive while the winner was
more exclusive overall. As discussed above, the attempt at predicting whether more or
less inclusive strategies should win according to the context in this case ran into difBculties
because two of the four political opportunity variables pointed toward more inclusive
strategies while two predicted to the favorability of more exclusive strategies. Since the foregoing analysis showed that more exclusive strategies were used by the leader who
effected a mobilization shift, it would appear that in this final case the most important variables are repression and international involvement. It can be argued that the hypotheses about the relationship between the regional integration and the change in stability of elite alignment variables are not valid—for example, that regional integration promotes more exclusive strategies. Given that regional integration was not significant in
304 the other two cases means no such conclusion can be drawn. Also, the fact that regional integration and the success of inclusive strategies were correlated in some of the Ireland cases weighs against that idea.
Instead, there are plausible reasons that the regional integration variable and the change in stability o f elite alignment factor simply may not have played a role in this final case. First, although regional economic and political cooperation had deepened in southern Africa since 1980, it had not progressed as far as European integration; instead, cooperation is perhaps the strongest word that should be used when describing this phenomenon in 1990 southern Afiica.
To understand why the change in stability of elite alignment variable may not have been significant in 1990, it is necessary to return to the original hypothesis. It was stated that if the alignment between controlling power and colonial power and between controlling power and regional power both loosen, then more inclusive strategies are favored. The logic behind the proposition was that these two alignments helped maintain a status quo which in turn bolstered a system objected to by the political minority. A dissolution of these relationships means that the status quo (the system that keeps the political minority in its place) is undermined and a leader making more inclusive appeals is more credible; when whites keep blacks out of the political system because of their color, a black leader trying to convince blacks to be conciliatory and that whites are part of “us” is expected to be less successful in this context than a black leaders trying to convince blacks of the need to overthrow the system and rule by a foreign “them” (whites).
305 In the 1990 case, the actors in Zimbabwe did not actually have much of any
relationship with Britain or with South Africa. Though applicable in 1979 and 1980, the
argument makes less sense that the relationship of South Africa and Britain helped
Mugabe maintain the status quo in 1990. Unlike the Northern Ireland case, these actors
stay less involved over time, a fact that is symbolized by Britain’s failure to ante up all of
the development funds for land purchase (from willing white sellers) that had been
promised at Lancaster House. Indeed, a loosening of both of those alignments had little
do with the fortunes of Tekere or any other voice of Zimbabwean political minorities. If
anything, these relationships have shifted into the same category as other international
actors, and their lack of involvement means that more exclusive strategies were actually favored by this variable (see propositions for international intervention/mediation). From this case, the proposition concerning the elite alignment variable should be qualified: in order for the change in stability of elite alignment variable to be a significant factor in the mobilization context, there must exist a relationship between the controlling and colonial and the controlling and regional powers which helps the controlling power maintain the status quo. The next section compares the results across cases.
6.5 Looking Across Cases: Learning from the Unexpected
The objective of this discussion is to derive several qualifying propositions for future research from what has been learned in this study. The next chapter brings together
306 the findings and qualifications fi'om the Rhodesia/Zimbabwe mobilization episodes and the
Northern Ireland cases for an overall assessment of the propositions. Table 6.22
summarizes the case studies.
Shift 1 Shift 2 Shift 3 Mugabe 8943 4943 7385 total/% change (45% more inclusive) (49% more exclusive) Opponent 6556 14,701 8862 total/% change (124% more (40% more exclusive) inclusive)
Expected winning Exclusive Inclusive Indeterminate strategj- type
Winner/type of Mugabe Mugabe Tekere strategy/movement from more exclusive more inclusive more exclusive previous period moved toward moved toward more predicted winning inclusive direction Most frequently used Storytelling Storytelling Governance Positive strategies by winning Enemy Image Governance Positive Governance leader Injustice Enemy Negative Image/Injustice Enemy Image Were expectations Supported Supported (n/a) supported?
The higher numbers indicate that the leader showed greater exclusivity in his appeals. For the percent change, the old index is subtracted from the old and this number is divided by the old index.
Table 6.22: Comparison of Change Between Shifts, Totals of Exclusivity Index “
In the first case, there was some disagreement between the predictions of the repression and change in stability o f elite alignment variables and the international involvement variable. In that case, “change in stability of elite alignment” and “repression" predicted that more exclusive strategies would win (the regional integration was not
307 significant). The variable concerning the involvement of the international community, on the other hand, predicted that more inclusive strategies would win because the situation was defined and treated as an international concern. As the content analysis results show, the kind of strategies the winning leader (Mugabe) employed were more exclusive.
Because the international community was not neutral toward the situation and was supporting Mugabe’s right to play a role in the political system, the “inclusive” effect of international involvement appears to have been overridden. In fact, this fits with the hypotheses developed in Chapter 2.
The logic behind the proposition was that international involvement which questions the repressive political system sends a signal (to potential followers of the political minority leaders) that attention is being paid to their plight. For example, people are more likely to respond to Ally, Governance Positive, or Justice Positive strategies when external actors are endorsing those kinds of “definitions of the situation.” On the other hand, as discussed in Chapter 2, when actors in the international system ignore the situation or state explicitly that it is an internal issue, more exclusive strategies make sense
(why cooperate or be conciliatory when the situation is so desperate?). However, the third proposition stated in reference to this variable was that if international involvement and/or intervention support one leader’s view, that leader is more likely to be successful in mobilizing people within the targeted community to support him, whether his strategies are more inclusive or exclusive.
Table 6.23 below takes each contextual variable individually, shows whether the propositions related to it predicted more or less inclusive strategies, and then lists if the
308 winning leaders’ appeals were more inclusive or exclusive. The objective here is to explore when the expectations were supported and try to observe patterns of combinations. The concluding chapter will compare all seven cases.
Case 1: Case 2: Case 3: Expectations/ Expectations/ Expectations/ Outcome Outcome Outcome Change in stability of Exclusive/ not significant Inclusive/ elite alignment Exclusive Exclusive
Repression Exclusive/ Inclusive/ Exclusive/ Exclusive Inclusive Exclusive
International Inclusive, but Inclusive/ Exclusive/ involvement/ favoring Mugabe/ Inclusive Exclusive mediation Exclusive
Regional integration not significant not significant Inclusive/ Exclusive
' Where the font is in bold, it denotes cases where the prediction and the finding matched.
Table 6.23 ; Individual Assessment of Contextual Variables Across Zimbabwe Cases “
The most striking aspect about the African cases is how unimportant the regional integration variable appears to be in terms of predicting winning strategies; possible reasons for this were discussed above. Still, there does seem to be a positive relationship between the increase in regional cooperation and the inclusion of Africa ingroup references; in 1990, both Mugabe and his competitor increases the percentage of ingroup references that included an African level of identity. This is in line with the findings of the
Irish cases.
309 A second unexpected finding is the apparent insignificance of the change in stability of ehte alignment variable in the second and third Zimbabwe cases. The importance of changes in an elite alignment which undergirds a particular political system is equated to the opportunity for change in the status quo. When the status quo is supported, more exclusive strategies appeal for apparent lack of more conciliatory avenues. When that status quo is undermined, there is an opening for power sharing or other forms of cooperative resolution of conflict. In the second case, there was simply no change in the elite alignment, and in the third case there really had not been the former status quo alignment for several years. As noted above, the set of Zimbabwe cases points to an important lesson that the powers which constitute the alignment must be rather seriously involved with each other and with the institutional system whose restructuring is demanded by the political minority. The British-unionist and British-Irish relationships meet this criteria to a greater degree than the British-Rhodesian or the British-South
African relationships.
In the Zimbabwe cases, the repression and the international intervention/mediation variables were important. Even more interesting is that there seems to be a relationship between the two; the next chapter will elaborate on this observed relationship and compare other aspects of all seven cases.
310 CHAPTER 7
CONCLUSIONS
7.1 Introduction
In 1982, Gerry Adams and his Sinn Fein party worked to draw international attention to the colonial domination of Northern Ireland by Britain, linking the plight of nationalists with the blacks persecuted by South Africa’s apartheid regime. He decried the abuses by the British army and their Protestant operatives in the RUC (the Northern
Ireland police force); he condemned the evils of the capitalist sell-outs in Dublin. He cursed the economic crisis facing West Belfast residents and proclaimed that negotiation or constitutional solutions to the conflict were dead-ends. The only answers were “Brits
Out!” and likewise the overthrow of capitalist Dublin in favor of a new socialist republic.
His success was breathtaking to his followers—and shockingly horrific to many others given the advocacy of political violence that underlay his party’s stance.
In 1985, Adams’s tirade against imperial tyrarmy continued, but success this time went to John Hume. This leader of the SDLP was the polar opposite of Adams: the crisis in Northern Ireland was a problem of representation, not colonialism. If the nationalists/Catholics could only have a voice in the political system and be allowed equal access to jobs and housing, tensions would defuse. Better governance and a focus on
311 common goals—only possible with the involvement of the British and Irish
governments—would alleviate the country’s ills. Hume and Adams were each making the
same appeals in 1985-1987 that he was making in 1982-1983. Why did the mobilization
strategies of Adams have such success in the earlier case not in the latter? On the other
hand, why did Hume’s appeals draw so many people his way in the latter period but not in
the former?
A “world away,” in the late 1970s and 1980, another battle for the hearts and
minds of “nationalist brethren” was taking place in a different British colony. In the governance crisis in Rhodesia in 1979, Robert Mugabe successfully carried the anti
imperialism banner against the white Rhodesians and “sell-out” Muzorewa. He proclaimed the need for national unity (meaning all Africans) against the “foreign threat,” and his radio broadcasts and grass-roots campaigns both kept people away from the polls and supporting the guerrilla war. In 1990, Mugabe continued his “national unity in the face of threat” appeals, highlighting the gains that had been and could be won if people would continue to support his vision of a single-party (ZANU PF-led) state. He also condemned his challenger’s willingness to ally with “traitors” in an all-white party. This time, however, the Father of Zimbabwe was shocked by the challenge of Edgar Tekere and the Zimbabwe Unity Movement. Though Mugabe stayed in power, Tekere’s rhetorical attacks forced significant changes in ZANU PF’s approach. Tekere attacked the repression of the government and decried the inconsistencies between Zimbabwe’s path and the democratization just then taking hold on Africa and Eastern Europe. Why did
Mugabe have such immense success with his mobilization efforts—against all odds—in
312 1979 and 1980, but then was vulnerable to challenge by an “underdog” with few
mobilizing resources in 1990? What accounts for the variation in follower reaction to
leaders using similar strategies across time? This study argues that the answers to these
questions about both countries lie in the relationships among the context, the way the
leaders viewed this context, and the strategies they employed to frame this context for
potential followers.
This project set out to examine the efforts o f competing leaders to mobilize people
around particular definitions of group identity and mission: are there differences in the strategies used by leaders who are more and less successful in their mobilization efforts?
The seven case studies have shown the importance of context in answering this question.
Some strategies (both taken individually and collapsed into more exclusive/more inclusive) appear to be more successfiil when particular conditions exist in the domestic and international arenas. To understand the role of leadership, this study shows that it is essential to consider how leaders work within their domestic and international political environments. But even more importantly, as this chapter will discuss, understanding the
“objective” environment and investigating how the political opportunities constrain or empower leaders is not enough. What we can leam from the results here is that leaders may prioritize different aspects of the context: sometimes one or two variables are more important than others to them. Therefore, what becomes truly important in understanding the strategies the “successful” leaders use is the match between the context and the way in which the leader interprets these situational conditions to his audience. To explore ±ese and several other issues, this chapter begins with a comparison o f the seven cases. The
313 second part of this chapter discusses the contributions this study makes to several bodies of literature from which the framework was drawn. Both parts of the chapter point out directions for future research, which are summarized in the concluding section.
7.2 Comparisons and Contrasts
Table 7.1 presents the predictions for each political opportunity variable across the cases. Where the prediction matched with the outcome, the cells are shown in bold type.
The next to the last column shows which contextual variables were most emphasized by the winning leader, and the final column states whether the findings about the individual strategies were significant.
314 Change in Repression International Regional Overall Strategies of Contextual Findings stability of involvement integration prediction winning factors about indiv elite /mediation leader/leader emphasized strategies alignment by winner ’’ significant? Case 1 Exclusive Exclusive Exclusive Not sig Exclusive Exclusive/ Elite align, Yes (.018) N. Ireland Adams Repression, 1982-83 Internall. Case 2 Inclusive Inclusive Inclusive Inclusive Inclusive Inclusive/ Elite align, Yes (.018) N. Ireland Hume Internat’l, 1986-1987 Regional Case 3 Not sig Inclusive Inclusive/ Inclusive Inclusive Exclusive Elite align, No (.24) N. Ireland (no change) Adams but move/ Internat’l 1996 Adams Case 4 Exclusive Exclusive Inclusive/ Not sig Exclusive Exclusive/ Elite align. Yes (.014) N. Ireland Adams (no change) Adams Repression, 1997 Internat’l Case 5 Exclusive Exclusive Inclusive/ Not sig Exclusive Exclusive/ Elite align. No (.24) LA Rhodesia Mugabe Mugabe Repression, 1979 Internat’l Case 6 Not sig Inclusive Inclusive Not sig Inclusive Inclusive/ Elite align. Yes (.014) Zimbabwe Mugabe Repression, but both 1980 Internat’l, leaders Regional match Case 7 Inclusive Exclusive Exclusive Inclusive Indeterm. Exclusive/ Repression, Yes (.014) Zimbabwe Tekere internat’l 1990 ’ Boldface type indicates that the prediction in the cell matched the outcome. Boldface type in the next to last column indicates tliat the contextual factor emphasized by the winning leader was a significant predictor of the winning strategies. In other words, the bolded cells in the other columns matched with what the winning leader was saying. The last column refers to the results of the exact tests for significance.
Table 7.1 ; Summary of Northern Ireland and Zimbabwe Cases ' Change in Stability o f Elite Alignment
For the change in stability of elite alignment variable, the hypotheses were shown
to be valid in four of five cases in which they were expected to be significant. Looking at
the major political opportunities that the leaders tended to emphasize in their appeals (see
final column in Table 7.1), the winning leader in fact emphasized this variable. For
example, in Northern Ireland/1997, Gerry Adams often talked about how the British
reaction to Catholic marchers at Drumcree (see Chapter 3) epitomized a return to the tight
British-Orange/Protestant relationship. He denounced British bad faith in 1997 after that
government’s earlier vows that they were neutral about a resolution in Northern Ireland.
Also, in Northern Ireland/1986-87, Hume focused on the changing relationship between
Britain and Ireland and Britain and the unionists; he even argued that the change was
beneficial for the community of Northern Ireland. His argument was similar to the
hypothesized relationship: efforts at negotiation are promoted when the two countries
work together and the British move toward a more neutral role vis-a-vis the parties in
Northern Ireland.
In two of the three Zimbabwe cases, this variable did not seem to be important.
For example, in 1980 this variable was not expected to be significant because there was no change from 1979. In 1990, this variable was hypothesized to promote more inclusive strategies but more exclusive strategies were used by the winning leader. In Rhodesia
1979 when the exclusive prediction does seem to be supported, Mugabe talked about this contextual variable in a way that placed emphasis on the hypothesized relationship stated in Chapter 2. For example, in 1979 he highlighted both the “loosening” relationship
316 between Britain and the Rhodesians and the “tightening” relationship between the
Rhodesians and South Africa. This combination was expected to encourage followers to
respond to more exclusive appeals because the South African support was enough to
continue support of the status quo (a system in which Mugabe’s audience was repressed
and disenfranchised). As with the Northern Ireland cases, it does appear that the
expectations of this variable were more likely to be supported when the leader emphasized
the role the elite alignment was playing in terms of opportunities for the ingroup.
Repression and International Involvement/Mediation
Across all of the cases, there appears to be a relationship between the repression
and the international involvement variables. When both the repression and international
variables predict toward more exclusive or more inclusive strategies (NI/1982-83,
NI/1986-87, NI/1996, Zim/1980, and Zim/1990), the expectations match with the
outcome. Further, with the exception of NI/1986-87 and NI/1996 (in which the winning
leaders do not play up both of these contextual aspects), the leaders do emphasize the
importance of these two political opportunities in their appeals. Still, in NI/1986-87 and
NI/1996, the leader plays up the opportunities offered by the one variable, international
involvement.
The two cases where these two variables make opposing predictions are NT/1997
and Rhodesia/1979. In Northern Ireland 1997 and Rhodesia 1979, the repression variable predicts toward more exclusive strategies and this dimension was indeed emphasized by both Adams and Mugabe (the winning leaders). However, the international variable for these cases played out in a more complicated way that offers interesting insight into the
317 relationship between these contextual variables, leaders’ interpretations of them, and the
kind of strategies the winning leaders use. In both cases, the role of the international
community was to treat the situation as an international issue—hypothesized to promote a
leader with more inclusive strategies (such as Governance strategies). Yet in these cases
the international community’s involvement was such that it was not neutral' and instead
was hypothesized to favor Adams in one case and Mugabe in the other. Because the
prediction here was then more in favor of an individual rather than a type of appeal, this
may suggest that the international variable weighed more heavily than the other contextual
factors in the Northern Ireland case. In the Zimbabwe case, this reinforced the prediction
that Mugabe’s exclusive strategies would be more successful, but in the Northern Ireland
case, the prediction for more inclusive strategies and for support for Adams went in
different directions (because Adams was more exclusive overall).
Comparing these two cases (NI/1997 and Rhodesia/1979) with Nl/1982-83 and
Zim/1990—all four cases where the repression variable promoted more exclusive strategies—it seems that when important actors in the international arena got involved, the exclusive strategies emphasizing repression may be “tempered.” However, when there is repression that is emphasized by a leader and the international community stays out of the situation, the more exclusive strategies such as Enemy Image, Injustice, and Governance
‘ As discussed in a footnote in Chapter 2. it is recognized here that "neutral” is a tricky term and that, in fact, no involvement fits the dictionary definition of neutral. However, the term in this study is used to denote involvement that pays little or no attention to one leader or another, but simply asserts the need for a resolution. Also, note that whether involvement favors one leader or another in this smdy refers to the leaders of the political minority. This means that involvement in favor of the political minority may not be neutral at all with regard to the unionist and nationalist parties in Northern Ireland, but can still be neutral (as defined here) in the contest between leaders of the nationalists (Muzorewa and Mugabe, Adams and Hume). 318 Negative are more likely to “resonate” with the audience of the leaders. (Table 7.2 shows
the three major strategies used by the winning leader and which of the four political
opportunity variables he tended to emphasize.^)
■ In terms of other findings from Table 7.2, it would be especially useful if patterns in the strategies used and the contextual variables could be found. One pattern is tliat the use of the Governance Positive strategies is accompanied by emphasis on international involvement In the two exceptions, Nl/1982-83 and Rhodesia/1979, the leaders emphasis on this dimension was achieved in an exclusive manner. What this means is that for example, when Gerry Adams discusses the international dimension of Ireland’s conflict he is reaching out to other paramilitary' organizations around the world, and seeking aid only for the Catholic community as opposed to focusing his efforts on Northern Ireland as a whole. There do not seem to be other correlations.
319 Strategies used by Change in stability Repression International Regional winning leader of elite alignment involvement integration Case 1 Enemy Image did emphasize did emphasize did emphasize, but N. Ireland Justice Negative in a very exclusive 1983-83 Storytelling way Case 2 Governance Pos did emphasize did emphasize did emphasize N. Ireland Justice Negative 1986-1987 Storytelling Case 3 Governance Pos did emphasize did emphasize N. Ireland Storytelling 1996 Justice Negative Case 4 Governance Pos did emphasize did emphasize did emphasize N. Ireland Storytelling 1997 Justice Negative Case 5 Storytelling did emphasize did emphasize did emphasize, but Rliodesia Enemy Image in exclusive way 1979 Justice Negative Case 6 Storytelling did emphasize did emphasize did emphasize Zimbabwe Governance Pos 1980 Enemy Case 7 Governance Pos did emphasize did emphasize Zimbabwe Governance Neg 1990 Enemy Image Where applicable, the first lino in each cell denotes that the leader did talk about these aspects of context in his speeches.
Table 7.2: Comparison of Individual Strategies and Contextual Variables Highlighted by Winning Leader" This cross-case comparison suggests the following hypotheses about the relationships among these variables which should be explored in future research— especially given the importance accorded to these two variables in the literature on social movements and nationalist group behavior. (1) When repression of the minority community is lessened and therefore promotes more inclusive strategies, and actors in the international community are involved in treating the situation as an international concern and involvement is “neutral,” the leader with more inclusive strategies is favored by that context. (2) When repression is increased and therefore promotes more exclusive strategies, and the international community defines the situation as an internal/not international concern, then more exclusive strategies are favored. (3) When repression is lessened or increased (and therefore promotes either more inclusive or more exclusive strategies, respectively), and actors in the international community are involved in treating the situation as an international concern but involvement favors one leader in particular, that leader is more likely to be successful. The contribution made here to extant work is elaborated in section 7.3 below.
Regional Integration
For the regional integration variable, it turned out that there were only three cases in which it was expected to be significant. In all three, it was expected to promote more inclusive strategies; the hypothesis was only valid in one case. Despite this low correspondence, when certain aspects of the data are examined the role of this variable may still shed some light. First, the only case in which the hypothesis relating this variable to a winning strategy type is valid is Northern Ireland, 1986-87. Note also that this is the
321 only case in which the leader chose to highlight this aspect of the context (see Table 7.1, where cell is in bold type). Indeed, in this period, John Hume presented European integration and the governance processes related to it as an opportunity for the community of Northern Ireland, Britain, and Ireland to overcome their differences. He often used the example of France and Germany; after centuries of fighting the two countries were now the mainstay of peace and stability in Europe.
The data for the Zimbabwe cases also offer insight about the way in which leaders may play up the contextual variables. In Zim/1980, from an analytic standpoint the research showed that regional integration was not expected to be significant in supporting more inclusive or more exclusive strategies. However, in this case Mugabe did in fact spent a great deal of time talking about Zimbabwe’s future role in the region—in the fight against apartheid, in the Front Line States alliance, and in the economic integration of
Zimbabwe with Zambia, Tanzania, and Mozambique. This case hints at the importance of understanding the representation of the situation by the leader since the winning leader emphasized this dimension (despite the “objective” assessment that it was not important).
In contrast, Muzorewa, the less successful leader in 1980, did not talk much at all about the regional arena in his appeals.
A second finding about the regional integration variable is apparent, looking over time. As discussed in Chapters 4 and 6, as regional integration increased, the percentage of group references that were at the more inclusive regional level (either “people in
Europe” or “people in Afiica”) increased as well. The importance of this correlation is that it suggests that a leader who is trying to be more inclusive in terms o f the Identity
322 strategy—which is a core aspect o f promoting multiple identities and loyalties—is likely to be more successful if the leader’s country is involved in regional initiatives that make liis more inclusive definition of the situation appear more plausible to potential followers within his country. Although this finding is only tentative, it warrants future research because the potential contribution of this line o f research to early ideas offered by Haas
(1958) becomes apparent. That author suggested that regional integration in Europe would promote “micronationalism” because previously isolated ethnic/national groups (for example, the Irish) would be more likely to interact across borders (North and South); exclusive strategies toward other groups (such as the Protestants or unionists) would be more appealing. The example of the Northern Ireland case shows that instead of becoming more exclusive toward others in the region, the appeals that have been
“successful” have had to be more inclusive. The next sections discuss additional ways in which this study contributes to extant work.
7.3 Contributions to Work on Social Movements, Nationalism, and International Relations
Although limited to the category of cases meeting the scope conditions for this study,^ the findings here speak to several existing literatures. A broader, overall finding in this comparative case study that is of significance to literature on both social movements and nationalism is the fact that contextual factors, especially the individual factors as they combine with each other, are important to the competition among nationalist leaders for
^ Recall that this study chose cases that are characterized by a "legitimatioa crisis” and in which two central leaders with political parties compete to lead people who consider themselves as part of a national or ethnonational group. The final section of this chapter discusses the generalizability of the findings. 323 the hearts and minds of followers. Few studies have looked comparatively or
systematically at the role of these four variables, even though a large body of literature
argues that each is important (for a review of this literature, see the chapters in McAdam,
McCarthy and Zald, 1996). Even fewer studies have examined how these variables interrelate, but as discussed, this dissertation suggests specific relationships between repression and international involvement, for example. Here there is also a relatively narrow and unique definition of the dependent variable—mobilization “direction” toward more inclusive or exclusive leaders. This is considered an important focus of research among pohtical scientists (especially those who study nationalism, identity, and conflict resolution), but tends not to be one of the foci of social movement theorists. Because the findings relate the effects of the “political opportunities” of repression and change in the stability of elite alignment to the role of international actors in influencing the debate between nationalist leaders, this project shows the possibilities for both the social movements and international relations fields to benefit from true cross-fertilization and not merely the borrowing of concepts.
More specifically, the three propositions about the relationship between repression and international involvement (see p. 321 above) that emerged from this study help to qualify and broaden the general discussion among social movement theorists about repression as a particular kind of political opportunity for mobilization, especially since there has been much debate over what kinds of factors should be classified as
“opportunities” (for example, see McAdam in McAdam, McCarthy, and Zald, 1996, pp.
25-29). This study suggests that changes in degree of repression on the wider community
324 and parties within that community do seem to affect how leaders using more exclusive or
inclusive appeals are able to step in and mobilize people around their particular idea of the
group’s identity and mission.
Further, the suggestion that the international involvement variable can moderate
(see proposition 3 above) or intensify (see propositions 2 and 3 above) the effect of
repression on the direction people in which people are mobilized is not a relationship that
has been explored. In light of the fact that this dissertation shows how changes in ideas of
conflict resolution and sovereignty (manifested in whether an ethnonational conflict is
defined as an internal or international issue) can influence popular mobilization efforts, the
lack o f research is unfortunate. Recent work in social movement theory has argued generally that the role of international factors on mobilization should be studied. For example, McAdam (1996, p. 34) states, “...movement scholars have, to date, grossly undervalued the impact of global political and economic processes in structuring the dom estic possibilities.. The propositions above offer a good place to start that ties in with extant literature on repression (contributions to social movements theory on the concept of framing are discussed separately in the following section).
Another significant implication emerges from the findings about the international dimension. An underlying assumption of this research was that the international community should try to act to promote leaders who are more inclusive (and empirical research will help us discover how to do so), for this should help make the audience more susceptible to these more inclusive definitions of group identity and mission vis-à-vis other groups in the society and/or the region. For example, if international actors can affect the
325 situation so that the exclusive appeals of Justice Negative and Enemy Image are less likely to resonate, leaders will be pushed toward using more Governance Positive strategies.
More inclusive definitions of group identity and mission are in turn building blocks of a civil society in which—over time—people have more of a sense of multiple loyalties.
When we consider that Hume himself reached out to Adams and Sinn Fein (first in 1988 and then more successfully in 1993) and also that the role of the international community in 1996 was one of several factors getting the parties to the table in Belfast, this assumption is called into question. This research shows that there may be particular points in time when it is necessary to support a more exclusive leader so that he can pull his followers along toward negotiations and get them to the “place” where they are forced by the institutions of a power-sharing assembly to define themselves more in terms of what they have in common with the unionists.
Of course, foreign policy makers in other states will have to have a keen awareness that the goals of the more exclusive leader are to move toward negotiations and not simply to get international support for a negative platform. This issue is addressed by a corollary to the above. While international involvement does not necessarily promote a leader using more inclusive strategies relative to his competitor, the involvement of international actors in a way that brings a more exclusive leader into a position of legitimacy can push that leader to become significantly more inclusive relative to his previous definition o f the situation. For example, Adams showed a huge increase in the percentage of Governance strategies he used fi'om 1987 to 1996; at the same time the Enemy Image strategies and
Justice Negative strategies decreased significantly as a percentage of all strategies used.
326 In terms of the exclusivity index scores, Adams was more exclusive than Hume in both
1996 and 1997, but compared to himself became increasingly more inclusive from 1987-
1997. Indeed, by talking to and not isolating an exclusive leader, actors in the international community can push that leader to broaden his constituency (this point is elaborated upon in section 7.5 below).
Although the Zimbabwe case can only be looked at counterfactually, the argument would be that if the international community would have paid more attention to the
Zimbabwe Unity Movement in 1990, the tensions that have been building in Zimbabwe against Mugabe since then may have been defused by giving dissidents a voice in the political process. The current explosive political atmosphere in that country might have been channeled toward an institutional resolution. A study of the competition between
Mugabe and the grass-roots opposition movement since 1993 to the present could shed light on this issue and enrich our understanding even further.
Indeed, in terms of avenues for future research, the need arises to explore ifrhow international behavior affects the degree to which a more exclusive leader is able to become more inclusive and still mobilize people toward negotiations once he is brought to the negotiating table. Because the repression and stability of elite alignment variables changed so dramatically in NI/1997 to promote exclusive strategies, it is not possible to shed much light on this question in the present study. Northern Ireland politics in 1998- present, and Zimbabwean politics from 1993-present offer rich opportunities to do so. In addition, the finding that the international community can be important to the identity debate makes it necessary to explore the role of other categories of actors beyond states,
327 regional organizations, and even the United Nations as it existed during the Cold War
years. In recent years, a plethora of non-governmental actors has come onto the scene in
a more active way, and in some cases may make a difference in affecting the mobilization
strategies used by leaders and the success of various strategies. In post-Cold War Europe,
the Organization for Security and Cooperation in Europe has taken on a relatively
interventionist role in states with recognized political minorities (for example, see
DeVries, 1999), and the effects of that organization on mobilization around more
exclusive or inclusive ideas remain to be seen. Even with the United Nations, the impetus
given to the “new” mission of peace building (see Boutros-Ghali, 1995) means that this
international organization may play an increased role in divided societies in the future. At
present, the UN has limited its missions to post-conflict peace building, so those cases that
have not moved all the way to civil war may not be affected by the involvement of this
actor, but in the future the UN may play a more active role. Ultimately, the effects of
involvement of both non-governmental and intergovernmental actors on mobilization
episodes raise empirical questions that need to be explored in comparative case studies.
A final section of this chapter returns briefly to the issue of the policy relevance of the findings on the international dimension.
In addressing the nationalism literature, this study does not test competing theories of nationalism but instead argues that a unique—or at least seldom employed—focus of research on nationalist conflicts/debates should be adopted. The focus on competing leaders within the “same” nationalist group is shown to be a valid and useful approach because it demonstrates the range of ways in which the “same” situation can be interpreted
328 so differently. This in turn underscores the importance o f leaders in the mobilization
process, as opposed to the predominance of structural factors. The context taken more
“objectively” may be crucial, but it is necessary to understand how leaders are interpreting
any situation to have a full picture of why they adopt particular strategies and why these
strategies are more or less likely to be successful with different kinds of audiences (this
point is elaborated below in the discussion of leadership).
Also, much literature on nationalism does not explore the role of external
variables, and the subset that does is limited to cases of military intervention, irredentism,
the tension between sovereignty and self-determination, the role of United Nations
peacekeeping, or explications of the details of a particular case, for example (for a
sampling, see Brown, 1996; Dashwood, 1997; Esman and Telhami, 1995; Ganguly and
Taras, 1998; Goodman, 1996; Guelke, 1988; MacGinty, 1997; Mayall, 1990; Ryan, 1995).
Though increasingly important in the post-Cold War era and certainly contributing to work in international relations, this work tends not to look at how the international arena affects definitions o f national/group identity. This is despite the fact that how people are mobilized around different definitions of identity and mission is regarded as an important and policy relevant phenomenon that needs to be understood more comprehensively. This study takes a first step in that direction, and relates external factors to the “definitions of the situation” suggested by leaders and the popularity of these definitions with audiences.
More generally, when the subject of investigation is societal identity, this study shows that the issues involved blur all lines between domestic and foreign arenas. Leaders in every case appealed to people in their own states by linking their fates with people
329 abroad who share similar dimensions of identity: Europeans, oppressed, democrats. That the idea of societal identity can coincide little with state borders undermines most basic assumptions of international relations theory; though many scholars have made this argument in recent years, the research here demonstrates the point. Leaders in every case also employed the events and actors in the international community to make their case about the role of the ingroup at home and abroad.
In reference to another international dimension, this study shows how the mediation and intervention behavior of other states can push a leader to consider these foreign actors as a part o f his constituency that needs to be mobilized so that he can acquire or maintain his position as a “legitimate” voice of the nation. Most literature on nationalist mobilization focuses on the ingroup as constituency. It does not explore the implications for group identity and mission when the elites who seek to create images of the latter must also be responsive to the international community or particular actors within it. Also, this study revealed that leaders do not only have to be responsive to the international community. The “other side of the coin” discovered here is that leaders may be aggressive in pulling in the international community, calling on external actors to be responsible for the resolution of societal strife. In the seven cases here, a few of the important actors were the United States, the European Community, and the Front Line
States. In more recent and future cases of this sort, however, these and a wider range of states, international organizations, and non-govemmental organizations may become increasingly involved— or be pulled in by the leaders. Especially for Afiica, groups such
330 as the World Bank and IMF, Amnesty International, and Afiica Watch come to mind
Further research on current cases is needed to explore the role of these kinds of actors, as
noted above.
Because this study shows how the external dimension can affect competing
nationalist leaders’ mobilization strategies and their success, it makes a contribution to
work on two-level games and the broader literature that emphasizes the importance of the
effects of international variables on domestic politics (for example, Evans, Jacobson and
Putnam, 1993; Gourevitch, 1978; Putnam, 1988). Putnam’s (1988, p. 434) work shows
that when leaders engage in negotiations, they have to pay attention to their “win-sets” at
two “game boards”: the domestic and the international. Leaders’ preferences at the
international game cannot be understood as “rational decisions” unless the preferences of
the domestic constituency the leader must also satisfy are understood. The larger the win-
set for the domestic game, the less constrained the leader will be (1988, p. 437). One of
Putnam’s (1988, p. 442) propositions is that the size of this win-set at home (“Level IT’)
“depends on the distribution of power, preferences, and possible coalitions among Level 11
constituents.” He argues that at times international issues provide the opportunity for
minority voices to become more powerful on the domestic fi"ont.
The case studies here show that when two leaders compete to be the
representative of a nationalist group and to define the identity and mission of the group in
more inclusive or exclusive ways, they may form “alliances” with actors at the international game board that help boost their definition of the situation even if other contextual variables favor the opposite appeals. This is argued to be a contribution for
331 two related reasons. First, the situations discussed here are not about a leader at an international game board trying to deal with domestic constituencies, but are cases where leaders are competing to get to the game board. Because this is not an uncommon phenomenon in situations of international involvement in ethnonational politics, it is important to consider how two-level game literature can offer insight (about how the international dimension affects this kin d of domestic politics). Second, the findings suggest that the “distribution of power, preferences, and possible coalitions...” at home in this category of situations can be significantly altered by the behavior of actors at the international game board. The dramatic effects of international support for Gerry Adams in 1996 is a case in point. Symbolic recognition provided by the granting of a visa by the
United States and the increasing communication with Adams by Britain and the Republic of Ireland signaled to the public that this erstwhile terrorist was in fact a legitimate leader.
7.4 Contributions to Work on Framing
The findings about the relationships between the political opportunity variables and the kinds of strategies successful leaders use also make an important contribution to work on fi'aming, both in the social movements literature and the political psychology literature on fi'aming. First, for the social movements literature (Gamson, 1992; McAdam,
McCarthy and Zald, 1996; Oberschall, 1996; Snow et al., 1986), the focus on framing is relatively new and only a handful of empirical, comparative studies have been conducted.
In the studies that have been done, the concept of “fi'aming processes” is often used more broadly than it is used here. Drawing on early work of Goffinan (1974), framing
332 processes are conceptualized broadly as the way in which movements develop a shared
social construction of ideas about grievances. As noted in Chapter 1 of the present study,
a more specific definition has come to the fore in the social movements field: fi'aming
processes are the “conscious strategic efforts by groups of people to fashion shared
understandings of the world and of themselves that legitimate and motivate collective
action” (McAdam, McCarthy, and Zald, 1996, p. 7). Even though there is the emphasis
on conscious strategy, movement leaders as political actors receive little treatment in this work. It is shown in this dissertation that the focus on leaders in this broader process is indispensable (see section 7.5 below), even though movement theorists pay less attention to the role of these entrepreneurs.
A second issue is that most of the extant work takes a particular case where successful mobilization by a single leader or social movement has occurred and looks at the fi*ames that were employed. Prominent examples are Oberschall’s (1996) work on the democracy movements in Eastern Europe and Snow’s (1986) work on the Iranian revolution. While these studies point out the importance of framing on mobilization, they do not consider situations in which there was actually competition to define frames or where leaders (or movements) were both successful and unsuccessful. Without variation on the frames used or on the outcome, it is not possible to know much about the actual effects o f the frames; this work is primarily descriptive."* In the present study, in most
^ I am not arguing that purely descriptive studies are not important. My point is that in order to make a contribution to understanding about the effects of frames in the mobilization process, it is imperative to have a more appropriate research desigru 333 cases the successful and unsuccessful leaders used very different strategies (exceptions were NI/1996 and Zim/1980), allowing a greater understanding of the role of the strategies themselves—and also the way the contextual variables affected the success of the strategies employed.
On a more fundamental level, the method used in this study to derive types of frames from work on nationalism is shown to be a very useful way to supply a substantive dimension to the idea of frames. Because the strategies were derived (from theory about nationalism, identity, and collective behavior) and the indicators for the strategies were developed before any coding was done (and then minor adjustments were made from the intercoder tests), it was remarkable to this researcher that the strategies were so frequently and apparently visible in the words of leaders. The fact that they were not imputed in the data by the researcher is shown by the intercoder reliability scores.
Work on framing in political psychology is most developed in the context of laboratory experiments (see Iyengar and Kinder, 1991; Iyengar and Kinder, 1987; Nelson,
Clawson and Oxiey, 1997), though some work has been done with survey data (for example, Zaller, 1994). This work demonstrates how the perception and interpretation of information is influenced by differences in framing but is limited in the insight it can offer to an understanding of leader framing, the persuasiveness of these frames, and the factors which influence how these frames become shared. The present study shows that these are important issues that need to be considered in future experimental work.
Most relevant to the arena of international politics is the work on problem representation (Sylvan and Voss, 1998; Sylvan, Ostrom and Gannon, 1994). Although
334 this work tries to get at the perceptual dimension (how decision makers represent a
particular issue), the idea here is that the propaganda or the public, persuasive problem
representation is important. To this literature, the comparative case approach that takes
into account the role of contextual factors offers insight about how external variables
promote the “believability” of some problem representations over others. If future
experiments on problem representation could take the variables found here to be important
and work them into an experimental manipulation, more could be learned about the
relationship between the way people see their environment and how they respond to elites
positing different descriptions of that environment. For example, it may be possible to
manipulate the contextual dimensions by providing the subjects with different scenarios
about repression and what international actors are doing (since these were two of the
more interesting factors), ask them to read campaign literature from competing leaders
using different kinds of strategies, and then have them choose a candidate. Since case
studies are limited in what they tell us about the “mechanisms” in the relationships among
leaders, followers, and context, this method may offer a more precise understanding of the
political psychology of mobilization.
7.5 Contributions to Leadership Studies
Previous work by Shamir (1994), Smith and Smith (1994), De Waal (1990), and
Campbell and Jamieson (1992) has shown the importance of leaders’ rhetoric in their relationship with followers; in crisis situations or other times of uncertainty, political talk is not cheap; words are deeds. As Campbell and Jamieson (1992) note, leaders do not
335 merely adapt to their audiences, they transform those who hear them into the audience the
leader desires. The kinds of cases in the scope of the present study show that this
situation is not as straightforward as these authors imply, though: in many cases of
nationalist conflict there is competition betv^'een or among leaders to define this audience,
which includes not only telling them who they are but also what their destiny is.
To lend greater specificity to the assertions o f the authors above, a major effort in
this study is to look comparatively at the relationship between leaders, (potential)
followers, their competitors, and the contexts within which they attempt to mobilize. As
illuminated by Hermann (1986), there are several images of leadership—each warranting a
research focus on different elements. The “pied piper” image calls for a focus on the
characteristics of the leader and implies that the leader who succeeds in recruiting
followers has particular traits that endow him with leadership. This approach has difficulty
explaining mobilization shifts away fi'om this leader, however. The second image is the
“salesman.” The leader has to be sensitive to people’s needs and persuade them that he can meet those needs. Neglected here is the situation of great uncertainty; as with the cases in the present study, people may be in crisis or even despondent about their sense of the future: What can we do? What do we need? And, who are “we” anyway?
In a third image o f leadership, the leader is the “puppet”—to understand leadership we have to know the goals and expectations of the people, and the individual who becomes the leader is the agent of these desires. The cases in the scope of this study show the shortcoming of this approach: it is problematic to take nationalist groups as a whole and sort out their expectations. Not only may potential followers lack a clear sense of
336 goals, but they also may be very divided in their desires. The final image is the “fire fighter.” Here it is necessary to understand the environment because that is what provides the demands on the leader, the constraints the leader faces, and even the choices for the leader and his followers. The comparative case study here does seem to point out the significance of the environment, but also shows that we need to know some of what the other models emphasize as well.
The Northern Ireland and Zimbabwe case studies demonstrate the significance of understanding a combination of these elements because it was discovered that (1) there is a relationship between the kinds of strategies the leader used and the success he had in persuading people to follow him, and (2) there is a relationship between the more objective treatment of the four political opportunity variables and the leader’s emphasis on aspects of this context. Previous sections focus on the former, therefore this section will expand on the second discovery.
If a leader was more inclusive in his strategies, when he mentioned any of the four aspects of context (see Table 7.2) he would also do that in a more inclusive way. For example, with the change in stability of elite alignment, the more inclusive leaders discussed this part of the environment when the change was favorable to altering the status quo alignment keeping the political minority down (which was hypothesized in
Chapter 2 to favor inclusivity). In contrast, more exclusive leaders only talked about this variable when it changed in a way to harm the opportunities of the political minority. A second example is the international mediation/involvement variable. More exclusive leaders played up the negative aspects. For instance, Gerry Adams from 1982-1987 really
337 focused on the common situations of Northern Ireland Catholics and blacks under
apartheid, Palestinians, and anti-capitalist insurgencies in Central America. He painted the
conflict in global terms: capitahsts versus the people, the West versus small states
(Ireland). In contrast, Hume emphasized how the international community could help
Northern Ireland fix its problems. In 1996, when Adams became dramatically more
inclusive, he was also talking about the international context in a much more positive,
integrative way.
On the other hand, the case studies also demonstrated how the context may have pushed the leaders to adopt some strategies over others as they react to what “worked” and what did not “work” in previous mobilization episodes; they may prioritize diSerent aspects of the context. An effort was made in the case studies to predict the strategies that each political opportunity variable was expected to promote (by making the audience more “susceptible” to either more inclusive or more exclusive strategies). In Zimbabwe
1990, the predictions were indeterminate because the repression and the international involvement variables were expected to favor a leader using more exclusive strategies, while the regional integration and the change in stability of elite alignment variables were expected to favor a leader using more inclusive strategies. The more exclusive leader was the “winner,” which offers some insight that speaks to the problem of having an indeterminate context.
When the substance of Tekere’s appeals are examined and compared with the substance of Mugabe’s appeals, an interesting pattern is visible. When Tekere discussed elite ahgnment, his focus was very negative. His focus was on Mugabe and how he had
338 monopolized the government in a way that was ignoring the demands of the nascent
opposition. Mugabe’s discussion o f this variable was o f course much more positive,
emphasizing the necessity of maintaining the stams quo to meet the needs of the nation.
As for the repression variable, Tekere’s appeals put a very negative spin on the Mugabe
security regime’s tactics. Mugabe did not focus on repression at all. Tekere said httle
about the regional arena while Mugabe played up the inclusive aspects (his own leadership
within the Front Line states as they worked together to boost their economies and to fight
apartheid, for example) while ignoring the trends toward multipartyism taking hold there.
Finally, Tekere placed much more emphasis on the international arena than Mugabe;
Tekere highlighted that Mugabe’s one-party, socialist state plans would take Zimbabwe
down the destructive path Eastern Europe’s former communist governments were on at
that time. Mugabe, in contrast, insisted on Zimbabwe’s unique situation. Looking back,
from an analytic standpoint, a strong argument can be made that Tekere’s emphases
provided a closer fit to what was really going on in Zimbabwe and the world and to what
people were really feeling in Zimbabwe. The gains of ten years that Mugabe spent so
much time highlighting were not “felt” or observed by the people at all. Cases in Northern
Ireland also demonstrate this relationship.
These matches show that understanding a leader’s pubhc problem representation
tells us how the more “objective” characteristics of a situation get interpreted or weighted
differently by each leader. In line with Brecher (1972), this study lends support to the idea that we cannot fully understand the roles of elites unless we focus on the match between the objective and perceived environments; though the argument here is not that strategies
339 measure leaders’ perceptions, the fact that they are ways the leader tries to get the
audience to perceive the situation makes this insight equally important. In future
research, expectations about the match between more exclusive or inclusive strategies and
particular contexts should be informed by the analyst’s comparison of an “objective”
assessment of the values o f the four variables with the way in which each leader frames those contextual factors.
One last contribution to leadership studies is important because it directly ties together leaders, audiences, and a particular aspect of context that can change this relationship. As noted, to understand leadership two important ingredients are knowledge about whom the leader sees as his constituency (Hermann, 1986, p. 180) and to whom the leader is accountable (Hermann, 1986, p. 170). Though it has done so indirectly, this study has shown that a leader may use dramatically different strategies and mobilizing resources depending on the particular audience he seeks to mobilize. The clearest example is the contrast between Hume and Adams in 1982-1983. In this period, Adams was considered the more successful leader; when the demographics of Sinn Fein and
SDLP voters are compared, SF mobilized working-class Catholics who tended to be younger and were hardest-hit by the economic inequahties in Northern Ireland at that time.
Hume tended to get the vote of middle-class Catholics (Irvin and Moxon-Browne, 1989).
Adams used rallies in working class areas and party supporters painted slogans on walls to mobilize people, while Hume used newspaper editorials and canvasses in more middle
340 class areas. Even in the international arena, there were contrasts: Hume appealed to governments and the European Community for help; Adams condemned governments and the EC and appealed to the camaraderie of the ANC and PLO.
Once Adams was more accepted as a legitimate political actor in the international community, however, his constituency broadened and therefore so did the range of actors to whom he was accountable. In 1996 and 1997, more inclusive strategies used by Adams
(compared to the two previous periods) coincided with his need to convince external actors such as the Clinton Administration that he was keeping up his end of the deal by trying to get his domestic audience behind a permanent peace agreement.
A tough issue arises from this consideration of audiences and accountability, concerning the limits on leaders such as Adams who came to power through exclusive appeals. The sanctioning of Adams by the international community puts him and leaders like him in a dilemma: they need support from the international community to give them legitimacy at the negotiating table, and in turn generally have to moderate their strategies to some extent—using more Governance strategies, fewer Enemy Image strategies, and more inclusive Identity strategies, for example. At the same time, they need to increase support at home, because as they have become more inclusive they find themselves in greater competition with the leader who had been more moderate in earlier times.
Another layer of complexity is added because this leader who initially came to power with more exclusive appeals has to maintain support of the original followers, essentially persuading them that the definition of the situation and the means to address it are different than they were initially. We see these challenges to Adams at this writing: the
341 jury is still out on if and how fully the April 1998 peace agreement will be implemented,
because the IRA and SF hardliners refuse to give up their weapons before other aspects of
the agreement are secured.
Though a counterfactual, we could surmise that more attention to democratization
in Zimbabwe in early 1990 may have garnered even greater support for Tekere, and then
in later years may have pushed him to be more inclusive (as discussed above). Future
work on democracy movements in Zimbabwe since 1990 can explore this issue, but for
now it does appear that the basic conundrum facing Adams would also face Tekere and
the other pro-multiparty forces. They take support away from Mugabe by being
exclusive; specifically for Tekere, this meant playing up urban problems and economic
deprivation. His emphasis on the lack of progress on land redistribution did not seem to
pull most peasants away from Mugabe, however. To really take power from Mugabe,
competitors will have to find appeals that get everyone behind them.
In sum, this section has discussed several major contributions the present study
makes to studies of leadership and has raised additional questions to be explored in future
comparative work. The next section returns to the international aspect of the study and its
foreign policy imphcations.
7.6 The International Dimension and Policy Relevance
A significant aspect of discovering (and exploring further) the role of the international dimension is its policy relevance to foreign policy decision makers in states and organizations not party to the conflict. The United States and others oftea have an
342 interest in whether a leader promoting a more or less exclusive view of national identity
and mission wins broad support; for example, fear that Zhirinovsky would come to power
in a post-Soviet Russia fueled a substantial US commitment to Yeltsin. In recent years the
West’s interest in electoral competition has skyrocketed, as free and fair, multiparty
elections have become a condition for admittance to the “international community” and
access to foreign aid (Sisk, 1998; Sisk, 1996). We expect elites in states emerging from
civil wars and recent peace negotiations “to legitimate their rule through the ballot box”
(Sisk, 1998, p. 146). This trend of international intervention through promoting electoral
competition—or using foreign policy to create potential mobilization episodes—makes
understanding the role of external actors increasingly important. One of the most
significant contributions of this study is the finding that the behavior of the international
community as a “whole” and actors within it appear to have many charmels through which
to shift the political opportunities toward one leader or another. This is not an argument
that international actors have the power to manipulate which leaders win, but from these
cases there do appear to be circumstances where the behavior of foreign actors play an
important role in reinforcing one leader’s “definition of the situation” over another’s.
The international dimension of the political opportunities consisted of two factors: international involvement/mediation and regional integration. The former in fact has several dimensions: norms of conflict resolution; norms concerning sovereignty and intervention; and international opinion, especially on human rights issues. All of these were important to varying degrees in the seven cases. Recent memory points out the role of the US when Gerry Adams was granted a visa for the first time. Although we cannot
343 attempt a counterfactual analysis here, it may be argued that Adams would not have received such a high percentage of the vote in 1996 if he had not been recognized as a legitimate actor—and not a terrorist—by the United States. At the same time, this behavior of the United States pushed Gerry Adams to broaden the boundaries of his constituency.
In terms of the elite alignment variable, the international dimension enhanced the prospects of a leader who was more inclusive because the actions of third actors such as the United States and the European Union at times pressured Britain to work more with the Republic of Ireland and to loosen ties to unionists when nationalists were suffering because of those ties. In Zimbabwe, the international involvement in 1979 and 1980 may have had the same effect; though the change in stability o f elite alignment taken as a single variable promoted exclusive strategies, in combination with the international dimension we see that the international community in fact pressured those actors maintaining the status quo through alignment to go to the negotiating table.
The international dimension may have also affected the way repression became a political opportunity for a leader promoting more exclusive strategies in 1997, but constrained the same leader in 1986-1987. In 1997, Adams drew on international opinion to condemn state violence; similar claims in 1986-1987 were ignored as the international community promoted negotiation among parties. Indeed, in the late 1980s, lessened repression was praised by the United States and Europe, perhaps giving greater validity to
Hume’s more inclusive strategies.
344 Political processes in the international arena can also affect whether a leader’s
Identity strategy resonates with potential followers. The progress of integration in
Europe, paired with the fact that Hume has been successful in bringing so much European economic support to Northern Ireland, surely made his appeals stronger that all people in the British isles share the European identity and have many things in common based on it.
Though the regional variable was not as important in the Zimbabwe cases, increases in regional integration did promote more inclusive Identity strategies. This tells us that regional initiatives should be supported in order to pull a nationalist conflict “out of itself,” as people in Northern Ireland often say about the role of the EU.
7.7 Conclusion
In summary, scholars and foreign pohcymakers can learn from this study how actors in the international arena who want to see more inclusive leaders succeed in places where there are protracted social conflicts can have the greatest chance of affecting the identity debate. The findings indicate that such actors can (1) change relationships among the parties involved in the conflict, (2) push leaders toward thinking of their constituencies more broadly, and (3) lessen the pressures on the leaders' followers that make them vulnerable to exclusive appeals. Exclusive strategies like appeals to injustice are more likely to fall on deaf ears when the society as a whole is prospering.
The present research has shown that leaders are not only "pushed" by their international and domestic contexts to adopt particular strategies but they can manipulate aspects of the international arena to draw in external actors and gain legitimacy for their
345 own "definitions of the situation." Indeed, in situations of uncertainty, leaders can use the
power o f fi'aming to make a situation "real" for their followers. A good example o f how
leaders can engage in this behavior is evidenced in this study by John Hume. He
consistently emphasized strategies that proposed tolerance among the various disputants
in Northern Ireland, which would be more feasible if living conditions were better and the
institutions o f governance could be changed to more equitably regulate political life. He
spent his career lobbying the United States Congress and the European Union to support
Northern Ireland as a whole, both economically and in any negotiations to bring about
peace. To the degree that these international actors got involved, as he argued they
should, leaders were forced to use more inclusive strategies.
While the study described here focused on two countries, the findings are
generalizable to an important set of cases. First, they are relevant to countries that have
been democratized to the extent that there are contested elections with genuine
competition for power - that is, there are observable electoral shifts. Second, they are
applicable when a crisis of legitimation has led to a period of uncertainty and there is
competition for the votes of the political minority. The number of cases meeting these two criteria is increasing as the holding of democratic elections is viewed as a prerequisite for entry into the club of aid-deserving states (Sisk, 1998). Further, the Carter Center and other election watch groups around the world have been documenting recent elections in states where political minorities have an increasing role in politics; this makes it relatively straightforward to judge whether the elections are viewed by area specialists to
346 be fairly contested. Examples of a few cases that meet these criteria include the 1994
election in South Africa and the upcoming (1999) elections there, the post-civil war
elections in Nicaragua where Chamorro defeated the Sandinista government, and the first
multi-party elections in Zambia where Chiluba defeated Kaunda.
This study represents an initial step in exploring how leaders use inclusive and
exclusive fi*ames to mobilize followers during crises of legitimacy. The results suggest a
number of possible directions for future research. First, given the ever increasing number
of international actors playing roles in internal mobilization contests, it is necessary to consider the effects of involvement by a range of actor types. In this study, the most involved international actors were states, followed by international organizations
(especially the European Union, the Front Line States, and the United Nations). Non governmental organizations such as Amnesty International and other types of international organizations such as the IMF and World Bank have become more visible in protracted social conflicts, as have small groups of scholars who work on “peacebuilding” or intercommunal reconciliation through small group interaction. It is unclear whether or how these other actors affect the mobilization process.
In addition, these different international actors may work at cross-purposes; that is, the behavior of the IMF may encourage more exclusive strategies while Amnesty
International’s involvement may promote more inclusive strategies. We will not have a complete understanding of the international influences on mobilization until we conduct empirical studies documenting the activities and testing hypotheses about the combined
347 influences of these actors. Updating the Northern Ireland and Zimbabwe cases and exploring other current cases will shed light on these questions.
Second, one of the most important follow-up studies to this dissertation will be a contingency analysis of the leaders’ speeches (for the cases already examined here).
Basically, this kind of analysis would provide a more clearcut idea of the associations among the strategies the leaders were using and the targets of those strategies. The end result would be similar to a conceptual map of the leaders’ problem representations, allowing better inferences to be made about how the leaders saw their strategies as more or less exclusive in particular circumstances. Following on the basic mapping for each case would be a comparison of these maps over time, to see when and how leaders began to change the way they posit who the allies and enemies are, as well as the identity o f the ingroup and its situation. Again, similarities between changes in the problem representations with changes in the context will allow a finer examination of how the leaders were manipulating the situation and how the situation was pushing the leader as well.
A third possible direction for future research would explore the relationship between the international involvement variable and the repression variable. Current work in International Relations is concerned with linkages between internal politics and international action/reaction. This study is well-placed to directly address this work.
Based on the findings across the seven cases, three propositions were stated about the conditions under which international involvement can temper or exacerbate the effects of repression on promoting more exclusive strategies. Especially intriguing is the idea that
348 repression on the community can increase (which generally promotes more exclusive
strategies), but the situation can still remain “ripe” for a leader using more inclusive strategies if important actors in the international community consider the situation an international concern and draw attention to the repression. Because this proposition was drawn out of the cases here, it is necessary to explore its validity in additional case studies.
Given the great variation of international involvement in cases with similar high degrees of repression, these issues lend themselves to comparative study.
A fourth avenue for future research involves the effect of the regional integration variable, especially on the use of more inclusive Identity strategies. In the African case, this variable was rarely expected to be significant. In fact, it may be necessary to study cases (from Afiica and Europe) from the last few years (and perhaps in Latin America in the future if the Free Trade Area of the Americas plan proceeds) to compare the role of increasing integration. The best way to explore the role of regional integration may be in a comparative study of European cases (from the Single European Act in 1985 to the present) such as Northern Ireland, Belgium, Spain, and Italy, for example.
Overall, the present research has shown that research considering only the context or only leaders’ appeals is likely to fall short o f understanding why leaders are or are not successful in their mobilization efforts. Instead, it is crucial to know more about the match among leaders, constituents, and situation to truly know why a more inclusive or exclusive message will resonate. In the present era arguably more than any other, the international dimension has become a significant part of the domestic mobilization process.
One of the major contributions of the present study is a method with which to explore this
349 phenomena. Indeed, in some circles of International Relations scholars, the domestic/international or internal/external divide is decried as artificial and obstructive to our theorizing about the world.^ The approach taken here offers an empirical "first step" toward understanding the reciprocal relationships of domestic and international processes.
^ In fact, this problem was the focus of a keynote address by James Rosenau at a recent conference attended by this author ("International Institutions: Global Processes~Domestic Consequences,” Duke University, Durham, NC, April 9-11, 1999). 350 APPENDIX A
HANDBOOK FOR ASSESSING MOBILIZATION STRATEGIES
I. Texts
The first issue to consider in any content analysis is the type of text that will be selected for coding. Public pronouncements, such as speeches, articles, editorials, in which the leader appeals to potential constituents are the appropriate texts for the fi'amework presented here. The ideal study might investigate appeals in a number forms (interviews, speeches, editorials) and to a range of audiences (domestic, international), and then compare results according to audience. However, for any case there may be data limitations, as there were for the cases investigated in this study. The most comparable set of speeches both over time and between leaders for the Northern Ireland cases are speeches at the annual party conferences (see below). For the Rhodesia/Zimbabwe cases, speeches, radio broadcasts, and manifestos delivered around campaign times were the most feasible texts. Although I argue that the texts used in this dissertation are parallel within the country cases, there are difierences between Northern Ireland and Rhodesia/Zimbabwe. These differences are unavoidable given data availability and the fact that there are simply differences in the political systems in terms of the channels through which leaders appeal.
The main texts that are used for the Northern Ireland leaders are the speeches given at the parties’ annual conferences. These were consistently available for every year and most clearly establish the parties’ views of the current situation and future agenda. Further, in Northern Ireland the content of these speeches is often reported in the media so that the central messages from the leader often is heard beyond those attending the conference. ‘ Occasionally, especially around significant events in the course of the conflict, the leaders delivered addresses which were attended by significant numbers and widely reported. These additional sources are also coded; however, the coding takes note of these different kinds of texts and audiences so that they may be compared to each other. There were not differences in the strategies used across these two kinds of speeches.
' These are also ideal sources of rhetoric for future analyses because each party has a website on which they post these armual addresses as well as other speeches by party leaders.
351 For the Rhodesia/Zimbabwe cases, the texts used include radio addresses, speeches made to election rallies, and manifestos. These “venues” are more prominent than annual addresses made to party conferences, especially since those leaders competing with Mugabe after independence did not hold annual conferences.
EL Unit of analysis
The second issue to be addressed in any content analysis approach is the unit of analysis for the coding. The unit here is the topic; for each topic delineated by the speaker, the coder will record the ingroups, outgroups, and strategies connected with these groups. The statistical analysis then allows comparisons (over time, between leaders, and by topics) of which groups are deemed relevant, how people are depicted in terms of group identities, and which strategies appear with particular group delineations. In trial coding, speeches have had a range of approximately 18-25 topics each. The topic is the optimal unit of analysis, as opposed to the sentence or paragraph, because it allows the analyst to take a more inductive approach in observing the strategies and groups the leader pulls together when discussing a topic.
A hst of topics has been compiled as part of the coding rules, to help the coder determine which passages are of interest (see below). About ninety-five percent of the speeches tend to address these topics and therefore should be coded; passages that fall outside the material of interest tend to deal with very specific internal party business. The kinds of topics related to mobilization attempts deal with the situation in the country, characterization of others, current negotiations, and the policies of the current government, for example.
Every time the leader changes to a new or different topic from the one immediately preceding, the coder should begin a new coding sheet. This makes the coding more tractable and makes it possible to see which groups and strategies are used together for each particular topic. On the coding sheet, there is a space to enter the code for the particular topic discussed. The numbers merely identify the topic; they have no ordinal or scalar values. The speaker may return to the same topic several times; each should be given its own coding sheet as long as the one topic is interrupted hy others, or the subject of the topic changes. In other words, in some of the topics, such as “characterizing competitors,” the coder should consider it a change in topic when the identity of the competitor(s) changes.^ Often a passage may appear to cover more than one of the topics; still, trial coding has shown that one topic is generally emphasized more than others and coding of topics shows a high degree o f intercoder reliabihty. The coder should select the topic heading that best describes the most dominant theme. There is another column of the coding sheet for the number of words within each topic. This word count provides a rough measure of the importance of that topic to that speaker, along with the number of times he uses the topic.
■ In one topic the speaker may be focusing on the role of actor A but then it should be considered a change of topic when he changes the subject to the role of actor B. 352 The following list of topics was created from both theoretical interests (drawing on mobilization and group conflict literatures) and from a reading of a broad sample of speeches.
Topics and codes :
1 Situation in country, where main focus is party’s position within scheme of things: may be discussing problems in the country; code 1 when emphasize party’s role (go to 2 if emphasis on blaming others). Also use code 1 when leader discusses performance of party, such as accomplishments or obstacles remaining. Special cases/notes: Often Gerry Adams will greet and/or declare solidarity with prisoners and the families of prisoners. When a leader does this, code 1 because he is expressing the party’s view. However, do not code those passages where the leader gives long lists of names of prisoners. When the leader discusses the internal situation or politics in the Republic of Ireland, code 1 when he talks about either the party’s role there or connects the situation in the Republic to the situation in Northern Ireland; do not code when the speaker’s discussion does not include one of these connections.
2 Situation in country, where main focus is its origins (such as blaming others, structural causes, etc.); use code 2 if the leader talks about proximate causes or looks back to history for the origins of the conflict/situation. Also use this code if the leader characterizes other groups (may be as allies or competitors), because he will generally do so in the context of discussing the situation. The exception is if the leader discusses other parties other groups in context of others’ role in talks, which i? code 4.
3 Links with the international dimension: international allies or enemies, making analogy of Ireland’s situation with others. Use code 3 if the leader mentions groups, regions, parties, or other international actors. Analogy o f Ireland’s situation/problems with global problems or problems in other areas: if the leader talks about a situation in the world and implies that something about that situation is similar to the situation in Northern Ireland/Ireland (similar kind of enemy, for example). Also use code 3 if leader makes an analogy of dealing with conflict between Ireland and other areas (regional, single countries, globally).
4 Current talks/proposals: whether or not speaker’s party involved, code for this topic when the leader discusses any talks among actors in the conflict, or any proposals. To be coded for this topic, the speaker may be describing the talks/proposals, others’ roles in the talks/proposals, or others’ reactions to them. Also use code 4 when leader discusses party’s/ingroup’s role in talks/proposals. Here this is distinguished from 1 above because here the leader will be focusing on talks or proposals instead of the party’s stance on the situation in general.
353 [Always start a new topic if the general subject of talks remains the same BUT the emphasis shifts, for example to a different group. This is to keep the size of the topics manageable, since these passages may be very long.]
5 Future/facing challenges/looking at the “next generation”: Use code 5 when the leader talks about the young, the next generation, and the future challenges faced by the country or groups in the country. The leader may mention the past in a qualifying clause, but if main emphasis is on the future, use code 5 and not 1, above.
NC Use this notation when the subject of a passage falls outside any of the topics described above. m . Overview
Once the coder reads through the speech and marks off each topic, the coder examines each passage for information about the following variables and enters the information on the coding sheet (some of the variables such as speaker and time can be entered initially):
topic speaker time (year of speech) number of words sheet number outgroup or ingroup? group reference group delineation Justice strategy/negative Justice strategy/positive Outgroup Image strategy/negative (Enemy) Outgroup Image strategy/positive (Ally) Governance strategy/negative Governance strategy/positive Storytelling strategy
Speaker
1 Hume 2 Adams 3 Mugabe 4 Muzorewa 5 Tekere
354 Time period
1 1982-1983 2 1986-1987 3 1996 4 1997 5 1978-1979 6 1980 7 1989-1990
Number of words in topic
Sheet number (numbered continuously throughout dataset) Outgroup or ingroup (for data entry only)
1 outgroup 2 ingroup
Ingroup is Leader’s Party (for data entry only)
99 group is outgroup 0 party 1 not party
IV. Identity Strategy: Group reference and group delineation
Because the objective is to compare which people/groups the leaders deem relevant and the labels or bases of identification that different leaders (and the same leaders over time) use to describe potential groups in the society, the coder should underline all the groups the leader mentions within each unit (topic). For this project, instead of using the particular labels (“Catholics,” “politicians,” “working class)”, a method more generalizable across time and cases is used to record the group delineation. Lists of labels that tend to be used to describe people in each country were created inductively, and then the lists were categorized according to degrees of exclusivity: very exclusive, moderately exclusive, moderately inclusive, and very inclusive.
It is not suggested here that there is an objective definition of “inclusive” and “exclusive,” but in order to observe differences between leaders and changes over time, it is necessary to assert a reference point. This project approaches this issue by considering the group reference: who are the people the leader refers to when using a particular label? The second code is the group delineation, for which the labels are assigned degrees of exclusivity based on the “perspective” of the group reference used with the labels. For example, consider two groups with the same group reference named by a leader:
355 “Nationalists in Northern Ireland” and the “working class folks in Northern Ireland.” From the perspective of the Northern Ireland group reference, “Nationalists” is considered “very exclusive” (because it is completely exclusive of those claiming British identity) and “working class” is considered “moderately inclusive” (since all are not in the working class in this area, it is not coded as “very inclusive”). For this study, the labels “Catholic” and “Protestant” are considered “moderately exclusive” because rehgion does not preclude membership in either state, where as “unionist” (and also “loyalist”) and “nationalist” (and also “republican”) are mutually exclusive.
Therefore, using the rules and codes below, the coder is to record which people the leader is grouping together (Group reference). This is an attempt to be as objective as possible; the coding then records the diSerent ways in which groups or potential groups in these areas are bounded and categorized (Group delineation). This section explains the coding rules for these two variables.
How the leader describes both ingroups and outgroups (or “self’ and “others”) is of interest. “Outgroup” is used here to mean any group that the leader distinguishes as an “other,” or not part of the “self.” “Ingroup” is used here to mean the group with which the leader identifies. For example. President Clinton may use the following ingroup descriptions: “Americans,” “Democrats.” He may name outgroups such as “French government,” “Republicans.”
An important issue is that identities are often nested, so that some outgroups may be discussed as subgroups of a larger ingroup. For example, “Irish firom the Protestant tradition” is surely an outgroup when spoken by the leader o f a predominantly Catholic party; but the fact that he is suggesting the shared dimension o f “Irish” may be crucial to understanding a dynamic conflict. This portrayal is at least more conciliatory than “Unionists” or “Protestants” alone, which emphasize no shared dimensions of group membership. The idea of nesting and presenting a picture o f group as multidimensional and crosscutting may be a key way of encouraging multiple loyalties. In this project, intercoder reliability could not be achieved when the coder is asked to code when an outgroup is talked about as a subgroup of a larger ingroup. Still, some sense of how leaders posit multiple levels of identity for the ingroup can be understood in the analysis, but examining the various ingroup references made.
An additional set of guidelines for coding are discussed here and then the codes are presented. Sometimes the leader may make important references to ingroups and outgroups in forms other than proper nouns. For example, “Irish repubhcanism” and “British repression” should be coded as references to groups. On the other hand, words that connote forces and not humans should not be coded as groups. An example here is “apartheid.” This would not be coded, but “Botha’s apartheid junta” would. The following list summarizes how to know which references to groups should be coded:
356 —Group names; “Irish republicans” or “African nationalists” —A more specific description of the pronouns “we” and “they” : “We are republicans”; “they are unionists” —Reference to a group that can be made explicit: “Our enemies in high places”; “the party who supported the legislative reforms”; note that contextual knowledge is necessary here to know who the groups are who may have supported or opposed particular measures. —Indications of ownership, actions, or intentions of groups: “British presence in this country”; “fi'eedom fighters”; “Dublin’s refusal”; “our position” (when it is clear fi*om a previous sentence who the “our” refers to) —Capital cities or political leaders when used to talk about governments: “Thatcher and Reagan support Dublin’s position” —Individuals as exemplars of groups: “John Hume attended the New Ireland Forum”
Group reference: This section reviews the discussion above with another example. Before the groups are coded in terms of labels, it is essential to know which people the leader is trying to label. For example, when someone says “Protestants” it is important to know if this label is given to what is considered the political majority in Northern Ireland, or if the speaker is using the island of Ireland as a reference point and therefore seeing this group as a minority. If the group reference is to people on the island of Ireland, “Protestants” is a very exclusive way of describing people. On the other hand, calling the majority in Northern Ireland “Protestant” is moderately exclusive (since more exclusive labels are commonly used: Orangemen, unionists, loyalists). This is important because at different times a leader may use the labels “Protestant majority” or “Irish of a different tradition.” In these other terms, the leader is potentially including and excluding different people.
As this example may indicate, coding decisions for the group reference may not be straightforward and require using the context of the speech or topic to determine which people are being labeled a certain way. Trial coding shows that this method is effective. In summary, the Identity variables (both for ingroups and outgroups) are linked, or have two levels. The first level (Group Reference) denotes the people to whom a leader is referring, and the second (Group Delineation) is the label a leader attaches to those people.
For the Northern Ireland cases, possible group references include people living in the borders of Northem Ireland (which may be called British or Irish or may be broken down into Catholics/Protestants or Nationalists/Unionists, for example), those living in the Republic of Ireland, those in the island of Ireland, those living in mainland Britain, people in both areas, those in Europe, and people in parts of the world or universal references (so people in Northem Ireland may be referred to as Christians, which has no geographic reference). For the Rhodesia/Zimbabwe cases, these include several subgroups in the borders of Rhodesia/Zimbabwe, those in Afiica, and universal references or those in other parts of the world. For example, the former may be broken down into Rhodesians meaning whites or Zimbabweans meaning blacks; it may be referred to as Zimbabweans meaning all people in the
357 borders. Note that this coding leaves room to record when those in particular borders may be defined as part of something larger. For the Irish cases, “Europeans” is a potential label; for Rhodesia/Zimbabwe, part of a greater Afiican nation may come into play; none of these are predefined as mutually exclusive. It should be apparent that for the Identity strategy, the coding is set up to allow the data to speak for itself so that the analysis is as interpretive as possible in recording the way leaders suggest group boundaries be drawn.
Group delineation: The “label lists” were created firom research on these cases and from trial coding on material that was not coded for the dissertation. For each set of cases, a scale was assigned to the label lists, fi"om very exclusive definitions to more inclusive labels, so that for the delineation column the possible values for both ingroups and outgroups are 0-4 (O if it is too ambiguous to code).
The exception is when the Group Reference is to “people in the world.” Because there are such a wide range of possibilities, it did not make sense to try to set up a scale. Instead, a list of labels with, numbers for each is given. This will help us see which international actors are talked about and how. For example, one party may always talk about international allies who are always called “oppressed people,” while another may often mention “democrats.” Note that for the group reference and delineation, there are codes for ambiguous references. For example, these may be used when a leader uses “we” or “them” but the coder cannot discern with high confidence fi"om surrounding sentences to whom the leader refers.
The following are the codes for the first two variables, presented first for Northem Ireland and then Rhodesia/Zimbabwe.
Northem Ireland
Group reference (which people is leader describing?)
1 present but ambiguous/indeterminate (use 0 for delineation scale) 4 people in Northern Ireland’s borders (go to 4 for delineation scale) 5 people in the (26-county) Republic o f Ireland (go to 5 for delineation scale) 6 people on the (32-county) island or in “region” of Ireland (go to 6 for scale) 7 people in Britain (go to 7 for delineation scale) 8 people in Britain and Ireland (go to 8 for delineation scale) 9 people in Europe (go to 9 for delineation scale) 10 people in world or universal references (go to 10 for delineation code list)
Notes: When there is mention of Sinn Fein, it is should be coded 6 here (it is an all- Ireland party), UNLESS the speaker is distinguishing that it is not. For the SDLP, use 4 because this party claims to be a non-sectarian party (but is only in Northem Ireland, not a 32-county party or an all-Great Britain and Northem Ireland party).
358 Group delineation (how is the leader describing the people?)
Using the hsts below, the coder should find the label that the leader uses. Then use the codes to find the degree of exclusivity the label has been assigned (1-4). If the label does not appear in the list below (which may be the case if it is a very specific group that only appears once), the code “0” should be used.
[See below for match between group labels and the exclusivity scale to be used here, except for 10 above]
0 indistinguishable very exclusive group boundaries moderately exclusive group boundaries moderately inclusive group boundaries very inclusive group boundaries
4 People in Northern Ireland’s borders
Very exclusive: British military RUC, UDR British, agents of the British pohtical prisoners/prisoners of war fireedom fighters/IRA Nationalists Unionists parties/representatives of parties (UUP/Alhance or APNI/DUP/PUP/SDLP/SF, etc.) terro rists/IRA/UVF Loyalists Republicans perpetrators of violence (when it is not clear they are speaking specifically about the IRA, UVF, RUC, British army, or other militant groups)
Moderately exclusive: Protestants Catholics our electorate (check context) other tradition middle class/establishment pohticians opponents (when not clear to whom referring) multiple parties
359 civil rights movement working class Catholics or Protestants sociahsts
Moderately inclusive: anti-imperialists ordinary people/working class six counties/people in the six counties young people/next generation democrats victims of violence unemployed
Very inclusive: people of Northem Ireland two traditions, both traditions colonized people oppressed people Irish European people of this community (check context)
5 People in the Republic of Ireland
Very exclusive: ERA political prisoners/prisoners of war freedom fighters parties (such as Fianna Fail, Fine Gael, etc.) Church
Moderately exclusive: Irish govemment/Dublin establishment/middle class multiple parties political leadership republicans Cathohcs
360 Moderately inclusive: oppressed people working class/ordinary people nationalists victims of violence
Very inclusive : Irish society/Irish people European
6 People in the region or island of Ireland (32 County)
Very exclusive: Northem nationalists/nationalists in the Six Counties IRA political prisoners freedom fighters party (Sinn Fein) Catholics Protestants Church Unionists/other tradition Loyalists families of prisoners perpetrators of violence/men of violence
Moderately exclusive: North/Six Counties republicans socialists middle class/Irish establishment political leaders/hip multiple parties anti-imperialists opponents, when not specified
Moderately inclusive: 26 Counties South oppressed people working class/ordinary people
361 nationalists/other tradition young people democrats victims of violence
Very inclusive: Irish people/nation/society our divided society (6-4 with a Justice Negative and Storytelling strategy) both Traditions/people of different traditions Europeans People in the 32 counties all the Irish parties (when include unionist/Protestants)
7 People in Britain
Very exclusive: British army British imperialists, or when it is clear focus is on Britain as an occupier pohtical prisoners
Moderately exclusive: British establishment/middle class British government/London pohtical leaders/hip multiple parties
Moderately inclusive: British people/nation/society ordinary people/working class democrats
Very inclusive: European
8 People in Britain and the island of Ireland
Very exclusive: socialists republicans prisoners
362 freedom fighters/anti-imperialists divided peoples British (Great Britain and Northem Ireland)
Moderately exclusive: middle class/establishment specific governments individual parties
Moderately inclusive: working class/ordinary people victims of war/violence multiple parties
Very inclusive: European people of both islands
9 People in Europe
Very exclusive: republicans individual nations
Moderately exclusive: middle class/establishment capitalists socialists EEC separate regions (when set in opposition)
Moderately inclusive: region working class/ordinary people democrats
Very inclusive: European members of the European Union
363 10 People in world
1 political prisoners 2 terrorists 3 United States 4 Proper names of particular parties (e.g. ANC, FLO), states (other than US), or peoples, when no explicit link is made in terms of ideology (see 12). **Use this code when leader talks about Irish America or Irish in other countries, and also mark “similar culture” on the Outgroup positive strategy 5 imperialists 6 the establishment 7 freedom fighters/anti-imperialists 8 socialists 9 nationalists 10 democrats 11 Europeans 12 Parties from other countries when a link is made in terms of ideology (the conservatives in all countries; but use 8, 9, or 21 if use those more specific labels) 13 oppressed people 14 totalitarian regimes/repressive regimes 15 capitalists 16 Western powers/the West 17 small nations/developing world 18 superpowers 19 ordinary people/people of goodwill 20 divided societies 21 republicans
Rhodesia/Zimbabwe
Group reference (which people is leader describing?)
1 ambiguous/indeterminate (use 0 for delineation scale) 2 people in Zimbabwe (go to 2 for delineation scale) 3 people in Afiica (go to 3 for delineation codes) 11 people in world or universal references (go to 11 for delineation scale)
364 Group delineation (how is the leader describing the people?)
2 People in Zimbabwe
very exclusive (code I):
—soldiers/militants/hreedom fighters/ZANLA forces/military cadres/comrades-in-arms —terrorists —prisoners/detainees —Smith regime, if not described as settler or imperialist (otherwise, see “world” list) —rebel leaders if only whites (2 if include Muzorewa, etc.) —proper names of parties (ZANU/ZAPU/UANC/RP or Rhodesian Front/Patriotic Front/Revolutionary Council, etc.) —party leadership, names of politicians when used to represent parties —the enemy (only if clear that the speaker means an enemy in the political system; otherwise see “world” list below; when this is used, also use “threat” code, outgroup negative) —puppets, puppet regime (when this is used, also use “betrayal” code, outgroup negative) —combination of Muzorewa, Sithole, Chirau —tribalists/sectionalists/regionalists —police/army —Shona/Ndebele —Executive/Parliament/Judiciary
moderately exclusive (code 2):
—Rhodesia/white Rhodesians —Zimbabwe-Rhodesia —Executive Council/transitional government —combination of Muzorewa, etc. and Smith or puppets and whites —rebel leaders if include whites and blacks —nationalist leadership —working class —establishment/white privileged class/black privileged class (if clear that group reference is Zimbabwe, otherwise see world) —politicians —white Zimbabweans/minority —white landowners/landlords —white farmers —capitalists, if clear speaker means in Zimbabwe (otherwise see world list below) —socialists, if clear speaker means in Zimbabwe (otherwise see Africa or world list below) —women/men (gender)
365 moderately inclusive (code 3);
—peasants/peasantry/povo/sons of the soil/those in Tribal Trust Lands (TTLs) —poor/poor peasants —innocent victims/victims of violence/refugees —families of fighters/civilians —our traditional society/our people —nationalists/nationalist movement —struggling masses/indigenous people —oppressed people/colonized people —Africans —blacks —black Africans —Zimbabwe, if clear that is excluding whites —our nation, if clear that is excluding whites —democrats —youth/children
very inclusive (code 4):
—people of all races —all parties —Zimbabwe, if not excluding whites —our nation, if not excluding whites
3 People in Africa very exclusive:
—individual countries or parties (South Africa, Zambia, Mozambique, Tanzania, ZANU, etc.) —Zimbabweans not living in Zimbabwe (including fighters in above countries) —whites —racists (though probably “world” reference) —imperiahsts (though probably “world” reference) —Southern Rhodesia —liberation movements moderately exclusive:
—Front Line States/Southern Afiica —West Africa —multiple countries
366 moderately inclusive:
—OAU (Organization of Afiican Unity) —oppressed people/dominated people —nationalists —socialists
very inclusive:
—Afiicans/Afiican brothers and sisters —people o f all races —Southern Afiica
11 People in World
1 settlers/coloniahsts/settler-colonial regime/white settlers/racists/foreign minority 2 British/British party/Pearce Commission 3 United States/Carter administration/Kissinger, etc. 4 the enemy/enemies 5 imperialists/the estabUshment/capitalists/rich 6 totalitarian regimes/repressive regimes 7 Western powers/the West/superpowers/Europe 8 international opinion 9 Proper names of particular parties (e.g. ANC, PLO), states (other than Britain, US, or Afiican states), or peoples 10 friends/allies 11 fi'eedom fighters/anti-imperialists/oppressed people/liberation movements/indigenous peoples (unless clear that group reference is Zimbabwe) 12 socialists/communists/Marxists 13 nationalists/democrats 14 Non-Aligned Movement 15 Commonwealth countries 16 small nations/developing world/poor 17 United Nations 18 East if leader says “Anglo-American,” use two groups; 2 and 3 above
367 V. Strategies used with Identity: Justice, Outgroup Image, Governance and Storytelling
After coding for the group reference and delineation, the general coding method is to read through the topic in which these groups are discussed and record which mobilization strategies are used with the ingroups and outgroups. Care must be taken to record the strategy with the proper group. In general, this means that no outgroup image strategies will be recorded on ingroup lines, for example. Matching the strategy with the group to which it applies is generally straightforward, but may be slightly more difihcult with the Governance strategy (rules for this are elaborated below). A key issue is making coding decisions about the frequencies of strategy use. Since this is related to recording the ingroups and outgroups as discussed above, this section begins where coding for the Identity strategy ended.
Frequencies and multiple strategies: An important issue concerns what to do when the same group reference/group label pair is used multiple times, because it is related to strategy frequencies. This is important in recording the strategies used, because each return to an aforementioned outgroup may include the use of another strategy, or the second use of the same strategy used the first time the outgroup was mentioned. If the leader uses the exact same description for a group (same reference and delineation) one or more times within the same topic, the coder should enter the codes only once (on one “outgroup” or “ingroup” line). Multiple references include using “they” (them, their, etc.) when it is cear the leader is referring to a group previously described; “The British government should persuade the unionists. They need to work with Ireland too.”"
Again, it is crucial to keep track of frequencies or uses of additional strategies for each outgroup. Since the topics may be as long as 500 words (though the average is about 250), it will not be unusual to have several strategies marked for each outgroup. For example, in the following sentence, the first outgroup is the British and the second is the Irish government; but the second reference to the British is an additional strategy to be recorded (on the same outgroup line as the first):
“The British are destroying our culture and have no plans of easing up on us. The Irish government should promote talks. The monetary pohcies of the British have led to massive unemployment, which has only promoted further conflict.”
Here the topic is 4 (discussing situation where emphasis is blame); the two outgroups are the British and the Irish government. The first strategy used for the British is Outgroup Image (negative)/threat and the second is Outgroup Image (negative)/blame.
The general rules for coding frequencies and multiple strategies are as follows. If several sentences serve to paint the picture of one strategy, the code is recorded only once.
^ None of these examples come from any speeches. I have made them up. 368 However, if there is then any interruption in this thought before the same strategy is seen again, such as a turn to another group or use o fanother strategy, the retian to any o f the strategies usedpreviously are recorded as another usage. Another case where multiple usage is recorded is when the leader begins a new paragraph using the same strategy with which the previous paragraph ended. This latter case is straightforward but the former (in italics) needs clarification. Take the following example:
“We, the Northem Irish people, suffer because of Britain. The injustice in our situation cannot be tolerated. People are without jobs, without decent housing, without the rights that every human should have. British policy for decades has taken every opportunity away. The British government’s monetary practices have made this a hopeless life. Our people deserve better than this.”
In this sentence, the ingroup is the Northern Irish people. There is only one outgroup, British government. In the first sentence, the speaker uses the Outgroup Image (negative)/blame strategy with Britain and the Justice/negative strategy with the Northem Irish people. The next two sentences are also the Justice/negative strategy, but do not count this as a second usage because it follows on the first. The fourth and fifth sentences go back to blaming Britain, and so a second usage of the Outgroup Image (negative)/blame strategy is recorded. The final sentence goes back to the first ingroup and thus a second usage of the Justice/negative strategy is recorded.
The codes used for entering the data for multiple strategies differs from those below, to make the analysis more tractable. These other codes are not given here but instead are discussed in the results section of the dissertation.
The subsequent sections take the mobilization strategies individually, describing them and giving the codes for each.
Justice strategy (negative/injustice)
99 absent 1 present
Justice strategy (positive)
99 absent 1 present
369 Key indicators for either of the Justice strategies include discussion of the situation of the group being described. This may be more common with the ingroup but not necessarily: leaders often talk about other groups or subgroups of the larger ingroup as well.
The negative variant o f the Justice (or Injustice) strategy is being used when the leader dwells on the fact that the group “deserves more than we are getting.” Here the group as a group (not just individuals) is subject to unfairness, is making unreasonable sacrifices, or is not being adequately rewarded for its efforts. Also code for this strategy when the leader discusses the struggle of the group and/or when the leader dwells on the poor situation in which a group finds itself. Phrases such as “the right to self- determination” and the “right to armed struggle” are also coded as this strategy.
For the positive variant of this strategy, look for mentions that the group is being treated fairly or more fairly than in the past, and/or of the recent gains that have been or are being made. Code for this strategy when the main theme is achieving success and overcoming particular obstacles.
Note: When the leader’s discussion fits into these descriptions, but the main emphasis is on institutions or policies that have caused the situation, see codes for the Governance strategy.
Outgroup Image—negative strategy (Enemy Image)
99 absent 1 present/type indistinguishable 2 threat 3 blame 4 betrayal
Outgroup Image—positive strategy (Ally Image)
99 absent 1 present/type indistinguishable 2 goals or values similar 3 situation similar 4 culture similar
These strategies will appear when the leader discusses any groups other than the ingroup and is not characterizing the situation o f the group in a way that is described as the Justice Negative or Justice Positive strategies. The emphasis is on the character of other groups, the groups’ roles in the conflict or in the situation of the ingroup, etc. The leader may emphasize positive or negative aspects.
370 For the Enemy Image strategy, there are several categories. Use code I if the leader is talking about a group in a negative way but the other three codes do not adequately categorize his discussion. Use code 2 if the emphasis is on the threatening nature o f the group. For this, the leader may not use the term “threat” directly, but may be creating a sense of fear or darkness about the group, that its presence is ominous. A good rule of thumb for distinguishing between threat and blame is to code for threat if the situation is ongoing, but if an enemy did something hurtful in the past, use blame. Use code 3 if the leader is blaming the group for any given situation (this may appear commonly with the Justice/negative strategy). Use code 4 if the leader discusses the outgroup’s role as a betrayal. This may require a contextual reading, but a general guideline is that this strategy is being used when the outgroup is a subgroup of a larger ingroup and is being blamed.
For the Ally Image strategy, there are also four categories. Use code 1 if the other codes do not adequately categorize the leader’s discussion of the outgroup but he seems to be characterizing it in a positive hght. Use code 2 if the leader emphasizes that the outgroup and the ingroup have common goals or values (both want to fight poverty, both want to get the British out of one place). Use code 3 if the reference goes beyond identification of common goals to a characterization that the two groups are in similar situations (both are poor because of same reason, both have British occupier and want to get that power out). Use code 4 if the emphasis is that the groups have a similar culture. Also use 4 if leader is talking about Irish Americans or Irish populations in other countries.
Governance negative strategy
99 absent 1 present/type indistinguishable*’’' 2 past institutions or alliances/local level 3 past/state level 4 past/international level 5 present/local level 6 present/state level 7 present/international level 8 future/local level 9 future/state level 10 future/international level
Governance positive strategy
99 absent 1 present/type indistinguishable** 2 past institutions or alliances/local level
371 3 past/state level 4 past/intemational level 5 present/local level 6 present/state level 7 present/international level 8 future/local level 9 future/state level 10 future/international level
The two Governance strategies are indicated by discussions of government institutions, laws, alliances with other elites or parties, proposals for change, taking part in elections, and/or talks with other parties (including a peace process), individuals, or governments (“governance factors”). The negative variant should be coded when ±e emphasis is on the harm that has been done, is being done, or will be done to a group by particular governance factors. Also, if the leader is discussing talks that failed, code for Governance Negative. The positive variant should be coded when the leader focuses on the benefits in the past, present, or future of particular governance factors.
The main emphases for which it is important to code (for theoretical interests) are the temporal dimension (past, present, future) and the level of the situation or relationships the speaker addresses (local, state, or international). These can generally be discerned but may be vague; if it is unclear use code 1 for the Governance positive or negative. Note: “future” applies to any institutions/talks/alliances which are proposed in ongoing negotiations OR when the leader expresses a need for particular policies/institutions/alliances.
For the Irish case, consider the level local when the speaker discusses local councils or talks among parties/actors in Northem Ireland. When the focus is all-Ireland, code this for state when it is clear that the leader sees it that way (most references will be this way because of the fact that the study is of Irish nationalists). When the focus is on British-Northern Ireland issues, code for state. When the focus is Irish-British, code as international (also any mention of additional states, international organizations, or regional organizations). **Although the more specific codes were recorded, for the analysis only whether the strategy was “present” or “absent” was considered.
Because this strategy involves (by definition) the mention of multiple groups, a difficult issues is with which groups the codes should be recorded: the particular ingroup(s) or the outgroup(s) which is/are being talked about as well? A rule of thumb is that it depends on the emphasis. If the leader is placing blame specifically or talking about the policies or proposals of a particular outgroup, record the Governance strategy on the outgroup line. If the leader is discussing the ingroup’s proposals or alliances of an ingroup with others, record the strategy on the ingroup line. If the leader says that the proposals made by an outgroup (etc.) are good for the ingroup, write the codes on both lines.
372 Special case; In the Irish case, it is common to refer to “Anglo-Irish talks.” When this occurs, record “Anglo-Irish” as an outgroup with codes for Britain/Ireland for the group reference (8) and moderately exclusive for the delineation (2), then record the Governance strategy used on this outgroup line. If the speaker then breaks these groups down into the Irish government and the British government, the separate groups should also be noted as outgroups. The linkage in the “Anglo-Irish” label is important. If it is clear that the leader is mentioning only the law and not the process (most likely in saying how an aspect of the law affects some group), write the code on the line of the affected group, and do not create a group with the 8-2 label.
Storytelling strategy
99 absent 1 present
The coding of the final strategy. Storytelling, is conducted differently than the coding for the others. The Storytelling strategy is treated as a mode, of presentation or persuasion whereby the leader “embellishes” the discussion that uses the other strategies (Identity, Justice, Outgroup Image, and Governance) by drawing on the past or a vision of the group’s destiny. The leader weaves a “story” around the imagined community, connecting the people of the present with those in the past and/or future.
The Storytelling strategy should be coded when the leader discusses the sacrifices of the past, which may include mentioning martyrs/those who have died for the cause (these examples would also be the Justice/negative strategy). Code for this strategy if the leader discusses martyrs to show the tragedy of the situation; also if he uses martyrs to say they have pushed the nation’s cause forward. Code for this if the leader characterizes how bad the situation in the past was, how the roots of the current problems are in the past, or that the past must be avoided.
The Storytelling strategy should also be coded when the leader focuses on gaining strength fi'om the past, when the emphasis is on redeeming the past, when he makes calls to realize the ideal nation/the right to self-determination, or when the focus is on future gains/dreams/aspirations. Also code for the positive variant when the leader mentions heroes of the past, not as martyrs but as people who have had a positive force in moving the nation forward.
373 VL Data Analysis
Chapter 2 explains how the coded information will be analyzed. Important information to be stated here is that the data is entered into SPSS using each group as a case; for each group the columns of data (variables) are the time period, speaker, and other identifying codes (is the group an ingroup or outgroup, if ingroup is it the leaders own party, etc.); group reference, group delineation. Injustice, Justice positive. Enemy Image, Ally Image, Governance Negative, Governance Positive, and Storytelling. In transferring the codes from the coding sheets into SPSS, the following codes are used (the Governance Positive and Negative codes stated above are entered):
Justice strategy (negative/injustice)
99 absent 1 present/low (1 use) 2 present/moderate (2-3 uses) 3 present/high (4-7 uses)
Justice strategy (positive)
99 absent 1 present/low (1 use) 2 present/mo derate (2-3 uses) 3 present/high (4-7 uses)
Outgroup negative strategy/type ambiguous
99 absent 1 ambiguous type/low (1) 2 ambiguous type/moderate (2-3) 3 ambiguous type/high (4-7)
Outgroup negative strategy/Threat
99 absent 1 threat/low (1) 2 threat/moderate (2-3) 3 threat/high (4-7)
374 Outgroup negative strategy/Blame
99 absent 1 blame/low (1) 2 blame/moderate (2-3) 3 blame/high (4-7)
Outgroup negative Strategy/Betrayal
99 absent 1 betrayal/low (1) 2 betrayal/moderate (2-3) 3 betrayal/high (4-7)
Outgroup positive strategy/ambiguous type
99 absent 1 ambiguous type/low (1) 2 ambiguous type/moderate (2-3) 3 ambiguous type/high (4-7)
Outgroup positive strategy/Goals or values similar
99 absent 1 goals or values similar/low (1) 2 goals or values similar/moderate (2-3) 3 goals or values similar/high (4-7)
Outgroup positive Strategy/Situation similar
99 absent 1 situation similar/low ( 1 ) 2 situation similar/moderate (2-3) 3 situation similar/high (4-7)
Outgroup positive strategy/Culture similar
99 absent 1 culture similar/low ( 1 ) 2 culture similar/moderate (2-3) 3 culture similar/high (4-7)
375 APPENDIX B
LIST OF CODED DOCUMENTS
Northern Ireland^
Adams, Gerry. 1981. ‘The Way Forward,” Address at the National H Block/Armagh C ommittee
Adams, Gerry. 1984. Address to the 1984 Ard Fheis.
Adams, Gerry. 1985 (November). Address to the 1985 Ard Fheis.
Adams, Gerry. 1986 (October/November). Address to the 1986 Ard Fheis.
Adams, Gerry. 1987 (October). Address to the 1987 Ard Fheis.
Adams, Gerry. 1996 (March). Address to the 1996 Ard Fheis.
Adams, Gerry. 1996 (May). “Making Real the Vision of 1916,” Address to March and Rally to Commemorate the Easter Rising.”
Adams, Gerry. 1996 (June). “Transforming Hope into Reality: Negotiating a New Beginning,” Speech written for All-Party Talks.
Adams, Gerry. 1996 (October). Speech to party activists in Belfast about peace talks.
Adams, Gerry. 1997 (April). Address to the 1997 Axd Fheis.
Adams, Gerry and Sinn Fein. 1982. Election Manifesto. An Phoblacht/Republiccm News, 30 September 1982.
Hume, John. 1981 (November). Address to the 1981 Annual SDLP Convention.
' All of the Northem Ireland documents listed here, except for the 1997 documents, are courtesy of the Linen Hall Library in Belfast.
376 Hume, John. 1982 (March). “The Way Forward As I See It,” Address to St. Anne’s Cathedral, Belfast and Servite Priory, Benburb.”
Hume, John. 1983 (January). Address to the 1983 Annual SDLP Convention.
Hume, John. 1985 (January). Address to the 1985 (14*^) Annual SDLP Convention.
Hume, John. 1985 (November). Address to the 1985 (15'*') Annual SDLP Convention.
Hume, John. 1986 (November). Address to the 1986 Annual SDLP Convention.
Hume, John. 1987 (November). Address to the 1987 Annual SDLP Convention
Hume, John. 1995 (November). Address to the 1995 Annual SDLP Convention.
Hume, John. 1996 (November). Address to the 1996 Annual SDLP Convention.
Hume, John. 1997 (May). “Representation is the Issue,” Address before May elections.
Hume, John. 1997 (May). “Critical Election,” Address before May elections.
Rhodesia/Zimbabwe^
Mugabe, Robert. 1977 (circa end of year). Radio Address: “The Meaning of Our Revolutionary Struggle.”
Mugabe, Robert. 1978 (January). New Year Message.
Mugabe, Robert. 1978 (February). Radio Address Concerning the Malta Conference.
Mugabe, Robert. 1979 (January). New Year Message to the People of Zimbabwe.
Mugabe, Robert. 1979 (April). “Ignore the April Bogus Elections,” Radio Address from Maputo (Mozambique) on the Eve of the April Bogus Elections.
Mugabe, Robert. 1980 (January 1). New Year Message.
Mugabe, Robert. 1980 (January 26). Radio Broadcast.
* These documents come from a range of sources: the Hoover Archives, Hoover Library, theH erald (Salisbury/Harare), the Chronicle (Bulawayo), the Foreign Broadcast Information Service, and the Baumhogger seven-volume documents collection (see bibiliography). 377 Mugabe, Robert. 1989 (April). Ninth Anniversary of Independence Speech.
Mugabe, Robert. 1989 (December). State of the Nation Address.
Mugabe, Robert and ZANU (PF). (1990). Election Manifesto.
Muzorewa, Abel. 1977. (March). Presidential Address to the National Consultative Assembly of the UANC in Salisbury.
Muzorewa, Abel. 1978 (July). Broadcast to the Nation.
Muzorewa, Abel. 1979 (December). Address to the Nation.
Muzorewa, Abel. 1980 (January). Speech Launching the Election Campaign of the UANC.
Muzorewa, Abel. 1980 (February 1). Radio Broadcast.
Muzorewa, Abel. 1980 (February 22). Radio Broadcast.
Radio Truth. 1989. Position Statement.
Tekere, Edgar. 1989. “Revolution Hijacked,” Mo/o.
Tekere, Edgar and Zimbabwe Unity Movement. 1989 (November). Election Manifesto
378 APPENDIX C
CONff ARISON OF STRATEGY FREQUENCIES,
WITHOUT STORYTELLING
I. Explanation
As noted in Chapter 6, the initial analysis did not treat the Storytelling strategy different from the other strategies. Because the former is a mode of using the others, passages coded with the Storytelling strategy are also coded with another. This may inflate the value of the Storytelling appeal in the end. This appendix presents the strategy frequencies for each case without Storytelling. First, tables similar to those used in the chapter present the data, then the section compares the strategies predicted to be most successful, the results as shown in the case chapters, and the results without the
Storytelling strategy.
Fortunately, in every case the new results strengthen the original findings (with the exception of those cases where one of the predicted winning strategies was the
Storytelling strategy). Those results were the following. In the first, second, and fourth
Northem Ireland cases, the predicted strategy profile was used by the winning leader. In the third case, there is no pattern that differentiates between the leaders (both Hume’s and
379 Adams’s strategy profiles are close to the expectations). In the first and third
Rhodesia/Zimbabwe cases, the strategy profile expected to be used by the more successful
leader was indeed used by that leader and was unique to him. In the second case, the two
leaders had similar strategy profiles which both fit with the expectations.
n. Northern Ireland
Hume Adams (n=520) (n=530) Identity (Northem Ireland 7.9% .4% references)’’ Outgroup Negative (Enemy Image) 14.6% 27.9%
Justice Negative 28.3% 25.8%
Govemance Negative 9.8% 16.6%
Govemance Positive 27.7% 13%
Outgroup Positive (Ally Image) 7.1% 9.1%
Justice Positive 4.6% 7.2%
Total %' 100% 100%
The "n" here is the total number of strategies and the number given in each cell is the percentage of this total. The strategies in bold are tlie strategies hypothesized to be used most frequently by the winning leader, in this case that is Gerry Adams.
Because Northem Ireland is the focus of the research and considered to be the society under exam ination, tliese references are the focus (Adams tends to focus much more on Ireland as the society). Tliis figure excludes references to the leader’s party because this is misleading about the groups in the society the leader is defining.
' Due to rounding off of the numbers, these totals do not equal exactly 100%.
Table 8.1: Comparison of Strategies without Storytelling, Northem Ireland Shift 1
380 Expectation; Justice Negative Enemy Image Storytelling
Results with Storytelling:
Hume Adams TVinner”') Justice Negative 22.5% Enemy Image 23.9% Govemance Positive 22.1% Justice Negative 22.2% Storytelling 20.2% Storytelling 14.7%
Results without Storytelling:
Hume Adams C'winnefl Justice Negative 28.3% Enemy Image 27.9% Govemance Positive 27.7% Justice Negative 25.8% Enemy Image 14.6% Govemance N egative 16.6%
Hume Adams (n=858) (n=712) Identity (Northem Ireland 5.1% 2% references) Outgroup Negative (Enemy Image) 13.8% 30%
Justice Negative 21.4% 25.7%
Govemance Negative 8.5% 17%
Govemance Positive 29.4% 9.8%
Outgroup Positive (Ally Image) 12.7% 9%
Justice Positive 9.1% 6 .6%
Total % 100.1% 100%
Table 8.2: Comparison of Strategies without Storytelling, Northern Ireland Shift 2
381 Expectation: Governance Positive Justice Negative Storytelling
Results with Storytelling:
Hume (“winner^’') Adams Governance Positive 24.8% Enemy Image 24.5% Justice Negative 18.1% Justice Negative 21% Storytelling 15.7% Storytelling 18.2%
Results without Storytelling:
Hume C'winner'l Adams Governance Positive 29.4% Enemy Image 30% Justice Negative 21.4% Justice Negative 25.7% Enemy Image 13.8% Governance Negative 17%
Hume Adams (n=226) (n=523) Identity' (Northern Ireland 4.9% .6% references ) Outgroup Negative (Enemy Image) 2.2% 12.6%
Justice Negative 19% 18%
Governance Negative 3.1% 13.2%
Governance Positive 51.3% 42.6%
Outgroup Positive (Ally Image) 8.4% 8.8%
Justice Positive 11.1% 4.2%
Total % 100% 100.1%
Table 8.3: Comparison of Strategies without Storytelling, Northern Ireland Shift 3
382 Expectation: Governance Positive Ally Image Justice Negative Storytelling
Results with Storytelling:
Hume Adams ('“winner’’') Governance Positive 43.1% Governance Positive 32.9% Justice Negative 16% Storytelling 22.7% Storytelling 16% Justice Negative 13.9% Justice Positive 9.3% Governance Negative 10.2%
Results without Storytelling:
Hume Adams ('“winner”^ Governance Positive 51.3% Governance Positive 42.6% Justice Negative 19% Justice Negative 18% Justice Positive 11.1% Governance Negative 13.2% Ally Image 8.4% Enemy Image 12.6%
Hume Adams (n=469) (n=400) Identity (Northern Ireland 5.1% 2% references) Outgroup Negative (Enemy Image) 12.4% 14%
Justice Negative 15.1% 18%
Governance Negative 13.6% 12%
Governance Positive 36.2% 38.3%
Outgroup Positive (Ally Image) 7% 6.5%
Justice Positive 10.4% 9.3%
Total % 99.8% 100.1%
Table 8.4: Comparison of Strategies without Storytelling, Northern Ireland Shift 4
383 Expectation; Enemy Image Governance Positive Justice Negative Storytelling
Results with Storytelling:
Hume Adams rSvinner'l Governance Positive 31.4% Governance Positive 30.4% Storytelling 13.5% Storytelling 20.6% Justice Negative 13.1% Justice Negative 14.3% Governance Negative 11.8% Enemy Image 11.1%
Results without Storytelling:
Hume Adams r'winner'l Governance Positive 36.2% Governance Positive 38.3% Justice Negative 15.1% Justice Negative 18% Governance Negative 13.6% Enemy Image 14% Enemy Image 12.4% Governance Negative 12%
384 m . Rhodesia/Zimbabwe
Mugabe Muzorewa (n=689)“ (n=313) Identity (Zimbabwe)*’ 8% 12.1%
Outgroup Negative (Enemy Image) 21.3% 14.4%
Justice Negative 18.9% 19.5%
Governance/negative 15.8% 11.8%
Governance/positive 16.3% 32.3%
Outgroup Positive (Ally Image) 9.4% 7%
Justice Positive 10.3% 2.9%
Total % ' 100 100
The "n” here is the total number of strategies and the number given in each cell is the percentage of this total. The strategies in bold are the strategies hypothesized to be used most Grequently by the winning leader, in this case that is Mugabe.
^ Because Zimbabwe is the focus of the research and considered to be the society imder exam ination, this reference is the focus. This figure excludes references to the leader’s party because this is misleading about the groups in the society the leader is defining.
" Due to rounding off of the numbers, these totals do not equal exactly 100%.
Table 8.5: Comparison of Strategies without Storytelling, Rhodesia Shift 1
385 Expectation; Justice Negative Enemy Image Governance Negative Storytelling
Results with Storytelling:
Mugabe (‘Viimer”’) Muzorewa Storytelling 19% Governance Positive 26.5% Enemy Image 17.3% Storytelling 17.8% Justice Negative 15.3% Justice Negative 16% Governance Positive 13.2% Enemy Image 11.8%
Results without Storytelling:
Mugabe TVinner”') Muzorewa Enemy Image 21.3% Governance Positive 32.3% Justice Negative 18.9% Justice Negative 19.5% Governance Positive 16.3% Enemy Image 14.4% Governance Negative 15.8% Identity (Zimbabwe) 12.1%
Mugabe Muzorewa (n=287) (n=939) Identity (Zimbabwe) 8% 9.9%
Outgroup Negative (Enemy Image) 18.5% 19.1%
Justice Negative 17.4% 16.7%
Governance/negative 11.8% 11%
Governance/positive 19.9% 27.1%
Outgroup Positive (Ally Image) 11.1% 6.5%
Justice Positive 13.2% 9.8%
Total % 99.9 100.1%
Table 8.6: Comparison of Strategies without Storytelling, Zimbabwe Shift 2
386 Expectation; Governance Positive Storytelling Enemy Image Justice Negative
Results with Storytelling:
Mugabe ('‘^winner”') Muzorewa Storytelling 25.8% Governance Positive 23.4% Governance Positive 14.7% Enemy Image 16.5% Enemy Image 13.7% Justice Negative 14.4% Justice Negative 12.9% Storytelling 13.6%
Results without Storytelling:
Mugabe ('“winner”! Muzorewa Governance Positive 19.9% Governance Positive 27.1% Enemy Image 18.5% Enemy Image 19.1% Justice Negative 17.4% Justice Negative 16.7% Justice Positive 13.2% Governance Negative 11%
Mugabe Tekere (n=978) (n=619) Identity (Zimbabwe) 10.7% 6 .8%
Outgroup Negative (Enemy Image) 5.3% 14.7%
Justice Negative 13.1% 13.4%
Governance/negative 3.6% 19.1%
Governance/positive 46.5% 35.7%
Outgroup Positive (Ally Image) 9.6% 7.1%
Justice Positive 11.1% 3.2%
Total % 99.9 100%
Table 8.7: Comparison of Strategies without Storytelling, Zimbabwe Shift 3
387 Expectation; Governance Positive and Negative Justice Negative Enemy Image
Results with Storytelling:
Mugabe Tekere Chvinner'n Governance Positive 41.3% Governance Positive 32.4% Justice Negative 11.6% Governance Negative 17.3% Storytelling 11.2% Enemy Image 13.3% Justice Positive 9.9% Injustice 12.2%
Results without Storytelling:
Mugabe Tekere r“winner”'> Governance Positive 46.5% Governance Positive 35.7% Justice Negative 13.1% Governance Negative 19.1% Justice Positive 11.1% Enemy Image 14.7% Identity (Zimbabwe) 10.7% Injustice 13.4%
388 BIBLIOGRAPHY
“Hume Gets Sinn Fein Deal Challenge.” Irish News, November 12, 1986.
“Criticism by Africa Watch.” 4/^^ca Tv'ew.y, December 1989: 5.
“Zimbabwe: Unpopular.” Africa Confidential 30, no. 21 (1989).
“Zimbabwe: Tekere's Challenge.” Africa Confidential 30, no. 23 (1989): 5-6.
“Socialist Thrust ReafiBrmed.” The Chronicle, December 20, 1989: 1, 11.
“President Optimistic Over Economy and Political Climate.” H erald, December 8, 1989: 1.
Editorial: “We are Socialist Oriented by Nature.” H erald, December 9, 1989: 4.
“Zimbabwe: AED Special Report on Politics—Mugabe, Secure in Position, Drops One- Party Dream.” African Economic Digest, November 5, 1990.
“Mugabe Says Multiparty System Leading Zimbabwe to Disaster.” The Reuter Library Report, April 6, 1990.
“President Pledges Free, Fair Poll.” The Chronicle, March 9, 1990: 1.
“Various Associations Hold Discussions with President.” H erald, March 15, 1990: 1.
“ZCTU Backs President and Ruling Party.” H erald, March 15, 1990: 1, 7.
“4 ZUM Candidates Resign to Rejoin Ruling ZANU (PF).” H erald, March 27, 1990: 1,5.
“ZANU (PF) Election Manifesto.” H erald, March 5, 1990: 2-3.
389 Anderson, Benedict. Imagined Communities: Reflection on the Origin and Spread o f Nationalism. Second revised edition. New York: Verso, 1991.
Bal, Mieke. On Story-Telling: Essays in Narratology. Sonoma, CA: Polebridge Press, 1991.
Barber, Benjamin. J ih a d vj. M cW orld. New York: Random House, 1995.
Baumhogger, Gosuan, Telse Diedrichsen, and Ulf Engel, eds. The Struggle for Independence: Documents on the Recent Development o f Zimbabwe, Volumes 1- 7. Hamburg, 1984.
Baynham, Simon, ed. Zimbabwe in Transition. Stockholm, Sweden: Almqvist & Wiksell International, 1992.
BBC/Lexis-Nexis. “Advice to the Voters in Rhodesia.” Revolutionary Voice o f Zimbabwe, 19 January 1980.
Bean, Kevin. “The New Departure: Recent Developments in Irish Republican Ideology and Strategy.” MA, University of Liverpool, 1996.
Bennett, Vanora. “Zimbabwe Starts Second Decade With Promises of Land and Peace.” The Reuter Library Report, May 2, 1990.
Bew, Paul, and Gordon Gillespie. Northern Ireland: A Chronology o f the Troubles, 1968-1993. Dubhn: GiU and MacMillan, 1993.
Bew, Paul, Peter Gibbon, and Henry Patterson. Northern Ireland 1921-1994: Political Forces and Social Classes. London: Serif, 1995.
Bourdieu, Pierre. Language and Symbolic Power. Cambridge, MA: Harvard University Press, 1991.
Boutros-Ghah, Boutros. An Agenda fo r Peace, Second Edition. New York: United Nations, 1995.
Bowen, John R. “The Myth of Ethnic Conflict.” Journal o fDemocracy 7, no. 4 (1996): 3-14.
Brass, Paul R. Ethnicity and Nationalism: Theory and Comparison. Newbury Park, CA: Sage Publications, 1991.
Brecher, Michael. The Foreign Policy System o f Israel: Setting, Images, Process. New Haven: Yale University Press, 1972.
390 Breen, Suzanne. “Fun Around the Comer.” Fortnight, April 1997; 7.
Brennock, Mark. “Soul-searching Sinn Fein Accepts Number of. Hitherto Ignored, Realities.” Irish Times, June 23, 1992.
Breuilly, John. Nationalism and the State. Chicago: University of Chicago Press, 1994.
Brewer, Marilynn, and Sherry K. Schneider. “Social Identity and Social Dilemmas: A Double-Edged Sword.” In Social Identity Theory, edited by Dominic Abrams and Michael A. Hogg. New York: Harvester/Wheatsheaf 1990: 169-184.
Brewer, Marilynn. “The Social Self: On Being the Same and Different at the Same Time.” Personality and Social Psychology Bulletin 17 (1991): 475-482.
Brewer, Marilynn. “When Contact is Not Enough: Social Identity and Intergroup Cooperation.” International Journal o f Intercultural Relations 20, no. 3/4 (1996): 291-303.
Brown, Michael E., ed. Ethnic Conflict and International Security. Princeton, NJ: Princeton University Press, 1993.
Brown, Michael E., ed. The International Dimensions o f Internal Conflict. Cambridge, MA: The MIT Press, 1996.
Brubaker, Rogers. Nationalism Reframed. Cambridge: Cambridge University Press, 1996.
Bums, James McGregor. Leadership. New York: Harper & Row, PubUshers, 1978.
Bush, K.D. and E.F. Keyman. “Identity-Based Conflict: Rethinking Security in a Post- Cold War World.” Global Governance 3 (1997): 311-328.
Buzan, Barry. People, States, and Fear. New York: Harvester/Wheatsheaf, 1991.
Callaghy, Thomas M. “Africa and the World Economy: Caught Between a Rock and a Hard Place.” In. A frica in W orld Politics, edited by John W. Harbeson and Donald Rothchild, 39-68. Boulder, CO: Westview Press, 1991.
Campbell. “Interim Report by Sir John Boynton, MC, British Election Commission.” Oxford University: Rhodes House Library, 1980.
Campbell, Kathleen Kohrs, and Kathleen Hall Jamieson. D eeds Done in Words: Presidential Rhetoric and the Genres o f Governance. Chicago: University of Chicago Press, 1992.
391 Carter, Neal, Andrea Grove, and Kent Kilie. “Nationalism and the Importance of Identity: A Systematic Assessment.” Paper presented at the International Studies Association-Midwest, Cleveland, OH, October 3-5, 1997.
Carter, Neal A. “Ethnoterritorial Justice: Arguments Linking Social Identities to Constitutional Reform in Belgium and Canada.” Dissertation, Syracuse University, forthcoming.
Centre for African Studies. “South Africa: Perspectives After F.W. De Klerk's February 2 Speech.” Southern Africa Political and Economic Monthly 3, no. 6 (1990): 3-6.
Connor, Walker. Ethnonationalism: The Quest for Understanding. Princeton: Princeton University Press, 1994.
Coogan, Tim Pat. On the Blanket. Dublin: Ward River, 1980.
Coogan, Tim Pat. The Troubles: Ireland's Ordeal 1966-1996 and the Search for Peace. Boulder, CO.: Roberts Rinehart Publishers, 1996.
Coser, Lewis. The Function o f Social Conflict. New York: The Free Press, 1956.
Cottam, Richard W. Foreign Policy Motivation: A General Theory and a Case Study. Pittsburgh: University of Pittsburgh Press, 1977.
Cottam, Martha. Foreign Policy Decision Making: The Influence o f Cognition. Boulder, Co.: Westview Press, 1986.
Dashwood, Hevina. “The Peacebuilding Process in Zimbabwe: External Threats and the Enemy Within.” Paper presented at the International Studies Association, Toronto, Canada, 1997.
Davies, James C. Human Nature in Politics: The Dynamics o f Political Behavior. New York: John Wiley, 1963.
Davis, Richard. Mirror Hate: The Convergent Ideology o f Northern Ireland Paramilitaries, 1966-1992. Brookfield, VT: Dartmouth Publishing Company, 1994.
De Sola Pool, IthieL, ed. Trends in Content Analysis. Urbana, EL.: University of Illinois Press, 1959.
DeVries, Robert L. “Protecting National Minorities Through Building Civil Societies in Eastern Europe: The Role of the OSCE and NGOs.” Paper presented at the International Studies Association, Washington, D C., 1999.
392 DeWaaL, Victor. The Politics o f Reconciliation: Zimbabwe's First Decade. London: Hurst, 1990.
Deutsch, Karl W. Nationalism and Social Communication. Cambridge, MA: MIT Press, 1953.
DeYoung, Karen. “Key Accord Signed on Northern Ireland: Pact with London Gives Dublin a Role in Troubled Ulster.” Washington Post, November 16, 1985.
Doty, Roxanne Lynn. “Sovereignty and the Nation: Constructing the Boundaries of National Identity.” In State Sovereignty as a Social Construct, edited by Thomas J. Biersteker and Cynthia Weber. Cambridge, UK: Cambridge University Press, 1996: 121-147.
EUemers, Naomi. “The Influence of Socio-structural Variables on Identity Management Strategies.”European Review o f Social Psychology 4 (1993): 27-57.
Elliott, Sydney, and Richard A. Wilford. The 1982 Northern Ireland Assemby Election, Studies in Public Policy. Glasgow: University o f Strathclyde (Center for the Study of Public Policy), 1983.
Elliott, Sidney. “A Post Mortem on an Unprecedented Poll.” Fortnight, 10 February 1986; 6-7.
Elliott, Sidney. “And Then There Were Thirteen.” Fortnight, luly/August 1987; 11-13.
Elliott, Sidney. “Changes.” Fortnight, June 1997: 9.
Esman, Milton, I. “Political and Psychological Factors in Conflict.” In Conflict and Peacemaking in Multiethnic Societies, edited by Joseph Montville. Toronto: Lexington Books, 1990: 53-64.
Esman, Milton J., and Shibley Telhami, eds. International Organizations and Ethnic Conflict. Ithaca, NY: Cornell University Press, 1995.
Evans, Peter B., Harold K. Jacobson, and Robert D. Putnam, eds. Double-Eldged Diplomacy: International Bargaining and Domestic Politics. Berkeley, CA: University of Cahfomia Press, 1993.
Farrel, Jack. “Adams' Unsmiling Irish Eyes.” Fortnight, April 1996: 16.
Fay, Marie-Therese. “The Battle for Control of the Nationalist Agenda: SDLP and SF Competition, 1982-1996.” Dissertation, Queen's University, 1996.
393 Feeney, Brian. “Stoops Down.” Fortnight, July/August 1997; 22.
Finlay, David J., Ole R. Holsti, and R. Fagan. Enemies in Politics. Chicago: Rand McNally, 1967.
Fisk, Robert. The Point of No Return: The Strike Which Broke the British in Ulster. London: Andre Deutsch, 1975.
Frederikse, Julie. None But Ourselves: Masses V5. Media in the Making o f Zimbabwe: 1964-1981. Harare: Zimbabwe Publishing House, 1982.
Freedom House Observation Group. “The 1980 Election in Zimbabwe (Rhodesia).” New York: Freedom House, 1980.
Fuller, Graham. “Redrawing the World's Borders.” World Policy Journal 14, no. 1 (1997): 11-21.
Gaertner, Samuel L., John F Dovidio, Phyllis A Anastasio, Betty A. Bachman, and Mary C. Rust. “The Common Ingroup Identity Model: Recategorization and the Reduction of Intergroup Bias.” European Review o f Social Psychology 4 (1993): 1-24.
Gagnon, V.P. Jr. “Ethnic Nationalism and International Conflict: The Case of Serbia.” International Security 19, no. 3 (1994/95): 130-166.
Gamson, William. Talking Politics. New York: Cambridge University Press, 1992.
Ganguly, Rajat, and Ray Taras. Understanding Ethnic Conflict: The International D im ension. New York: Longman, 1998.
Gellner, Ernest. Nations and Nationalism. Ithaca, NY: Cornell University Press, 1983.
George, Alexander. “Quantitative and Qualitative Approaches to Content Analysis.” In Trends in Content Analysis, edited by Ithiel De Sola Pool. Urbana, II.: University of Illinois Press, 1959: 7-32.
George, Alexander. “Case Studies and Theory Development: The Method of Structured, Focused Comparison.” In Z)/p/o/nac>’.- New Approaches in History, Theoiy, and Policy, edited by P.G. Lauren. New York: Free Press, 1979: 43-68.
Gofifinan, Erving. Frame Analysis: An Essay on the Organization o f Experience. New York: Harper Colophon, 1974.
394 Goodman, James. Nationalism and Transnationalism: The National Conflict in Ireland and European Union Integration. Brookfield, VT: Avebury, 1996.
Gordon, Michael R. “Post-Communist Russia Plumbs Its Soul, in Vain, for New Vision.” New York Times, March 31 1998: 1, 6.
Gourevitch, Peter. “The Second Image Reversed: The International Sources of Domestic Politics.” International Organization 32 (1978): 881-911.
Graham, William. “Accord Advances Real Says Hume.” Irish News, November 11, 1986.
Graham, William. “Hume Secures New US Jobs Initiative.” Irish News, March 27, 1987.
Greenstein, Fred. “Chapter 2.” In Personality and Politics. Princeton: Princeton University Press, 1987.
Grundy, Kenneth W. “Southern Africa: The Revolution Prolonged.” la. A frica in World Politics, edited by John W. Harbeson and Donald Rothchild. Boulder, CO: Westview Press, 1991: 95-118.
Guelke, Adrian. Northern Ireland: The International Perspective. New York: St. Martin's Press, 1988.
Guetzkow, Harold. Multiple Loyalties: Theoretical Approach to a Problem in International Organization. Princeton: Center for Research in World Political Institutions, 1955.
Gunther, Richard, and Anthony Mughan. “Political Institutions and Cleavage Management.” In Do Institutions Matter? Government Capabilities in the United States and Abroad, edited by R. Kent Weaver and Bert A. Rockman. Washington, D.C.: Brookings Institution, 1993: 272-301.
Gurr, Ted Robert. Why Men Rebel. Princeton: Princeton University Press, 1970.
Gurr, Ted Robert. Minorities at Risk: A Global View o fEthnopolitical Conflict. Washington, D C.: United States Institute of Peace Press, 1993.
Gurr, Ted Robert. “People Against States: Ethnopolitical Conflict and the Changing World System.” International Studies Quarterly 38, no. 3 (1994): 347-377.
Haas, Ernst B. The Uniting o f Europe: Political, Social, and Economic Forces, 1950- 1957. Stanford, CA: Stanford University Press, 1958.
395 Harbeson, John W., and Donald Rothchild, eds. Africa in World Politics. Boulder, CO; Westview Press, 1991.
Harbeson, John W., and Donald Rothchild. “Africa in Post-Cold War International Politics: Changing Agendas.” Africa in World Politics, edited by John W. Harbeson and Donald Rothchild. Boulder, CO: Westview Press, 1991: 1-15.
Hargrove, Erwin. “Two Conceptions of Institutional Leadership.” In Leadership and Politics, edited by Bryan D. Jones. Lawrence: University Press of Kansas, 1989: 57-83.
Herbst, Jeffrey. “Racial Reconciliation in Southern Africa.” International Affairs (1989): 43-54.
Herbst, Jeffrey. State Politics in Zimbabwe. Berkeley, CA: University of California Press, 1990.
Herbst, Jeffrey. “The United States and Africa: Issues for the Future.” \n. A frica in World Politics, edited by John W. Harbeson and Donald Rothchild. Boulder, CO: Westview Press, 1991: 161-178.
Hermann, Margaret G. “When Leader Personality Will Affect Foreign Policy: Some Propositions.” In In Search o f Global Patterns, edited by James Rosenau. New York: Free Press, 1976: 326-333.
Hermann, Margaret G., ed. A Psychological Examination o f Political Leaders. New York: Free Press, 1977.
Hermann, Margaret G. “Handbook for Assessing Personal Characteristics and Foreign Policy Orientations of Political Leaders.” Columbus, OH: Mershon Center, 1983, 1987.
Hermann, Margaret G. “The Ingredients of Leadership.” In. P olitical Psychology: Contemporary Problems and Issues, edited by Margaret G. Hermann. San Francisco: Jossey-Bass, 1986.
Hermann, Margaret G., and Joe D. Hagan. “International Decision Making: Leadership Matters.” Foreign Policy 110 (1998): 124-137.
Herrmann, Richard K. Perceptions and Behavior in Soviet Foreign Policy. Pittsburgh: University of Pittsburgh Press, 1985.
Herrmann, Richard K. “The Empirical Challenge of the Cognitive Revolution.” International Studies Quarterly 32, no. 2 (1988): 175-204.
396 Herrmann, Richard K ., and Michael P. Fischerkelier. “Beyond the Enemy Image and Spiral Model; Cognitive-Strategic Research After the Cold War.” International Organization 49, no. 3 (1995): 415-450.
Hobsbawm, E.J. Nations and Nationalism Since 1780: Programme, Myth, Reality. Second ed. Cambridge: Cambridge University Press, 1990.
Holsti, Ole. “Foreign Policy Formation Viewed Cognitively.” In Structure o f Decision: The Cognitive Maps o f Political Elites, edited by Robert Axelrod. Princeton: Princeton University Press, 1976: 18-54.
Hoover, Dean, and David Kowalewski. “Dynamic Models of Dissent and Repression.” Journal o f Conflict Resoluton 58 (1992): 637-658.
Horowitz, Donald. Ethnic Groups in Conflict. Berkeley: University of California Press, 1985.
Huntington, Samuel. The Third Wave: Democratization in the Late Twentieth Century. Norman, OK: University of Oklahoma Press, 1991.
Huntington, Samuel. “The Clash of Civilizations.” Foreign Affairs 72, no. 3 (1993): 22-49.
Ickes, William, and Richard Gonzalez. “"Social" Cognition and Social Cognition: From the Subjective to the Intersubjective.” Small Group Research 25, no. 2 (1994): 294-315.
Irvin, Cynthia, and Eddie Moxon-Browne. “Not Many Floating Voters Here.” Fortnight, May 1989: 7-9.
Irvin, Cynthia. “Terrorists' Perspectives: Interviews.” In Terrorism and the Media, edited by David L. Paletz and Alex P. Schmid. Newbury Park, CA: Sage Publications, 1992: 62-85.
Irvin, Cynthia. “Paramilitary Politics in Parliamentary Democracies: Militant Nationalism in Ireland and Spain.” Dissertation, Duke University, 1993.
Ishiyama, John T., and Marijke Breuning. Ethnopolitics in the New Europe. Boulder, CO: Lynn Rienner, 1998.
Iyengar, Shanto, and Donald Kinder. New That Matters: Television and American Opinion. Chicago: University of Chicago Press, 1987.
397 Iyengar, Shanto, and Donald Kinder. Is Anyone Responsible? How Television Frames Political Issues. Chicago; University of Chicago Press, 1991.
Jackson, Robert. Quasi-States: Sovereignty, International Relations, and the Third fPbr/ûf. Cambridge: Cambridge University Press, 1990.
Jenkins, J. Craig. “Resource Mobilization Theory and the Study of Social Movements.” Anmial Review o f Sociology 9 (1983): 527-553.
Kaplan, Robert D. “The Coming Anarchy.” Atlantic Monthly (1994): 44-74.
Kennedy-Pipe, Caroline. The Origins o f the Present Troubles in Northern Ireland. New York: Longman, 1997.
Kertzer, David 1. Politics and Symbols: The Italian Communist Party and the Fall o f Communism. New Haven: Yale University Press, 1996.
Kohn, Hans. The Idea o f Nationalism: A Study in its Origins and Background. New York: Collier Books, 1969.
Kowalski, Robin M., and Randall Wolfe. “Collective Identity Orientation, Patriotism, and Reactions to National Outcomes.” Personality and Social Psychology Bulletin 20, no. 5 (1994): 533-540.
Krause, Keith, and Michael C. Williams. “Broadening the Agenda o f Security Studies: Politics and Methods.” Mershon International Studies Review 40, no. Supplement 2 (1996): 229-254.
Kriger, Norma J. Zimbabwe's Guerrilla War: Peasant Voices. New York: Cambridge University Press, 1992.
Lancaster, Carol. “The Lagos Three: Economic Regionalism in Sub-Saharan Africa.” In Africa in World Politics, edited by John W. Harbeson and Donald Rothchild. Boulder, CO: Westview Press, 1991: 249-267.
Lederer, Gerda. “Protest Movements as a Form o f PoHtical Action.” la P olitical Psychology, edited by Margaret G. Hermann. San Francisco: Jossey- Bass, 1986: 355-378.
Lijphart, Arend. Democracy in Plural Societies: A Comparative Exploration. New Haven: Yale University Press, 1977.
Lijphart, Arend, Ronald Rogowski, and R. Kent Weaver. “Separation of Powers and Cleavage Management.” In Do Institutions Matter? Government Capabilities in
398 the United States and Abroad, edited by R. Kent Weaver and Bert A Rockman. Washington, D.C.: The Brookings Institution, 1993: 302-344.
Lijphart, Arend. “The Puzzle of Indian Democracy: A Consociational Interpretation." Americofi Political Science Review 90, no. 2 (1996): 258-268.
Lyons, Gene M., and Michael Mastanduno, eds. Beyond Westphalia? State Sovereignty and International Intervention. Baltimore and London: Johns Hopkins University Press, 1995.
MacGinty, Roger. “American Influences on the Northern Ireland Peace Process.” The Jotimal o f Conflict Studies (1997): 31-50.
Makumbe, John. “The 1990 Zimbabwe Elections: Implications for Democracy.” In The One Party State and Democracy, edited by Ibbo Mandaza and Lloyd M. Sachikonye, 179-188. Harare: SAPES, 1991.
Mandaza, Ibbo, ed. Zimbabwe: The Political Economy o f Transition 1980-1986. Dakar, Senegal: CODESRIA, 1986.
Mandaza, Ibbo. “The Post-White Settler Colonial Situation.” In Zimbabwe: The Political Economy o f Transition 1980-1986, edited by Ibbo Mandaza. Dakar, Senegal: CODESRIA, 1986: 21-74.
Mandaza, Ibbo. “Perspectives on Regional Economic Cooperation in Southern Africa.” Southern Africa Political and Economic Monthly 2, no. 1 (1988): 11-13.
Mathema, Cain. “Marxism-Leninism and the State: Only Socialism Can Meet the Wishes of the People.” Herald, January 12, 1990.
Mayall, James. Nationalism atid International Society. New York: Cambridge University Press, 1990.
McAdam, Doug, John D. McCarthy, and Mayer N. Zald. “Social Movements.” In Handbook o f Sociology, edited by Neil J. Smelser. Newbury Park, CA.: Sage Publications, 1988: 695-737.
McAdam, Doug, John D. McCarthy, and Mayer N. Zald, eds. Comparative Perspectives on Social Movements. New York: Cambridge University Press, 1996.
McAdam, Doug. “Conceptual Origins, Current Problems, Future Directions.” In Comparative Perspectives on Social Movements: Political Opportunities,
399 Mobilizing Structures, and Cultural Framings, edited by Doug McAdam, John D. McCarthy and Mayer N. Zald. New York; Cambridge University Press, 1996: 23-40.
McGany, John, and Brendan O'Leary, eds. The Politics o f Ethnic Conflict Regulation: Case Studies o f Protracted Ethnic Conflict. New York: Routledge, 1993.
McGarry, John, and Brendan O'Leary. Explaining Northern Ireland. Cambridge, MA.: BlackweU Publishers, 1995.
McCarty, Craig, S. Alexander Haslam, Karen J. Hutchinson, and John C. Turner. “The EfSects of Salient Group Memberships on Persuasion.” Small Group Research 25, no. 2 (1994): 267-293.
McSweeney, Bill. “Identity and Security: Buzan and the Copenhagen School.” Review o f International Studies 22 (1996): 81-93.
Mercer, Jonathan. “Anarchy and Identity.” International Organization 49, no. 2 (1995): 229-252.
Midlarsky, Manus I., ed. The Internationalization o f Communal Strife. London: Routledge, 1992.
Miller, David. Don't Mention the War: Northern Ireland, Propaganda, and the Media. Boulder, CO: Pluto Press, 1994.
Mitchell, Paul. “Party Competition in an Ethnic Dual Party System.” Ethnic and Racial Studies 18, no. 4 (1995): 773-795.
Montville, Joseph, ed. Conflict and Peacemaking in Midtiethnic Societies. Toronto: Lexington, 1990.
Morrison, John. “Zimbabwe's Tekere Vows to Run for President Against Mugabe.” The Reuter Library Report, February 23, 1990.
Moyo, Jonathan N. “Democracy in Zimbabwe: A Critical Review.” Southern Africa Political and Economic Monthly 2, no. 5 (1989): 8-11.
Moyo, Jonathan N. “Zimbabwe: A Critical Appraisal of the ZUM Challenge.” Southern Africa Political and Economic Monthly 2, no. 11 (1989): 24-28.
Moyo, Jonathan N. Votingfor Democracy: Electoral Politics in Zimbabwe. Harare, Zimbabwe: University of Zimbabwe Publications, 1992.
400 Mugabe, Robert. “Campaign speech.” 1980.
Mummendey, Amelie, and Hans-Joachim Schreiber. “Better or Just Different? Positive Social Identity By Discrimination Against or By Differentiation From Outgroup.” European Journal o f Social Psychology 13 (1983); 389-397.
Muzorewa, Abel. “Campaign speech.” 1980.
Ncube, Welshman. “Constitutionalism, Democracy and Political Practice in Zimbabwe.” In The One Party State and Democracy, edited by Ibbo Mandaza and Sachikonye. Harare: SAPES, 1991; 155-177.
Nelson, Thomas E., Rosalee A. Clawson, and Zoe M. Oxley. “Media Framing of a Civil Liberties Conflict and Its Effect of Tolerance.” American Political Science Review 91, no. 3 (1997): 567-583.
Nordlinger, Eric A. Conflict Regulation in Divu^dSocieties. Cambridge, MA.: Center for International Affairs, Harvard University, 1972.
Nyang'oro, Julius E ., and Timothy M. Shaw. “The African State in the Global Economic Context.” In The African State at a Critical Juncture: Between Disintegration and Reconfiguration, edited by Leonardo A. Villalon and Philip A. Huxtable. Boulder, CO: Lynn Rienner Publishers, 1998: 27-42.
O'Connor, Fionnuala. In Search o f a State: Catholics in Northern Ireland. Belfast: The Blackstaff Press, 1993.
O'Leary, Brendan. “Beyond the Blocs.” Fortnight, March 1990: 16-17.
O'Leary, Brendan, and John McGarry. The Politics o f Antagonism: Understanding Northern Ireland. Atlantic Highlands, N.J.: The Athlone Press, 1993.
Oberschall, Anthony. “Opportunities and Framing in the Transition to Democracy.” In Comparative Perspectives on Social Movements, edited by Doug McAdam, John D. McCarthy and Mayer N. Zald. New York: Cambridge University Press, 1996: 93-121.
Ollapally, Deepa M., and Alexander Cooley. “Identity Politics and the International System.” Nationalism and Ethnic Politics 2, no. 4 (1996): 479-499.
Osgood, Charles E. “The Representational Model and Relevant Research Methods.” In Trends in Content Analysis, edited by Ithiel De Sola Pool. Urbana, II.: University of Illinios Press, 1959: 33-88.
401 Petty, Richard E., and John T. Cacioppo. Attitudes and Persuasion: Classic and Contemporary Approaches. Dubuque, Iowa; W.C. Brown, 1981.
Petty, Richard £., and John T. Cacioppo. Communication and Persuasion: Central and Peripheral Routes to Attitude Change. New York: Springer-Verlag, 1986.
Purdie, Bob. Politics in the Streets: The Origins o f the Civil Rights Movement in Northern Ireland. Belfast: The Blackstaff Press, 1990.
Putnam, Robert D. “Diplomacy and Domestic Politics: The Logic of Two-Level Games.” International Organization 42 (1988): 427-460.
Quantin, Patrick. “The 1990 General Elections in Zimbabwe: Step Towards a One-Party State?” In Zimbabwe in Transition, edited by Simon Baynham. Stockholm, Sweden: Almqvist & Wiksell, 1992: 24-44.
Ravenhill, John. “Africa and Europe: The Dilution o f a 'Special Relationship'.” \xi A frica in World Politics, edited by John W. Harbeson and Donald Rothchild. Boulder, CO: Westview Press, 1991: 179-201.
Risse, Thomas. “A Europeanization of Nation-State Identities?” Paper presented at the American Political Science Association, Boston, MA, 1998.
Rolston, Bib. “Alienation or Political Awareness? The Battle for the Hearts and Minds of Northern Nationalists.” In Beyond the Rhetoric: Politics, The Economy, and Social Policy in Northern Ireland, edited by Paul Teague. London: Lawrence and Wishart, 1987: 58-80.
Rukobo, Andries Matenda. “Misplaced Emphasis in the Democracy Debate.” In The One Party State and Democracy, edited by Ibbo Mandaza and Lloyd M. Sachikonye. Harare: SAFES, 1991: 121-131.
Ryan, Stephen, ed. Ethnic Conflict and International Relations. Brookfield, VT: Dartmouth Publishing Company, 1995.
Sachikonye, Lloyd M. “The Context of the Democracy Debate.” In The One Party State and Democracy, edited by Ibbo Mandaza and Lloyd M. Sachikonye. Harare: Southern AJfrica Political Economy Trust, 1991: 43-60. Sakaike, Tonic. “Zimbabwean Leader Launches Election Campaign to Small Crowd.” The Reuter Library Report, March 4, 1990.
SAPA. “President Mugabe Issues New Year's Message.” Foreign Broadcast Information Service, 3 January 1990: 15.
402 SAPA. “ZUM Promises to End State of Emergency.” FBIS, 13 March 1990; 12.
Sawey, Marion. “Hume in Running for International Peace Award.” Irish News, April 14, 1987.
Schütz, Barry M. “Political Change and the Management of Ethnic Conflict in Zimbabwe.” In Conflict and Peacemaking in Multiethnic Societies, edited by Joseph Montville. Lexington, MA: Lexington Books, 1990: 433-447.
Shafer, Boyd C. Faces o f Nationalism. New York: Harcourt, Brace, Jovanovich, 1972.
Shamir, Boas, Michael B. Arthur, and Robert J. House. “The Rhetoric of Charismatic Leadership: A Theoretical Extension, A Case Study, and Implications for Research.” Leadership Quarterly 5, no. 1 (1994): 25-42.
Shamuyarira, Nathan M., Narendra Kumar, and Titivafi John Kangai, eds. M ugabe's Reflections. New Delhi, India: Har-Anand Publications, 1994.
Sisk, Timothy D. Power Sharing and International Mediation in Ethnic Conflicts. Washington, DC.: U.S. Institute of Peace, 1996.
Sisk, Timothy D. “Elections and Conflict Management in Aftica: Conclusions and Recommendations.” In Elections and Conflict Management in Africa, edited by Timothy D. Sisk and Andrew Reynolds, 145-171. Washington, D C.: United States Institute of Peace, 1998.
Sithole, Masipula. “The General Elections: 1979-1985.” In Zimbabwe: The Political Economy o f Transiton 1980-1986, edited by Ibbo Mandaza. Dakar, Senegal: CODESRIA, 1986: 75-97.
Sithole, Masipula. “Ethnicity and Democratization in Zimbabwe: From Confrontation to Accommodation.” In Ethnic Conflict and Democratization in Africa, edited by Harvey Glickman. Atlanta, GA: The African Studies Association Press, 1995: 121-160.
Sithole, Masipula. “Zimbabwe's Eroding Authoritarianism.” Journal o f Democracy 8, no. 1 (1997): 127-141.
Smith, Anthony. “Conflict and Collective Identity: Class, Ethnie, and Nation.” In International Conflict Resolution: Theory and Practice, edited by Edward E. Azar and John W. Burton. Boulder, CO.: Lynn Rienner Publishers, 1986: 63-84.
403 Smith, Anthony. “The Ethnic Sources of Nationalism.” \n Ethnic Conflict and International Security, edited by Michael E. Brown. Princeton: Princeton University Press, 1993 : 27-41.
Smith, Craig Allen, and Kathy B. Smith. The White House Speaks: Presidential Leadership as Persuasion. Westport, CT.: Praeger (Greenwood Publishing Group), 1994.
Smith, Anthony. “Gastronomy or Geology? The Role of Nationahsm in the R.econstuxcû.on oï'HdLÛons.’’'' N ations and N ationalism 1, no. 1 (1995): 2-23.
Smock, David R., and Audrey C. Smock. The Politics o f Pluralism: A Comparative Study o f Lebanon and Ghana. New York: Elsevier, 1975.
Snow, David A., E.B. Jr. Rochford, S.K. Worden, and R.D. Benford. “Frame Alignment Processes, Micromobilization, and Movement Participation.” Am erican Sociological Review 5 \ (1986): 464-481.
Snyder, Jack, and Karen Ballantine. “Nationahsm and the Marketplace of Ideas.” International Security 21, no. 2 (1996): 5-40.
Stedman, Stephen John. Peacemaking in Civil War: International Mediation in Zimbabwe. 1974-1980. Boulder, CO: Lynn Rienner, 1991.
Stem, Paul. “Why Do People Sacrifice for their Nations?” Political Psychology 16, no. 2 (1995): 217-235.
Sylvan, Donald A., Thomas M. Ostrom, and Katherine Gannon. “Case-Based, Model- Based, and Explanation-Based Styles of Reasoning in Foreign Policy.” International Studies Quarterly 38, no. 1 (1994): 61-90.
Sylvan, Donald, and James Voss, eds. Problem Representation in Foreign Policy Decision Making. New York: Cambridge University Press, 1998.
Sylvester, Christine. “Zimbabwe's 1985 Elections: A Search For a National Mythology.” The Journal o fModem African Studies 24, no. 1 (1986): 229-255.
Sylvester, Christine. “Unities and Disunities in Zimbabwe's 1990 Election.” The Journal of Modem African Studies 1^, no. 3 (1990): 375-400.
Sylvester, Christine. Zimbabwe: The Terrain o f Contradictory Development. Boulder, CO: Westview Press, 1991.
404 Sylvester, Christine. “Whither Opposition in Zimbabwe?” The Journal o f Modem African Studies 33, no. 3 (1995): 403-423.
Tajfel, Henri, and John C. Turner. “The Social Identity Theory of Intergroup Behavior.” In Psychology o f Intergroup Relations, edited by Worchel and Austin. Chicago: Nelson-Hall, 1986: 7-24.
Tamarkin, M. The Making o f Zimbabwe: Decolonization in Regional and International P olitics. Savage, MD: Frank Cass, 1990.
Tarrow, Sidney. Struggling to Reform: Social Movements and Policy Change During Cycles o f Protest, Western Societies Program Occasional Paper No. 15. Ithaca, New York: New York Center for International Studies, Cornell University, 1983.
Taylor, Rupert. “Running on Empty.” Fortnight, February 1990: 10-11.
Thaler, Peter. “How to Measure Identity: Austrian National Consciousness in the Mirror o f Public Opinion.” Nationalism and Ethnic Politics'}), no. 4 (1997): 69-93.
Tilly, Charles. From Mobilization to Revolution. Reading, MA.: Addison-Wesley, 1978.
Turner, John C., Penelope J. Oakes, S. Alexander Haslam, and Craig McGarty. “Self and Collective: Cognition and Social Context.” Personality and Social Psychology B ulletin 20, no. 5 (1994): 454-463.
Van Hattem, Margaret. “SDLP Faces a Poll-weary Electorate.” Financial Times, October 18, 1989.
Vasquez, John A. The War Puzzle. New York: Cambridge University Press, 1993.
Waever, Ole, Barry Buzan, Morten Kelstrup, and Pierre LeMaitre. Identity, Migration, and the New Security Agenda in Europe. New York: St. Martin's Press, 1993.
Walker, I., and T.F. Pettigrew. “Relative Deprivation Theory: An Overview and Conceptual Critique.” British Journal o f Social Psychology 23 (1984): 301-310.
Weiss, Ruth. Zimbabwe and the New Elite. New York: British Academic Press, I B. Tauris and Co. Ltd., 1994.
Weitzer, Ronald. Transforming Settler States. Berkeley, CA: University of California Press, 1990.
Wendt, Alexander. “Collective Identity Formation and the International State.” Am erican Political Science Review 88 (1994): 384-396.
405 White, Barry. John Hume, Statesmen o f the Troubles. Belfast: Blackstaff, 1984.
Wilson, Robin. “The Best Laid Plans.” Fortnight, December 1987: 7-8.
Wilson, Andrew J. “From the Beltway to Belfast: The Clinton Administration, Sinn Fein, and the Northern Ireland Peace Process.” New Hibernia Review 1, no. 3 (1997): 23-29.
Winter, David G., and Abigail J. Stewart. “Content Analysis as a Technique for Assessing Political Leaders.” In/I Psychological Examination o f Political Leaders, edited by Margaret G. Hermann. New York: Free Press, 1977: 27-61.
World Bank. Sub-Saharan Africa: From Crisis to Sustainable Growth. Washington, D C.: World Bank, 1989.
Wright, Stephen C., Donald M. Taylor, and Fathali M. Moghaddam. “Responding to Membership in a Disadvantaged Group: From Acceptance to Collective Protest.” Journal o f Personality and Social Psychology 58, no. 6 (1990): 994-1003.
Zald, Mayer N. “Culture, Ideology, and Strategic Framing.” In Comparative Perspectives on Social Movements, edited by Doug McAdam, John D. McCarthy and Mayer N. Zald, 261-274. New York: Cambridge University Press, 1996.
Zaller, John. “Elite Leadership of Mass Opinion: New Evidence from the Gulf War.” In Taken By Storm: The Media, Public Opinion, and U.S. Foreign Policy in the G u lf War, edited by Lance Bennett and David L. Paletz. Chicago: University of Chicago Press, 1994: 186-209.
406 IMAGE EVALUATION TEST TARGET (Q A -3 ) y
y
/ Vf
%
1.0 “ la 12.5 ta B= 112.2
12.0 l.l 1.8
1.25 1.4 1.6
150mm
yqPPUED ^ IIWIGE . Inc . = 1653 East Main Street • Rochester. NY 14609 USA .3 = % = Phone: 716/482-0300 Fax: 716/288-5989
0 1993. Applied Image. Inc.. Ail Rights Besenied