Crisis in Autocratic Regimes 9781626377257

Citation preview

CRISIS IN AUTOCRATIC REGIMES

CRISIS IN AUTOCRATIC REGIMES edited by

Johannes Gerschewski and Christoph H. Stefes

b o u l d e r l o n d o n

Published in the United States of America in 2018 by Lynne Rienner Publishers, Inc. 1800 30th Street, Boulder, Colorado 80301 www.rienner.com and in the United Kingdom by Lynne Rienner Publishers, Inc. 3 Henrietta Street, Covent Garden, London WC2E 8LU © 2018 by Lynne Rienner Publishers, Inc. All rights reserved Library of Congress Cataloging-in-Publication Data A Cataloging-in-Publication record for this book is available from the Library of Congress. ISBN 978-1-62637-673-1 (hardcover : alk. paper) British Cataloguing in Publication Data A Cataloguing in Publication record for this book is available from the British Library. Printed and bound in the United States of America The paper used in this publication meets the requirements of the American National Standard for Permanence of Paper for Printed Library Materials Z39.48-1992. 5

4

3

2

1

Contents

vii

Acknowledgments 1 Analyzing Crises in Autocratic Regimes Johannes Gerschewski 2 Anticipating Crises in Autocracies Martin K. Dimitrov 3 Disturbing the Dictator: Peaceful Protest Under Authoritarianism Andreas Schedler

1

21 43

4 Force Versus Institutions in Authoritarian Politics Milan W. Svolik

75

5 The Military and Autocratic Regimes: Roles of the Armed Forces Aurel Croissant, Tanja Eschenauer, and Jil Kamerling

89

6 The Impact of Elections: The Case of Uganda Svein-Erik Helle and Lise Rakner

111

7 Coordination and Crisis Prevention: The Case of Post-Soviet Eurasia Max Bader

135

8 The Resilience of Authoritarianism: The Case of the Chinese Developmental State Thomas Heberer

155

9 Historical Legacies and the Institutionalization of Military Rule: Lessons from Latin America Felipe Agüero and Julian Brückner

175

v

vi

Contents

10 Successes and Failures of Authoritarian Governance: The Case of the Arab Spring Frédéric Volpi

199

11 Crisis? What Crisis? Concluding Thoughts Christoph H. Stefes

217

References The Contributors Index About the Book

229 261 263 269

Acknowledgments

This volume resulted from a research project on the stability of autocratic regimes that the research unit Democracy and Democratization at the Social Science Research Center Berlin (WZB) hosted in 2011–2015. We thank the German Research Foundation, which provided funding for the project. We also received financial assistance from the Fritz-Thyssen Foundation for an international authors’ workshop in 2015. We’d like to express our gratitude to the contributors for sharing their ideas and providing insightful comments. We thank our collaborators Wolfgang Merkel, Alexander Schmotz, and Dag Tanneberg, who provided invaluable feedback on earlier drafts of the material. Furthermore, during the two years that we worked on the book, we could always rely on the efforts of several research assistants: Judith Wenner, Marcus Spittler, Bettine Josties, Jan Stuckatz, and Zoé Canal-Brunet. We are particularly grateful to Simon Haux for his unflagging support for this project. We’d also like to thank Gudrun Mouna for her many encouraging words and much appreciated administrative support. We offer our gratitude to Maria Stephan, Emmanuel Gyimah-Boadi, and Anvar Rahmetov for their important input and helpful comments at the first of two authors’ workshops that laid the foundation for the book. Johannes Gerschewski would like to thank his former colleagues at the WZB for their helpful input and kind encouragement. He learned a lot from long discussions with Julian Brückner, Kressen Thyen, and Saara Inkinen. Christoph Stefes would like to thank his colleagues in the Department of Political Science at the University of Colorado Denver who provided unwavering support throughout his extended sabbatical at the WZB. In particular, he is indebted to Betcy Jose for her encouragement and feedback.

vii

1 Analyzing Crises in Autocratic Regimes Johannes Gerschewski

In 1978, Juan Linz introduced his eminent work on “crisis, breakdown, and reequilibration” with the following words: A change of political regime affects millions of lives, stirring a spectrum of emotions from fear to hope. The March on Rome, the Machtergreifung by Hitler, the Spanish civil war, Prague in February of 1948, the coup against Allende—such dramas, symbolizing the transfer of power, become fixed in the memory of people as pivotal dates in their lives. Yet the events themselves are in truth the culmination of a longer process, an incremental political change that has evolved over a more or less prolonged period. Is there a common pattern in the process that has led to changes of the regime, or is each a unique historical situation? (Linz 1978, p. 3)

Our common research endeavor in this book is motivated by questions similar to those raised by Linz almost four decades ago, revolving around how and why a regime slides into crises. Linz and his colleagues focused on democracies, “with no attempt here to extend the study to authoritarian, totalitarian, or traditional political systems” (Linz 1978, p. 5). This is, however, exactly what the contributors to this book have in mind: while Linz’s seminal contribution identified the Achilles’ heel of young and mature democratic regimes, we embark on a similar endeavor for nondemocratic regimes. We ask what makes autocratic regimes prone to crisis situations. Why do some stumble in moments of crisis while others do not, and to what extent are we able to generalize beyond the idiographic case? Can we locate common patterns of vulnerability shared across different autocratic regimes? Are there parallels to be drawn with regard to the structures and functions that (de)stabilize autocratic rule? What roles do actors play in the emergence of crises? To formulate theoretically informed and empirically meaningful answers to these questions, we brought together chapters from leading scholars in the field of comparative autocracy research. The selected studies are firmly grounded in historical 1

2

Johannes Gerschewski

research, providing a detailed and in-depth view of empirical cases spanning all geographic regions of the world. In this chapter, I provide an overview of how to analyze crises in autocratic regimes. Adopting an institutionalist perspective, I outline common ground and define the concepts that the contributors use throughout the book. Our common analytical framework not only enhances the consistency of the collected work, but also is a necessary condition for the comparability across the cases studied. In essence, then, this chapter provides the building blocks for that common analytical framework. I want to stress that I do not aim here to offer concrete answers to the question of why autocratic regimes get into crisis situations. That task is left to Christoph Stefes, who in the concluding chapter builds on the rich insights found in the empirical chapters. Drawing on the empirical cases, we are able to put more empirical “flesh” on the theoretical “bones.” The rest of this chapter is organized in four parts. First, I discuss what a regime crisis actually means in the context of this book. Next, I review the state of research and identify connections to existing academic literature. In the third section, I put forth basic elements of how to analyze the emergence of regime crisis from an institutionalist perspective, and in the final section I outline the overall structure and contributions of the volume. What Is a Crisis? What is a regime crisis? Crisis appears to be an almost omnipresent term in academia. We currently discuss economic, Euro, and financial crisis, myriad ecological crises, demographic and immigration crises, the crisis in Syria, Libya, and Ukraine. Many academics almost routinely wonder whether or not democracy is in crisis, debating the implications of such a crisis (or its absence). Whenever a serious challenge or a malfunctioning occurs, we invariably announce a crisis. The German historian Reinhart Koselleck argues that the term “crisis” serves as nothing less than the “temporally elastic umbrella term of modern times” (Koselleck 1982, p. 631, my translation). It is important to remember that the semantics of the term crisis have changed considerably over time. Not only has its meaning changed, but usage of the term has also transcended disciplinary boundaries. This leaves us with the need to define what we actually mean when we refer to a crisis. Etymologically, the Greek krisis implied layers of different possible meanings spanning from judgment, choice, and the weighing of pros and cons to a concrete decision. It was used in the context of laws, courts, and sentences. In medicine, the Corpus Hippocraticum introduced the term to refer to the diagnostic findings and judgment about the process

Analyzing Crises in Autocratic Regimes

3

that eventually led to a decision determining whether a patient could be expected to recover or die. This original meaning of a decision that is due and urgent, but that has not yet been made, was translated to the social sciences. The medical usage of the term crisis was the gateway to social sciences. Most interestingly, its historic entry point was the anthropomorphic comparison of a political system to a body and its organs (a metaphor not uncommon to autocratic political thought). It has since gained unparalleled prominence in the academic language of social science (Koselleck 1982). The political usage of the term has historically centered around the aspect of decision, emphasizing that crises redefine the space of possibility and constrain alternative behavioral options. However, over time, the analytical clarity of the term crisis has suffered. In contrast to its former meaning, it has also been applied to situations that are not unique, but repeat themselves due to inherent structural malfunctions. Crises have come to no longer describe just short and bounded periods of time, but also are applied to capture whole epochs. The entire century seemed to be in crisis. Further confusing matters, another strand of literature has equated crisis with revolutions; that is, short-term (and often violent) ruptures in history (Koselleck 1982). In this light, it is important to delineate different types of crisis and I propose doing so along three dimensions. First, does crisis refer to a brief moment or a longer phase? It is more precise to speak then of an acute versus a chronic crisis. Second, is the crisis present (manifest) or is it building up in the background and about to break out (latent)? In Chapter 3 of this volume, Andreas Schedler adds a third dimension. He suggests differentiating between a functional crisis in which the partial performance of a regime is put into question and an existential crisis in which larger survival is at stake. This constitutes a typological space with duration, manifestation, and scope as the three dimensions. In this book, we concentrate mainly on one type of crisis: acute, manifest, and existential crisis. We also focus on regime crises. A regime is here understood as the set of formal and informal rules that govern the access to and the exercise of power. A regime crisis is therefore defined for us as an urgent situation requiring immediate action by the incumbents to maintain power. These are the life-and-death moments that the original medical usage of the term suggests. Crisis and Existing Research Crisis has been a catchword for the social sciences for decades. Nevertheless, I argue that the emergence of crises in autocracies has received less

4

Johannes Gerschewski

attention than it deserves. I approach the subject of this book from three different angles: the democratization literature, the comparative autocracy research, and the neoinstitutionalist turn, placing particular emphasis on gradual forms of change that accumulate to a crisis moment. A deeper understanding of how the “termites in the basement” (Mahoney and Thelen 2010, p. 31) hollow out the autocratic substance is a crucial task. It sheds light on the long and evolving processes that eventually make an autocratic regime slide into existential crisis. The first corpus of work is associated with the democratization paradigm (Huntington 1991; O’Donnell and Schmitter 1986; Whitehead 2002). Not surprisingly, its point of departure is the end of the autocratic regime and the subsequent initiation of a complex democratization process. Accordingly, it disregards to a large extent the inner dynamic of the autocratic predecessor regime. The general assumption underpinning this literature is that an autocratic regime eventually reaches a tipping point, after which an uncertain democratization path unfolds. Adam Przeworski famously claims that the “crucial moment in any passage from authoritarian to democratic rule is the crossing of the threshold beyond which no one can intervene to reverse the outcome of the formal political process” (1991, p. 14). The end of the autocratic regime is typically portrayed as being a (given) historic caesura. The scholarly focus in the 1980s and 1990s was therefore not on longer-term trends explaining why and how an autocratic regime might have been hollowed out. These scholars instead explained the final fall of a regime through close analysis of deliberate political action within and against the incumbent regime, tracing the subsequent complex transition to, and consolidation of, democracy. Over the past few years, however, a new research agenda has emerged that explicitly deals with the internal stability and performance of autocratic regimes. Barbara Geddes’s (1999) groundbreaking contribution is often quoted as the starting point of this new research endeavor. She argues that some types of autocratic rule persist longer than others with one-party rule proven to be the most robust regime form. Geddes’s work triggered other academic studies that have since supported her findings on autocratic regime stability (e.g., Hadenius and Teorell 2007). Building on her analysis, in recent years the study of autocratic regimes has strongly focused on institutions (Pepinsky 2014; see also Schedler 2013). Most prominently, the roles of political parties in autocratic regimes have been thoroughly reassessed. In classic accounts, their role is reduced to a mere power structure whose function is to provide a stable and reliable constituent base (Friedrich and Brzezinski 1956). More recently, scholarly focus has shifted to the various ways by which a political party can stabilize autocratic rule. Jason Brownlee, for instance,

Analyzing Crises in Autocratic Regimes

5

makes a notable contribution by advancing an internal cohesion argument. He argues that autocratic regimes last longer when backed up by a powerful party because the party “curb[s] leaders’ ambitions and bind together political coalitions” (2007, p. 37). Parties not only are mobilization vehicles for creating mass support, but also have an internal disciplinary effect by mediating and settling intra-elite conflict. They prolong time frames and make actions in general more predictable, including the facilitation of smooth leadership transitions within an incumbent autocratic regime. Political parties not only manage elite rivalries, but also “restrain the conflicts of actors in power” (Brownlee 2007, p. 38, emphasis in original). In a similar vein, Beatriz Magaloni and Milan W. Svolik examine the impact of institutions on the longevity of autocratic rule by expanding the analysis to focus on internal power-sharing arrangements and showing how institutions make commitments among the elite more credible. Institutions diminish the moral hazard problem (Magaloni 2008; Svolik 2009, 2012,). Jennifer Gandhi and Adam Przeworski also demonstrate that institutions neutralize threats and make autocracies last longer (Gandhi 2008; Gandhi and Przeworski 2007). Besides the role of political parties, other institutions such as cabinets (Arriola 2009) and elections (Blaydes 2011; Gandhi and Lust-Okar 2009; Heberer and Schubert 2006; Schedler 2013) have been found to stabilize autocratic regimes. These diverse academic studies focus on different institutions, but they converge in the research objective of explaining autocratic regime stability. However, less emphasis is paid to the gradual decline and the emergence of crises. These studies focus more on the puzzling question of why autocratic regimes survive and less on asking how their once stabilizing institutions may have been slowly undermined. This book is meant to connect to the nascent literature on institutional evolution in autocratic settings (Bunce 1999; Geddes 2004; Smith 2005; Slater 2010). Our collective analytical interest primarily lies in the emergence and decay of institutions. We are interested in the dynamic process of how institutions come into being in the first place and how they gradually lose their constraining force on actors’ behavior. Against that backdrop, this volume should be understood as a complement to these important studies by providing insights into the long-term development of institutions. By doing so, this book also speaks to a third, more general, strand of literature associated with the neoinstitutionalist turn (Thelen, Steinmo, and Longstreth 1992; Hall and Taylor 1996; Immergut 1998; Thelen 1999; Pierson 2004). More concretely, we are interested in gradual institutional change that takes place over longer periods of time. The core theorem of (historical) institutionalist explanations is the dualism of long phases of institutional stability that are interrupted by critical junctures,

6

Johannes Gerschewski

which then again are the origins of far-reaching and stable developments (Capoccia and Kelemen 2007; Collier and Collier 1991). When it occurs, rapid change is assumed to be externally induced. Newer research has proposed two major modifications to this basic model (Mahoney and Thelen 2010). On the one hand, more emphasis has been placed on endogenous sources of change. On the other hand, the focus has been shifted from sudden rupture to long-term change that gradually undermines the status quo. In the following section, I outline these modifications and discuss the emergence of crisis within autocratic regimes. How Do Crises Emerge? Analytical Building Blocks The challenge in theorizing the emergence of crises in autocratic regimes is a classic one: How can actors change an institution (or a structure more generally) that constrains them? How can actors overcome institutional straightjackets? Or, from a macro perspective, why do structures lose their influence on people’s behavior? In the pages that follow, I first outline where crises occur. I then discuss how they occur, distinguishing between sudden ruptures and a gradual lead-up to crisis; and consider the deinstitutionalization process, its potential causal drivers, and concrete mechanisms. Locus: Horizontal Versus Vertical Games

Where do crises occur? Svolik (2012) has introduced an apt distinction between two forms of challenges that every autocracy faces: a vertical and a horizontal challenge. Autocratic regimes need to control the opposition and the population with the instruments available. They can either try to repress the opposition and instill fear among their citizens or they can try to generate mass support (Gerschewski 2013). This game between the ruler and the ruled corresponds to a vertical axis. The horizontal axis, in turn, pertains to existing power-sharing arrangements among members of the ruling elite. Here, the regime needs to maintain unity within the elite, and power sharing must be credible and relatively predictable. Moral hazard problems must be avoided to achieve and maintain elite cohesion (Magaloni 2008; Svolik 2009). As Figure 1.1 shows, crises in autocratic regimes can occur along both the vertical and horizontal axes. They can occur from below (i.e., in the interaction between regime opponents and members of the elite) and from above (i.e., between elite members). In the former case, we observe a mobilization of political protest from those that turn against the regime and join the opposition. In the latter case, splits within the ruling elite serve as an apt

Analyzing Crises in Autocratic Regimes

7

Figure 1.1 Loci of Crisis in Autocratic Regimes

Observable indicators for

Elite level

Horizontal game of maintaining intra-elite unity

Intra-elite splits

People’s level

Vertical game of repressing dissent and creating legitimacy

Protest mobilization

Source: Author’s illustration.

indicator for internal disunity among regime decisionmakers. Of course, this distinction is only for analytical purposes to help pinpoint the origin of a regime crisis as defined in this book. In reality, spillovers, mutual infections, and contagions can be observed during complex transformation processes. For example, a protest movement that is perceived as either threatening or legitimate by elite members might serve as a catalyst for internal splits. This can also work vice versa: disunity in the elite might be interpreted as encouragement for the opposition to form a protest movement. These interactive dynamics often reinforce each other and deserve the close attention of political scientists. Yet accounting for the horizontal-vertical distinction contributes to a more precise analysis of the locus of crisis and improves the detection of a starting point for later political contagion effects. Tempus: Sudden Ruptures Versus Gradual Lead-up to Crisis

How do crises emerge with regard to time? The bread-and-butter theorem of historical institutionalist explanations provides a first answer. The dualism between path dependency and critical junctures is a prominent analytical lens in comparative politics. Steven D. Krasner has called this the “punctuated equilibrium” (1984, p. 226) approach. It suggests a branching tree metaphor. After a juncture appears and a decision is made, that first step forecloses other options. Such thinking has also been prominent in the social sciences where it is usually referred to as path dependency. Once a particular path is taken, it becomes increasingly difficult to alter it. Instead, the first step shapes future choices and sets up a long-term trajectory. Against this background, Ruth Collier and David Collier have proposed a related theoretical framework. They

8

Johannes Gerschewski

set forth a sequential theory of (1) structural antecedent conditions that lead (2) to a cleavage or crisis that again (3) triggers a critical juncture. This is a brief moment in time, a watershed, in which rapid change is possible. Through (4) a crystallization process of reactions and counterreactions a legacy is produced which again leads (5) to a stable equilibrium core of attributes (Collier and Collier 1991).1 Critical junctures open up space for change and represent moments in which new paths can be generated, that again become difficult to alter (Capoccia and Kelemen 2007; Mahoney 2000; Pierson 2004). However, the critical juncture versus path dependency discourse has remained largely silent on the central questions of how crises emerge in the first place and how change becomes possible. This approach brackets out the emergence of critical junctures, avoiding such questions by referring vaguely to structural antecedents that make change possible “in distinct ways in different countries” (Collier and Collier 1991, p. 29). The neighboring rational choice approach to institutions faces the same challenge.2 While this approach has the advantage of paying greater attention to actors and allows for the refining of historical insights with the help of formal modeling techniques, structural sources for change have only been found exogenously and in the form of rapid and sudden moments (Shepsle 1986, 1989). Peter A. Hall and Rosemary C. Taylor have called it a “contradiction” (1996, p. 953) to reconcile institutional stability and institutional change in such a rational choice fashion. The contradiction can be resolved only by reference to an external shock that profoundly affects the rules of the game. While in rational choice approaches actors have no incentive to deviate, even sociological explanations have generally “outsourced” (Mahoney and Thelen 2010, p. 5) their explanation of change. Like the former, these are also better suited to explain stability rather than change. Typical recourses are made to concepts such as socialization and reification, the taken for grantedness as well as the routine of cognitive scripts (DiMaggio and Powell 1983; Meyer and Zucker 1989; W. W. Powell and DiMaggio 1991; Clemens and Cook 1999). However, these are all means of institutional reproduction that do not clarify how and why political change occurs. The three neoinstitutional approaches (the historical, the rational choice, and the sociological) share an intrinsic strength in explaining why some institutions exhibit inertia and remain persistent over time. However, they also all struggle to explain the emergence of crisis situations. Loosely following the Gould-Eldredge type of change in which sudden leaps bring about change, these three approaches all have difficulties in explaining why these crises occur in the first place. Against this backdrop, two major critiques have emerged that are important for our own theoretical reasoning on crisis in autocracies. On the

Analyzing Crises in Autocratic Regimes

9

one hand, the research on sudden ruptures needs to be complemented with closer consideration of the potential for more gradual change. On the other hand, the reliance on an external and largely unexplained deus ex machina that triggers sudden ruptures also needs to be complemented (Weyland 2008). James Mahoney and Kathleen Thelen (2010) argue that all three branches of neoinstitutionalism take compliance with institutional rules and norms as given. In contrast, Mahoney and Thelen argue that compliance is contested and that there is always a struggle over the proper meaning of an institution; the institution’s “guiding expectation often remains ambiguous and always are subject to interpretation” (2010, p. 11). If institutional demands are contested and compliance is not given due to emerging cognitive disequilibria, but needs to be achieved instead, then this opens up new space for endogenous sources of change (Aoki 2001). Compliance with institutional demands for action should be taken as the central variable of inquiry. This not only makes room for endogenous sources of change, but also provides an opportunity to think more systematically about gradual forms of change. If an institution is always contested, the struggle between actors over its proper meaning and effect can be a lasting and incremental one. The one-sided view on critical junctures and path dependency should be complemented with a parallel understanding of endogenous and gradual forms of institutional degeneration. A regime crisis can occur in both ways: either as the result of an exogenous shock that leads to a rapid deinstitutionalization (see also Chapter 10 by Frédéric Volpi in this volume), or as the product of a long and gradual hollowing out of an institution. Exogenous shocks are by definition difficult to explain. They arrive almost unnoticed and can potentially develop a sudden and overwhelming outcome. Revolutions or spontaneous mass protests come to mind, for example. They are unanticipated, by both political actors and political scientists. However, most shock events do not fall from the sky. They are almost always foreshadowed by various signals and warnings that often go unnoticed, however, due to their brevity, tempo, and acceleration. The gradual hollowing out of institutions that become defunct over time is usually easier for observers to track. The challenge here lies in detecting piecemeal change. It happens over a longer period, is slower, and usually has a constant velocity without acceleration. However, what the sudden ruptures and the gradual hollowing out share is a decline prior to the crisis. I try to make sense of this decline by using the idea of deinstitutionalization processes. The Deinstitutionalization Process

A crisis usually results from the prior decline of institutional capacities. An institution, defined as the “rules of the game” (North 1990, p. 3), is able to

10

Johannes Gerschewski

shape and influence individual and collective actions. Institution should be understood as a placeholder in this general discussion, not intended to denote a concrete meaning. In specific empirical investigations, the placeholder “institution” can refer to formal and informal institutions ranging from political parties, parliaments, courts, and elections to corruption, patronage, and other informal practices. The important common denominator uniting formal and informal institutions, however, is that they “exert patterned higher-order effects on the actions” of individuals (Clemens and Cook 1999, pp. 444–445). To be sure, an institution does not determine what action ought to be taken, but it does provide a powerful incentive structure that changes the individual calculation of risk and probabilities in the formation of preferences, the construction of actor coalitions, and the eventual action itself (Katznelson and Weingast 2005). Institutionalization is the process by which an institution gains behavioral power over actors. The classic definition proposed by Samuel Huntington in his early work describes it as a “process by which organization and procedures acquire value and stability” (1968, p. 12). The process of institutionalization unfolds over a longer period of time and, as it progresses, the behavioral demands on actors tend to intensify. From this perspective, successful institutionalization can be characterized as the process by which an institution exerts its patterned higher-order effect on more actors, in more circumstances, more often. In contrast, deinstitutionalization can be defined as the reverse process. Deinstitutionalization is then a process by which the behavior-shaping power of an institution declines. Deinstitutionalization can be characterized as the process by which the actions of fewer actors in fewer circumstances are constrained less often by institutional pre- and proscriptions. It creates new space for alternatives; for the purpose of this book, deinstitutionalization in autocratic regime contexts opens up space for oppositional demands. This deinstitutionalization process can be driven either by actors or by shifts in the structural parameters. Moreover, I distinguish here between two mechanisms: an inherent design error at institutional inception; or a later occurrence that puts an institution on an erroneous development that creates perverse incentives undermining the original institutional design. In the following I discuss, first, the different causal drivers, and, second, the mechanisms. Causal Drivers: Change Agents and Structural Quasi-Parameters

What are the causal drivers behind such a deinstitutionalization process and the subsequent emergence of crises? Among the major developments of his-

Analyzing Crises in Autocratic Regimes

11

torical institutionalism has been the inclusion of power in the explanation of outcomes (Pierson 2015). Institutions should be understood as vehicles for both power allocation and power (re)distribution (Streeck and Thelen 2005). In a recent contribution, Giovanni Capoccia (2016) pushes the debate further by proposing that academics think more systematically about microfoundational perspectives and more closely observe bottom-up processes. In this light, Capoccia (2016) emphasizes the role of shifting social coalitions and points to the specific role of “change agents” (Mahoney and Thelen 2010; see also Bernhard 2015). These agents are the major drivers for change. Since institutions are always contested and compliance with their norms and rules cannot be taken for granted, these change agents are capable of creating a crisis moment. Because coalitions are often built on initial power asymmetries, losers of the arrangement might have a strong incentive to defect at a later stage, implying constantly shifting coalition membership. This is in line with most causal accounts that view actors and actor coalitions as being the initiator and source for change. From this point of view, it is actors that bring about crisis—and nothing else. In our context of autocratic regime crisis, the fundamental texts by Guillermo O’Donnell and Philippe C. Schmitter (1986) as well as Przeworski (1991) already argue that splits in the elite between hard- and soft-liners and changing loyalties are a prerequisite of democratization processes. While this research program has focused on intra-elite struggles, Svolik (2012) adds an important second axis, the vertical one. Change agents can also arise from the vertical game between the ruling regime on the one hand and the opposition on the other hand. If the opposition is able to mobilize and organize dissent with the regime, then a crisis looms large. Moreover, the role of ordinary citizens who are not formally part of the mobilized opposition should not be underestimated. An autocratic regime needs to provide an effective justification formula to legitimate its grip on power. Autocracies need to control the masses as well (Gerschewski 2013; Gerschewski et al. 2013). Protest by normal citizens often emerges as a result of popular perceptions of declining regime legitimacy. With this in mind, the most important (collective) change agents come from the elite, the opposition, and ordinary people. However, to a certain extent, the explanatory route via change agents relegates the reasons for crisis and change to the black box of actors and their myriad competing identities, interests, and preferences. In particular, structuralists find it unsatisfying to merely refer to actors, likely wondering: Why do actors change their motivations in the first place? Why do they want to change an existing institution? As such, the explanatory factor for crisis, (i.e., emergence of change agents and changing actor coalitions) is already the outcome of what is actually studied. Rule defection and the

12

Johannes Gerschewski

noncompliance of rule-takers with the demands of rule-makers is not the explanation. In other words, endogenous change has already taken place that has prompted important actors to reconsider their choices—and it is this that deserves to be explained. If we take the alternative route and focus on structural factors (instead of actor motivations), the explanatory dilemma is, however, turned upside down. One needs to find a solution for a seemingly paradoxical situation: an institution that once contributed to stabilizing autocratic rule has now turned into its opposite, becoming a reason for regime crisis. In other words, how does what once worked in one direction (institutionalization) now turn in the opposite direction (deinstitutionalization)? How can such a paradox be tackled in a way that goes beyond reference to actor motivations and change agents? An interesting idea of explaining structural institutional change was proposed by Avner Greif and David D. Laitin (2004, p. 633), the crux of which was their introduction of “quasi-parameters.” They proposed a tripartition into (1) parameters that are fixed and exogenous to the institution under study; (2) variables that vary and are endogenous to the institution; and (3) quasi-parameters that are fixed in the short run, but can vary over time and therefore become endogenized. Quasi-parameters thereby serve as an explanatory bridge. If a quasi-parameter changes, it either reinforces an institution (and, hence, makes the institution more robust), or it triggers the opposite effect when the institution refers afterward to a smaller set of situations. It then undermines the behavioral effect of the institution on actors (Greif and Laitin 2004). The main idea is that small changes in these quasiparameters are not recognized by the relevant actors and therefore are perceived as parametric (i.e., exogenous and fixed) in the short run. In the long run, however, they can vary and may turn into an endogenous force of change. In this sense, the introduction of quasi-parameters provides us with more leeway to solve the central paradox of how something that has long been a stabilizing factor can turn—at least in the medium and long run— into a source of its own demise. This idea is well illustrated by an example in this book. In their case study on electoral politics in Uganda (Chapter 6), Svein-Erik Helle and Lise Rakner argue that multiparty elections can have—depending on the quasi-parameters—both a stabilizing and a destabilizing effect on nondemocratic regimes. The case of Uganda is particularly apt for such a research endeavor as it hosted three different institutional landscapes during the presidency of Yoweri K. Museveni. This provides the authors with enormous analytical leverage because they can follow a comparative withincase design. They show that the adoption of multiparty elections can be interpreted as a response to the decaying no-party rule. The introduction of

Analyzing Crises in Autocratic Regimes

13

double-edged multiparty elections was therefore a strategic choice by the incumbents. Helle and Rakner are able to show with their rich empirical material that these elections indeed had a stabilizing effect for the regime. However, they also argue that this success might be seen only in the shortrun and only as long as rural strongholds continue to supersede the influence of the urban population. This rural bias in Ugandan electoral politics is what we would define as a quasi-parameter. The rural voter advantage enjoyed by the regime is likely to change with an increasing number of Ugandans heading toward the cities. This ongoing rural exodus will have important future repercussions for the power equilibrium in Uganda. While the institution of multiparty elections did indeed stabilize the Museveni regime, it may turn out to be the source for a deinstitutionalization process in the foreseeable future. The factor of rural bias in this light can be interpreted as the hinge between a potential stabilizing and destabilizing effect of the institution. (In treating it as a quasi-parameter, we might be able to find a way out of the paradox of what stabilized the regime for some time might contribute to its eventual decay in the long run.) Mechanisms: Construction Error Versus Perverse Incentives

As of now, we have clarified locus (horizontal vs. vertical game), tempus (rapid vs. gradual), process (deinstitutionalization), and drivers for change (change agents vs. structures). In the last step, I focus on how institutional compliance with an established rule can be undermined over time. I propose here two mechanisms. On the one hand, an institution can be designed erroneously from the very beginning in a way that subverts its intended objective in the long run. Such an institution is impaired by an inherent fundamental flaw. Such cases refer to institutions afflicted by an initial construction error that later leads to a crisis. On the other hand, a soundly designed institution can also evolve over time to develop in a manner different from what was originally envisioned. This deviation from its original developmental path then distorts the institution and leads to unforeseen pathological and perverse incentive structures. To illustrate this distinction between initial construction errors on the one hand and the development of perverse incentives on the other, consider the following empirical examples. The breakdown of the Ferdinand Marcos regime in 1986 is a good case in point to illustrate the construction error mechanism (Lande 1986). Marcos’s rule was personalistic in style, and his inability to build a reliable group of followers supportive of his regime was a major factor in his downfall. He empowered new economic elites and sought to maintain their loyalty by fostering dependence on him and his goodwill. However, Marcos also alienated and repressed

14

Johannes Gerschewski

some traditionally influential family clans such as the Lopez, Aquino, and Osmeña families (Kang 2002). This exclusionary style of co-optation was his regime’s Achilles’ heel. It can be effectively argued that Marcos’s cooptation efforts were ill designed from the very beginning. Instead of incentivizing elite cohesion, his approach bred fierce and powerful internal opposition from the first days of his regime. While his exclusionary style worked temporarily and could be compensated in the short term by other stabilizing mechanisms, it remained a vulnerability throughout his tenure before he finally fell. While the Marcos regime’s co-optation efforts can be interpreted as an example of institutional construction errors at the point of inception, a wellknown example of how perverse incentives develop over time is illustrated in modernization theory more generally and the “developmental state” (Johnson 1982) more specifically. In Chapter 8, Thomas Heberer uses the concept of developmental states to explain the remarkable endurance of the Chinese autocratic regime over time. The core idea of the developmental state is a strong state that directs and steers economic and social policies largely independent from societal demands. This feature brings about at least an elective affinity to authoritarianism (Woo-Cumings 1999). Heberer argues with great insight that the Chinese regime is now experiencing a new version of the developmental state, in which the local-state nexus is strengthened and greater emphasis is placed on innovations and sustainability. In so doing, the regime successfully adapts its legitimacy formula to a fluctuating environment, but this adaptation can also result in unintentional changes to existing incentive structures. Political demands for greater transparency and less corruption, which might be encouraged by the new policy direction, could also backfire in the long run. Heberer lists concrete demands ranging from the disclosure of incomes and performance evaluation of senior officials and party cadres, to the abolishment of torture and reeducation camps and the liberation of political prisoners. This sets perverse incentives that were unintended at the beginning of the developmental era under Deng Xiaoping. To sum up, theoretical reasoning about the emergence of crises in autocracies consists of several building blocks introduced above. I defined deinstitutionalization as the process by which an institution gradually loses its influence over people’s behavior. This creates new space for oppositional demands, potentially resulting in a regime crisis. While sudden ruptures have traditionally been the central research focus, given the classical interest in analyzing critical junctures and path dependency, newer research has pointed to the gradual and endogenous origins of regime crises. The first step of such a gradual change can be located in the quasi-parameters discussed above. This is not a prerequisite for the explanation. However,

Analyzing Crises in Autocratic Regimes

15

accounting for quasi-parameters often provides context essential to understanding why the ambiguity of an institution becomes visible in the first place, and why the resistance of change agents and shifting actor coalitions is initiated. In some cases, the origin of the deinstitutionalization process can be traced back to its initial design. An institution can sometimes be afflicted with a fundamental flaw from the very beginning, meaning that the institution and its behavior-shaping influence never fully develops as foreseen. In other cases, deinstitutionalization sets in after a certain period in which the institution worked properly, but then develops perverse incentives that lead to unintentional consequences—notably sliding into a regime crisis. Table 1.1 summarizes the building blocks of analyzing crises in autocratic regimes. Structure of the Book The next four chapters (Chapters 2–5) make theoretical and conceptual contributions, while Chapters 6–10 present in-depth case studies. In all of these chapters, a distinction is made with regard to the sources of the

Table 1.1 Crises in Autocratic Regimes Type of crisis Locus of crisis Indicators for crisis Tempus of crisis

Causal drivers

Mechanism

Acute and chronic Latent and manifest Existential and functional/partial

Horizontal game within elite and vertical game between elite and people/opponents

Intra-elite splits (horizontal game) and protest mobilization (vertical game)

Sudden rupture vs. gradual lead-up

Actor-driven (change agents, coalitions of actors) Structural factors and quasi-parameters as bridge between exogenous, long-term, and fixed parameters and short-term endogenous variables Spillover and contagion effects between institutions under study

Construction error from the very beginning or erroneous development of perverse incentives in institutions under study

16

Johannes Gerschewski

crisis, clarifying whether they stem from oppositional challengers to existing rule or from within the regime’s elite. These are the two weak spots of authoritarian rule, corresponding with the vertical and horizontal powersharing games. Chapter 11 summarizes our findings and suggests avenues for further research. In Chapters 2 and 3, Martin Dimitrov and Andreas Schedler focus on the vertical axis between the ruler and the ruled. They discuss the relevance of public protest. While Schedler (Chapter 3) analyzes protest and asks to what extent the opposition becomes a change agent that challenges autocratic rule, Dimitrov (Chapter 2) goes on to the interesting distinction between hidden and overt forms of discontent. To avoid regime crisis, authoritarian regimes need to undercut the transformation of the former into the latter. To avoid crisis, an authoritarian regime needs to create and maintain effective “anticipatory ex-ante governance.” This effectiveness is measured with a pioneering and novel approach: via archival citizen complaints, which are interpreted as the goodwill of the population to voluntarily transfer their discontent to the regime. However, this goodwill depends on material demands that the regime can or cannot fulfill. Schedler’s contribution is thought provoking in the best sense and challenges what we know about protests in autocracies (Chapter 3). He contrasts two dominant perspectives: a first strand of literature that argues that protests are rare in autocratic regimes—and if they appear lead to severe crisis and disruptions; and a second strand that argues that we should perceive protests as frequent events with an information function for the ruling regime. If protest occurs, it remains mostly within preaccepted limits and does not trigger an existential regime crisis. In light of these contrasting views, Schedler explores a new dataset and highlights the “disturbing normality” of protest. He cannot find empirical support either for protests being an apt indicator of crisis or for an “anticipatory obedience of risk-averse protesters” that limit themselves to certain domains. Instead, it appears to be genuine protest, but not disruptive protest. It is not self-censored protest, but rather open dissent that aims at disturbing the legitimacy narrative of the ruling regime. In contrast to Dimitrov’s and Schedler’s focus on key themes of the vertical game between ruler and ruled, Milan W. Svolik is interested in horizontal threats that emerge from inside the elite. His main research interest revolves around how to interpret institutions in authoritarian settings, and he proposes that we rethink our common understanding of institutions. Compliance with institutions is a puzzle that needs to be solved and cannot be easily assumed or discounted. Instead, we need to ask how institutions shape actors’ behavior. Do they set the incentives that were originally intended or do they also facilitate the development of

Analyzing Crises in Autocratic Regimes

17

perverse incentives over time? What are the consequences of an institution such as term limits and what incentives do they provide for key actors? Answering these questions, as Svolik points out, also has methodological implications. Our focus on formal political institutions might need a corrective because “the man who gives orders might not reside in the presidential palace but rather across the street from it.” Svolik’s contribution to this volume can be applied to different kinds of institutions. Empirically, he demonstrates his idea on an elite-level institution that seeks to ensure cohesion among the party cadres. Aurel Croissant, Tanja Eschenauer, and Jil Kamerling take up this perspective of the horizontal game in Chapter 5. They apply it to the second elite group that matters most in creating and avoiding crisis: the military. The authors develop a novel and innovative dataset on military influence. They differentiate between the military as a hidden ruler on the one hand and the military as a supporting structure to the ruling regime. They systematically test whether these different military regimes are more prone to crises phenomena and if these crises occur rather at the horizontal intra-elite axis or at the vertical elite-people nexus. They provide a novel picture to help understand why and how different types of military regimes slide into crisis. Croissant, Eschenauer, and Kamerling’s contribution should be understood as a more nuanced discussion of an important subtype. Military regimes differ from each other perhaps as much as they differ from other autocratic subtypes. The contributions of Schedler and Dimitrov focus on key themes of the vertical game between the ruler and the ruled, while Svolik and Croissant et al. expand the view to the horizontal game by respectively discussing the role of the party and the military in nondemocratic rule. Svein-Erik Helle and Lise Rakner (Chapter 6) examine the inner tension of elections as an institution. As discussed above, Helle and Rakner’s contribution can be read against the theoretical question of why an institution’s effect on actors can change over time. How can we explain that institutionalized multiparty elections in Uganda have a stabilizing effect (in the short run) while also providing potential for destabilization (in the medium and long run)? To tackle this paradox, the authors look at the horizontal challenge to the regime as well as to the challenges that slowly started to emerge from the population. That elections are Janus-faced and may have crisis-inducing effects has also been noted by observers of the postelection protests of the color revolutions in the post-Soviet space. Max Bader focuses on this type of regime crisis in Chapter 7. He adds a transnational perspective to understanding why autocratic regimes slide into crisis. The regimes of Azerbaijan, Belarus, Kazakhstan, Russia, Tajikistan, and Uzbekistan are all consolidated autocracies in the region that cooperate via various formats, most

18

Johannes Gerschewski

notably the Shanghai Cooperation Organization (SCO) and the Commonwealth of Independent States (CIS). These intergovernmental forums allow for mutual learning, through three mechanisms highlighted by Bader. First, these regimes learn from each other about how to prevent elections from having unintended consequences. Second, they provide a forum for political solidarity and support for each other during times of crisis. Third, Bader is able to trace the process of legal emulation: the autocratic regimes in the region even coordinate their legal activities for best practice. Thomas Heberer (Chapter 8) studies the Chinese case. With rich empirical material from original and secondary sources as well as from his intensive field work, he argues that the Chinese state has been quite successful in avoiding a regime crisis since 1989 by institutionalizing a developmental state with strong local roots. However, as noted above, Heberer also makes clear that while the institution has been a successful formula in the short and medium run, it might also lead to unintended perverse incentives in the long run. Avoiding crises by being responsive to innovations and local developments might turn out to be the beginning of a longer deinstitutionalization phase in which the autocratic regime increasingly loses its grip on power. Turning to Latin America, Felipe Agüero and Julian Brückner (Chapter 9)—like Croissant, Eschenauer, and Kamerling in Chapter 5—argue that military regimes are not internally coherent, but should be internally differentiated. “Military regimes,” despite being a common and widely used concept in comparative authoritarianism, should be disaggregated. By reviewing and reviving the classic works of the 1970s and 1980s in the field, they highlight the role of historical legacies in shaping the institutional choices of military rulers. They illustrate their reasoning with two typical case studies that shed light on the inner working mechanisms of these regimes. While Peru has followed an inclusionary ruling style by incorporating different actors, Chile is a prime example for the exclusionary role model. The former attempts to attract a large coalition of peasants (via land reforms) and political parties as well as organizations like trade unions that are mostly placed on the political left. The latter is more restrictive and bans these parties and organizations. With this distinction, Agüero and Brückner demonstrate the weak spots of these regime subtypes and show how they gradually slide into crisis. Frédéric Volpi analyzes the causes of the Arab Spring in Chapter 10. Volpi argues that standard models of explanations have not only missed a causal factor when accounting for the Arab Uprisings. Instead, he calls our attention to “eventful sociology” and the idea of “transformative events.” We need to interpret the role of agency differently in times of rapid deinstitutionalization versus times of routine governance. This compelling idea

Analyzing Crises in Autocratic Regimes

19

is applied to the horizontal and the vertical games in autocracies. Moreover, he provides insightful suggestions about the origins and nature of regime crisis and the role that different actors play in them. In the final chapter, Christoph H. Stefes connects and summarizes the various contributions of this volume. Here, he picks up Svolik’s suggestion that institutional stability fundamentally rests on distinct balances of power that in turn are engendered by rules and norms. Slowly changing structural factors might upset this reinforcing cycle of institutions and balances of power, leading to the emergence of autocratic regime crises. Stefes shows how this has been the case in all instances of actual and potential regime crises analyzed in this volume. He concludes the book with a call for further research into the structural parameters and roles of actors in bringing about and shaping regime crises, borrowing from Volpi in Chapter 10. Notes I would like to express my gratitude to the generosity and patience of the colleagues who collaborated with us on this book. We had two fruitful and inspiring workshops at the WZB Berlin Social Science Center. This introduction benefited immensely from the discussions that we had. I am particularly grateful to Julian Brückner, Martin Dimitrov, Andreas Schedler, Lise Rakner, and Farrah Hawana for insightful and detailed comments. 1. This sequential theory of institutional change has later been refined by James Mahoney. He presents an analytical frame in which also (1) antecedent conditions define the range of historical options that are available for actors. The actors decide during (2) critical junctures, which option is taken that sets in motion a consequential chain. At this point, he expands the definition of critical junctures by stressing that different options must be on the table. Otherwise, we could not speak reasonably of a critical juncture. This critical juncture is then followed by an institutional pattern that is characterized by (3) a structural persistence in which this pattern is produced and reproduced and by (4) a reactive sequence in which reactions and counterreactions to the institution take place. Finally, (5) the regime outcome represents the solution of the conflict; it is a new institutional pattern, for example, a distinct regime type (Mahoney 2001). 2. The following discussion is particularly informed by the insightful article by Mahoney and Thelen (2010). As it is central to my own perspective, I take up their main argument on the blind spots of neoinstitutionalist explanations of gradual change.

2 Anticipating Crises in Autocracies Martin K. Dimitrov

All autocracies experience crises, some of which precipitate collapse. One explanation for the emergence of such crises is the inability of autocrats to collect information on levels of popular discontent (Kuran 1991; Lohmann 1994). The scarcity of reliable information on popular discontent has been identified as a basic problem of governance in autocracies (C. Friedrich and Brzezinski 1965; Kuran 1991; Wintrobe 1998). A fundamental obstacle to acquiring such information is preference falsification, which involves citizens publicly declaring support for the regime while privately being opposed to it (Kuran 1995; J. Jiang and Yang 2016; Wedeen 1999). The literature has argued that the ubiquity of preference falsification explains why coups and revolutions occur so suddenly (Kuran 1991). In this chapter, I acknowledge that system-destabilizing crises can emerge unexpectedly. However, I maintain that regimes that survive such crises can develop techniques that allow them to circumvent the problem of preference falsification and to collect information on popular preferences. In turn, such information can be used to anticipate future crises. Autocratic regimes create a range of bureaucracies that can be mobilized to compile the necessary information on the public mood. Therefore, autocrats have the capacity to authorize the collection of information that allows them to assess private preferences. Cognizant that these preferences cannot be observed directly but that they are correlated with levels of discontent (i.e., private opposition to the regime is likely to be highest among the most highly discontented citizens), authoritarian leaders foster channels that can supply information on popular discontent. In this chapter, I focus on the collection of information on private preferences in communist regimes, which are the longest-lasting type of nondemocratic regime to emerge since World War I.1 How was this information collected? One piece of literature that addresses this question is the new research on electoral autocracies, which has highlighted how competitive elections can provide information to the regime 21

22

Martin K. Dimitrov

about its level of mass support (Gandhi and Lust-Okar 2009; Magaloni 2006; M. K. Miller 2015). Ongoing research has identified protests as a second channel for transmitting information about levels of discontent to the leadership in autocracies, with a recent study arguing that the Chinese government encourages protests because of its inability to gather information on discontent through other channels (Lorentzen 2013). A third line of inquiry has focused on news content published in commercialized media in postTiananmen China as a valuable source of information about public discontent (Stockmann 2013). In this chapter I argue that, although they do reveal popular perceptions of governance problems in communist regimes,2 elections, protests, and commercialized media cannot function as the main avenues for collecting information because all of these channels have the potential to usher in systemic crisis rather than to help manage discontent. With regard to elections, the experience of Poland in 1989 demonstrates how competitive elections can precipitate regime collapse. Eastern Europe also illustrates how unconstrained mass protests can usher in regime instability and, eventually, regime collapse (Bunce 2003). Commercialized media also present a danger because, to the extent that the information about discontent is disseminated, it undermines stability by creating a coordination mechanism for discontented citizens (Kuran 1991). For example, the liberalization of Soviet media during perestroika hastened regime collapse by revealing to citizens the pervasiveness of discontent with the Communist Party. Communist regimes want to foster channels that allow for the regularized transmission of information on popular discontent without endangering regime stability. On the basis of archival materials and internal government documents from the German Democratic Republic (GDR) and from post-1949 China, I maintain in this chapter that there were two types of discontent in the leaders’ own understanding and each required a different monitoring strategy. One was hidden and the other overtly expressed. Overt discontent involved public acts of opposition such as the distribution of antiregime pamphlets, the creation of unauthorized civil society groups, and participation in unsanctioned protests and demonstrations. Hidden discontent (latent discontent) was limited to privately held dissatisfaction that typically stemmed from the regime’s inability to meet its redistributive commitments to the population. Leaders were aware that latent discontent could also arise from frustrated expectations for political concessions, but believed that those expectations could be kept at bay through a relaxation of repression and by reallocating budgetary spending from heavy industry to consumer goods, welfare, and services. In this chapter, I argue that information on the sources of latent discontent can reveal the private preferences of citizens under communism whereas information on overt discontent allows the regime to track

Anticipating Crises in Autocracies

23

publicly expressed preferences. Leaders thought that the two types of discontent were related and that the emergence of overt discontent represented a visible failure to identify and neutralize brewing latent discontent through redistributive concessions. Because overt discontent could threaten the stability of the system, leaders prioritized tracking and responding to latent discontent prior to its transformation into overt discontent. What channels were used to monitor discontent? On the basis of a systematic analysis of a large corpus of state security and Communist Party archival documents from pre-1989 Eastern Europe and from post-1949 China, I argue in this chapter that the tracking of overt discontent did not present a serious logistical challenge: because it was visible, state security political mood reports, situation reports issued by the Communist Party, and internal (not-for-publication) journalistic reports were all used to provide detailed assessments of its spread. In contrast, because it was unobservable, latent discontent presented a difficulty for the regime (Dimitrov 2015). To uncover private preferences, both involuntarily collected and voluntarily provided information was used. The involuntary collection of information was executed through the same channels that were employed to monitor overt discontent, though these channels were assigned new functions to identify incipient discontent and produce early warning about its possible escalation into overt discontent. Yet whereas these channels could supply accurate information on visible acts of opposition to the regime, the archival documents reveal that communist leaders were aware that their utility for identifying latent discontent was limited by the existence of preference falsification. Therefore, they valued highly information that was provided voluntarily (Dimitrov 2014a, 2014b). Transmitted in the form of citizen complaints, this information was used to identify the main types of grievances in society. However, incentivizing citizens to continuously provide such information required both a switch from massive terror to selective repression (because if citizens are afraid of retaliation, they will not complain) and sustained responsiveness (citizens will not complain unless a minimally acceptable probability of a positive resolution exists). This complex bargain tied the hands of communist leaders: if they wanted to continue to receive information from the public, they could not counteract discontent through brute force but instead had to maintain responsiveness through costly social spending. In this chapter, I advance three central claims. The first concerns the conditions under which autocracies begin to monitor latent discontent. The general answer lies in the inability of state security, which is the main institution tasked with tracking overt discontent, to forecast when a mass uprising might take place. In East Germany, for example, it was the failure of the Stasi, the internal security apparatus of the GDR, to pick up

24

Martin K. Dimitrov

the warning signs of the June 1953 uprising that led to the creation of the Zentrale Auswertungs- und Informationsgruppe (ZAIG; Central Evaluation and Information Group), which was charged with comprehensive monitoring of the public mood. In Bulgaria, the leadership was surprised when textile workers in Khaskovo and tobacco-processing workers in Plovdiv went on strike in May 1953 due to poor work conditions, inadequate compensation, and massive layoffs. 3 These domestic protests, as well as the June 1953 events in the GDR, helped to reorient Bulgarian state security from repression to surveillance.4 This resulted in the establishment of the Central Information-Organizational Directorate (TsIOU), a state security division responsible for compiling reports on the public mood for the central leadership. Similarly, an effort was made in China to establish comprehensive systems for the collection of information following the surprise eruption of discontent in the spring of 1989. In short, an unanticipated regime crisis can serve as a catalyst for the decision to start monitoring latent discontent. My second claim focuses on the element of surprise in 1989. I argue that when the channels for monitoring latent discontent are consolidated, as they were in Eastern Europe well before 1989, communist regimes have abundant information on growing dissatisfaction with their performance. Logically then, the transformation of latent discontent into overt discontent prior to the revolutions of 1989 was not a surprise (the exact timing of the revolution, of course, results from stochastic events and cannot be predicted). In contrast, when the channels are not fully established, the available information on latent discontent is limited. Thus, the underdeveloped state of the information-gathering systems in China prior to 1989 meant that the transformation of latent into overt discontent in 1989 came as a surprise. My third claim in this chapter is that force would not be employed to defuse protests in societies where the adoption of selective repression has made it possible for the systems for the voluntary provision of information to become fully developed. Compliance in such regimes is instrumental, not driven primarily by fear. Citizens comply for as long as the regime provides them with material rewards. Once these can no longer be supplied, citizens exit the system, engaging in protest. Crushing this type of protest is costly because levels of fear have declined when compared to the harshly repressive period of initial communist rule. In China in 1989 levels of fear were significantly higher than in Eastern Europe, which on the one hand meant that the channels for the voluntary provision of information were not yet fully operational, but also that leaders could use brute force when the surprise Tiananmen protests emerged. Thus, from the leaders’ perspective, the tragic brilliance of a fully fledged system for the voluntary collection of

Anticipating Crises in Autocracies

25

information is that it can weaken their ability to withstand regime crisis through the use of force. At a more general theoretical level, in this chapter I argue that the building of institutions tasked with the collection of information on popular discontent develops at a different speed in different single-party communist regimes. These institutions begin to be established as a result of unanticipated regime-destabilizing mass protests. I make no assumption that all communist regimes would experience such protests or that these protests would occur at a similar temporal juncture in the lifespan of each communist regime. In the GDR, for example, the relevant protests occurred in 1953. In China, they did not take place until 1989. Thus, the 1989 protests had a very different meaning and called for different responses in China and in the GDR. These synchronous events have to be examined from a diachronic perspective (Dimitrov 2016). A brief discussion of case selection, hypothesis testing, and scope conditions is necessary. My study offers a paired comparison of one communist regime that eventually collapsed (the GDR) and another that has survived to the current day (China). By choosing one case from each group of communist regimes—namely, the group of ten regimes that eventually collapsed in 1989−1991 and the group of five regimes that are still extant)—I demonstrate that learning from crisis is not unique to either group. In both groups, learning from crisis has helped extend the lifespan of the regimes. The paired comparison allows us to use one case (the GDR) as the theory-generating case and the other case (China) as a test of the theory. This test can only be provisional, as institutional evolution is ongoing in China. Finally, my argument is developed on the basis of information from single-party communist regimes, but it may apply to another type of authoritarian regime as well; namely, to single-party noncommunist regimes (Dimitrov and Sassoon 2014). The rest of this chapter is organized as follows. First, I focus on the GDR from 1953 to 1989. Then, I analyze the role of information in the lead-up to the Tiananmen Square massacre and Chinese learning from the unanticipated events of 1989. These two sections draw on East German and Chinese archival materials that allow us to see how crises led to learning with regard to the techniques that could be used to assess popular discontent in these two countries. In the final section, I conclude and discuss some of the broader theoretical implications of the argument. This chapter connects with some of the main concerns of the volume such as the emergence of crises (Chapter 1 by Johannes Gerschewski and Chapter 11 by Christoph H. Stefes); the emergence, stabilizing effects, and decay of institutions (Chapter 1 by Gerschewski and Chapter 4 by Milan W. Svolik); and the strategies that have allowed China to avoid crisis since 1989 (Chapter 8 by Thomas Heberer).

26

Martin K. Dimitrov

Crisis and the Monitoring of Latent Discontent in the GDR, 1953−1989 The case of the GDR allows us to generate the key hypotheses of this chapter: namely that crisis can serve as an impetus for developing the systems for monitoring latent discontent and that, because these systems were developed well before 1989, they should have provided ample warning to the regime about the probability of this discontent being transformed into acts of overt opposition to the regime. The 1953 Crisis

The German Democratic Republic experienced its first crisis less than four years after the establishment of the communist regime. About a million people participated in protests and strikes that engulfed as many as 500 cities and villages throughout fourteen of the fifteen regional districts (Bezirke) of the GDR in the summer of 1953 (Schroeder 2013). The uprising stemmed from discontent that mounted throughout 1952 as a result of forced collectivization, shortages of staples (butter, margarine, meat, vegetables, and sugar), and the difficulty of securing coal for winter heating (Engelmann 2013). The spark was the political opening created by the death of Joseph Stalin in March 1953 as well as the raising of prices for sugar products, and, especially, the increase in work norms coupled with a decrease in salaries. New laws raising the severity and frequency of criminal punishment also played a role (Engelmann 2013). Though Soviet tanks successfully quelled the uprising, the 1953 events demonstrated that the GDR regime was deeply unpopular and that the party could not base its long-term survival on repression alone. In the summer of 1953, the leadership assessed the reasons for the uprising and charted a new course of governance. One of the key findings of the leadership was that the uprising had come as a surprise (Mitter 1991; Ostermann 2001). The Stasi was blamed for this intelligence failure and Wilhelm Zaisser was relieved of his duties as minister of state security. How can we explain the inability of state security to anticipate this outcome? In the first four years after the establishment of the GDR, the emphasis of state security was on monitoring overt discontent. This type of governance is retrospective, focusing on neutralizing discontent that is already visibly expressed. The standard mechanism of rule is heavy repression. Because information gathering is crude and incomplete, surprises are likely. By contrast, the monitoring of latent discontent is aimed at detecting inchoate trends in popular opinion and making forecasts about the conditions under which these might be transformed

Anticipating Crises in Autocracies

27

into overt discontent. Sophisticated information-gathering systems are used to send early warning signals to the leadership, which guide the prophylactic use of targeted social spending and selective repression. Thus, one of the conclusions of the leadership was that it needed comprehensive information on the popular mood (Engelmann 2013; Ostermann 2001). The monitoring of latent discontent in the GDR began in the immediate aftermath of the June 1953 Berlin uprising with the establishment of the ZAIG of the Stasi, which was tasked with producing timely assessments of the public mood that could be used to prevent the rise of another unanticipated expression of popular discontent through the deployment of targeted redistribution and selective repression.5 In addition, the public mood was studied through party reports (Partei Informationen), internal journalistic not-for-publication reports, and surveys conducted by the Central Institute for Youth Research and the Opinion Research Institute of the Central Committee (W. Friedrich, Förster, and Starke 1999). These channels generated detailed information on the popular mood. However, because all of these channels focused on the involuntary collection of information, there was a non-negligible probability that the information was marred by preference falsification. For this reason, like their counterparts in other communist regimes, East German leaders valued highly the voluntary transmission of information through citizen complaints (Eingaben) (Mühlberg 2004; Staadt 1996). When aggregated, complaints information could reveal trends in latent discontent.6 The party took Eingaben so seriously that districts were encouraged to compete for handling the highest volume of complaints (Betts 2010). Although the regime valued complaints, incentivizing citizens to voluntarily provide information required a decline in repression (citizens will not complain if they are afraid) and a reasonable level of responsiveness to the grievances articulated in their complaints. Repression indeed became less frequent: between 1953 and 1989, the number of individuals prosecuted for political crimes by the Stasi declined 4.8-fold, despite a sevenfold increase in the size of the full-time Stasi officer corps during that period (Gieseke 2006). Expectations of responsiveness rested on a bargain that scholars have described as a socialist social contract (L. Cook 1993; L. Cook and Dimitrov 2017; Pravda 1981), which involved the mutual expectation of the citizens and the leadership that citizens would remain quiescent as long as the regime provided them with benefits including secure jobs, social services, subsidized housing, and consumer goods. Although the threat of repression never disappeared, the leadership aimed to ensure the quiescence of the population through a package of benefits and price subsidies. Instead of repression, the social contract formed the wellspring of regime stability in the GDR after 1953.

28

Martin K. Dimitrov

In the aftermath of the June 1953 uprising and especially after the building of the Berlin Wall in 1961, the East German population came to expect a growing set of entitlements and an expanding variety of consumer goods. The seriousness with which the leadership took these expectations is highlighted by the fact that the Stasi spent most of 1977, which was an otherwise “normal” year (Allinson 2009), trying to solve “the coffee shortage crisis” that had befallen the GDR (Bispinck 2012). The Stasi was also involved in securing the goods that were sold both in the Intershop/Genex chains (which operated with hard currency) and the Delikat/Exquisit shops, where one could use GDR Marks to buy hard-tofind domestic and imported goods (Judt 2013). Much like the rest of the party state apparatus, the state security system was tasked with assessing and satisfying the consumption preferences of the population. When they were dissatisfied, citizens sent in Eingaben. The issues about which East German residents complained bore a striking resemblance to those raised in complaints in Bulgaria and the Soviet Union, focusing on problems with the fulfillment of the social contract.7 Taken as a whole, the analysis of citizen complaints provided rich information on the nature of latent discontent in the GDR. This information was then used to ensure quiescence through the targeting of benefits and services and the provision of consumer goods. Moving Toward a New Crisis

Information gathering in the GDR after 1953 was aimed at identifying trends in discontent and forecasting how they might change over time. The archives produce a major discovery about the ability of the Stasi to do precisely that. The Bulgarian Committee for State Security wrote a brief on an exchange of views with the Stasi regarding the potential of the incipient political opposition to lead to future political destabilization in the GDR.8 According to the Stasi, most dangerous was the peace movement, which developed under the auspices of the evangelical church. In the future, the Stasi opined, the number of opposition groups was likely to increase, with new movements centered on environmental issues, human rights, and countercultural arts and music emerging. Subsequent Stasi reports from 1988 to 1989 present powerful evidence that this 1983 forecast was exceedingly accurate (Joestel 2010; Mitter and Wolle 1990). The inaction of the Stasi in light of the forecast attests to its growing inability to counteract discontent by the early 1980s. By the 1980s, economic difficulties made it increasingly difficult for the regime to respond to citizen preferences regarding social spending and consumer goods provision. The housing crisis, which had been a source of

Anticipating Crises in Autocracies

29

constant concern throughout the existence of the GDR, worsened. Although the GDR imported automobiles from other socialist countries and produced its own Trabant and Wartburg, there remained a vast unmet demand for cars. And despite efforts to keep supermarkets and department stores well stocked, shortages became more palpable in the 1980s. From the public’s perspective, the regime had reneged on its commitments under the socialist social contract by failing to deliver the anticipated volume of benefits and consumer goods. The tenor of complaints became angrier and citizens began to withdraw from the system of citizen petitions. Trends in citizen complaints provided distressing information to the leadership. Subnationally, complaints were declining in the second half of the 1980s, as they did elsewhere in Eastern Europe and China. 9 Complaints sent to the Central Committee and the State Council, however, increased in the second half of the 1980s.10 Some have interpreted this to mean that citizens lost trust in local governments, but trusted the center (Staadt 1996). A closer look reveals a peculiar feature of the East German categorization of complaints: in contrast to Bulgaria, requests for permanent exit from the GDR and for visits to West Germany, West Berlin, and nonsocialist countries were included in the total count of Eingaben. These exit requests started to increase at a double-digit rate in the early 1980s and by the second half of 1987 became the most frequent type of complaint to reach the Central Committee, accounting for 25.3 percent of the complaints; a year later, the share of these requests was 32.2 percent of all complaints.11 Germans were exiting the system by applying for permits to leave the GDR. The opening of the Berlin Wall, we now know, was an accident. Politburo member Günter Schabowski misspoke at a press conference on the evening of November 9, 1989, and misrepresented the scope of the travel liberalization decision made by the Central Committee earlier that day and the time frame for implementing it,12 claiming that it would enter into effect sofort, unverzüglich (immediately and without delay), 13 even though the changes were supposed to be announced in a press release the next day and to be implemented only afterward. 14 This mistake brought the crowds onto the streets and into West Berlin, effectively dismantling the communist regime three weeks after Erich Honecker’s ouster from the Politburo on October 18, 1989. Although of huge consequence, Schabowski’s lapsus was a stochastic event that could not have been foreseen. Yet the general rise of latent discontent was visible well before 1989. Why did the leadership remain inactive in the face of clear evidence about the erosion of support? One possibility is that this information was not transmitted to top leaders. Honecker is often caricatured as a hapless and uninformed individual.

30

Martin K. Dimitrov

This image is convenient because it allows us to see inaction in the final years and months of the GDR as reflecting a lack of information. However, the archival evidence indicates that Honecker was well informed: he received ZAIG reports (Joestel 2010; Mitter and Wolle 1990), reports on citizen complaints,15 and reports from grassroots party organizations and from the Worker-Peasant Inspection (Andert and Herzberg 1990). In addition, Honecker received information directly from Stasi head Erich Mielke during their weekly tête-à-tête meetings (Vieraugengespräche) after Politburo meetings on Tuesdays (Bispinck 2012). Thus, like his counterparts elsewhere in Eastern Europe, Honecker was kept apprised of trends in popular discontent. The leadership in countries with fully consolidated systems for tracking latent discontent knew about the problems that confronted them, but stagnating economic growth meant that the funds available for redistribution were getting more limited and the leadership was gradually becoming incapable of defusing citizen discontent through redistributive concessions. Using massive repression to stave off discontent stemming from frustrated expectations for political rights and freedoms was not an option either. Recent research has powerfully demonstrated that the received wisdom that the East German leadership was actively considering using a “Chinese solution” is not supported by documentary evidence (Schäfer 2012). It is also relevant that Mikhail Gorbachev made clear in early October 1989 that Soviet troops could not be relied on to put down unrest; however, as scholars have convincingly argued, GDR leaders knew at the latest by the summer of 1989 that they were losing control of the system (Münkel 2009). Like a paralyzed but fully lucid patient, the East German leadership was well informed of the impending systemic crisis occasioned by the rise in discontent, yet could not react to it. The East German case provides evidence that selective repression is not an effective tool for counteracting widespread overt discontent when levels of fear among the general population are low. Quelling nationwide overt discontent in a low-fear equilibrium (like the one that existed in the GDR in 1989) would have required a considerably greater use of harsh force than did responding to nationwide unrest in 1953, when levels of fear were considerably higher. Simply put, the East German regime could not repress in 1989. In light of the foregoing argument, we can make sense of the seemingly paradoxical decision of the leadership of the party, the Sozialistische Einheitspartei (SED), and the Stasi to avoid using force against the growing wave of discontent in 1988−1989 and to allow “the carnival of revolution” to engulf the GDR as well (Kenney 2003). This was a rational decision, given the available information and the type of governance that was practiced in East Germany at the time.

Anticipating Crises in Autocracies

31

China Before and After Tiananmen: Learning from the 1989 Crisis In this section, I examine how China learned from the unanticipated nationwide crisis of 1989. The Chinese case provides a test of the empirical implications of the theoretical argument that I develop in this chapter; namely, that high levels of repression prevent the establishment and consolidation of institutions for monitoring latent discontent. Moreover, China demonstrates that regimes can learn from unexpected crises and develop the institutions needed to assess discontent. The Unexpected Crisis of 1989

China, in contrast to the GDR, did not have institutions for the monitoring of latent discontent on the eve of the Tiananmen Square incident. Overt discontent was tracked mainly by state security, with the party and journalistic reporting playing secondary roles.16 Until the establishment of the Ministry of State Security (MSS; guojia anquan bu) in 1983 (Guo 2012; Zhongguo 1997), domestic counterintelligence was handled by the Political Protection Bureau (zhengzhi baowei ju) of the Ministry of Public Security (MPS) (DeVore 1999), which relied on both full-time employees and Soviet-style informants (ermu, or eyes and ears) to track overt discontent (Schoenhals 2013). The information collection system suffered various strains during the tumultuous 1950s and 1960s and faced extinction when Mao Tse-tung decided to abolish spying as an “OGPU revisionist line” and to publicly expose many domestic agents (guonei teqing) in 1967.17 Because it was established so late, the MSS had subnational representation only in the major cities and in key provinces in 1989; it was not until 1998 that the MSS had established offices in all Chinese provinces and provincial-level cities (Mattis 2011). This fragmentation of the information-gathering bureaucracy impacted the quality of the information that was collected and transmitted to the leadership. Analysis of a large corpus of archival materials (over 1,000 party, government, and journalistic reports that I collected at the Shanghai Municipal Archives, supplemented with 119 published reports from other provincial archives), 18 internal circulation public security publications (Renmin gong’an and Renmin gong’an zengkan), and internal circulation news bulletins (especially Neibu cankao, [Internal Reference]) reveal that information during the Mao period (1949−1976) was fragmentary. Facts reported were rarely contextualized with regard to how representative they were of general trends. Although such lack of analysis might be understandable for reports produced at and below the provincial level (where the

32

Martin K. Dimitrov

quality of the bureaucracy was lower), this was the norm at the national level as well. Let us take the example of the secret (jimi) bulletin aimed at Central Committee members entitled Neibu cankao (Internal Reference). The reports contained in this bulletin create the impression that leaders were rarely presented with comprehensive information about national conditions. Reports from the 1950s and early 1960s are geographically specific, discussing events in a single large city (Beijing, Shanghai, Tianjin, Chongqing, Chengdu, Wuhan, Nanjing, Guangzhou, or Nanning), part of a province (southern Jiangsu), a single province, or at the highest level of aggregation, a macroregion consisting of several provinces (the Southwestern Region, North China, or Northeast China). It is impossible to get a sense of national variation from these reports; in the GDR, by contrast, the Stasi aimed to provide leaders with mood assessments for each of the fifteen provinces (Bezirke) in the immediate aftermath of the June 1953 worker uprising.19 Perhaps most distressing is that even party reporting was fragmentary: the most comprehensive report about hunger during the Great Leap Forward that has become available thus far covers only sixteen provinces (Zhou 2012), about three-fifths of China’s twenty-eight provinces, provincial-level municipalities, and ethnic autonomous regions at the time. This might explain why, when confronted with evidence of hunger in some provinces, Chairman Mao reportedly interjected that the situation was better in other provinces (Dikötter 2010). Given the fragmented nature of information-gathering institutions in China, it is doubtful that even Mao had complete information on the spread of overt discontent (Garnaut 2013). After the death of Mao, information gathering slowly changed. Party and journalistic reporting was gradually being reoriented toward the monitoring of latent discontent. One early example is provided by a November 1976 investigation report commissioned by the Shanghai Municipal Party Committee on citizen reactions to a proposed change in the system for pork rationing (pork prices are so vital to the Chinese economy that they are still monitored by the government today and used as a bellwether of food price inflation).20 A second area that experienced significant change was the content of the secret bulletins prepared for the Chinese leadership. Although coverage remained region specific (rather than presenting comprehensive nationwide information), by 1980 both Qingkuang huibian and Neibu cankao had started to print longer, more in-depth reports that aimed to focus on important issues of broad significance. In contrast to earlier styles of reporting in such bulletins, in the post-Mao period efforts were made to contextualize the specific incident that was described and to showcase how it might be representative of general trends; coverage of official corruption and of retaliation against petitioners is exemplary of this new tendency.21

Anticipating Crises in Autocracies

33

Finally, the creation of the MSS in 1983 and the introduction of opinion polling in the mid-1980s further extended the ability of the state to eventually make a successful transition toward the involuntary extraction of information on latent discontent (Manion 1993). The one area where progress was a lot more uncertain was the voluntary provision of information through citizen complaints. Vice-Premier Xi Zhongxun opined in 1961, “Is it better to have more letters and visits or fewer? Of course it is better to have more. This means that the masses support us, love us, and trust us.”22 Yet building that trust was not a simple task. One problem was the relatively low use of the system: in 1979, the peak year for complaints during the entire 1949−1989 period, the central complaints offices in China received 769 complaints per million people, which was 5.8 times lower than central-level complaints in East Germany (4,464 complaints per million people) and in Bulgaria (4,494 complaints per million people).23 Even more troubling, the complaints volume decreased steadily throughout the 1980s: in Gansu, there was a 5.6-fold decline between 1979 and 1989; in Tianjin, a 3.3-fold decline; also nationwide complaints declined from 13.4 million in 1979 to 800,000 in 1989.24 The initial reason for this decline was that the rectification of the Cultural Revolution cases involving injustices and false accusations (pingfan yuanjia cuo’an), which fueled the high tide of complaints in 1979−1981, was officially declared to be complete by 1982 and such complaints were no longer accepted and their authors were not protected against retaliation. The intensification of repression (three rounds of the strike-hard anticrime campaign took place in 1981, 1984, and 1986) further diminished citizen trust in the regime. The continued decline in the mid- to late 1980s also reflected widespread discontent among China’s urban residents, who were distressed that the 1986 Law on Enterprise Bankruptcy and the 1988 Law on Industrial Enterprises Owned by the Whole People (which stipulated that stateowned enterprises could be abolished through dissolution or bankruptcy) signaled an end to the system of lifetime employment and generous benefits known as the iron rice bowl (tiefanwan). In addition, workers worried because urban inflation rose to double-digit rates. In Beijing, consumer prices grew by 30 percent in 1987−1988, leading to panic among those with fixed salaries (Vogel 2011). Feeling that the regime had reneged on the socialist social contract, urbanites signaled their discontent by complaining less often. This drop in complaints in the lead-up to Tiananmen represents an important parallel with the decline in complaints in Eastern Europe prior to 1989. There is a key difference, however: although we have evidence that the reports on trends in complaints were made available to and closely studied by the leadership in Eastern Europe, no such evidence exists for

34

Martin K. Dimitrov

China in the archival materials that I collected, in the Tiananmen Papers, or in the recently published biography of Deng Xiaoping and the diaries of Zhao Ziyang and Li Peng (C. Li 2010; Vogel 2011; Zhao 2009). Complaints data on the pre-1989 period reported in the previous paragraph are extracted from internal circulation (neibu) materials that were published in the mid- to late 1990s, not contemporaneously in the 1980s. The one contemporaneous source that contained information on the decline in complaints is Renmin Xinfang but it was an interdepartmental serial for complaints office personnel, not a government document aimed at the central leadership. Importantly, according to the argument of this chapter, leaders would have ignored these statistics even if they had been passed along to them because they were not yet sensitized to the importance of monitoring latent discontent through the systematic analysis of complaints. The surprise of 1989 focused their attention on anticipating the rise of overt discontent by monitoring trends in latent discontent. This explains the sustained and extensive leadership involvement with the complaints system immediately after Tiananmen.25 My most important argument in this section is that due to high levels of repression, the institutions for monitoring latent discontent were still in a nascent state on the eve of Tiananmen. Some might question this claim. It is of course true that China was a lot less repressive in the 1980s than, for instance, in 1950: according to statistics published in a neibu compendium, no fewer than 1.8 million bandits (tufei) were exterminated (jiaomie) during the first eleven months of that year (Jianguo 2003). However, China was still practicing capital punishment against those accused of counterrevolutionary crimes in the 1980s and rates of execution had been increasing since the death of Mao (Scobell 1990). Contrast this with the GDR, which carried out its last execution in 1981 (curiously, of Werner Teske, a Stasi officer who attempted to defect) and where a grand total of 166 judicial executions took place in the four decades of communist rule (1949−1989).26 In China, the shelter and investigation, reeducation through labor (laojiao), and the gulag-like reform through labor (laogai) systems were all operating at a high volume throughout the 1980s. Though repression had declined by comparison with the 1950s or 1960s, this decline had not been sustained for a long enough period to translate into a diminution of fear among the population comparable to that in East European countries like the GDR. This explains why the Chinese regime was able to contemplate defusing the surprise of Tiananmen through the use of force. We should note that many factors explain Chinese success in using repression to avoid collapse. It is, of course, relevant that proponents of political reform were a minority within the Chinese leadership and had not managed to co-opt the armed forces to their side (Zhao 2001). This meant

Anticipating Crises in Autocracies

35

that the army would abide by the orders of the leadership to disperse protesters through the use of force. Another important consideration was that the MSS believed that, despite turmoil in urban areas, the actual level of support for the government outside major cities with universities was very high (Zhao 2001). This meant that by using force the leadership would lose legitimacy only with urban residents, who at the time constituted no more than 20 percent of the population. However, one key difference from Eastern Europe is that China had not yet entered the lowrepression equilibrium that is a necessary precondition for having a fully consolidated system for the collection of information on latent discontent. This allowed China to react to the surprise protests of 1989 in the same way that regimes in Eastern Europe reacted to protests in 1953, 1956, and 1968—through brute suppression. Learning from Tiananmen

In the immediate aftermath of Tiananmen and preceding the collapse of communism in Europe by several months, the leadership swiftly turned its attention to petitions. Between June and September 1989, Shanghai mayor Zhu Rongji, General Secretary Jiang Zemin, Politburo Standing Committee member Qiao Shi, and Premier Li Peng each made separate statements about the importance of providing prompt and detailed responses to citizen complaints.27 In August 1989 the top leadership convened a meeting in Beijing of ten directors of provincial-level letters and visits bureaus,28 and in September 1989 the Central Bureau of Letters and Visits sent out a notice to all provincial-level party and government offices regarding the need to strengthen letters and visits work.29 The petitions system had never before received as much sustained attention from the top leadership as it did in the summer of 1989. This attention is not surprising, considering that the leadership wanted to regain the trust of the masses through increased responsiveness to petitions. The leadership also thought that greater responsiveness would maintain stability by preventing the escalation of individual complaints into group petitions or into visits to Beijing. A series of measures implemented throughout the 1990s, including vertical expansion of the network of letters and visits offices to the township level and eventually to the villages,30 helped reverse the trend, as illustrated by Figure 2.1. Then, in 2004, there was another drop in the volume of citizen complaints. Documenting the precise magnitude of this decline is difficult and time series statistics do not exist in the scholarly literature in either Chinese or English. However, relevant statistics on the volume of subnational petitions can be culled from the annual reports of the Central Bureau of Letters and Visits and from other internal government publications. These documents

36

Martin K. Dimitrov

Figure 2.1 Citizen Complaints in China, 1990–2005 (&$!!!$!!!" (%$!!!$!!!" (#$!!!$!!!" (!$!!!$!!!" '$!!!$!!!" &$!!!$!!!" %$!!!$!!!" #$!!!$!!!" !" ()''" ())!" ())#" ())%" ())&" ())'" #!!!" #!!#" #!!%" #!!&"

Sources: Renmin Xinfang. 1993–2003; annual work reports of central government departments, 2005–2006.

allow us to shed light on how the volume of petitions fluctuated through 2009 (see Figure 2.2). What is especially worrisome about the decline registered in Figure 2.2 is that it has occurred in conjunction with two other developments: a rise in the number of petitions that were lodged at the central level in Beijing and a dramatic increase in the number of mass incidents, most of which took place at the subnational level (on mass incidents, see Figure 2.3). Both petitioning Beijing and participating in mass incidents indicate that citizens do not believe that their grievances can be resolved through the formal petitioning channels available at the local level. Since 2004, the central government has implemented various measures that make it easier for citizens to petition online. Yet even though it is now simpler than ever to complain just by sending an e-mail or filling out an online form, citizens have continued to exit the petitions system and to use other channels such as online chatrooms, blogs, and microblogs (weibo). Exiting the petitions system and turning to online public opinion platforms is a source of concern for the regime. These channels transmit information as well, of course, but they do so differently from petitioning. Simply put, petitioning is an act of individualized trust in the system; participating in public fora is a visible act of frustration that can threaten the paramount imperative of stability maintenance (weiwen). In the aftermath of the unexpected crisis of 1989, China has been developing its systems for tracking latent discontent and for anticipating crises. Various channels for assessing overt discontent also exist such as the

37 Figure 2.2 Citizen Complaints in China, 2004–2009

(&$!!!$!!!" (%$!!!$!!!" (#$!!!$!!!" (!$!!!$!!!" '$!!!$!!!" &$!!!$!!!" %$!!!$!!!" #$!!!$!!!" !" !""#$

!""%$

!""&$

!""'$

!""($

!"")$

!""*$

!"+"$

Sources: Annual work reports of central government departments, 2005–2008; Renmin Xinfang, February 2010.

Figure 2.3 Mass Incidents in China, 1993–2010 #!!ǡ!!!" #!!ǡ!!!" '&!ǡ!!!" '&!ǡ!!!" '%!ǡ!!!" '%!ǡ!!!" '$!ǡ!!!" '$!ǡ!!!" '#!ǡ!!!" '#!ǡ!!!" '!!ǡ!!!" '!!ǡ!!!" &!ǡ!!!" &!ǡ!!!" %!ǡ!!!" %!ǡ!!!" $!ǡ!!!" $!ǡ!!!" #!ǡ!!!" #!ǡ!!!" !" !""#$ !""%$ !""&$ !""'$ #((($ #((#$ #((%$ #((&$ #(('$ #(!($ #(!#$

Sources: Andrew Wedeman. 2009. “Enemies of the State: Mass Incidents and Subversion.” Paper presented at the meeting of the American Political Science Association, Toronto, September 2009; Sun Liping. 2011. “Shehui shixu shi dangxia de yanjun tiaozhan.” Jingji Guancha Bao, February 25.

38

Martin K. Dimitrov

monitoring of all telecommunication platforms (mail, phone, e-mail, social media), opinion polling, and internal journalistic reporting. China today is where Eastern Europe was in the early 1980s: it has the full complement of systems that are needed to forecast a future crisis. However, another much less encouraging parallel to Eastern Europe is that, because it has now switched to a low-repression low-fear equilibrium, the Chinese regime will not be able to resolve the next regime-destabilizing crisis through the use of force. The multiple signs of an impending crisis (chief among them citizens exiting the formal complaints system) suggest that the future of the Chinese regime is uncertain.

Conclusion In this chapter, I offered a theory of the channels through which communist regimes collect information on overt and latent discontent. I argued that information on latent discontent is especially valuable because it allows communist autocracies to minimize the likelihood of the rise of overt discontent through the use of redistributive concessions and selective repression. I documented the mounting inability of East European regimes to handle latent discontent in the 1980s. Such evidence calls into question the extent to which the emergence of overt discontent in 1989 was a surprise for the European communist regimes. I also argued that the leadership made an informed decision that the use of force was not a rational option for counteracting discontent. In contrast, China had underdeveloped systems for monitoring latent discontent in 1989, which meant that the Tiananmen protests were a surprise for the leadership. However, as levels of fear were still high, the protests could be counteracted with brute force. My argument is consistent with recent studies that have emphasized that politics take place in time and that we should be mindful of the tempo of political processes (Grzymala-Busse 2011; Pierson 2004). The tempo of institutional development across two countries might vary significantly. For example, although China and the German Democratic Republic were established as communist states only six days apart in 1949, their institutions for collecting information on popular discontent developed at a very different speed. These differences mandate that we theorize the synchronous events of 1989 from a diachronic perspective, paying close attention to the sequencing of the processes of institutional development. On a broader theoretical level, in the chapter I demonstrated that communist regimes can learn from unexpected nationwide crises and can gradually build the institutions that are needed to assess latent discontent. The

Anticipating Crises in Autocracies

39

stability of the system is predicated not simply on collecting such information, but on being able to use redistributive concessions to prevent the rise of a new crisis. When responsiveness declines, communist regimes will be able to anticipate but not to avert a regime-destabilizing crisis. Archival Sources Bulgarian

AMVR f. 1 op. 1 a. e. 2811 (1954). AMVR f. 1 op. 1 a. e. 2922 (1953). AMVR f. 1 op. 12 a. e. 501 (1983). TsDA f. 1 op. 24 a. e. 125 (1953). TsDA f. 1 op. 24 a. e. 188 (1953−1954). TsDA f. 1B op. 64 a. e. 185 (1953).

Chinese (Archival and Neibu Materials)

Gansu xinfang zhi, 1949−1989. 1991. Lanzhou: Gansu Minzhu Chubanshe. Jianguo yilai gong’an gongzuo dashi yaolan. 2003. Beijing: Qunzhong Chubanshe. Neibu cankao, 1949−1964 and No. 73/1981. Qingkuang huibian, 1980. SMA B92-972-1 (1965). SMA B92-2-918-19 (1964). SMA B92-2-972-1 (1965). SMA B92-2-972-4 (1965). SMA B92-2-972-12 (1965). SMA B92-2-1560 (1972). SMA B246-2-940-40 (1973). SMA B246-2-944-83 (1973). SMA B246-3-134-62 (1976). SMA B248-2-924 (1976). SMA B250-2-329-37 (1971). SMA B250-2-1024-35 (1977). SMB B326-2-35-201 (1975). Renmin gong’an. 1958−1963. Renmin gong’an zengkan. 1951. Renmin xinfang. 1985−1990. Tianjin tongzhi xinfang zhi. 1997. Tianjin: Tianjin Shehui Kexueyuan. Zhongguo renmin gong’an shigao. 1997. Beijing: Jingguan Jiaoyu Chubanshe.

East German

BStU MfS SdM 249 Bl. 102-107 (Meldung Nr. 7/53) SAPMO DC 20/16563 (1981). SAPMO DC 20/16634 (1989). SAPMO DC 20/16635 (1989).

40

Martin K. Dimitrov

SAPMO SAPMO SAPMO SAPMO

DY30/913 (1981). DY30 6265 (1988). DY30 6266 (1989). DY 30/IV 2/1/529 (1976).

Soviet

Rubinov Collection, Library of Congress TsKhSD F. 89 per. 6 d. 11, 12, 13, 14, 15, 16, 17, 18, 19, 23, 24, and 25.

Notes 1. Communist regimes are the most durable type of nondemocratic regime, outlasting both noncommunist single-party regimes and nondemocratic monarchies. As of 2000, the average lifespan of noncommunist single-party regimes was 28.51 years and that of nondemocratic monarchies was 34.75 years. In contrast, communist single-party regimes had an average lifespan of 46.2 years as of 2000. My dataset includes thirty-nine noncommunist single-party regimes (based partially on Smith 2005), twenty nondemocratic monarchies, and fifteen communist regimes. As of 2017, the five remaining communist regimes have an average lifespan of 60.0 years. 2. On information extracted through the analysis of elections, see Zaslavsky and Brym (1978); on protests, see TsKhSD F. 89 per. 6 d. 11-19 and d. 23-25; on journalistic monitoring, see the Rubinov Collection. 3. AMVR f. 1 op. 1 a. e. 2922 (1953), p. 5; AMVR f. 1 op. 1 a. e. 2811 (1954). 4. TsDA f. 1 op. 24 a. e. 125 (1953); TsDA f. 1 op. 24 a. e. 188 (1954−1955); TsDA f. 1b op. 64 a. e. 185 (1953). 5. See Stasi documents from 1953, 1961, 1976, 1977, 1988, and 1989 in the BStU series Die DDR im Blick der Stasi; see also Mutter and Wolle (1990). 6. SAPMO DY30/913; SAPMO DC 20/16563; SAPMO DC 20/16634; SAPMO DC 20/16635. 7. See SAPMO DY30/913 (1981); SAPMO DC 20/16563 (1981); SAPMO DC 20/16634 (1989); SAPMO DC 20/16635 (1989). On Bulgaria, see Dimitrov (2014a). On the Soviet Union, see Dimitrov (2014b). 8. AMVR f. 1 op. 12 a. e. 501 (1983), p. 178. 9. No comprehensive subnational data exist. For data on Karl-Marx-Stadt, see Mühlberg (2004, p. 176). 10. SAPMO DY30 6265 (1988); SAPMO DY30 6266 (1989). 11. SAPMO DY30 6265 (1988), p. 26; SAPMO DY30 6266 (1989), p. 3. 12. See relevant part of the Central Committee meeting transcript in Hertle and Stephan (2012, pp. 303−306). 13. “Pressekonfenrenz über die neue Reiseregelung,” November 9, 1989. 14. On this see also Sarotte (2014). 15. SAPMO DY 30/IV 2/1/529 (1976). 16. For examples of party and journalistic reporting, see Shanghai Municipal Archive (SMA) B92-972-1, B92-2-918-19, B92-2-972, B92-2-972-4, B92-972-12, B92-2-1560, B250-2-1024-35, B326-2-35-201, B246-2-944-83, B250-2-329-37, B246-2-940-40. 17. Schoenhals (2011, esp. pp. 1368−1369). Domestic spying resumed in 1973.

Anticipating Crises in Autocracies

41

18. These 119 documents are collected in Zhou (2012). 19. BStU MfS SdM 249 Bl. 102-107 (Meldung Nr. 7/53). 20. SMA B248-2-924 (1976). 21. On corruption, see Neibu cankao (No. 73/1981), esp. pp. 10-14 and Qingkuang huibian (No. 549/1980). On retaliation, see Qingkuang huibian (No. 365/1980). 22. Vice-Premier Xi Zhongxun (as quoted in Diao 1996, p. 158). 23. Calculations based on data in Diao (1996, p. 260); Mühlberg (2004, p. 177); TsDA f. 1B op. 55 a. e. 940 (1983), p. 5. 24. Calculations based on data in Gansu 1991, p. 49; Tianjin 1997, p. 188; Diao (1996, pp. 260−261, missing provincial data for 1979−1981 imputed by the author); and Renmin Xinfang, 1985−1990. 25. See reports on the involvement of Jiang Zemin, Zhu Rongji, Li Peng, and Qiao Shi in Renmin Xinfang, No. 10/1989, pp. 2−8; Renmin Xinfang, No. 11/1989, pp. 2−4; Renmin Xinfang, No. 2/1990, pp. 2−3. 26. Only some of these executions were for political crimes. See Marxen and Werle (2000). 27. Renmin Xinfang, No. 10/1989, p. 2; Renmin Xinfang, No. 11/1989, pp. 2−4. 28. Renmin Xinfang, No. 10/1989, pp. 3−8. 29. Renmin Xinfang, No. 2/1990, pp. 2−3. 30. Renmin Xinfang, No. 2/1991, pp. 17−19; Renmin Xinfang, No. 5/1995, p. 19.

3 Disturbing the Dictator: Peaceful Protest Under Authoritarianism Andreas Schedler

What is the role of peaceful mass protest in the generation of authoritarian regime crises? Comparative scholars tend to agree that the “predominant political conflict in dictatorships” does not unfold between rulers and masses, but “among regime insiders” (Svolik 2012, p. 5, emphasis removed). The primary threats to the political survival of authoritarian rulers are horizontal; they arise from within the ruling coalition. Vertical challenges from below, by ordinary citizens, rarely succeed in toppling dictators.1 Students of comparative authoritarianism also tend to agree on one basic fact that accounts at least in part for the relative irrelevance of popular protest in the downfall of nondemocratic regimes: the exceptional nature of antiauthoritarian mass challenges. While the horizontal “competition among rival factions” (Geddes 1999, p. 121) is endemic under dictatorship, vertical challenges to authoritarian rule are supposed to be rare occurrences. Authoritarian regimes strive to either preempt or repress them and are usually successful in doing so. In equilibrium, they generate popular quiescence. Due to repression, contentment, uncertainty about the preferences of others, or problems of collective action, most of the time nearly all citizens comply with the behavioral demands of the regime (see, e.g., Chen 2012; Kurzman 2004a; Stein 2007). Inversely, citizen protest is supposed to flourish under democracy. Subdued under dictatorship, “contention greatly increases with democratization” (Tarrow and Tilly 2009, p. 448). The “same processes that promote democratization also promote social movements [and] democratization itself also promotes social movements directly” (Tarrow and Tilly 2009, p. 449). The literature disagrees, though, on the causal relevance of peaceful protests once they do occur under dictatorship. Some scholars emphasize their disruptive potential: when popular protests pierce the surface of authoritarian tranquility, they tend to escalate and push regimes into crises of survival (see, e.g., Kuran 1995; Lohmann 1994). “By their very nature, [such] disruptive acts pose a challenge to autocratic regimes” (Ulfelder 43

44

Andreas Schedler

2005, p. 312). Others, by contrast, emphasize their reactive nature: mass protests tend to irrupt in response to emergent troubles of the political regime. In and by themselves, they are unlikely to cause major troubles (see, e.g., O’Donnell and Schmitter 1986, p. 48; Tarrow 1991). However, as an accumulating body of case evidence suggests, the assumption that popular protests are rare under dictatorship as well as the related debate about their disruptive or reactive nature appear to be misleading. In many autocracies and most prominently in contemporary China, citizens seem to be willing and capable of generating contentious challenges on a regular basis; and authoritarian rulers seem to be willing and capable of normalizing and absorbing these challenges into their operating routines. Rather than rare disruptive events that provoke existential crises, or reactive moves that aggravate such crises, recurring popular protests seem to form an integral part of political normality in many authoritarian regimes. Country experts tend to read regular protests in autocracies such as China in functionalist terms: rather than putting authoritarian regimes into question, they contribute to their smooth operation. Acts of public protest under dictatorship are conventionally seen as acts of antiregime challenge, as instances of popular contention, confrontation, defiance, resistance. Contemporary students of “contentious authoritarianism” (Chen 2012) by contrast, interpret them as acts of tacit collaboration between protesters and authorities that do not challenge the authoritarian status quo, but actually serve to perpetuate it by rendering it more legitimate and more efficient. In quantitative terms of protest frequency, I suspect that the accumulating case evidence of frequent protest under authoritarianism may hold beyond a few prominent countries. (Tertium datur between quiescent equilibria and contentious crises of dictatorship.) The intermediate possibility of recurrent ordinary protest may be more widespread than our theories suggest. To date, the paucity (and poverty) of extant cross-national datasets on peaceful political contention have prevented us from evaluating empirical claims about the relative infrequency of protest events in authoritarian regimes. After first outlining some broad theoretical propositions on the dynamics of antiregime protest, the meaning of regime crises, and the logic of authoritarian repression, in this chapter I make use of newly available event data from the Social Conflict in Analysis Database (SCAD) to assess the incidence of protest demonstrations in democratic and nondemocratic regimes. The dataset covers Africa, Mexico, Central America, and the Caribbean from 1990 through 2012. In qualitative terms of the nature of protest, I suspect that extant theoretical perspectives misconstrue both the causal force and the motivational bases of citizen protest against dictatorship. As I hypothesize, the modal contentious action against authoritarian regimes is neither disruptive, nor

Disturbing the Dictator

45

reactive, nor submissive. Neither does it bring down the regime (or accelerate its downfall in the wake of crises), nor does it serve its smooth functioning. If antiauthoritarian citizens are competent observers of their political environment, they should know it is hard to topple a dictatorship through public protest. If they are committed to the cause of political freedom, they should be reluctant to drop their quest for regime change in favor of minor, more acceptable policy demands. I therefore expect them to be neither quiescent, nor revolutionary, nor opportunistic—but disturbing. Even if citizen protest fails to bring down the regime or to purchase particularistic concessions from the regime, it is not futile. It serves to communicate dissent and herein to irritate the official story of authoritarian legitimacy. Its main role is informational. Protest demonstrations do what they are supposed to do: they demonstrate. They publicize popular discontentment and show that the surface of citizen quiescence is deceptive. By tearing the veil of generalized “preference falsification” (Kuran 1995), they may not ignite revolution. Yet they do not play into the hands of the dictator either. Neither frightening nor submissive, they still unsettle the authoritarian standard script of social harmony and popular gratitude. They disturb the dramaturgical self-complacency of the regime. The SCAD dataset also allows to probe into the nature of peaceful contention under dictatorship. Do protesters generally try to look harmless, rather than threatening, by sticking to nonpolitical issues, avoiding challenging political authorities, keeping their numbers low, organizing fleeting events rather than sustained campaigns, and staying away from the capital city? The SCAD data broadly support the empirical plausibility of my casebased intuitions about the “disturbing normality” of contentiousness under authoritarianism. As they suggest, peaceful, yet still provocative protests are remarkably common under dictatorship. More often than not, they appear to constitute acts of principled opposition rather than calculated acts of “protest opportunism” (Chen 2012, line 3040). Mapping the Terrain: Theoretical Considerations The motivating question of this chapter is X-centered. Rather than comprehensively tracing the causes of authoritarian regime crises (the dependent variable Y), I selectively focus on the effects that popular protests (my independent variable X) bear on such crises.2 State responses to protest (lethal repression) serve as my measure of governmental crisis perceptions. None of these variables are self-explanatory, nor are their relationships. Let us thus briefly review their logic one by one.

46

Andreas Schedler

Disturbing Contention: The Logic of Protest

Under dictatorship, the absence of antiregime protest does not indicate the presence of regime support. To the degree that governmental repression induces citizens to falsify their preferences (Kuran 1995), we cannot know (for certain) how happy or unhappy they are about the political status quo. The legitimacy of authoritarian regimes is one of the things that we know we cannot know. Yet while the nonoccurrence of protest does not allow us to draw any inferences on popular preferences, what does the occurrence of protest allow us to infer? For citizens to engage in collective protest, they need to overcome daunting obstacles to coordinate (Olson 1965; Schelling 1960). The literature on contentious action fills libraries. One of its core insights, though, can be summarized in a rather simple manner. Individual decisions to participate in public protest are a function of three rational emotions: anger, fear, and hope. When individuals subject to authoritarian rule decide whether to participate in public protest, they cannot simply act on their grievances (their anger in the face of injustice). They also need to ponder the risks that governmental responses pose to their physical integrity (the probability of repression) as well as the chances of inducing genuine political changes through collective action (the probability of collective success). When we see public protest erupting against authoritarian rule, we observe the triumph of anger and hope over fear.3 Since authoritarian regimes specialize in generating fear and depleting hope, popular discontentment often remains hidden under the surface of public compliance. “Independent collective action is unusual in closed autocracies” (Trejo 2012, line 1110). “In the most repressive regimes resistance is largely limited to ‘the weapons of the weak’” (O’Brien 1996, p. 47).4 Furthermore, even when individuals are neither satisfied with the regime, nor fearful, nor hopeless, they do not know to what extent their fellow subjects share their preferences and beliefs. A rich game theoretic literature analyzes the problems of protest coordination under conditions of imperfect information. At its core are the dynamics of protest escalation under uncertainty. Game theory strives to explain the possibility of transitions from one equilibrium to another: from the ordinary equilibrium of popular quiescence in which no one protests except a hard core of radicals to an extraordinary equilibrium of mass protest in which large numbers of discontented citizens flood the streets. Often, such “waves” of “democratic contention” (Weyland 2014) are induced by external shocks such as economic crises, elite divisions, or regime change in neighboring countries.5 Overall, the dominant theoretical expectation about citizen behavior under dictatorship is compliance. Fear, resignation, uncertainty, and perhaps

Disturbing the Dictator

47

even a measure of contentment should lead the subjects of authoritarian rule to remain quiescent and bow to the behavioral demands of the regime. In many repressive regimes, this expectation of citizen quiescence is fully borne out. We have no reports, for instance, about antiregime demonstrations in totalitarian North Korea. However, we do have countless reports about acts of protest and resistance in many nondemocratic regimes. In contemporary China, for example, “the number of protests and other ‘mass incidents’ increased from approximately 10,000 in 1994 to over 80,000 in 2008 by the regime’s own count and a rumored 180,000 in 2010” (Lorentzen 2013, p. 128). In its last quarter of a century (1974–2000), electoral authoritarian Mexico registered over 3,500 indigenous protest events, most of them “tiny acts of micro mobilization that involved approximately 200 rural indigenous villagers . . . and that usually went unnoticed by the international press” (Trejo 2012, line 553). Each year from 1997 through 2000, Russia’s competitive electoral autocracy lost millions of working days to labor strikes (Robertson 2010). Even in hegemonic party regimes that look invincible to their supporters as well as their opponents, opposition actors stage public protests against electoral processes and outcomes in well over a third of all national elections (Schedler 2013). While the eruption of mass protest during the so-called Arab Spring of 2011 took analysts as well as actors themselves by surprise, it did not happen in a vacuum of popular passivity, but rather against the background of strong “traditions of militant mass action” (Clancy-Smith 2013, line 442). “The dramatic and rapid upsurge of popular mobilization did not come out from nowhere . . . for most of the decade of the 2000s, popular protest had increasingly characterized Arab politics” (Lynch 2014, p. 264). These disperse pieces of (mostly) case evidence suggest that various types of contentious action happen in various kinds of authoritarian regimes on a regular basis. Rather than constituting extraordinary events that mark acute states of legitimacy crisis (see Chapter 1 by Johannes Gerschewski), social and political protest seems to form part of normal political life in many, if not most, nondemocratic regimes. Given the fact that in contemporary China high rates of social protest and high levels of political stability seem to coexist in a rather harmonious manner, scholars of Chinese politics have been developing functional interpretations of popular protest. In general, these protests appear to be carefully self-contained in their goals, targets, and framing. They refrain from criticizing the political regime and its ideological, organizational, and personal pillars. Appealing to official goals and national law and professing respect to central authorities, they articulate local grievances against local figures within the language of “rightful resistance” (O’Brien 1996). As they accept the political red lines drawn by Beijing, even when they test and

48

Andreas Schedler

expand them (Chen 2012), these acts of self-restrained resistance are seen to fulfill multiple functions for the rulers in the capital city. In addition to serving as early warning systems about popular discontent, they work as mechanisms of political accountability toward local officials. By providing information on local wrongdoing and by activating political as well as legal mechanisms of redress, social protests help central authorities to solve the huge principal-agent problems they face in their relations with subnational authorities (see, e.g., Lorentzen 2013). In his analysis of labor strikes in electoral authoritarian Russia, Graeme B. Robertson (2010) also undermines the notion that civilian protests should be seen as vehicles of anti-regime resistance. His account differs, though, from the Chinese story of popular protest as a mechanism through which national elites constrain their subnational agents. As he argues, instead of subjecting local elites to some minimal form of societal control, orchestrated social protests in Russia serve as instruments that local elites use to capture the attention (and resources) of national officials. Although I find such accounts of either self-restrained or heteronomous contention entirely plausible, they may not be capturing the common spirit of normal citizen protest under dictatorship. Even when citizens know that it is hard to trigger spirals of revolutionary protest, rational resignation may not be their preferred course of action. Contentious action is instrumental. It aims at achieving tangible political goals such as policy changes or regime change. Yet contentious action is also communicative. It aims at communicating discontent with existing policies and regimes. By participating in protest demonstrations, citizens do what the term suggests they do: demonstrate the breadth and intensity of discontentment to rulers as well as to their fellow citizens (see, e.g., Tilly 2004). More often than not, though, the instrumental goals of contentious action are distant and elusive. Protest participants can optimistically hope to achieve them, but they know that all challenges to existing power structures have low odds of success. Yet they still can realistically expect to accomplish the communicative function of protest: to inform authorities as well as the general public about the existence of popular discontentment. In authoritarian regimes, the communicative role of contentious action opens up an intermediate road between disruptive protest on the one hand and quiescence or submissive protest on the other. If citizens know that the chances of revolutionary mass protests are low, they may stay passive or limit their protest activities to regime-licensed grievances, targets, and repertoires (rightful resistance). Yet even if their ultimate instrumental goal of regime change seems out of reach, they still may decide to pursue more modest communicative goals. In particular, by demonstrating the existence of antiregime sentiment among the population, they can hope perhaps not

Disturbing the Dictator

49

to destroy, but at least to irritate the official narrative of legitimate governance that authoritarian regimes commonly construct. When such communicative antiregime protest happens on a regular basis, it creates a disturbing normality to authoritarian rulers. They would almost certainly wish to prevent it, but may well decide to tolerate it. On the one hand, they would probably know to distinguish it from disruptive protest that threatens to plunge the regime into an existential crisis. On the other, they would probably know that the repression might be counterproductive, as it reinforces the delegitimizing message of communicative protest. We may sum up these empirical findings and theoretical intuitions by concluding that authoritarian regimes apparently differ in their protest equilibria. Some do as dictatorships are expected to do and suppress all public expressions of discontentment. Others institutionalize certain levels of tolerance toward certain forms of regular collective protest that respect the disciplinary red lines drawn by the regime. Still others, however, seem to accommodate recurrent episodes of genuine contention that punctuate periods of relative quiescence. Such chronic contentious irritations do not seem to conform to our notion of (Nash) equilibria in which no player has incentives to unilaterally deviate from his or her strategy. They place regimes in a permanent state of disequilibrium. If this is indeed the case, if authoritarian regimes accommodate varying levels and types of mass contention, we should expect popular protest under authoritarianism to be causally heterogeneous. Whether it is capable of provoking a regime crisis depends on its context. It depends on the ordinary level of protest the regime has been experiencing over the previous years. Only unusual protest that appears as extraordinary against the background of ordinary protest carries the potential of triggering disruptive spirals of mass mobilization that puts the resilience of the authoritarian regime to test. Existential Uncertainties: The Meaning of Crisis

While the literature on social movements in established democracies has been largely concerned with explaining the emergence of protest campaigns, the literature on contentious politics in established autocracies has focused its attention on tracing its consequences.6 In the comparative study of political regimes, the consequence of interest has been regime change. Over the past decades, we have witnessed some dramatic instances of movement success—such as the Philippine people power movement of 1986, the mass demonstrations that swept away the communist gerontocracies of East Germany and Czechoslovakia in 1989, the subsequent wave of protest that destabilized long-standing African dictatorships, and the post-Soviet color revolutions of the 2000s. Inspired by these epic events, scholars have

50

Andreas Schedler

inquired into the extent to which, and the conditions under which, mass protests (as well as other forms of contentious action) contribute to the downfall of authoritarian regimes.7 In this chapter, by contrast, I do not explore sources of regime change, but instead sources of regime crises. In modern politics, the notion of crisis often serves to denounce states of severe systemic malfunctioning that require urgent remedial intervention. Crisis talk in this sense is ubiquitous. We talk of economic crises; environmental crises; educational crises; crises of the family, of political parties, of the state, and so forth. The conception of crisis I employ here, however, is not functional but existential. In functional crises, the performance of a system is in doubt; in existential crisis, its survival is in doubt. In medical scholarship of ancient Greece, crises marked those painful moments of uncertainty that decided “whether the patient would survive or die” (Koselleck 2004, p. 619). In an analogous manner, crises of authoritarian regimes are moments of dramatic uncertainty in which both trajectories seem possible: the regime may fall or may regain control. What defines regime crises is not one outcome or the other, but the uncertainty of outcomes. Their defining feature is not regime change but indeterminacy, the possibility of regime change. Yet how can we tell situations of relative certainty (regime stability) from situations of dramatic uncertainty (regime crisis)? Institutional uncertainty (or indeterminacy) is a rather elusive dependent variable. It is not a matter of empirical facts we can ascertain with relative objectivity, but a matter of intersubjective perceptions. When observers ascertain changes in political regimes, they make ex post claims about the empirical world; when they observe regime crises, they make ex ante claims about possible future worlds. The observation of regime changes is a matter of descriptive inferences, the observation of regime crises a matter of predictive inferences. Although we can try to reconstruct actor perceptions through interviews and archival studies (see Alexander 2002; Ermakoff 2008; Weyland 2014), subjective expectations are not observable. They are private knowledge. In the limited public spheres of authoritarian regimes, contending actors will offer competing narratives and performances. Regime representatives will try to uphold appearances of strength and normality while their adversaries will introduce diagnoses of acute crisis.8 Regime crises are relevant, however, not because of their capacity of creating mental confusion and public controversy, but because of their capacity of inducing behavioral changes. When actors come to believe that the institutional constraints of the regime are loosening, they will be doing things they would not be doing under normal conditions. In particular, I contend, in the face of mass protests that threaten to destabilize the regime, rulers will be willing to resort to their last means of self-defense: lethal repression.

Disturbing the Dictator

51

Measures of Last Resort: The Logic of Repression

According to the “the law of coercive responsiveness,” nondemocratic “authorities respond to behavioral threats with repression” (Davenport 2007, pp. 7–8).9 They are generally thought to do so in a roughly proportional manner. Striving to navigate, on rough seas, between the Scylla of excessive and the Charybdis of insufficient violence, they seek to choose levels of repression that are “appropriate to” or “consistent with” the levels of threat they face (Gartner and Regan 1996, p. 277).10 The positive lineal relation between threats and repression allows for major exceptions, though. Regimes may resort to excessive repression against mild threats and show leniency in the face of formidable threats. Excessive repression. Due to cognitive factors (misperceptions), psy-

chopathologies (paranoia), strategic foresight (preventive repression), dramaturgical calculations (propagandistic threat exaggeration), or structural factors (regime weakness), authoritarian governments may respond with extreme risk aversion to apparently innocuous forms of protest. They may “overreact” and respond with massive brutality to small acts of dissidence “that fall short of anything like a robust revolutionary challenge” (Boudreau 2005, p. 34). Similarly, autocrats may respond to threatening, yet peaceful, demonstrations through military action that exceeds any notions of rationality or proportionality of means. Syrian president Bashar al-Assad exemplifies such brutal overreach in the face of nonviolent mass protest. In early 2011, when peaceful demonstrators called for the expansion of civil liberties and political rights, he refused to defuse the situation through political reform and instead led his country, with the self-righteous blindness of the philosopher king, into mass murder, civil war, and disintegration.

Leniency. In contexts of mild contentious challenges, regimes may renounce the use of repression as long as they have other means at their disposal. Repression is only one strategy within the repertoire of authoritarian threat management and usually not the first one. Before resorting to repression, governments will often try to contain contentious actions through more accommodative means such as material side payments. Yet even in situations of significant contentious threats, autocrats may renounce the use of repression for a variety of reasons. They may not have the physical or human means of controlling huge crowds (resource scarcity). They may expect repression to trigger even larger or less peaceful protests (counterproductive repression). Or they may fear that higher- or lower-level officials in the security apparatus will refuse shooting at their co-citizens (agent defection).

52

Andreas Schedler

Within the broad literature on protest and repression, some scholars conceive both variables in continuous terms and study (linear as well as nonlinear) associations between levels of protest and repression (see, e.g., Regan and Henderson 2002). Others introduce qualitative distinctions between types of protest, in particular between violent and peaceful protest. They tend to find self-reinforcing circles of oppositional violence and governmental repression.11 Here, by contrast, I emphasize a distinction between types of repression, which has found little systematic attention in the recent literature: the distinction between lethal and nonlethal repression. Above, I suggested that authoritarian rulers may possibly respond to existential contentious threats with lethal repression. Here, I wish to defend a stronger, less commonsensical claim: authoritarian rulers are even likely to respond to existential contentious threats (but not to normal, nonthreatening protest) with lethal repression. The strong nexus I posit between critical threats and lethal repression rests on three assumptions: 1. As repression is both costly and risky, it is a measure of last resort. Authoritarian rulers use it only when they hold it necessary. In the face of peaceful protest, repression is unnecessary as long as protests look innocuous and as long as other means of threat management, such as propaganda, co-optation, policy concessions, or cosmetic reform, are available. 2. When repression does seem necessary, because mass demonstrations look threatening and regimes have depleted their reservoir of alternative approaches, rulers will prefer to exercise nonlethal violence over lethal violence because the costs and the risks of backlashes are lower (Francisco 2005). 3. When contentious challenges reach critical proportions, non-lethal repression may be insufficient to contain them. And although lethal repression can be counterproductive, authoritarian rulers commonly do seem to embrace “the dictator’s illusion” (Francisco 2005, pp. 62) that holds ruthless repression to be a reliable technique for suffocating mass protest. As a matter of fact, more often than not, their illusions come true. Just think of the Prague Spring, Tiananmen, or Egypt under General Abdel Fattah el-Sisi.

In sum, I am proposing to take, not the occurrence of popular protest, but its lethal repression, as an observable indicator of authoritarian crises (as perceived by authoritarian rulers). As I argued above, we cannot take the occurrence of mass protests as a reliable sign of authoritarian crises. In some authoritarian regimes, a certain level of political contention may be a normal part of politics. However, when rulers respond to peaceful mass

Disturbing the Dictator

53

protests with lethal repression, we can infer that they perceived these protests as genuine threats to the stability of the regime. Deadly repression is authoritarian rulers’ last line of defense. Contentious crisis indicates that all other lines of defense have been broken. In the face of critical contention, lethal repression will therefore be their strategy of choice. Nothing else may grant them survival in power. Lethal repression of contentious challenges may be preventive or reactive. It may be a response to protests that carry the risk of spiraling out of control (a potential crisis) or to protests that are already spiraling out of control (an acute crisis). The former take place in those “critical regions” in which “a seemingly innocuous social protest is transformed into complete mobilization” (Yin 1998, p. 546). The latter are already part of full-blown regime crises. In either case, repression responds to perceived threats to regime stability. To read repressive responses to popular protest as signs of governmental crisis perceptions seems to be a reasonable rule of inference. It is not an infallible rule, though. False positives (cases of lethal repression in normal times) exist as well as false negatives (cases of regime crisis without lethal repression). For the explorative purposes of this chapter, however, I simply assume they are rare enough to be ignored, for the time being. Surveying the Terrain: Empirical Explorations My (largely) inductive hypothesis of the disturbing normality of peaceful popular protest in authoritarian regimes carries three simple empirical implications. First, if mass demonstrations indeed are ordinary phenomena in many authoritarian regimes, rather than exceptional disruptive events, we need to see them happening with some frequency. Second, if normal protest strives to communicate genuine political dissent, rather than performing a delicate dance of self-discipline and good conduct for central authorities, we should see it go beyond small scales and local concerns. Third, if a certain level of contentious politics forms an integral part of authoritarian normality, disturbing yet not disruptive, we should expect rulers to treat them with guarded indifference rather than alarm, and to ordinarily tolerate them rather than repressing them. Above all, in the face of mere contentious irritations we should not expect rulers to resort to lethal repression, their measure of last resort in the eventuality of contentious crises. Data

To test these simple hypotheses, I employed the recently released Social Conflict in Analysis Database by Idean Salehyan from the University of North

54

Andreas Schedler

Texas and Cullen Hendrix from the University of Denver (www.scaddata .org). The database is the successor to the Social Conflict in Africa Database, extending its coverage to Mexico, Central America, and the Caribbean (1990–2013). Its units of analysis are individual conflict events, rather than the conventional country-years. Its comparative advantage to other crossnational data sources lies in its event coverage, its level of disaggregation, and the richness of event information. While its registry of conflict events is certainly not complete, it is significantly less incomplete than the standard source of cross-national conflict data, the Arthur Banks Cross-National Time-Series Data. Based on LexisNexis searches of Associated Press (AP) and Agence France-Presse (AFP) newswires, SCAD contains information on a broad set of social and political conflicts: organized and spontaneous demonstrations, organized and spontaneous riots, general and limited strikes, nonlethal and lethal state repression (pro-governmental violence ), rebellion (antigovernmental violence), private repression (extragovernmental violence), and elite violence (intragovernmental violence). For each event, it seeks to register, among some other items, its duration and location, the number of participants, the identity of its protagonists and its targets, their grievances, and the number of deaths (see Salehyan and Hendrix 2014). On the side of governmental responses to conflict events, SCAD is also superior to extant datasets. Rather than estimating the intensity of state repression on a yearly basis, it records governmental responses to each conflict event. In contrast to annual aggregates of repression, these event data allow us to relate repressive stimuli and responses in a direct manner. SCAD offers an additional advantage of disaggregation by distinguishing between lethal and nonlethal acts of repression. The Frequency of Demonstrations

The SCAD dataset does not distinguish between regime types. It registers conflict events regardless of whether they take place in democratic or nondemocratic contexts. To identify authoritarian country-years, I drew the dividing line between democracy and dictatorship following the operational proposal advanced by Michael Wahman, Jan Teorell, and Axel Hadenius in their Authoritarian Regimes Dataset (2013). As they explain, Using the mean of each country’s Freedom House and Polity scores, converted to a scale from 0 (least democratic) to 10 (most democratic), we distinguish democracies from autocracies at a score of 7.5—the authoritarian family consisting of all regimes with a score below that point. We chose this threshold value by estimating the mean cutoff point separating democracy from autocracy in five well-known categorical measures of

Disturbing the Dictator

55

democracy—those of Adam Przeworski et al., Scott Mainwaring et al., and Garry Reich, along with the categorical thresholds for democracy of Freedom House and Polity (Hadenius and Teorell 2007, pp. 145–146).12

The Appendix to this chapter contains the resulting political regimes covered by the SCAD dataset (1990–2012). It also lists the duration of each regime as well as the total number and annual average of (organized as well as spontaneous) protest demonstrations as captured by SCAD.13 Given my exclusive interest in protest demonstrations, my data exclude all demonstrations that were staged in support of the government.14 In addition, given their extraordinary nature as well as their causal interdependence (within an international wave of contentious diffusion), I also excluded the protest events of the Arab Spring.15 From 1990 through 2012, the resulting SCAD data cover 32 democracies and 66 autocracies, spanning a total of 1,326 regime years. More than two-thirds of these correspond to authoritarian regimes (929 years or 70 percent). For the sample under examination, SCAD registered 3,888 protest demonstrations, more than three-quarters of which took place in authoritarian contexts (3,006 or 77.3 percent). Table 3.1 displays the frequency distribution of different categories of conflict events within both regime types.16 Peaceful demonstrations (35.9 percent) and riots (21.8 percent) are the most frequent conflict event types in autocracies. In democracies, the relative incidence of private repression Table 3.1 Conflict Events by Regime Type, 1990–2012 Conflict Events

Demonstrations Riots

Strikes

Rebellion

Private repression

Elite violence

All events

N % N % N % N % N % N % N %

Democracies

894 25.2% 556 15.7% 463 13.0% 507 14.3% 1,113 31.4% 15 0.4% 3,548 100.0%

Autocracies 2,962 35.9% 1,798 21.8% 1,027 12.4% 701 8.5% 1,659 20.1% 106 1.3% 8,253 100.0%

All Regimes

3,856 32.7% 2,354 19.9% 1,490 12.6% 1,208 10.2% 2,772 23.5% 121 1.0% 11,801 100.0%

Source: Social Conflict in Analysis Database (SCAD). Note: Conflict events: Demonstrations = organized or spontaneous. Riots = organized or spontaneous. Strikes = general or limited. Rebellion = antigovernmental violence. Private repression = extragovernmental violence. Elite violence = intragovernmental violence. My figures exclude pro-government violence (repression), pro-government events, and Arab Spring country years. For country and regime coverage, see the Appendix.

56

Andreas Schedler

(31.4 percent) is substantively higher than in autocracies (21.1 percent), which probably reflects the inclusion of Mexico and its ongoing drug war in the dataset (see Schedler 2014).17 Figure 3.1 shows the frequency distributions of mean annual protest demonstrations for all democracies and autocracies in my sample.18 On both sides, a majority of regimes show low annual protest averages. Autocracies, however, host greater variance. A higher portion of authoritarian regimes have experienced five or more protest demonstrations per year and the overall annual event average is higher in autocracies (3.2 demonstrations per year) than in democracies (2.2 demonstrations per year). These simple descriptive data provide strong empirical support for my initial normality hypothesis. Rather than rare and unusual events, protest demonstrations (as well as other types of contentious actions) seem to belong to the regular political landscape of numerous authoritarian regimes. As it appears, rather than calmly presiding over quiescent silent subjects, many autocrats have to manage a constant stream of contentious challenges from below. The Scale of Demonstrations

How significant are the protest demonstrations that autocrats face on an ordinary basis? Are we talking about fleeting minuscule events in remote regions, or large prolonged protest campaigns in major cities? Table 3.2 indicates broad empirical variance.

Size. While large protests involving over 100,000 participants are rather rare (3.2 percent), the portion of very small protests (with less than 100 participants) is not large either (16.8 percent).19 Well over a third of all demonstrations under authoritarianism sponsor between 100 and 1,000 participants (36.2 percent) and another third between 1,000 and 10,000 (33.6 percent). Rather than self-sacrificial performances of superminoritarian radicals, these seem to be quite broad and significant expressions of dissidence. The frequency distribution of protests by size is similar in democracies (see Table 3.2). Duration. In autocracies (just as in democracies), only few protest demon-

strations seem to be part of sustained nonviolent resistance campaigns (Chenoweth and Stephan 2011b). The overwhelming majority consists of short-lived one-day events (80.1 percent). Only a fraction last longer than a week (6.4 percent) (see Table 3.2).

Territorial reach. Popular rebellions do not necessarily start in the capital

city. The 2011 mass protests in Tunisia, Libya, and Syria, for instance, began in provincial cities. Yet unless protests reach a country’s political center, they

Figure 3.1 Protest Demonstration by Regime Type, Annual Averages, 1990–2012 Democracies

Regimes (N = 32), regime years (N = 397), protest demonstrations (N = 882), overall Ø = 2.2.

Authoritarian Regimes

Regimes (N = 66), regime years (N = 929), protest demonstrations (N = 3,006), overall Ø = 3.2.

Source: Social Conflict in Analysis Database (SCAD). Note: Demonstrations = organized or spontaneous (excluding pro-government demonstrations). For country and regime coverage, see the Appendix.

58

Andreas Schedler

Table 3.2 Scale of Demonstrations by Regime Type, 1990–2012 Number of participants

Less than 100 101 to 1,000

1,001 to 10,000

10,101 to 100,000 Over 100,000

Total

Duration

1 day

2–7 days

8–14 days

15 days or more

Total

Territorial reach

One or multiple rural areas

One or multiple cities Capital city

Nationwide

Total

Democracies

Autocracies

122 18.2% 267 39.9% 204 30.4% 63 9.4% 14 2.1% 670 100.0%

363 16.8% 781 36.2% 725 33.6% 219 10.2% 69 3.2% 2,157 100.0%

Chi2 = 6.4

N % N % N % N % N % N %

N % N % N % N % N %

N % N % N % N %

p = .172

Chi2 = 4.3

p = .227

Chi2 = 31.5

p = .000

744 83.2% 100 11.2% 20 2.2% 30 3.4% 894 100.0%

128 14.9% 212 24.7% 480 55.9% 38 4.4% 858 100.0%

2,373 80.1% 399 13.5% 75 2.5% 115 3.9% 2,962 100.0%

238 8.5% 769 27.4% 1,642 58.5% 157 5.6% 2,806 100.0%

Source: Social Conflict in Analysis Database (SCAD). Note: Rural areas < 100,000 inhabitants. For country and regime coverage, see the Appendix. Boldface percentages highlight key differences between autocracies and democracies.

are more easily contained. As Table 3.2 indicates, rural protests are either rare events in autocracies or they do not make it into the headlines. Only 8.5 percent of protest demonstrations under authoritarianism took place in villages or cities with less than 100,000 inhabitants (by comparison, democracies

Disturbing the Dictator

59

seem to be more amenable to protest in small places). Yet at the other end of the territorial threat scale, genuinely national protests are uncommon as well (5.6 percent). The bulk of protests either take place in the capital city (58.5 percent) or in major cities outside the capital (27.4 percent). In sum, even if most protest demonstrations under authoritarianism are on-and-off events that happen on a specific day and seldom survive their first week, they do seem to assemble significant numbers of participants (between 100 and 100,000) at significant places (the capital city or other major cities). The Motives of Demonstrations

Above all, students of contentious action in contemporary China have been reading popular protest under authoritarianism as essentially regime supportive: useful and domesticated, tolerated, encouraged, or even managed by political elites. Do such functional diagnoses of regime-friendly protest apply to other authoritarian regimes as well? Are the peaceful protest demonstrations that take place in autocracies protest demonstrations at all? Do they really count as autonomous acts of defiance, challenge, resistance, confrontation, rebellion, protest, dissidence, and so forth, as we usually describe them? Or are they no more than sophisticated forms of tacit collaboration between subnational protest entrepreneurs and national authorities? SCAD does not (and cannot) contain fine-grained information on the discursive framing of conflict events. Yet it does register in a rough manner what kind of grievances protesters articulated (“the first issue that was mentioned at the source of the tension/disorder”) and whether either “the central government” or subnational authorities (“a regional, provincial or local government”) were “the target of the event” (Salehyan and Hendrix 2014, pp. 4–5). Table 3.3 shows the respective frequency distributions within the event category of peaceful demonstrations (by regime type, from 1990 through 2012).20

Grievances. As the figures in Table 3.3 indicate, while in almost oneeighth of registered demonstrations under authoritarianism the motives of protest remain unclear (13.4 percent), a fair number of protests indeed limit themselves to articulating more innocuous complaints about economic issues (19.8 percent) or foreign affairs (14.0 percent). However, at least half of protest demonstrations under authoritarianism fail to respect the taboo of criticizing the political regime and do articulate grievances about democracy, elections, and human rights (47.3 percent). By comparison, in electoral democracies, economic grievances are more relevant (31.8 percent) while regime complaints are less salient (29.4 percent).

60

Andreas Schedler

Table 3.3 Motives of Demonstrations by Regime Type, 1990–2012 Grievances

Economic issues Political regime

Minority discrimination Foreign affairs

Other or unknown issues

Total

Targets

Central government Other Total

Subnational government Other Total

N % N % N % N % N % N % N % N % N % N % N % N %

Democracies

Autocracies

284 31.8% 263 29.4% 40 4.5% 140 15.7% 167 18.7% 894 100.0%

585 19.8% 1,402 47.3% 162 5.5% 416 14.0% 397 13.4% 2,962 100.0%

Chi2 = 110.7

p = .000

Chi2 = 39.9

p = .000

Chi2 = .060

.806

504 56.4% 390 43.6% 894 100.0%

47 5.3% 847 94.7% 894 100.0%

2,010 67.9% 952 32.1% 2,962 100.0%

162 5.5% 2,800 94.5% 2,962 100.0%

Source: Social Conflict in Analysis Database (SCAD). Note: Economic issues = social, economic, and environmental issues. Regime issues = democracy, elections, and human rights. Minority issues = religious or ethnic discrimination. Foreign affairs = international relations and foreign affairs. For country and regime coverage, see the Appendix. Boldface percentages highlight key differences between autocracies and democracies.

Targets. Contrary to the image of local movements bringing local issues to

the attention of local authorities, two-thirds of all protest demonstrations under authoritarianism target the central government (67.9 percent). Only a small minority of reported demonstrations address local authorities (5.5 percent). In democracies, the dynamics of protest do not seem more decentralized (a similar share of protests target subnational governments), but somewhat less politicized (a lower portion of protests address the central

Disturbing the Dictator

61

government). Of course, international news agencies may be seriously underreporting demonstrations that do not target the national centers of power. The Frequency of Repression

As I posited, perceptions of crises depend on perceptions of normality. To the extent that authoritarian rulers perceive peaceful protest demonstrations to represent manageable, nonthreatening, perhaps even useful, elements of ordinary politics rather than alarming symptoms of manifest crises, they should tolerate rather than repress them. The percentage of protest demonstrations that authoritarian governments meet with repression indeed tells a story of relative tolerance (even though democracies are even more tolerant of peaceful protest). According to my figures, two-thirds of peaceful protest demonstrations in authoritarian regimes do not trigger repressive responses (65.1 percent). A good quarter meet with nonlethal repression (27.0 percent) and less that one in twelve with lethal repression (7.9 percent). The corresponding figures for democracies are 80.3 percent (no repression), 15.4 percent (nonlethal repression), and 4.3 percent (lethal repression).21 Threat and Repression

Anybody who has ever taken part in one of those cheerful protest demonstrations in which a chanting handful of participants is outnumbered by the watchful police force stationed at the other side of the street knows that not all protest demonstrations are equal. Dependent on their size and demands, some are more threatening than others. We should expect authoritarian rulers to adjust their responses to the scale as well as to the motives of protest.

The scale of protest. Students of nonviolent contentious action commonly assume the number of participants to be the crucial determinant of its success. For instance, most game theoretic models of mass participation in antiregime protest rest on explicit assumptions about the determinate power of numbers: “The regime falls if enough people mobilize” (E. Bueno de Mesquita 2010, p. 447). Following this well-established line of reasoning, it seems to be reasonable to assume a positive lineal relationship between the size of demonstrations and the perceived intensity of threat they pose to an authoritarian government. The bigger the show, the more threatening it looks. Temporal and spatial variables are likely to reinforce the suggestive force of numbers: the longer a protest campaign lasts, and the closer it gets to major cities (and in particular the capital city), the more menacing it is. The bar graphs in Figure 3.2 cross-tabulate these three dimensions of protest scale (size, duration, and location) with repressive responses. As the

62

Andreas Schedler

upper pair of graphs shows, repression in democracies is insensitive to the size of protest. In autocracies, nonlethal repression becomes less, and lethal repression more frequent, as protest demonstrations grow in size. When less than 10,000 people show up at a demonstration, in one occasion out of four, someone gets sprayed with water cannons, beaten up, or arrested. In larger protests, only one out of ten demonstrations is met with nonlethal repression. By contrast, lethal repression is rare in very small protests. Yet when more than 10,000 people participate, in one occasion out of ten, someone gets killed (see Figure 3.2). The often cited notion that protesters find “safety in numbers” applies only to nonlethal repression. Large crowds are more likely to provoke lethal responses than small gatherings. Although the differences in repressive responses to peaceful protest are both statistically and substantively significant, it is still remarkable that repression is not the rule, but the exception. If, as I hypothesized, lethal repression is a good indicator of perceived authoritarian crises, less than a tenth of even large protest demonstrations can count as such. As it seems, authoritarian governments are able to normalize not just the odd gathering of a few dissidents, but sizable public protests. In four out of five cases, they let them happen and let them pass without applying public force. The duration of events does not carry a lineal impact on repression. Fleeting protest events that do not last beyond a single day are least likely to provoke deadly repression. When demonstrations carry on into the first week, the rate of lethal responses rises from 6.0 to 16.3 percent. For longer campaigns, it remains at similar levels. Democracies show similar associations, although at lower levels of repressiveness. Of course, event duration may be endogenous to repression, if early repression creates fresh grievances that prolong the protest movement. So far, these figures confirm the hypothesized association between levels of threat (protest size and duration) and rates of lethal repression. The data on protest location, by contrast, yield a more mixed picture. As expected, under dictatorship, nationwide protests are most likely to suffer deadly repression (19.7 percent) and least likely to go without repression (53.5 percent). Demonstrations that take place in villages or small towns and those that take place in the capital city are treated in apparently inconsistent ways. The rate of deadly repression is four times higher against the former, which are supposed to be more innocuous (17.2 percent), than against the latter, which are supposed to be more threatening (4 percent). Apparently, the frequency of lethal repression in rural settings reflects the infrequency of rural protest, rather than some generic territorial notion of threat intensity. As we saw above, most protests take place either in the capital city (58.5 percent) or in other major cities (27.4 percent) (see Table 3.2). As it seems, authoritarian rulers tolerate them as normal urban events. In

Figure 3.2 Repressive Responses to Demonstrations, by Scale, 1990–2012

63 continues

64

Figure 3.2 Continued

Disturbing the Dictator

65

rural settings, by contrast, they treat them as what they are, as exceptional and thus potentially more disruptive incidents. The motives of protest. The demands articulated by protest movements vary in the levels of systemic threat they pose to authoritarian governments. When dissidents direct their protests against the central government or when they call for the transformation of the political regime, they are more likely to put authoritarian rulers on alert. Figure 3.3 disaggregates official responses to demonstrations by motives and targets of protest. As it shows, authoritarian governments are rather egalitarian in their treatment of demonstrations. They are happy to see their citizens protesting against a foreign government. Yet otherwise, they subject them to similar rates of repressive treatment regardless of their protest agenda. While apparently indifferent to the substantive grievances their subjects articulate, authoritarian governments are highly sensitive to the targets they choose. Protest demonstrations that address the central government are twice as likely to suffer from nonlethal repression (32.6 percent) than those that target other actors (15.2 percent). Their rates of lethal repression (9.0 percent) are 60 percent higher than those of other demonstrations (5.6 percent). By contrast, protests that target subnational authorities suffer from similar rates of nonlethal repression as other protests, yet triplicate the incidence of lethal repression (21.6 percent) in comparison to other events (7.1 percent). The staggering death rates of subnational protests suggest that peaceful protest demonstrations in autocracies are more likely to provoke local crises than national crises. They also suggest that we need to rethink the role of popular protests at different levels of authoritarian governance. As it seems, the idea that has crystallized from research on contentious action in contemporary China, according to which national elites tolerate and even encourage popular protest to keep subnational elites in check, does not hold outside its context of discovery. Most protest demonstrations under authoritarian rule articulate regime grievances, target the central government, and take place in the capital city. Within this general context, those protests that do target subnational governments are not more likely to be tolerated. Quite to the contrary, they are much more likely to fall victim to lethal repression. To assess the individual impact that the various characteristics of protest demonstrations have on the likelihood of repression, I ran two multinomial logistic regressions (for the split sample of democracies and autocracies), which put our preceding bivariate explorations in perspective.22 In addition to all previously examined features of protest demonstrations, I added to my list of independent variables the absolute number of protest demonstrations that took place in the previous calendar year. I expected authoritarian rulers to perceive protests as more threatening when they are uncommon events. When demonstrations

Figure 3.3 Repressive Responses to Demonstrations, by Motives, 1990–2012

66

Figure 3.3 Continued

67

68

Andreas Schedler

are repeat performances from previous years, rulers should be more willing to renounce repression and to tolerate protests as normal occurrences. Table 3.4 reports our results.23 Perhaps the most intriguing finding concerns the master variable of peaceful contentious politics, the size of protest. The regressions essentially confirmed the asymmetric patterns of association we saw before: in democracies, repression is unrelated to the number of participants. In autocracies, larger crowds make regimes more reluctant to exercise nonlethal repression, yet more willing to apply lethal repression. As the size of peaceful protest grows, participants are less likely to be clubbed (eb = .77), yet more likely to be shot (eb = 1.56). Protest events that last longer are more likely to provoke lethal repression (or possibly the other way around). Their territorial reach bears no effect on rates of repression. When protesters express socioeconomic grievances, yet not when they articulate regime issues, democratic governments are more likely to repress them through nonlethal means. Authoritarian regimes, by contrast, step up both forms of repression in both types of events. When autocrats face socioeconomic protest, the odds of nonlethal repression rise by 68 percent (eb = 1.68) and the odds of lethal repression almost 80 percent (eb = 1.79). When they confront demands for regime change, the odds of nonlethal repression (eb = 1.90) as well as the odds of lethal repression (eb = 2.27) basically duplicate. Similar repressive dynamics kick in when demonstrators direct their demands to the central government. In these cases, the odds of nonlethal repression go up by almost 100 percent (eb = 1.92) and the odds of lethal repression by over 150 percent (eb = 2.53). When dissidents target subnational authorities, the odds of nonlethal repression rise by two-thirds (eb = 1.64) while, dramatically, the odds of lethal repression more than triplicate (eb = 3.72). My thin measure of protest routines, the number of peaceful demonstrations that were reported in the calendar year previous to each event, turned out as expected in democracies but not in dictatorships. Previous experience with protest demonstrations consistently augments the tolerance of democratic regimes. In authoritarian regimes, by contrast, it slightly lowers the rate of nonlethal repression, yet leaves the rate of lethal repression unaffected. As it seems, quite remarkably, the frequency of peaceful demonstrations in the recent past does not alter authoritarian threat perceptions. Neither do authoritarian regimes repress demonstrations more harshly after a period of tranquility, nor do they treat them more leniently if they are anteceded by more agitated times. Conclusion

The empirical patterns that emerged in my explorative analysis of peaceful protests in Africa, Central America, and the Caribbean are inconsistent with

Disturbing the Dictator

69

Table 3.4 Explaining Repressive Responses to Democracies, by Regimes, 1990–2012

Democracies

No. of participants (5 categories) Duration (4 categories) Territorial reach (4 categories) Socioeconomic issues Regime issues Minority discrimination Foreign affairs Target central government Target subnational government No. of demonstrations in previous year Constant

Nagelkerke pseudo R2 = .161 N = 643

Authoritarian regimes

No. of participants (5 categories) Duration (4 categories) Territorial reach (4 categories) Socioeconomic issues Regime issues Minority discrimination Foreign affairs Target central government Target subnational government No. of demonstrations in previous year Constant Nagelkerke pseudo R2 = .130 N = 2038

Nonlethal Repression p eb

Lethal Repression p eb

.430 .007 .777 .029 .201 .791 .848 .672 .323 .000 .305

.350 .003 .307 .296 .482 .373 . .123 .015 .021 .000

.000 .061 .740 .013 .001 .017 .699 .000 .054 .035 .049

.903 1.498 1.044 2.212 1.611 .805 .914 1.121 .466 .934

.777 1.151 1.028 1.682 1.907 1.995 .909 1.962 1.642 .989

.000 .000 .278 .157 .030 .799 .090 .004 .001 .997 .070

1.301 2.003 .716 2.427 1.857 3.174. .000 3.077 8.012 .911

1.564 1.763 .857 1.793 2.270 1.164 .336 2.534 3.721 1.000

Sources: Conflict data: Social Conflict in Analysis Database (SCAD). Regime data: Michael Wahman, Jan Teorell, and Axel Hadenius. 2013. “Authoritarian Regime Types Revisited: Updated Data in Comparative Perspective.” Contemporary Politics 19, no. 1: 19−34. Note: Multinomial logistic regressions. Reference category: no repression. eb = odds ratio. Statistically significant associations are in boldface (p ≤ .05). All independent variables (except the first three and the last) are dummies. For country and regime coverage, see the Appendix.

the notion that mass contention is rare under authoritarianism. They also disconfirm the association between contentious action and regime crises. Most peaceful protest demonstrations seem to be part of normal authoritarian life. Neither do they trigger regime crises in any systematic fashion, nor do they emerge in the wake of such crises. My findings also run counter to the notion of anticipatory obedience by risk-averse protesters. Rather than staging innocuous forms of protest that limit themselves to preapproved themes and nonthreatening strategies, contentious actors regularly articulate sensitive grievances in challenging ways. If they know what scholars of comparative

70 Appendix Regimes and Protest Demonstrations, Case Coverage (Social Conflict in Analysis Database [SCAD], 1990–2012) Country

Algeria Angola Benin Benin Botswana Burkina Faso Burundi Cameroon Central African Republic Chad Congo Congo, Dem. Republic of Costa Rica Côte d'Ivoire Cuba Dominican Republic Dominican Republic Dominican Republic Egypt El Salvador Eritrea Ethiopia (−1992) Ethiopia (1993−) Gabon Gambia Ghana Ghana Guatemala Guatemala Guatemala Guatemala Guatemala Guatemala Guinea Guinea-Bissau Haiti Honduras Honduras Jamaica Kenya Kenya Kenya Kenya Kenya Lesotho Liberia Libya

Regime Years

Political Regime

1990–2010 1990–2012 1990 1991–2012 1990–2012 1990–2012 1990–2012 1990–2012

Authoritarianism Authoritarianism Authoritarianism Democracy Democracy Authoritarianism Authoritarianism Authoritarianism

1990–2012 1990–2012 1990–2012 1990–2012 1990–1993 1994–1995 1996–2012 1990–2010 1992–2012 1993–2012 1990–1992 1993–2012 1990–2012 1994–2012 1990–2000 2001–2012 1990–1995 1996–2001 2002–2005 2006–2008 2009–2010 2011–2012 1990–2012 1990–2012 1990–2012 1990–2008 2009–2012 1990–2012 1990–2002 2003–2008 2009 2010–2011 2012 1990–2012 1990–2012 1990–2010

Authoritarianism Democracy Authoritarianism Authoritarianism Democracy Authoritarianism Democracy Authoritarianism Democracy Authoritarianism Authoritarianism Authoritarianism Authoritarianism Authoritarianism Authoritarianism Democracy Authoritarianism Democracy Authoritarianism Democracy Authoritarianism Democracy Authoritarianism Authoritarianism Authoritarianism Democracy Authoritarianism Democracy Authoritarianism Democracy Authoritarianism Democracy Authoritarianism Authoritarianism Authoritarianism Authoritarianism

1990–2012 1990–2012 1990–2012

Authoritarianism Authoritarianism Authoritarianism

Duration Demonstrations Demonstrations (in years) (total number) (per year) 21 23 1 22 23 23 23 23

100 22 1 16 7 25 35 37

23 23 23 23 4 2 17 21 21 20 3 20 23 19 11 12 6 6 4 3 2 2 23 2 23 19 4 23 13 6 1 2 1 23 23 21

79 7 110 151 4 4 30 204 32 7 5 30 24 5 17 5 7 13 21 2 3 5 40 19 388 32 16 33 48 31 5 5 7 20 42 38

23 23 23

31 17 13

4.8 0.9 1.0 0.7 0.3 1.1 1.5 1.6

1.3 0.7 0.6

3.4 0.3 4.8 6.6 1.0 2.0 1.8 9.7 1.5 0.4 1.7 1.5 1.0 0.3 1.5 0.4 1.2 2.2 5.3 0.7 1.5 2.5 1.7 9.5 16.9 1.7 4.0 1.4 3.7 5.2 5.0 2.5 7.0 0.9 1.8 1.8 continued

71 Appendix continued Country

Madagascar Madagascar Madagascar Malawi Malawi Malawi Mali Mali Mali Mauritania Mauritius Mexico Mexico Morocco Mozambique Mozambique Mozambique Namibia Nicaragua Nicaragua Nicaragua Nicaragua Niger Niger Niger Niger Niger Nigeria Panama Rwanda Senegal Senegal Sierra Leone Sierra Leone Somalia South Africa South Africa South Sudan Sudan (−2011) Sudan (2012−) Swaziland Tanzania Togo Trinidad and Tobago Tunisia Uganda Zambia Zambia Zambia Zambia Zimbabwe

Regime Years

1990–1991 1992–2008 2009–2012 1990–1993 1994–2000 2001–2012 1990–1991 1992–2011 2012 1990–2012 1990–2012 1990–1999 2000–2012 1990–2010 1990–2007 2008 2009–2012 1992–2012 1990–1991 1992–1995 1996–2010 2011–2012 1990–1992 1993 1994–2003 2004–2006 2007–2012 1990–2012 1990–2012 1990–2012 1990–1999 2000–2012 1990–2006 2007–2012 1990–2012 1990–1993 1994–2012 2012 1990–2011 2012 1990–2012 1990–2012 1990–2012 1990–2012 1990–2010 1990–2012 1990 1991–1992 1993–2007 2008–2012 1990–2012

Political Regime

Authoritarianism Democracy Authoritarianism Authoritarianism Democracy Authoritarianism Authoritarianism Democracy Authoritarianism Authoritarianism Democracy Authoritarianism Democracy Authoritarianism Authoritarianism Democracy Authoritarianism Democracy Democracy Authoritarianism Democracy Authoritarianism Authoritarianism Democracy Authoritarianism Democracy Authoritarianism Authoritarianism Democracy Authoritarianism Authoritarianism Democracy Authoritarianism Democracy Authoritarianism Authoritarianism Democracy Authoritarianism Authoritarianism Authoritarianism Authoritarianism Authoritarianism Authoritarianism Democracy Authoritarianism Authoritarianism Authoritarianism Democracy Authoritarianism Democracy Authoritarianism

Duration Demonstrations Demonstrations (in years) (total number) (per year) 2 17 4 14 7 12 2 20 1 23 23 10 13 21 18 1 4 21 2 4 15 2 3 1 10 3 6 23 23 23 10 13 17 6 23 4 19 1 22 1 23 23 23 23 21 23 1 2 15 5 23

2 14 23 5 15 36 2 16 30 61 3 136 213 70 18 1 5 12 7 5 19 1 4 3 46 7 22 267 27 26 0 51 25 7 95 76 239 1 76 21 49 45 61 19 12 39 1 1 24 6 151

1.0 0.8 5.7 0.4 2.1 3.0 1.0 0.8 30.0 2.6 0.1 13.6 16.4 3.3 1.0 1.0 1.3 0.6 3.5 1.3 1.3 0.5 1.3 3.0 4.6 2.3 3.7 11.6 1.2 1.1 0.0 3.9 1.5 1.2 4.1 19.0 12.6 1.0 3.4 21.0 2.1 2.0 2.6 0.8 0.6 1.7 1.0 0.5 1.6 1.2 6.6

72

Andreas Schedler

politics know—namely, that popular protests are unlikely to shake or even shatter an authoritarian regime—we have to conclude that their expressions of dissent are principled rather than opportunistic. Finally, the data that I examined contradict the idea that authoritarian rulers tend to tolerate wellbehaved protesters while repressing more serious challengers. Indeed, the odds of both nonlethal and lethal repression roughly duplicate when demonstrators either address sensitive issues or target the central government. Yet even so, their chances of getting away without repression are about fifty-fifty while their odds of suffering lethal repression are about 1:6. Despite their rough and preliminary quality, my empirical explorations suggest the possibility that the current comparative literature on authoritarian regimes has been developing a large blind spot. In the neofunctionalist emphasis that it places on strategic equilibria, institutional safeguards, and distributive policies, it tends to overlook the contentious quality of authoritarian governance. As it seems, there is more to the politics of authoritarian governance than the design of institutions and the distribution of rents. It involves constant public engagement with multifarious public challenges by principled opponents who do not wish to signal their fundamental conformity through calibrated acts of self-constrained protest, but rather to communicate serious dissent through genuine acts of defiance. Even when democratic citizens know they cannot bring down the regime, they can at least rumple its pretensions of popular consent. Even when they cannot chase the dictator out of the presidential palace, they can at least disturb his or her self-complacent narrative of dictatorial popularity.

Notes I thank Max Bader, Jason Brownlee, Martin Dimitrov, Johannes Gerschewski, SveinErik Helle, Wendy Hunter, Anna Lührmann, Wolfgang Merkel, Christoph Stefes, Eitan Tzelgov, and Kurt Weyland for their valuable comments on earlier versions of this chapter. 1. See also Gandhi (2008), Geddes (2005), Schmitter and O’Donnell (1986), Tullock (1987). 2. On X-centered versus Y-centered research designs, see Gerring (2001). 3. For a brief critical discussion of classic problems of collective action, see Schedler (2013, pp. 305–309). 4. See Scott (1985). On various forms of “everyday transgression” under dictatorship, see also Wedeen (1999, p. 151): “Although sustained collective rebellion may be rare in . . . stable authoritarian regimes . . . individual triumphs that call on shared conditions of suffering or unbelief are visible everywhere.” 5. See, for instance, E. Bueno de Mesquita (2010), Granovetter (1978), Kricheli, Livne, and Magaloni (2010), Lohmann (1993, 1994), Yin (1998). 6. Of course, there are many exceptions. Many of the debates on the Arab Spring have revolved on its origins. The literature on contentious politics in China

Disturbing the Dictator

73

has little illusions about its destabilizing effects. It focuses instead on its stabilizing effects. Weyland (2014) offers a single theory of organizational rationality that explains the emergence and the success of contentious waves. 7. See, for instance, Aspinall (2005), and Wolchik (2010), Chenoweth and Stephan (2011b), Schedler (2013, chap. 10), Kricheli, Livne, and Magaloni (2010), Levitsky and Way (2010, pp. 354–356), O’Donnell and Schmitter (1986, chap. 5), Przeworski et al. (2000), Trejo (2012, chap. 8), and Ulfelder (2005). 8. See Schedler (2013). On the contested nature of crises, see Milstein (2015). On “framing contests” in the face of external shocks, see Boin, Hart, and McConnell (2009). On performative strategies of crisis prevention by authoritarian governments, see Schedler and Hoffmann (2015). 9. Democratic authorities also do so, yet on average at lower levels except in Latin America. The mean level of violations of physical integrity rights is higher in Latin American democracies than in the autocracies of the rest of the world (M. A. Rivera 2010, p. 24, with data for 1981–2014). 10. Often, authoritarian rulers establish “divided structures of contestation” (Lust 2005) that create two classes of opposition forces: licensed, permissible, loyal opposition forces that they tolerate versus intolerable dissidents that they exclude and repress. 11. States tend to respond violently (outside the bounds of legal and humanitarian restrictions) to societal violence, and dissidents tend to respond violently (outside the bounds of peaceful contention) to oppressive state violence. Unbound state violence and societal violence reinforce each other in circles of positive feedback. See, for example, Goodwin (2001), Lichbach (1987), M. A. Rivera (2010, pp. 110–111), Trejo (2012). 12. The debate on the conceptual and operational difficulties of regime classification is extensive (among many others, see Schedler 2013, pp. 181–191). Wahman, Teorell, and Hadenius (2013) updated their dataset until 2010; see https://sites.google .com/site/authoritarianregimedataset. In this chapter, I use mean Freedom House and Polity values (with imputed Polity values for years with missing data) from the January 2015 Gothenburg Quality of Government (qog) dataset (Teorell et al. 2015). 13. The practice of distinguishing democratic and authoritarian regimes on the basis of rather arbitrary cutoff points along continuous scales such as Polity and Freedom House tends to create rather arbitrary regime transitions for regimes that lie close to the borderline. In the Appendix, some countries like Guatemala, Kenya, and Nicaragua quite clearly display such artificial regime transitions. 14. That is, I eliminated all 554 “conflict events” from our calculations whose “first issue” was recorded as “pro-government” (category 11 of scad variable “Issue 1”). 15. That is, we eliminated the following North African country years from the dataset (figures in parentheses indicate the number of protest demonstrations registered by scad in the corresponding year): Algeria 2011 (18) and 2012 (12), Egypt 2011 (33) and 2012 (177), Libya 2011 (11) and 2012 (38), Morocco 2011 (24) and 2012 (26), and Tunisia 2011 (17) and 2012 (54). A previous version of this chapter (available at request from the author) included these regime years. Their inclusion, though, does not significantly alter either my descriptive results (patterns of protest) or my estimation of causal effects (sources of repression). It only inflates average numbers of annual protest demonstrations under authoritarianism (from 3.2 to 4.5). Failing to exclude these extraordinary events would have given unfair advantage to my claim about the normality of protest under dictatorship. 16. Since I am interested in repressive responses to conflict events, my event counts exclude “pro-government violence (repression)” (category 7 of scad variable “Etype”).

74

Andreas Schedler

17. These differences in relative frequencies are statistically significant (Chi2 = 367.6, p = .000). 18. Of course, annual regime data that indicate the nature of political regimes at the end of each calendar year do not match well with event data that record precise start and end dates of conflict events. In years of regime change, I attribute all conflict events to the new regime, regardless of the date of regime change. 19. Of course, very small events are most likely to go unreported by international news agencies. 20. Original scad data are more disaggregate than mine. I fuse “organized” and “spontaneous” demonstrations. My aggregate category of “economic issues” includes scad items 2 (economy, jobs), 3 (food, water, subsistence), 4 (environmental degradation), 7 (education), and 12 (economic resources/assets). My “political regime” category aggregates items 1 (elections) and 10 (human rights, democracy). The variable “minority discrimination” merges items 5 (ethnic discrimination, ethnic issues) and 6 (religious discrimination, religious issues). The residual category of “other or unknown issues” contains items 11 (pro-government), 13 (other), and 14 (unknown, not specified). 21. Throughout this chapter, I have been treating authoritarian regimes as a homogenous category, ignoring the profound differences between authoritarian subtypes that have animated the study of authoritarian regimes over the past years (see, e.g., Geddes 1999; Hadenius and Teorell 2007; Schedler 2013). Yet quite obviously, different types of authoritarian regimes may differ in the level and nature of collective contention they engender as well as in the either conciliatory or repressive treatment they apply to contentious challenges. While I leave these questions to future research, in a previous version of this chapter, I did compare responses to peaceful demonstrations by democracies and four authoritarian regime subtypes (monarchies, military regimes, single-party regimes, and electoral autocracies). As I found, these subtypes of authoritarian regimes do not differ systematically among each other in the relative frequency of their repressive responses to demonstrations. They only differ from democratic regimes. All authoritarian subtypes (except single-party regimes) exercise nonlethal repression more often than democracies. Surprisingly, though, only military and electoral authoritarian regimes resort more often to lethal violence than democracies do. Note, however, that my coarse, binary measure of “lethal repression” (which registers the occurrence of one death or more due to the repression of a protest event) may hide wide variation among regimes in the actual number of fatal victims produced by state repression. 22. Given the ordinal nature of the scale of no repression, nonlethal repression, and lethal repression (my dependent variable), ordinal regression would have been the method of choice. Yet the required assumption of proportional odds was met only for democracies, as assessed by a full likelihood ratio test comparing the residual of the fitted location model to a model with varying location parameters (Chi2 [10]= 8.1, p = .610), but not for autocracies (Chi2 [10]= 46.4, p = .000). 23. Rather than logistic regression coefficients, which are hard to interpret, the table reports odds ratios (eb), which indicate the extent to which the odds of occurrence of the outcome variable change if the independent variable undergoes a oneunit change. Close to 1 they tell us that nothing changes, below 1 they indicate decreases, above 1 they indicate increases in the dependent variable. In democracies, the baseline odds for nonlethal repression are .182 (15.4/84.6) and for lethal repression .045 (4.3/95.7). Under authoritarianism, the baseline odds for nonlethal repression are .369 (27.0/73.0) and for lethal repression .085 (7.9/92.1).

4 Force Versus Institutions in Authoritarian Politics Milan W. Svolik

Xi Jinping, the “paramount” leader of China, assumed that post in 2012 with an authority that may appear curiously circumscribed for a dictator. He is expected to serve no more than two five-year terms and be accountable to a set of institutions within the Communist Party of China that balance major political coalitions as well as regional and organizational interests within the Chinese political system (C. Li 2010). Members of these institutions—primarily the Standing Committee of the Politburo and the full Politburo—will themselves be similarly constrained by term limits and mandatory retirement age provisions (A. L. Miller 2011; C. Li 2016). Xi inherited a political machinery engineered in the 1980s by Deng Xiaoping. Under Deng’s leadership, the newly revised Constitution of the People’s Republic of China (hereafter PRC constitution) prohibited certain officials from serving concurrently in more than one leadership post (Baum 1997), adopted mandatory retirement ages at various levels of the government hierarchy (Manion 1992; A. L. Miller 2008), and limited tenure at top government posts to two consecutive five-year terms (Baum 1997). At the same time, informal norms developed according to which analogous term limits and retirement age provisions applied to members of key party bodies (A. L. Miller 2008). China’s term limits and retirement provisions raise a number of questions for the study of authoritarian politics: Do such institutions contribute to the stability of dictatorships and, if so, then how? Are these institutions effectively binding Chinese leaders or are they mere reflections of a factional balance of power within China’s political system? And why would any authoritarian leader accept formal institutional constraints on his or her authority? These questions point to a central theoretical puzzle in the study of dictatorships: Can institutions serve any politically consequential role in a political system that is ultimately governed by force? In this chapter, I approach this puzzle by contrasting authoritarian and democratic politics and highlight how differences between the two shape 75

76

Milan W. Svolik

and limit our efforts to understand the role of institutions in authoritarian stability. I suggest that authoritarian politics is distinctive in two key ways: first, in dictatorships, no independent authority has the power to enforce agreements among key actors; and, second, in authoritarian politics, violence is the ultimate arbiter of political conflicts. These features result in a number of challenges to theory building, inference, and measurement in the study of authoritarianism. I illustrate these challenges by discussing the potential and the limits to the role in authoritarian stability of executive term limits—with a special attention to the case of China. More generally, I emphasize the complex and sometimes complementary contribution to authoritarian stability of shared interests, brute force, and institutional incentives. Institutions and Authoritarian Stability Consider the first of the two differences between authoritarian and democratic politics that I just previewed: unlike democracies, dictatorships lack an independent authority with the power to compel key actors to comply with their commitments. Authoritarian high courts, for instance, although de jure supreme, are de facto subservient to the incumbent, rarely ruling against the rulers.1 This is because the presence of a formal authority with the power to bind key actors in dictatorships would imply a check on the very powers that many of these actors aim to acquire. Whether it is the regime’s promise to play fair in elections, the leader’s promise to share power with allies, or the repressive agents’ promise to remain loyal in the face of mass opposition, in authoritarian regimes neither can be realistically expected to be enforced by an independent third party. As a result, commitment problems abound: pacts, rules, and even constitutions agreed on at one point may be broken later, when they become inconvenient. This concern is compounded by the prominent role that violence plays in resolving political conflicts in authoritarian politics. By my count, about two-thirds of all leadership changes in dictatorships between 1946 and 2008 were nonconstitutional—they departed from official rules or established conventions. Furthermore, almost one-half of all leadership changes involved the military, and about one-third of them were accompanied by overt violence.2 Hence, in authoritarian politics, the option of violence never seems to be off the table: political conflicts may be, and indeed frequently are, resolved by brute force. Thus, when formal rules and institutions appear to govern authoritarian politics, it may not be because of their binding power, but because the alternative of resolving political conflicts by brute force looms in the background.

Force Versus Institutions in Authoritarian Politics

77

These differences between authoritarian and democratic politics imply a number of distinct challenges for the study of institutions in authoritarian politics. Consider theory building. When I say that dictatorships lack an authority with the power to enforce agreements among key actors and that violence is the ultimate arbiter of conflict in authoritarian politics, I am not suggesting that—because of these two features—all dictatorships resolve conflict violently, no promises are ever kept, and formal institutions are irrelevant. Neither am I suggesting that the exact opposite holds under democracy. Rather, I propose that the lack of an independent authority with the power to enforce agreements and the pervasive use of violence critically limit the assumptions that we can reasonably make when we build explanations of authoritarian politics. In the study of democratic politics, institutions and rules that presumably allocate power can be realistically expected to do so. When Gary Cox (1997) studies the coordination dilemmas that electoral systems create for voters and parties, he can safely assume that the rules that govern electoral competition indeed do so. When Michael Laver and Norman Schofield (1990) study the politics of coalition governments, they need not consider the possibility that one faction within the legislature would eliminate another by force.3 Such a “partial equilibrium” analysis is warranted because the relevance of electoral and constitutional rules for allocating power in democracies is rarely in question. By the definition of democracy, any government that would resolve conflicts by violence or circumvent a major constitutional provision would no longer be considered democratic. In the study of authoritarian politics, by contrast, compliance with institutions is as much of a puzzle as are the consequences of those institutions. When it comes to theory building, explanations of authoritarian politics must therefore examine the “full” rather than the “partial” political equilibrium: we must explain not only the political consequences of rules and institutions but also why, given their consequences, key actors have an incentive to comply with them. Put in the jargon of contemporary political science, both the behavior of key actors and the institutions that presumably govern it must be self-enforcing.4 When it comes to the role of political institutions in the stability of dictatorships, the above discussion implies that their political relevance cannot be taken at face value. This does not amount to saying that institutions are inconsequential. Institutions may have the potential to facilitate power sharing among authoritarian elites (Boix and Svolik 2013) or the co-optation of the opposition (Gandhi 2008). Yet because for every institutional resolution of a political conflict there is a crude alternative in which brute force plays a decisive role, the limits to the potential of institutions to contribute

78

Milan W. Svolik

to authoritarian stability may critically hinge on the distribution of power and the congruence of interests among key actors. When and Why Do Institutions Bind Dictators? The Case of Executive Term Limits Consider term limits. This is an institution that is frequently studied in democracies, but rarely in dictatorships.5 This is in spite of the fact that executive term limits have been evoked as the cornerstone of authoritarian stability in several prominent cases and, equally intriguingly, that the circumvention of term limits played a prominent role in the rise of a number of autocrats. In the era of the Institutional Revolutionary Party (PRI) in Mexico (1929−2000), the principle of “no reelection” was credited with establishing a mechanism that prevented the emergence of the next Porfiriato and, thus, stabilizing the PRI’s seventy-one-year “perfect dictatorship.”6 Yet in many more cases, term limits were abandoned or circumvented as soon as they became an obstacle to the rise of an ambitious autocrat. The Chilean junta that came to power in the coup of 1973 was initially supposed to govern by unanimous consent and its presidency was to rotate among its four members.7 Soon, however, Pinochet came to dominate: in 1974, he compelled other members of the junta to appoint him president, replaced unanimous decisionmaking by a majority rule, and foreclosed any further considerations of rotation of the presidency. In 1978, Pinochet expelled from the junta Gustavo Leigh, the air force representative and his most vocal opponent. From that moment on, according to Genaro Arriagada (1988, p. 37), Pinochet began to act as “the de facto, if not the de jure, Generalissimo of the Armed Forces.”8 How come term limits prevented the emergence of a personal dictatorship in Mexico, but failed in so many other cases? Consider again the case of China. Term limits have been a central feature of the political machinery that has governed Chinese leadership politics since Deng Xiaoping’s reforms in the 1980s. After Deng emerged victorious from the struggle over the succession to Mao Tse-tung in 1980, he supervised an official appraisal of Mao’s leadership. Adopted in 1981, the Resolution on the Party History condemned Mao’s leadership for the “over-concentration of Party power in individuals,” “the development of arbitrary individual rule,” and the maintenance of a “personality cult” (MacFarquhar and Schoenhals 2006, p. 458). A key section of the Resolution on Party History reads: “[We] failed to institutionalize and legalize inner-Party democracy . . . we drew up the relevant laws but they lacked due authority. This meant that conditions were present for the over-concentration of Party power in

Force Versus Institutions in Authoritarian Politics

79

individuals and for the development of arbitrary individual rule and the personality cult in the Party.” In a similar vein, the revised 1982 Constitution of the People’s Republic of China declared that “no party member, whatever his position, is allowed to stand above the law or . . . make decisions on major issues on his own” (Baum 1997, p. 348). As a symbolic gesture of departure from Mao’s leadership style, the simultaneously revised party constitution abolished the post of party chairman that Mao held, and Deng himself avoided any titular confirmation of his powers (MacFarquhar 1997b). Consistent with the stated purpose of these reforms—that is, to prevent a return to the arbitrary autocratic leadership that prevailed under Mao—key party bodies began meeting with “metronomic regularity” (A. L. Miller 2008, p. 62), according to provisions of the new party constitution. At the same time, the 1982 PRC constitution limited the tenure in leading government posts, including the presidency, premiership, and chairmanship of the National People’s Congress, to two consecutive fiveyear terms (Baum 1997). This is a political system that differs markedly from the one in which Deng rose to prominence after Mao’s death. Toward the end of Mao’s life, institutions of collective leadership within the party and the government either were abandoned or, when they did survive, were staffed by Mao’s courtiers and governed by his personal authority rather than formal rules.9 There were no formal restrictions on lifelong tenure of political leaders. In fact, the expectation had been that, barring a political purge, leading officials would retire only when incapacitated by old age (Manion 1992; Baum 1997). Although Deng Xiaoping’s personal authority also transcended the binding power of the institutional reforms that he initiated, these have effectively constrained subsequent generations of Chinese leadership. Deng’s successor, Jiang Zemin, at first politically exploited mandatory retirement age provisions by invoking them to retire opponents within the leadership in 1997. Yet the same term and age provisions eventually came to limit Jiang’s own time in office when he was compelled to step down from his posts in the early 2000s (Huang 2008). The tenure of Jiang’s successor, Hu Jintao, as well as the selection of Xi Jinping has in turn been fully governed by the set of rules introduced under Deng and first implemented under Jiang (A. L. Miller 2011). How do term limits and retirement-age provisions help to perpetuate China’s dictatorship? At first sight, the implications of term limits appear obvious: a term limit on a leader’s tenure amounts to a line in the sand. Its violation is easily observable and, thus, reveals a leader’s true ambitions to both those within and outside the regime. It is an alarm bell that ensures that

80

Milan W. Svolik

a leader’s attempts to usurp power at the expense of others will be caught before it is too late. More conceptually, term limits facilitate the monitoring of a dictator’s commitment to share power with the rest of the political elite. From this perspective, China’s term limits and retirement age provisions suffer from one important deficiency: unlike the limits on government posts, term limits and retirement age provisions for leadership posts in the party and the military emerged by convention and precedent rather than an explicit formal adoption. Although the 1982 PRC constitution did stipulate that leadership-party cadres were “not entitled to lifelong tenure” (Manion 1992, p. 11; A. L. Miller 2008, p. 63), even Deng lacked the weight to institute any specific constraints on party or military veterans with revolutionary credentials (Vogel 2011). Instead, informal rules regulating term limits and mandatory retirement age gradually developed for leading party posts. As a first step Deng appealed to the generation of revolutionary veterans, to which he belonged, to create room for a new generation of leaders and attempted to smooth their retirement from official posts by creating a new consultative body, the Central Advisory Commission (Baum 1997; Vogel 2011). Indeed, this generation of elites—including Deng—retired almost entirely from the highest party posts by 1987. An informal norm barring the reappointment of Politburo members past the age of seventy emerged at the fifteenth Party Congress in 1997, when Jiang Zemin forced one of his competitors, Qiao Shi— who was seventy-five years old at the time and a member of the Politburo Standing Committee and chairman of the National People’s Congress—into retirement by invoking his advanced age (A. L. Miller 2008; Dittmer 2002; Huang 2008; Nathan 2003). A test of the binding power of both formal term limits and informal retirement age norms came at the time of Jiang Zemin’s expected retirement as China’s paramount leader. Since Jiang’s tenure, political leadership of the PRC came to be associated with three posts: the general secretary of the Communist Party, the chairman of the Central Military Commission, and the president of the People’s Republic of China. Only the last and most ceremonial of these, however, was subject to explicit constitutionally mandated term limits. The informal nature of rules concerning retirement from leading party or military posts opened the door for speculation about their exact nature (Bo 2007) and for maneuvering by Jiang, who in fact campaigned to retain the remaining two posts (Shirk 2002). Eventually, Jiang stepped down from the post of party general secretary in 2002 and from the presidency in 2003, in accordance with informal norms and the formal constitutional two-term limit, respectively. However, Jiang did attempt to hold on to the chairmanship of the Party and State Central Military Commissions for another term. When he retained

Force Versus Institutions in Authoritarian Politics

81

those posts, Jiang claimed to be keeping them because of the “complicated international situation” and the “pressing demands of army-building” (Mulvenon 2005, p. 5) and to ensure “the smooth transition from the old to the new generation [of leaders]” (Mulvenon 2003b, p. 21). Jiang also appealed to a precedent set by Deng Xiaoping, who held on to the same posts for several years after he retired from the Politburo (Vogel 2011). The contrast between explicit limits on tenure in government posts and tacit norms about limits on party and military tenures highlights the benefit of formal institutionalized rules in authoritarian governance. Because term limits and retirement age provisions for leadership posts in the party and the military were based on implicit expectations rather than formal rules, they allowed for speculation about their exact nature. The ambiguity allowed for by such informal norms not only creates incentives for maneuvering to circumvent them, but also fosters suspicions that such attempts are taking place—even when they may not be. In turn, tacit unwritten norms or expectations entail the risk of unnecessary regime-destabilizing confrontations. Jiang Zemin’s reluctance to step down from all of his posts led to precisely such suspicions. The People’s Daily announcement of the leadership change that occurred at the sixteenth National Congress of the Chinese Communist Party in 2002 placed Jiang’s photograph and resume above the newly elected general secretary Hu Jintao’s, leaving the latter barely visible in subsequent media announcements.10 Observers speculated that Jiang intended to use his Central Military Commission posts to exert continued political influence from behind the scenes.11 These suspicions along with questions about Hu’s actual political authority and signals of disagreements between the two leaders on key policy issues raised concerns about the potential instability that could attend a pronounced leadership struggle.12 Following growing criticism from within the party and the military (Huang 2008; Mulvenon 2003a, 2003c), Jiang eventually resigned from both Central Military Commissions by 2005, before the expiration of his term.13 In his resignation letter, Jiang stated that he was stepping down because it served “the long-term peace and stability of the party and state” and ensured “the institutionalization, standardization, and proceduralization of the succession of new high-ranking party and state leaders.”14 Can Xi Jinping be expected to follow the same institutionalized rules of collective leadership that have governed the past two generations of Chinese leaders? While term limits may facilitate the monitoring of an agreement to share power among the political elite, that alone is not enough. Recall my earlier claim that dictatorships lack an independent authority with the power to enforce agreements among key actors and that violence is never off the table. To be successfully deterred, a leader’s

82

Milan W. Svolik

attempt to consolidate power not only must be detected, but also credibly punished. The latter critically depends on the rest of the elite’s ability to remove the leader from office—by brute force, if necessary. As Mao put it, political power grows out of the barrel of a gun. After all, a notable feature of the institutionalization of collective leadership initiated by Deng Xiaoping is that it occurred only after Mao’s death and, according to most observers, failed to effectively constrain Deng himself (see, e.g., Baum 1997). In fact, formal institutions for collective leadership already were put in place under Mao, but their political relevance quickly eroded. The eighth Party Congress in 1956 established the Politburo Standing Committee as the highest institution of collective leadership, charged the party’s Secretariat and several “small leadership groups” with implementing the Politburo Standing Committee’s decisions, and appointed members of institutions representative of the PRC’s broader leadership (A. L. Miller 2008). These institutions nevertheless failed to prevent Mao’s arbitrary exercise of power, purges of recalcitrant party members, and their own eventual decline into political irrelevance. After 1959, key party bodies met only sporadically (A. L. Miller 2008) and, following the launch of the Cultural Revolution in 1966, most became inconsequential until Mao’s death a decade later (MacFarquhar and Schoenhals 2006). Put simply, institutions in dictatorships cannot be expected to constrain leaders unless backed by a credible threat of force. This is why, in China, formal institutions of collective leadership successfully governed the tenures of Jiang Zemin and Hu Jintao, but failed to constrain Mao Tsetung and Deng Xiaoping. The latter two leaders commanded personal authority grounded in revolutionary achievements and charismatic personalities that eclipsed any of their contemporaries. As evidenced by the Cultural Revolution, Mao in particular was capable of annihilating any opposition—whether individual or institutional—by exploiting his almost divine popular status (MacFarquhar and Schoenhals 2006). By contrast, Jiang and Hu possessed neither charisma nor revolutionary credentials; they instead owed their careers to bureaucratic and administrative skills. Both have been regarded as firsts among equals within two evenly balanced political coalitions in the Chinese leadership and, rather than dominating those coalitions, they depended on their support (A. L. Miller 2004; C. Li 2010; Huang 2008). The successful adoption of term limits under dictatorship thus requires—at least initially—a permissive balance of power among the regime’s key players. Yet once in place for a sufficiently long amount of time, term limits may successfully perpetuate precisely the kind of balance of power among the elite that allows for credible power sharing. A term limit does not merely place a sharp limit on a leader’s time in office. The

Force Versus Institutions in Authoritarian Politics

83

political retirement of an authoritarian leader typically implies the departure of an entire generation of officials. Thus, once in place, term limits coordinate the political horizons of multiple generations of authoritarian elites: they encourage ambitious political clients to invest their careers in their own generation of leaders rather than the current, but only temporary, cohort of elites. The resulting incentives help us understand why conditions required for the establishment of politically consequential term limits may differ from those needed for their perpetuation. Credible term limits emerge when power is distributed evenly among authoritarian elites: only then can the first generation of leadership facing term limits be realistically expected to step down and, thus, initiate the expectation of future alternations in power among the army of clients at lower ranks of the political hierarchy. In the case of China, the effective adoption of term limits was made possible by the even balance of power that emerged within the Chinese political elite after the departure of Mao’s and Deng’s revolutionary generation. But once term limits are in place, they coordinate over time the political investments of a large number of clients and, thus, reproduce the existing balance of power.15 In turn, a dictator who is intent on overstaying an established term limit must anticipate opposition not only from his or her heir apparent, but also from the multitude of clients who have invested their careers in patrons belonging to the next generation of leadership. This is why PRI-era Mexicans were able to retire their dictators every six years, as Frank Brandenburg (1964) eloquently puts it. Challenges to the Empirical Study of Authoritarian Institutions My discussion so far has focused on the challenges that two distinctive aspects of authoritarian politics—the lack of an independent authority with the power to enforce commitments among key actors and the role of violence as the ultimate arbiter of conflicts—present for theory building. The ensuing requirement that my explanations account for both the consequences of institutions and the compliance with them also points to a number of challenges for empirical research on authoritarian politics. Most prominently, the need to think of institutions as self-enforcing equilibria significantly limits the number of factors that can be considered exogenous. In democracies, major constitutional provisions—say whether the executive is bound by a term limit—can be considered both binding and given, at least in the short run. My discussion of term limits in dictatorships, by contrast, suggests that compliance with them is at least initially

84

Milan W. Svolik

contingent on a permissive balance of power among the authoritarian elite—and, thus, endogenous to it. An empirical study of term limits in dictatorships that would ignore this endogeneity might naively conclude that their adoption would automatically prevent the emergence of personal autocracy in any dictatorship. The distinctive features of authoritarian politics thus amplify concerns about the endogeneity of presumed causes that we frequently encounter in other subfields of political science.16 The two distinguishing features of authoritarian politics—the lack of an independent authority with the power to enforce agreements among key actors and the pivotal role of violence—also complicate the measurement of the institutional makeup of dictatorships. A major dilemma in authoritarian politics is not only whether institutions matter for the conduct of authoritarian politics, but also which institutions and leaders should matter in the first place. By now, for instance, it has become apparent that the effective head of the Russian government is neither the president of the Russian Federation nor its prime minister. Rather it is Vladimir Putin—regardless of the official post that he confers on himself. Putin’s political transubstantiation has parallels across the world of authoritarian politics. The Great Benefactor Rafael Trujillo formally led the Dominican Republic during only eighteen of the thirty-one years of his de facto rule. Fearing criticism by the United States and the Organization of American States, he interspersed his years in power with four pliant substitutes, including his brother Héctor. Meanwhile, Deng Xiaoping, who is universally regarded as the paramount leader of China between 1978 and 1992, avoided any titular confirmation of his powers in an attempt to distance himself from his domineering predecessor (MacFarquhar 1997a; Vogel 2011). The nominal resemblance of many institutions in dictatorships—especially legislatures, parties, and even some elections—to institutions in democracies therefore provides only limited guidance for their conceptualization in authoritarian politics. Consider again the case of term limits. When political scientists study term limits in the context of democratic politics, their focus most often is on how term limits affect electoral accountability and legislative representation.17 By contrast, my earlier discussion suggests that the primary role of term limits in dictatorships is to serve as an alarm bell for usurpation of power by any single leader or faction—a very different and uniquely authoritarian concern. Likewise, rather than coordinate the political activities of like-minded citizens (see, e.g., Aldrich 1995), regime parties in dictatorships appear to instead co-opt the most capable and opportunistic among the masses to strengthen the regime. Thus, while many institutions in dictatorships nominally mirror their democratic counterparts, their political ends are distinctively authoritarian.

Force Versus Institutions in Authoritarian Politics

85

The questionable relevance and function of many political institutions in dictatorships is compounded by their diversity. By most definitions, the world of authoritarian politics ranges from cases like PRI-era Mexico, whose institutions posed as democratic in form, to traditional polities like the neofeudal Saudi Arabia, to idiosyncratic regimes like Iran with its overlapping system of republican and religious authorities, to contemporary China with its Leninist institutional hardware.18 As Barbara Geddes concludes, “Different kinds of authoritarianism differ from each other as much as they differ from democracy” (1999, p. 121). This institutional diversity obtains partly because dictatorship is a residual category that contains all countries that do not meet established criteria for democracy and partly because of dictatorship’s richer and longer pedigree. Whereas democracy has historically followed a few institutional blueprints, dictatorship combines institutional models from multiple centuries and levels of development. Geddes’s (1999) classification of dictatorships into personalist, military, and single-party types is one of the first and most productive efforts to map and organize the institutional makeup of authoritarianism. The wave of substantive research on authoritarian politics as well as competing data collection efforts that followed Geddes’s original work is evidence of the catalyzing effect that publicly shared data can have on comparative political research.19 In spite of this progress, however, our large-N data on authoritarian politics are mostly confined to the post−World War II period. This temporal limitation may be significantly biasing our conclusions about the political organization of dictatorships, the process of regime change, and the consequences of authoritarian institutions. Consider how the Cold War affected the political organization of dictatorships: in much of Eastern Europe, Asia, and Africa, the institutional infrastructure of authoritarianism followed the Leninist single-party blueprint. When it was not directly dictated from Moscow, the Leninist blueprint was encouraged as a part of the Soviet package for prospective third world clients and even managed to inspire a few Ba’athist copycats. Meanwhile, the emergence of the highly bureaucratic, conservative, and exclusionary military regimes among many US clients was a parallel reaction to Vladimir Lenin’s, Mao Tse-tung’s, and Fidel Castro’s improbable revolutions.20 In turn, our conclusions about the distribution of regime types, their longevity, and the process of regime change may be unduly shaped by the limits of our data. The emergence and demise of single-party and military dictatorships may have as much to do with Cold War geopolitics as with the intrinsic features of these regimes’ political organization.21 Meanwhile, when Larry Diamond (2002), Steven Levitsky and Lucan A. Way (2010), Valerie Bunce and Sharon L. Wolchik (2011), and Andreas Schedler (2013)

86

Milan W. Svolik

observe the rise of hybrid regimes after the end of the Cold War, they may be correctly identifying a shift in the distribution of regimes in the post−Cold War period but at the same time mischaracterizing its unprecedented nature. By most historical accounts, the late nineteenth century and the first half of the twentieth century were replete with competitive dictatorships and defective democracies (see, e.g., Knutsen, Møller, and Skaaning 2016). Hence, rather than being new and exceptional, hybrid regimes may be the historical norm and, rather than representative, the sharply delineated single-party and military dictatorships of our existing data may be an aberrant institutional by-product of the Cold War. Conclusion I have argued that—especially when compared to democratic politics— authoritarian politics takes place under distinctively toxic conditions. In dictatorships, no independent authority has the power to enforce commitments among key actors and violence is the ultimate arbiter of conflicts. While neither of these concerns is unique to authoritarian politics, their combination and severity amplifies many of the challenges to theory building, inference, and measurement that we encounter in other areas of political science. In theory building, the lack of an authority with the power to enforce commitments and the pervasive use of violence place a high bar on what reasonably counts as an explanation. I have suggested that when we propose explanations of authoritarian politics, we must examine the full rather than the partial political equilibrium—we must explain why both behavior and the institutions that presumably govern it are self-enforcing. This is because for every institutional resolution of a political conflict under dictatorship, there is a crude alternative in which force plays a decisive role. In turn, when we evaluate our claims empirically, we cannot take authoritarian institutions as given – we must instead confront concerns about their endogeneity. Meanwhile, the questionable relevance of formal political institutions in dictatorships results in distinct challenges to measurement and data collection. Because of the potential disconnect between de jure institutions and de facto power, which authoritarian institutions and leaders matter is frequently far from obvious. In many dictatorships, the leader who gives orders may not reside in the presidential palace, but rather across the street from it.22 A final challenge arises out of the limited scope of our large-N data on authoritarian politics. Today’s oligarchs of the United Russia Party and the antiliberal populists of Latin America may be closer to the aristocratic republics of the nineteenth century and the imperfect democracies of the interwar years than the Leninist single parties and reactionary juntas that

Force Versus Institutions in Authoritarian Politics

87

we so often encounter in our existing data. The latter may be by-products of the Cold War and, thus, distorting our image of authoritarianism. Notes 1. Ginsburg and Moustafa (2008), Ginsburg and Simpser (2014). 2. This data can be accessed at http://campuspress.yale.edu/svolik/the-politics -of-authoritarian-rule/. 3. Acemoğlu, Egorov, and Sonin (2008) examine the challenges to authoritarian stability created by precisely this possibility. 4. Przeworski (1991), Geddes (1999), and B. Bueno de Mesquita et al. (2003) were among the first to emphasize the need for an “equilibrium analysis” of institutions in their study of self-enforcing democracy, authoritarian breakdowns, and selectorate theory, respectively. For a review of formal models of authoritarian politics, see Gehlbach, Sonin, and Svolik (2016). 5. For exceptions, see Baturo (2014), Frantz and Stein (2013). 6. I am borrowing Mario Vargas Llosa’s oft-quoted characterization of Mexico’s PRI regime. Remarks by Vargas Llosa during a debate on Mexican television station Televisa on August 30, 1990; Vargas Llosa, “México es la dictadura perfecta,” El País, September 1, 1990. 7. The Argentine junta of 1976–1983, motivated by the marginalization that the military experienced after bringing General Juan Carlos Onganía to power a decade earlier, came up with similar rules—including the rotation of the presidency and the sharing of government posts among the multiple branches of the military—only to abandon them amidst conflicts over the control of over the presidency and key ministries (Fontana 1987; Remmer 1989). 8. For an account of Pinochet’s consolidation of power within the junta, see also Constable and Valenzuela (1993), Spooner (1999). According to Barros (2002), Pinochet never attained the absolute dominance commonly attributed to him. 9. See MacFarquhar (1974/1997), MacFarquhar (1997a), MacFarquhar and Schoenhals (2006). 10. See Joseph Kahn, “Officially, Jiang Is History; in News, He’s Still on Top,” New York Times, November 17, 2002. 11. See Erik Eckholm, “Chinese Leader Gives Up a Job but Not Power,” New York Times, November 16, 2002. 12. See Joseph Kahn, “China’s 2 Top Leaders Square Off in Contest to Run Policy,” New York Times, September 2, 2002; Erik Eckholm, “China’s Leader Won’t Hold On, Anonymous Author Says,” New York Times, September 5, 2002; Joseph Kahn, “Analysts See Tension in China Within the Top Leadership,” New York Times, July 1, 2003; Joseph Kahn, “Former Leader Is Still a Power in China’s Life,” New York Times, 16 July 2004. 13. See Joseph Kahn, “China Ex-president May Be Set to Yield Last Powerful Post,” New York Times, September 7, 2002; Joseph Kahn, “Hu Takes Full Power in China as He Gains Control of Military,” New York Times, September 20, 2002. 14. “China Publishes Jiang Zemin’s Letter of Resignation,” Xinhua News Agency, September 19, 2004 (as cited in Mulvenon 2005, p. 2). 15. Similar incentives may account for why most aging dictators avoid anointing a successor: once the sucessor’s identity is clear, ambitious clients may switch

88

Milan W. Svolik

their loyalties to the successor, effectively retiring the incumbent leader before his time is up. 16. On endogeneity concerns in comparative politics and the study of authoritarianism, see also Pepinsky (2014), Przeworski (2009). 17. See, for example, Besley and Case (1995), Carey (1996). 18. I am paraphrasing McGregor’s (2010) observation that contemporary China is still running on “Soviet hardware” (p. 12). 19. As of March 2017, Geddes’s (1999) article, which is the primary reference for her data, had been cited more than 1,400 times. Subsequent and alternative sources of data on authoritarian politics include Gandhi (2008), Geddes, Wright, and Frantz (2014), Magaloni and Kricheli (2010), Przeworski et al. (2000), Svolik (2012). 20. See also, Boix (2011). For a parallel argument about how the Cold War affected the nature of civil wars, see Kalyvas and Balcells (2010). 21. On authoritarian parties see, for example, Boix and Svolik (2013), Brownlee (2007), Geddes (2008), Gehlbach and Keefer (2010), Greene (2007), Levitsky and Way (2010), Magaloni (2006), Magaloni and Kricheli (2010), Malesky and Schuler (2010), Smith (2005), Svolik (2012, chap. 6). On military dictatorships see, for example, Debs (2009), Geddes (1999), Remmer (1989), Rivero (2011), Rouquié (1987), Svolik (2012, chap. 5). 22. I am paraphrasing a saying about Plutarco Calles’s continuing influence after he resigned from the Mexican presidency; see Krauze (1997, p. 430).

5 The Military and Autocratic Regimes: Roles of the Armed Forces Aurel Croissant, Tanja Eschenauer, and Jil Kamerling Politics in any society involves the management of coercive power. Therefore, the interactions between the armed forces and the structures, organizations, and actors of the political system are crucial for the stability and survival of all forms of political regimes (Feaver 2003). However, consolidating political control of the military (i.e., the subordination of the military under the authority of a regime’s political leaders) is especially arduous in nondemocratic regimes. Autocracies tend to enjoy less popular support and legitimacy than democracies, and must therefore lean on the agents of coercion of which the military is the largest and most powerful (Barany 2012). While most dictatorships rely on the police and specialized internal security agencies for everyday repression, the military is the final guarantor of regime security as it is the only coercive institution capable of defeating a mass-based, mass-organized, and potentially violent opposition movement (Skocpol 1979). Hence, no authoritarian government can hold onto the reins of power without the expressive approval of military leaders. This became evident in the 2011 crises of Arab autocracies. The capacity of dictators to survive the challenges of mass mobilization, regime collapse, and civil war during the Arab Spring cannot be understood without the regime leader-military nexus (Albrecht, Croissant, and Lawson 2016). However, a strong military is a doubleedged sword for autocrats. A powerful military is more effective in repressing political conflicts between the ruling elite and the masses, but may at the same time create threats to regime survival that emerge from within the regime coalition (Croissant and Kuehn 2015). Moreover, powerful armed forces are in a much better position to demand greater political and economic concessions in exchange for their role in maintaining the regime (Acemoğlu and Robinson 2006). Thus, most scholars agree that the relationship between political elites and the military establishment has important consequences for the internal dynamics and the survival of autocratic regimes. 89

90

Aurel Croissant, Tanja Eschenauer, and Jil Kamerling

However, scholars have thus far paid insufficient attention to politicalmilitary relations and the different political roles of the military in contemporary autocracies. Instead, the literature has mainly focused on two attributes of this relationship, which seriously limit our understanding of the political role of the military in autocracies. First is the preoccupation with military regimes; that is, the “rule by a group of high-ranking officers who can limit the dictator’s discretion” (Geddes, Wright, and Frantz 2014, p. 147). The focus on overt and direct military rule by a group of military officers—often institutionalized through a ruling council or junta—is problematic as this form of authoritarian rule has become less frequent since the 1980s. According to data by Barbara Geddes, Joseph Wright, and Erica Frantz (2014), in 1978 every fifth nondemocracy worldwide qualified as a “military regime” (twenty-one out of ninety-seven). In 2010, only two of fifty-nine autocracies were military regimes. The dwindling numbers of military regimes could be a sign for considerable reduction in the political power, influence, and role of the military in contemporary authoritarian regimes. Empirical studies suggest, however, that these developments are not the result of military professionalization and a retreat from the political sphere. Rather, autocrats in Asia, Africa, and the Middle East often build mixed supporting coalitions in which civilian elites and social groups flank the security forces and the military in maintaining the status quo (Basedau and Elischer 2013; S. A. Cook 2007; Mietzner 2011; Roehrig 2013). Second, much of the current political science literature on politicalmilitary relations in authoritarian regimes studies military coups d’état. Undoubtedly, the coup d’état is the most dramatic form of military involvement in politics. Yet seizure of the reins of government is by no means the only way in which the armed forces influence politics (Huntington 1968; Zimmermann 1979). Another, and often more relevant for the long term, is the impact of military organizations on the political recruitment, socialization, and communication, and on the articulation and aggregation of political interests and demands (Lovell and Kim 1967). Furthermore, military interventions can range from brief and limited military incursions into political affairs to complete military control of the state, and may manifest themselves as blackmailing, obstruction, insubordination, going public, or mutinies (Finer 1962). In this chapter, we aim to shed light onto two sets of questions for comparative study of autocratic regimes. First, how can different roles of the armed forces in authoritarian politics be conceptualized and measured, and are there different patterns of military influence across autocracies? Second, do different political roles of the armed forces help explain why and how autocratic regimes slide into crises? More precisely, we are interested in whether certain military roles make autocracies more or less vulnerable to

The Military and Autocratic Regimes

91

crises and if autocratic regimes exhibiting different political roles of the military differ with regard to their chances to survive regime crises. The chapter is organized in five sections. First, we introduce a systematic analytical concept of political-military relations. This is based on previous work by Matthias Basedau and Sebastian Elischer (2013) and Aurel Croissant et al. (2010, 2013), and differentiates between two roles of the armed forces, military rulers and supporters. Second, we operationalize this concept in a number of measurable criteria and indicators. The complete dataset includes information on six variables of military influence aggregated as indexes of military rulers and supporters. Third, we present data for seventy-one authoritarian regime spells for the period 1999−2012 across different types of autocracies and regions of the world. Hereby, we demonstrate that focusing on regime type alone is not sufficient to identify the role of armed forces in authoritarian politics and stress the need for alternative measures. Fourth, we assess whether different roles of the military impact the onset and outcome of different types of authoritarian regime crises, resulting from “horizontal” (coming from within a dictator’s ruling coalition) or “vertical” (coming from below) threats (Schedler 2015, pp. 33−36). In the final section, we summarize our findings and discuss future research avenues. Political Roles of the Military: Theoretical Concepts and Operationalization In this section, we present our concepts of military rulers and supporters and lay out the operationalization and the aggregation rules for our indexes. Theoretical Concept

Despite the key role of the military for the inner working and survival of autocratic regimes, political science still lacks convincing concepts to conceptualize and measure the role of militaries between overt military rule and perfect subordination under civilian supremacy. This especially counts for those forms of political-military relations in which military elites have moved away from the direct exercise of political power and have assumed a guardian role in the wings of the political stage (“ruling but not governing”; Cook 2007). In these regimes, which are not classified as military regimes, the role of the military in selecting rulers or securing the status quo has not yet been systematically examined. Basedau and Elischer (2013) propose a concept of military influence to assess these roles of the military. According to them, the decrease in the

92

Aurel Croissant, Tanja Eschenauer, and Jil Kamerling

number of military coups in post-1980s sub-Saharan Africa is no indication of a military withdrawal from politics. Instead, these developments reflect new influential roles of the armed forces (Basedau and Elischer 2013, p. 358). The militaries of sub-Saharan Africa have become what Basedau and Elischer call military rulers and military supporters.

Military ruler. Regimes with a military ruler originate from military coups d’état or civil wars. Coup leaders or rebel leaders become key actors in the ruling coalition of the postcoup or postwar regime. “Rebel” regimes—that is, regimes that achieved control via insurgency—are included in this category (see also Lyons 2016; Wahman, Teorell, and Hadenius 2013). Multiparty elections can occur on a regular basis, but lack real competition (Basedau and Elischer 2013). Military supporter. Yet the military does not need to rule to remain

influential. Regimes that are heavily dependent on military support will offer their repressive agents privileges to ensure their compliance and secure their own survival. Being pivotal to the regime’s survival—and fully aware of this—the military uses the government’s need to its own advantage. As a veto actor it shapes political decisions to its will, retains considerable autonomy, and enjoys impunity (Basedau and Elischer 2013). In return, the armed forces lend their support to the regime and suppress antigovernment protests. Operationalization

Basedau and Elischer’s (2013) approach goes beyond the mere assessment of the military’s role by regime types and coups d’état. Their concepts propose a classification of the military’s function within the regime and not distinct regime types. Nonetheless, this approach is not without deficits, which we address in the operationalization of our indexes of military roles.1

Variables. We capture military role as two separate aggregated indexes: military ruler and supporter (Figure 5.1). Military rulers (mi_mruler) are composed of three variables: military origin (mi_origin), political leader (mi_leader), and minister of defense (mi_mod). A regime has a military origin when it results from a military coup or a victory in civil war, without legitimizing the new regime in free and competitive elections according to Polity IV (coding 0/1). We assume that the shadows of the past become less influential over time, and set the origin of a regime back to “not military” (0) if the civil war or the military coup happened more than a generation ago (i.e., twenty-five years ago).

The Military and Autocratic Regimes

93

Figure 5.1 Concepts Regime origin Military ruler

Political leader Minister of defense Veto power

Military supporter

Impunity Repressive agent

Political leader (mi_leader) measures the connection of the regime leader, the dominant political executive, to the armed forces. We differentiate between no member of the military/rebel leader (0), retired (0.8), or active (1) member of the military or rebel leader. We include rebel leaders based on the assumption that victorious guerrilla armies of a rebel movement will become the regime’s new military. Like mi_leader, the variable minister of defense (mi_mod) checks for direct military affiliation. If the defense minister is an active or retired member of the armed forces, mi_mod is coded 1. For coding, we consider only January 1 of a given year. References for these variables are the Database of Political Institutions (DPI; Beck et al. 2001), Freedom House Reports, Bertelsmann Transformation Index (BTI) Country Reports, and US Department of State Human Rights Reports (USHRR), supplemented by a variety of secondary sources. The concept of a military supporter (mi_msupport) is measured by the military’s veto power (mi_veto), its deployment as an agent of repression (mi_repress), and impunity for the members of the armed forces (mi_impun). These variables are dichotomous; we code them as 1 if the military possesses veto power, is employed as a repressive agent, or enjoys impunity, and 0 if not. Whether the military enjoys veto power is determined by the BTI Report questions 2.2 “effective power to govern” and 16.2 “anti-democratic actors.” We provide a score of 1 only if these reports explicitly mention the military as a relevant veto player. Military repression takes on the value 1 for all forms of de facto repression of political protests, separatists, and preemptive repression by official military actors subordinated to the government. The military enjoys impunity if it engages

94

Aurel Croissant, Tanja Eschenauer, and Jil Kamerling

in illegal activities and members of the military do not have to fear prosecution and conviction for human rights violations, hazing, and other criminal actions. This is also the case if prosecution must be considered highly disproportional to the number and severity of crimes committed. References are USHRR and other secondary sources.

Research sample. The complete version of the dataset comprises information on all regimes of the BTI sample, regardless of whether they are democratic or autocratic (Croissant, Eschenauer, and Kamerling 2016). The original dataset contains information on eight variables of military influence aggregated as indexes of military rulers and supporters for 120 regimes for the period 1999−2012. For the purpose of this book, we focus solely on autocratic regimes, resulting in a research sample of sixty-seven states.2 We gather data on a yearly basis and aggregate indicators over regime spells as defined by the Autocracies of the World dataset (Magaloni, Min, and Chu 2013), which shows the largest intersection with our research sample. This has two advantages. First, it allows for the possibility that the degree and form of military influence changes during our research period. Second, regime spells offer greater validity to assess the structural aspects of military influence—yearly data may vary considerably, even though the actual degree and manner of military influence did not change fundamentally. Overall, our sample of sixty-seven autocracies translates into seventytwo autocratic regime spells. The aggregation rules for all six indicators of military influence—origin, leader, minister of defense, veto power, repression, and impunity—are as follows. If the indicator displays military influence, a coding of 1, for less than 25 percent of the years of the regime spell it receives a score of 0; coding of 1 for more than 25 percent but less than 50 percent of the regime spell results in a score of 1; coding of 1 for 50 percent to 75 percent of the regime spell results in a score of 2; and coding of 1 for 75 percent or more of the regime spell results in a coding of 3.3 Following the theoretical concepts, the indexes of military influence, military ruler and supporter, are aggregated according to the following formulae based on Boolean algebra.4

mi_ruler = mi_supporter =

mi_origin * (mi_leader + mi_mod) (mi_veto * mi_repress) + (mi_veto * mi_impun) + (mi_repress * mi_impun)

We identify military rulers and/or supporters if the value of the respective index is equal to or bigger than 2. Yet it needs to be emphasized that these

The Military and Autocratic Regimes

95

indexes of military influence do not represent regime types. Rather, they reflect distinct forms of political-military relations, which are characterized by specific roles of militaries and their influence on the internal organization and leadership structure of a given political regime. The Role of the Military in Autocratic Politics Aside central and southeast Europe, and Latin America and the Caribbean, with zero and three regime spells, respectively, all world regions are more or less equally represented in our sample; ranging from ten in West and central Africa to nineteen in the Middle East and North Africa (MENA).5 Differences result from both the varying number of autocracies and the persistence of political regimes. We first compare our measures of military influence to the regime typology of Beatriz Magaloni, Eric Min, and Jonathan Chu (2013) and demonstrate the relevance of political roles of the armed forces for the assessment of political-military relations in autocracies (Figure 5.2). In line

Figure 5.2 Political Roles of the Military in Autocracies

Source: Magaloni, Beatriz, Eric Min, and Jonathan Chu. 2013. Autocracies of the World Dataset, 1950−2012. Version 1.0. Stanford: Stanford University.

96

Aurel Croissant, Tanja Eschenauer, and Jil Kamerling

with their data we find, first, a worldwide decline of military influence in autocratic polities. Second, and more importantly, our measures also detect military roles as rulers and supporters in regimes not classified as “military regimes” in Magaloni’s dataset. Military influence in the form of military rulers and supporters is more common than military regimes; that is, regimes where “effective control is primarily held by the military” (Magaloni, Min, and Chu 2013, p. 8). As expected, patterns of military influence do indeed offer valuable additional information to common regime typologies and broaden our understanding of political-military relations in autocracies.6 A first differentiation of ruling and supporting militaries shows military supporters to be considerably more common than rulers. We identify eighteen military rulers (25 percent) and thirty-three military supporters (46 percent; see Figure 5.3); that is, regime spells with a score of two or higher on the respective index. Across different regions, we find obvious differences in the degree of military influence (Table 5.1). Ruling militaries are most common in West and central Africa as well as in southern and East Africa, where they account for 33 percent of all regime spells. In contrast, none of the regime spells in Latin America has a ruling military while there are two spells with ruling militaries in the Middle East (Algeria, 1999−2012; Sudan, 2000−2009), post-Soviet Eurasia (Azerbaijan, 1999−2012; Tajikistan 1999−2012), and Asia (Myanmar, 1999−2011; Pakistan, 1999−2007). Figure 5.3 Distribution of Military Rulers and Supporters

The Military and Autocratic Regimes

97

Table 5.1 Regional Distribution of Military Rulers and Supporters BTI Regions

Latin America and the Caribbean West and Central Africa

Middle East and North Africa South and East Africa Post-Soviet Eurasia Asia and Oceania Total

Military Ruler 0 1

3 5.66% 4 7.55% 17 32.08% 10 18.87% 8 15.09% 11 20.75% 53 100%

0 0% 6 33.33% 2 11.11% 6 33.33% 2 11.11% 2 11.11% 18 100%

Note: BTI = Bertelsmann Transformation Index.

Total

3 4.23% 10 14.08% 19 26.76% 16 22.54% 10 14.08% 13 18.31% 71 100%

Military Supporter 0 1

2 5.26% 2 5.26% 9 23.68% 7 18.42% 9 23.68% 9 23.68% 38 100%

1 3.03% 8 24.24% 10 30.3% 9 27.27% 1 3.03% 4 12.12% 33 100%

The highest frequency of regimes that rely on military support (10) can be found in the MENA region (30.3 percent),7 followed by southern and East Africa, and West and central Africa with nine and eight spells, respectively. In Asia and Oceania four regime spells have military supporters, while we identify only one each in post-Soviet Eurasia and Latin America and the Caribbean (Venezuela, 2002−2012). These findings compare well to existing regional studies on politicalmilitary relations: while the military still exerts considerable influence on access to and persistence of rule in many sub-Saharan countries in Africa (Basedau and Elischer 2013; Harkness 2017; Ouédrago 2014), the results for Latin America indicate that the historically influential role of the armed forces has changed (Bruneau 2005; Mares and Martinéz 2014; Pérez-Liñán and Polga-Hecimovich 2016). In Asia, the military seldom finds itself in a ruling position, though it often enjoys veto powers, has impunity, or plays a crucial role in repressing opposition to the regime (Croissant 2013a). In most postSoviet countries, however, there has been a tradition of military submission under civilian leaders since the Soviet times (Avagyan and Hiscock 2005; Barany 2012; Bremmer and Charap 2007; Fluri and Cibotaru 2008; D. W. Rivera and Rivera 2014). Our findings on Eastern and southeastern Europe correspond to current scholarly literature on postcommunist countries, which ascribes armies only little or no importance in the internal organization and maintenance of the regime (Born et al. 2006; Edmunds, Cottey, and Forster 2006; Marat 2009; Sehring and Stefes 2010; Stykow 2013).

98

Aurel Croissant, Tanja Eschenauer, and Jil Kamerling

Our data further support the assumption that political-military relations in the Arab world underwent drastic changes since the mid-1980s. As a consequence, the role of the armed forces in countries like Egypt, Iraq, Yemen, and Syria has shifted to regime support instead of being the actual ruler or challenger to the regime (S. A. Cook 2007; Rubin 2002; Schlumberger 2008; Springborg 2016). During the Arab Spring, the relationship between political and military leadership as well as the structures of political-military relations proved to be central to the varying resilience of Arab autocracies (Albrecht, Croissant, and Lawson 2016). Military Influence in Authoritarian Regime Crises The identification of different political roles of the military evidently presents a new dimension to assess political-military relations, but does it help explain why and how autocratic regimes slide into crises? Do certain roles of the military make autocracies more or less vulnerable to crises, and do they affect their chances to survive crises? Authoritarian rule and regime survival are commonly contrasted to two types of challenges: horizontal threats from within the ruling elite and vertical pressures arising from the populace (Schedler 2015). At the elite level, competition for power and resource distribution threatens the powersharing agreement of the ruling coalition. Meanwhile within the populace, the threat of dissatisfaction to boil over and bring down the regime is ever present, and information on both discontent with their rule and strength of opposition is not easily available to autocrats. Milan W. Svolik (2012) labels these challenges as the problems of authoritarian control and authoritarian power sharing. Horizontal crises often result in the forced removal of the incumbent from office, a coup d’état. There are various forms of coups, including those instigated by civilian elements of the state apparatus (J. M. Powell and Thyne 2011; Svolik 2012). However, putsches, which are led by a military officer and executed by the military as institution (corporate coups) or segments of the armed forces (factional coup), are the most common form of horizontal crises in autocracies (Croissant 2013a; Svolik 2012). Vertical regime crises, such as mass mobilization, challenge the regime and the dictator’s political survival from below and do not originate from the ruling coalition. Scholars of popular mass unrest have found that a dictator’s ability to guarantee the loyalty of the armed forces is crucial for the outcome of such vertical crises8 (Chenoweth and Stephan 2011a; Goldstone 2001; Lee 2015; Nepstad 2011; Pion-Berlin, Esparza, and Grisham 2014; Schock 2005; Skocpol 1979).

The Military and Autocratic Regimes

99

We combine these different research strands to investigate the relation of military rulers and supporters to horizontal and vertical crises. Hereby, we present theoretical expectations on their impact on the onset of horizontal crisis, and the link between military supporters and the outcome of vertical crises in authoritarian regimes. Theoretical Expectations: Military Ruler and Horizontal Crises

By definition, military rulers are established either through a military coup or when victorious rebel movements transform into post−civil war governments. The structural and proximate causes for military coups—as the outcome of a horizontal threat—are many, both known and unknown (Cheibub 2007; Croissant and Kuehn 2015). The comparative literature provides ample evidence that regimes that emerge from violent military overthrows are more coup-prone than other forms of autocracies (Croissant 2013a; Geddes 2003; Geddes, Wright, and Frantz 2014; Svolik 2009). As a result, military regimes have a significantly shorter lifespan than most other types of authoritarian rule (Geddes, Wright, and Frantz 2014; Wahman, Teorell, and Hadenius 2013). This predisposition to horizontal crises can be traced back to birth defects inherent to regimes with ruling militaries. Scholarly literature on military regimes and horizontal crises argues that these forms of autocratic rule “carry within them the seeds of their own disintegration” as “transitions from military rule usually begin with splits within the ruling military elite” (Geddes 1999, p. 122).9 Therefore, we assume that these “construction errors from the very first days” (Gerschewski 2017, p. 11) render regimes with ruling militaries particularly prone to deinstutionalization processes and political crises. Birth defects of regimes with ruling militaries unfold their effect on the emergence of horizontal regimes in three ways. First, successful military takeovers do not necessarily imply that coup leaders enjoy the support of the military as a whole. More often than not, military coups are staged by “small groups of officers” while the rest decide “to roll with it” (Geddes 2009). A corporate coup is usually conducted by senior military leaders and supported by the military as an institution. The factional coup, in contrast, is staged by a military faction, often led by mid-level or junior officers, with dire consequences for postcoup politics. While most military coups are staged by relatively small groups of conspirators, the factional coup is “a sign of weakness in a key and distinctive aspect of the military’s capacity to seize power—its highly centralised and disciplined organisational structure” (Brooker 2009, p. 94). Precisely because the group seizing power lacks solid support from the rest of the military, the factional coup runs a

100

Aurel Croissant, Tanja Eschenauer, and Jil Kamerling

much greater risk that disaffected officers will launch “counter-coups” (Brooker 2008, p. 134). A similar logic applies to post−civil war regimes led by a victorious rebel movement. Insurgent groups that won civil wars are transformed into regular military forces of a post−civil war regime. Militaries that originate from victorious insurgent groups retain legacies of previous institutional structures and patterns of behavior impeding internal cohesion of the “new” armed forces. Thus, we should find the same birth defects in regimes resulting from a military victory in civil war as in those stemming from a military coup. Second, the coordination costs for coup plotters are reduced, as Erica Frantz and Natasha Ezrow argue, because most members of the elite coalition in regimes with a military origin have control over at least some parts of the security forces (2011). Coups simply become more feasible and easier to carry out in regimes with military rulers, making them particularly coup-prone. Third, we assume that regimes with ruling militaries often lack an institutional setting that would enable them to co-opt key actors of the regime. In autocracies with military rulers the top brass of the armed forces dominates important institutional channels. This privileged military elite often co-opts only some factions of the armed forces and individual officers into the ruling coalition, creating fissures within military institutions and conflicts between privileged and marginalized military factions (“losers” and “winners”; cf. Lee 2015). Summing up, because of corroding effects on military cohesion and weak institutionalization, regimes with military rulers are assumed to be less efficient in co-opting military and civilian elites. These arguments result in the following hypothesis: Hypothesis 1 (H1): Regimes with military rulers are particularly prone to horizontal crises.

Theoretical Expectations: Military Supporters and Horizontal Crises

An alternative view that connects the threat of a military coup and vertical threats from below is provided by Svolik (2012). According to Svolik, military coups are the result of failed brinkmanship bargaining by authoritarian and military leaders. The probability for such a scenario depends on the magnitude of mass threats to the regime, and the regime’s and its political leader’s reliance on the armed forces for domestic repression. Autocracies at low levels of mass threat do not depend on the military for domestic repression. Regimes with a military that is not pivotal in

The Military and Autocratic Regimes

101

domestic repression are more likely to institutionalize political control; thus, the likelihood of military intervention is low. In contrast, autocracies that face high levels of mass threat, and in which the military is indispensable for regime survival, are effectively under military tutelage. Because the military exerts its supremacy over the government, political leaders will accommodate their institutional and policy preferences. Since the military already enjoys extensive economic or political privileges in these regimes, it abstains from an overt intervention in politics, which would bear the risks of military fractionalization (Svolik 2012). It is only in the third scenario, characterized by medium intensities of mass threats and government reliance on the military in domestic repression, where military intervention is highly likely. In this scenario, military supporters will seek to increase their prerogatives and concessions while dictators aim to retain control over the armed forces. Horizontal crises, in the form of military coups, thus emerge “if one of the two parties rocks the boat too much” (Svolik 2012, p. 8) and constitutes the ultimate failure of a bargaining process between the dictator and the military over concessions and prerogatives. Building on Svolik’s (2012) theory, we expect regimes with military supporters to be particularly vulnerable to military coups. Due to their reliance on military support, they grant the armed forces prerogatives and concessions in exchange. If the bargaining between both sides fails, the military has great incentives to intervene into politics to secure its privileged position and maintain its benefits. Thus, if regimes grant their militaries a supporting role, they create “perverse incentives” (Gerschewski 2017), encouraging them to intervene into politics and assume power for themselves when they perceive a risk of losing their privileges. These thoughts lead to the following hypothesis: Hypothesis 2 (H2): Regimes with military supporters are particularly susceptible to horizontal crises.

Theoretical Expectations: Military Supporters and Vertical Crises

The third focus of contemporary research on the political role of the military and authoritarian regime crises is the outcome of vertical regime crises, in particular the military’s reaction to popular uprisings. One of the empirically more robust findings in the literature on contentious politics, nonviolent revolutions, and “anti-incumbent mass-protests” (Hale 2013, p. 336) is that a dictator’s ability to guarantee the loyalty of the armed forces is crucial for the outcome of such vertical crises (see, e.g., Barany 2016; Chenoweth and Stephan 2011a; Lee 2015; Nepstad 2011). While there is

102

Aurel Croissant, Tanja Eschenauer, and Jil Kamerling

consensus among researchers that military responses to popular uprisings frequently determine their outcome, what explains the military’s decision to defend a dictator, or to defect from the regime either by staying quartered, by shifting loyalty to the opposition, or by organizing a coup remains contested (Croissant and Selge 2016; Lutscher 2016). Regimes with supporting militaries combine characteristics that are discussed in the literature as being conductive to generate military loyalty during such vertical regime crises. As we mentioned above, military supporters serve as repressive agents against civilian opposition and in return enjoy ample veto power and impunity from the regime. In contrast to armed forces that are never deployed for internal security, military supporters are accustomed to employing force against civilians and therefore cracking down on peaceful protesters does not necessarily contradict the military’s self-conception and identity as guardian of the regime (Pion-Berlin, Esparza, and Grisham 2014). In this case, staying loyal and securing the regime’s status quo violently is rather likely (Bellin 2012; Huntington 1957; Nordlinger 1977). Additionally, if a military is responsible for human rights violations, it will probably defend the regime’s survival to avoid criminal prosecution (Hunter 1997; O’Donnell and Schmitter 1986). Impunity may therefore induce loyalty in the eye of vertical regime threats. Finally, a military that enjoys political or economic privileges is more likely to repress protests for fear of losing these privileges after a regime breakdown (Pion-Berlin, Esparza, and Grisham 2014). For this reason, veto power may serve as a factor that increases a military’s willingness to defend the regime with brute force. In sum, military supporters have real incentives to protect the incumbent regime in times of crises. Faced with vertical crises, military supporters should thus remain loyal to the incumbent regime. We therefore hypothesize: Hypothesis 3 (H3): Regimes with military supporters are likely to politically survive vertical regime crises.

Empirical Evidence

A first glance at the distribution of coups10 in our sample reveals that regimes with military rulers experience 42.9 percent of all attempted and successful coups (Table 5.2). This is a remarkable finding, considering that military rulers can be found in only 25 percent of our sample. Regimes with military supporters, which make up 46 percent of our sample, experience 57.1 percent of all coups. Table 5.3 shows that based on the Archigos dataset by Hein E. Goemans et al. (2009), regime leaders without military supporters are more

103

The Military and Autocratic Regimes Table 5.2 Political Roles of the Military and Frequency of Coups No military ruler Military ruler Total

Number of Coups 4 57.14% 3 42.86% 7 100%

No military supporter

Number of Coups

Military supporter

3 42.86% 4 57.14% 7 100%

Table 5.3 Military Supporter and Regime Leader Exit Regimes Without Military Supporters

Regimes With Military Supporters

Exit by popular protest

Exit by popular protest

Other forms of exit Total

34 89.47% 4 10.53% 38 100%

Other forms of exit Total

32 96.97% 1 3.03% 33 100%

often toppled by popular protest (10.5 percent) than regimes with military supporters (3.0 percent). These findings provide tentative support for our theoretical assumptions. To further investigate the relationship between military influence and regime crises, we provide three short plausibility probes: the military coups of April and July 2001 in Burundi, the military ousting of the Egyptian president Hosni Mubarak in February 2011, and the military refusal to crack down on protesters during the Georgian “Rose Revolution” in November 2003 that heralded the end of Eduard Shevardnadze’s presidency. Military Ruler and Horizontal Crises: The Burundian Military Coups of April and July 2001

Due to weak institutionalization and danger of further fractionalization of the armed forces, we expect regimes with military rulers to be particularly prone to horizontal crises (H1). This expected mechanism is illustrated by the two failed coup attempts of 2001 in Burundi. While Burundi experienced prolonged episodes of ethnic violence and military rule since independence from Belgian colonial rule in 1962, the country held its first free and fair elections in 1993. The election led to a

104

Aurel Croissant, Tanja Eschenauer, and Jil Kamerling

sweeping victory of the Hutu-led Front pour la démocratie au Burundi (FRODEBU) party (Reyntjens 2005; Vandeginste 2009). However, transfer of power from Major Pierre Buyoya (a Tutsi) to the elected Hutu president Melchior Ndadaye was aborted when elements of the defeated Tutsi elite staged a coup d’état in October of that year, killing President Ndadaye and other members of his government. Though the coup ultimately failed because François Ngeze (a Hutu) succeeded the murdered Ndadaye as president (Boshoff, Vrey, and Rautenbach 2010), the country plunged into renewed civil war (Boshoff, Vrey, and Rautenbach 2010; International Crisis Group 1998; Reyntjens 2005). In the ensuing chaos, army leaders called for a military intervention (International Crisis Group 1998) and Buyoya staged another military coup in 1996 and assumed power (Rumin 2012). Peace negotiations commenced in 1998 under the mediation of Tanzania and South Africa (Daley 2007). While Buyoya signed the Arusha Peace Agreement in August 2000 (International Crisis Group 2002), the government soon came under the suspicion of stalling the peace process (International Crisis Group 2001a). Within the armed forces, the government was believed to prolong the conflict because the ruling Bururi11 military-civilian coalition benefited from the status quo (International Crisis Group 1998; 2001a). These resentments deepened fissures in the highly factionalized armed forces. Non-Bururi cadres in particular had “the feeling that they [were] being used as cannon fodder for the sole benefit of the superior officers from Bururi . . . busy making the most of their Congolese business dealings or sharing out the few public contracts still allocated by the state” (International Crisis Group 2001a). In addition, the Arusha agreement demanded rigorous restructuring of the armed forces (Banal and Scherrer 2008) and, though officers were likely to keep their positions, the normal rank and file had to fear future unemployment (International Crisis Group 2001b). Cadets felt that they were the losers of the peace talks and saw “themselves as the victims of the reform of the army” (International Crisis Group 2001b). The resentments with the military leadership culminated on April 18, 2001, when Lieutenant Pasteur-Gaston Ntakarutimana and some forty soldiers took control of Burundi’s national radio and television station (RTNB), declared Buyoya and his military government to be “relieved of their responsibilities,” and defamed the government as collaborators of the rebel movement (International Crisis Group 2001a, p.13). The putschists were ultimately detained by loyalist forces, but the coup received support from various sources within the Tutsi community revealing further splits in the ruling elite (International Crisis Group 2001a). About three months later, the government faced a second coup attempt when two battalions of

The Military and Autocratic Regimes

105

soldiers headed toward the capital of Bujumbura (International Crisis Group 2001b). They were denied entrance to the city by police forces who engaged the roughly 400 soldiers in fire. The soldiers fled after they failed to help the mutineers of the April coup escape from prison (International Crisis Group 2001b). While it was indeed a broad coalition of army officers that brought Buyoya back to power in 1996, his government serves as a textbook example of horizontal crises induced by a military ruler and therefore supports Hypothesis 1: due to the weak institutionalization after the coup, Buyoya’s government was mainly based on the co-optation of a small circle of military elites that was held together by regional ties. His exclusive rule and privileged treatment of the elite widened existing fissures in the military and disenfranchised large parts of the armed forces who felt they were losers and staged two consecutive coups to place their demands back on the government’s agenda. Military Supporter and Horizontal Crises: Egypt’s 2011 Military Coup d’État

We argue that regimes with military supporters are particularly prone to horizontal crises. Militaries that are engaged in protecting autocracies against horizontal threats are granted concessions and prerogatives by the dictator in exchange (H2). However, this bargaining over concessions fails if one side—either the dictator or the military—exaggerates its demands and the military overtly intervenes in politics (Svolik 2012), which is why such autocracies are particularly prone to military coups. The Egyptian military has been tightly intertwined with politics ever since the 1952 revolution and perceives itself as the guardian of the nation (Demmelhuber 2013). Following the military coup d’état in 1952, all presidents came from the Egyptian armed forces and were de facto selected by the officer corps (Lutterbeck 2013). After the Yom Kippur War in 1973 and the ensuing start of peace negotiations with Israel, however, then president Anwar Sadat increasingly civilianized the regime, curtailed the military’s representation in the cabinet, and nearly halved the military’s personnel capabilities (Springborg 1987). In exchange, the military received extensive economic prerogatives, which developed into an opaque military economic empire during the Hosni Mubarak presidency (Springborg 2011, 2016). Under Mubarak (1981−2011), the military pursued the regimesupporting role it had already carried out under his predecessors. Repression and surveillance of political opponents were primarily the responsibility of the internal security apparatus (Droz-Vincent 2011; Hashim 2011;

106

Aurel Croissant, Tanja Eschenauer, and Jil Kamerling

Lesch 2011). Although it was not regularly involved in everyday repression and torture, the military fulfilled a key function in the provision of internal security. Since the early 1990s an Islamist insurgency had flared up in Egypt, posing a serious threat to the country’s internal security (Gerges 2000; International Crisis Group 2007). To battle Islamist insurgents the military conducted counterinsurgency operations (Bellin 2012; Chams El-Dine 2016), and martial law enabled officers to convict suspected Islamists in military courts (Lesch 2011). In return for its loyalty and support, the military maintained vast political and economic privileges and exerted a major influence on security and foreign policy-related policies (S. A. Cook 2007; Springborg 2016). Furthermore, Mubarak approved the military’s nontransparent business practices (Hashim 2011). The Egyptian armed forces thus executed a regime-supporting role and served as the dictatorship’s “ultimate coercive backbone” (Albrecht and Bishara 2011, p. 14). During the final years of his reign, the symbiotic relationship between the regime and the military deteriorated. Mubarak allowed for the advance of a new business elite around his son Gamal (al-Sayyid 2013; Nepstad 2013). The new business elite pervaded the regime’s National Democratic Party (NDP), ascended to political positions, and used its swelling political weight to influence economic policies to its advantage (Demmelhuber 2013). Rumor had it that Gamal would inherit his father’s position as president (Hale 2013), which would have been a major blow for the military’s economic interests and cut its “traditional hold on the presidency” (Lutterbeck 2013, p. 39). Besides the business elite, a second potential rival of the military loomed as the regime nurtured and relied on a mighty internal security apparatus capable of outnumbering the military and challenging its prominent role within the regime (Barany 2011). When the regional wave of mass contention reached Egypt in January 2011, a window of opportunity opened for the military: it could retaliate on Mubarak for its advancing political and economic marginalization in favor of his son Gamal’s economic entourage and the internal security apparatus. With the security apparatus overwhelmed by the mere size and insistence of the popular upheaval, Mubarak had to fall back on his repressive agent of last resort to secure his maintenance in office (Blanga 2014; Pion-Berlin, Esparza, and Grisham 2014). To his surprise, the military detracted its loyalty from the dictator and declared that it would not violently crack down on protesters to ensure the survival of the regime (“Egypt Set for Mass Protest as Army Rules Out Force” 2011). On February 11, Mubarak finally resigned as president and handed over power to the Supreme Council of the Armed Forces (SCAF) (“Hosni Mubarak Resigns as President” 2011). This move was pushed forward by the mili-

The Military and Autocratic Regimes

107

tary’s top brass who had held a SCAF meeting the previous day to discuss the situation, already excluding its commander-in-chief Mubarak (“Hosni Mubarak May Step Down” 2011; Barany 2011). Thus, bargaining between Mubarak and the military over concessions and prerogatives failed during the years prior to the Arab Spring, prompting the military to intervene in politics and reverse the power gap to its advantage. By staging a coup and steering the transition process, the military ensured and even enlarged its economic and political influence. These findings lend support to Hypothesis 2 and indicate that regimes that lean on their militaries for support pay with a higher proneness to coups. Military Supporter and Vertical Crises: Georgia’s 2003 Rose Revolution and the End of Shevardnadze

We assume that regimes that rely on military support enjoy a comparative advantage during situations of “anti-incumbent mass-protests” (Hale 2013, p. 336); that is, primarily nonviolent mass movements demanding regime change but not necessarily achieving it (H3). The authoritarian regime of Georgia and its collapse during the Rose Revolution of November 2003 serves as a negative case that illustrates the plausibility of our argument. Shevardnadze, former head of the Georgian Communist Party and minister of foreign affairs of the USSR, became president of the newly independent Republic of Georgia in 1995. His neopatrimonial rule was based on the skillful manipulation of corrupt client networks and semiprivatized state structures, the co-optation of competing factions of nonmilitary elites, electoral fraud, and calibrated doses of soft repression (Jones 2011; Wheatley 2005). For day-to-day repression, the regime did not employ the military but primarily relied on the police. The police force remained heavily involved in violations of human rights, torture, illegal arrests, extortion of money from businesspeople and criminals, bribery, falsification of the results of investigations, organized crime, and assassinations. In contrast, the military had been deployed only to defend the territorial integrity of the newly independent republic against separatist groups over which the central government had lost control. Although the Georgian military had become a major recipient of US military aid under Shevardnadze, the armed forces remained dramatically underfunded; military expenditures declined from 2.2 percent of gross domestic product (GDP) in 1996 to 1.1 percent in 2003. By the end of Shevardnadze’s term in office, Georgia’s military had seen a steady deterioration by years of corruption among senior officers, dire economic conditions, and widespread evasion of military service (Mikaberidze 2015). Moreover, the military did not enjoy the

108

Aurel Croissant, Tanja Eschenauer, and Jil Kamerling

same opportunities for self-enrichment and illegitimate businesses as the police and the National Guard, both under control of the Ministry of the Interior (Darchiashvili 2005). All in all, the military did not fulfill a supporting role under Shevardnadze. Although President Shevardnadze had been successful in weathering a number of storms during his tenure, from the late 1990s on a series of developments signaled a weakening of his power, including economic crises, declining domestic and international support, and skyrocketing corruption (Bunce and Wolchik 2011). When the president’s party (the Citizens’ Union of Georgia) started to disintegrate, members of his ruling coalition joined the opposition and civil society groups began to mobilize against Shevardnadze. On November 2, 2003, the government held legislative elections, which were denounced as manipulated and fraudulent by the opposition and international observers. This sparked mass demonstrations in the capital city of Tbilisi and elsewhere (Karumidze and Wertsch 2005). An important turning point in the Rose Revolution occurred when the police and military began to withdraw their support from Shevardnadze and declared their allegiance to the opposition. When protesters broke into parliament on November 22, a couple of elite military units joined the opposition (Karumidze and Wertsch 2005). The next day, a cascade of army units declared loyalty to the opposition. They were followed by police units and, at last, the Tbilisi chief of police (Karumidze and Wertsch 2005) and the nonviolent protest reached its peak. When negotiations between the president and the opposition failed, Shevardnadze resigned on the evening of November 23, 2003. As Steven Levitsky and Lucan A. Way (2010, p. 225) argue, the downfall of Shevardnadze was rooted less in the scale of the mass protest and more in the “decomposition of the coercive apparatus.” When Shevardnadze became aware that he no longer controlled the military and security forces, he resigned (Mitchell 2004). To some degree, opposition tactics played a role as key individuals in the opposition tried to convince the military leadership to defect from the regime (Karumidze and Wertsch 2005). However, the military’s willingness to shift its loyalty can be attributed to the fact that it had never been courted by Shevardnadze. Instead, Shevardnadze had treated the armed forces with a surprising degree of negligence. This produced strong discontent among soldiers even before the events of November 2003 (Mitchell 2008, p. 81). Hence, when the crisis peaked in late November, the military did not make any effort to enforce the state of emergency and to defend the dictator. The findings on Shevardnadze’s downfall thus support Hypothesis 3 because they illustrate that the absence of a supporting military might

The Military and Autocratic Regimes

109

prove to be lethal to the dictator’s political survival during instances of mass contention. Conclusion Our chapter was motivated by two interrelated research questions: First, how can different roles of the military in authoritarian politics be conceptualized and measured? Second, do certain roles of the armed forces make autocracies more or less vulnerable to crises, and do they affect the chances of autocrats to survive such critical episodes? Summarizing the contributions of this chapter, three points should be retained. First, our modification and enlargement of earlier work shows that it is possible to compose a dataset of military roles in politics for a large number of cases, which breaks with the fallacy of coupism and offers greater insight into the gray zone of military influence. Second, the data illustrate that militaries remain important and mighty actors in the diverse forms of authoritarian rule despite the declining number of coups and military regimes. As our global and regional comparisons showed, militaries still exert influence on politics in a significant share of authoritarian regimes by acting as overt or hidden rulers and/or by maintaining dictatorial survival as a supporter. Third, we argued that the form of military influence has an impact on the emergence and the outcome of authoritarian regime crises, which was demonstrated in three short plausibility probes. The 2001 military coups in Burundi as well as the coup by the armed forces to oust Mubarak in 2011 indicate that the presence of a military ruler and/or a supporter increases a regime’s proneness to horizontal regime crises. Shevardnadze’s political end during Georgia’s Rose Revolution revealed that refusing the military a supporting role can encourage a military to reject a violent crackdown of protesters, risking the downfall of the regime. Regimes with military supporters thus seem to have better chances to survive a vertical regime crisis due to the privileged position of their militaries, which creates incentives to protect the regime against mass protest. And fourth, by investigating the relationship between military influence and regime crises, we presented an approach to combine studies on the stability of autocracies with current research on military influence. Future analyses on crises in authoritarian regimes should bear in mind that measuring military influence in terms of coups and military regimes fails to capture the multifaceted and diverse roles that armed forces play in contemporary dictatorships. Advancing the research and measurement techniques of military influence widens and deepens our knowledge of authoritarian crises and how militaries contribute to their emergence and outcome.

110

Aurel Croissant, Tanja Eschenauer, and Jil Kamerling

Notes 1. See Croissant, Eschenauer, and Kamerling (2016) for a more detailed explanation of data generation and aggregation of our indexes. 2. States which are considered “failed states” or possess no regular armed forces are excluded. 3. The original dataset contains both yearly and aggregated data. 4. * denotes a logical “AND,” + a logical “OR.” 5. The regional coding is based on the regions of the BTI. 6. See Croissant, Eschenauer, and Kamerling (2016) for comparisons with the Autocratic Regimes dataset by Geddes, Wright, and Frantz (2014) and Varieties of Democracy measures on military influence (Coppedge et al. 2016). 7. This includes Algeria (1999−2012), Egypt (1999−2011), and Lebanon (1999−2012). 8. Vertical threats can be differentiated into threats that are brought forward by armed nonstate actors and those that are resulting from nonviolent mass unrest or civil resistance. While intrastate armed conflict is a frequent phenomenon in nondemocratic regimes, autocratic regime crises resulting from nonviolent vertical threats are less frequent, though the relative predominance of leader exits precipitated by the masses has increased since the 1990s (Kendall-Taylor and Frantz 2014). 9. We borrow theoretical arguments from the study of military regimes since these regimes share common characteristics with regimes that possess a ruling military. Military regimes and regimes with ruling militaries alike originate from military takeovers and are ruled directly or indirectly by influential militaries. We assume that military regimes and regimes with ruling militaries suffer from the same birth effects, destabilizing consequences, and proneness to horizontal crises. 10. We consider all failed and successful coups of the dataset provided by J. M. Powell and Thyne (2011) except the origin coup of a new regime spell. 11. Members of the political and military elite are predominantly recruited from the southern region of Bururi.

6 The Impact of Elections: The Case of Uganda Svein-Erik Helle and Lise Rakner

The people of Africa—the people of Uganda—are entitled to democratic government. It is not a favor from any government: it is the right of the people of Africa to have democratic government. —President Museveni’s swearing-in address, January 29, 19861 Democracy means the people support you. If they don’t support you, you don’t win. That’s all. —President Museveni at a presidential debate live on Ugandan TV, February 13, 20162

In February 2016, President Yoweri K. Museveni and his ruling National Resistance Movement party (NRM) were reelected for a fifth time through national elections in Uganda. The official electoral count showed that Museveni won 60 percent of the vote in the presidential race and the NRM won 70 percent of the seats in parliament. While the results for Museveni dropped slightly compared to the 2011 electoral race, the 2016 elections signal the continued challenges involved in defeating an electoral authoritarian regime at the polls. Observers have pointed to different factors when accounting for the continued electoral successes of Museveni and the NRM after thirty years in office. Some attribute the electoral success of the incumbent to a disorganized and uninspired opposition and satisfaction with improved growth and security (Conroy-Krutz and Logan 2012). Others link the repeated incumbent victories to massive preelection spending, vote buying, and intimidation (Carbone 2008; Izama and Wilkerson 2011). Both explanations of the NRM’s electoral victory fail to provide a convincing explanation of why or how it transpired because the singular focus on opposition weakness or incumbency advantages do not address the underlying structures that enable the authoritarian incumbent—the NRM under President Museveni—to minimize the institutional uncertainty that a 111

112

Svein-Erik Helle and Lise Rakner

multiparty election poses. As we explain, under Museveni’s leadership, the NRM has been able to entrench and routinize their powers through multiparty elections by shaping the playing field to their advantage. The understanding of electoral authoritarian regimes and the ways that institutions, such as elections, are employed to maintain control has increased significantly over the past two decades as the number of these regimes has multiplied. The literature concludes that institutions in general—and elections in particular—can be drivers of both democratization and autocratization processes and that we need to be more specific in terms of scope conditions for our theories to identify when, where, and why they play these roles (Brancati 2014; Cassani 2014; Haggard and Kaufman 2016; Morgenbesser 2014; Morse 2012). We contribute to this debate with a case study of Uganda, a regime that has voluntarily and strategically formalized the role of elections over time, a strategic decision that has stabilized the regime and prevented potential crises. Our analysis of electoral policies in Uganda under NRM rule shows how elections can be employed to strengthen the power of the incumbent, and that the underlying conditions that enable this can be found in already established institutions catering to the demands of the incumbent. The incumbent’s continued—and in some cases increasing—use of a superior state-supported organization and monetary advantage effectively creates an uneven playing field that undermines the opposition’s ability to compete. The NRM’s catering to rural demands through the manipulation of local government structures created and maintained in a no-party setting illustrates how a persistent rural bias in African politics creates formidable barriers to political transitions via elections. The NRM’s entrenched electoral hegemony is achieved through systematic use of local government structures that have not been restructured to accommodate multiparty electoral competition. The Ugandan case highlights that rural constituencies continue to offset the electoral weight of urban populations (Boone 2013). In Uganda, a largely urban-based opposition has limited space to mobilize support when facing a regime with a formidable ability to maintain its electoral base in the rural areas. However, as our analysis demonstrates, the same mechanisms that have contributed to this stability may lead to possible erosion and decay in the longer term through changes in what can be considered quasi-parameters endogenous and exogenous to the regime. The NRM’s continued success in multiparty elections depends on their ability to maintain and strengthen control over state and party structures, not least by successfully managing the question of succession. At the same time, changes in opposition strategy and demographic developments may challenge the stabilizing effect of multiparty elections and contribute to deinstitutionalization.

The Impact of Elections

113

We begin our analysis with a review of the literature on elections in nondemocracies, before analyzing the role of elections in Uganda from the start of the NRM rule up until the decision to reintroduce multiparty elections in 2005. We then turn to compare the electoral strategies and success of the incumbent and the opposition in three subsequent multiparty elections (2006, 2011, and 2016). We show that the NRM and President Museveni instituted a number of formal and informal reforms between the 2006 and 2016 elections that closed the space for the opposition to establish party organization and mobilize resources. The combined processes of recentralizing power away from local councils, and at the same time increasing the number of administrative districts, has enhanced the executive power over resource distribution and made it more difficult for the opposition to mobilize resources and votes. A concluding section finalizes the chapter. Electoral Competition in Nondemocracies: Stability or Crisis? The jury is still out on the effect of elections in authoritarian and hybrid regimes. A substantial literature understands elections as an institutional tool in the hands of the ruling regime serving to stabilize their rule through providing tools for legitimation, co-optation, or repression (Gandhi and Lust-Okar 2009; Gerschewski 2013; Schedler 2013). Another strand of literature highlights that elections, despite taking place in a context of informal autocratic practices, provide opposition forces with unprecedented opportunities if certain preconditions are in place (Bunce and Wolchik 2010; Hadenius and Teorell 2007; Howard and Roessler 2006). Some conclude that elections contribute to democratization (Lindberg 2009), some that they ensure autocratic stability (Bogaards 2013; Magaloni 2006). Other works again find that elections in and by themselves have little effect (Wahman 2013). A range of other, intermediary, variables conditions the effect of elections in hybrid regimes (Donno 2013; Kaya and Bernhard 2013; Seeberg 2013). The diverging scholarship is attributed to flawed conceptualization resulting from a failure to understand the underlying dynamics of institutions under authoritarian rule (Brancati 2014; Haggard and Kaufman 2016; Morgenbesser 2014; Morse 2012; Pepinsky 2014). Thomas Pepinsky (2014) holds that since much of the literature claiming that institutions have an independent effect on regime stability and crisis is premised on institutions put in place with the consent of the rulers to stabilize their rule by constraining actors’ choices, they cannot at the same time have a destabilizing effect as their ability to constrain choices depends on the goodwill of

114

Svein-Erik Helle and Lise Rakner

the ruler. Institutions such as multiparty elections must therefore be seen as epiphenomenal to regime interests and wider societal issues and, as a result, studies of authoritarian regimes and their institutional underpinnings should be expanded to include a wider range of variables. We acknowledge Pepinsky’s (2014) fundamental points about institutions in general, but find that this perspective overlooks a central aspect introduced by Johannes Gerschewski in Chapter 1 of this volume: the purpose and function of institutions might change as a result of flawed design or contestation over the role and meaning of the institution. This is particularly important with regard to the institutions of multiparty elections because, as argued by Andreas Schedler (2013), by definition institutions entail an element of uncertainty. For Schedler, elections in authoritarian systems should be understood as asymmetric games, which are defined by “the unequal distribution of power: one of the players (or teams of players) holds the power over rule definition, rule application, or dispute adjudication. Such asymmetries of power . . . are bound to subvert the ‘spirit of the game,’ the spirit of fair competition among equals” (2013, p. 113). In other words, elections serve a clear purpose for the incumbent. But this does not mean that, once introduced, elections serve the interest of the incumbent forever. Since the electoral game is voluntary and the incumbent depends on having an opponent for the game to be seen as real, the incumbent cannot manipulate the process to the degree that it is rendered meaningless for the opposition to participate. As a result, authoritarian regimes cannot remain electoral and, at the same time, totally eliminate uncertainty (Schedler 2013, p. 132). This fundamental uncertainty may open for electoral situations where the original intentions of the institutions are rendered void through a deinstitutionalization process. Whereas formal institutions in democracies work as a constraint on power in that they set the rules of the game and are binding for all actors, in a noncompetitive autocracy the opposite is the case: formal institutions are epiphenomenal to the interests of key actors and, therefore, do not work as effective constraints on power. But what makes competitive electoral authoritarian regimes hybrid is that we do not know if and when formal institutions will work as a constraint on power and when they will not. To identify how elections and electoral competition contribute to regime stabilization or crisis (deinstitutionalization), a range of quasi-parameters may be identified that may contribute to elections serving a stabilizing or destabilizing role. The diverging arguments regarding the effects of the election are further linked to the level of framing theories. In separate reviews, Dawn Brancati (2014), Stephan Haggard and Robert R. Kaufman (2016), and Yonatan L. Morse (2012) all argue that the literature focuses on case universes that are too broad, and that it is necessary to move beyond the focus

The Impact of Elections

115

on generalizability to a focus on either midrange theories and scope conditions or equifinality and combinations of causal factors and effects. In this chapter, we follow the former approach by focusing on regimes where multiparty elections were introduced voluntarily by an incumbent regime in control of the process, and largely without foreign interference. The effect of the form of political transition for post-transition politics is well documented (O’Donnell and Schmitter 1986; Przeworski 1991). In the case of Africa, Nicholas van de Walle (2002) points to diverging short-term outcomes and long-term effects when transitioning from single-party to multiparty politics. In cases where the incumbent regime was able to control the transition to multiparty politics, it was typically also able to maintain control of the state apparatus, hereunder the electoral process (van de Walle 2002). Regimes remaining in control of the transition to multiparty elections could therefore be expected to use elections as an autocratic tool—at least in the short run. Illustrating a case of elections introduced and formalized in a controlled top-down fashion, Uganda offers an interesting example of how and why elections serve a both stabilizing and destabilizing effect. The incumbent NRM regime believed that introducing multiparty elections would strengthen their hold on power in a situation where the previous system was decaying at an alarming rate (Makara, Rakner, and Svåsand 2009). Uganda also represents a case of gradual adoption of formal institutions as the NRM regime and President Museveni have ruled in an autocratic fashion for thirty years, through three different institutional arrangements that have featured an increasing formalization of political competition. The institutional landscape in Uganda under NRM rule has evolved at least partly as a response to institutional decay, ending up in a multiparty system. Multiparty elections have stabilized the MuseveniNRM regime in the short to medium term, in particular through a local government structure created and maintained in the no-party setting. However, the same mechanisms that have contributed to the stability of Museveni’s autocratic rule can lead to possible erosion and decay in the longer term. From Inclusiveness to Decay: The Crisis of the No-party Model in Uganda, 1986−2003 Museveni was sworn in as president of Uganda in January 1986 after emerging victorious from a five-year guerilla war where his National Resistance Army (NRA) defeated both the civilian regime of Milton Obote, who came to power through a fraudulent election, and the subsequent military regime

116

Svein-Erik Helle and Lise Rakner

that briefly replaced Obote in 1985. In his inauguration speech Museveni announced that one of the primary objectives of the Bush War had been to restore democracy in Uganda, and therefore this was one of the primary goals of his new regime (Museveni and Kanyogonya 2000). The solution he proposed was not a standard democracy where parties and party candidates competed for votes through elections: Museveni blamed the previous political conflicts and political violence that had plagued Uganda since independence on sectarianism that “divide [sic] people along ethnic and religious lines” and laid a large part of the blame on the old political parties, the Democratic Party (DP) and Uganda People’s Congress (UPC) (Museveni and Kanyogonya 2000, p. 6). He advocated for a gradual approach to building a new democratic Ugandan state that would replace the old corrupted state institutions. Formally, what the system created was a bottom-up approach to democracy, with the backbone of the state anchored in the local rural councils modeled on the Resistance Councils (RCs) established by the NRA in occupied territories during the war. The system, which was legalized by a government statute as early as 1987, established elections at the local level, but the national level remained in the hands of the unelected ruling NRM elite and President Museveni (Carbone 2008). The elections for the RCs were hostile toward alternative political organizations: elections were to be conducted based on the principle of individual merit and no party affiliation was allowed. Political parties, while nominally allowed to exist, could not practice party politics outside their own party headquarters. Instead, all Ugandans were declared as members of one nonpartisan political organization called the National Resistance Movement (NRM-S). While it was de jure a noparty system, de facto the system separated between official candidates and those who did not possess such credentials (Carbone 2003). The system has therefore been described as a single-party system (OlokaOnyango 2000), a dominant party system (Kasfir 1998), and hegemonic party system (Carbone 2008). Initially, this system was popular with the Ugandan population. The decentralized RC system provided the Ugandan population with unprecedented opportunities for political participation, and the local presence provided by the RCs gave access to state institutions that also was unprecedented. In the words of Joshua Rubongoya, “During the first years of NRM rule, the LC system was without question the most important legitimizing strategy” (2007, p. 69). The system, combined with continuous economic growth and a general fear in the population of returning to war, provided the NRM with sufficient legitimacy to build the foundations of the regime without experiencing undue criticism. The system carried additional advantages as it effectively barred the opposition parties

The Impact of Elections

117

from building organizations, and it exposed them to the dangers of cooptation (Carbone 2008). Precisely because the NRM depicted itself as an all-inclusive movement rather than a partisan organization, it was effectively protected from political attacks from actors outside the organization, and this made it rewarding for potential challengers to work inside the structures. With no separation between the state of Uganda and the NRM, the movement utilized its nonpartisan state apparatus to build a wide-reaching network of local government organizations that were de facto party structures (Tripp 2010). Institutional Decay, 1996−2003

Elections to the local councils (LCs; later LC1, etc.) constituted the grassroots foundations of the governance system in Uganda until 1996, when the first presidential, parliamentary, and local elections were held. The decision to open up electoral competition for executive as well as legislative offices came partly as a result of institutional decay within the NRMS. By the early 1990s, participation in local councils had reduced significantly (Tripp 2010), and attention had shifted to the national arena where elections for an assembly that was to debate a new draft constitution had created clear divisions between those in favor of a return to multiparty politics and those in favor of keeping the movement system (Carbone 2008). The movement supporters in the Constituent Assembly, backed by Museveni, were able to defeat the multiparty supporters backed by the old political parties (Carbone 2008). The constitution therefore established a noparty system at the national level as well, and Museveni subsequently defeated multipartyist Paul Ssemogere in the first presidential election. The NRM formalized no-party competition at all levels once they were confident that they could win over their opponents who were forced to be legally undefined “outsiders” inside the NRM-S. In 1997−1998 the NRM took further steps toward formalizing the NRM as a party within a no-party system through the Movement Act that specified the nexus between the state and the official NRM faction (Carbone 2003), and by establishing an NRM caucus in parliament (Muhumuza 2009). But as the process of institutionalizing the NRM movement as a party gathered momentum, the early emphasis on inclusion and the all-encompassing nature of the NRM coalition gave way to internal elite struggles and institutional decay (Carbone 2003; Makara, Rakner, and Svåsand 2009). Toward the late 1990s, several internal conflicts emerged exposing the movement to factions that challenged the leadership of President Museveni. Central NRM politicians began to openly question the government about corruption and the leadership of Museveni (Tangri and

118

Svein-Erik Helle and Lise Rakner

Mwenda 2001). The 2001 elections marked a turning point in Ugandan politics as former regime insider Colonel Kizza Besigye mounted a very personal campaign against the president. While Museveni received 69.3 percent of the vote, Besigye’s strong electoral performance thus formalized the previously informed rift within the NRM (Table 6.1 highlights that his support dropped by 5 percent compared to the 1996 election) (Carbone 2003). The internal weaknesses of the movement system exposed in the 2001 elections showed clear signals of deinstitutionalization, as the system provided opponents with the opportunity to attack it from within. The role of ideology had declined, and incidences of corrupt behavior of key movement politicians became publicly exposed. Assessing the status of the movement after the 2001 elections, the NRM National Executive Committee concluded that the all-inclusive nature of the movement had provided space for opponents bent on destroying the movement system from within (NEC 2002). In 2003 to the astonishment of most international observers, President Museveni announced that the NRM National Executive Committee would meet to discuss the reintroduction of multiparty politics. Introducing Multiparty Elections to Address Institutional Decay

The fact that Besigye won 29 percent of the vote in the 2001 presidential elections sent a strong signal to NRM party leaders and the decision to reintroduce multiparty politics in Uganda must be understood in this light (Makara, Rakner, and Svåsand 2009). Museveni characterized the 2005 referendum on multiparty politics as a housecleaning exercise that would

Table 6.1 National Elections Under NRM-S

1996 2001

Executive Elections: Yoweri K. Musevini’s Share of Vote 74.33% 69.33%

Executive Elections: Largest Challengers’ Share of Vote (Paul Ssemogere/ Kizza Besigye) 23.61% 27.82%

Parliamentary Elections: Declared Movement Candidatesa Share of Total Seats 56.52% (150 of 276) —b

Source: African Elections Database (2014). Notes: a. Figures based on African Elections Database (2014). The percentage of candidates that supported official Movement policy and President Museveni is likely to be higher. Numbers based on directly elected representatives only. b. No estimates found on percentage of declared candidates at time of election. NRM-S = National Resistance Movement-State.

The Impact of Elections

119

“rid the movement of saboteurs” suggesting that the NRM executive leadership was willing to formalize a multiparty system in exchange for extended executive control (Makara, Rakner, and Svåsand 2009, p. 193). Through its control of the government, parliament, and the public sector, the NRM was able to manage the transition process and minimize the uncertainty of losing power in the transition process (Makara, Rakner, and Svåsand 2008). The NRM used its government status to shape the electoral rules to its favor and its incumbency position to tilt the playing field to its advantage. By reintroducing multiparty elections, the NRM and President Museveni were able to prevent the deinstitutionalization of the NRM-S, caused by endogenous change agents and popular apathy, from leading to a regime crisis. By linking multiparty reforms to the removal of term limits from the constitution, Museveni and the NRM regime not only took away the ability of their opponents to mobilize on popular disenchantment with the forced inclusivity of the NRM-S, but also to delay and overcome the critical issue of leader succession within the NRM. Across sub-Saharan Africa, electoral authoritarian regimes have been challenged as a result of failed attempts to remove term limits, either through popular uprisings as in Burkina Faso in 2014, or through electoral losses under a new presidential candidate as in Kenya in 2002. With President Museveni still in power, the formalization of multiparty politics in Uganda in 2006 prevented two potential sources of regime crisis. The Routinization of Incumbency Rule Through Elections: Uganda, 2006−2016 The joint decision of linking the return of multiparty politics with the removal of presidential term limits ensured an important electoral advantage for the NRM and Museveni in subsequent elections (Makara 2010). In the immediate aftermath, however, this also carried some costs for the regime. By explicitly linking the return of multiparty politics to his continued rule, the first multiparty elections in 2006 became a contest about the popularity of the regime rather than a regular electoral contest (Makara, Rakner, and Svåsand 2009). As a result, Besigye and the opposition were able to campaign on the basis of being advocates of democracy, at least in the presidential race. Museveni and the NRM won the majority vote in the presidential and parliamentary elections, but the opposition gained support in both races (see Tables 6.2 and 6.3) despite a campaign characterized by fraud and violence to the degree that it was a 3−2 decision in the Supreme Court to uphold the election results

120

Svein-Erik Helle and Lise Rakner

(Gloppen et al. 2006). The 2006 experience led to important changes in the electoral strategies of the incumbent regime (Golooba-Mutebi and Hickey 2016). Poor governance performance and declining support of NRM candidates and President Museveni from 2001 to the 2006 elections led many to believe that the opposition would continue to progress in the 2011 elections, perhaps even forcing a runoff in the presidential race (Conroy-Krutz and Logan 2012; Izama and Wilkerson 2011). But as the campaign gathered momentum, it became clear that the opposition, and particularly the Forum for Democratic Change (FDC), were lagging behind the incumbent (de Torrenté 2013). As Table 6.2 shows, President Museveni was reelected for a fourth term with 68.4 percent of the national vote while his main opponent, Besigye, obtained 26 percent of the vote. President Museveni and the NRM reversed the downward trend of previous elections, and for the first time the NRM and Museveni won a plurality of votes in every region of Uganda (Perrot et al. 2014). The 2016 elections showed many similarities to the previous multiparty contests. The two main candidates were still Museveni and Besigye. The fact that former prime minister and secretary general of the NRM, Amama Mbabazi, ran as an independent candidate introduced a new element of uncertainty. While he failed in the election, securing a dismal 1.4 percent of the vote, Mbabazi’s bid attracted attention and raised concerns among members of the incumbent regime, who had spent most of their energy in the year leading up to the polls identifying, isolating, and reintegrating potential Mbabazi supporters within their own camp. This allowed the FDC and Besigye a freer role than in previous elections, at least until the campaigns gathered momentum in November 2015 and Besigye documented his popular appeal. Besigye bounced back in the

Table 6.2 Votes in Executive Elections in Uganda, 2006–2016 NRM (Yoweri K. Museveni)

2006 2011 2016

4,109,000 (59.26%)a 5,428,000 (68.38%) 5,972,000 (60.62%)

FDC (Kizza Besigye) 2,593,000 (37.39%) 2,065,000 (26.01%) 3,509,000 (35.61%)

DP (John S. Kizito/ Norbert Mao)

UPC (Milton Obote/ Otunnu)

110,000 (1.58%) 148,000 (1.86%) —

57,000 (0.82%) 125,000 (1.58%) —

Sources: African Elections Database (2014); Ugandan Electoral Commission (2016). Note: a. Percentages are of the total vote. NRM = National Resistance Movement; FDC = Forum for Democratic Change; DP = Democratic Party; UPC = Uganda People’s Congress.

The Impact of Elections Table 6.3 Parliamentary Elections in Uganda, 2006–2016

121

Total Seats

NRM Seats (% total)

FDC Seats (% total)

DP Seats (% total)

UPC Seats (% total)

Independent Seats (% total)

2011

350

2016

Final tally not yet announced

250 (71.42%)

34 (9.71%)

12 (3.42%)

10 (2.86%)

41 (11.71%)

2006

284

191 (67.25%)

37 (13.02%)

8 (2.81%)

9 (3.17%)

36 (12.68%)

Source: African Elections Database (2014). Note: a. Directly elected seats only. NRM = National Resistance Movement; FDC = Forum for Democratic Change; DP = Democratic Party; UPC = Uganda People’s Party.

presidential election, increasing his share of the official vote count by 10 percent compared to the 2011 electoral race. However, as can be seen in Table 6.3, the opposition parties made little progress in the parliamentary elections, as the NRM secured a two-thirds majority and the largest group of non-NRM members of Parliament were NRM-leaning independents. This parliamentary majority may become crucial because Museveni will needs a two-thirds majority in parliament to remove the age limit for presidential candidates specified in the constitution if he wants to run again in 2021. Both the 2011 and 2016 election results signal an entrenchment of the NRM regime under multiparty politics. In the aftermath of the 2011 and 2016 elections, divergent explanations have been offered as to why the NRM maintained the momentum in Ugandan politics. The first, advocated in many of the international monitoring reports, claims that the NRM’s victory was primarily a result of the extremely uneven playing field between opposition and incumbent (Abrahamsen and Bareebe 2016; COG 2011; DEMGroup 2011; EUEOM 2011, 2016; Izama and Wilkerson 2011; Perrot et al. 2014). The second explanation emphasizes the relative success of and support for the NRM government and the weak options provided by the opposition (Conroy-Krutz and Logan 2012; Vokes and Wilkins 2016). Both observations hold merit, as the 2011 and 2016 elections were characterized by a distinctly uneven playing field between the incumbent and opposition as well as a splintered and relatively weak opposition. However, these issues are merely symptoms of underlying structural and institutional factors nurtured by the incumbent

122

Svein-Erik Helle and Lise Rakner

regime since it attained power in the 1980s. In particular, Ugandan demographics and the local government structure created and maintained in a one-party setting have enabled the NRM to entrench their hegemonic powers through the strategic use of multiparty elections. New NRM Tactics Under Multiparty Rule in the 2011 and 2016 Elections

In the aftermath of the 2006 elections, President Museveni and the top leadership of the NRM conducted a thorough evaluation of the elections, and decided on important reforms. In 2006, NRM candidates were taken by surprise when the citizens blamed them for poor service delivery (Hickey 2013) and the regime lost support both domestically and internationally for their widespread use of detention, direct repression, and force to combat their opponents (Murrison 2013; Perrot et al. 2014). After the postelection evaluation, their focus shifted to building their electoral organization and offering voters a choice between the carrot and the stick. When the 2011 campaigns gathered momentum, the NRM had a cohesive network of cadres in place to protect the NRM’s interest at the local level. In each village area, nine-person campaign committees were responsible for mobilization, mirroring the setup of the RCs. Before the 2016 elections, these NRM committees were increased to thirty individuals in each village.3 The committees were often headed by heads of the village councils, the lowest level of the decentralized governance system in Uganda. These structures were last elected in 2002 under the NRM-S and, when multiparty politics was reintroduced in 2006, most of them switched loyalty from the system to the NRM party.4 These local electoral structures effectively provide the NRM with a partly state-sponsored local electoral organization, which in turn makes the NRM an attractive party for candidates running for office because they can run with the backing of an organization.5 The strength of the NRM election apparatus was witnessed in the heavily contested 2010 and 2015 party primaries. While the primaries were criticized for widespread rigging and led to several splits, they highlighted the enormous mobilizing apparatus of the incumbent party. At the top, President Museveni has centralized control over the state and party apparatus. He wanted to run a more professional campaign in the 2011 elections, with a focus on outreach and modernizing the campaign style (Perrot et al. 2014).6 To achieve this, the president and his close circle have gradually centralized control of key aspects of the political process. Control over important government programs such as the agriculture subsidy program, the National Agricultural Advisory Services (NAADS), has been centralized either at the State House or in the military

The Impact of Elections

123

(Kjær and Joughin 2012; Nabaho and Kiiza 2013). As an indication of the increasingly more centralized process, the State House and Office of the President budgets have grown at a higher rate than most other ministries. According to Sam Hickey (2013, p. 196), centralization and political control was ensured as a “means of mobilizing electoral support.” A large budget for presidential donations, a bloated cabinet, a seemingly endless number of positions as presidential advisers, the resident district commissioners (RDCs), and the sprawling local government network provide the president with ample positions to distribute as patronage (Awortwi and Helmsing 2014). Starting in earnest in 2014, President Museveni has also asserted his authority and centralized control within the NRM party structures. Having left the party to languish in the hands of other people for almost a decade after the reintroduction of multiparty politics, the president assumed firm control after Secretary General Mbabazi emerged as a potential challenger to his position in 2013. After a brief and clandestine struggle where Museveni used a younger generation of party cadres to promote a sole candidature position for him within the party, Mbabazi and several potential loyalists were ousted from key positions and made to challenge Museveni from outside the party. In return for their loyalty, the younger party cadres received Museveni’s backing to replace Mbabazi’s team and occupy key positions within the party and state. The emerging succession battles witnessed in the run-up to the 2016 elections illustrate once again how Museveni and the NRM regime have utilized the institutions of multiparty elections to strengthen and centralize their control over the carrots that are key to ensuring electoral victories: the state and the party. In addition to these incentives, the regime has created a climate of fear by applying selective and premeditative violence and repression against key groups while reminding the population that the regime can and will resort to violence and massive repression, if necessary (Perrot et al. 2014). At regular intervals between 2008 and 2016, newspapers and TV and radio stations have been closed because they have reported on issues considered threatening to the interest of the NRM. Journalists and editors are frequent targets of threats when reporting on protests or politically sensitive issues (Grönlund and Wakabi 2015; Human Rights Watch 2016). This has contributed to a widespread culture of self-censorship among Ugandan media practitioners. 7 Both Museveni himself and army spokespersons frequently criticize the opposition and sow seeds of doubt about whether they will accept an opposition victory.8 Finally, the regime has a long tradition of employing paramilitary forces. In the 2001 and 2006 elections, these forces were used directly to harass and frighten opposition supporters and politicians. In 2011 and especially in 2016, the approach has been slightly different, as the NRM government initiated a

124

Svein-Erik Helle and Lise Rakner

large-scale recruitment program of so-called Crime Preventers—unemployed youth given training by the government to act as a neighborhood watch at the village level. The presence of such groups has no legal basis, and their mandate is uncertain. They have been accused of being partisan agents of the NRM, something partly confirmed by the NRM’s attempt to use them as polling agents.9 Portrayed as a semiformal institution without a specific mission, the Crime Preventers were understood by the public as a potential source of violence and as a source of employment for unprivileged youth. Both the positive and negative view of the Crime Preventers strengthened the hand of the NRM in the elections (Tapscott 2016). Similar to electoral authoritarian machines across the world, in Uganda the combination of funds and positions as carrots and threats of violence as sticks fundamentally tilts the playing field against the opposition. The Opposition: The Challenge of Internal and External Cohesion

After the 2006 elections, the opposition also conducted an evaluation. They acknowledged that failure to agree on common candidates had hurt them in the presidential race and the parties also saw the need for building stronger organizations to increase outreach and enable them to compete in all constituencies.10 This indicates that the opposition saw the need for both internal and external cohesion. However, while FDC managed to field more candidates throughout the country in the 2011 and the 2016 elections, the opposition did not have sufficient resources to enable their candidates to run efficient campaigns, and were left mobilizing on antiregime questions rather than alternative policies or promises of resources. For the smaller opposition parties, the situation was worse. They failed to field candidates in many constituencies across the country, as they have struggled to act as coherent organizations. The two traditional parties, the Democratic Party and the Uganda People’s Congress arguably have the greatest organizational potential among the opposition parties since they have latent organizations spread across the country from before the movement era. However, internal wrangles, partly exacerbated by the NRM and the FDC, have prevented them from proper functioning as parties. The DP has at least two different leadership factions outside the official leadership who lean on either the NRM or the FDC. As a result, several high-profile DP-affiliated candidates have run as independents instead of as party representatives, taking part of the party organization and support with them.11 The UPC disintegrated as a party after the 2011 election, with one faction of the party pledging allegiance to the last elected party president Olara Otunnu, who pledged support for Mbabazi in the 2016 presidential elec-

The Impact of Elections

125

tion. The other faction, led by Obote’s son Jimmy Akena, said that he personally would support Museveni. The UPC and, to a lesser extent, the DP have been paralyzed by internal factionalism. Internal bickering has plagued even the larger FDC. Factions disagreeing on style and strategy emerged both during the 2012 party presidential and executive elections and the 2015 nomination exercise for the general elections, and the factions have remained visibly divided after the elections. The opposition’s failure to mobilize voters is primarily linked to their lack of organizational capacity: lack of internal and external organizational cohesion, access to key areas of the country, and resources. At the elite level, frequent advances from the NRM threatened the internal cohesiveness of the opposition parties. Defection of high-level party officials and opposition MPs were common in the 2011 and 2016 election campaigns,12 and contributed to the above-mentioned factionalism. Losers in internal party wrangles will not stay and fight for the party, they will leave and run as independents or for another party, often with the encouragement of the NRM and even other opposition parties. This scenario presents the opposition with a dilemma: by participating in the election, they legitimize the process and strengthen the legitimacy of the NRM. However, if they do not participate in the elections, the parties may not survive because individual candidates had strong incentives for running without the approval of the parties. This illustrates how—paradoxically—multiparty elections in an autocratic setting decrease the uncertainty for the incumbent party, as the opposition is torn between the dilemma of legitimizing the regime or sacrificing party cohesion. In the 2011 as well as the 2016 elections, the opposition parties came together to develop a joint platform with the backing of international donors. However, interest in the project faded because donor funding was restricted to the dialogue, cooperation efforts, and internal party matters, and could not be used for external mobilization.13 The Interparty Co-operation (IPC) that was created with donor funding in the aftermath of the 2006 elections broke down before the 2011 campaigns gathered momentum when the DP and later the UPC withdrew because they feared that the alliance would effectively make them a subsidiary of the FDC and that they would effectively lose their independence and privileges.14 Before the 2016 election, opposition parties came together again under the umbrella of the Democratic Alliance to agree on joint candidates for presidential, parliamentary, and local elections. After lengthy talks and support from both civil society actors within Uganda and external actors, the talks broke down over the issues of candidate selection and use of human and financial resources. The main fault line was between former NRM-stalwart Mbabazi, who received the support of the smaller opposition parties, and the FDC and Besigye,

126

Svein-Erik Helle and Lise Rakner

who felt that Mbabazi offered too little to the other parties for them to relinquish their position as the head of the opposition. The failure to agree on a joint candidate at the presidential level effectively spelled the end of the alliance, as the opposition failed to coordinate at the lower-level elections as well.15 Individual candidacy incentives thwarted opposition cohesion and prevented the opposition from effectively challenging the NRM. The NRM played a role in orchestrating this, partly through denying donors the opportunity to provide more comprehensive funding, and partly through sponsoring different factions within the opposition parties with diverging interests.16 In addition, the NRM used their organizational advantage to more directly deny the opposition the possibility of building nationwide organizations. How Control of Rural State Structures Strengthens Incumbents and Weakens Opposition

The Ugandan opposition parties are vulnerable to internal disagreements. Yet the main reason for their failure to build nationwide organizations is that they are denied access to the countryside and to resources, especially between elections. As argued by the treasurer of the Democratic Party, Issa Kikungwe, “For an election, you need the candidate, the structures, the message and the money. We had the candidate and the message, but we did not have the money and the structures. For us, these two issues are linked.”17 The story repeated itself before the 2016 election. Both the FDC and the DP had plans to focus on building structures in the aftermath of the 2011 elections, but were hampered by delays in anticipated donor funding and repeated harassment by local-level state structures when they sought to establish party offices across the country.18 This partly explains the predominantly urban nature of the support for the Ugandan opposition. The opposition parties have managed to mobilize urban citizens in the past three elections (Brisset-Foccault 2013), but they have not been able to reach the population in the countryside outside their traditional strongholds. This is linked to the NRM structures established at the time of the movement system, and maintained and extended after the introduction of multiparty elections. The NRM has always maintained a stronghold in the rural areas of western, central, and eastern Uganda (Muhumuza 2009; Vokes and Wilkins 2016). Today, the NRM controls most of the rural local-level government structures in Uganda. In 2011, it won over 90 percent of the local-level elections at the lower levels (LC3 and LC5), as the opposition struggled to field and fund candidates in many races because it lacked organizations and resources.19 Emphasizing how the multiparty constitutional reforms have not translated to institutional change

The Impact of Elections

127

at the local levels, local government elections at the lowest (village, or LC1) have not been held since 2001. This means that leaders elected through the NRM system run the lowest level of government in Uganda, the village level. While there are no official records that detail how many of the village councils are intact thirteen years after the last election, the NRM are confident that they control most of them and, if they are not intact, they deem it likely that a person with ties to the party has the authority in the village.20 Despite the elections being postponed since 2001, the NRM nevertheless held primaries for party candidates for all LC1 councils ahead of the 2016 elections. Since no general LC1 elections seem to be forthcoming, the winners of these primaries are now seen as the legitimate LC1 chairpersons in some areas.21 In cases where the opposition manages to win control at the district level, the NRM has other mechanisms to counter this potential power. These include the RDCs, who are the personal representatives of the president at the district level. These nonelected agents are more powerful than the district chairpersons and act as control agents with regard to the activities of the elected representatives (Meyers 2014). In sum, the NRM has ensured that uncertainty in the rural areas is kept to a minimum by controlling important access points through the maintenance of a local government system created during the one-party era. In the period running up to the 2011 and the 2016 elections, the NRM sought to reduce electoral uncertainty in the urban areas through the appointment of administrative authorities that removed much of the power of the elected leaders. An example of this is the Kampala City Authority, which was established in 2010 to counteract the influence of the Kampala mayor who has continuously been a member of the opposition (Lambright 2014). This means that NRM de facto control almost all local-level government structures in Uganda and, as Genevieve Meyers (2014) highlights, the NRM has a long history of utilizing the local government structures for partisan purposes. This advantage has been compounded by a recent trend in Ugandan politics: the vast expansion of the number of local districts. When the NRM came to power in 1986, there were thirty-three districts in Uganda. This increased to fifty-five by 2004, and just before the 2006 election twenty-two new ones where added. The so-called districtification gathered momentum after the 2006 elections. By the end of 2010, Uganda consisted of 112 districts (Singiza and De Visser 2011), and 25 new ones were promised during the campaigns and tabled in parliament (Meyers 2014). While the social benefits of increases in districts remain unclear, it paid dividends for the NRM and Museveni in the elections. Several analyses have shown that the new districts are usually granted by NRM officials, and particularly the president, during campaigns (Green 2010;

128

Svein-Erik Helle and Lise Rakner

Tumushabe 2009). This provides significant advantages to the ruling party because it creates a large number of new positions that can be used for patronage purposes (Meyers 2014) and because the creation of districts tends to be popular and enable the ruling party to win these seats (Tumushabe 2009). The trend continued in 2016, when government approved the establishment of twenty-three new districts six months before the election, and over forty new constituencies were up for grabs in 2016 relative to 2011 elections. The NRM government claimed that this was done to bring representation and services closer to the people. The opposition, however, claimed that the creation of new districts and constituencies was done to maximize the NRM representation in parliament.22 The hegemony of the NRM over the local government apparatus has both direct and indirect consequences for the opposition’s ability to build organizations and compete effectively in elections. The NRM uses this organizational advantage to maintain their hegemonic position in the countryside (Muhumuza 2009) and deny the opposition access. When the opposition seeks to conduct party activities between elections, the local councils, RDCs, and local police frequently stop them. 23 RDCs, local councils, and police also frequently deny the opposition access to local radio stations and other local news media, even though they have paid for access (Makara 2010). Illustrating how nominally democratic institutions in hybrid regimes are employed to entrench the power of the incumbent, infringements on the right to assembly controlled by local structures hinder the opposition from reaching voters between elections, making it unnecessary for the NRM to “get their hands dirty” during the campaigns.24 The local government system further enables the NRM to employ the threat of cohabitation. As highlighted by Douglas K. Singiza and Jaap De Visser (2011) and Meyers (2014), local governments receive the bulk of their funding from central government grants. The bulk of the funding is conditional grants, or grants made on the basis of an agreement between the central government and the local government. Through these agreements the central government effectively controls what the local government can do (Singiza and De Visser 2011). President Museveni and the NRM frequently warn the citizens of Uganda of the negative consequences for their district if they do not vote for the NRM.25 Finally, the resources of the opposition in Uganda are stretched by the size of the government system. Opposition parties in Uganda are poor and underresourced compared to the NRM, which enjoys access to both state and private resources (ACFIM 2016). The structure of government and elections in Uganda make it extremely costly for an opposition party to participate meaningfully in electoral politics: more districts mean that the

The Impact of Elections

129

opposition has to find and fund more candidates. In 2011 there were 18,629 elected offices, each carrying a significant nomination fee (DEMGroup 2011). In 2016, all nomination fees were increased significantly. The fee for contesting MPs for example was hiked fifteen times from roughly US$60 to $880, meaning that a party running a candidate for MP in every constituency would have to pay over $350,000 in nomination fees alone. Given that the opposition generally provides little monetary support for candidates other than their presidential candidates,26 candidates for MP and other offices need to fund their own nomination fees and campaigns, which can often be exceedingly expensive.27 The opposition relies on attracting candidates who can fund their own campaign. At the parliamentary level sixty-six new directly elected seats were contested in the 2011 election relative to the 2006 election, with forty more being created between the 2011 and 2016 elections. The winner-takes-all nature of the presidential competition also drains the opposition’s meager resources. The opposition parties spend most of their focus and resources on just keeping their presidential candidate in the field.28 President Museveni, on the other hand, can legally utilize the infrastructure attached to the executive office during the campaign period. Potential Challenges to the Stabilizing Effect of Multiparty Elections At a rally in Wakiso outside Kampala during the 2016 campaign, President Museveni argued: “The reason why leaders like UPC’s Milton Obote lost power twice through coup d’états was because they failed to read the situation properly, thereby failing to change accordingly.”29 As highlighted above, Museveni has proved adept at both tweaking the institutional setup of his regime when facing challenges of legitimacy and deinstitutionalization, and adapting his electoral tactics to new institutional setups. As a result, the gradual formalization of electoral competition culminating in the establishment of multiparty elections from 2006 onward has stabilized the NRM regime. However, Museveni and the NRM party are now facing challenges that in the long run may be difficult to contain within the setup of a multiparty regime. Changes in what Gerschewski identifies as quasiparameters in Chapter 1 might distort their stabilizing effect. These include a parameter that is endogenous to the institution and, therefore, might be seen as a birth defect: the financial and administrative costs of maintaining the carrots and sticks identified above. However, it also includes an exogenous factor that links to the uncertainty of elections as an institution: demographic changes in the electorate. In time, these quasi-parameters can affect

130

Svein-Erik Helle and Lise Rakner

and potentially alter the stabilizing effect of multiparty elections. We briefly discuss each in turn below. The uneven playing field in Uganda’s elections is key to understanding its stabilizing effect, but it also holds within it the seed for change because it is extremely costly to maintain. The 2011 election has been dubbed “Uganda’s most expensive ever” (COG 2011, p. 19), yet the 2016 election seems to top it as vast sums of money, material, and promises of club goods were dished out by the ruling party before and during the campaign (ACFIM 2016). Doing systematic monitoring of campaign spending in 16 out of 112 districts in Uganda, the Ugandan anticorruption organization Alliance for Finance Monitoring (ACFIM) found that two months into the campaign and with fifty days left, a total of just over $40 million had been spent by parties and candidates. The NRM had spent 90 percent of this (ACFIM 2016). And this was before the NRM spending kicked into high gear: a week before the election, each of the more than 60,000 village-level branches was given approximately $75 for mobilization.30 While such spending certainly works to grease the machinery of the NRM and mobilize voters, it also has two important and increasingly negative effects for the regime. First of all, it is increasingly hard for the NRM to get anyone to do anything without having to pay them. The NRM is quickly turning into a money-generating machine, which people see primarily as a vehicle for getting rich, making it increasingly difficult and expensive to run the party. People are also demanding increased payments for each election. This means that the NRM has to increase spending for each election to mobilize the vote. There is a great worry within the party that this cycle is nonsustainable and that the commercialization of politics in Uganda could make the system break down. In our interviews, party and government officials consistently mentioned the commercialization of politics as the single greatest challenge for the party and regime in power.31 Second, election spending has already created wider economic effects, as the 2011 election was followed by a record hike in inflation. Public anger at rising food and fuel prices allowed the opposition to mobilize large protests against the regime (“Walk to Work” campaigns; see Goodfellow 2014). While the protests were suppressed heavy-handedly and inflation normalized within a year, they illustrate the potential downstream effects that the commercialization of politics can have for the electoral authoritarian regime. Arguably, the uneven playing field is an effective mechanism for minimizing the risk of multiparty politics, but it also carries the potential for creating a crisis. Regimes that are becoming overly reliant on patronage typically break down when facing a disturbance to their income stream (Magaloni 2006). The other challenge facing President Museveni and the NRM is the changing demographics of the Ugandan electorate. In addition to the

The Impact of Elections

131

gradually shifting urban-rural divide, another parameter is starting to affect electoral politics: age. Uganda has the second-youngest population in the world, with an estimated median age of 15.6 years (CIA 2016). For each election cycle in Uganda, a new generation of voters is becoming an increasing part of the electorate. This generation differs from the previous generation, which grew up under Museveni and NRM rule, in that it has no personal experience of the social turmoil associated with previous regimes in Uganda. The threat of returning to war and conflict therefore rings more hollow to the younger generation. Polling data consistently shows that younger voters are more concerned with the high rate of corruption and unemployment and, perhaps most importantly, that they are more likely to support and vote for the opposition (Afrobarometer 2015; RWI 2016). The changing composition of the electorate was one of the drivers behind the change in leadership within the NRM. 32 Despite this, they are not confident that they will be able to contain the youth. Perhaps illustrating this, the NRM secretary general appealed to parents to keep young people away from protests in the aftermath of elections, warning, “The state will kill your children if they come to disorganize and destabilize the peace and security in Kampala and Wakiso.”33 Combined with the increasing cost of winning elections, the changing demographics of the Ugandan electorate suggest that multiparty politics may not stabilize the autocratic regime in the future. Conclusion About 30 percent of the regimes in the world today fall in the category of electoral authoritarian, defined as regimes that allow the opposition to run in elections without allowing them to challenge the hegemony of the incumbent. Paralleling accounts of electoral authoritarian regimes across the globe, our discussion has shown that the NRM regime in Uganda has entrenched its hegemony through excessive use of state resources (hyperincumbency advantages). Fitting the case category, Uganda illustrates how electoral dynamics, at least in minimally competitive nondemocracies such as electoral authoritarian regimes, cannot be understood either as fair competitions with certainty of procedures like those in democratic regimes or as meaningless exercises with certainty of outcomes like those in closed authoritarian settings. In electoral competitions, uncertainty is the name of the game in terms of both procedures and outcome, and incumbents choose strategies that reduce the uncertainty to the extent possible. As our analysis has highlighted, Uganda’s incumbents have learned to play the game of multiparty elections and utilize it to reduce uncertainty. Again reiterating

132

Svein-Erik Helle and Lise Rakner

patterns from authoritarian electoral regimes across the continent, the case of Uganda underscores the persistent rural bias in African electoral policies. The local government structure challenges analytical perspectives that regard the manipulation of the ruling parties in hybrid regimes as ineffective. The NRM has developed a form of rural and regional populism, linked to redistrictification and patronage. While opposition forces can mobilize support in urban areas and gain legitimacy through protesting against the autocratic regimes, elections are won in the countryside. As Beatriz Magaloni (2006) and Catherine Boone (2013) highlight, low urbanization signals that the electoral competition is primarily decided in the countryside. The case of Uganda illustrates how competitive elections and uncertainty create possibilities for regime stabilization in the short run, but potential deinstitutionalization in the long run. In Uganda multiparty elections were reintroduced in a controlled top-down fashion by the ruling powers, in large part because the incumbents believed it would strengthen their hold on power in a situation where the no-party system was in decay. The NRM regime and President Museveni have ruled in an autocratic fashion for almost thirty years, through three different institutional systems that have featured an increasing number of formal institutions. A conscious decision to reintroduce multiparty elections in 2003 was intended to address the institutional decay that had become evident in the no-party rule. Since the first multiparty election in 2006, the incumbent regime has used the formal electoral institutions as a mechanism to entrench its legitimacy and power. The 2001 and 2006 elections were marked by overt actions of violence and repression, but the 2011 and 2016 elections were characterized by fewer overt incidents of violence. Instead, the excessive financial muscle of the state-supported NRM and the fear generated by threats of violence and repression ensured that, come voting day, rigging was not necessary. However, the same quasi-parameters that have contributed to this stability can lead to possible erosion and decay in the longer term. An important insight from the study of electoral authoritarian regimes is that support to the regime is an indicator of regime stability, as citizens have an incentive to support the regime only if they expect it to last and continue to distribute benefits (Magaloni and Kricheli 2010). Our analysis of the entrenchment of authoritarian rule in Uganda post-2006 underscores this finding. The patronage network created in Uganda is substantial and sophisticated, yet expensive and vulnerable in times of economic recession. As Magaloni (2006) and Kenneth F. Greene (2010) emphasize, extensive party-based patronage systems are an advantage only as long as the incumbent is able to maintain control of a well-funded state apparatus. Faltering state finances, as a result of corruption, waste, or overspending, may challenge the elaborate patronage network established by the NRM and create

The Impact of Elections

133

openings for opposition parties to mobilize voters and challenge the incumbent at the ballot in future elections. Coupled with external factors such as the changing composition of the electorate, the role of multiparty elections as a stabilizing force in the electoral authoritarian regime of President Museveni and the NRM may well give way to destabilization in the nottoo-distant future. Notes We wish to thank Jeffrey Conroy-Krutz, Sam Hickey, Milan Svolik, Catherine Boone, and Johannes Gerschewski for valuable comments on earlier versions of the chapter. 1. Yoweri K. Museveni (as quoted in Museveni 2000, p. 3). 2. Yoweri K. Museveni, as quoted in Jill Craig, “A First: Uganda’s Museveni Takes Part in Presidential Debate,” Voice of America, February 13, 2016. 3. High-level NRM party officials, interviewed by the authors, January 2016. 4. NRM mobilizers (2) and high-level party officials (2), interviewed by the authors, December 2013 and January 2016. 5. As shown by Wilkins (2016), this state-sponsored patronage centrifugal patronage model is not as important in the NRM heartlands, where internal competitions drive local candidates to rely on their own resources to a much larger degree. However, Wilkins (2016) and Vokes and Wilkins (2016) also show the central role of President Museveni in greasing this machinery. 6. Key informant interviews with NRM high-level party officials (2), December 2013 and January 2016. 7. Twenty-two radio editors and radio journalists, interviewed by the authors, Uganda, October 2015−February 2016. See also Human Rights Watch (2016). 8. Freedom House (2013). For a recent example, see Robert Muhereza. “I Cannot Leave Power to Wolves, Says Museveni,” Daily Monitor, January 18, 2015. 9. Organizer of nationwide polling agent trainings, interviewed by the authors, Kampala, February 2016. 10. Secretaries general of FDC, UPC, and DP, interviewed by the authors, December 2013. 11. Repeated interviews with three high-level officials in DP, December 2011, December 2013, October 2015, and January 2016. 12. See, for example, Mwesigye, “NRM Converts Work Their Socks Off for Museveni,” The Observer, November 17, 2010; Lutukomoi, “Defections Are Part of Politics,” The Observer, November 6, 2013; Uganda Radio Network “Bukenya to Back Museveni 2016 Election,” The Observer, October 9, 2015. 13. Secretaries general of DP, FDC, and UPC, interviewed by the authors, December 2010; DDP officials, interviewed by the authors December 2010 and December 2013. 14. Secretaries general of DP and UPC, interviewed by the authors, December 2010. 15. High-level members of DP and FDC who participated in TDA (the Democratic Alliance) talks, interviewed by the authors, October 2015 and January 2016. 16. Donor and party representatives involved in election process, interviewed by the authors, December 2010, December 2013, and October 2015.

134

Svein-Erik Helle and Lise Rakner

17. Issa Kikungwe, Party Treasurer of DP, interviewed by the authors, December 2013. 18. Secretaries general of DP and FDC, interviewed by the authors, December 2013, October 2014, October 2015, and February 2016. 19. Secretaries general and party treasurers of DP, FDC, and UPC, interviewed by the authors, December 2013. 20. NRM mobilizers, interviewed by the authors, December 2013; high-level NRM party officials, interviewed by the authors, January 2016. 21. High-level NRM party officials, interviewed by the authors, January 2016 and October 2016. 22. For quotes, see Atuhaire, “New Districts Debate,” The Independent, August 31, 2015. 23. Mobilizers and secretaries general of DP, FDC and UPC, interviewed by the author, December 2010 and December 2013. Also noted in EUEOM (2011) and Freedom House (2013, p. 3). 24. It is easier for the opposition to reach the countryside during the campaigns, as “the local level obstacles that are usually present are not as visible because of the attention the campaigns receive” in the words of UPC vice president Joseph Bossa. Interviewed by the authors, December 2013. 25. See Okino, “NRM Warn Alebtong Residents on Opposition,” New Vision, January 4, 2010; Musoke, “Don’t Vote Opposition, Museveni Tells Masaka,” New Vision, September 10, 2010; Nambogga, “Don’t Vote Opposition, Museveni Tells Jinja,” New Vision, December 20, 2010; Wandera, “Kick Out Unproductive Leaders, Says Museveni,” Daily Monitor, October 15, 2015. 26. In 2016 FDC paid the nomination fees for their parliamentary candidates, partly as a result of the government releasing some public funds for parties for the first time. 27. For example, Issa Kikungwe, party treasurer of DP, estimated that he spent approximately $30,000 of his private money on his own and a few other DP campaigns in 2011. Kikungwe contested and won the Kyadondo South MP seat. 28. Party treasurers of DP, FDC, and UPC, interviewed by the authors, December 2013. Note that this changed somewhat in the 2016 elections, as FDC ran a tighter financial ship in their presidential campaign and DP and UPC did not field their own presidential candidate. 29. Yoweri K. Museveni, as quoted in Musisi, “Why Museveni Hasn’t Kept His Promise of Retiring,” Daily Monitor, January 24, 2016. 30. High-level officials in NRM, February 2016. See also Rumanzi, “NRM Cash Splits Ntungamo Candidates,” Daily Monitor, February 13, 2016. 31. Five high-level NRM officials, interviewed by the authors, December 2010, December 2013, October 2014, February 2016. 32. High-level NRM officials, interviewed by the authors, February 2016. 33. NRM secretary general, Justine Kasule Lumumba, as quoted in Wesonga, “Lumumba Shoot-to-Kill Threat Sparks Outrage,” Daily Monitor, February 1, 2016. Sound recordings of the speech in Luganda are published online, https://soundcloud .com/masake-anthony/justine-kasule-lumumba.

7 Coordination and Crisis Prevention: The Case of Post-Soviet Eurasia Max Bader Recent years have seen a deepening divide in the postcommunist world between Central and Eastern Europe on the one hand, and the postSoviet or Eurasian countries on the other. While democracy is consolidated in close to all Central and Eastern European states, authoritarian rule is the norm in most of post-Soviet Eurasia. Four countries in the region (Georgia, Moldova, Kyrgyzstan, and Ukraine) have taken notable steps toward democracy, but their democracies remain feeble. Armenia is seen as having a competitive authoritarian regime (Levitsky and Way 2010), with unfair elections but a serious degree of competition and pluralism. The seven remaining countries in the region (Azerbaijan, Belarus, Kazakhstan, Russia, Tajikistan, Turkmenistan, and Uzbekistan) are ruled by consolidated authoritarian regimes (Freedom House 2014). These regimes have seen no alternation in power for many years, little or no meaningful political competition at the national level, and embattled opposition movements. For the authoritarian regimes of post-Soviet Eurasia, the color revolutions, the revolutions in Kyrgyzstan in 2010 and in Ukraine in 2014, and the revolutions of the Arab Spring have been a reminder that staying in power cannot be taken for granted. Like all authoritarian regimes, the authoritarian regimes of the post-Soviet area need to permanently adapt to ensure their survival. To this end, the regimes employ a survival strategy consisting of an often intricate mix of policies and practices aimed at, on the one hand, winning popular and elite consent and, on the other hand, suppressing real and potential opposition. Despite the wide range of corrective policies that are available to the authoritarian regimes of the region, crises occur in the form of large-scale postelection protests and violent local uprisings. Since such events have led to regime turnovers elsewhere in the region, the regimes invest significant effort into suppressing them and forestalling their reoccurrence. In this chapter, I investigate how the consolidated authoritarian regimes of the region have employed intergovernmental coordination to limit their 135

136

Max Bader

susceptibility to crises. I argue that the central theme in the intergovernmental coordination consists of decoupling the authoritarian regimes from Western linkages. The regimes routinely accuse Western actors as well as domestic actors with links to the West of instigating regime turnover. Through coordination in intergovernmental organizations such as the Commonwealth of Independent States (CIS), Collective Security Treaty Organization (CSTO), and Shanghai Cooperation Organization (SCO), the regimes seek to establish a community of nondemocratic status quo powers, which are immune to exposure to supposedly Western norms and rules. I begin the chapter by describing the types of crises that occur in the consolidated authoritarian states of post-Soviet Eurasia. I then discuss the strategies that the regimes employ to stay in power, and ask why crises occur nonetheless. In the last section of the chapter, I demonstrate how the regimes have coordinated efforts to limit their susceptibility to crises. Crises in Post-Soviet Eurasia In November 2003, the regime of President Eduard Shevardnadze of Georgia organized parliamentary elections. When preliminary results suggested an unlikely victory for the pro-presidential For a New Georgia bloc, thousands of people gathered to protest. A wave of further protests over the course of three weeks culminated in the ouster of Shevardnadze. The events amounted to a comprehensive turnover of power: in the following months, new presidential and parliamentary elections brought to power the former opposition politician Mikheil Saakashvili and his party (Wheatley 2005). The scenario of this so-called Rose Revolution was repeated twice more in the post-Soviet area over the next two years. In 2004, significant fraud in presidential elections in Ukraine incited protests that led to the Orange Revolution. Similarly, fraudulent elections in Kyrgyzstan in 2005 brought down President Askar Akayev after large-scale protests. These color revolutions apparently caused great concern among the other authoritarian regimes of the region: in the years since, they have taken a wide range of measures to specifically prevent the scenario of an electoral revolution from unfolding in their countries (Ambrosio 2007; Silitski 2010). Large-scale postelection protests that had been organized throughout the region after the color revolutions were successfully broken up or disarmed by the authorities. Protests in Azerbaijan following the November 2005 parliamentary election copied many of the techniques that had been employed during the color revolutions, but the regime proved resilient (Valiyev 2006). In 2008, peaceful postelection protests in Armenia that had gone on for two weeks were broken up by the police, leaving

Coordination and Crisis Prevention

137

eight dead (International Crisis Group 2008). In December 2010, hundreds of people were detained at a rally on the night of the presidential election in Belarus (Potocki 2011). The December 2011 parliamentary election in Russia was followed by a wave of the protests that were the biggest the country had seen since the collapse of the Soviet Union. The wave of protests effectively ended after dozens of people were detained in skirmishes with the police in May 2012. The Russian authorities reacted to the unrest by adopting a range of new legislation in 2012 and 2013, further curtailing the freedoms of assembly, association, and speech (Freedom House 2013). On April 6, 2010, protestors stormed the regional government headquarters in the city of Talas in northern Kyrgyzstan, and deposed the governor. Seemingly inspired by these events, thousands of protestors attempted to storm the presidential palace the next day in the capital Bishkek. After a day of fighting between protestors and regime-loyal coercive forces, in which some eighty-six people were killed, the president fled the capital on the same night (Collins 2011). The events marked the second regime turnover in Kyrgyzstan in five years. Since the beginning of the century, there have been similar cases of violent uprisings outside national capitals in a number of post-Soviet states. Only in the case of Kyrgyzstan in 2010, however, did such an uprising bring about a revolution. In May 2005, hundreds of people were killed by police forces in a local uprising in the provincial city of Andijan in Uzbekistan (Hill and Jones 2006). In December 2011, violence erupted in the city of Janaozen in Kazakhstan after oil workers had gone on strike for seven months. During the violence, local government buildings were set on fire and at least fifteen people were killed by the police (Human Rights Watch 2012). A number of provincial cities in Azerbaijan have been the scene of spontaneous outbursts of violence in recent years. Most prominently, thousands of local citizens engaged in a violent confrontation with the police in January 2013 in the city of Izmayıllı.1 Finally, dozens of people were killed in clashes between local citizens and forces from the central government in the city of Khorog, Tajikistan, in July 2012. The region, then, has seen two types of unrest with a potentially fatal outcome for the authoritarian regimes: postelection protests and local uprisings. There are three differences between these two types of events. First, postelection protests are specifically a reaction to fraudulent elections while local uprisings are unrelated to elections. Second, postelection protests have sometimes featured violence, but most of their participants have tended to profess a commitment to nonviolence. The local uprisings, by contrast, tend to be violent, with the violence coming initially from either protestors or the authorities. Third, the postelection protests have taken place predominantly in the capital; the violent uprisings, by contrast, started

138

Max Bader

in cities outside the capital. In few instances have these events become truly life-threatening to the respective authoritarian regimes. Because of the precedents of the color revolutions and the example of the 2010 revolution in Kyrgyzstan, however, the protests and uprisings have had sufficient impact to provoke the authorities to suppress them with force and to take measures to forestall their recurrence. A notable difference between the postelection protests and local uprisings that did end in regime turnover, and those that did not, is the political context in which these events took place. The regime turnovers invariably took place in a context with a substantial measure of pluralism and a relatively low degree of repression (McFaul 2005; Radnitz 2010). This rather competitive authoritarian context in which the regime turnovers took place created wider opportunities for the opposition to organize, to protest, and to present alternative views in the media. The regime turnovers, accordingly, happened in countries that were classified as Partly Free rather than Not Free by Freedom House’s Freedom in the World index. Where postelection protests and local uprisings did not lead to a regime turnover, by contrast, the country was typically Not Free and had a hegemonic authoritarian political regime, in which protests were suppressed quicker and opposition movements found it more difficult to organize in the first place. Altogether seven such hegemonic authoritarian regimes in post-Soviet Eurasia have been successful in consolidating authoritarian rule, and in fighting off threats to regime survival when they occurred. How do these regimes engineer their survival? Regime Survival

Most authoritarian regimes in the world today, including those in postSoviet Eurasia, are electoral authoritarian in which access to power is organized through popular but unfair elections rather than, for instance, hereditary succession or military force (Diamond 2002; Schedler 2006). Traditional authoritarian regimes, which do not hold elections, stay in power as long as they ward off the triple threat of a popular revolution, a coup, and regime change imposed by a foreign power. The latter threat— regime change imposed by a foreign power—is rare and not considered here. Like traditional authoritarian regimes, electoral authoritarian regimes have to prevent revolutions and coups but, in addition, they periodically open themselves up to the challenge of electoral competition. Just to stay in power, then, electoral authoritarian regimes have to manage three tasks: preventing revolutions, preventing coups, and winning elections. When electoral authoritarian regimes mismanage either one of these three tasks, a crisis occurs.

Coordination and Crisis Prevention

139

To prevent a revolution, authoritarian regimes can utilize two strategies: first, make sure that people do not want a revolution to happen in the first place; and, second, when people do want a revolution to happen, make sure that they are too intimidated to undertake one. While the former strategy rests on winning the consent of the people, the latter strategy rests on using repression. To prevent a coup, defined here as regime change driven by domestic elites, authoritarian regimes can equally utilize two strategies: first, providing incentives to elites that make them want to stay loyal to the regime; and, second, raising the cost of defection from the elite. The first strategy, again, rests on winning consent while the second strategy implies repression. To win elections, finally, electoral authoritarian regimes also have two strategies at their disposal: first, actually get the vote; and, second, steal the vote. Actually getting the vote can be achieved, again, by winning popular consent while stealing the vote entails repression in the sense that the expressed will of the people is subverted. Electoral authoritarian regimes, in short, rely on a mix of consent and repression. Whether they rely primarily on consent or, alternatively, on repression, and how they win consent and use repression shape the nature of electoral authoritarian regimes. All measures and policies that an electoral authoritarian regime undertakes with a view of enabling its survival can be seen as corresponding either with a logic of consent, or with a logic of repression. In some areas, the regime may be inclined to act according to a logic of consent while in other areas a logic of repression dictates policy. The survival strategies of winning consent and using repression often are not stable over time: electoral authoritarian regimes frequently adapt them to account for new circumstances and available information. In doing so, they learn from their own experience, but also from the experience of other authoritarian regimes. The survival of these regimes hinges on how well they learn and adapt. The biggest threat to authoritarian regimes, from this perspective, is being isolated from credible information, most crucially about the preferences and sentiments of both elite actors and the general population. The sort of consent that electoral authoritarian regimes strive for comes in two forms: one implies aligning the interests and preferences of elites with those of the regime (co-optation) while the other implies being viewed by the general population as having the legitimate right to rule (legitimacy) (Gerschewski 2013). In the consolidated authoritarian regimes of the postSoviet area, the co-optation of elites is achieved primarily by drawing them into patronage networks through which rents are distributed (Hale 2006; Wallander 2007). Legitimation strategies that are applied in the region are diverse and include securing economic growth; providing welfare, stability, and security appealing to national pride and projecting international power.

140

Max Bader

Repression, as used by electoral authoritarian regimes to increase their chances of survival, can also be seen as coming in two principal types: reactive repression and proactive repression (Moore 1998). Reactive repression, in this context, is aimed at thwarting a (perceived) direct threat to the regime by arresting, prosecuting, or killing people. A risk of this type of repression for the regime is that, due to its highly visible nature, it can be met with an immediate and furious response. Proactive repression, on the other hand, is aimed at making sure that people will not engage in the type of activity for which, once the activity is undertaken, the regime would arrest, prosecute, or kill them. Verbal intimidation is one element in the menu of proactive repression. Another element is legislation that imposes high costs on oppositional activity. Examples of this include provisions for excessively high fines for violations at protest rallies, or excessively long prison sentences for misappropriating foreign funds by a civil society organization. When oppositional activity is effectively thwarted as a result of proactive repression, an authoritarian regime may be left with little insight into the nature and extent of antigovernment sentiments. The risk involved with proactive repression, then, is that when it ultimately fails, protest may erupt suddenly and uncontrollably such as in Uzbekistan in 2005 or in Kazakhstan in 2010. Why Do Crises Occur?

Why do crises occur in the consolidated authoritarian states, despite the wide range of opportunities for authoritarian leaders to engineer their survival? One likely reason is that, in managing their survival, the leaders face inevitable trade-offs. They face such trade-offs in each of the three tasks that they need to manage to stay in power: preventing a coup, preventing a revolution, and winning elections. Authoritarian leaders often prefer to win elections by a wide margin because this sends a signal of strength (Simpser 2013). The electoral manipulation including the outright fraud that they typically have to muster for this, however, can incite people to protest the official results. If the regime chooses not to commit substantial fraud and instead opt for a smaller margin of victory, it risks appearing weak. A small margin of victory can make it seem to elites that the regime does not have the requisite administrative capacity to organize large-scale electoral manipulation while to the population a small margin of victory can convey the impression that the regime can actually be defeated. In a sense, the Vladimir Putin regime experienced the worst of both worlds in the 2011 legislative elections; the ruling United Russia won by a comparatively small margin (49 percent to 19 percent for the runner-up party), and was only able to achieve a margin of this size through significant fraud

Coordination and Crisis Prevention

141

(Shpil’kin 2011). A second type of trade-off that authoritarian leaders face relates to choices about taxation and rent distribution. Gaining the consent of elites and of the general public typically involves the provision of material benefits. With control over limited resources, delivering private goods to select elites can win their support, but damage the regime’s popular legitimacy. Delivering more public goods on the other hand can win popular support, but reduce the incentives of self-interested elites to stay loyal to the regime (De Mesquita et al. 2001; Grzymala-Busse 2008). Another reason why crises still occur in consolidated authoritarian states is that the authoritarian leaders and their opponents have incomplete information. The leaders may not have reliable information about how people will actually vote in elections and about the commitment and capacity of regional operatives to manipulate the vote. Authoritarian leaders may generally underestimate their level of support among the population as well as the people’s readiness to resist the regime, especially outside the capital. On the other hand, opponents of the regime may not have a realistic image of the risks involved with resisting the regime. The crises that have occurred in the consolidated authoritarian states of the region generally have not made their perpetrators better off, with many of their participants having been killed, injured, arrested, or prosecuted. Moreover, the regimes have often reacted to the crises by imposing greater repression. It is safe to say, therefore, that most protestors and revolutionary leaders probably would have refrained from opposing the regime if they would have been cognizant of the consequences. Yet another reason why crises occur in the consolidated authoritarian states of post-Soviet Eurasia likely has to do with the poor performance of the regimes. The regimes score low on good governance and invariably are seen as highly corrupt.2 It is, altogether, not surprising that crises emerge in these countries from time to time. Some types of crisis seem more likely to emerge than others. Postelection protests are a common manifestation of crisis because elections are a focal point of opposition activity while they also highlight the corruption and manipulative practices of the regime. The reason why violent protests more often occur outside the capital is likely because coercive forces are concentrated in the capital and, thus, protests there would be more readily suppressed. Moreover, local or regional elites seem an easier target for protestors; while going against the central government is a tall order, removing a local strongman often seems more feasible. The Impact of Linkages with the West

The inevitable trade-offs in managing authoritarian rule, limited information, and poor regime performance can be seen as proximate causes for why

142

Max Bader

crises occur in the consolidated authoritarian states of post-Soviet Eurasia. The more fundamental reason behind the outbreak of events that threaten regime survival is a loss of legitimacy on the part of the regimes coupled with failings in repressive capabilities. In many of their public declarations, however, the regimes point to a different type of explanation for the outbreak of crisis: the role of foreign, and in particular Western, forces. During and following the crises, the authoritarian leaders of the post-Soviet states have often accused the West, as well as domestic actors with links to the West, of inciting, organizing, or financing the protests. After the clashes in Andijan, for instance, President Islam Karimov claimed that the crisis had been incited by extremists with financial support and backing from Western actors (Shermatova 2006). Similarly, after the Janaozen events, pro-government media in Kazakhstan asserted that the people who were at the forefront of the protests had been paid by Western forces seeking to destabilize the country (Imas 2012). Russian leaders have often accused the West, and the United States in particular, of orchestrating the color revolutions of 2003−2005 and the overthrow of President Viktor Yanukovych in Ukraine in 2014 (Gorenburg 2014). It is widely believed that the new legislation, which requires Russian nongovernmental organizations (NGOs) who receive foreign funding and are engaged in political activities to register as foreign agents, is inspired by the conviction that Western-funded NGOs have been instrumental in the color revolutions (Crotty, Hall, and Ljubownikow 2014; Daucé 2014). Opposition leaders in the post-Soviet states have been routinely accused of receiving Western money or operating on the orders of Western governments. Such accusations have also been voiced in the Russia-led intergovernmental organizations of the region. At a 2013 summit of the CSTO, for instance, Secretary-General Nikolai Bordyuzha accused the West of manipulating election observation missions, waging information campaigns on the Internet, and manipulating public opinion through media and NGOs in CSTO countries, and announced that the CSTO would take active measures to counteract these practices (Tarasenko 2013). The leaders of the authoritarian regimes in the post-Soviet area exaggerate the weight of Western-funded NGOs and political activists as the instigators of unrest. The linkages of the authoritarian regimes with the West, however, can make the regimes more vulnerable. Linkage with the West, whether in the form of economic linkage, geopolitical linkage, communication, social contacts, or transnational civil society networks, has been identified as a crucial explanatory factor for democratization in the postcommunist world (Way and Levitsky 2007). Most of the current linkages between the West and the post-Soviet states are a legacy of the first decade after communism. All the countries in the region became member

Coordination and Crisis Prevention

143

states in the Organization for Security and Co-operation in Europe (OSCE) and some joined the Council of Europe. These two organizations in particular have observed elections in the region, have reviewed new legislation, and have addressed violations of international standards by the postSoviet states. The countries of the region have also developed relations with the European Union (EU) and other intergovernmental organizations. Their economies after 1991 have to varying degrees become integrated into the world economy, with most of the countries joining the World Trade Organization. Many companies in the authoritarian states, including state-controlled companies, have expanded trade with Western companies. The populations of the countries have become more exposed to Western culture, especially through the Internet. Large numbers of people from the authoritarian post-Soviet states have traveled, worked, or studied in the West. Finally, starting in the early 1990s, significant numbers of NGOs in these countries started to receive grants from Western governments and intergovernmental organizations. The regimes in post-Soviet Eurasia have increasingly come to see the Western linkages as a threat: they fear that the exposure of their populations to Western values and culture can increase demands for good governance and accountability; they realize that trading with Western companies on the basis of international standards and regulations limits opportunities for corruption and other practices that are intrinsic to their patrimonial capitalist order (Robinson 2011); and they know that political influence from organizations dominated by Western governments can increase the cost of authoritarian governance. In the following section, I detail how coordinated efforts by the authoritarian regimes of the region have aimed at reducing the negative externalities of these linkages. Intergovernmental Coordination in Crisis Management There are a number of reasons to expect a high degree of intergovernmental coordination in the post-Soviet area. The first and most obvious reason is that the region has one of the largest concentrations of authoritarian regimes in the world. All of the economically powerful states in the region in particular have stable regimes that do not allow serious political competition. There are theoretical reasons to believe that autocracies prefer the presence of equally authoritarian neighbors over democratic neighbors (J. Bader, Grävingholt, and Kästner 2010). Second, due to a legacy of centuries of shared statehood in the Russian empire and then the Soviet Union, linkages in the region run deep. Russia wields significant soft power through its culture and through its media outlets, social contacts

144

Max Bader

across borders are common, trade relations are often dense, and the Russian language is a widely spoken lingua franca in much of the region. The homogenizing experience of the Soviet Union in particular has left a deep imprint on the region’s societies. Third, the regional landscape is shaped by the existence of a regional power (Russia) that itself represents one of the most prominent authoritarian regimes in today’s world and that aims to reintegrate the countries of the region (Cameron and Orenstein 2012). As the former imperial center and the current hegemonic power of the region, Russia is a point of reference for the other Eurasian countries in more than one regard. The relative weakness of other countries in the region often makes these countries dependent on Russia and vulnerable to its pressure. The region is thus characterized not just by strong linkages, but also by the significant leverage Russia has over the affairs of most of the region’s states. Finally, the region is home to a panoply of intergovernmental initiatives, agreements, and organizations. The intergovernmental organizations are arenas in which the political regimes of the region interact. The adoption of agreements and treaties by these organizations, moreover, entails a diffusion of norms, rules, and regulations across the relevant states. Next to the multilateral mechanisms of intergovernmental coordination, autocratic rule in the region is also advanced through the countries individually. Much of the focus has been on Russia’s purported efforts to promote authoritarianism in the post-Soviet area (Ambrosio 2009; Babayan 2015; Tolstrup 2009; Way 2015). Less often described are processes of authoritarian diffusion through which countries in the region learn from the best practices of other states in the region (primarily Russia) and copy those states’ legislation or policies (Ambrosio 2010; M. Bader 2014; Vanderhill 2012). The intergovernmental organizations through which most coordination takes place are the Commonwealth of Independent States, the Collective Security Treaty Organization, and the Shanghai Cooperation Organization. The CIS was formed right after the dissolution of the Soviet Union with the goal to maintain a degree of coordination and cooperation among the newly independent post-Soviet states. All newly independent states joined, except for the Baltic states; Georgia left in 2008. The CIS has failed to become particularly influential (Kubicek 2009). One of the areas in which the organization has been active throughout, via its Parliamentary Assembly, is relevant to my analysis: efforts to harmonize the legislation of the member states through the publication of model laws that serve as a template for adoption by the member states. The CSTO is a military alliance of Armenia, Belarus, Kazakhstan, Kyrgyzstan, Russia, and Tajikistan, which was established in 2002. Uzbekistan was a member of the alliance between 2006 and 2012. The alliance organizes joint military exercises and has cre-

Coordination and Crisis Prevention

145

ated a Collective Rapid Reaction Force. The SCO, finally, is an organization that deals primarily with security issues in and around Central Asia. Its members include Russia, China, and the Central Asian states of Kazakhstan, Kyrgyzstan, Tajikistan, and Uzbekistan. Insulating Against Electoral Revolutions

International election observation missions, most prominently those organized by the OSCE, carried great political significance in the color revolutions (Bunce and Wolchik 2011). The OSCE in particular was seen by many as a neutral arbiter in the election process. Moreover, people learned about fraudulent practices in the elections in part through the reports of international election observation missions. Since the beginning of the century, the CIS and its member states have taken a range of measures to shield themselves from the potential damage of international election observation to their regimes. The OSCE has observed elections in the region since the early 1990s. Given the political context in which elections generally are held in the region, it is not surprising that reports from the OSCE election observation missions have often been highly critical of the quality of the electoral process and the degree of compliance with international standards. Seemingly in response to these critical assessments, the consolidated authoritarian states of the region have started to organize parallel election observation missions that invariably issue favorable statements about the elections that they observe (Fawn 2006; Walker and Cooley 2013). The primary organization that has been used for the purpose of shadow election observation is the CIS, with smaller missions having been organized by the SCO. The first CIS election observation mission was deployed in 2001 to the presidential election in Belarus.3 Since 2001, CIS missions have been organized in all CIS states. Documentation from the CIS asserts that CIS election observation now has equal standing with that of the OSCE.4 As Roy Allison observes, “Monitoring activity of this type, prompted by Russia and conducted under the aegis of CIS and SCO structures, now contests the OSCE’s terms of democratization. It is seeking to create alternative rules and practices for democratization and election observation and to confer legitimacy on this basis to the leadership of those states under political scrutiny” (2008, p. 190). The CIS also asserts that the organization has developed a methodology that is at least as sophisticated as that of the OSCE.5 The CIS methodology mimics that of the OSCE in several ways. While OSCE observers, however, fill out forms at each polling station containing some sixty to seventy questions, CIS observer forms contain just twelve questions that are broadly phrased (“Are there violations of international standards for democratic

146

Max Bader

elections?”) and that do not allow for establishing a comprehensive picture of the quality of election procedures.6 In addition to organizing joint CIS election observation missions, a more recent trend in shadow election observation has been to invite individual election observers who, like the CIS missions, issue favorable statements and praise the elections in local media outlets. It appears that Belarus initiated this practice in 2010. For the 2010 presidential election in Belarus, the Central Election Committee accredited ninety individual observers, whose favorable comments on the election processes were widely cited by the Central Election Committee. 7 In 2011, Russia and Kazakhstan followed the Belarusian example, inviting 51 and 112 observers respectively to their elections.8 The fact that many of the same individual observers are deployed to different countries in the region suggests coordination among these countries. Election observation by the OSCE is based on provisions made in the 1990 Document of the Copenhagen Meeting of the Conference on the Human Dimension of the CSCE.9 Despite the fact that all CIS member states are also members of the OSCE, the CIS has produced its own Convention on Standards of Democratic Election, Voting Rights and Freedom in the States of the Commonwealth of Independent States.10 The two documents reveal different points of view on the importance and value of elections. While the Copenhagen document focuses on elections as an instrument that expresses the will of the people and which forms the “basis of the authority and legitimacy of all government,” the CIS 2002 draft convention on election standards, electoral rights and freedoms argues that “elections are one of the political and legal instruments of a stable civil society and a solid development of the state.” OSCE election observation missions are formally invited by the recipient country. As part of each mission, the OSCE publishes statements and reports during different stages of the election process. The core idea behind the missions is that the OSCE, as a collective organization, observes shortcomings in the election process, on the basis of which it issues recommendations that the recipient country is expected to subsequently implement. Rather than being receptive to the OSCE’s observations and conclusions, however, the authoritarian states of the region in recent years have started to actively put forward rebuttals to the OSCE statements and reports. Belarus, Kazakhstan, and Russia in particular now have adopted the practice of delivering point-by-point rebuttals of OSCE reports.11 Organizing shadow election observation missions and publishing rebuttals to OSCE reports have been ways of defusing OSCE missions. The consolidated authoritarian states of the post-Soviet area have also been assertive in criticizing OSCE election observation in general. Most prominently, the

Coordination and Crisis Prevention

147

CIS issued a statement in 2004 “about the state of affairs in the OSCE,” which among other things addresses election observation.12 In the statement, the CIS leaders berate the politicized nature of OSCE election observation in particular. A common strand in the criticism of CIS leaders about OSCE election observation is that the OSCE has no clear criteria in its election observation, opening the door to the application of double standards in the observation.13 Both Russian and Belarusian officials have argued that, by contrast, the CIS does have the necessary unambivalent standards and criteria to carry out adequate election observation (Churov 2007; Lozovik 2011). CIS secretary-general S. N. Lebedev has proposed developing joint standards for the international observation of elections based on the 2002 CIS convention and other CIS documents (Lebedev 2013). Political Solidarity

The violent uncontrolled uprisings of protestors outside the capital city of a number of post-Soviet states have made it clear to the regimes that they all face similar threats. The events have also alerted the authorities of these countries to the need of having capable coercive forces that can be used to stifle unrest. Through the CSTO and the SCO, the regimes have worked on enhancing their coercive capabilities while indicating that they will support each other in case of future unrest. The coercive capabilities of the regimes to suppress unrest have been enhanced in three different ways. First, as part of a CSTO program, officers from Central Asia have received military training in Russian academies to counter threats such as the Andijan uprising (Frost 2009). Second, both the CSTO and the SCO have organized military exercises that mimick local uprisings (Silitski 2010). According to one report, for instance, an SCO military exercise in 2007 was modeled on the events in Andijan, “suggesting that future SCO operations may in fact be used for quashing rebellion, or dealing with political instability” (McDermott 2007, p. 3). Finally, the CSTO has created a collective force that, at least in theory, can be rapidly deployed when deemed necessary. A 2002 agreement first formed the Collective Rapid Deployment Force, which later was transformed into the Collective Rapid Reaction Force. These forces also partake in collective military training (Tolipov 2009). The member states of the CSTO and the SCO have expressed commitments to support each other in times of crisis. The founding document of the CSTO in particular contains an article that resembles NATO’s Article 5: “[The signatories] view an armed attack (aggression), other challenges and threats aimed at one or several Signatories as an armed attack (aggression), other challenges and threats aimed at all Signatories and will

148

Max Bader

undertake corresponding countermeasures using all powers and means that are at their disposal.”14 In a more general sense, the CSTO and the SCO have served to express political solidarity among the leaders of the region (Allison 2008). When the Kazakh government faced the crisis in Janaozen in 2011, for instance, the CSTO expressed its support for the government.15 Rather than values such as democracy and human rights, the documents of the SCO and the CSTO emphasize state sovereignty and stability. The organizations are inherently conservative in that they seek to preserve the political status quo in the region. The SCO in particular speaks out against imposing foreign norms through democracy promotion, and criticizes liberal interventionism (Ambrosio 2008). The SCO and the CSTO are often used as platforms from which criticism against the West is voiced. According to some, being an anti-Western alliance is actually the main rationale behind the existence of the SCO (Melnykovska, Plamper, and Schweickert 2012). As Flemming Splidsboel Hansen notes, “In the future, the SCO is likely to serve still more as a forum for the rejection of Western criticism and for the justification of current domestic rule” (2008, p. 223). Notably, when one of the former leaders of a CSTO member state was violently removed from power, the CSTO did not deploy its rapid reaction force or otherwise come to Kyrgyz President Kurmanbek Bakiyev’s rescue. The CSTO also did not intervene in the country when several hundred people were killed in interethnic violence in the south of Kyrgyzstan a few months later. The failure to intervene in Kyrgyzstan in 2010 is seen as an indication that the actual commitment of the member states to maintaining stability in the CSTO region is limited (Özdal 2011). Model Legislation

Participants in the crisis events in the post-Soviet area have often been labeled as terrorists and extremists. After the 2005 events in Andijan, Uzbekistan, for instance, the authorities claimed that people who had taken the local government headquarters were “terrorists.”16 The crackdown against the people who participated in the events in Khorog, Tajikistan, was presented by the authorities as a counterterrorism operation (Mirzobekova 2013). Protestors who rallied against the regime in Minsk on election night in 2010 were portrayed as “extremists” (Kuznetsov 2010). The Inter-Parliamentary Assembly of the CIS has advocated the harmonization of legislation on extremism and terrorism through the publication of model laws, from which counterextremism and counterterrorism legislation has spread to most of the consolidated authoritarian states of the region. Accusations of extremism in particular have increasingly become widespread in the prosecution of regime opponents. The first counterex-

Coordination and Crisis Prevention

149

tremism law in the region was adopted in Russia in 2002.17 Subsequent amendments to the law have widened the scope of what is considered as extremism, and expanded opportunities to use the law for repression. As Stephen White argues, “Amendments to the law on extremism allowed any politically or ideologically motivated crime to be designated in this way, and the law was itself being used arbitrarily to initiate cases against NGOs, activists and the independent media, including internet sites and blogs” (2010, p. 273). According to the 2006 amendments, extremist activity now also included, for example, slander against government officials and impeding the work of government bodies and election commissions through violence or the threat thereof.18 The Inter-Parliamentary Assembly of the CIS agreed on a model law on counterextremism in 2009, which largely mirrors the 2002 Russian law including the 2006 amendments.19 Since 2002 Belarus, Kazakhstan, Tajikistan, and Turkmenistan have adopted counterextremism laws that borrow heavily from the 2002 Russian law and the 2009 CIS model law.20 The authorities in these countries, as in Russia, have increasingly utilized the counterextremism legislation in criminal prosecutions, including those against political opponents. In Kazakhstan, for instance, the oppositional political party Alga as well as several media outlets have been banned because of alleged extremist activity (Freedom House 2013). In Russia, videos of the band Pussy Riot, several members of which were jailed after staging an impromptu performance in Moscow’s main cathedral, have been banned for extremism (Balmforth 2012). Independent media outlets in Belarus have also been banned after allegedly publishing extremist material (Aliaksandrau and Bastunets 2014). CIS member states have also harmonized their counterterrorism legislation. The first counterterrorism law in the region was adopted by Russia in 1998.21 Months later, the Inter-Parliamentary Assembly of the CIS agreed on its first counterterrorism model law, based largely on the original Russian law. 22 Similar counterterrorism laws were subsequently adopted in all other consolidated authoritarian states of the region between 1999 and 2003. In 2006, Russia adopted a new counterterrorism law containing a revised definition of “terrorism” that has been seen as problematic from a human rights perspective. 23 While the definition of terrorism in the 1998 law focused on “violence or the threat of violence,” the definition in the 2006 law understands terrorism primarily as an “ideology of violence.” In 2009, the Inter-Parliamentary Assembly of the CIS adopted a new model which, inter alia, also contains this definition. 24 Both Belarus and Kazakhstan amended their counterterrorism laws in 2012 to incorporate the definition found in the CIS model law.25 The new definition is regarded as problematic because it lends itself to potential abuse by authorities who may label (political) opponents as “terrorists”

150

Max Bader

(Omelicheva 2009). Moreover, the new definition is incompatible with definitions from existing international conventions (International Federation for Human Rights 2009). For the participating states of the OSCE and the member states of the Council of Europe, it is customary practice to request these organizations to comment on new legislation. CIS member states, however, have generally circumvented the OSCE or the Council of Europe when adopting counterextremism and counterterrorism legislation. Only the government of Kazakhstan once requested that the OSCE comment on its counterextremism law. In their legal review, the OSCE experts argued that the definitions in the law are imprecise and therefore open to abuse, and suggested that a special counterextremism law is not really necessary. The review also warns that “counteractive measures on extremism must not serve as a pretense for suppression of oppositional political or religious groups” (OSCE/ODIHR 2004). Conclusion Post-Soviet Eurasia is, for the most part, a bastion of authoritarian rule. In Georgia, Kyrgyzstan, and Ukraine, however, authoritarian leaders have been removed from power on five occasions since 2003. The authoritarian leaders of the region tend to point to Western influences as the core reason behind the regime turnovers in Georgia, Kyrgyzstan, and Ukraine. While it is unlikely that Western forces actively seek to engender regime change in post-Soviet states, it is widely believed that linkages with Western democracies make authoritarian regimes more vulnerable, and that such linkages can help explain some of the regime changes in post-Soviet Eurasia. Unlike past regimes in Georgia, Kyrgyzstan, and Ukraine, the political regimes of Azerbaijan, Belarus, Kazakhstan, Russia, Tajikistan, Turkmenistan, and Uzbekistan have successfully resisted pressure for democratization in this century and now seem virtually immune to linkages with Western democracies. This is in part a result of coordinated efforts to decouple the regimes from linkages with the West. The regimes have shielded themselves from the potentially costly impact of international election observation by organizing alternative election observation missions and by establishing alternative norms for holding elections. Through the CSTO and the SCO, they have strengthened their coercive capabilities, created an alliance with a major nondemocracy (China), and pledged commitment to preserving political stability in the region. They also have adopted and harmonized new repressive legislation through the CIS while circumventing the OSCE and the Council of Europe in the process.

Coordination and Crisis Prevention

151

The combined effect of these efforts has been the establishment, to some extent, of a community of nondemocratic status quo powers in the region. This grouping is far from internally coherent and consistent in its operations as, for example, the failure to act in Kyrgyzstan in 2010 shows. There is variation in the degree to which the states participate in the coordination efforts, with some regimes in the region choosing to opt out altogether. Moreover, despite the coordinated efforts to prevent crisis in the region, crises still occur. Cutting linkages with the West makes sense from the perspective of the authoritarian regimes: strong linkages with the West arguably undercut the legitimacy of the regime, as more people would be likely to have some insight into both the normative shortcomings and the poor performance of the authoritarian neopatrimonial regimes. The efforts to reduce linkages with the West, however, are undertaken at the expense of addressing domestic sources of instability and discontent. Whether intergovernmental coordination in the region will continue depends largely on Russia. The Russian authorities are currently the driving force behind most, if not all, of the coordination efforts. The regimes are more affected by their linkages with Russia than by linkages with the West. In a world that has become increasingly more democratic and globalized, it is questionable whether the post-Soviet autocracies will be able to maintain authoritarian rule without a nearby major political ally. The future of authoritarian rule in the region therefore hinges on the political trajectory of Russia. Notes

1. “İsmayıllıda xalq üsyani,” see www.azadliq.info/siyast/26416-ismayillida-xalq -usyani.htm. 2. Data from the Worldwide Governance Indicators project demonstrate the poor performance of the regimes in terms of good governance. See http://info .worldbank.org/governance/wgi/index.aspx#home. The Corruption Perceptions Index of Transparency International shows that the countries are seen as highly corrupt. See www.transparency.org/cpi2014. 3. “Elections of the President of Belarus 2001,” see http://e-cis.info/page.php ?id=19795. 4. “Reshenie o praktike raboty Missii nabliudatelei ot SNG na prezidentskikh i parlamentskikh viborakh, a takzhe na referendumakh v gosudarstvakh—uchastnikakh Sodruzhestva Nezavisimykh Gosudarstv ot 10 aprelia 2009,” see http://e-cis .info/page.php?id=22079. 5. Ibid. 6. “Rekomendatsii dlya mezhdunarodnykh nablyudateley ot Sodruzhestva Nezavisimykh Gosudarstv po nablyudaniyu za vyborami i referendumami (novaya redaktsiya),” see http://old.iacis.ru/data/prdoc/R18.rar. 7. “Inostrannye nablyudateli o pervom ture vyborov. Po materialam informatsionno-politicheskogo tsentra ZIS,” see www.rec.gov.by/sites/default/files/pdf/Archive -Elections-PPNS2-Soob6.pdf.

152

Max Bader

8. “Sprisok inostrannykh (mezhdunarodnykh) nablyudateley, akkreditovannykh TsIK Rossii ot ekspertov, obladayushchikh priznannym avtoritetom v oblasti izbiratel’nogo prava,” see www.cikrf.ru/banners/duma_2011/international/npo.doc; “Pochti 700 nablyudatelei akkreditovany na vybory v Kazakhstane,” see http://ria.ru /world/20120106/533802666.html#ixzz3Eo4g4ruo. 9. Document of the Copenhagen Meeting of the Conference on the Human Dimension of the CSCE is available at www.oscepa.org/publications/reports/special -reports/election-observation-reports/documents/1344-osce-copenhagen-document -1990-eng/file. 10. “O standartakh demokraticheskikh vyborov, izbiratel’nykh svobod i prav v gosudarstvakh—uchastnikakh Sodruzhestva Nezavisimykh Gosudarstv” (2002), see http://cis.minsk.by/reestr/ru/index.html#reestr/view/text?doc=1312. 11. “Svodnyi kommentarii na predvaritel’nye zaklyucheniya i vyvody missii OBSE po nablyudeniyu za vyborami Prezidenta Belarus’ 19 marta 2006 goda,” see www.rec.gov.by/sites/default/files/pdf/Archive-Conference-2006-Inf4.pdf; “Kommentarii i zamechaniya kazakhstanskoy storony k promezhutochnomu otchyotu Missii BDIPCh/OBSE po nablyudeniyu za vyborami (12 avgusta 2007 goda),” see www.zakon.kz/4452296-kommentarijj-i-zamechanija.html ixzz3Eo4g4ruo; “Kommentarii k Predvybornomu otsenochnomu otchyotu BDIPCh OBSE 17-22 avgusta 2011 goda Vybory v Gosudarstvennuyu Dumu 4 Dekabria 2011 goda,” see http:// cikrf.ru/news/relevant/2011/09/29/churov.html ixzz3Eo4g4ruo. 12. “Zayavlenie gosudarstv-uchastnikov SNG otnositel’no polozheniya del v OBSE,” see http://ecis.info/page.php?id=2096. 13. “Aspekty mezhdunarodnogo nabliudeniia BDIPCh OBSE za vyborami: praktika i puti sovershenstvovanniya: Doklad N. I. Lozovika na mezhdunarodnoi nauchno-prakticheskoi konferentsii Vybory i razvitie demokratii v Kyrgyzstane,” see www.rec.gov.by/sites/default/files/pdf/Archive-Conference-2006-Inf3.pdf; “O sotrudnichestve Respubliki Kazakhstan s Biuro OBSE po demokraticheskim insitutam i pravam cheloveka (BDIPCh/OBSE),” see www.contur.kz/node/1049; “Dvoinye standarty BDIPCh OBSE na primere Irlandskikh vyborov,” see http:// cikrf.ru/international/events/dv_stand.html. 14. “Soglashenie o poryadke formirovaniya i funktsionirovaniya sil i sredstv sistemy kollektivnoi bezopasnosti Organizatsii Dogovora o kollektivnoi bezopasnosti,” Article 2, see www.odkb-csto.org/documents/detail.php?ELEMENT_ID=1681. 15. See “Kazakhstan polnost’yu kontroliruet situatsiiu v Zhanaozene, shchitaiut v ODKB,” see http://ria.ru/politics/20111220/521693454.html. 16. “Genprokuror Uzbekistana: terrorirsty v Andizhane ubili zalozhnikov,” see http://ria.ru/incidents/20050517/40370371.html. 17. “O protivodeystvii ekstremistkoi deyatel’nosti,” No. 114-F3, July 25, 2002. 18. “O vnesenii izmeneniy v stat’i 1 i 15 Federal’nogo zakona O protivodeystvii ekstremistkoi deyatel’nosti,” No. 148-F3, July 27, 2006. 19. “Model’nyi zakon O protivodeystivy ekstremizmu,” No. 32-9, May 14, 2009, see http://base.spinform.ru/show_doc.fwx?rgn=30827. 20. “Zakon respubliki Belarus’ O protivodeystivy ekstremizmu,” No. 203-3, January 4, 2007; “Zakon Respubliki Kazakhstan O protivodeystivy ekstremizmu,” No. 31, February 18, 2005; “Zakon Kyrgyzskoi Respubliki O protivodeystvii ekstremistkoi deyatel’nosti,” No. 150, August 17, 2005; “Zakon Turkmenistana O protivodeystivy ekstremizmu,” March 1, 2015, see www.turkmenistan.gov.tm/?rub=12. 21. “O bor’be s terrorizmom,” No. 130-F3, July 25, 1998. 22. “O bor’be s terrorizmom,” see www.iacis.ru/upload/iblock/d89/075.pdf. 23. “O protivodeystviy terrorizmu,” No. 35-F3, March 6, 2006.

Coordination and Crisis Prevention

153

24. “O protivodeystviy terrorizmu,” see http://www.iacis.ru/upload/iblock/e81/ass _33_15a.pdf. 25. “O vnesenii izmeneniy i dopolneniy v nekotorye zakony Respubliki Belarus’ po voprosam bor’by s terrorizmom i protivodesyvtiya ekstremizmu,” No. 435-3, October 26, 2012; “O vnesenii izmeneniy i dopolneniy v nekotorye zakonodatel’nye akty Respubliki Kazakhstan po voprosom protivodeystviy terrorizmu,” No. 63-V 3RK, January 8, 2013.

8 The Resilience of Authoritarianism: The Case of the Chinese Developmental State Thomas Heberer For quite a long period of time, social scientists argued that authoritarian systems were per se economically inefficient, noninnovative, and lacking in legitimacy and political authority (e.g., Barma and Vogel 2008; Calvert 2005; Habermas 1997; Merkel 2010). Moreover, such states are classified as weak states “inherently fragile because of weak legitimacy, overreliance on coercion, over-centralization of decision-making, and the predominance of personal power over institutional norms” (Nathan 2003, p. 6). However, China’s economic success and relative stability in recent decades have shown that such postulates are far too simplistic. Among China scholars, we find a broad discussion of the reasons why the power of the Chinese Communist Party (CCP)—in contrast to that of the former Soviet Union—not only has not collapsed, but in fact thus far appears to be relatively stable. Twenty-five years after the urban protest activities and unrest of 1989, the Chinese political system seems to be more stable than in the 1980s, and the reform process in the meantime has also gone much farther and been more far-reaching than at the end of the 1980s. Western scholarship has tried to find an answer to the question of why China is managing to develop so smoothly and what are the reasons for its stability. US scholar Andrew J. Nathan speaks of a “resilient authoritarianism” and argues that the key factor in China is “an autocratic system responsive enough to societal demands to keep itself in power for a long time” (2003, p. 16). US policy adviser Robert A. Kagan (2009, p. 3) speaks of a new “authoritarian consolidation.” Political scientist Bruce Gilley (2009), in turn, declares that China’s political leadership attains legitimacy by means of continuous institutional change and performance. Gary Sigley (2006) speaks of the regrouping of the party state, David Shambaugh (2008) of atrophy and adaptation of the party state,1 and Sebastian Heilmann and Oliver Melton (2013) of the successful transformation of the planning system (adaptive planning) and the incorporation of experimental programs into 155

156

Thomas Heberer

macroplans. For Baogang He and Mark E. Warren, in turn, the development of a deliberative authoritarianism (He and Warren 2011; He and Thogersen 2010), and for Thomas Heberer and Gunter Schubert (2012) the agency of “strategic groups” of local cadres, along with the policy process in the local state, are crucial for policy implementation, resulting in systemic adaptiveness (Ahlers, Heberer, and Schubert 2016) and resilience.2 For their part, Jennifer Gandhi and Adam Przeworski raise the question of why “some autocrats survive for decades, and others fall soon after taking power” (2007, p. 1279). A major question therefore is whether the concept of the “authoritarian state” is appropriate for describing and analyzing such complex institutions, structures, and processes as those existing in China.3 Peter Evans (2004, p. 30) alerts us to be cautious of “institutional monocropping”; that is, “the imposition of blueprints based on idealized versions of Anglo-American institutions” that neglect the distinct features and institutional-organizational particularities of specific countries. A major assumption of traditional modernization theories was, for instance, that authoritarian states would inevitably turn into democratic states. Johannes Gerschewski et al. (2013) note that forecasts about the inevitable victory of democracy over all other regime forms were of a “speculative nature.” On the one hand, authoritarianism does not explain the mode of operation of a system and the quality of its institutions. On the other, it cannot explain why we find both stable and unstable authoritarian systems. With regard to China, social, political, and institutional processes in the meantime have become more complex, diverse, and fragmented, involving a broad variety of actors. We therefore need to analyze the inner dynamics and the logic and function of this system’s operation. Adam Przeworski (2016) sums up this dilemma in the phrase that the Soviet Union failed but China flourishes, arguing that the one-party state cannot be the cause of this difference. While Western scholarship was focusing on the question of which systems are the best ones, emphasizing the salient role of steering by laws and rights, the crucial question in traditional Chinese political philosophy was the question of how to preserve order so as to avoid disorder. The (Confucian) rites, it was taught, should be used to steer society and people by an entirety of etiquettes and should impact on individuals and social groups to make them think and behave in a specific way. The latter, of course, is beyond the horizon of this analysis. Rather, my point here is that political science apparently has not yet developed appropriate tools and categories for classifying the economic and social development and stability of authoritarian political systems, such as that of China, in a more differentiated way. In this chapter, I therefore start out from the function of the Chinese state. I argue that China is functioning as an effective developmental state at both the central and the local level. I classify the current type of devel-

The Resilience of Authoritarianism

157

opmental state as an advanced model, calling it “developmental state 3.0.” It is a strong state exhibiting state capacity. Local governments figure also as developmental agencies, which steer development according to the specific local features of a locality. I further argue that, due to its successful developmentalism, the Chinese state is capable of effectively tackling crises and displays both (relative) stability and (relative) legitimacy. Relative stability and relative legitimacy are key characteristics of developmental states. Finally, I show that the local developmental state also contributes to the stability and legitimacy of the entire political system. This chapter is structured in the following way. Beginning with a characterization of the Chinese state at the central and the local level as a developmental state, I trace the features of this state and examine developmentalism by analyzing the role of the local state in steering and developing the private sector. I scrutinize why the Chinese state exhibits legitimacy and stability. After that, I outline the modification of Chinese developmentalism at the end of 2013 as a case in which institutions are adapted to processes of change. I conclude the chapter with an outlook on China’s further prospects of its developmental state. The Perspective of the Central State: China as an Effective Developmental State This chapter commences primarily with a functional approach; that is, with a focus on the role of the Chinese party state in steering, shaping, and developing the country and its people. In a second step, I examine the specific operations and structures in the context of this functional role. Accordingly, I take up the concept of the developmental state to better explain and localize the behavior of the Chinese state, and I argue that developmental states exhibit specific features that contribute to explaining the resilience of the Chinese political system. Other authors have also dealt with the role of the state in a developing country and, therefore, with the issue of how state power is used in the developing processes. Dan Slater and Sofia Fenner, for example, maintain that “state power is the strongest foundation for authoritarian regimes in power” (2011). Strong states, so they argue, are the optimal bedrock for a durable and stable authoritarian regime. This, however, does not apply to every strong autocratic regime, but to only those that manage to be effective in policy formulation, decisionmaking, and policy implementation. Moreover, even strong states are not immune to leadership errors and failures. The concept of the developmental state was originally advanced by Chalmers Johnson in the 1980s and first applied to Japan’s development.

158

Thomas Heberer

Johnson (1982) argues that, in late-industrializing countries, the state would act as a developmental driving force. Developmental states were classified as those in which the state has taken over macroeconomic planning and steering, continuously intervenes in the economy, and acts relatively autonomously (Johnson 1982). Later, the concept was taken over as a tool for analyzing the development of other countries in East Asia (e.g., South Korea, Taiwan, and Singapore). With this as a basis I argue here that, in spite of all its problems and difficulties, the Chinese state can be classified at both the central and local levels as a typical “developmental state.”4 Linda Weiss (2000) posits that developmental states are characterized by three criteria: transformational development priorities, the existence of a developing elite, and a strong interaction between government and the economy. These criteria can be further refined. Developmental states differ from so-called developing countries in that the former are purposeful states characterized by the will and consent of the political elites to develop in a structured way, in a predominantly top-down manner, and under a unified leadership (in China the CCP).5 They are capable of promoting development successfully and against all odds and obstacles in the domains of politics, economics, and society; they are effective at intervening in the economy and controlling labor; and they exhibit a strong interaction and a symbiosis between government and enterprises. State intervention can even alleviate market failures.6 Such states can effectively adapt institutions to socioeconomic changes.7 Moreover, they display the ability to enforce their policies and are relatively independent of the interests of societal groups (Evans 1995). “Without such an autonomy from societal actors, the state cannot be seen to control society and regulate social relations across its territories” (Seufer and vom Hau 2008, p. 224), and states cannot make use of power to “discipline their societies” (Kohli 2004, p. 381). They also impose repressive measures against potential opponents of their development goals. Therefore, developmental states are strong states exhibiting state capacity and political stability and legitimacy and have an effective and highly skilled bureaucracy, which is constantly being professionalized, at their disposal. I call the current Chinese developmental state “developmental state 3.0” since it differs from previous forms of developmental states in China. The first (developmental state 1.0), during the Mao Tse-tung era, was characterized by industrialization without modernization; and the second (developmental state 2.0), after 1978, by reform modernization. But developmental state 3.0, since 2013, exhibits a shift of the development model marked by a fostering of policy innovations, an upgrading of industries, a stronger focus on sustainability, and the ability to steer the economy and society by institutionalization and guidance.

The Resilience of Authoritarianism

159

There can be no doubt that the Chinese state meets the criteria of a developmental state. With regard to development, China’s economy has advanced rapidly and successfully in the past twenty-five years, with an average annual economic growth rate of 10 percent since the 1980s. Whereas the United Kingdom required sixty years to double its gross national product and the United States forty years, China needed only twelve years. The Chinese leadership is well aware that central planning tends to exclude people who want to work for a common goal while a market economy integrates them. China effectively implements policies by which political power can contribute positively and effectively to economic well-being—for instance, by focusing on long-term growth and structural change as pivotal goals, along with shrewd political management of the economy, institution building, and innovation. The state acts in a pragmatic way by abandoning traditional organizational patterns whenever this is required by the developmental process. Strong indications of political pragmatism in China are the abandonment of mandatory planning and its replacement by macroeconomic indicative planning, the development and fostering of market economic structures, and the abandonment of the class character of the CCP when it argued that from now on it would represent the interests of the entire Chinese people and no longer those of particular classes (Jiang 2001). Whereas for Chalmers Johnson authoritarianism apparently was not a necessary marker of developmental states, the successful ones in East Asia exhibited or (in the case of China) exhibit strong features of an authoritarian developmentalism. Concurrently, Johnson admits that developmental states are “normally authoritarian.” The reason, he argues, is that they want to preserve political stability and long-term predictability since their legitimacy is based on their performance and effectiveness (Johnson 1989, p. 143). Authors such as Kenichi Ohno speak of the developmental state as an “authoritarian state with economic capability” (2013, p. 38), and Adrian Leftwich notes that such states are characterized not only by successful development but also by a mixture of political repression, a poor human rights record, and the absence of democracy (2000). However, it is not the character of the regime that was and is decisive, but “the character of the state and its associated politics” (Leftwich 1993, p. 614). In weakly institutionalized states with a weak civil society, state capacity requires a strong state that in turn may display strong features of an authoritarian state. This concurrently shows that authoritarianism is not necessarily an obstacle to development, but can—under specific conditions and given an effective state capacity—become an authoritarian developmentalism. I believe that an effective development road map and its successful implementation, institutionalization, and adaptiveness in terms of institutional changes are much

160

Thomas Heberer

more crucial in classifying a developmental state than the pure attribute of “authoritarian.” In addition, autocratic regimes exhibit greater durability, since parties such as the CCP are able to “curb leaders’ ambitions and bind together political coalitions” (Brownlee 2007, p. 37). They are not only successful at mobilizing mass support, but also have an internal disciplinary effect. State capacity is a crucial feature of developmental states. From the perspective of a centralized state, the term state capacity refers to the capability of a state to advance and enforce political decisions and programs. In enforcing its policies, the Chinese developmental state has proven to be a strong state that has the resources and the capacity to steer and implement its policies throughout the country. This kind of state capacity encompasses six major elements. First is transformational capacity, the capability to translate national goals into a stringent development concept and effectively implement this concept top down and by mobilization power (Gandhi 2008). Second is legitimacy, the enhancement of political output through acceptance and support by society, and stability of the political system. Third is capacity for social regulation and social control, the gradual institutionalization of the political entity, with both steering and control capacity concerning the entire economy and society. Fourth is enforcement capabilities (e.g., availability of financial and personnel resources and effective use of these resources), along with means for coercion (coercive capacity,8 including repression), the ability to get things done, to the point of disciplining social groups and opponents to bring them into line with reforms. Fifth is bargaining capacity, the state’s ability to manage relations with social actors effectively; to incorporate new social groups, associations, and organizations into bargaining processes; to co-opt potential opposition forces; and to find a balance between various particularistic interests. And sixth is learning capacity, the ability to adapt when formulating policies and promoting social innovations, and the ability to adjust to institutional and political changes (Tsai and Dean 2013). “Learning” here means learning from previous mistakes and failures and the ability to find new solutions for problems. State capacity in this sense is important for implementing an effective development program and for effectively coping with domestic crises, problems, and conflicts. Interaction Between Two Kinds of Developmental States: The Central State and the Local State The role of the local state for state capacity and authoritarian resilience has long been underestimated. However, it is not only the central state that acts as a developmental agency, enforcing “high modernism,”9 but also the local state (cities, counties, towns, and townships). The local state in China is

The Resilience of Authoritarianism

161

tasked with implementing the policies of higher authorities within the bounds of the central guidelines, but it has gained more discretionary power in that policy implementation (i.e., it is to execute development according to specific local resources and features). It must primarily solve local problems and ensure social stability. Thus, the central state transfers decisionmaking powers to the local level on the one hand while closely monitoring the performance of leading local cadres on the other. Policy modeling and policy experimentation at the local level are crucial for China’s local cadres for them to be noticed by higher authorities and improve their opportunities to be promoted to higher positions. Accordingly, they are under considerable pressure from the higher authorities to create local models. From the perspective of the center, local governments should be innovative and should produce effective policy experiments that can be replicated elsewhere. Policy experiments are to lead to the creation of models (shifan) and experiments (shidian); that is, the formal pursuit of political innovation (Heilmann 2008). To become a model, a location must abide by the respective rules and regulations of the center, the province or the prefectural city, whereby the latter will regularly check the model character and its development.10 Evaluation reinforces the pressure to devise such policy experiments locally and to copy from successful ones since they indicate how advanced a locality and its leadership are. Models are evaluated by higher authorities in an annual process of program evaluation, and the creation of successful models may positively affect the standing of the local leadership and its evaluation scores (Heberer and Trappel 2013). In counties that I investigated between 2009 and 2013, for instance, I encountered a broad variety of top-down local developmental models. There were environmental or ecological models such as Deqing County (Zhejiang Province), a national ecological model; Shouguang City (Shandong Province), a model for green agriculture; and tourism development models—Meitan County (Guizhou Province), a model for agrotourism, and Xifeng County (Guizhou Province), a red tourism model.11 Suining City (Sichuan Province) in turn was a model for religious tourism, Meigu County (Sichuan Province) for developing tourism by instrumentalizing the traditional Bimo culture,12 and others such as Nanfeng County (Jiangxi Province) were models for afforestation and preservation of the traditional Nuo culture. Finally, Jiangyin City (Jiangsu Province) was a model for creating a welfare-oriented society and Shiquan County (Shaanxi Province), a model for establishing education for left-behind children of rural migrant workers. The particularity of this modeling is that all these counties represent models that have been developed according to a given local environment. In recent years, higher authorities expected each city or county to advance a model for its development based on its material environment, resources,

162

Thomas Heberer

economic structure, and financial opportunities. Such models or experiments inform higher echelons about the development path that a local government intends to follow. Thus, they are crucial for acquiring resources, political appreciation, and optimal evaluation by higher authorities. Such innovations were in fact being produced in great numbers. Evaluation regulations for local cadres were largely responsible for this. As my colleague Christian Göbel and I argue elsewhere, “Leaders have to identify a problem, come up with a policy, and, most importantly, market this policy by convincing other departments and the general public to participate” (Göbel and Heberer forthcoming).13 Local Developmentalism: The Case of Private Entrepreneurship Developmental modeling at the local level also extends into the economic sphere. During my field research in southern China between 2012 and 2015, I discerned five primary models of private sector development:

1. The Pearl River Delta model, characterized by growth by means of foreign investment. 2. The southern Jiangsu model: development of native entrepreneurship through privatization of former collective-owned township and village enterprises (TVEs); and strong connections between local enterprises and the local government. Many private entrepreneurs had previously been managers in the former collective TVEs that prevailed up to the 1990s and, thus, are party members. Accordingly, there exists a strong intertwining between local enterprises on the one side and local party and government organizations on the other. The local governments are responsible for structural policies and crisis management. 3. The Jinjiang model in Fujian Province: TVEs did not exist here. Rather, a native entrepreneurship, which initially was labeled as illegal, has emerged bottom up. The majority of entrepreneurs are not party members; nonetheless, there is a strong network of relationships between local enterprises and the local government and party organization. Local governments are responsible for structural policies and crisis management. 4. The Wenzhou model, characterized by intensive export production on the part of native entrepreneurs. Originally, the local government took a rather laissez-faire approach toward the private sector here. Due to economic crises in 2008 and 2010, however, the government began intervening and steering the private sector more strictly.

The Resilience of Authoritarianism

163

5. Less developed areas: as a rule, few industrial enterprises exist here (primarily in the areas of agriculture and agricultural product processing). The local government attempts to attract external investment and to promote the tertiary sector (e.g., tourism).

These diverging models exemplify the fact that private sector development at the local level in China does not follow a uniform design, but is rather tailored according to specific local conditions. That is, local governments steer local development by a multitude of means: restructuring industries, supporting the upgrading of products and industries, land administration, helping enterprises to get access to credits and loans, providing tax subsidies for innovations and brand products of enterprises, implementing environmental standards, improving the infrastructure, providing public goods, controlling entrepreneurial associations, providing market information, and promoting the further training of workers and staff in private enterprises. As a leading local official argues, “Government is not administering the enterprises but paying attention to both local and business development environments.” 14 The aim of the local governments is therefore to improve the competitive capacity of enterprises on the world and domestic markets and to establish a more effective administration system. Entrepreneurs (whether party members or not) tie themselves closely to the party state, from which they expect political protection and support in return. They interact with the local government on an individual basis rather than as a coherent interest group. Since private entrepreneurs tend to be co-opted by local governments, the interrelationships and interaction between local governments and private entrepreneurs since the late 1990s can be classified as a new pattern of local state corporatism.15 Because the private sector has become increasingly crucial for tax income, positive performance evaluation, and cadre careers, the local state has been forced to retreat from its entrepreneurial activities and become a regulator—a shift of function from immediate leadership to guidance and service provision. Effective state intervention is based on close cooperation between the state and the private economic sector (Kohli 2004). In sum, effective local developmentalism contributes to effective policy implementation in tackling crises and in generating stability and legitimacy in the political system. The Relative Legitimacy of the Chinese Developmental State As argued above, legitimacy is a salient attribute of developmental states. According to the research findings of Chinese and Western scholars

164

Thomas Heberer

(including my own), a sizable majority of the urban and rural population in China supports the political leadership and the regime.16 This support is both specific and diffuse in character. Specifically, the system is perceived by the population as performing efficiently and is supported accordingly; diffuse support, on the other hand, manifests itself as a type of advance payment in the form of trust in the institutions and foundations of political order, independently of concrete system output (Easton 1965). In this context, trust and support appear to be functionally analogous terms: specific trust (specific support) emerges in the form of concrete demands on the system; diffuse trust (e.g., diffuse support) is brought about through sustained experience with institutions and continuous internalization of the system’s demonstrations of trustworthiness (i.e., the presence of reliability). Both types of support have been generated by successful economic development; the capacity to achieve national goals such as reunification with Hong Kong, Macao, and (possibly) Taiwan; the creation of a strong China on an equal footing with the United States; the preservation of a peaceful and stable political order; and the conviction that the CCP has rescued China from a fate similar to that of the former Soviet Union. In the case of China, however, it is less “input legitimacy” (i.e., input from social interest groups, political parties, the legal system) that counts, but rather “output legitimacy” in the sense of policymaking in the sense of utility for Chinese society (Scharpf 1999, pp. 17−18). Gunter Schubert shows that the “structural imperatives of system adaptability and legitimate authority require that a political system identify policies that are relevant for solving critical problems which the system faces” (2014, p. 605). Wolfgang Streeck and Kathleen Thelen define “regime legitimacy” as follows: “A regime is legitimate in the sense and to the extent that the expectations it represents are enforced by the society in which it is embedded” (2005, p. 108). Such a definition is linked to the concept of output legitimacy, emphasizing that the latter is strongly related to the social expectations of citizens and the corresponding political decisions made by a government. In particular, public good provision is a salient factor in creating output legitimacy. Take, for instance, recent tangible benefits for rural residents: in 2005, the agricultural tax was removed, so that peasants no longer have to pay taxes. Rural infrastructures across the country were extensively improved, school tuition fees in rural areas were abolished, a cooperative rural health insurance program was established on a nationwide scale, and a rural old age pension program is currently being implemented throughout the country. All of these programs are heavily subsidized by all levels of government. In addition to output legitimacy, input legitimacy is also gradually increasing in China fostered, for example, by the Internet and microblogging. Concurrently, state organizations at all levels participate in discus-

The Resilience of Authoritarianism

165

sions about the new media, which they conceive of as a kind of input mechanism. Some authors classify this as part of a deliberative or consultative authoritarianism (e.g., Noesselt 2014). Nevertheless, the Chinese people distinguish between the legitimacy of the central authorities and that of local authorities. As the perceived benign state, the central government possesses trust while the malign local authorities are frequently blamed for all grievances and enjoy only a minimum of trust or none at all. Of course, such a dichotomy is by far too superficial. Heberer and Schubert (2012) have shown that in many locations in rural China, effective policy implementation17 and trust in local governments are undeniably present. However, both legitimacy and stability are relative in China: the high growth rates may not last forever, the political system is highly fragmented, and social disparities and social conflicts are on the rise. Legitimacy can also decline due to a performance crisis or value changes (Gilley 2008). In the long run, therefore, economic growth alone is no guarantee for the endurance of legitimacy and stability. Even the party leadership cannot rule out that it might lose its legitimacy if it fails to meet the demands of the people, for example, for improved governance. This is one reason why almost a decade ago the CCP developed a new goal: the creation of a harmonious society and a harmonious legitimacy in which policies are oriented toward the needs of the population, social interests are harmonized, and social justice is attained through a redistribution of wealth and assets (the view of the party state). Whether or not this will be successful remains to be seen (Resolution of the State Council Regarding the Implementation of the Program 2013). Operational errors of the political leadership cannot be excluded (i.e., wrong policy decisions could still have fatal consequences for China’s development, legitimacy, and stability). The Developmental State’s Push for Preventing Crises and Preserving Stability Stability is another crucial attribute of developmental states. Institutionalization—as a precondition for stability—can neutralize threats and render autocracies more durable. The CCP views stability as the crucial basis for avoiding crises and generating legitimacy. One of the most prominent Chinese political scientists even argues that stability is “the crucial value” in China, above all else, and maintains that it is the central objective of public governance and a focal point for avoiding crises (Yu 2013). Development theories elucidate, however, that developmental states by no means develop in a stable harmonious manner. Modernization processes, particularly under conditions of authoritarian state structures and a lack of

166

Thomas Heberer

checks and balances and in the absence of sufficient institutionalization of laws and the legal system, inevitably give rise to social and political conflicts and even crises (Huntington 1968). He shows that states passing through a fundamental transformation from a premodern to a modern society have a huge potential for crisis. This potential encompasses such phenomena as the erosion of traditional social structures, the rise of a stratum of nouveau riche who strive for political power, an increase in geographical mobility, migration from rural to urban areas, growing disparities between rich and poor, improvement of educational standards, and general access to mass media. All of these factors lead to expectations that cannot be met, along with regional and ethnic conflicts regarding the allocation of investments and consumption goods, a growing knack for social organization, and as a consequence increased unsatisfiable expectations from the government. These phenomena all apply to the current situation in China and help to explain the social frustration and dissatisfaction prevalent in its populace (Huntington 1968). This could be an exact depiction of the current conflict lines in China. Avoiding crises and ensuring stability of the developmental state is therefore a top priority of China’s political leadership. I argue here that the Chinese political system currently appears to be relatively stable. In this chapter, the term stability refers to stability of the (political) system. It does not mean an absence of any structural change (persistence), but rather refers to the capability of a system to restore social balance in the case of social disruption, to adjust its institutions and structures to cope with processes of change, and to resolve crises in a way that benefits the regime (Deng Xu, and Shen 1998). The capacity for adaptation leads to incremental changes and therefore gradual transformations in contrast to abrupt fundamental changes with the result of institutional breakdown and a systemic crisis (Streeck and Thelen 2005). “The ultimate form of stability”—Slater and Fenner argue—“does not entail meeting and overcoming crises, but avoiding and . . . resolving crises decisively in the regime’s favor” (Slater and Fenner 2011, p. 17). Institutionalizing is conducive in this regard since institutionalized regimes “also produce a degree of stability in policy-making” (Gandhi 2008, p. 182). The issue and notion of stability may differ according to specific situations. The Chinese development process has consistently been characterized by its capability and capacity to switch back and forth between a normal mode of preserving stability and a crisis mode of policy management. The normal mode is characterized by extensive procedures for finding compromises in decisionmaking. In contrast, the crisis mode becomes effective in acute and destabilizing situations of real or perceived crises. The latter mode is marked by (re)centralization, ideologization, and the

The Resilience of Authoritarianism

167

strict enforcement of political decisions in a top-down manner. In this case, the central leadership of the party claims exclusive jurisdiction over the entire decisionmaking process. The switch between both is by no means predefined. The Communist Party has, however, a set of options available (e.g., recentralization) that can easily be imposed and enforced across the country by means of the party network. Facing critical junctures, the Chinese leadership invariably has switched from a normal to a crisis mode of stability preservation, political decisionmaking, and policy formation and then back to normality, as Sebastian Heilmann convincingly shows (2002). This not only was the case at the end of the Cultural Revolution and during the urban protest movement in 1989. It was also the case in 1992, when then paramount leader Deng Xiaoping traveled to some special economic zones in southern China to put reform policies and reform stability, both of which had widely been stopped after suppression of the protest movement in June 1989, back on track again. In a speech in Shenzhen and other locations, he declared that the reform process and a policy of opening to the outside should be widely expanded, that China should increasingly learn from capitalist countries, and that this should be directed to the maintenance of stability (Deng Xiaoping 1993, 1994). In moments of political crisis, argue Edward Friedman and Joseph Wong, “Political entrepreneurs have to make ‘hard’ choices and potentially effect ‘big’ changes, with learning consequences for the future of the party and the political system” and “adaptive choices . . . reflect the lessons learned from losing” (Friedman and Wong 2008, p. 266). Edward Friedman further contends that China is a “model” of “learning not to lose” based on “effective leadership” and learning to transform (2008, p. 266). During the global financial crisis of 2008–2009, moreover, China also switched its policies to crisis management to maintain stability. At the national level, the Chinese government announced a stimulus package of 4 trillion yuan (US$586 billion) to tackle the impacts of the global financial crisis in China. China’s central bank lowered the base rate, loan guarantees were given for small and medium enterprises, and cheap credits for housebuilding were offered. Exports were stimulated by increased refunds of sales tax (in the interests of peasants), and the prices of grain purchased by the state were raised. Moreover, massive investment packages were adopted for establishing a social security system, enhancing the infrastructure, and improving the educational system. In addition, China used the crisis to restructure its economy from the world’s workbench toward high-tech industries and sustainable growth. Not only was the infrastructure improved, but efforts were made to upgrade and promote the growth of important and promising industries.

168

Thomas Heberer

During my fieldwork in a county in southern Jiangxi Province in 2009, I was able to observe how the local developmental state strove to overcome the financial crisis at the local level to tackle the crisis and to maintain stability. The local leadership there had the task of solving the problem of migrant workers who had become unemployed due to the breakdown of their enterprises in coastal areas and had returned to their relatively poor hometowns. The central government had ordered that all counties facing such a problem were to take the following steps. First, they were directed to attract external investments to create new jobs by approaching successful entrepreneurs who originated from their county, but had become successful entrepreneurs elsewhere in China. These entrepreneurs were asked to invest in new local branch enterprises. By doing so, they gained special access to credits and loans, a preferential choice of land, tax relief or redemption, and government allowances and subsidies. The central leadership provided local governments with specific target numbers of jobs that they had to create. These targets were to be met under any circumstances to ensure local government leaders of career advancement. Fulfillment of the targets was rewarded, nonfulfillment resulted in salary reductions. Local officials told me that these conditions put tremendous pressure on them since success in meeting them was a precondition for further promotion. Second, local enterprises were ordered to employ additional people (jobless migrant workers). And third, the county government organized professional training to qualify jobless migrant workers for work in local enterprises or to qualify them for self-employment. Such (authoritarian) steering measures helped to rather rapidly solve the unemployment problem of returning migrant workers in 2009. Apart from a normal and a crisis mode, we find at times a third mode that I call the “transitional mode.” The focus of a transitional mode is related to the enforcement of a new principle policy by the central leadership. It is directed against every form of resistance against this policy, be it by the bureaucracy, intellectuals, or various interest groups, and aims at bringing these groups into line. Currently, the phase of enforcing the new development and growth model and the related policies adopted in November 2013 stands for such a transitional mode. Transitional modes are characterized by a certain kind of repression that is exerted against all forces opposing the new policies and a major political campaign of enforcement (see below). A major aim here is the unification of thinking and acting among both party members and social groups. In the minds of the people of China, however, there is a growing feeling of social instability due to social inequality, uncertainties, corruption, disintegration of traditional communities, and the decay of values. Therefore, the political leadership strives to portray the CCP as a central guarantor of sta-

The Resilience of Authoritarianism

169

bility and crisis avoidance. Since the organizational net of the party covers the entire country, the CCP is able to intervene continuously in economic and social processes at all levels and to resolve conflicts.18 Hence, I argue that the Chinese political system currently is capable of avoiding crises or at least dealing with them in an effective way and therefore, in spite of a growing feeling of instability, still appears to be relatively stable. The relative stability and robustness of the Chinese system rests on three pillars: (1) a successful developmental state ensuring economic development;19 (2) the capacity and ability for institutional adjustment and flexible policy implementation; and (3) legitimacy. Over and above effective economic development, strong state capacity (including co-optation efforts) and coercive capacity (including repression), China’s authoritarian political culture (meaning that a high percentage of the people favor authoritarian structures),20 and governmentality constitute the bedrock under these pillars. Governmentality refers to the political technologies used by the party state to invoke among leading cadres a “will to improve” (T. M. Li 2007), that is, a will to develop a specific field or location (e.g., by using disciplining tools to achieve compliance). Michael Foucault also calls these techniques “governmentality” (1991), in my case a policy of instilling into local cadres and the people this will to improve, meaning that they consciously implement central policies and regard the projects of the central leadership as their own self-determined projects. The outcome of this entire process can be classified as “authoritarian consolidation,” which might explain why authoritarian systems like China do not necessarily collapse but prove to be capable of tackling crises and therefore remain relatively stable. Legitimacy, the recognition and acceptance of the legality of power and of a political system, is a salient feature and precondition of stability. After experiencing the period of turbulence after 1949, the Chinese people today are primarily interested in prosperity and social and political security, and not in political experiments or radical political change. China’s rapid remedies for the global economic and financial crisis of 2008−2009 in the form of distinct steering programs has undoubtedly reinforced the legitimacy of the Chinese developmental state. This will be substantiated in the following section. Outlook The long-awaited third Plenary Session of the eighteenth Central Committee in November 2013 adopted a comprehensive sixty-point program specifying far-reaching reforms in fifteen policy fields (Decision on Major

170

Thomas Heberer

Issues Concerning Comprehensively Deepening Reforms 2013). These targets will be implemented by the year 2020. This program is labeled the “Fifth Modernization”21 (i.e., a modernization of governance in contrast to a program of political reforms). It comprises the following modernizations of governance. First is governance of the state (service function of state; reforms of financial, fiscal, tax, and land systems; sustainability; and reforms of the legal system, i.e., “governing by means of law”). Second is governance of the market (expansion of the etatist market economy and the regulating role of the state; private sector development; strengthening domestic consumption and domestic markets; upgrading industries and improving the efficiency of enterprises; and reform of state-owned enterprises). Third is governance of society (urban-rural integration; establishing social security systems; and abolition of the residence permit system, the system of reeducation by labor, and one-child birth planning). This undoubtedly constitutes the boldest set of economic and social reforms since 1978 and contains most of the major reform changes that have been demanded for years by Chinese and foreign experts. Such an ambitious program can be implemented only incrementally and step by step. A central management team for comprehensive deepening of the reform program has been established to oversee the program, which is expected to be fully implemented by 2020. Implementing this program will not be an easy task. The program must still be translated into concrete policies, which will then have to be put into practice. In addition, the central leadership is facing various challenges posed by the vested interests of state-owned enterprises, security authorities, local governments, banks, and other interest groups. On the other hand, the new reform program demonstrates that the current leadership under Xi Jinping is following Deng Xiaoping’s reform philosophy. And it elucidates anew that the party conceives of itself as the crucial developmental agency in bringing about the modernization of the Chinese nation. While the stability and legitimacy of the Chinese party state are based on a specific concept of developmentalism, many ordinary people and even officials argue in support of political reforms (but not of political change). According to a report in the Global Times, a newspaper survey published by the party’s daily Renmin Ribao revealed in 2012 that 88.4 percent of 1,200 people interviewed in eighteen cities supported such reforms (Bai 2012). Currently, the political leadership is focusing on relentlessly combating corruption and implementing new economic reforms adopted at the end of 2013. To accomplish these two goals, the regime needs a stable political system. Moreover, developmental states have to focus on specific policies. It would be risky to concurrently conduct economic and political reforms while simultaneously fighting corruption. To ensure a stable political sys-

The Resilience of Authoritarianism

171

tem for achieving its primary goals, the regime is at the same time dealing sternly with all kinds of particularistic interests and collective actions. This is why I argue that China is in a phase of transition in which repression toward oppositional forces is temporarily reinforced. In the short and medium term, a regime change is not to be expected. Despite further reforms, even in the political domain, great reforms such as an independent legal system or a separation of powers are highly improbable. Reforms can be expected that will contribute to political institutionalization and to the improvement of governance. Suisheng Zhao speaks of political reforms that are “driven by efficiency,” meaning that the purpose of such reforms is to improve the efficiency of the system in the interest of China’s further economic and social development (S. Zhao 1997). These might include greater transparency in the actions of both the party and the government (from the central right down to the local level), increased participation by the people regarding checks and balances, disclosure of the incomes of senior officials and deputies of the People’s Congresses, tougher evaluations of cadres by higher administrative echelons, elimination of torture, abolishment of administrative reeducation through labor camps, and enforcement of compliance with existing laws. However, we should understand that a breakdown of the political system would not lead automatically and necessarily to a stable democracy in China: the conditions of the current transformation process are too difficult and complex, and there are too many conflicting interests between various social strata and regions. The outcome of overthrowing the rule of the CCP might well be political turmoil, inner power struggles, a protracted economic crisis, or even a fascist or military dictatorship—with severe consequences not only for the Chinese people but also for Asia, the global system, and the world economy. Many Chinese scholars are fully aware that democracy cannot simply be imported and that it requires more than merely the establishment of democratic institutions like the separation of powers, an independent law system, and a multiparty system. Rather, a stable political entity requires specific preconditions for democratic structures to develop and stabilize. Among these are the emergence of structures that are basic to a civil society such as greater liberty for the media and for NGOs and the development of citizen values and attitudes. Moreover, this requires an operational legal system that protects the rights of individuals and groups against arbitrariness by the state; legal awareness among state officials and the people; and, in particular, civilizational competence (Heberer 2012). In China, civil society is in its initial stages, and the development of an operational legal system still has a long way to go. By “civilizational competence,” a term coined by Polish sociologist Piotr Sztompka (1993, pp. 88−89) referring to “human modernization,” I

172

Thomas Heberer

mean the cognitive preconditions for a stable democracy: the emergence of a citizenry with a society-oriented public spirit and a sense of civic responsibility; the acceptance of diverging opinions and political criticism by the state and by society; peaceful management of conflicts (by the state and society) through the establishment of patterns of conflict management; and, finally, the development of empathy (i.e., a capability for understanding the feelings, emotions, and thinking of others). In the discussion of preconditions for a transition of authoritarian states to democracy, these cognitive aspects generally are widely neglected. Recent developments in other world regions (Arab countries, Afghanistan) illustrate that democracy cannot simply be introduced. Moreover, democratization does not consist merely of the establishment of democratic institutions. Cognitive preconditions are required for the development and stabilization of democratic structures and institutions. US political scientist Joel Migdal shows that, when weakly developed civil society structures and a weak society face a strong state, states can function as “political architects” (1988). This is exactly the case in China’s current party state. Therefore, the party state is not merely a “development dictatorship” but can rather be classified as a “developmental agency” with “infrastructural power” and effective governing tools such as coercive power, the power to extract revenues, and so forth (Mann 1993, p. 5). Infrastructural power requires not only pure authoritarian mechanisms of enforcement, but also the inclusion of social groups in decisionmaking processes and the establishment of a corresponding set of institutions. Here, the local state in China plays a crucial role as an effective and successful development agency; that is, as a developmental state thus contributing to the regime’s stability and legitimacy. The World Development Report 1997: The State in a Changing World has already emphasized that, under conditions of a weakly developed civil society, the state has to function as an “activating state” (World Bank 1997). I argue that the Chinese party state has adopted this function and acts as a political entrepreneur that is purposefully pursuing modernization of the country. In the interests of social and political development, gradual political reforms such as the reinforcement of a rule of law and citizen rights, the fostering of a new societal consensus, and new patterns of efficient social management are indispensable. Avoiding societal fragmentation by finding solutions for burning issues such as growing income disparities, environmental and ecological disasters, and corruption is crucial for the further progress, stability, and legitimacy of the CCP. These are precisely the tasks of a developmental state 3.0. Whether the new leadership will be capable of providing fresh answers to these questions remains to be seen.

The Resilience of Authoritarianism

173

Notes 1. Shambaugh altered his stance, speaking about an “endgame of Communist rule” in an article in the Wall Street Journal (Shambaugh 2015). He elaborated this idea in more detail in his book China’s Future (Shambaugh 2016). 2. Some of these authors have recently grown more pessimistic. See, for example, Gilley (2015). 3. Thus, I agree with Fukuyama’s argument that regarding nondemocratic systems comparative politics cannot provide a sufficient framework for analysis. See Fukuyama (2012, pp. 17−18). 4. On developmental states see, for example, Kohli (2004), Woo-Cumings (1999); and from a different perspective, Ong (2012). 5. In contrast to Acemoğlu and Robinson (2012), I argue that not only institutions but also actors play a crucial role in developmental processes. Przeworski points out quite rightly that institutions and agencies—that is, the impact of actors—are inseparable and interactive. See Przeworski (2004). 6. It is frequently argued that state intervention in the economy is detrimental to economic development and leads to a decline in economic growth. Kohli (2004) in contrast notes that such arguments lack evidence and that intervention by strong states rather triggers development. 7. In this chapter, I distinguish between “ineffective” and “effective” policy implementation by referring to outcomes that are consistent with development goals as laid out in the policy guidelines and work reports of China’s county governments. This does not mean, however, that these outcomes reflect implementation efficiency, understood as the best possible results in terms of resource allocation and administrative capacity. See Heberer and Schubert (2012). 8. Levitsky and Way (2010) point out that the role of coercive capacity for the regime stability of authoritarian states has long been overlooked. Repressive measures are also part of the coercive capacity of a regime. 9. Scott (1998, pp. 89-90) argues that authoritarian developmental states justify their modernizing policies with a “scientific concept” which must be enforced in a top-down manner across all social barriers and obstacles. He called this developmental ideology “high modernism,” meaning that a political elite with a technocratic belief in progress, assumes that modernization is a planned act, a “rational design of social order” by means of which a society can be bureaucratically reorganized and modernized. Both the scientific concept of development and the construction of a new socialist countryside of the Hu Jintao era correspond precisely to this type of high modernism. 10. Cf. Renmin Ribao, September 6, 2011; vice party secretary, interviewed by the author, Jiangyin, August 15, 2011. 11. “Red tourism” is a label for historical sites related to the history of the Communist Party and now figuring as tourist attractions. Prior to 1949, the Guomindang regime had established the second-largest concentration camp for Communist prisoners in Xifeng County. The former camp has meanwhile been rebuilt as a large museum and visitors park and it attracts large crowds of tourists. 12. Bimo are priests and shamans of the Yi nationality. 13. Meanwhile, such models and experiments have to be authorized by higher authorities. See “What Does Xi Jinping’s Top-down Leadership Mean for Innovation in China?” (2017).

174

Thomas Heberer

14. Vice party secretary of Jinjiang, interviewed by the author, Jinjiang, September 6, 2012. 15. I call it “local state corporatism 2.0.” In traditional local state corporatism (“local state corporatism 1.0”), local governments and private enterprises were conceptualized as components of a larger entity (a corporate firm) with the goal of maximizing the profits of the local corporate state. In this concept, the state figured as an entrepreneur itself, establishing and operating collective-owned enterprises (TVEs) to develop a locality. Under the conditions of local state corporatism 2.0, however, the local state no longer figures as a corporate head, but functions rather as a regulator. See, for example, Schubert and Heberer (2015). 16. Compare Gilley (2009). See also Gilley (2008), Gilley and Holbig (2009), Heberer and Schubert (2006), Schubert (2008). On trust, cf. Tang (2005, 2007a, 2007b) as well as Heberer and Schubert (2011). 17. Effective policy implementation refers to the outcomes that are consistent with local development goals as laid out in the policy guidelines and the work reports of county governments. This does not mean, however, that these outcomes reflect implementation efficiency (understood as the best possible result in terms of resource allocation and administrative capacities). 18. An article in the influential newspaper Nanfang Zhoumo [Southern Weekend] has specifically pointed out that the participation of citizens should be the focal point in resolving street violence (Dai 2014). 19. On the one hand, the party state provides the population with economic benefits and public goods; on the other, it prevents people from organizing collective action. 20. By “political culture” I refer here less to values than to a toolkit of habits, customs, world views, beliefs, and so forth, which help to construct and organize specific actions. See, for example, Swidler (1986). Shi (2008) has shown that even democratic values can support an authoritarian system such as that of China. See also Vollan et al. (2013). 21. Following the “four modernizations” of industry, agriculture, science and technology, and national defense.

9 Historical Legacies and the Institutionalization of Military Rule: Lessons from Latin America Felipe Agüero and Julian Brückner Building on “the [new] institutional turn in comparative authoritarianism” (Pepinsky 2014), in this chapter we address the unique survival challenges faced by military regimes, a distinct type of autocracy that the literature has deemed highly unstable compared to other autocratic forms like monarchies, single-party regimes, or personalist dictatorships (Geddes 1999; Hadenius and Teorell 2007). In accounting for this instability, scholars in the past have claimed that the military’s very professionalism, or its institutional interests, leads it to seek extrication sooner rather than later (Fitch 1986; Geddes 1999). From another angle, scholars have argued that military intervention seeks only limited goals by playing merely a moderator role, which ultimately results in underinstitutionalization (Huntington 1968; Stepan 1971). However, the 1964 military coup in Brazil marked the abandonment of the moderator role and the inauguration of a new kind of military intervention in Latin America. The military governments that emerged in the 1960s and 1970s set out to execute lasting changes, to intervene on long-accumulated structural problems that required long-lasting military rule (Garretón 1983) and viewed some degree of institutionalization as a desirable goal. Nevertheless, there was substantial variation in the duration and stability of these regimes that none of the existing accounts can help us understand, which is what this chapter addresses and explains.1 We do this by focusing on the Latin American cases, which represent the “standard model” of the military junta that continues to inform scholars’ thinking about the characteristic weaknesses of military rule to this day (Croissant 2013a, p. 409). These cases may be grouped in three categories in terms of duration: long duration (Brazil, twenty-one years, 1964−1985, and Chile, seventeen years, 1973−1990); intermediate (Peru and Uruguay, twelve years, 1968−1980 and 1973−1985, respectively); and short (Argentina and Ecuador, seven years, 1976−1983 and 1972−1979, respectively). 175

176

Felipe Agüero and Julian Brückner

We account for this variation by assessing different levels of success in the efforts of military regimes to attain some degree of institutionalization. Institutions play a fundamental role in helping dictators confront the survival challenges that Milan W. Svolik identified as the “problem of authoritarian control” of those excluded from power and the “problem of authoritarian power-sharing” with critical elites in the military leadership or the business sector (2012, p. 2). If designed or accommodated properly, the literature holds, institutions may contribute to the survival of autocratic regimes by limiting conflicts among dictators, their ruling coalitions, and the opposition (Pepinsky 2014). Hence, recent scholarship on authoritarian survival has focused on the battery of institutions to which authoritarian regimes may resort to cement alliances and effectively broaden their support such as semicompetitive elections (Schedler 2002; Magaloni 2006), constitutions (Albertus and Menaldo 2012), courts (Moustafa 2009), parties and legislatures (Gandhi 2008; Boix and Svolik 2013), and intelligence services (Policzer 2009). We believe, however, that these studies have tended to view autocrats in both military and civilian regimes as largely free to make choices in their efforts to consolidate power. For instance, Jennifer Gandhi and Adam Przeworski argue that dictators, faced with the problem of power sharing and control, “must accurately perceive the strength of the threat and respond with a sufficient degree of institutionalization” (2007, p. 1293). In this view, then, autocratic rulers are able to create a wide array of custom-made institutions in response to challenges. Here, we share Thomas Pepinsky’s critique that the new institutional turn in the comparative study of autocracies suffers from excessive voluntarism: “institutions do exactly what their creators want them to do, and leaders adjust institutional forms when doing so is in their interest” (2014, p. 637). We argue that autocrats’ institutional choices are not entirely free, but that they are in fact influenced by enabling or constraining legacies from the period prior to authoritarian installation. We maintain that historical legacies, such as the historical development of military professionalism and the military’s prior relation to party politics, weigh in on the ability of military rulers to achieve regime institutionalization and avoid crisis and instability. The military itself is at the root of this instability, threatening coups on military rulers more often than on civilian authoritarian rulers (Croissant 2013b). In fact, military rule presents a distinct challenge; namely, the inherent contradiction of performing both military and government functions, which is the contradiction between the hierarchical nature of professional military organizations and the contested and divisive realm of political decisionmaking (Clapham and Philip 1985). This is the Achilles’ heel of military authoritarian rule (Fitch 1998), and it explains why military

Historical Legacies and the Institutionalization of Military Rule

177

leaders must first solve internal tensions and challenges before readying the institution for the demands of running government.2 What, then, are the kind of institutions with which ruling militaries may avail themselves to counter threats from below, from allies, or from within? And what constraints do they meet in seeking the support, empowerment, and protection of institutions? In this chapter, we address these questions and present the different ways in which the military leadership will attempt to cope with those challenges. We argue that ideally a military regime will seek to institutionalize its rule, preferably by embedding it in a new constitution. Furthermore, we maintain that the route to achieve this goal is a proper solution to the issue of succession (i.e., the question of who the ruler will be and for how long). We point out that this interest in institutionalization is particularly critical to the Latin American regimes of the 1960s through the 1980s, which sought long-term foundational goals. Finally, we make the case that the weight of enabling or constraining historical legacies strongly influences the ability of military rulers to achieve institutionalization and avoid crisis and instability. This chapter is organized as follows. First, we identify three potential sources of crisis and a set of crucial challenges that all military regimes face when seeking institutionalization at different stages of their tenure. The issue of succession arises during the installation phase, followed by the problem of organizing rule: anticipating the “right” amount of institutionalization needed and the decision on the regime’s grand policy orientation. Then, we introduce our main contribution on the role of historical legacies in enabling and constraining rulers’ choices. Finally, we illustrate our argument with a view toward the Latin American cases, especially Chile and Peru. Dilemmas of Succession and Institutionalization Long-term rule highlights the challenge of institutionalization. But what does institutionalization mean for a military regime? We borrow from Barbara Geddes, Joseph Wright, and Erica Frantz (2014, p. 147) that a military regime is “rule by a group of high-ranking officers who can limit the dictator’s discretion” and participate in critical decisionmaking. Institutionalization, then, is the formalization of those limits and modes of participation (an internal dimension) as well as the formalization of a strategy for governing the society (an external dimension) expressed in administrative procedures and power-sharing mechanisms aimed at implementing decisions and securing support from forces outside the military. Ideally, if military intervention seeks broad political, social, and economic transformations, institutionalization would be best attained in a new constitution, as it

178

Felipe Agüero and Julian Brückner

enables obedience in the name of the law, legal recognition from the outside, and regulation of access to power within the authoritarian elite (Negretto 2014). However, military regimes must first confront the challenges of the installation phase: “how to survive past the initial moments of the overthrow of the old regime in order to establish a pattern of rule that will last” (Domínguez 2002, p. 1). This is followed by the challenges of choosing “institutional means” and then a “strategy to govern the society” (Domínguez 2002, p. 1). These choices and moments capture what we see as the main challenges of access to power and organization to rule. The installation phase entails the adoption of measures to establish total control of public order, and the elimination of all forms of resistance to the new rulers. At the same time, early coordination decisions among the top leaders will lead to the question of who is in charge, given the collective nature of military juntas. Decisions initially are collectively made but, in practice, one leader will predominate. Sooner or later an explicit decision must be made that establishes who will rule and for how long, and whether there will be rotation or some form of succession, the most critical issue during the installation phase. Agreement on this issue grants regimes the ability to dispel coups or forceful threats against the leader from other leaders or factions and, thus, attain some degree of stability. The need for such an agreement becomes in fact a political survival strategy for the regime (Croissant and Kamerling 2013; Domínguez 2002; Huntington 1968), and may result in different possibilities. Keeping the military junta as the top executive decisionmaking body is a cumbersome alternative and, in fact, military juntas themselves often tend to designate or permit the development of a junta chief or president. This leads to the issue of rotation: whether the chiefs of staff will take turns in assuming the position of head of the junta. Rotation demands definition of the duration of the term, which may develop concurrently with a separation of the junta from executive positions. At this point, the succession rule becomes intertwined with the first set of broader issues about the organization of rule: not only who is in charge and for how long, but also who is in charge of what. The junta head may take on the role of chief executive while the rest are charged with leading specific sectors of the administration or participating in the most important strategic decisions. There also exists the alternative of no rotation at all, which may mean that a military leader’s de facto leading position at the start of the regime is maintained over time with no specification of a set term and without a formal decision that validates this path. This alternative is equivalent in fact to no succession rule at all, and is eminently unstable,

Historical Legacies and the Institutionalization of Military Rule

179

subject to changing winds among the group of high-ranking officers. However, no rotation may take a different course as well, which offers more promise for stability: having power formally assigned to one individual with no rotation over time. This could develop into a military strongman scenario, but this need not be the case. While a military leader who becomes the only ruler over time may see his personal power enhanced, he may simultaneously remain accountable to a military junta or council, or at least bound by rules that limit his discretion.3 Whichever course is taken, regulating leadership turnover increases the chances of regime survival and becomes an early measure of a military regime’s institutionalization. Organizing rule confronts military leaders with another critical decision: the extent of involvement of the military as institution in the political roles of government, which may or not lead to setting up a strict separation between military officers in military functions and those assigned to government responsibilities. The point here is to avoid the Achilles’ heel of military rule: a clash between the hierarchic nature of the military and the divisive nature of politics and political decisionmaking. A related decision concerns the division of labor in the organization of government, primarily the assignment of legislative and executive functions. One possibility is to maintain a functioning, albeit controlled congress. This is a rare option for a military regime and, among the Latin American cases, it happened only in Brazil. Another option is to give the military junta all legislative functions and the top military chief the executive role. The organization of power also entails a strategy for governing the society (Domínguez 2002). One crucial factor concerns the organization of repression and the implantation of fear, which is critical for ruling over previously mobilized societies by regimes intent on demobilization (Policzer 2009). Military leaders may resort to the traditional police organization, the military barracks, and the use of existing intelligence branches of the armed forces for detention, interrogation, torture, and suppression of resistance, which will cement complicity across ranks and services. In addition, they may create a specialized agency closely connected to the head of the government (Weiss Fagen 1992) to conduct these tasks albeit at the risk of encouraging criticism and resentment among the ranks, even if it liberates them from the performance of gruesome tasks. If overdeveloped, these agencies would generate a new source of imbalance within the military and become a source of disagreement and tension, as Alfred Stepan (1988) convincingly argues for the case of Brazil. The other crucial factor concerns securing support among social sectors outside the military to promote the transformational changes pursued, which may mean the creation of a political party or the accommodation of

180

Felipe Agüero and Julian Brückner

an existing one (Negretto 2014), besides resorting to expert civilians (Biglaiser 1999). Recent literature has addressed the options of military leaders for creating special security forces or political parties as a support basis. Barbara Geddes (2009), for instance, analyzed the ways in which dictators deal with rivals in the uncertain early phase in which no binding rules exist. She argues that dictators will opt for power-sharing agreements if the military is professionalized because as a unitary actor it may produce a credible enough threat for the dictator to consent. In the case of less professionalized militaries weakened by factionalism, in contrast, a dictator’s survival is best guaranteed by less dependence on the military and the creation of a mass civilian party or a special security force. We find Geddes’s (2009) view unsatisfactory for our concern with regime stability. It focuses too narrowly on individual survival, underestimating collective preferences of militaries that come to power imbued with grand transformational goals. It also emphasizes the early choices and less so the longer-term consequences of those choices for stability. But perhaps most importantly, the otherwise useful distinction of professionalized and less professionalized militaries fails to capture varying degrees of factionalism within professional militaries. All Latin American cases of the 1960s and 1970s that we studied are examples of professional militaries, yet each displayed different degrees of internal cohesion. Finally, we believe that the considerations that shape the choice of strategies in the early phases are influenced by the grand policy orientation with which the armed forces come to power. Stepan (1978) suggested that military regimes can be placed along a continuum between inclusionary and exclusionary corporatism depending on the degree to which they rely on the support or the suppression of organized labor, on the one hand, and capital, on the other. We believe that an exclusionary orientation, such as characterized most of the Latin American military regimes, will result in a heavier repressive effort and the devaluing of the alternative of creating a mass civilian party. In contrast, regimes with an inclusionary intent (e.g., Peru), deprived of the kind of natural support from the dominant social and economic sectors that exclusionary regimes may count on, will feel the need to organize social support to favor their reformist goals, via either party creation or other means. The military regimes of the 1960s through the 1980s in Latin America varied considerably in most of the elements reviewed in this section, which are relevant to the prospects and processes of institutionalization and that help to understand differences in the longevity of these regimes. They also highlight the fact that, in facing decisions on power arrangements and institutions, autocrats were not unconstrained. Choices were contextually embedded, and influenced by legacies from the preceding period or regime.

Historical Legacies and the Institutionalization of Military Rule

181

Historical Legacies Historical legacies from preceding periods and regimes have been a prominent topic in the (democratic) transition and consolidation literature on Latin America and southern Europe (Cesarini and Hite 2004; J. S. Valenzuela 1992) as well as postcommunist regimes in Eastern Europe and the former Soviet Union (Jowitt 1992; Crawford and Lijphart 1997). Specifically with regard to Latin America, scholars note the lasting influence of different founding conditions (Agüero 1998) created during the transition and of previous longer-term factors on civil-military relations in postauthoritarian regimes (Loveman 2001; Weeks 2004). No such concern with the weight of the past is found in the literature on autocratic regimes and among scholars studying the effects of institutions on autocratic crisis and stability. This lacuna is especially surprising in the case of military regimes that are organized around “a ready-made institution” (Gandhi and Przeworski 2007, p. 1284) closely connected to important historical antecedents like processes of state formation. The Military and Professionalization

The most obvious legacy factor acting on military authoritarian regimes is the military institution itself, especially its professionalization processes. Legacies on the military side therefore need to be based on a proper understanding of military professionalization and professionalism. Geddes (1999) rightly assumes that military professionalization leads soldiers to value corporate interests—especially discipline, hierarchy, and autonomy—above all else. But viewed by itself, the desire to maintain military unity explains little about (post)authoritarian civil-military relations beyond the armed forces’ initial decision to intervene or withdraw from power. In fact, recent contributions to the literature on institutions and power sharing in (military) authoritarian regimes have tended to reduce professionalism to the armed forces’ capacity for collective action (Geddes 2009; Croissant and Kamerling 2013). Whereas Samuel Huntington (1957) was rightly criticized for assuming that professionalization would render the military apolitical, the new institutional turn in the study of (military) authoritarian regimes has rendered the concept of professionalism apolitical by neglecting the fact that the preferences and behavior of the armed forces are also shaped by different founding conditions. In the past, military professionalism has been linked to the establishment or breakdown of (military) authoritarian regimes (Nordlinger 1977; O’Donnell and Schmitter 1986), although the “bureaucratic authoritarian” nature of military rule itself was seen as reflecting more the socioeconomic

182

Felipe Agüero and Julian Brückner

crisis and the severity of the threat that led the military to intervene in the first place (O’Donnell 1978). Arguing against the socioeconomic determinism of the literature on bureaucratic authoritarian regimes, Karen L. Remmer first suggested a connection between “the specific institutional history of the armed forces” and the “structure of military regimes” (1989, p. 38). Whether power would continue to be exercised collectively by a group of high-ranking officers representing the different branches of service and whether a clear separation between traditional military and newly acquired government functions would be maintained depended less on socioeconomic context conditions or professionalism in the narrow sense than on the legacies of the armed forces’ institutional development. Thus, in the case of Chile, a long-standing tradition of nondeliberation and the lasting influence of Prussian military training gave rise to a “corporate ethic of constitutionalism” with authoritarian undertones. Once the military assumed power, this legacy and the armed forces’ aloofness from partisan politics supported the emergence of a strong presidency and the appointment of civilian technocrats to key governing positions (Remmer 1989, pp. 121−145). Remmer showed that the institutional choices of military rulers and the resulting variation in the durability of military regimes are conditioned by the lasting influence of military historical legacies. Unlike more recent contributions to the literature on authoritarian institutions and power sharing that reduce military professionalization to growing discipline, hierarchy, and autonomy, a focus on the legacies of the armed forces’ institutional development can help us understand why highly professionalized militaries choose very different institutions to organize their rule. For instance, the Peruvian and the Brazilian armed forces could be found at similar levels of cohesion at the outset of military rule in 1964 and 1968, respectively. Nevertheless, the generals in Brazil agreed on fixed terms for a succession of military presidents, whereas in Peru no limit was given to Army chief general Juan Velasco’s presidency. These institutional choices and their consequences cannot be explained by the greater capacity of leaders of professionalized militaries to enforce power sharing among the officer corps as has been suggested by Geddes (2009) and others. Yet a focus on military historical legacies can also contribute to answering the question of when institutions in military regimes are likely to “bite” or lose their grip (Capoccia 2016). The Chilean authoritarian constitution of 1980 and prior decree laws that defined the powers of the junta and its president were instrumental in helping the regime survive several internal and external challenges. Yet despite its stabilizing effects, the 1980 constitution ultimately bound the regime to its own rules, forcing

Historical Legacies and the Institutionalization of Military Rule

183

General Augusto Pinochet to accept defeat in a plebiscite in 1988 that paved the way for a controlled transition to democracy (Barros 2003). In Argentina, in contrast, the rotating presidency that military leaders had agreed on at the outset of the regime quickly gave way to a series of palace coups after the first military president, General Jorge Videla, had served out his term and transferred power to a successor in 1981. Rather than helping to keep factionalism at bay, the institution of presidential succession itself became a source of dispute that the civilian opposition could exploit (Pion-Berlin 1985). Much like other regions of the world today, the professionalization of the Latin American armed forces in the late nineteenth and early twentieth century was often undertaken with the help of military missions from Europe and the United States. Growing security needs in South America made apparent by two major wars—the Paraguayan War (1864−1870) and the War of the Pacific (1879−1883)—prompted several governments to contract German and French military missions to help with the modernization of their countries’ armed forces (Resende-Santos 2007). 4 Widely considered to be the most advanced armies of their time, decades of German and French military training between the arrival of the first Prussian officers in Chile in 1886 and the outbreak of World War II achieved many of its professional goals, albeit with some unintended consequences. In terms of successful military modernization, “European-style professionalization helped create an institutional identity, a separateness based on education, mission and attainment of higher rank through merit” (Nunn 2001, p. 28). However, the different societal context in which professionalization took place also gave rise to a highly politicized military elite that soon came to blame their countries’ underdevelopment on “corrupt” politicians (Loveman 1999, p. 69). But the European military influence also proved to be a critical juncture in another important way. According to Frederick M. Nunn (1975), competing ideas about military professionalization had the potential of reducing personalist orientations in the officer corps that had been carried over from the age of the caudillos in postindependence Latin America. In the case of the Chilean and Brazilian officer corps, “Franco-German professional ‘differences of opinion’ kept political action from taking on a personalist tone” (Nunn 1975, p. 3), whereas in Argentina and Peru personalism persisted. Thus, military professionalization in Latin America assisted by European military missions either created a legalist orientation and a sense of duty to the state that Huntington (1957) considers to be one of the hallmarks of military professionalism, or it reinforced older personalist orientations. Therefore, military professionalization and its political effects must be understood in historical context, with an eye to military-society relations.

184

Felipe Agüero and Julian Brückner

As J. Samuel Fitch puts it, “The political content of military professionalism is not constant, but variable, depending on both military doctrine and the specific relationships between the military, the state, and the rest of society” (1998, p. 3). We believe that the other crucial historical legacy relevant to the military in its attempts to institutionalize military rule is its historical relation to parties and party politics, which may lead to higher or lesser levels of factionalism. The Military, Party Systems, and Party Politics

The military comes to power with varying degrees of vulnerability to factionalism, which affects its ability to take the steps necessary to attain institutionalization as a military regime. Early steps toward institutionalization, such as agreement on the issue of succession or rotation, can provide mechanisms to prevent internal divisions, but divisions themselves may perversely stand in the way of reaching such agreements. Factionalism may have different sources, many of them internal, but here we maintain that a crucial source of factionalism is to be found in the legacy of military exposure to partisan politics. Early efforts to promote military professionalization were aimed precisely to insulate the armed forces from the partisan divide, granting it organizational and professional distinctness. However, the intensity of the partisan divide in the context of weak political institutions often could not buffer the military from the pursuit of civilian factions seeking military support. Only a well-established, institutionalized party system could permit adequate representation and negotiation without recourse to the military. Militaries historically less exposed to the pressures of party factionalism, because of well-established functioning party systems, were less prone to factionalism at the time of military takeover. But since institutionalized party systems in Latin America were rather scarce, it makes sense to trace the actual kind of relationships that the military established with political parties. Three patterns may be discerned: aloofness, hostility, and enticement. A military may stay aloof from partisan politics over the course of time in the context of a well-established party system. Where no such party system developed, the military may enter into relations of hostility toward parties or navigate situations of enticement from existing parties. These patterns developed mostly as a result of the critical junctures in the formative decades of the 1920s and the 1930s. The decisive juncture that placed the military in Chile in a trajectory of aloofness was the end of a period of reformist military interventions in the 1920s. The military intervened in 1924, in the context of the emergence of the “social question,” to unblock oligarchic opposition to

Historical Legacies and the Institutionalization of Military Rule

185

reforms. But military factionalism and the shock of the Great Depression that hit Chile especially hard, in 1932 led to the return to democratic civilian rule, uninterrupted for the following four decades. Democratization unfolded with two relevant features: an explicit policy of subduing the military—a so-called “civilianist reaction”—and the emergence of a new well-structured party system sustaining parties of the left, the center, and the right (Scully 1995; Valenzuela 1978). The military remained aloof from party politics and unnoticed by party elites across the political spectrum, until they watched in disbelief the first signs of military rereassertion in politics in the early 1970s (Agüero 2002). When the military came to power in 1973 it was, for historic reasons, less vulnerable to threats of factionalism. The hostility pattern is clearly found in Peru and Argentina, in which the consequences of the impact of the Great Depression worked in very different ways than it did in Chile. Whereas in the latter, the crisis unleashed by the Great Depression led to democratic civilian rule, in Peru it precipitated a military uprising and turned the army into a “decisive political factor in the defense of order, assuming the role of ‘watchdog of the oligarchy’” (Cotler 1995, p. 328). Military strongman Colonel Luis Sánchez Cerro took over in 1930 and held elections in 1931, of which his major opponent, the leader of the American Popular Revolutionary Alliance (APRA) party, claimed it had been robbed. APRA, defeating communist and socialist competition, took control of the mobilized masses with which it led several revolts. In 1932, APRA captured a military garrison in a northern city and massacred its occupants. “The military recaptured and counter-massacred, even more ferociously, and a section of the military became Anti-APRA for the next generation” (Philip 1985, pp. 137−138). The pattern of hostility of the military toward APRA was clearly established then for decades to come, moving the army to suppress any attempts of APRA to resurface and gain legitimacy through electoral means. The dynamics of Peruvian politics would be marked by the frustrated attempts by APRA to reenter the political system. In this context, no well-structured party system developed: in a study, Scott P. Mainwaring and Timothy R. Scully (1995) ranked Peru at the bottom in party system institutionalization in Latin America. APRA’s efforts to reenter the system included the pursuit of military factions for support, leading to their suppression by opposing dominant military factions and to the rise of several military presidents.5 Military hostility toward the largest mass party thus did not translate into cohesion. APRA’s pursuit of legitimacy led it gradually to shift to centrist positions and ultimately to an alliance with the oligarchy under the presidency of General Manuel Odría, who had previously suppressed it vigorously. The

186

Felipe Agüero and Julian Brückner

demand for nationalist and antioligarchic reforms was thus left without representation. And when the armed forces themselves shifted, disillusioned by the repressive use which the oligarchy made of them to suppress peasant rebellions, it moved to lead and channel these reformist demands, albeit in utter isolation and battered by divisive civilian partisan politics. The reformist military coup of 1968 and its inclusionary intent stood on the basis of the armed forces’ historical hostility toward APRA and their new hostility toward the oligarchy and its parties. In Argentina, the early 1930s also saw the emergence of the military to power, reacting to the Radical (middle-class) Party it had overthrown. Military hostility toward the Radicals took the form of political suppression, in turn countered by Radical-led efforts to obtain military support for the coup attempts that they in fact, unsuccessfully, staged. Between 1931 and 1943, military-backed regimes dominated Argentine politics, in alliance with conservatives (Philip 1978). But the clearest pattern of hostility in the military developed against Peronism after the ouster of General Juan Perón in 1955. Perón had risen to power in 1943 sustained by internal military organizations, and then was elected president in 1946. His maneuverings to obtain personal control of the unions, the church hierarchy, and the military backfired with increasing church opposition and resentment from the military. His ouster by the military led to a purge of Peronist officers and a prohibition of former Peronist officials or anyone taking orders directly from Perón from assuming important political posts (McGuire 1995). Military hostility translated, as in Peru with APRA, into impeding any resurfacing of Peronism, now a structured and massive civilian organization, into open politics (McGuire 1995; O’Donnell 1986). Each time there was a threat of a political resurfacing of Peronists, the military would intervene. In the period between Perón’s exit in 1955 and the next major coup in 1966, political parties, in an enticement pattern, “played the military card when it suited them, conniving for coups in the hope that a military government would appoint their leaders to top posts or shift the balance of political forces in their favor” (McGuire 1995, p. 219). The result was extreme factionalism, to the point of armed confrontation between two military factions in 1963. This background of factionalism under the more general pattern of hostility was the main factor leading to military rule in the period 1966−1973 which aimed, with no success, to eliminate the praetorianized context that pressured and fractured the military. The vastly exclusionaryoriented coup of 1976 thus unfolded against the backdrop of a recent history of deeply harmful military factionalism. The 1930s were a decisive turning point also in Brazil, expressed in the junior officers’ revolt of 1930, clearly on the left, and further developed by successively more radical revolts that were severely crushed. Ultimately, it

Historical Legacies and the Institutionalization of Military Rule

187

gave rise to the protracted authoritarian period of Getulio Vargas until his ouster in 1945. The period between 1945 and the exclusionary-oriented coup of 1964 was Brazil’s “first experience with open, democratic, competitive politics” (Stepan 1971, p. 85) and yet one that saw five military coups. The party system was weak and unable to limit the high levels of personalism that dominated Brazilian politics (Mainwaring 1995). The military, pressured by all sectors of society to further their own particular interests, struggled to pursue its own interests. This enticement or “moderating pattern” is best captured by Stepan: “The very absence of strong political institutions in a country such as Brazil has meant that all major actors attempt to co-opt the military as an additional supportive force in the pursuit of their political goals” (1971, p. 61). Highly politicized, the military was wooed by different political groups and given legitimacy to intervene from its own positions. The consequence was a legacy of factionalism bestowed on the military, which ultimately led to the new type of intervention that took place in 1964, albeit one devoid of the type of historical hostility that characterized the relation with one of the parties in the Peruvian and Argentine cases. We summarize the argument developed so far by presenting it in Table 9.1. To provide a clearer account of the connections among the factors highlighted in the table, we present in greater detail the unfolding of the military regimes in Chile and Peru (space limitations compel us to present only these two). They illustrate different paths and trajectories, the ways in which different challenges were confronted and choices made as well as the impact of historical legacies. Chile’s military regime dealt successfully with the issue of succession and attained institutionalization by setting a constitution under authoritarian aegis. The Peruvian regime, in turn, failed to solve the problem of succession and to attain some degree of institutionalization.

Table 9.1 Historical Legacies, Institutionalization, and Regime Duration A. Historical legacies Military legacy Party legacy B. Institutionalization Succession Role separation Repression Outside support Constitution C. Duration (outcome)

Brazil

Chile

Peru

Argentina

Legalist Enticement

Legalist Aloofness

Personalist Hostility

Personalist Hostility

Yes Yes Less harsh Yes Yes 21 years (1964−1985)

Yes Yes Harsh Yes/No Yes 17 years (1973−1989)

No No Mild Yes/No No 12 years (1968−1980)

Yes/No No Harsh Yes/No No 7 years (1976−1983)

188

Felipe Agüero and Julian Brückner

Exploring the Paths to Stability and Crisis: Historical Legacies and Military Rule in Chile and Peru The Case of Chile

By the time the military moved against President Salvador Allende in the coup of 1973, its most important internal divisions had been removed. Cohesiveness was influenced by the legacy in the relational patterns between the military and political parties, which in turn influenced the stance on parties that the regime would take. A ban on political parties of the center and left was decreed from day one while parties of the right consented to go into recess. The pattern of aloofness in the military’s relation to parties anticipated this go-it-alone position that so clearly separated it from the Brazilian authoritarian stance toward parties. The military made a series of decisions addressing the issue of succession. The initial idea that the headship of the junta would rotate among its members was soon replaced by a decision to make the head of the army, General Pinochet, the permanent president of this collective body, a position from which he labored to have the junta agree to his nomination as president of the republic. Ultimately, the military opted for the model of no rotation; that is, that Pinochet would lead the executive power for the entire duration of the regime. Once a constitution was adopted in 1980 and until it would become fully enforced in 1990, its transitory clauses constitutionalized the presidency of General Pinochet and established that the junta would unanimously submit to a plebiscite in 1988 the single candidate for president for the next eight-year term (1989−1997), with the possibility of reelection. Thus, the issue of succession was settled by establishing a system of no rotation, with one man in charge, and the possibility of change (or revalidation) of the top leader in 1989. This mode of institutionalizing the top leadership gave enormous stability to the regime. This particular junta agreement, rather arduously pushed by Pinochet himself, was justified on grounds that administrative procedures demanded such centralization. The military could not dismiss this view given the value attached to hierarchy with single unipersonal command. Subsequently, Pinochet stayed out of the junta, being replaced by the army’s deputy service chief. The junta would then completely and exclusively take on the function of the legislative organ, and maintain constituent power, operating under the principle of unanimity (Arriagada 1985). The unusual centralization of power in one person has led many to follow Remmer’s (1989) influential characterization of the Pinochet regime as one-man personalist rule. However, we disagree and argue with Robert Barros (2002) and other experts on the case of Chile (Huneeus 2000; Valenzuela

Historical Legacies and the Institutionalization of Military Rule

189

1995) that, despite the preeminence of Pinochet by means of his control of the executive, he was still bound by rules and the power of the military junta. “Ultimately, Pinochet was limited by the very institutions that gave him power—the armed forces and the state—institutions that were subject to the framework of a highly regularized legal process” (Valenzuela 1995, p. 22). The power of the junta as a limit to Pinochet’s personal powers was further enhanced by the unanimity decision rule and the autonomy it acquired by Decree Law 527 passed in 1974 that stipulated that junta members as commanders-in-chief of their services would hold their posts for life. Behind these early steps in institutionalizing rule were the professional tradition of obedience and hierarchy as well as the presidentialist tradition within which professionalism had evolved. The military took this tradition from the constitution and continued to apply it as a means to regularize its power. The legalistic tradition expressed itself in the expedient move to organize executive-legislative roles around the president and the junta, and to start equipping the latter with advisory committees staffed by military and civilian advisers to handle an increasingly heavy load of legislative activity. The endogenously driven traditions operating through the military to assert themselves in these early institutionalizing features were reinforced by learning from outside experiences. The regime moved to promote a clear separation between the military as government and the military as institution, aiming to avoid the pitfalls of politicization that it saw eroding stability in the military regimes of Peru and the previous military dictatorship in Argentina. Those divisive situations were fully visible to the Chilean military, as were the tensions in the 1974 Brazilian succession. These lessons reinforced the need for the Chilean military to stick to their professional traditions and an emphasis on nonpoliticization and nondeliberation. Still, military participation in government was not negligible. Half of all ministers serving at different points during the length of the regime were military officers (Huneeus 2000), and many more served in the junta’s legislative support staff and in newly created agencies to support and advise the work of the presidency. Deliberation and political decisionmaking, however, were confined to the top-level service chiefs. Officers serving in the government did not perform any military representation function, and General Pinochet was the bridge between the military as government and the military as institution in his double role as president and army commander-in-chief.6 In addition, the military forged a strategic alliance with a group of economists (the so-called Chicago Boys) that provided the substantive content and long-term perspective for the regime’s exclusionary orientation. This group provided the military leadership with a buffer from particular interests

190

Felipe Agüero and Julian Brückner

of bourgeois sectors that saw themselves as natural allies of the military. They helped provide a rational, universal, and “scientific” economic discourse to promote elite cohesion and insulate the military from outside pressures (Biglaiser 1999; Garretón 1983, p. 138), fitting perfectly within the previous pattern of aloofness. This alliance was possible due to a higher degree of autonomy of the military from dominant economic groups than was seen in other cases, especially in Argentina (Pion-Berlin 1986). The constitution was the other fundamental pillar of institutionalization. A committee of notables was charged early on, in 1973, with drafting a new constitution. A slow start shifted to fast-track in 1977. General Francisco Franco of Spain had died in 1975 and sympathetic observers in Chile witnessed in dismay the speedy unraveling of Franquist institutions as the transition unfolded. Franco had not left things tied up as tightly as he had alleged, and the regime associates in Chile wanted to prevent such an outcome for their regime. The constitutional draft was approved by the junta and made ready for a referendum in 1980. To be fully enforced in 1990, the constitution contemplated participatory institutions with limited pluralism and full tutelary powers for the military. During the intermediate period, presidential powers were enhanced by exceptional prerogatives for General Pinochet. The regime thus assured longevity with direct military rule for ten years following the referendum, and with tutelary roles after that. Equipped with these layers of institutionalization, the regime could confront severe crises unscathed. The first severe challenge came from junta member General Gustavo Leigh, the most active and aggressive instigator of the coup and a hard-liner with stern views on repression, who opposed with growing publicity the increasing hegemony of the Chicago Boys in government policy. Pinochet, in agreement with the other two members of the junta, decided in 1978 to act swiftly: Leigh was summoned and asked to submit his resignation. Only after refusing to sign a decree intended to legalize his removal from the junta and upon threatening to convene his air force generals, Leigh and his most senior commanders were ousted and retired (Cavallo et al. 1997). General Fernando Matthei, the tenth-ranking general, then assumed as commander-in-chief of the air force and joined the military junta. Here, a semi-institutionalized military junta served to confront a cohesion crisis swiftly and successfully, and Pinochet’s presumably enhanced power after winning this round did not lead to any diminishing of the junta’s powers that limited those of Pinochet. The next major stumble was the economic crisis and recession of 1982−1983 in which gross domestic product (GDP) fell 14.5 percent. This caused the emergence of massive popular demonstrations against the regime and severe problems with economic groups. Pinochet appointed a succession of finance ministers that were not part of the Chicago Boys

Historical Legacies and the Institutionalization of Military Rule

191

who proved unable to solve the crisis. Had the military not been protected by the 1980 constitution, which restrained all competing groups, the regime likely would not have survived unscathed. But the regime could resist outside pressure, protracted mobilization by the opposition, and internal fragmentation on the grounds that rules had already been legitimately established. However, the social and political context changed dramatically after that, with ominous consequences for the regime. The constitutional clause that had established a plebiscite in 1988, which had elicited much intraregime debate about its necessity, had been seen as a mere formality (albeit a necessary one) to validate the continuity of Pinochet at least until 1997. But with the surge in the opposition, the plebiscite clause started to look worrisome. The opposition managed to conduct a protracted and skillful campaign, first by deciding to abide by the constitution—which it had previously deemed illegitimate—and call the regime up on the plebiscite. Then, it mobilized international pressure to make sure that the referendum would be held under minimally fair conditions. The result was Pinochet’s defeat in the plebiscite. It is important to note that constitutional legalism prevailed in the military under the weight of its traditions. By accepting defeat, the junta stood its ground before a reluctant Pinochet. Then, following constitutional directives, competitive elections for president were held in 1989, which the opposition won, leading it to inaugurate a weak democracy in 1990 under heavy military tutelary powers. Learning the lesson from the Spanish transition, the military did in fact leave things tied up tightly. Pinochet continued as commander-in-chief of the army until 1998, and serious constitutional reforms that weakened military prerogatives did not take place until 2005. As in Brazil, authoritarian legacies pervaded the new democracy. But in Brazil nothing kept Congress from replacing the constitution three years after the transition to civilian rule. Chile, on the other hand, continues (as of this writing) to live under the constitution inherited from the military regime, although many changes were subsequently introduced. The Case of Peru

After ousting President Fernando Belaúnde, a centrist reformer, in October 1968, the military surprised by the determination with which it swiftly unveiled its nationalistic, popular, anti-oligarchic, and anti-imperialist agenda (Thorp and Bertram 1978). Nationalization of oil facilities was followed by the expropriation of banks, companies, and large rural estates; the promotion of national industry; and the organization of popular sectors in what Stepan labeled “inclusionary corporatism” (Stepan 1978, pp. 76−77).

192

Felipe Agüero and Julian Brückner

As an inclusionary regime and marking the difference with neighboring military regimes in the region, it did not engage in the kind of atrocities that characterized its neighbors. Furthermore, although parties ceased to function, varying degrees of political activity were tolerated. This reformist inclination had been developing for years within the Center for High Military Studies (CAEM) and the Military Intelligence System, influenced by French military doctrine, concluding in the need for structural reforms to confront social unrest and violence and lead the country to modernization and national unity. Because APRA (initially a staunchly reformist party) had shifted to the right and joined the conservatives in opposing reforms, the demand for political reforms was left politically void. The military thought it had the responsibility to fill the space and assume the national task of initiating deep structural reforms. Under the APRA-targeted hostility pattern, the military (as in Argentina) also thought it had to intervene precisely so that the cycle of military interventions would come to an end. This presented the Peruvian military regime with the heavy additional task of having to address the creation of outside bases of support. The military had created an enemy in the oligarchy and the reform parties were not available, either because they had abandoned the commitment to reform or because they had annulled each other in mutual rivalries. Still, those parties continued to carry mass support, making it difficult for the military to create their own support party where no potential followership was available (Stepan 1978). Ultimately, the military decided to organize potential beneficiaries of the reforms in the rural and urban sectors, under the supervision of an especially created state agency where this support would be encapsulated: the National System of Social Mobilization (SINAMOS) (Contreras and Cueto 2007). At the start of the regime, the military cohered around the need for reforms and the blame of the oligarchy. However, given the legacy of personalism in the military, there existed several internal currents, the most dynamic of which was the group of more radical reformers who came from CAEM and coalesced around the president, General Velasco and his Presidential Advisory Committee (COAP). On the other end was a group of conservatives, present in the army but mostly consisting of admirals in the navy. There also emerged a third, “developmentalist,” group united by the objective of reform program implementation and more sympathetic to state capitalism than to the more radical pro-distribution group around Velasco (Philip 1978). Against this background of factionalism, how did the military organize to rule? The first decision, three months into the regime, was that the head of the army and coup leader would be president and would remain

Historical Legacies and the Institutionalization of Military Rule

193

beyond the end of his commission. That decision, however, was made almost by default when a cabinet meeting meant to address the issue failed to even discuss it, leaving it to the military junta, which issued the respective decree. This settled an initial uncertainty and killed any jockeying for power by possible alternatives. However, no succession mechanism was established: General Velasco would remain until the end of the regime or until he was pushed out. Second, government policy decisionmaking resided with the all-military cabinet, the military junta, and the general president that chaired it. These decisionmaking bodies were strongly backed by other military-staffed entities such as the COAP, the general staff, and the intelligence services (Cleaves and Pease Garcia 1986). Although there were officers exclusively dedicated to professional military tasks, especially at a time in which tensions with Chile rose, the lines dividing military as government and military as institution were significantly blurred. General Velasco initially proved to be quite skillful in managing and conciliating different positions at the all-military cabinet level. Tensions began to emerge during the fifth year of military rule among the various tendencies confronting policy dilemmas such as greater confrontation or compromise, more participation, or firmer authoritarianism (Vásquez Huamán 2000). The economy began to slow down partly as a result of policy hesitation: attacking business interests and the right, but without going decisively left either (Thorp and Bertram 1978). And most unexpectedly to military reformers, reforms had failed to generate public support. To the contrary, support for APRA and other opposition forces had not receded and the failure of the top-down bureaucratic attempt to organize support via SINAMOS became evident. The revolutionary clique had cornered itself into a situation of extreme isolation, with no outside allies or social support. Inclusionary policies had alienated both the bourgeoisie and the forces that those policies were supposed to benefit. The military-inspired logic of inclusion through control from above had turned inclusion and mobilization into complete isolation (Kruijt and Tello 2002). And with no institutionalization of military rule, the early decision to appoint Velasco for the duration turned into the issue of who decided what that duration was: whoever garnered the power inside the military to do so. General Velasco was ousted in August 1975 by the leading general in the cabinet, Francisco Morales Bermúdez. Facing increasing manifestations of public discontent, unease with the problems created by the strategy of inclusion/incorporation, frank disagreement with the ideological orientation that had prevailed so far, and fearing army divisiveness out of politicization and personalization of power, Morales obtained army support for the coup.

194

Felipe Agüero and Julian Brückner

The government eventually switched tracks: it ousted the reformist generals; reconnected with the alienated bourgeoisie; gradually terminated the inclusion strategies steered from SINAMOS; and, urged by escalating economic difficulties, initiated stabilization policies. Sixteen months after taking over from Velasco, General Morales announced a process of military extrication and return to democracy, although holding on to power until 1980. Facing growing antimilitary sentiments by popular sectors frustrated by the new policies, and the largest demonstrations seen in a long time, the government called elections for a Constitutional Assembly in 1978, in which deliberations would be conducted freely, albeit still under the aegis of military rule. APRA obtained a plurality and parties of the left came in stronger than expected. The assembly elected APRA’s longtime historic leader—Víctor Raúl Haya de la Torre—to preside over the deliberations, and to head a coordination committee to meet regularly with the military executive. The assembly approved a new democratic constitution, with military consent as it granted it ample autonomy, and a presidential election was held in 1980 that was won by Fernando Belaúnde, the reformist candidate that the military had ousted in 1968. The military regime’s most lasting legacy was the significant transformation of the countryside by eliminating the rural oligarchy and the creation of a national industrial sector where none existed before 1968. However, its legacy did not include the much sought-after end of military intervention, as the military supported President Alberto Fujimori’s selfcoup against democracy in 1992, nor the end of violence in the countryside, which would reemerge in full force in the 1980s. Conclusion We opened this chapter with the observation that military regimes like the Latin American juntas established in the 1960s and 1970s vary significantly in terms of their duration and stability. Building on the new institutional turn in the study of authoritarian regimes, we have argued that this variation can be explained by the ability of military rulers to achieve a sufficient level of institutionalization—ideally in the form of a constitution—that allows them to deal with problems of authoritarian power sharing and control that emerge in military regimes as well. Military rule, however, presents the distinct challenge of reconciling the inevitable tensions that arise between a professional military hierarchy (the military as institution) on the one hand, and the group of high-ranking officers in charge of political decisionmaking (the military as government) on the other.

Historical Legacies and the Institutionalization of Military Rule

195

Against this background, we identified three potential sources of crisis that emerge at different stages in the tenure of military regimes. In the installation phase, immediately after coming to power, military leaders face the issue of succession. Failure to regulate the access to power, whether formally or informally, is likely to give rise to coups and forceful threats that breed instability. Following installation, military leaders must organize power. This entails a decision on the degree to which the military as institution will be involved in the day-to-day running of government. The critical challenge here is to avoid clashes between the military hierarchy and the divisive realm of politics and political decisionmaking, a conflict that constitutes the Achilles’ heel of military rule. Finally, the organization of power also requires military leaders to choose a strategy for governing society. Here, two factors are particularly relevant. One is the organization of repression, which is critical for exclusionary regimes intent on demobilizing previously mobilized societies. The other is garnering sufficient outside support to promote the implementation of policies and transformational changes in broader society. Outside support may come in the form of civilian technocrats as in the case of the Chicago Boys in Chile or the creation or accommodation of a mass civilian party as in the case of the newly created pro-regime party National Renewal Alliance (ARENA) in Brazil. The extent to which military regimes must rely on repression or outside support, however, is to a large degree shaped by their inclusionary or exclusionary orientation. The Latin American military regimes of the 1960s through the 1980s that we studied in this chapter varied significantly in their ability to meet the challenges of institutionalization and, as a result, were more or less prone to crisis and early regime breakdown. As we have argued, building on a critique by Pepinsky (2014), the current literature on authoritarian institutions in military regimes has difficulties in accounting for this variation because it suffers from excessive voluntarism: military dictators are seen as largely unconstrained in their ability to create institutions in anticipation of survival challenges that arise from the issue of power sharing and control. In this view, underinstitutionalization and regime breakdown essentially become a matter of strategic miscalculation on the part of the dictator. However, we have shown that variations in the institutional structure of military regimes and their ability to deal with crisis cannot be explained by the greater capacity of professionalized militaries to enforce power sharing among the officer corps. The military regimes that came to power in Argentina, Brazil, Chile, and Peru in the 1960s and 1970s were all established by militaries that had achieved sufficiently high levels of discipline, hierarchy, and autonomy to qualify them as professional in the narrow sense of the concept used by Geddes (2009) and others. Yet as Table 9.1

196

Felipe Agüero and Julian Brückner

illustrates, only Brazil and Chile achieved a sufficient level of institutionalization that would allow them to rule for an extended period of time. In contrast, Peru and Argentina failed to resolve even the most basic challenge of succession, leaving them vulnerable to factionalism, crisis, and a quicker return to the barracks. But the generals in Peru and Argentina did not simply underestimate the “right amount” of institutionalization needed to stay in power and achieve their goals. They were, in fact, constrained by historical legacies that originated in different founding conditions. These legacies included, on the military side, a personalist outlook of the officer corps as a result of an idiosyncratic adaptation of European military training, and, on the societal side, a long-standing hostility toward established parties. While the exact relationship between historical legacies, institutions, and regime outcomes will require further elaboration, our exploration of the cases of Chile and Peru shows the need to pay greater attention to the antecedents of authoritarian crisis and stability which, in the case of military regimes, must include a return to a broader, more political understanding of military professionalism. Notes The authors would like to thank the participants of the workshops on “Crises in Autocratic Regimes” held at the WZB Berlin Social Science Center in July 2014 and June 2015, and especially the discussants Emmanuel Gyimah-Boadi and Andreas Schedler, for their insightful comments on earlier versions of this chapter. Earlier versions also benefited from the feedback provided by Svend-Erik Skaaning and other members of the research project “Conflict and Democratisation in the 21st Century” during a research stay by Julian Brückner at Aarhus University in May and June 2016, as well as from the feedback provided by Jonathan Hiskey and the participants of the panel “Political Institutions and Participation Under Authoritarianism, Past and Present” at LASA 2017 in Lima. The authors are grateful for thoughtful comments from David Pion-Berlin, Augusto Varas, and Claudio Fuentes as well as those from participants in the University of Chile’s Instituto de Asuntos Públicos’s research seminar where this chapter was presented as a paper. Finally, they thank the members and associates of the research project “Critical Junctures and the Survival of Dictatorships” at the WZB Berlin Social Science Center for enlightening discussions throughout the process of writing this chapter as well as Maxine Lowy and Simon Haux for their help with language and style editing. 1. Arcenaux (2001) addressed differences in the endurance of military regimes in Latin America, but with a focus on the military’s ability to control the transition. 2. Stepan (1971) aptly captured this central tension with his observation that armed forces in military regimes seek to function in a twofold capacity: the “military as an institution” and the “military as government.” 3. The case of General Pinochet in Chile resembles this alternative.

Historical Legacies and the Institutionalization of Military Rule

197

4. For an overview of major European military missions in South America, see Resende-Santos (2007). 5. In fact, only one civilian president between 1912 and 1980 (Manuel Prado, 1939−1945) could complete his term in office and transfer power to another democratically elected president (Cotler 1995). 6. However, Pinochet did get the army involved in politics when he activated his campaign to pressure junta members to select him for the 1988 plebiscite, bringing army chiefs to express public support. See Varas (1995).

10 Successes and Failures of Authoritarian Governance: The Case of the Arab Spring Frédéric Volpi Regional specialists and comparativists have repeatedly highlighted over the years the shared features that underpinned postcolonial authoritarian regimes in the Middle East and North Africa (MENA). By and large, by the 2000s, the dominant debate regarding the MENA region centered on explaining the resilience of authoritarianism (Posusney and Angrist 2005; Schlumberger 2007). While the scholarship highlighted some continuing problems of governance in these polities, the most common assessment was that the regimes of the region had sufficient domestic and international resources and support to contain any challenge to their rule in the medium term. Before the 2011 Arab uprisings, the focus of much of the research was about understanding the variations in the mechanisms of authoritarian governance that produced regional stability through slightly different means in each country. At the time, accounts of “upgrading authoritarianism” (Heydemann 2007) had complexified the issue of authoritarian resilience; and in so far as they focused on the institutionalized mechanisms of routine authoritarian governance, these explanations of resilience were quite sound. In this context, any process of deinstitutionalization leading to regime change was expected to be a slow, long-term, and mainly elite-driven transformation. Hence, after the Arab uprisings, critics may have had some good reasons to say that Middle East experts underestimated the hidden forces driving change (Gause 2011). Yet such criticism is somewhat tautological. Analysts underestimated them precisely because they were hidden. Had scholars been focusing on other factors of change, they could have weighted them better. But then again, so would have the authoritarian elites of the region, and the fate of the protesters that mobilized against these regimes would have been different too. The generalized failure to see what was coming in the region cannot be attributed simply to the fact that analysts missed some objective significant causes of change. Mainly, these factors were missed because they were not there at the time. Most analyses did not consider the dynamics of a sudden and sizable transformation 199

200

Frédéric Volpi

of grassroots social and political behavior within and across countries— and many still fail to do so today (Schwedler 2015). For post-2011 accounts of politics in the Middle East and North Africa, a main challenge in explaining both authoritarian resilience and authoritarian failure is to make room for the causal factors of the uprising process itself (Volpi 2017). Hence, the task is not simply to reconfigure known explanatory variables with the benefit of hindsight to attribute greater weight to those factors that proved to be important during these particular crises. Rather, there is a need to better understand the systemic role played by crisis dynamics in the longer-term evolution of political resilience of the authoritarian regimes of the MENA region (and beyond). One way to deal with the contingency of these crisis episodes is to examine the strategies and responses of authoritarian regimes to past crises to obtain insights into the changing interactions between different proregime and antiregime social and political actors. The trajectories of authoritarian upgrading, erosion, or decay of the MENA regimes prepared the ground for the Arab uprisings without determining their timing, modalities, or outcomes. The 2011 uprisings started as events that were not dramatically different from past crises that these regimes had experienced. While they were not a mere repetition of past crises, the patterns of interactions between pro-regime and antiregime actors revealed during earlier episodes of contestation provide useful signposts for an analysis of the protests’ options and dynamics at the time of the Arab uprisings, and after it. In the following, besides briefly considering those authors and approaches that have seriously taken into account the dilemmas of contingency, I examine three categories of interactions between the authoritarian state and its subjects that have been significantly reshaped by crisis dynamics. First, I consider the legitimacy-repression nexus, and the narratives proposed by different regimes to confer on state institutions a degree of ideological legitimacy that requires to be only minimally backed by force. Then, I address the issue of patron-client dynamics, and how specific socioeconomic incentives have been provided by different regimes in exchange for political quiescence. Finally, I investigate the role of institutionalized politics in providing an outlet and framework for power sharing and for reform processes that can be engaged by political actors and ordinary citizens. Taking Contingency Seriously Before the Arab uprisings, Charles Kurzman’s (2004a, 2004b) account of another much unexpected regime change, the 1979 Islamic revolution in

Successes and Failures of Authoritarian Governance

201

Iran, highlighted some of the main difficulties in explaining unexpected political transformations after they have occurred. In the case of Iran, common post hoc explanations posited some initial crisis conditions that they retrospectively presented as the causes for the revolution and the fall of the Shah. Yet Kurzman notes that the revolutionary crisis itself was rarely examined in detail to evaluate how far protest mobilization created the conditions of its own success. Hence, generic structuralist and rational choice arguments repeatedly ran the risk of turning into mere rationalization. Countering this trend, Kurzman highlights the centrality and causal importance of shifting perceptions, interpretations, emotions, and ad hoc behaviors of the protesters during this period of rapid deinstitutionalization of the regime. The case of the Iranian revolution illustrates one more time that looking at preexisting conditions to unearth the causal factors that possibly can explain regime change only tell part of the story. The new political dynamics that gain momentum during crises as a result of the interactions between the different pro-regime and antiregime actors at the time provide another element of answer. While these crisis dynamics derive from preexisting interactions—be they repression, co-optation, or support— they do not have to follow, and cannot be subsumed under, those preexisting political opportunity structures enabled by routine governance. Regarding the actors’ strategic choices and behaviors, in particular, it may be counterproductive to use the framework of routine politics to evaluate options because what makes sense in periods of routine governance is unlikely to remain equally relevant during episodes of rapid institutional breakdown (and reconstruction). Guillermo O’Donnell and Philippe Schmitter (1986) already noted in the cases of the democratization processes in Latin America that the challenges created by situations of political fluidity were poorly captured by analyses grounded in an explanation of established authoritarian rule. In situations of routine authoritarianism—or even during routine protest events (i.e., non-system-challenging unrest)—the habitus (Bourdieu 1977) of social and political actors contribute to produce behaviors that entrench the status quo. These are cases when we can witness a slow process of regime decay and erosion that gradually reshapes social and political interactions at the margins. In periods of crisis and rapid deinstitutionalization, by contrast, the interactions between pro-regime and antiregime actors produce far more unintended consequences that undermine the stability of the political system. In the “eventful sociology” suggested by William H. Sewell (2005), it is precisely this new impact of multiple forms of agency on established social structures and political institutions that turns mundane crises into transformative events.

202

Frédéric Volpi

While political crises and revolutionary situations can be a direct consequence of internal or external shocks to the system, they need not always be the outcome of concrete events. Processes of deinstitutionalization can also gain momentum as the result of behavioral changes induced by reframing processes shared by many people (Kurzman 2012). In terms of diffusion, the causal mechanisms that underpin such reinterpretations of political normality remain nonetheless difficult to structure as they are grounded in time-bound intersubjective interpretations (Weyland 2012). Though the Tunisian uprisings and the fall of the Zine el-Abidine Ben Ali regime may have constituted some kind of cognitive shock to the regional model of authoritarian governance, its relevance for other mobilization processes across the Arab world could not have been assumed. Grievances against the regimes, as well as socioeconomic and political conditions, may have been quite similar across the region. But since this was a given, it could hardly be presented as a specific explanation of the 2011 wave of protest mobilization and regime change. How were specific actors, interactions, and discourses rearticulated in a way that facilitated continued support for or acquiescence to a particular authoritarian order? Revisiting the dynamics of authoritarian resilience of different regimes in times of crisis or at critical junctures can help to better understand the range of crisis interactions that have been produced over time. Considering past crises, significant changes of policy, cases of transmissions of power, and so forth give some insights into the types of reframing and behavioral change that have repeatedly (or occasionally) been relevant for both government and opposition, and for the elites and the population. While a comparative analysis of the authoritarian trajectories of different regimes provides some insights into crisis dynamics, it can reveal only part of the spectrum of possible change. In the end, each crisis episode itself introduces another source of variation in the range of possibilities for change. The Legitimacy-Repression Nexus How does the regime manage to appear legitimate to a large enough proportion of the population so that governance is actually achievable without an inordinate amount of state resources being spent on repressing an unruly people? Juan J. Linz suggests that typically each authoritarian regime generally follows a “legitimacy formula” to cope with this predicament. He presents legitimacy in a Weberian perspective as “the belief that in spite of shortcomings and failures, the existing political institutions are better than any others that might be established, and that they therefore can demand

Successes and Failures of Authoritarian Governance

203

obedience” (Linz 1978, p. 16). I use a slightly less ambitious definition which, rather than saying that legitimacy beliefs enable political institutions to demand obedience, considers that legitimacy facilitates acquiescence to the existing status quo. As Johannes Gerschewski notes in Chapter 1 of this volume, institutionalization means placing some structural constraints on individual and collective choices to encourage particular behaviors and strategies and discourage others. Directly or indirectly, it also involves fostering a habitus that shapes people’s expectations and dispositions toward the regime in a self-reproducing and self-reinforcing fashion. In the MENA context, the institutionalization of traditional political and religious authority, particularly in monarchic regimes, provides a good illustration of this process of entrenchment of so-called traditional political rule. However, institutionalization processes can also have perverse (or self-defeating) effects that over time weaken institutional arrangements by fostering views and practices that undermine longer-term institutional stability. Since the very beginning of the postcolonial period, stability in MENA has been explained in relation to the recurrent legitimacy crises of the regimes of the region. Considering the trajectories of the early postcolonial state, Michael C. Hudson (1977) also uses a Weberian typology to highlight some of the features that would constitute central elements of the debate in ensuing decades. These include the processes of modernization of traditional models of governance (particularly in relation to monarchies), and of adaptation of liberal or socialist institutional models of governance (particularly in the new Arab republics). Using these categorizations, it is relatively easy to identify the punctuated evolution of different combinations of authority types in postcolonial North Africa. In the neotraditional monarchies of Morocco and Libya, King Mohammed V and King Muhammad Idrīs al-Sanūsī can be portrayed as having used a mixture of traditional authority and charisma to buttress their political leadership in the aftermath of independence (Hammoudi 1997). In the new Arab republics of Algeria and Tunisia, charisma was clearly a key element in Presidents Habib Bourguiba’s and Ahmed Ben Bella’s political leadership, as were more mundane rational legal types of authority. Then, like now, the challenge for authoritarian leaders, and the regimes that they headed, was the institutionalization of their political authority in a way that facilitated transmission over time to themselves or to their chosen successors. In Morocco, a legitimacy boost to the authority of the monarchy was the direct consequence of the role of the sultan during the anticolonial struggle—a process that helped create what Clifford Geertz (1971, p. 74) calls a “maraboutic king.” This context-dependent charismatic and traditional authority could not be transferred easily to the king’s son a few years

204

Frédéric Volpi

after independence when Mohammed V died. To cope with a charisma and traditional authority deficit, the new Moroccan king, Hassan II, strategically used the repressive apparatus and the legal administrative authority of the state to contain the political challenges emanating from nationalist and leftist opposition movements. A similar decay and erosion of charisma and traditional authority could also be witnessed in the Libyan monarchy at about the same time. In Libya, in the 1960s, King Idris increasingly centralized the administrative and political functions of the state to manage the growing oil wealth of the country (Vandewalle 1998). In doing so, Idris progressively moved away from his traditional religious sources of legitimacy—notably his leadership of the Cyrenaica-based Senussi Sufi brotherhood—and toward a legal administrative model of state authority. In both Libya and Morocco, these efforts at strengthening the authority of the monarchy through a strengthening of the state institutions implied a greater reliance on the repressive apparatus to quell dissenting voices. This process had intrinsic costs for these monarchs, as in these circumstances royal authority could more easily be challenged by a modernizing military institution that began to rethink its role in state building. Failed (Morocco 1971, 1972) or successful (Libya 1969) military coups in the neotraditional monarchies illustrated at the time the importance of intra-elite competition over legitimacy in the making and in the resolution of authoritarian crises. In Algeria, in the absence of monarchy, similar dilemmas centered on the legitimacy of the political role that the military had constructed for itself during the war of decolonization. After independence, revolutionary legitimacy was rearticulated on rational legal sources of authority as the Algerian revolution became institutionalized and the joint rule of the Algerian Liberation Front (FLN) and Algerian Liberation Army (ALN) became normalized (Entelis 1986). This semipraetorian system in which the military institution acted as a guardian for the political elite generated its own dilemmas by producing a particular understanding of political legitimacy. Luis Martinez (2000) emphasizes the relevance of a militaristic tradition in Algerian politics, where seizing state power through armed struggle was repeatedly seen as legitimate by many social and political actors in times of crisis—with the 1992 coup being the latest illustration of these dynamics. It was precisely to downplay this praetorian tendency that Colonel Houari Boumedienne promoted “Arab socialism” in the country after his 1965 coup. In neighboring Libya, Colonel Muammar Qaddafi would do the same, albeit briefly, a few years later after his 1969 coup. The endorsement of Arab socialism not only facilitated the merging of leftist and nationalist legitimacy claims, but it also helpfully prioritized socioeconomic issues over electoralist considerations. Despite this socialist narrative, however,

Successes and Failures of Authoritarian Governance

205

the troubled issue of civil-military relations and the continuing influence of military actors in politics would periodically generate political crises in these polities—as illustrated by the repeated coup attempts against Qaddafi and the difficult succession after the death of Boumedienne in 1978. In Tunisia, the minimal role played by the armed forces during the process of decolonization, and the weak claims advanced by the king, strengthened by default the legal administrative authority of the regime grounded on the main independentist party (Neo-Destour Party) of President Bourguiba (Ware 1985). However, internal challenges from within the state apparatus and external challenges from organized opposition movements, in the circumstances, were the price to pay for the institutionalization of this party-based model of authoritarian governance. Writing in the mid-1970s, Michael C. Hudson observes that, “Tunisia too has its legitimacy problem; but compared to the hidden, simmering politics of the monarchies, the volatile, bitter and tragic politics in the pan-Arab core, and the unformed, often capricious and praetorian politics in the peripheral states, Bourguiba’s republic appears as a relatively well legitimized political system” (1977, p. 377). Nonetheless, in 1978, one of the main political crises in postcolonial Tunisia nearly brought down Bourguiba as a result of a confrontation between the regime and the Tunisian General Labor Union (UGTT; Union Générale Tunisienne du Travail) over the political and economic direction of the country. In all these polities, elite-level legitimacy dilemmas took on a new shape principally from the 1980s onward when the rise of Islamist activism forced MENA regimes to update their ideological and security responses (Burgat 1997). What repositioned more centrally the relevance of Islamism at that time was the growing importance of a global democratization discourse no longer shaped as a direct response to the communist threat, which modified realpolitik and the criteria of international legitimacy (Volpi 2010). This new international context was all the more relevant for Arab regimes whose economies were increasingly characterized by the developmental failures of Arab socialism and other forms of state dirigisme. In ideological and institutional terms, an important side effect of these regimes’ adoption of the neoliberal economic models promoted by international organizations like the International Monetary Fund (IMF) was to reduce the appeal of those left-wing movements that had relied on state support to promote their views among the more traditional sectors of society (Kepel 2002). The downgrading of the developmental state and the privatization of social services facilitated the growth of the educational and welfare activities of Islamist movements. Grassroots mobilization through religious and social activism provided a foundation on which Islamists were able to construct a

206

Frédéric Volpi

more political agenda for governance, locally first and then nationally. In the late 1980s and early 1990s, the political implications of the increased legitimacy of political Islam and electoral democracy were on display in Algeria during a hectic episode of democratic political reforms that resulted in a military coup (Volpi 2003). Tunisia went through a similar reform process in the same period, albeit in a more euphemized form, as the country’s new president Ben Ali sought to stabilize his rule by making a temporary truce with the Islamist opposition. In both cases, failed political liberalization resulted in a hardening of the regime, in the form of a military regime in Algeria during the civil conflict of the 1990s and in the form of a police state in Tunisia. During that decade, the Libyan regime also responded to the political and armed challenges of the Islamists principally by a hardening of its security strategy. In Morocco, by contrast, grassroots Islamic activism was partially co-opted into the state-controlled multiparty system in the mid-1990s when the monarchy legalized an Islamist party as part of its strategy to maintain a divided parliamentary opposition (Zeghal 2008). In the mid-1990s, Olivier Roy (1996) could understandably talk about “the failure of political Islam” considering the lack of success of most Islamist movements seeking to seize state power in the region. However, the growing recognition by the population of the claims to legitimacy put forward by the Islamists meant that the political and security successes of authoritarian regimes came at the cost of an increased delegitimization of their own political claims. Beyond intra-elite competition, and the place of the Islamists in relation to the ruling elites, state-society interactions were characterized by the reduced role played by ideational support in the regimes’ legitimacy formula. After the al-Qaeda attacks of September 11, 2001, the reorientation of the international agenda by the United States to address violent transnational Islamism provided MENA regimes with new external ideational and material resources. Although this new international context did not enable regimes to fix their legitimacy deficit, it (temporarily) facilitated the circumvention of some problematic aspects of legitimacy. In those circumstances, new security discourses and practices served as a fig leaf to cover up those instances where the regimes’ legitimacy deficit was most blatant. The war on terror contributed to the survival of authoritarian regimes with increasingly limited domestic support by maximizing the opportunities for repression domestically and for political and financial leverage internationally. Considering the international dynamics of authoritarianism, Steven Levitsky and Lucan A. Way (2006) stress the importance of linkage and leverage for Western governments to promote their liberal democratic agenda abroad. In the 2000s, however, as the United States and European countries were looking for increased security cooperation from the MENA

Successes and Failures of Authoritarian Governance

207

regimes, linkage and leverage repeatedly worked the other way around as authoritarian regimes used demands for security cooperation to obtain political, financial, and security rewards. The rehabilitation of Qaddafi’s regime by the international community provides a vivid illustration of the reverse leverage induced by this global securitization process. Although war on terror policies provided an external source of validation as well as material benefits for the MENA regimes, they did not significantly modify the underlying dynamics of delegitimization. In particular, continuing international pressure in favor of neoliberal economic forms did not abate in the 2000s and kept creating difficulties for ruling elites still legitimized at least in part by authoritarian social contracts promising socioeconomic development in exchange for political quiescence (Hinnebusch 2006). The Patron-Client Nexus In the MENA region, rentier theory repeatedly outlined how the postcolonial governance dilemmas of oil-rich rent-supported regimes differed from those of countries where the economy was not primarily based on wealth redistribution by the state (Beblawi 1990). Yet by the 2000s, the evolving strategies of Maghrebi rentier states seeking to diversify their economy (Libya, Algeria) and that of nonrentier states seeking to develop a more competitive market economy (Tunisia, Morocco) led the regimes to make similar compromises. Repeatedly, the successful upgrading of authoritarianism meant the endorsement of some free-market reforms to co-opt emerging economic elites into the clientelist networks of state governance (Perthes 2004). Invariably, by passing on some of the state responsibility for socioeconomic development to private actors, regimes sought to reduce the political accountability of the state in the everyday management of domestic economic issues (Cammett 2007). In this respect, the strategies of the four North African regimes to harness neoliberal economic reforms to the strengthening of the status quo can be positioned on a spectrum ranging from conservative (Libya) to innovative (Morocco). In Morocco, through a controlled process of privatization, the king ensured that traditionally pro-monarchy local notables directly benefited from market reforms. In their turn, these local notables contributed to the continuing dominance of the pro-monarchy parties in the more pluralistic parliamentary system that was engineered by the king in the 1990s (Liddell 2010). Traditional forms of patrimonialism, particularly present in the countryside, were thus strengthened through the process of economic liberalization as the regime selectively attributed market opportunities to the relevant elites. The monarchy also retained its influence over the urban

208

Frédéric Volpi

bourgeoisie by ensuring that crony capitalist groups owed their business success to their good relations within the administration as much as to their competitiveness vis-à-vis independent mercantile networks (Catusse 2009). In Algeria, during the civil conflict of the 1990s, the economic choices of the regime were primarily dictated by immediate political and security considerations. As a result, the economic orientations of the regime repeatedly changed over the years, from dirigisme in 1992−1994, to liberalization and privatization under the impulse of the IMF structural adjustment plan in 1994−1998, to a return toward more state interventionism after 1999 during the first Abdelaziz Bouteflika presidency (Volpi 2003). In the 2000s, the revenues obtained from buoyant oil and gas markets ensured that the regime could meet a wide range of domestic demands without having to implement painful structural economic reforms. In practice, to retain a modicum of popular support, the regime redistributed the revenues of the oil rent through its patron-client networks and subsidized an ever larger public sector. In addition, as in Morocco, the Algerian regime opened up political spaces for private economic actors doubling up as political entrepreneurs to supplement its clientelist networks by more competitive forms of business politics. In Tunisia, a mixture of socioeconomic incentives and repressive measures were provided throughout the 1990s to foster acceptance of the modernist vision of President Ben Ali for the country. As Béatrice Hibou (2011) highlights, this authoritarian contract required explicit public expressions of compliance from citizens to have full access to the public goods dispensed by the regime and the state administration. Although the regime’s modernist orientation was meant to favor the urban middle class, the increasingly kleptocratic system that became entrenched over time eroded the confidence of those very segments of population that had initially acquiesced to this authoritarian governance (Erdle 2004). In Tunisia, the intrinsic problems of concentrating the state distributive capacities within an ultradominant party system were not compensated by a relaxation of political controls, as the regime did not allow the traditional bourgeoisie or new crony capitalists to organize themselves as political entrepreneurs. Libya was one of the last regimes of the region to consider a turn toward a neoliberal economic model. The economic sanctions imposed by the United States and the United Nations in the 1990s had contributed to a greater concentration of resources and, therefore, of patronage inside elite groups tied to the Qaddafi family. With the end of the international embargo in the 2000s, Libya witnessed the emergence of a small group of crony capitalists and private entrepreneurs associated with a younger generation within the ruling elite—notably Qaddafi’s son Saif-al-Islam (Haimzadeh 2011). As had happened before in the other countries of the

Successes and Failures of Authoritarian Governance

209

region, such a partial liberalization process led to an increase in the socioeconomic disparities between different parts of the population, as well as between Libyan regions, which undermined the equalitarian rhetoric of the regime. And as in Tunisia, the regime was extremely reluctant to loosen its grip on the political sphere and to allow controlled dissent by this emerging group of economic entrepreneurs. Considering the role of clientelism across the region, Ellen Lust (2009) suggests that viewing electoral processes as business opportunities helps us to better understand the dynamics of controlled electoralism in these regimes. The degree to which such opportunities were proposed in each polity affected the width and depth of support for the regime or the status quo. At the end of the 2000s, we therefore had at one end of this politicoeconomic spectrum the complex networks of royal influence that permeated Morocco’s fragmented multiparty system, and at the other end the crude reward mechanisms of Tunisia’s ultradominant party model. Algeria uneasily maintained a halfway compromise, and the Libyan regime mostly opted out of business politics altogether. While crony capitalism and other forms of neoliberal economic reforms supporting the authoritarian status quo had been useful in partially delinking the legitimacy of the regime from socioeconomic failure, they also contributed to the further erosion and decay of the authoritarian social contract that had initially entrenched these elites. Not only did this neoliberal evolution of the patron-client system undermine the ability of the regime to directly control the degree of quiescence of the population (without a direct use of repression), but it also made more volatile the allegiance of co-opted economic actors, especially when their interests diverged from that of the ruling elites. Institutionalized Politics Before as well as after the 2011 Arab uprisings, the Moroccan political system provides one of the best illustrations of the complex pseudodemocratic arrangements that can be found in the region. The monarchy had devised from early on a strategy of alliance building with different elites—economic, religious, military, rural, and so forth—to remain the main power broker in the country via the reproduction of an ever changing system of divide and rule (Boukhars 2011; Waterbury 1970). Despite the regime’s formal endorsement of multipartyism—including some Islamists—and electoralism, the monarchy had set up a parliamentary system that limited the powers of elected representatives and had established a network of royal appointees to oversee the administration and policy implementation. In addition, the devising of electoral laws that worked against the possibility

210

Frédéric Volpi

of a strong parliamentary majority enabled the king to widen participation in the formal political sphere without handing over a significant amount of control to the opposition (Denoeux and Maghraoui 1998; Sater 2013). For authoritarian elites in the region, the setting up of such an institutionalized system of partial co-optation of the political opposition has proved to be a long and complex process fraught with dangers, as illustrated by the failed Algerian democratic experiment of the 1990s. Although a monarchic system does not necessarily guarantee the establishment of a more resilient pseudodemocratic system, the separation of royal authority from elected institutions does facilitate the setting up of an authoritarian supervision of electoral politics (Yom and Gause 2012). In Algeria, the emergence of an Islamist parliamentary majority unbridled by pro-regime counterpowers was the main reason behind the 1992 coup that precipitated the collapse of the democratic transition. Subsequently, the buttressing of a strong presidential function, and the control over access to that function, became the main political solution for the Algerian regime (Volpi 2006). However, the weakness of parliament notwithstanding, in a republican system even the presidential function remains more open to political challenges than the function of king in a monarchy. In Tunisia in the late 1980s, Ben Ali briefly considered the possibility of ruling in conjunction with a divided parliament before deciding on an ultradominant party system. He made good use of the failures of the democratic transition in neighboring Algeria to defer any political opening and to repress an Islamist opposition that was seen as a main contender for state power (Camau and Geisser 2003). Over the years, the near-complete closure of the Tunisian political field ensured that the regime could more easily repress any emerging dissent in the polity. At the same time, this strategy heightened the personalistic nature of Ben Ali’s rule since he was both president and party leader. Unavoidably, a more personalistic rule implied a more direct responsibility of the autocrat in the country’s developmental failures as well as heightened the dilemmas of succession within the ruling elites. In Libya too, rather than setting up a set of pseudodemocratic institutions supportive of the regime, Qaddafi appeared to be moving in the direction of a republican dynasty that would enable the transmission of power from father to son (Sadiki 2010). In the 2000s, as the country initiated some (modest) neoliberal economic reform, the regime sought to silence the voices of those who were losing out in the process of economic liberalization or who were demanding that political liberalization follow (Pargeter 2006). Despite the rise of a new group of social and economic entrepreneurs mainly associated with Saif-al-Islam, the regime did not seek to develop institutional arrangements that would give them official recognition. At most, Saif-al-Islam’s

Successes and Failures of Authoritarian Governance

211

reform initiatives facilitated the emergence of a tolerated civil society activism that remained removed from networks of governance. Commenting on the evolution of Arab authoritarianism over the decades, Burhan Ghalioun (2004) suggests that by the 2000s these regimes had lost the popular support that once hid their authoritarian core. The populist policies that had underpinned various forms of Arab socialism and that had empowered many in the middle and lower-middle classes after independence increasingly were no longer effective. At the same time, nationalistic and anti-imperialist forms of popular mobilization increasingly ran out of credit. This shift in material capacities and ideational strength led the more perceptive authoritarian elites of the region to institutionalize pseudodemocratic systems of political representation to provide mitigating circumstances for their rule. The others estimated that the costs of changing trusted methods of authoritarian control were greater than the benefits that they could derive from it. Transformative Episodes and Evolving Resilience The different trajectories of authoritarian upgrading, erosion, and decay outlined above prepared the ground for the Arab uprisings without determining their timing, modalities, or outcomes. When the uprisings started in 2011, they generated a set of circumstances that at first were not dramatically different from past crises that these regimes had experienced. The slow processes of deinstitutionalization via decay and erosion and shocks to the system—from the cognitive reassessment of the effectiveness of grassroots protests to foreign military interventions—as well as new grassroots interactions between the regime and the protesters all contributed to produce new crisis dynamics. The issues of the legitimacy of the ruling elites, of the nature of the authoritarian social contract, and of the role of the political opposition all came into play to create new configurations of crisis in each polity. In each country, past experiences of contentious politics were rearticulated into new behaviors that shaped new arenas of contentions, which in their turn shaped specific dilemmas of repression and of governance for the regimes. As in earlier crises, the attempts by the ruling elites to reassert a working legitimacy formula by using tools of repression and co-optation produced mixed results. Such mixed outcomes reflected the partial adequacy of the regimes’ standard answers to political challenges—designed in the light of past crises—to the circumstances created by the 2011 uprisings. In the years preceding the uprisings, as authoritarian stability became constitutive of authoritarianism, political actors from the top of the international community all the way down to the street level came to acquire

212

Frédéric Volpi

certain expectations about politics in the region. These included the assessments that dissent will be effectively repressed, that the opposition will be co-opted, that electoral processes will be corrupted, and that protesters will be bought off (Lust 2005). In their turn, these expectations reinforced particular behavioral trends—not participating in elections, not participating in political protests, not overtly voicing opposition to the regime, using patron-client networks to obtain favors, and so forth. These political understandings and the resulting (a)political behaviors became locked into a selfreinforcing set of strategic choices that indirectly supported the material and ideational tropes of authoritarianism. Over time, however, some gaps opened up in the regimes between the material factors underpinning authoritarianism and the causal beliefs about authoritarian resilience that contributed toward stabilizing these systems. These gaps were made apparent by the first stirrings of protest, and those regimes that were unable to close them rapidly enough then faced an ever growing protest cycle. Kurzman suggests that in revolutionary situations, “the greater the degree of deinstitutionalization, the harder it is to argue that people are falling back on an established pattern of behavior without being aware of doing so . . . the greater the break from routine, the more likely that people will be aware of the break” (2004b, p. 347). Whatever may have been the causes of the initial outbreak of unrest in the countries of the Arab uprisings, as soon as the protests gained momentum it was to be expected that the citizens would start to question their routine obedience to the authorities. As they began to consider alternative courses of action, increasing confrontational interactions with the state apparatus repeatedly undermined the ability of the ruling institutions (including the security apparatus) to maintain the same degree of authoritarian control. As increasing cognitive dissonance facilitated the spread of new protest behaviors, previously tried and tested tools of repression and co-optation became less effective (Allal and Pierret 2013). While the processes of deinstitutionalization during the uprisings are structured by the longer-term erosion or decay of each authoritarian system, situations of rapid deinstitutionalization are not merely conjunctures that resolve known preexisting tensions. In these extraordinary revolutionary situations, the intersubjective reassessment of political opportunities is an important factor of change that is only loosely connected to the structural constrains of routine authoritarian governance. Transformative episodes correspond precisely to these sequences of rapid cognitive and behavioral changes when political order was restructured and new dynamics superseded old ones (Sewell 2005). As in earlier crisis episodes, the participants in the 2011 Arab uprisings (both on the side of the protesters and on the side of the regime) did not merely resolve preexisting tensions

Successes and Failures of Authoritarian Governance

213

and dilemmas, but they also created new ones alongside new identities, practices, and narratives. In relation to the legitimacy-repression nexus, old dilemmas took a new dimension as protest events made publicly explicit the legitimacy deficit of the ruling elites. This deficit deepened and widened as the protests spread in those countries where the regime failed to respond to the demands of the street and assuage the protesters’ anger quickly enough. In effect, the weak ideological credential and discursive resources of the regimes—as illustrated by the desperate calls for calm on national television by Tunisian president Ben Ali, and in Libya by Qaddafi and his son—meant that repression was repeatedly used by default to quell the protests. Repeated reliance on publicly visible forms of direct repression contributed to diffusion of the assessment among different publics, at home and abroad, that the security forces were making an illegitimate use of violence, which thereby further weakened the state’s claim to yield legitimate violence (Pearlman 2013). In its turn, deepening delegitimization of the state security apparatus led to an increase in the ineffectiveness of repression—be it entirely (revolution), partially (civil conflict), or marginally (temporary crisis). In a similar way, patron-client dynamics were affected by a process of rapid deinstitutionalization, which put in jeopardy the privileged position of patrons and thus their relevance for their clients. Protest episodes increased the pressure on the patronage networks of the regimes to operate at full capacity to incentivize protesters and would-be protesters to turn away from protest strategies in exchange for material rewards. In this context, the strategic choices made by both patrons and clients created particular local and national dynamics, as all these actors sought to maximize their gains by taking calculated risks. Even local big men who usually collaborated with the state authorities could in such extraordinary circumstances estimate that siding with the protesters was in their best interest—albeit temporarily if the tide of protest turned (Smaoui and Wazif 2012). As in the case of ineffective repression generating more dissent than it suppressed, attempts to maximize the advantages obtained through patron-client relations could generate unintended consequences, including the actual breakdown of established networks of patronage (especially when generalized unrest prevented the functioning of the basic tools of economic redistribution, as occurred in Libya and in Tunisia). Finally, a similar strategic process of give and take was at work in institutionalized politics. In these situations, the choices made and the outcomes produced were not necessarily a reflection of the cost-benefit analysis made by preexisting (individual and collective) political actors on the basis of known constraints and opportunities—though they could also be that. They were specifically too, as in most cases of severe political crises

214

Frédéric Volpi

(Dobry 2009), the result of new logics of action and logics of situation that were not relevant previously but that became causally important in these new circumstances. In hegemonic or quasi-hegemonic authoritarian orders (Libya, Tunisia), system breakdown and regime failure as an unintended consequence of leaderless/leaderful mass protests was facilitated by the absence or irrelevance of institutionalized opposition actors. In pseudodemocratic systems, established opposition actors directly or indirectly constrained and were constrained by the rise of revolutionary practices and discourses during these protest episodes. In the complex institutionalized system of Morocco, most clearly, the strategic choices of established political parties and unions contributed to the entrenchment of a reformist pathway over a revolutionary one (Buehler 2013). The different types of regime crises and the political trajectories generated by the 2011 Arab uprisings are a direct outcome of different logics of action and situations. While all the protests pointed to a similar range of demands and grievances, these were the products of particular trajectories of authoritarian decay, erosion, and resilience. The longer-term dynamics of authoritarian governance meant that different (or the same) state responses were more or less meaningful and satisfactory for the protesters in each country. And equally, the regimes themselves had a greater or lesser amount of resources available to invest in each type of strategy. Longer-term factors may have laid the ground for the start of these crises, but the uprisings themselves shaped the new logics that produced specific outcomes in each country. Rapid deinstitutionalization meant the emergence of new arenas of contention where previously marginal discourses and behaviors could become widespread. When and where regimes failed to contain these new practices efficiently, it led to ever deepening deinstitutionalization and to the normalization of protest behaviors instead of a return to (reformed) authoritarianism as in earlier political crises. Strategic success for the regimes meant adapting their responses swiftly enough to changing conditions to provide once again an effective mixture of incentives and repression that encouraged protesters and would-be protesters to retain or revert to non-system-challenging repertoires of contention (Volpi 2017). The strategic success of protest mobilization—and of the protestors insofar as this was their objective—was in breaking down established societal and governance routines. This success applied only to the dynamics of rapid deinstitutionalization that were instrumental in the fall of particular authoritarian elites. Yet regime crisis could also turn into a deeper state crisis whenever protest behaviors and deinstitutionalization became so entrenched that a new institutional order could not be firmly reestablished in the short to medium term. In this case, new governance models established themselves at the subnational level, as in the Libyan civil conflict,

Successes and Failures of Authoritarian Governance

215

and began to structure routine social and political governance in the polity. Whenever the episodes of rapid deinstitutionalization were more transient and accompanied by the emergence of a new consensus on governance among the politically relevant elites, as in Tunisia, they could produce an effective top-down process of reinstitutionalization—in this case, according to democratic rules. In authoritarian systems where the regime crafted a response that diverted the political and socioeconomic momentum of protests toward more or less substantial reform pathways, mobilization ended up reinforcing as before an authoritarian status quo.

11 Crisis? What Crisis? Concluding Thoughts Christoph H. Stefes

This chapter brings together some of the key empirical findings and theoretical discussions of our book and also offers some critiques and proposes ways to enrich the study of crisis in autocratic regimes. The focus here is on a few themes. First, what does the concept of crisis mean to the contributors of the book? Do they share a basic understanding of the concept? Second, what are the root causes of crises? Third, what are the factors that incrementally undermine institutions of autocratic regimes? Fourth, what do autocratic leaders do to fortify their regimes in general, and the underlying regime institutions in particular? Fifth, what does the research presented in this book tell us about the stability of the current regime in China? And how vulnerable is China to a severe economic downturn? Finally, how can we advance the study of crisis in autocratic regimes? Which questions have remained unanswered? Crisis as a Concept Throughout this book, it has become clear what the contributors do not call a “crisis.” It is not a period of prolonged conflict between the regime and society. In fact, as Andreas Schedler points out in Chapter 3, although recurring public protests might be a nuisance to the regime leadership, they do not necessarily usher in regime collapse. It also is not a longer period during which the regime fails to deliver public goods or hand out special rewards to its supporters. All of these latent failures and shortcomings might culminate in a crisis, but they are the causes of a crisis and not the crisis itself. For instance, the recurring periods of declining regime support that the Ugandan regime faced, during which it responded with institutional reforms, do not qualify as crises (see Chapter 6 by Svein-Erik Helle and Lise Rakner). In fact, the regime anticipated those looming crises and responded to them with reforms that have 217

218

Christoph H. Stefes

been quite effective in stabilizing the regime—at least, in the short to medium run. So what is a crisis? When the contributors to this book talk about crisis, they mean a short moment in history—maybe a few weeks or months—during which the very survival of a regime is at stake. This is not about a regime’s inability to secure pensioners a decent standard of living. This is about masses of people pouring into the streets and demanding fundamental democratic reforms, undeterred by thousands of guns being pointed at them. The occurrence of a crisis means that a tipping point has been reached. As Frédéric Volpi explains in Chapter 10, new actors suddenly enter the stage and, though previously marginalized and contained, these actors start to make a difference. They can now pose an existential threat to the regime. Crisis can also constitute a moment in which the interests of powerful and established actors rapidly change, causing them to withdraw their previous support for the regime and even turn against it. Usually, during a crisis both processes set in, mutually reinforcing each other so that regime decay rapidly accelerates. The reinforcing processes that undermine support for the political leadership among the elite and bolster mass opposition against the regime might actually be the defining moment of a crisis. That said, how do we know that an autocratic regime has reached a critical moment in which its survival is threatened? As Schedler argues in Chapter 3, there might be no objective way to know. Crisis is in the eye of the beholder. If the regime leadership feels that its rule is at stake, the regime is in a crisis even if the leaders overestimate the threats to the regime. This still begs the question of how we could know that the leadership feels severely threatened. By looking at its actions, Schedler proposes. Since violent and especially lethal repression is risky and costly, leaders use this form of repression only as a last resort. In fact, as Milan W. Svolik empirically demonstrates in Chapter 4, the solving of political conflicts through brute force is a recurring theme in autocratic regimes. It is important to note, however, that relying on lethal force is only occasionally successful. In fact, brutal repression often backfires, as it outrages the population and members of the security apparatus (Chenoweth and Stephan 2011b; Tanneberg, Stefes, and Merkel 2013). That might be the key reason that regime leaders abstain from using lethal repression even under the most threatening circumstances. They have simply given up, as did the leadership of the German Democratic Republic (GDR) in 1989 when facing hundreds of thousands of protesters on the streets of East Berlin, Leipzig, Dresden, and other cities and towns. Using lethal repression under these circumstances secures only lengthy jail sentences after the collapse of the regime. Why risk it?

Crisis? What Crisis? Concluding Thoughts

219

Yet if regime leaders give up, we will never know whether the leadership really knew about the severity of the challenge or whether they simply considered it fruitless to rely on brutal force to counter this threat. In that case, only the collapse of the regime, as an objective indicator, will tell us post hoc that the regime was actually in a crisis. In sum, reliance on lethal force and regime collapse are the indicators of autocratic regime crises. However, mass protests and elite defections that are not met by lethal force and that do not lead to regime collapse do not constitute a crisis. Looking at the cases that are investigated in this book, it appears that the contributors largely agree with this definition. The mass uprisings in the GDR and China in 1953 and 1989, respectively, were unsuccessful but involved lethal repression (see Chapter 2 by Martin K. Dimitrov), just like the Andijan uprising in Uzbekistan in which the Islam Karimov regime successfully countered with lethal repression (see Chapter 7 by Max Bader). The mass uprisings in Kyrgyzstan (2005 and 2010) as well as in Georgia (2003) led to regime collapse, but involved the use of lethal repression in only one instance (Kyrgyzstan in 2010; see Chapter 7 by Bader). The use of lethal repression was widespread during the Arab Spring and sometimes led to regime collapse and sometimes not (see Chapter 10 by Volpi). On the other hand, in Uganda no lethal violence was used in response to elite defection and the regime leadership nevertheless managed to stay in power. Therefore, in Chapter 6, Helle and Rakner do not consider these episodes moments of crisis. Principal Causes of Regime Failure Autocratic leaders face two key challenges: from the streets and from their own entourage among which the military often poses the biggest threat (see Chapter 5 by Aurel Croissant, Tanja Eschenauer, and Jil Kamerling as well as Chapter 9 by Felipe Agüero and Julian Brückner). To address these vertical and horizontal challenges, autocratic regime leaders have to legitimize their rule toward the masses and the elite. Justifying one’s rule with reference to noble causes such as God, national security, or justice undoubtedly helps. Yet ultimately, people want jobs, schools, and welfare and members of the elite demand to be rewarded handsomely for their willingness to put their valuable resources (guns, economic resources, and the loyalty of their followers) into service of the regime. Finally, autocratic regimes always face real or potential opposition, which it tries to contain through (the threat of) repression. In short, autocratic regimes have to secure the support of the masses, secure the continuous loyalty of the elite, and repress its opponents (Gerschewski 2013;

220

Christoph H. Stefes

Gerschewski et al. 2013; see also Chapter 1 by Johannes Gerschweski). As Bader puts it in Chapter 7, autocratic regimes’ survival strategies consist of “an often intricate mix of policies and practices aimed at, on the one hand, winning popular and elite consent and, on the other hand, suppressing real and potential opposition.” Continuous failure in either one of these areas will inevitably produce negative spillovers. As Volpi argues in Chapter 10, a regime that loses its legitimacy has to scale up repressive measures to contain a growing opposition. However, while masses and the elite might be willing to tolerate some level of repression as long as the regime provides rewards, they condemn out-of-control repression, especially if they are disappointed by the regime’s performance. In other words, repression cannot compensate for declining legitimacy in the long run. In fact, repression accelerates the decline in regime support and, thereby, makes repression ever more costly and hence less effective. Think about a car engine of which just one part is defective—for instance, a clogged air filter. While the driver might overcome the negative consequences of a soiled air filter by driving at high revolutions per minute (RPM), the engine eventually will break down, probably due to overheating. Hence, just like drivers who should maintain their cars, autocratic leaders have to address challenges early on to stay in the driver’s seat. This requires reliable monitoring systems of the kind that Dimitrov discusses in Chapter 2. Most importantly, and for reasons discussed elsewhere by numerous scholars (Boix and Svolik 2013; Gandhi and Przeworski 2007; Wright 2008), autocratic leaders attempt to institutionalize their rule. That is, autocrats establish formal and informal rules and norms that, among other things, provide credible commitments to the elite (e.g., through powersharing arrangements), divide the opposition (e.g., through semicompetitive elections), and secure the loyalty of the masses (e.g., through targeted public programs). It has been the goal of this book to analyze the causes that lead to the slow disintegration of institutions in autocratic regimes that are set up to shore up public support, create and maintain elite loyalty, and repress the opposition. Causes of Institutional Decline As Gerschewski and Svolik argue in Chapters 1 and 4 respectively, compliance with institutions can never be taken for granted. Rules and norms are inherently contested because institutions are often ambiguous, leaving room for different interpretations, especially if they are not formalized. In the end, institutions are only as stable as the underlying balance of power,

Crisis? What Crisis? Concluding Thoughts

221

which they partially but not entirely engender. As Svolik posits, two features characterize autocratic regimes: (1) no autonomous higher order exists that could enforce agreements among key actors; and (2) violence is therefore the ultimate arbiter. Those actors that have accumulated all the power create and change institutions, but they are not bound by them. To paraphrase a famous dictum, there will be rule by institutions, not rule of institutions. Only when power is somewhat distributed—not necessarily evenly though—do all actors fear that blatant violations of rules will incur painful reprisals. Only then do institutions do what they are supposed to do: constrain and enable actors’ choices and related behavior. Yet institutions are not just pawns in the hands of the powerful to be created, changed, and manipulated at will. First, as Agüero and Brückner correctly point out in Chapter 9, historical legacies constrain the choices that regime rulers have when setting up institutions. Second, once created, institutions have autonomous power. As Svolik finds out in his study of term limits in Chapter 4, institutions are able to maintain a balance of power. In other words, they are able to create the conditions under which they strive. Agüero and Brückner come to the same conclusion. Term limits and clear rules of succession stabilize military regimes. However, particular balances of power can be upset for various reasons to which Gerschewski alludes in Chapter 1. Institutions might have defects that from the start or later down the road cause developments that upset existing balances of power. Alternatively, or in conjunction with institutional defects, quasi-parameters might change that lead to an incremental redistribution of power. External shocks might do the job much more suddenly and radically (e.g., the loss of the Falklands War that led to a rapid delegitimization of the Argentine junta). Yet this book is primarily concerned with the slow erosion of institutions. In the following, I highlight some of the causes of institutional decay related to shifting balances of power that are caused either by institutional (birth) defects or changes in quasi-parameters. As Croissant, Eschenauer, and Kamerlin argue in Chapter 5, building on earlier work by Barbara Geddes (1999), military regimes are inherently unstable. Their birth defect lies in their inability to co-opt civilians and fellow officers. Since their primary interest does not lie in the maintenance of political power, but in the unity of the military organization and the defense of the country, military leaders retreat from power rather than risk divisions among the top military brass. In the words of Agüero and Brückner in Chapter 9, the military is confronted with “the inherent contradiction of performing both military and government functions.” Institutions that attempt to overcome factionalization within the military might not be well designed precisely because factionalization has made it difficult or even

222

Christoph H. Stefes

impossible to set up appropriate institutions in the first place. The birth defect of military regimes is therefore the inability to institutionalize their rule appropriately, leaving many questions unanswered such as duration and succession rules and the relationship between regime and society (e.g., ways and means of repression). Even when not officially in charge, the military often attempts to shape politics in autocratic regimes from the sidelines. This is facilitated by civilian leaders who try to secure the loyalty of the military by granting the military leadership veto power in certain matters (e.g., defense and foreign policy), reserve cabinet positions for members of the security apparatus, and provide the military with special economic privileges (as in Egypt). Yet as Croissant and colleagues conclude in Chapter 5, these institutional ruses are not always successful. In the end, the military is the most powerful actor in many autocracies. An even balance of power between civilian and military leaders is therefore hard to achieve even with the most elaborate institutions. In fact, buying the military creates perverse incentives because it invites the military to meddle in government affairs during times of societal upheaval, which provides the military with an opportunity to blackmail the civilian leadership. In the long run even though the military might initially be willing to clamp down on the opposition, the civilian leadership will be unable to fulfill the increasing demands of the military, thereby setting the stage for a coup. Dividing the security apparatus and playing the various branches off against each other therefore might be the only way that civilian leaders could achieve some level playing field. Setting up paramilitary groups, such as the Revolutionary Guard in Iran, might keep the power of the regular military in check and the institutions of repression under leadership control. Semicompetitive elections are another institutional means to keep the balance of power between regime incumbents and the opposition. Yet these elections must work in ways that favor the regime leadership. Otherwise, incumbents might quickly realize that elections can also turn against them. Ultimately, this fine balance depends on a stable socioeconomic environment. As Helle and Rakner argue in Chapter 6, the ability of Uganda’s leadership to keep winning elections hinges on several quasi-parameters. First, regime support is maintained through patronage networks. Yet patronage sets perverse incentives since followers of the regime expect increasing rewards, which the regime might not be able to deliver especially in times of economic downturns. Moreover, Uganda’s ruling party receives critical support from an older generation that feels grateful for the regime’s ability to restore and maintain order after years of civil unrest. It is also able to maintain its political dominance by containing opposition activities in the rural areas. However, these two parameters are changing. New generations of voters do not feel indebted to the regime, as they have only a faint rec-

Crisis? What Crisis? Concluding Thoughts

223

ollection of the Bush War. And increasing urbanization undermines the regime’s ability to marginalize the opposition with its urban strongholds. In short, parameters in Uganda are currently changing in a way that shifts the balance of power in favor of an opposition that is unlikely to defend current institutions that work so much to their disadvantage. Parameters are not only of a domestic nature. Foreign actors and international developments also shift balances of power within autocratic regimes. For instance, as Volpi shows in Chapter 10, the global rise of neoliberalism has shifted the balance of power in the Middle East away from the autocratic regimes in the region. For instance, institutions of legitimation were difficult to maintain, as social programs were cut and state companies privatized, undermining the social contract between citizens and the regimes. It has also made it more difficult for regime leaders to co-opt the military leadership that in Egypt and other countries has traditionally controlled large parts of the economy and increasingly faced domestic and foreign business competitors. Western economic and political liberalism, as described by Bader in Chapter 7, also made it harder for the autocratic regimes of the former Soviet Union to control their societies. Exposure to the West increased people’s expectations about the quality of governance of their own governments—expectations that the political leaders of post-Soviet countries were either unable or unwilling to meet. As a result, legitimacy declined. Moreover, being integrated into the global economy also made it more difficult to secure the loyalty of the business elite, as institutions of global finance offered them an exit option. Western economic, social, and political linkage was therefore a quasi-parameter that played an important role in maintaining a distinct balance of power, which in turn contributed to the stability of regime institutions. In recent years, autocratic leaders of post-Soviet countries have attempted to roll back Western influence. This shows that quasiparameters do not just change without regime leaders having no control over them. However, Western linkages are difficult and costly to disrupt, making it a quasi-parameter that cannot be manipulated easily. In sum, in autocratic regimes compliance with rules and norms should not be taken for granted. Institutions are always contested. Their survival depends on an intricate balance of power. While institutional dynamics maintain a preexisting balance of power, they cannot initially create these favorable circumstances at their birth. Moreover, exogenous shocks might disrupt the balance of power and, thus, undermine related institutions. For our book, however, more important are incremental changes in the domestic and international environment that accentuate preexisting institutional deficiencies. Changes in quasi-parameters thereby upset balances of power and in so doing undermine autocratic institutions.

224

Christoph H. Stefes

Anticipating and Preempting Crises Initial crises might hit nascent autocratic regimes unexpectedly. Yet as Dimitrov explains in Chapter 2, “an unanticipated crisis can serve as a catalyst for the decision to start monitoring latent discontent.” For autocratic regimes, it is crucial to learn and adapt—from crises in other autocratic regimes and, even more importantly, from their own experiences. Yet autocratic leaders need information to do so (see also Chapter 7 by Bader). In fact, the preemption of a crisis necessitates a close monitoring of masses and elites. It therefore is not surprising that autocracies maintain an extensive (and expensive) surveillance apparatus. Such an apparatus always entails some form of state security services. Yet Dimitrov also shows how citizens might voluntarily communicate their grievances with the regime through formal channels. While street protests are one way to do so, these protests are nuisances to the regime and often difficult to contain (see also Chapter 3 by Schedler). The regime therefore will attempt to contain collective expressions of grievances. Individual petitions and letters of complaint, on the other hand, are more manageable and provide equally useful insights into the public mood. Past and present autocratic regimes around the world therefore have invited citizens to submit their grievances via petitions (e.g., the GDR, Bulgaria, and China). Getting a sense of the elite’s stance toward the regime can be achieved by keeping its members close to the regime through various formal institutions such as cabinets, labor boards, and trade associations. In short, autocratic leaders need to set up institutions of surveillance and communication to monitor whether regime institutions are in peril or not and whether balances of power are slowly shifting due to incremental shifts in quasi-parameters. If balances of power and quasi-parameters shift, autocrats are not condemned to watch their regime’s institutions slowly disintegrate. They can take preemptive measures. One preemptive measure is to let street protests run their course and not respond with brutal repression, which often backfires. Yet autocratic regimes differ in the ways in which they react to public demonstrations. Some regimes let protests occur regularly, and some do not. Some are particularly nervous about antiregime protests while other regimes are equally repressive to antiregime protests and to demonstrations against particular policies (see Chapter 3 by Schedler). We know surprisingly little about when, how, and why autocratic regimes take preemptive measures against street protests. As Helle and Rakner show in Chapter 6, autocratic regimes might also manipulate electoral and party systems to preempt the rise of opposition forces within and outside of the regime. Uganda has moved from a no-party system to a multiparty system, but not due to massive pressure from below.

Crisis? What Crisis? Concluding Thoughts

225

Instead, reforming the party system has been one way for the Ugandan regime to marginalize nascent opposition groups within and outside of the ruling party which started off as an anti-party movement. Equally adaptive has been the Chinese Communist Party, as Thomas Heberer argues in Chapter 8. Building on strong state capacity, of which the ability to learn has been fundamental, the party state has frequently reformed its institutions to the processes of change. One of these reforms has entailed the strengthening of local state institutions to allow for policy modeling and experimentation under the close supervision of the communist leadership in Beijing. Heberer concludes that “effective local developmentalism contributes to effective policy implementation both in tackling crises and in generating stability and legitimacy in the political system.” Furthermore, as Heberer argues, the Chinese leadership has also shown considerable adaptability by switching back and forth between different governance modes (normal, transitional, and crisis). Rapid decisionmaking and a heavy-handed response to massive protests thereby characterize the crisis mode whereas the normal mode entails incremental decisionmaking processes and relative liberal policies. Bader’s study of autocratic regimes in the former Soviet Union in Chapter 7 demonstrates how leaders have recently attempted to roll back Western liberal influences in the regime. Close connections to the West and international organizations were sought during the 1990s as a way to legitimize the newly independent states. Yet following a wave of antiregime protests during the first half of the 2000s that were blamed on Western meddling, these regimes have created several new institutions of repression to deal with potential and actual opponents. They have also found new ways to legitimize those autocratic measures by referring to local customs and cultures (see also Colton 2016; Omelicheva 2015). It is noteworthy how they have learned from each other and engaged in bilateral and multilateral cooperation. China and the Economy Several contributors to this book assess the stability of China’s autocratic regimes. They conclude that China is an exceptionally stable regime. Even though public protests have increased rapidly in recent years (see Chapter 3 by Schedler), nobody contends that another Tiananmen Square is imminent. Svolik argues in Chapter 4 that the introduction of executive term limits has been exceptionally useful in keeping a favorable balance of power among the elites. It has allowed new elites to enter top leadership positions frequently. Executive term limits have thereby facilitated elite co-optation and stabilized the regime—just like they did in Mexico (Magaloni 2006). Dimitrov shows in Chapter 2 how the communist leadership responded to the

226

Christoph H. Stefes

unexpected 1989 uprising on Tiananmen Square by strengthening mechanisms that allow ordinary citizens to file complaints. And as mentioned above, Heberer argues in Chapter 8 that the central leadership has empowered local governments to react more flexibly to the particular needs of citizens in different parts of China. Moreover, the Chinese leadership has been successful in upholding the socialist social contract, which can be considered an informal institution of legitimation. It has maintained rising income levels across the population even during global financial upheavals in return for political acquiescence. Yet the last point reminds us that autocratic regimes rely heavily on economic resources to maintain their formal and informal institutions as well as underlying balances of power. Economic crises might lead to waves of discontent, as they weaken a regime’s legitimacy (see Chapter 3 by Schedler). Neoliberal reforms, which at least initially cause economic hardships for large parts of the population, are also detrimental to autocratic regimes’ attempts to bolster their legitimacy (see Chapter 10 by Volpi). Economic downturns further undermine a regime’s solicitation of feedback from its citizens whose willingness to express their grievances depends on reduced repression levels as well as the regime’s ability to respond and address grievances. Why would citizens complain if regimes were unresponsive? A regime’s responsiveness, however, critically hinges on the availability of economic resources (see Chapter 2 by Dimitrov). Likewise, a regime’s ability to co-opt elites through patronage networks depends on the availability of material awards, which are in scarce supply during economic downturns (see Chapter 6 by Helle and Rakner). In short, economic and material resources are often the glue that keeps formal and informal institutions of autocratic regimes intact. It therefore is not surprising that autocratic regimes are more vulnerable to prolonged economic downturns than are democratic regimes (Tanneberg, Stefes, and Merkel 2013). Venues for Further Research Previous studies of autocratic crises have focused on exogenous shocks that cause the rapid collapse of autocratic regimes. The contributors to this book do not deny that such external events can indeed cause a sudden regime collapse. Yet the empirical puzzle remains that some regimes are much better suited to deal with shock-like challenges than other regimes. In the end, these shocks just might tip the scale of already severely weakened regimes. This book, with its focus on incremental institutional decay, aims to understand the underlying sources of regime weakness.

Crisis? What Crisis? Concluding Thoughts

227

Quasi-parameters thereby serve as a central concept when analyzing slow-moving institutional disintegration processes. Yet what are quasiparameters, and what exactly do they do? There is a real danger that quasiparameters can become a fill-in to explain regime crises. Strictly speaking, quasi-parameters are endogenously changed—that is, by the institution under investigation—and do not directly condition behavior, which is what institutions do (Greif and Laitin 2004; see also Chapter 1 by Gerschweski). In this book, however, the concept has often been used more liberally for structural conditions that are seemingly constant, but in fact slowly change over time and thereby upset the specific balances of power that underlie the institutions of autocratic regimes such as the demographic changes in Uganda that weakened the support base of the ruling party (see Chapter 6 by Helle and Rakner). If we follow Svolik’s historical institutionalist argument in Chapter 4 that institutional stability ultimately hinges on institutions’ ability to maintain specific balances of power, it is nevertheless apt to apply the more liberal meaning of the concept. After all, these balances of power are configured by processes endogenous and exogenous to the institutions under investigation. Of course, a more liberal use of the concept invites an arbitrary selection of parameters. It therefore might be necessary to distinguish analytically between exogenous and endogenous quasi-parameters. It must also be specified how exactly endogenous quasi-parameters undermine the self-reinforcing processes of institutions that maintain specific balances of power. Since the introduction of the concept of quasi-parameters by Avner Greif and David D. Laitin in 2004, the concept has not gained much traction in the study of political institutions. More conceptual and empirical work is necessary to make it a more useful tool for the study of incremental institutional decay. Furthermore, the nature and role of actors inevitably deserves further consideration. If we indeed assume that institutions are inherently contested, it would be worthwhile to analyze who contests institutions, when they do it (and when their chances for success increase), and how they do it. In Chapter 10, Volpi takes a stab at highlighting the critical role of actors. Volpi reminds us that formerly neglected actors, such as smaller opposition groups, often accelerate crises. They do so by subverting institutions through reframing processes that break “the spell of institutional constraints” and, thereby, force institutional disintegration. Yet as Volpi suggests, these actors have to bide their time. Ultimately, their action bears fruit only if institutions are already weakened because their self-reinforcing dynamics have been upset. But could such actors also play a critical role in undermining these self-reinforcing dynamics? For instance, the history of nonviolent resistance shows us that opposition groups are able to engender

228

Christoph H. Stefes

the very circumstances under which they can sound the death knell for repressive regimes (Chenoweth and Stephan 2011b). Also, little work has been put forward that conceptualizes specific types of actors and their corresponding resources. In short, this book emphasizes explicitly the role of quasi-parameters and implicitly the role of actors to understand the incremental decay of institutions in autocratic regimes. Yet to advance the study of autocratic institutions, more conceptual and empirical work is needed to tell us what quasi-parameters exactly are and what they do. The same holds true for actors. This will be a fruitful exercise, for the study of autocratic regime crisis needs to go beyond the focus on sudden regime collapses that are seemingly caused by exogenous shocks. There is a reason that regimes disintegrate even without these shocks, and why some regimes weather them and some do not.

References

Abrahamsen, Rita, and Gerald Bareebe. 2016. “Uganda’s 2016 Elections: Not Even Faking It Anymore.” African Affairs 115, no. 461: 751−765. Acemoğlu, Daron, Georgy Egorov, and Konstantin Sonin. 2008. “Coalition Formation in Non-democracies.” Review of Economic Studies 75, no. 4: 987−1009. Acemoğlu, Daron, and James Robinson. 2006. Economic Origins of Dictatorship and Democracy. Cambridge: Cambridge University Press. ———. 2012. Why Nations Fail: The Origins of Power, Prosperity, and Poverty. New York: Crown. ACFIM (Alliance for Finance Monitoring). 2016. Monitoring Campaign Expenses During the Pre-election Period: Progress Report November−December 2015. Kampala: ACFIM. Africa Elections Database. 2011. Elections in Uganda. http://africanelections.tripod .com/ug.html. Afrobarometer. 2015. Summary of Results: Afrobarometer Round 6: Survey in Uganda. Afrobarometer. http://afrobarometer.org/publications/uganda-round-6 -summary-results-2015. Agüero, Felipe. 1998. “Legacies of Transitions: Institutionalization, the Military, and Democracy in South America.” Mershon International Studies Review 42, no. 2: 383-404. ———. 2002. “A Political Army in Chile: Historical Assessment and Prospects for the New Democracy,” In Political Armies: The Military and Nation Building in the Age of Democracy, eds. Kees Koonings and Dirk Kruijt. London: Zed Books, pp. 111-134. Ahlers, Anna L., Thomas Heberer, and Gunter Schubert. 2016. “Whither Local Governance in Contemporary China? Reconfiguration for More Effective Policy Implementation.” Journal of Chinese Governance 1, no. 1: 55–77. Albertus, Michael and Victor Menaldo. 2012. “Dictators as Founding Fathers? The Role of Constitutions Under Autocracy.” Economics & Politics 24, no. 3: 279306. Albrecht, Holger, and Dina Bishara. 2011. “Back on Horseback: The Military and Political Transformation in Egypt.” Middle East Law and Governance 3: 13−23. Albrecht, Holger, Aurel Croissant, and Fred Lawson, eds. 2016. Military Engagement in Mobilizing Societies: Armies and the Arab Spring. Philadelphia: University of Pennsylvania Press. Aldrich, John H. 1995. Why Parties? The Origin and Transformation of Party Politics in America. Chicago: University of Chicago Press.

229

230

References

Alexander, Gerard. 2002. The Sources of Democratic Consolidation. Ithaca: Cornell University Press. Aliaksandrau, Andrei, and Andrei Bastunets. 2014. “Belarus: Laws and Regulations Stifle Independent Media.” www.indexoncensorship.org/2014/02/belarus-legal -frameworks-regulations-stifle-new-competitors/. Allal, Amin, and Thomas Pierret, eds. 2013. Au Cœur des Révolutions Arabes: Devenir Révolutionnaires. Paris: Armand Colin. Allinson, Mark. 2009. “1977: The GDR’s Most Normal Year?” In Power and Society in the GDR, 1961−1979: The “Normalization of Rule”? ed. Mary Fulbrook. New York: Bergahn Books, pp. 253−279. Allison, Roy. 2008. “Virtual Regionalism, Regional Structures and Regime Security in Central Asia.” Central Asian Survey 27, no. 2: 185−202. al-Sayyid, Mustafa K. 2013. “What Went Wrong with Mubarak’s Regime.” In Egypt’s Tahrir Revolution, eds. Dan Tschirgi, Walid Kazziha, and Sean F. McMahon. Boulder: Lynne Rienner Publishers, pp. 11−28. Ambrosio, Thomas. 2007. “Insulating Russia from a Colour Revolution: How the Kremlin Resists Regional Democratic Trends.” Democratisation 14, no. 2: 232−252. ———. 2008. “Catching the ‘Shanghai Spirit’: How the Shanghai Cooperation Organization Promotes Authoritarian Norms in Central Asia.” Europe-Asia Studies 60, no. 8: 1321−1344. ———. 2009. Authoritarian Backlash: Russian Resistance to Democratization in the Former Soviet Union. Aldershot: Ashgate. ———. 2010. “Constructing a Framework of Authoritarian Diffusion: Concepts, Dynamics, and Future Research.” International Studies Perspectives 11, no. 4: 375−392. Andert, Reinhold, and Wolfgang Herzberg. 1990. Der Sturz: Erich Honecker im Kreuzverhör. Berlin: Aufbau-Verlag. Aoki, Masahiko. 2001. Toward a Comparative Institutional Analysis. Cambridge: MIT Press. Arceneaux, Craig L. 2001. Bounded Missions: Military Regimes and Democratization in the Southern Cone and Brazil. University Park: Pennsylvania State University Press. Arriagada, Genaro. 1985. La Política Militar de Pinochet. Santiago de Chile: Impresor Salesianos. ———. 1988. Pinochet: The Politics of Power. Boston: Allen & Unwin. Arriola, Leonardo R. 2009. “Patronage and Political Stability in Africa.” Comparative Political Studies 42, no. 10: 1339–1362. Aspinall, Edward. 2005. Opposing Suharto: Compromise, Resistance, and Regime Change in Indonesia. Stanford: Stanford University Press. Avagyan, Gagik, and Duncan Hiscock. 2005. Security Sector Reform in Armenia. London: Saferworld. Awortwi, Nicholas, and A. H. J. Helmsing. 2014. “In the Name of Bringing Services Closer to the People? Explaining the Creation of New Local Government Districts in Uganda.” International Review of Administrative Sciences 80, no. 4: 766−788. Babayan, Nelli. 2015. “The Return of the Empire? Russia’s Counteraction to Transatlantic Democracy Promotion in Its Near Abroad.” Democratization 22, no. 3: 438−458. Bader, Julia, Jörn Grävingholt, and Antje Kästner. 2010. “Would Autocracies Promote Autocracy? A Political Economy Perspective on Regime-type Export in Regional Neighbourhoods.” Contemporary Politics 16, no. 1: 81−100.

References

231

Bader, Max. 2014. “Democracy Promotion and Authoritarian Diffusion: The Foreign Origins of Post-Soviet Election Laws.” Europe-Asia Studies 66, no. 8: 1350−1370. Bai, Tiantian. 2012. “Survey Shows Most Support Steady Reform.” Global Times, November 7. Balmforth, Tom. 2012. “Moscow Court Designates Pussy Riot Video as Extremist.” Radio Free Europe/Radio Liberty. www.rferl.org/content/pussy-riot-video -extremist-russia/24784613.html. Banal, Laurent, and Vincenza Scherrer. 2008. “ONUB and the Importance of Local Ownership: The Case of Burundi.” In Security Sector Reform and UN Integrated Missions, eds. Heiner Hänggi and Vincenza Scherrer. Geneva: Geneva Centre for the Democratic Control of Armed Forces, pp. 29−66. Barany, Zoltan. 2011. “The Role of the Military.” Journal of Democracy 22, no. 4: 24−36. ———. 2012. The Soldier and the Changing State: Building Democratic Armies in Africa, Asia, Europe, and the Americas. Princeton: Princeton University Press. ———. 2016. How Armies Respond to Revolutions and Why. Princeton: Princeton University Press. Barma, Naazneen H., and Steven K. Vogel, eds. 2008. The Political Economy Reader: Markets as Institutions. New York: Routledge. Barros, Robert. 2002. Constitutionalism and Dictatorship: Pinochet, the Junta, and the 1980 Constitution. New York: Cambridge University Press. ———. 2003. “Dictatorship and the Rule of Law: Rules and Military Power in Pinochet’s Chile,” In Democracy and the Rule of Law, eds. Maravall, José María and Adam Przeworski. Cambridge: Cambridge University Press, pp. 188-219. Basedau, Matthias, and Sebastian Elischer. 2013. “Auf dem Rückzug in die Kasernen? Autoritäre Herrschaft und Militär im Subsaharischen Afrika.” In Autokratien im Vergleich, eds. Steffen Kailitz and Patrick Köllner. Baden-Baden: Nomos, pp. 354−383. Baturo, Alexander. 2014. Democracy, Dictatorship, and Term Limits. Ann Arbor: University of Michigan Press. Baum, Richard. 1997. “The Road to Tiananmen: Chinese Politics in the 1980s.” In The Politics of China: The Eras of Mao and Deng, ed. Roderick MacFarquhar. New York: Cambridge University Press, pp. 340−371. Beblawi, Hazem. 1990. “The Rentier State in the Arab World.” In The Arab State, ed. Giacomo Luciani. Berkeley: University of California Press, pp. 85−98. Beck, Thorsten, George Clarke, Alberto Groff, Philip Keefer, and Patrick Walsh. 2001. “New Tools in Comparative Political Economy: The Database of Political Institutions.” World Bank Economic Review 15, no. 1: 165−176. Bellin, Eva. 2012. “Reconsidering the Robustness of Authoritarianism in the Middle East: Lessons from the Arab Spring.” Comparative Politics 44, no. 2: 127−149. Bernhard, Michael. 2015. “Chronic Instability and the Limits of Path Dependence.” Perspectives on Politics 13, no. 4: 976−991. Besley, Timothy, and Anne Case. 1995. “Does Electoral Accountability Affect Economic Policy Choices? Evidence from Gubernatorial Term Limits.” Quarterly Journal of Economics 110, no. 3: 769−798. Betts, Paul. 2010. Within Walls: Private Life in the German Democratic Republic. New York: Oxford University Press. Biglaiser, Glen. 1999. “Military Regimes, Neoliberal Restructuring, and Economic Development: Reassessing the Chilean Case.” Studies in Comparative International Development 34, no. 1: 3-26.

232

References

Bispinck, Henrik. 2012. Die DDR im Blick der Stasi 1977: Die geheimen Berichte an die SED-Führung. Göttingen: Vandenhoeck & Ruprecht. Blanga, Yehuda. 2014. “Turmoil in Egypt 1968−2011: The Status of the Armed Forces in Citizen Uprisings in Egypt.” Contemporary Politics 20, no. 3: 365−382. Blaydes, Lisa. 2011. Elections and Distributive Politics in Mubarak’s Egypt. New York: Cambridge University Press. Bo, Zhiyue. 2007. China’s Elite Politics: Political Transition and Power Balancing. Hackensack, NJ: World Scientific. Bogaards, Matthijs. 2013. “Reexamining African Elections.” Journal of Democracy 24, no. 3: 151−160. Boin, Arjen, Paul ‘T Hart, and Allan McConnell. 2009. “Crisis Exploitation: Political and Policy Impacts of Framing Contests.” Journal of European Public Policy 16, no.1: 81–106. Boix, Carles. 2011. “Democracy, Development, and the International System.” American Political Science Review 105, no. 4: 809−828. Boix, Carles, and Milan W. Svolik. 2013. “The Foundations of Limited Authoritarian Government: Institutions and Power-sharing in Dictatorships.” Journal of Politics 75, no. 2: 300−316. Boone, Catherine. 2013. Property and Political Order in Africa. Cambridge: Cambridge University Press. Born, Hans, Marina Caprina, Karl Haltiner, and Jürgen Kuhlmannn, eds. 2006. Civil-military Relations in Europe: Learning from Crisis and International Change. New York: Routledge. Boshoff, Henri, Waldemar Vrey, and George Rautenbach. 2010. The Burundi Peace Process. From Civil War to Conditional Peace. Monograph 171. Pretoria: Institute for Security Studies. Boudreau, Vince. 2005. “Precarious Regimes and Matchup Problems in the Explanation of Repressive Policy.” In Repression and Mobilization, eds. Christian Davenport, Hank Johnston, and Carol Mueller. Minneapolis: University of Minnesotta Press, pp. 33–57. Boukhars, Anouar. 2011. Politics in Morocco: Executive Monarchy and Enlightened Authoritarianism. New York: Routledge. Bourdieu, Pierre. 1977. Outline of a Theory of Practice. Trans. Richard Nice. Cambridge: Cambridge University Press. Brancati, Dawn. 2014. “Democratic Authoritarianism: Origins and Effects.” Annual Review of Political Science 17, no. 1: 313−326. Brandenburg, Frank. 1964. The Making of Modern Mexico. Englewood Cliffs, NJ: Prentice Hall. Bremmer, Ian, and Samuel Charap. 2007. “The Siloviki in Putin’s Russia: Who They Are and What They Want.” Washington Quarterly 30, no. 1: 83−92. Brisset-Foucault, Florence. 2013. “Buganda Royalism and Political Competition in Uganda’s 2011 Elections.” Journal of Eastern African Studies 7, no. 3: 509−529. Brooker, Paul. 2008. “Authoritarian Regimes.” In Comparative Politics, ed. Daniele Caramani. Oxford: Oxford University Press, pp. 133−155. ———. 2009. Non-democratic Regimes. Basingstoke: Palgrave Macmillan. Brownlee, Jason. 2007. Authoritarianism in an Age of Democratization. Cambridge: Cambridge University Press. Bruneau, Thomas. 2005. “Civil-military Relations in Latin America: The Hedgehog and the Fox Revisited.” Revista Fuerzas Armadas y Sociedad Año 19, no. 1: 111−131. Buehler, Matt. 2013. “The Threat to “Un-moderate”: Moroccan Islamists and the Arab Spring.” Middle East Law and Governance 5: 231–257.

References

233

Bueno de Mesquita, Bruce, Alastair Smith, Randolph Siverson, and James D. Morrow. 2001. “Political Competition and Economic Growth.” Journal of Democracy 12, no. 1: 58−72. ———. 2003. The Logic of Political Survival. Cambridge: MIT Press. Bueno de Mesquita, Ethan. 2010. “Regime Change and Revolutionary Entrepreneurs.” American Political Science Review 104, no. 3: 446–466. Bunce, Valerie. 1999. Subversive Institutions: The Design and the Destruction of Socialism and the State. New York: Cambridge University Press. Bunce, Valerie. 2003. “Rethinking Recent Democratization: Lessons from the Postcommunist Experience.” World Politics 55, no. 2: 167-192. Bunce, Valerie, and Sharon L. Wolchick. 2010. “Defeating Dictators: Electoral Change and Stability in Competitive Authoritarian Regimes.” World Politics 62, no. 1: 43−86. ———. 2011. Defeating Authoritarian Leaders in Postcommunist Countries. Cambridge: Cambridge University Press. Burgat, François. 1997. The Islamic Movement in North Africa. Trans. W. Dowell. Austin: University of Texas Press. Calvert, Peter. 2005. “Changing Notion of Development: Bringing the State Back In.” In Palgrave Advances in Development Studies, ed. Jeffrey Haynes. Houndmills: Palgrave Macmillan. Camau, Michel, and Vincent Geisser. 2003. Le Syndrome autoritaire: Politique en Tunisie de Bourguiba à Ben Ali. Paris: Presses de Sciences Po. Cameron, David R., and Mitchell A. Orenstein. 2012. “Post-Soviet Authoritarianism: The Influence of Russia in Its ‘Near Abroad.’” Post-Soviet Affairs 28, no. 1: 1−44. Cammett, Melani Claire. 2007. “Business-government Relations and Industrial Change: The Politics of Upgrading in Morocco and Tunisia.” World Development 35, no. 11: 1889–1903. Capoccia, Giovanni. 2016. “When Do Institutions ‘Bite’? Historical Institutionalism and the Politics of Change.” Comparative Political Studies, 49, no. 8: 10951127. Capoccia, Giovanni, and Daniel R. Kelemen. 2007. “The Study of Critical Junctures: Theory, Narrative and Counterfactuals in Historical Institutionalism.” World Politics 59, no. 3: 341–369. Carbone, Giovanni. 2003. “Political Parties in a ‘No-party Democracy.’” Party Politics 9, no. 4: 485−501. ———. 2008. No-party Democracy? Ugandan Politics in Comparative Perspective. Boulder: Lynne Rienner Publishers. Carey, John M. 1996. Term Limits and Legislative Representation. Cambridge: Cambridge University Press. Cassani, Andrea. 2014. “Hybrid What? Partial Consensus and Persistent Divergences in the Analysis of Hybrid Regimes.” International Political Science Review 35, no. 5: 542-558. Catusse, Myriam. 2009. “Maroc: Un fragile état social dans la réforme néolibérale.” In L’Etat face aux “débordements” du social au Maghreb: Formation, travail et protection, eds. Myriam Catusse, Blandine Destremau, and Eric Verdier. Paris: Karthala, pp. 121−148. Cavallo, Ascanio, Manuel Salazar, and Oscar Sepúlveda. 1997. La Historia Oculta del Régimen Militar: Chile 1973-1988. Santiago de Chile: Grijalbo-Mondadori. Cesarini, Paola and Katherine Hite, eds. 2004. Authoritarian Legacies and Democracy in Latin America and Southern Europe. Notre Dame: University of Notre Dame Press.

234

References

Chams El-Dine, Cheerine. 2016. “Egypt: From Military Reform to Military Sanctuarization.” In Military Engagement in Mobilizing Societies, eds. Holger Albrecht, Aurel Croissant, and Fred Lawson. Philadelphia: University of Pennsylvania Press, pp. 185−202. Cheibub, José A. 2007. Presidentialism, Parliamentarism, and Democracy. Cambridge: Cambridge University Press. Cheibub, José A., Jennifer Gandhi, and James R. Vreeland. 2010. “Democracy and Dictatorship Revisited.” Public Choice 143, no. 1: 67−101. Chen, Xi. 2012. Social Protest and Contentious Authoritarianism in China. New York: Cambridge University Press. Chenoweth, Erica, and Maria J. Stephan. 2011a. “Why Civil Resistance Works: The Strategic Logic of Nonviolent Conflict.” International Security 33, no. 1: 7−44. ———. 2011b. Why Civil Resistance Works: The Strategic Logic of Nonviolent Conflict. New York: Columbia University Press. Churov, Vladimir. 2007. “Suverennaia demokratiia i vybory.” Rossiiskaia Gazeta, March 30. CIA (Central Intelligence Agency). 2016. The World Factbook. Washington, DC: CIA. Clancy-Smith, Julia. 2013. “From Sidi Bou Zid to Sidi Bou Said: A Longue Durée Approach to the Tunisian Revolutions.” In The Arab Spring: Change and Resistance in the Middle East, eds. Mark L. Haas and David W. Lesch. Boulder: Westview Press, Mobilereference e-book. Clapham, Christopher and George Philip. 1985. “The Political Dilemmas of Military Regimes,” In The Political Dilemmas of Military Regimes, eds. Christopher Clapham and George Philip. London: Croom Helm, pp. 1-26. Cleaves, Peter S. and Henry Pease Garcia. 1986. “State Autonomy and Military Policy Making,” In Armies and Politics in Latin America, eds. J. Samuel Fitch and Abraham F. Lowenthal, Rev. Ed. New York: Holmes & Meier, pp. 335-366. Clemens, Elisabeth S., and James M. Cook. 1999. “Politics and Institutionalisms: Explaining Durability and Change.” Annual Review of Sociology 25: 441–466. COG (Commonwealth Observer Group). 2011. Uganda Presidential and Parliamentary Elections 18 February 2011. Report of the Commonwealth Observer Group. London: Commonwealth Secretariat. Collier, Ruth B., and David Collier. 1991. Shaping the Political Arena: Critical Junctures, the Labor Movement, and Regime Dynamics in Latin America. Princeton: Princeton University Press. ———. 2011. “Kyrgyzstan’s Latest Revolution.” Journal of Democracy 22, no. 3: 150−164. Colton, Timothy J. 2016. Russia: What Everyone Needs to Know. New York: Oxford University Press. Conroy-Krutz, Jeffrey, and Carolyn Logan. 2012. “Museveni and the 2011 Ugandan Election: Did the Money Matter?” Journal of Modern African Studies 50, no. 4: 625−655. Constable, Pamela, and Arturo Valenzuela. 1993. A Nation of Enemies: Chile Under Pinochet. New York: Norton. Contreras, Carlos and Marcos Cueto. 2007. Historia del Perú Contemporáneo. 4th ed. Lima: Instituto de Estudios Peruanos. Cook, Linda J. 1993. The Soviet Social Contract and Why It Failed: Welfare Policy and Workers’ Politics from Brezhnev to Yeltsin. Cambridge: Harvard University Press. Cook, Linda J., and Martin K. Dimitrov. 2017. “The Socialist Social Contract Revisited: Evidence from Communist and State Capitalist Economies.” Europe-Asia Studies 69, no. 1: 8−26.

References

235

Cook, Steven A. 2007. Ruling but Not Governing: The Military and Political Development in Egypt, Algeria, and Turkey. Baltimore: Johns Hopkins University Press. Coppedge, Michael, John Gerring, Staffan I. Lindberg, Svend-Erik Skaaning, and Jan Teorell. 2016. “V-Dem Codebook v6.” Gothenburg: Varieties of Democracy (V-Dem) Project. Cotler, Julio. 1995. “Political Parties and the Problems of Democratic Consolidation in Peru.” In Building Democratic Institutions: Party Systems in Latin America, eds. Scott Mainwaring and Timothy R. Scully. Cambridge: Cambridge University Press, pp. 323-353. Cox, Gary. 1997. Making Votes Count: Strategic Coordination in the World’s Electoral Systems. New York: Cambridge University Press. Crawford, Beverly and Arend Lijphart, eds. 1997. In Liberalization and Leninist Legacies: Comparative Perspectives on Democratic Transitions. Berkeley: University of California Press. Croissant, Aurel. 2013a. “Militärregime im Autokratievergleich und in der Demokratisierungsforschung.” In Staatstätigkeiten, Parteien und Demokratie, ed. Klaus Armingeon. Wiesbaden: VS Verlag, pp. 407−430. ———. 2013b. “Coups and Post-Coup Politics in South-East Asia and the Pacific: Conceptual and Comparative Perspectives.” Australian Journal of International Affairs 67, no. 3: 264-280. Croissant, Aurel, Tanja Eschenauer, and Jil Kamerling. 2016. “Militaries’ Roles in Political Regimes: Introducing the PRM Data Set.” European Political Science. Online first. doi:10.1057/s41304-016-0083-6. Croissant, Aurel and Jil Kamerling. 2013. “Why Do Military Regimes Institutionalize? Constitution-making and Elections as Political Survival Strategy in Myanmar.” Asian Journal of Political Science 21, no. 2: 105-125. Croissant, Aurel, and David Kuehn. 2015. “The Military’s Role in Politics.” In The Routledge Handbook of Political Institutions, eds. Jennifer Gandhi and Rubén Ruiz. London: Routledge, pp. 258-277. Croissant, Aurel, David Kuehn, Paul W. Chambers, and Siegfried O. Wolf. 2010. “Beyond the Fallacy of Coup-ism: Conceptualizing Civilian Control of the Military in Emerging Democracies.” Democratization 17, no. 5: 950−975. Croissant, Aurel, David Kuehn, Philip Lorenz, and Paul W. Chambers. 2013. Civilian Control and Democracy in Asia. Basingstoke: Palgrave Macmillan. Croissant, Aurel, and Tobias Selge. 2016. “Should I Stay or Should I Go? Comparing Military (Non-) Cooperation During Authoritarian Regime Crises in the Arab World and Asia.” In Military Engagement in Mobilizing Societies, eds. Holger Albrecht, Aurel Croissant, and Fred Lawson. Philadelphia: University of Pennsylvania Press, pp. 97−124. Crotty, Jo, Sarah Marie Hall, and Sergej Ljubownikow. 2014. “Post-Soviet Civil Society Development in the Russian Federation: The Impact of the NGO Law.” Europe-Asia Studies 66, no. 8: 1253−1269. Dai, Zhiyong. 2014. “Yi gongmin canyu yu huajie jietou baoli” [Solving street violence by participation of citizens]. Nanfang Zhoumo [Southern Weekend], April 24. Daley, Patricia. 2007. “The Burundi Peace Negotiations: An African Experience of Peace-making.” Review of African Political Economy 34, no. 112: 335−352. Darchiashvili, David. 2005. “Georgian Defense Policy and Military Reform.” In Statehood and Security: Georgia After the Rose Revolution, eds. Bruno Coppieters and Robert Legvold. Cambridge: MIT Press, pp. 117−151.

236

References

Daucé, Françoise. 2014. “The Government and Human Rights Groups in Russia: Civilized Oppression?” Journal of Civil Society 10, no. 3: 239−254. Davenport, Christian. 2007. “State Repression and Political Order.” Annual Review of Political Science 10: 1–23. Debs, Alexandre. 2009. “Living by the Sword and Dying by the Sword? Leadership Transitions In and Out of Dictatorships.” Unpublished manuscript, Yale University. Decision on Major Issues Concerning Comprehensively Deepening Reforms. 2013. Abridged version: www.china.org.cn/china/third_plenary_session/2013-11/16 /content_30620736_4.htm; Chinese version: http://news.xinhuanet.com/politics /2013-11/15/c_118164235.htm. DEMGroup. 2011. Report on Money in Politics: Pervasive Vote Buying in Uganda Elections. Kampala: DEMGroup. Demmelhuber, Thomas. 2013. “Der Pharao, das Regime und der Staat: Regimeund Elitenwandel nach Mubarak.” In Revolution und Regimewandel in Ägypten, eds. Holger Albrecht and Thomas Demmelhuber. Baden-Baden: Nomos, pp. 43−62. Deng, Weizhi, Juezai Xu, and Yonglin Shen. 1998. Biange shehui zhong de zhengzhi wending [Political stability of changing societies]. Shanghai: Shanghai renmin chubanshe. Deng Xiaoping. 1993. “Zai Wuchang, Shenzhen, Zhuhai, Shanghai deng di de tanhua yaodian.” In Deng Xiaoping wenxuan. Tianjin: Tianjin renmin chubanshe, pp. 370−383. ———. 1994. “Excerpts from Talks Given in Wuchang, Shenzhen, Zhuhai and Shanghai (January 18−February 21, 1992).” In Selected Works of Deng Xiaoping. Vol. 3, (1982−1992). Beijing: Waiwen chubanshe, pp. 358−370. Denoeux, Guilain, and Abdeslam Maghraoui. 1998. “King Hassan’s Strategy of Political Dualism.” Middle East Policy 5, no. 4: 104–130. de Torrenté, Nicholas. 2013. “Understanding the 2011 Ugandan Elections: The Contribution of Public Opinion Surveys.” Journal of Eastern African Studies 7, no. 3: 530−548. DeVore, Howard O. 1999. China’s Intelligence and Internal Security Forces. Coulsdon: Jane’s Information Group. Diamond, Larry. 2002. “Thinking About Hybrid Regimes.” Journal of Democracy 13, no. 2: 21−35. Diao Jiecheng. 1996. Renmin xinfang shilüe, 1949−1995. Beijing: Beijing Jingji Xueyuan Chubanshe. Dikötter, Frank. 2010. Mao’s Great Famine: The History of China’s Most Devastating Catastrophe, 1958−1962. New York: Walker. DiMaggio, Paul J., and Walter W. Powell. 1983. “The Iron Cage Revisited: Institutional Isomorphism and Collective Rationality in Organizational Fields.” American Sociological Review 48, no. 2: 147–160. ———. 1991. “Introduction.” In The New Institutionalism in Organizational Analysis, eds. Walter W. Powell and Paul J. DiMaggio. Chicago: University of Chicago Press, pp. 1–40. Dimitrov, Martin K. 2014a. “What the Party Wanted to Know: Citizen Complaints as a ‘Barometer of Public Opinion’ in Communist Bulgaria.” East European Politics and Societies and Cultures 28, no. 2: 271−295. ———. 2014b. “Tracking Public Opinion Under Authoritarianism: The Case of the Soviet Union Under Brezhnev.” Russian History 41, no. 3: 329−353. ———. 2015. “Internal Government Assessments of the Quality of Governance in China.” Studies in Comparative International Development 50, no. 1: 50−72.

References

237

———. 2016. “1989 v diakhronna perspektiva: Kitai i Iztochna Evropa.” In Istoricheskoto Sravnenie—Metodi i Obekti: Kum Edna Sravnitelna Istoriia na Komunizma v Iztochna Evropa, ed. Ivailo Znepolski. Sofia: Maison des Sciences de l’Homme et de la Société, pp. 147−172. Dimitrov, Martin K., and Joseph Sassoon. 2014. “State Security, Information, and Repression: A Comparison of Communist Bulgaria and Ba’thist Iraq.” Journal of Cold War Studies 16, no. 2: 4−31. Dittmer, Lowell. 2002. “Chinese Factional Politics under Jiang.” In Gang Lin and Susan Shirk (eds.), The 16th CCP Congress and Leadership Transition in China. Washington, DC: The Woodrow Wilson Center Asia Program, pp. 20– 27. Dobry, Michel. 2009. “Critical Processes and Political Fluidity: A Theoretical Appraisal.” International Political Anthropology 2, no. 1: 74–90. Domínguez, Jorge I. 2002. “The Perfect Dictatorship? Comparing Authoritarian Rule in South Korea and in Argentina, Brazil, Chile, and Mexico.” Presented at the Annual Meeting of the American Political Science Association, Boston, August 29-September 1, 2002. Donno, Daniela. 2013. “Elections and Democratization in Authoritarian Regimes.” American Journal of Political Science 57, no. 3: 703−716. Droz-Vincent, Philippe. 2011. A Return of Armies to the Forefront of Arab Politics? Working Papers 11. Rome: Instituto Affari Internazionali. Easton, David. 1965. A System Analysis of Political Life. New York: Wiley. Edmunds, Timothy, Andrew Cottey, and Anthony Forster, eds. 2006. Civil-military Relations in Postcommunist Europe: Reviewing the Transition. London: Routledge. “Egypt Set for Mass Protest as Army Rules Out Force.” 2011. The Guardian, January 31. www.theguardian.com/world/2011/jan/31/egyptian-army-pledges-no -force. Engelmann, Roger. 2013. Die DDR im Blick der Stasi 1953: Die geheimen Berichte an die SED-Führung. Göttingen: Vandenhoeck & Ruprecht. Entelis, John P. 1986. Algeria: The Revolution Institutionalized. Boulder: Westview Press. Erdle, Steffen. 2004. “Tunisia: Economic Transformation and Political Restoration.” In Arab Elites: Negotiating the Politics of Change, ed. Volker Perthes. Boulder: Lynne Rienner, pp. 207−236. Ermakoff, Ivan. 2008. Ruling Oneself Out: A Theory of Collective Abdications. Durham: Duke University Press, Mobilereference e-book. EUEOM (European Union Election Observation Mission). 2011. Uganda: Final Report General Elections 18 February 2011. Brussels: EUEOM. ———. 2016. Uganda 2016: Preliminary Statement. European Union Election Observation Mission. Brussels: EUEOM. Evans, Peter. 1995. Embedded Autonomy: States and Industrial Transformation. Princeton: Princeton University Press. ———. 2004. “Development as Institutional Change: The Pitfalls of Monocropping and the Potentials of Deliberation.” Studies in Comparative International Development 38, no. 4: 30−56. Fawn, Rick. 2006. “Battle Over the Box: International Election Observation Missions, Political Competition and Retrenchment in the Post Soviet space.” International Affairs 82, no. 6: 1133−1153. Feaver, Peter. 2003. Armed Servants. Agency, Oversight, and Civil-military Relations. London: Harvard University Press.

238

References

Finer, Samuel E. 1962. Man on Horseback: The Role of the Military in Politics. Boulder: Westview Press. Fitch, J. Samuel. 1986. “Armies and Politics in Latin America: 1975-1985.” In Armies and Politics in Latin America, eds. Abraham F. Lowenthal and J. Samuel Fitch. 2nd. rev. ed. New York: Holmes & Meier, pp. 26-55. ———. 1998. The Armed Forces and Democracy in Latin America. Baltimore: Johns Hopkins University Press. Fluri, Filip, and Viorel Cibotaru, eds. 2008. Defense Institution Building: Country Profiles and Needs Assessments for Armenia, Azerbaidjan, Georgia and Moldova. Background Materials. Geneva: Geneva Center for the Democratic Control of Armed Forces. Fontana, Andres Miguel. 1987. “Political Decision Making by a Military Corporation: Argentina 1976–1983.” PhD diss., University of Texas at Austin. Foucault, Michel. 1991. “Governmentality.” Trans. Rosi Braidotti and revised by Colin Gordon. Burchell, Graham, Colin Gordon and Peter Miller eds. In The Foucault Effect: Studies in Governmentality. Chicago: University of Chicago Press, pp. 87−104. Francisco, Ronald A. 2005. “The Dictator’s Dilemma.” In Repression and Mobilization, eds. Christian Davenport, Hank Johnston, and Carol Mueller. Minneapolis: University of Minnesota Press, pp. 58–107. Frantz, Erica, and Natasha Ezrow. 2011. The Politics of Dictatorship: Institutions and Outcomes in Authoritarian Regimes. Boulder: Lynne Rienner. Frantz, Erica, and Elizabeth A. Stein. 2013. “The Benefits of Institutionalized Leadership Succession for Dictators’ Tenure.” Unpublished manuscript, Bridgewater State University and University of New Orleans. Freedom House. 2010. Uganda: Freedom of the Net.https://freedomhouse.org /report/freedom-net/2012/uganda. ———. 2013. “Nations in Transit 2013. Authoritarian Aggression and the Pressures of Austerity.” https://freedomhouse.org/report/nations-transit-2013/nations-transit -2013-authoritarian-aggression-and-pressures-austerity. ———. 2014. “Nations in Transit. Eurasia’s Rupture with Democracy.” https:// freedomhouse.org/report/nations-transit/nations-transit-2014. Friedman, Edward. 2008. “Why the Dominant Party in China Won’t Lose.” In Political Transitions in Dominant Party Systems: Learning to Lose, eds. Edward Friedman and Joseph Wong. New York: Routledge, pp. 252−268. Friedman, Edward, and Joseph Wong. 2008. “Learning to Lose: Dominant Parties, Dominant Party Systems, and Their Transitions.” In Political Transitions in Dominant Party Systems: Learning to Lose, eds. Edward Friedman and Joseph Wong. New York: Routledge, pp. 1−12. Friedrich, Carl J., and Zbigniew Brzezinski. 1956. Totalitarian Dictatorship and Autocracy. New York: Praeger. ———. 1965. Totalitarian Dictatorship and Autocracy. 2nd rev. ed. Cambridge: Harvard University Press. Friedrich, Walter, Peter Förster, and Kurt Starke, eds. 1999. Das Zentralinstitut für Jugendforschung Leipzig 1966-1990: Geschichte, Methoden, Erkenntnisse. Berlin: Edition Ost. Frost, Alexander. 2009. “The Collective Security Treaty Organization, the Shanghai Cooperation Organization, and Russia’s Strategic Goals in Central Asia.” China and Eurasia Forum Quarterly 7, no. 3: 83−102 Fukuyama, Francis. 2012. “The Patterns of History.” Journal of Democracy 23, no. 1: 14−26.

References

239

Gandhi, Jennifer. 2008. Political Institutions Under Dictatorship. New York: Cambridge University Press. Gandhi, Jennifer, and Ellen Lust-Okar. 2009. “Elections Under Authoritarianism.” Annual Review of Political Science 12: 403−422. Gandhi, Jennifer, and Adam Przeworski. 2007. “Authoritarian Institutions and the Survival of Autocrats.” Comparative Political Studies 40, no. 11: 1279−1301. Garnaut, Anthony. 2013. “Hard Facts and Half-truths: The New Archival History of China’s Great Famine.” China Information 27, no. 2: 223−246. Garretón, Manuel Antonio. 1983. El Proceso Político Chileno. Santiago de Chile: Facultad Latinoamericana de Ciencias Sociales (FLACSO). Gartner, Scott Sigmund, and Patrick M. Regan. 1996. “Threat and Repression: The Non-linear Relationship Between Government and Opposition Violence.” Journal of Peace Research 33, no. 3: 273–287. Gause, F. Gregory, III. 2011. “The Middle East Academic Community and the ‘Winter of Arab Discontent.’” In Seismic Shift—Understanding Change in the Middle East, ed. Ellen Laipson. Washington, DC: Henry L. Stimson Center, pp. 1−27. Geddes, Barbara. 1999. “What Do We Know About Democratization After Twenty Years?” Annual Review of Political Science 2, no. 1: 115−144. ———. 2003. Paradigms and Sand Castles: Theory Building and Research Design in Comparative Politics. Ann Arbor: University of Michigan Press. ———. 2004. “Minimum-winning Coalitions and Personalization in Authoritarian Regimes.” Paper presented at the meeting of the American Political Science Association, Chicago, September. 2004. ———. 2005. “Why Parties and Elections in Authoritarian Regimes.” Paper presented at the meeting of the American Political Science Association, Washington, DC. September 2005. ———. 2008. “Party Creation as an Autocratic Survival Strategy.” Unpublished manuscript, University of California at Los Angeles. ———. 2009. “How Autocrats Defend Themselves Against Armed Rivals.” Paper prepared for the American Political Science Association meeting, Toronto, September 2009. ———. 2014. “Autocratic Breakdown and Regime Transitions: A New Data Set.” Perspectives on Politics 12, no. 2: 313−331. Geertz, Clifford. 1971. Islam Observed: Religious Development in Morocco and Indonesia. Chicago: University of Chicago Press. Gehlbach, Scott G., and Philip Keefer. 2011. “Investment Without Democracy: Ruling-party Institutionalization and Credible Commitment in Autocracies.” Journal of Comparative Economics 39, no. 2: 123−139. Gehlbach, Scott G., Konstantin Sonin, and Milan W. Svolik. 2016. “Formal Models of Nondemocratic Politics.” Annual Review of Political Science 19: 565−584. Gerges, Fawaz A. 2000. “The End of the Islamist Insurgency in Egypt? Costs and Prospects.” Middle East Journal 54, no. 4: 592−612. Gerring, John. 2001. Social Science Methodology: A Criterial Framework. Cambridge: Cambridge University Press. Gerschewski, Johannes. 2013. “The Three Pillars of Stability: Legitimation, Repression, and Co-optation in Autocratic Regimes.” Democratization 20, no. 1: 13−38. Gerschewski, Johannes, Wolfgang Merkel, Alexander Schmotz, Christoph H. Stefes, and Dag Tanneberg. 2013. “Warum überleben Diktaturen?” In Autokratien im

240

References

Vergleich: PVS Sonderheft 47, eds. Steffen Kailitz and Patrick Köllner. BadenBaden: Nomos, pp. 106–131. Ghalioun, Burhan. 2004. “The Persistence of Arab Authoritarianism.” Journal of Democracy 15, no. 4: 126–132. Gieseke, Jens. 2006. Der Mielke-Konzern: Die Geschichte der Stasi, 1945−1990. Munich: Deutsche Verlags-Anstalt. Gilley, Bruce. 2008. “Legitimacy and Institutional Change. The Case of China.” Comparative Political Studies 41, no. 3: 259−284. ———. 2009. The Right to Rule: How States Win and Lose Legitimacy. New York: Columbia University Press. ———. 2015. “China Is Coming Closer to Ditching Its Corrupt Regime, and the US Can Help.” National Review. www.nationalreview.com/article/416163/china-coming -closer-ditching-its-corrupt-regime-and-us-can-help-bruce-gilley. March 30, 2015. Gilley, Bruce, and Heike Holbig. 2009. “The Debate on Party Legitimacy in China: A Mixed Quantitative/Qualitative Analysis.” Journal of Contemporary China 18, no. 59: 339−358. Ginsburg, Tom, and Tamir Moustafa. 2008. Rule By Law: The Politics of Courts in Authoritarian Regimes. New York: Cambridge University Press. Ginsburg, Tom, and Alberto Simpser, eds. 2014. Constitutions in Authoritarian Regimes. New York: Cambridge University Press. Gloppen, Siri., C. Atoo, E. Kasimbazi, A. Kibandama, J. Kiiza, S. Makara, G. Okiror, et al. (2006). “Uganda’s 2006 Presidential and Parliamentary Elections.” CMI Report. Bergen: Chr. Michelsen Institute. Göbel, Christian, and Thomas Heberer. Forthcoming. “The Policy Innovation Imperative: Changing Techniques for Governing China’s Local Governors.” In A Disciplined Dynamism: Ways of Governing and Being Governed in China Today, eds. Vivienne Shue and Patricia M. Thornton. Cambridge: Cambridge University Press. Goemans, Henk E, Kristian Skrede Gleditsch, and Giacomo Chiozza. 2009. “Introducing Archigos: A Datset of Political Leaders.” Journal of Peace Research 46, no. 2: 269-283. Goldstone, Jack A. 2001. “Toward a Fourth Generation of Revolutionary Theory.” Annual Review of Political Science 4: 139−187. Golooba-Mutebi, Frederick, and Sam Hickey. 2016. “The Master of Institutional Multiplicity? The Shifting Politics of Regime Survival, State-building and Democratization in Museveni’s Uganda.” Journal of Eastern African Studies 10, no. 4: 601−618. Goodfellow, Tom. 2014. “Legal Manoeuvres and Violence: Law Making, Protest and Semi-authoritarianism in Uganda.” Development and Change 45, no. 4: 753−776. Goodwin, Jeff. 2001. No Other Way Out: States and Revolutionary Movements, 1945–1991. Cambridge: Cambridge University Press. Gorenburg, Dmitry. 2014. “Countering Coor Revolution: Russia’s New Security Strategy and Its Implication for U.S. Policy.” Policy Memo 342. Washington: Program on New Approaches to Research and Security in Eurasia. Granovetter, Mark. 1978. “Threshold Models of Collective Behavior.” American Journal of Sociology 83, no. 6: 1420–1443. Green, Elliott. 2010. “Patronage, District Creation, and Reform in Uganda.” Studies in Comparative International Development 45, no. 1: 83−103. Greene, Kenneth F. 2007. Why Dominant Parties Lose: Mexico’s Democratization in Comparative Perspective. New York: Cambridge University Press.

References

241

———. 2010. “The Political Economy of Authoritarian Single-party Dominance.” Comparative Political Studies 43, no. 7: 807−834. Greif, Avner, and David D. Laitin. 2004. “A Theory of Endogenous Institutional Change.” American Political Science Review 98, no. 4: 633–652. Grönlund, Åke, and Wairagala Wakabi. 2015. “Citizens’ Use of New Media in Authoritarian Regimes: A case study of Uganda.” Electronic Journal of Information Systems in Developing Countries 67, no. 4: 1-23. Grzymala-Busse, Anna. 2008. “Beyond Clientelism: Incumbent State Capture and State Formation.” Comparative Political Studies 41, nos. 4−5: 638−673. ———. 2011. “Time Will Tell? Temporality and the Analysis of Causal Mechanisms and Processes.” Comparative Political Studies 44, no. 9: 1267−1297. Guo, Xuezhi. 2012. China’s Security State: Philosophy, Evolution, and Politics. New York: Cambridge University Press. Habermas, Jürgen. 1997. Legitimation Crisis. Cambridge: Polity Press. Hadenius, Axel, and Jan Teorell. 2007. “Pathways from Authoritarianism.” Journal of Democracy 18, no. 1: 143−156. Haggard, Stephan, and Robert R. Kaufman. 2016. “Democratization During the Third Wave.” Annual Review of Political Science 19: 125−144. Haimzadeh, Patrick. 2011. Au cœur de la Libye de Kadhafi. Paris: J.-C. Lattès. Hale, Henry E. 2006. “Democracy or Autocracy on the March? The Colored Revolutions as Normal Dynamics of Patronal Presidentialism.” Communist and PostCommunist Studies 39, no. 3: 305−329. ———. 2013. “Regime Change Cascades: What We Have Learned from the 1848 Revolutions to the 2011 Arab Uprisings.” Annual Review of Political Science 16: 331−353. Hall, Peter A., and Rosemary C. Taylor. 1996. “Political Science and the Three New Institutionalism.” Political Studies 44, no. 5: 936–957. Hammoudi, Abdellah. 1997. Master and Disciple: The Cultural Foundations of Moroccan Authoritarianism. Chicago: University of Chicago Press. Hansen, Flemming Splidsboel. 2008. “The Shanghai Co-operation Organisation.” Asian Affairs 39, no. 2: 217−232. Harkness, Kristen. 2017. “Military Loyalty and the Failure of Democratization in Africa: How Ethnic Armies Shape the Capacities of Presidents to Defy Term Limits.” Democratization 24, no. 5: 801-818. Hashim, Ahmed. 2011. “The Egyptian Military, Part Two: From Mubarak Onward.” Middle East Policy 13, no. 4: 106−128. He, Baogang, and Stig Thogersen. 2010. “Giving the People a Voice? Experiments with Consultative Authoritarian Institutions in China.” Journal of Contemporary China 19, no. 66: 675−692. He, Baogang, and Mark E. Warren. 2011. “Authoritarian Deliberation: The Deliberative Turn in Chinese Political Development.” Perspectives on Politics 9, no. 2: 269−289. Heberer, Thomas. 2012. “Creating Civil Society Structures Top-down?” In Civil Society and Governance in China, eds. Jianxing Yu and Sujian Guo. New York: Palgrave Macmillan, pp. 63−84. Heberer, Thomas, and Gunter Schubert. 2006. “Political Reform and Regime Legitimacy in Contemporary China.” Asien, no. 99: 9−28. ———. 2011. “Political Trust and the Case of China.” Oriental Archive. Journal of African and Asian Studies 79, no. 2: 189−202. ———. 2012. “County and Township Cadres as a Strategic Group: A New Approach to Political Agency in China’s Local State.” Journal of Chinese Political Science 17, no. 3: 221−249.

242

References

Heberer, Thomas, and Rene Trappel. 2013. “Evaluation Processes, Local Cadres’ Behaviour and Local Development Processes.” Journal of Contemporary China 22, no. 84: 1048−1066. Heilmann, Sebastian. 2002. Das Politische System der Volksrepublik China [The Political System of the People’s Republic of China]. Wiesbaden: Westdeutscher Verlag. ———. 2008. “From Local Experiments to National Policy.” China Journal, no. 59: 1−30. Heilmann, Sebastian, and Oliver Melton. 2013. “The Reinvention of Development Planning in China, 1993−2012.” Modern China 39, no. 6: 580−628. Hertle, Hans-Hermann, and Gerd-Rüdiger Stephan, eds. 2012. Das Ende der SED: Die letzten Tage des Zentralkomitees. Berlin: Ch. Links. Heydemann, Steven. 2007. Upgrading Authoritarianism in the Arab World. Analysis Paper 13. Washington, DC: Brookings Institution. Hibou, Béatrice. 2011. The Force of Obedience: The Political Economy of Repression in Tunisia. Cambridge: Polity Press. Hickey, Sam. 2013. “Beyond the Poverty Agenda? Insights from the New Politics of Development in Uganda.” World Development 43, 194-206. Hill, Fiona, and Kevin Jones. 2006. “Fear of Democracy or Revolution: The Reaction to Andijon.” Washington Quarterly 29, no. 3: 109−125. Hinnebusch, Raymond. 2006. “Authoritarian Persistence, Democratization Theory and the Middle East: An Overview and Critique.” Democratization 13, no. 3: 373−395. “Hosni Mubarak may step down.” 2011. Al Jazeera, February 10. www.aljazeera .com/news/middleeast/2011/02/20112101653445426.html. “Hosni Mubarak Resigns as President.” 2011. Al Jazeera, February 11. www.aljazeera .com/news/middleeast/2011/02/201121125158705862.html. Howard, Marc M., and Phillip G. Roessler. 2006. “Liberalizing Electoral Outcomes in Competitive Authoritarian Regimes.” American Journal of Political Science 50, no. 2: 365−381. Huang, Jing. 2008. “Institutionalization of Political Succession in China: Progress and Implications.” In China’s Changing Political Landscape, ed. Cheng Li. Washington, DC: Brookings Institution Press, pp. 80−97. Hudson, Michael C. 1977. Arab Politics: The Search for Legitimacy. New Haven: Yale University Press. Human Rights Watch. 2012. “Striking Oil, Striking Workers.” www.hrw.org/report /2012/09/10/striking-oil-striking-workers/violations-labor-rights-kazakhstans -oil-sector. ———. 2016. “’Keep the People Uninformed.’ Pre-Election Threats to Free Expression and Association in Uganda.” https://www.hrw.org/sites/default/files/report _pdf/uganda0116web.pdf. Huneeus, Carlos. 2000. El Régimen de Pinochet. Santiago de Chile: Editorial Sudamericana. Hunter, Wendy. 1997. Eroding Military Influence in Brazil: Politicians Against Soldiers. Chapel Hill: University of North Carolina Press. Huntington, Samuel. 1957. The Soldier and the State: The Theory and Politics of Civil-military Relations. New York: Vintage. ———. 1968. Political Order in Changing Societies. New Haven: Yale University Press. ———. 1991. The Third Wave: Democratization in the Late Twentieth Century. Norman: University of Oklahoma Press.

References

243

———. 1996. Political Order in Changing Societies. 2nd ed. New Haven: Yale University Press. Imas, Eugene. 2012. “Media Blame Game and the Zhanaozen Crisis in Kazakhstan.” Washington Review of Turkish and Eurasian Affairs, February. Immergut, Ellen M. 1998. “The Theoretical Core of the New Institutionalism.” Politics and Society 26, no. 1: 5–34. International Crisis Group. 1998. Burundi Under Siege: Lift the Sanctions; Relaunch the Peace Process. Burundi Report 1. New York: International Crisis Group. ———. 2001a. Burundi: Breaking the Deadlock. Africa Report 29. New York: International Crisis Group. ———. 2001b. Burundi: One Hundred Days to Put the Peace Process Back on Track. Africa Report 33. New York: International Crisis Group. ———. 2002. Burundi After Six Months of Transition: Continuing the War or Winning Peace. Africa Report 46. New York: International Crisis Group. ———. 2007. Egypt’s Sinai Question. Middle East/North Africa Report 61. New York: International Crisis Group. ———. 2008. “Armenia: Picking up the Pieces.” Crisis Group Europe Briefing 48. New York: International Crisis Group. International Federation for Human Rights. 2009. “Russian Society Under Control: Abuses in the Fight Against Terrorism and Extremism.” www.fidh.org/IMG/pdf /Russian_society_under_control.pdf. Izama, Angelo, and Michael Wilkerson. 2011. “Uganda: Museveni’s Triumph and Weakness.” Journal of Democracy 22, no. 3: 64−78. Jiang, Junyan, and Dali Yang. 2016. “Lying or Believing? Measuring Preference Falsification from a Political Purge in China.” Comparative Political Studies 49, no. 5: 600−634. Joestel, Frank. 2010. Die DDR im Blick der Stasi 1988: Die geheimen Berichte an die SED-Führung. Göttingen: Vandenhoeck & Ruprecht. Johnson, Chalmers. 1982. MITI and the Japanese Miracle: The Growth of Industrial Policy, 1925−1975. Stanford: Stanford University Press. ———. 1989. “Political Institutions and Economic Performance: The Governmentbusiness Relationship in Japan, South Korea, and Taiwan.” In The Political Economy of the New Asian Industrialism, ed. Frederic C. Deyo. Ithaca: Cornell University Press, pp. 136−164. Jones, Stephen. 2011. “Georgia’s “Rose Revolution of 2003: Enforcing Peaceful Change.” In Civil Resistance and Power Politics: The Experience of Non-violent Action from Gandhi to the Present, eds. Adam Roberts and Timothy Garton Ash. Oxford: Oxford University Press, pp. 317−334. Jowitt, Ken. 1992. New World Disorder: The Leninist Legacy. Berkeley: University of California Press. Judt, Matthias. 2013. Der Bereich Kommerzielle Koordinierung: Das DDR-Wirtschaftsimperium des Alexander Schalck-Golodkowski: Mythos und Realität. Berlin: Ch. Links Verlag. Kagan, Robert A. 2009. The Return of History and the End of Dreams. New York: Vintage Books. Kalyvas, Stathis N., and Laia Balcells. 2010. “International System and Technologies of Rebellion: How the Cold War Shaped Internal Conflict.” American Political Science Review 104, no. 3: 415−429. Kang, David C. 2002. Crony Capitalism: Corruption and Development in South Korea and the Philippines. Cambridge: Cambridge University Press.

244

References

Karumidze, Zurab, and James V. Wertsch. 2005. Enough! The Rose Revolution in the Republic of Georgia. New York: Nova Science. Kasfir, Nelson. 1998. “‘No-party Democracy’ in Uganda.” Journal of Democracy 9, no. 2: 67−81. Katznelson, Ira, and Barry R. Weingast. 2005. “Intersections Between Historical Institutionalism and Rational Choice Institutionalism.” In Preferences and Situations: Points of Intersection Between Historical and Rational Choice Institutionalism, eds. Ira Katznelson and Barry R. Weingast. New York: Sage, pp. 1–24. Kaya, Ruchan, and Michael Bernhard. 2013. “Are Elections Mechanisms of Authoritarian Stability or Democratization? Evidence from Postcommunist Eurasia.” Perspectives on Politics 11, no. 3: 734−752. Kendall-Taylor, Andrea, and Erica Frantz. 2014. “How Autocracies Fall.” Washington Quarterly 37, no. 1: 35−47. Kenney, Padraic. 2003. A Carnival of Revolution: Central Europe 1989. Princeton: Princeton University Press. Kepel, Gilles. 2002. Jihad: The Trail of Political Islam. Cambridge: Harvard University Press. Kjær, Anne Mette, and James Joughin. 2012. The reversal of agricultural reform in Uganda. Ownership and values.” Policy and Society 31, no. 4: 319–330. Knutsen, Carl Henrik, Jørgen Møller, and Svend-Erik Skaaning. 2016. “Going Historical: Measuring Democraticness Before the Age of Mass Democracy.” International Political Science Review 37, no. 5: 679−689. Kohli, Atul. 2004. State-directed Development: Political Power and Industrialization in the Global Periphery. Cambridge: Cambridge University Press. Koselleck, Reinhart. 1982. “Krise.” In Geschichtliche Grundbegriffe: Historisches Lexikon zur politisch-sozialen Sprache in Deutschland. Vol. 3, eds. Otto Brunner, Werner Conze and Reinhart Koselleck. Stuttgart: Klett-Cotta, pp. 617–650. ———. 2004. “Krise.” In Geschichtliche Grundbegriffe: Historisches Lexikon zur politisch-sozialen Sprache in Deutschland, eds. Otto Brunner, Werner Conze, and Reinhart Koselleck. Stuttgart: Klett-Cotta, pp. 617–650. Krasner, Stephen D. 1984. “Approaches to the State: Alternative Conceptions and Historical Dynamics.” Comparative Politics 16, no. 2: 223–246. Krauze, Enrique. 1997. Biography of Power: A History of Modern Mexico, 1810−1996. New York: HarperCollins. Kricheli, Ruth, Yair Livne, and Beatriz Magaloni. 2010. “Taking to the Streets: Theory and Evidence on Protests Under Authoritarianism.” Unpublished manuscript, Stanford University. Kruijt, Dirk and María del Pilar Tello. 2002. “From Military Reformist to Civilian Dictatorship: Peruvian Military Politics from the 1960s to the Present.” In Political Armies: The Military and Nation Building in the Age of Democracy, eds. Kees Koonings and Dirk Kruijt, London: Zed Books, pp. 35-63. Kubicek, Paul. 2009. “The Commonwealth of Independent States: An Example of Failed Regionalism?” Review of International Studies 35, no. 1: 237−256. Kuran, Timur. 1991. “Now Out of Never: The Element of Surprise in the East European Revolution of 1989.” World Politics 44, no. 1: 7−48. ———. 1995. Private Truths, Public Lies: The Social Consequences of Preference Falsification. Cambridge: Harvard University Press. Kurzman, Charles. 2004a. The Unthinkable Revolution in Iran. Cambridge: Harvard University Press. ———. 2004b. “Can Understanding Undermine Explanation? The Confused Experience of Revolution.” Philosophy of the Social Sciences 34, no. 3: 328−351.

References

245

———. 2012. “The Arab Spring Uncoiled.” Mobilization 17, no. 4: 377−390. Kuznetsov, Dmitrii. 2010. “Chto sluchilos’ v Belarusi.” http://samlib.ru/k/kuznecow _d_j/olbi.shtml. Lambright, Gina S. 2014. “Opposition Politics and Urban Service Delivery in Kampala, Uganda.” Development Policy Review 32, no. 1: 39−60. Lande, Carl H. 1986. “The Political Crisis.” In Crisis in the Philippines: The Marcos Era and Beyond, ed. John Bresnan. Princeton: Princeton University Press, pp. 114–144. Laver, Michael, and Norman Schofield. 1990. Multiparty Government: The Politics of Coalition in Europe. Oxford: Oxford University Press. Lebedev, S. N. 2013. “Tezisy vystupleniya Predsedatelya Ispol’nitelnogo komiteta Ispolnitel’nogo sekretarya SNG S. N. Lebedeva na zasedanii Postoyannogo Soveta OBSE g. Vena 27 iyunya 2013 goda.” http://ecis.info/page.php?id=23516. Lee, Terence. 2015. Defect or Defend: Military Responses to Popular Protests in Authoritarian Asia. Baltimore: Johns Hopkins University Press. Leftwich, Adrian. 1993. “Governance, Democracy and Development in the Third World.” Third World Quarterly 14, no. 3: 605−624. ———. 2000. States of Development: On the Primacy of Politics in Development. Cambridge: Polity Press. Lesch, Anne M. 2011. “Egypt’s Spring: Causes of the Revolution.” Middle East Policy 18, no. 3: 35−48. Levitsky, Steven, and Lucan A. Way. 2006. “Linkage Versus Leverage: Rethinking the International Dimension of Regime Change.” Comparative Politics 38: 379−400. ———. 2010. Competitive Authoritarianism: Hybrid Regimes After the Cold War. Cambridge: Cambridge University Press. Li, Cheng. 2010. “China’s Communist Party-state: The Structure and Dynamics of Power.” In Politics in China: An Introduction, ed. William A. Joseph. New York: Oxford University Press, pp. 165−191. Li, Cheng. 2016. Chinese Politics in the Xi Jinping Era. Washington, DC: Brookings Institution Press. Li, Tania Murray. 2007. The Will to Improve: Governmentality, Development, and the Practice of Politics. Durham: Duke University Press. Li Peng. 2010. Li Peng liusi riji. El Monte, CA: Xidian Chubanshe. Lichbach, Mark Irving. 1987. “Deterrence or Escalation? The Puzzle of Aggregate Studies of Repression and Dissent.” Journal of Conflict Resolution 31, no. 2: 266–297. Liddell, James. 2010. “Notables, Clientelism and the Politics of Change in Morocco.” Journal of North African Studies 15: 315−331. Lindberg, Staffan I. 2009. “The Power of Elections in Africa Revisited.” In Democratization by Elections: A New Mode of Transition, ed. Staffan I. Lindberg. Baltimore: Johns Hopkins University Press, pp. 25−47. Linz, Juan J. 1978. “Crisis, Breakdown, and Reequilibration.” In The Breakdown of Democratic Regimes, eds. Juan J. Linz and Alfred Stepan. Baltimore: Johns Hopkins University Press. Lohmann, Susanne. 1993. “A Signaling Model of Informative and Manipulative Political Action.” American Political Science Review 87, no. 2: 319–333. ———. 1994. “The Dynamics of Informational Cascades: The Monday Demonstrations in Leipzig, East Germany, 1989–91.” World Politics 47, no. 1: 42–101. Lorentzen, Peter L. 2013. “Regularizing Rioting: Permitting Public Protest in an Authoritarian Regime.” Quarterly Journal of Political Science 8: 127–158.

246

References

Lovell, John P., and C. I. Eugene Kim. 1967. “The Military and Political Change in Asia.” Pacific Affairs 40, nos. 1−2: 113−123. Loveman, Brian. 1999. For la Patria: Politics and the Armed Forces in Latin America. Wilmington: Scholarly Resources. ———. 2001. “Historical Foundations of Civil-Military Relations in Spanish America.” In Civil-Military Relations in Latin America, ed. David Pion-Berlin, Chapel Hill: The University of North Carolina Press, pp. 246-274. Lozovik, N. I. 2011. “O kriteriiakh otsenki vyborov missiami mezhdunarodnykh nabliudatelei SNG i OBSE.” http://www.rec.gov.by/sites/default/files/pdf/news /15-11-2011.pdf. Lust, Ellen. 2005. Structuring Conflict in the Arab World: Incumbents, Opponents, and Institutions. Cambridge: Cambridge University Press. ———. 2009. “Competitive Clientelism in the Middle East.” Journal of Democracy 20, no. 3: 122−135. Lutscher, Philip M. 2016. “The More Fragmented the Better?—The Impact of Armed Forces Structure on Defection During Nonviolent Popular Uprisings.” International Interactions 42, no. 2: 350−375. Lutterbeck, Derek. 2013. “Arab Uprisings, Armed Forces, and Civil-military Relations.” Armed Forces and Society 39, no. 1: 28−52. Lynch, Marc. 2014. “Introduction.” In The Arab Uprisings Explained: New Contentious Politics in the Middle East, ed. Marc Lynch. New York: Columbia University Press. Lyons, Terence. 2016. “From Victorious Rebels to Strong Authoritarian Parties: Prospects for Post-war Democratization.” Democratization 23, no. 6: 1026−1041. MacFarquhar, Roderick. 1974. The Origins of the Cultural Revolution. New York: Columbia University Press, 1997. ———. 1997a. The Politics of China: The Eras of Mao and Deng. New York: Cambridge University Press. ———. 1997b. “The Succession to Mao and the End of Maoism, 1969–82.” In Politics of China 1949–1989: Eras of Mao and Deng, ed. Roderick MacFarquhar. New York: Cambridge University Press, pp. 248–339. MacFarquhar, Roderick, and Michael Schoenhals. 2006. Mao’s Last Revolution. Cambridge: Harvard University Press. Magaloni, Beatriz. 2006. Voting for Autocracy: Hegemonic Party Survival and Its Demise in Mexico. Cambridge: Cambridge University Press. ———. 2008. “Credible Power-sharing and the Longevity of Authoritarian Rule.” Comparative Political Studies 41, nos. 4−5: 715–741. Magaloni, Beatriz, and Ruth Kricheli. 2010. “Political Order and One-party Rule.” Annual Review of Political Science 13: 123−143. Magaloni, Beatriz, Eric Min, and Jonathan Chu. 2013. Autocracies of the World Dataset, 1950−2012. Version 1.0. Stanford: Stanford University. Mahoney, James. 2000. “Path Dependence in Historical Sociology.” Theory and Society 29: 507–548. Mahoney, James. 2001. “Path Dependent Explanations of Regime Change: Central America in Comparative Perspective.” Studies in Comparative International Development 36, no. 1: 111-141. Mahoney, James, and Kathleen Thelen. 2010. “A Theory of Gradual Institutional Change.” In Explaining Institutional Change. Ambiguity, Agency, and Power, eds. James Mahoney and Kathleen Thelen. New York: Cambridge University Press, pp. 1–37.

References

247

Mainwaring, Scott P. 1995. “Brazil: Weak Parties, Feckless Democracy.” In Building Democratic Institutions: Party Systems in Latin America, eds. Scott Mainwaring and Timothy R. Scully. Stanford: Stanford University Press, pp. 354–398. Mainwaring, Scott P. and Timothy R. Scully. 1995. “Introduction: Party Systems in Latin America,” In Building Democratic Institutions: Party Systems in Latin America, eds. Scott Mainwaring and Timothy R. Scully. Stanford: Stanford University Press, pp. 1–34. Makara, Sabiti. 2010. “Deepening Democracy Through Multipartyism: The Bumpy Road to Uganda’s 2011 Elections.” Africa Spectrum 45, no. 2: 81−94. Makara, Sabiti, Lise Rakner, and Lars Svåsand. 2008. “Turnaround: The National Resistance Movement and the Reintroduction of a Multiparty System in Uganda.” In Electoral Democracy in Uganda: Understanding Institutional Processes and Outcomes of the 2006 Multiparty Elections, eds. Julius Kiiza, Sabiti Makara, and Lise Rakner. Kampala: Fountain, pp. 262−286. ———. 2009. “Turnaround: The National Resistance Movement and the Reintroduction of a Multiparty System in Uganda.” International Political Science Review 30, vol. 2: 185−204. Malesky, Edmund, and Paul Schuler. 2010. “Nodding or Needling: Analyzing Delegate Responsiveness in an Authoritarian Parliament.” American Political Science Review 104, no. 3: 482−502. Manion, Melanie. 1992. “Politics and Policy in Post-Mao Cadre Retirement.” China Quarterly, no. 129: 1−25. ———. 1993. Retirement of Revolutionaries in China: Public Policies, Social Norms, Private Interests. Princeton: Princeton University Press. Mann, Michael. 1993. The Sources of Social Power. Vol. 2. The Rise of Classes and Nation-states. Cambridge: Cambridge University Press. Marat, Erica. 2009. “Kazakhstan’s Military Reform and OSCE Chairmanship: Regional Implications.” Security and Human Rights 20, no. 1: 83−87. Mares, David, and Rafael Martinéz. 2014. Debating Civil-military Relations in Latin America. Eastbourne: Sussex Academic Press. Martinez, Luis. 2000. The Algerian Civil War 1990−1998. Trans. J. Derrick. New York: Columbia University Press. Marxen, Klaus, and Gerhard Werle. 2000. Strafjustiz und DDR-Unrecht. Vol. 5, pt. 2. Berlin: De Gruyter. Mattis, Peter. 2011. “Assessing the Foreign Policy Influence of the Ministry of State Security.” China Brief 11, no. 1. https://jamestown.org/program/assessing-the -foreign-policy-influence-of-the-ministry-of-state-security. McDermott, Roger N. 2007. The Rising Dragon: SCO Peace Mission 2007. Occasional Paper. Jamestown Foundation. https://jamestown.org/report/the-risingdragon-sco-peace-mission-2007. McFaul, Michael. 2005. “Transitions from Postcommunism.” Journal of Democracy 16, no. 3: 5−19. McGregor, Richard. 2010. The Party: The Secret World of China’s Communist Rulers. New York: Harper. McGuire, James W. 1995. “Political Parties and Democracy in Argentina.” In Building Democratic Institutions: Party Systems in Latin America, eds. Scott Mainwaring and Timothy R. Scully. Stanford: Stanford University Press, pp. 200–246. Melnykovska, Inna, Hedwig Plamper, and Rainer Schweickert. 2012. “Do Russia and China Promote Autocracy in Central Asia?” Asia Europe Journal 10, no. 1: 75−89.

248

References

Merkel, Wolfgang. 2010. Systemtransformation: Einführung in die Theorie und Empirie der Transformationsforschung [Systemtransformation: Introduction to the Theory and Empiricism of Transformation Research]. 2nd rev. and extended ed. Wiesbaden: VS Verlag. Meyer, Marshall W., and Lynne G. Zucker. 1989. Permanently Failing Organizations. Newbury Park, CA: Sage. Meyers, Genevieve F. 2014. “Decentralization in Uganda: Towards Democratic Local Governance or Political Expediency.” In Challenges to Democratic Governance in Democratic Countries, eds. Gedeon Mudacumura and Göktug Morcöl. Switzerland: Springer, pp. 95-110. Mietzner, Marcus. 2011. “Overcoming Path Dependence: The Quality of Civilian Control of the Military in Post-authoritarian Indonesia.” Asian Journal of Political Science 19, no. 3: 270−289. Migdal, Joel. 1988. Strong Societies and Weak States: State-society Relations and State Capabilities in the Third World. Princeton: Princeton University Press. Mikaberidze, Alexander. 2015. Historical Dictionary of Georgia. Lanham, MD: Rowman & Littlefield. Miller, Alice L. 2004. “Hu Jintao and the Party Politburo.” China Leadership Monitor, no. 9. http://media.hoover.org/sites/default/files/documents/clm9_lm .pdf. ———. 2008. “Institutionalization and the Changing Dynamics of Chinese Leadership Politics.” In China’s Changing Political Landscape, ed. Cheng Li. Washington, DC: Brookings Institution Press, pp. 61−79. ———. 2011. “The Politburo Standing Committee Under Hu Jintao.” China Leadership Monitor, no. 35: 1–9. Miller, Michael K. 2015. “Elections, Information, and Policy Responsiveness in Autocratic Regimes.” Comparative Political Studies 48, no. 6: 691−727. Milstein, Brian. 2015. “Thinking Politically About Crisis: A Pragmatist Perspective.” European Journal of Political Theory 14, no. 2: 141–160. Mirzobekova, Ramziya. 2013. “Khorog, 2012. God spustia.” Asia Plus. http://news .tj/ru/node/162907. Mitchell, Lincoln. 2004. “Georgia’s Rose Revolution.” Current History 103: 342−348. ———. 2008. Uncertain Democracy: U.S. Foreign Policy and Georgia’s Rose Revolution. Philadelphia: University of Pennsylvania Press. Mitter, Armin. 1991. “Die Ereignisse im Juni und Juli 1953 in der DDR.” Aus Politik und Zeitgeschichte, no. 5: 31−41. Mitter, Armin, and Stefan Wolle. 1990. Ich liebe euch doch alle! Befehle und Lageberichte des MfS, January−November 1989. Berlin: Basis Druck. Moore, Will H. 1998. “Repression and Dissent: Substitution, Context, and Timing.” American Journal of Political Science 42, no. 3: 851−873. Morgenbesser, Lee. 2014. “Elections in Hybrid Regimes: Conceptual Stretching Revived.” Political Studies 62, no. 1: 21−36. Morse, Yonatan L. 2012. “The Era of Electoral Authoritarianism.” World Politics 64, no. 1: 161−198. Moustafa, Tamir. 2009. The Struggle for Constitutional Power: Law, Politics, and Economic Development in Egypt. Cambridge: Cambridge University Press. Mühlberg, Felix. 2004. Bürger, Bitten und Behörde: Geschichte der Eingabe in der DDR. Berlin: Karl Dietz Verlag. Muhumuza, William. 2009. “From Fundamental Change to No Change: The NRM and Democratization in Uganda.” Le Cahiers 41, no. 2: 21−42.

References

249

Mulvenon, James. 2003a. “The Crucible of Tragedy: SARS, the Ming 361 Accident, and Chinese Party-army Relations.” China Leadership Monitor, no. 8: 1–12. ———. 2003b. “The PLA and the 16th Party Congress: Jiang Controls the Gun?” China Leadership Monitor, no. 5: 20–29. ———. 2003c. “Reduced Budgets, the ‘Two Centers,’ and Other Mysteries of the 2003 National People’s Congress.” China Leadership Monitor, no. 7: 1–8. ———. 2005. “The King Is Dead! Long Live the King! The CMC Leadership Transition from Jiang to Hu.” China Leadership Monitor, no. 13: 1–8. Münkel, Daniela. 2009. “Die DDR im Blick der Stasi 1989.” Aus Politik und Zeitgeschichte, nos. 21−22: 26−32. Murison, Jude. 2013. “Judicial Politics: Election Petitions and Electoral Fraud in Uganda.” Journal of Eastern African Studies 7, no. 3: 492−508. Museveni, Yoweri K. 2000. “Ours is Fundamental Change.” In What is Africa’s Problem?, eds. Museveni Yoweri K. and Elizabeth Kanyogonya. Minneapolis: University of Minnesota Press, pp. 3-10. Nabaho, Lazarus, and Alfred Kiiza. 2013. “Human Resource Management in Local Governments: An Analysis of Recruitment and Selection Practices in Uganda.” Journal of African and Asian Local Government Studies 2, no. 2: 1−14. Nathan, Andrew J. 2003. “Authoritarian Resilience.” Journal of Democracy 14, no. 1: 6–17. NEC (NRM National Executive Committee). 2002. Report of the Ad Hoc Committee Set Up by the 6th Meeting of the National Executive Committee. Negretto, Gabriel L. 2014. “Authoritarian Constitution-Making: The Role of the Military in Latin America,” In Constitutions in Authoritarian Regimes, eds. Tom Ginsburg and Alberto Simpser. Cambridge: Cambridge University Press, pp. 83–110. Nepstad, Sharon Erickson. 2011. Nonviolent Revolutions: Civil Resistance in the Late Twentieth Century. New York: Oxford University Press. ———. 2013. “Mutiny and Nonviolence in the Arab Spring: Exploring Military Defections and Loyalty in Egypt, Bahrain, and Syria.” Journal of Peace Research 50, no. 3: 337−349. Noesselt, Nele. 2014. “Microblogs and the Adaptation of the Chinese Party-state’s Governance Strategy.” Governance 27, no. 3: 449−468. Nordlinger, Eric. 1977. Soldiers in Politics: Military Coups and Governments. Upper Saddle River, NJ: Prentice Hall. North, Douglass C. 1990. Institutions, Institutional Change and Economic Performance. New York: Cambridge University Press. Nunn, Frederick M. 1975. “Effects of European Military Training in Latin America: The Origins and Nature of Professional Militarism in Argentina, Brazil, Chile, and Peru, 1890-1940.” Military Affairs 39, no. 1: 1-7. ———. 2001. “Foreign Influences on the South American Military: Professionalization and Politicization.” In The Soldier and the State in South America: Essays in Civil-Military Relations, ed. Patricio Silva. Basingstoke: Palgrave, pp. 13-37. O’Brien, Kevin J. 1996. “Rightful Resistance.” World Politics 49, no. 1: 31–55. O’Donnell, Guillermo. 1978. “Reflections on the Patterns of Change in the Bureaucratic-Authoritarian State.” Latin American Research Review 13, no. 1: 3–38. ———. 1986.” Modernization and Military Coups: Theory, Comparisons, and the Argentine Case.” In Armies and Politics in Latin America, eds. Abraham F.

250

References

Lowenthal and J. Samuel Fitch. 2nd revised Ed. New York: Holmes & Meier, pp. 96-133. O’Donnell, Guillermo, and Philippe C. Schmitter. 1986. Transitions from Authoritarian Rule: Tentative Conclusions About Uncertain Democracies. Baltimore: Johns Hopkins University Press. Ohno, Kenichi. 2013. “The East Asian Growth Regime and Political Development.” In Eastern and Western Ideas for African Growth, Diversity and Complementarity in Development Aid, eds. K. Ohno and Izumi Ohno. New York: Routledge, pp. 30−52. Oloka-Onyango, Joseph. 2000. “New Wine or New Bottles? Movement Politics and One-partyism in Uganda.” In No-party Democracy in Uganda: Myths and Realities, eds. Justus Mugaju and Joseph Oloka-Onyango. Kampala: Fountain, pp. 40−59. Olson, Mancur. 1965. The Logic of Collective Action. Cambridge: Harvard University Press. Omelicheva, Maria Y. 2009. “Russia’s Counterterrorism Policy: Variations on an Imperial Theme.” Perspectives on Terrorism 3, no. 1. http://www.terrorismanalysts.com /pt/index.php/pot/article/view/61/html. ———. 2015. Democracy in Central Asia? Competing Perspectives and Alternate Strategies. Lexington: University of Kentucky Press. Ong, Lynette H. 2012. “Between Development and Clientelist States: Local Business Relationships in China.” Comparative Politics 44, no. 2: 191−205. OSCE/ODIHR (Organization for Security and Co-operation in Europe/Office for Democratic Institutions and Human Rights). 2004. “Comments on the Draft Laws of the Kazakh Republic ‘On Counteractive Measures Against Extremist Activities’ and ‘On Amendments to Several Legislative Acts with Regard to Counteractive Measures Against Extremist Activities.’” Warsaw, October 20. Ostermann, Christian F., ed. 2001. Uprising in East Germany, 1953: The Cold War, the German Question, and the First Major Upheaval Behind the Iron Curtain. New York: Central European University Press. Ouédrago, Emile. 2014. Advancing Military Professionalism in Africa. Washington, DC: African Center for Strategic Studies. Őzdal, Habibe. 2011. “Putting the CSTO to the Test in Kyrgyzstan.” In USAK (Uluslararası Stratejik Araştırmalar Kurumu) Yearbook 4: 363-366 Pargeter, Alison. 2006. “Libya: Reforming the Impossible?” Review of African Political Economy 33, no. 108: 219−235. Pearlman, Wendy. 2013. “Emotions and the Microfoundations of the Arab Uprisings.” Perspectives on Politics 11, no. 2: 387−409. Pepinsky, Thomas. 2014. “The Institutional Turn in Comparative Authoritarianism.” British Journal of Political Science 44, no. 3: 631−653. Pérez-Liñán, Anibal, and John Polga-Hecimovich. 2016. “Explaining Military Coups and Impeachments in Latin America.” Democratization 24, no. 5: 839–858. Perrot, Sandrine, Sabiti Makara, Jerôme Lafargue, and Marie-Aude Fouéré, eds. 2014. Elections in a Hybrid Regime: Revisiting the 2011 Ugandan Polls. Kampala: Fountain. Perthes, Volker, ed. 2004. Arab Elites: Negotiating the Politics of Change. Boulder: Lynne Rienner. Philip, George. 1978. The Rise and Fall of the Peruvian Military Radicals 19681976. London: The Athlone Press. Pierson, Paul. 2004. Politics in Time: History, Institutions, and Social Analysis. Princeton: Princeton University Press.

References

251

———. 2015. “Power and Path Dependence.” In Advances in Comparative Historical Analysis, eds. James Mahoney and Kathleen Thelen. Cambridge: Cambridge University Press, pp. 123−146. Pion-Berlin, David. 1985. “The Fall of Military Rule in Argentina: 1976-1983.” Journal of Interamerican Studies and World Affairs 27, no. 2: 55–76. ———. 1986. “The Defiant State: Chile in the Post-Coup Era,” Armies and Politics in Latin America, eds. Abraham F. Lowenthal and J. Samuel Fitch. New York: Holmes and Meier, pp. 317–334. Pion-Berlin, David, Diego Esparza, and Kevin Grisham. 2014. “Staying Quartered: Civilian Uprisings and Military Disobedience in the Twenty-first Century.” Comparative Political Studies 47, no. 2: 230−259. ———. 1985. The Military in South American Politics. London: Croom Helm. Policzer, Pablo. 2009. The Rise and Fall of Repression in Chile. Notre Dame: University of Notre Dame Press. Posusney, Marsha Pripstein, and Michele Penner Angrist, eds. 2005. Authoritarianism in the Middle East: Regimes and Resistance. Boulder: Lynne Rienner. Potocki, Rodger. 2011. “Belarus: A Tale of Two Elections.” Journal of Democracy 22, no. 3: 49−63. Powell, Jonathan M., and Clayton L. Thyne. 2011. “Global Instances of Coups from 1950 to 2010.” Journal of Peace Research 48, no. 2: 249−259. Powell, Walter W., and Paul J. DiMaggio, eds. 1991. The New Institutionalism in Organizational Analysis. Chicago: University of Chicago Press. Pravda, Alex. 1981. “East-West Interdependence and the Social Compact in Eastern Europe.” In East-West Relations and the Future of Eastern Europe: Politics and Economics, eds. Morris Bornstein, Zvi Gitelman, and William Zimmerman. London: Allen & Unwin, pp. 162−187. Przeworski, Adam. 1991. Democracy and the Market: Political and Economic Reforms in Eastern Europe and Latin America. New York: Cambridge University Press. ———. 2004. “Institutions Matter?” Government and Opposition, no. 39: 527−540. ———. 2009. “Is the Science of Comparative Politics Possible?” In The Oxford Handbook of Comparative Politics, eds. Carles Boix and Susan C. Stokes. Oxford: Oxford University Press, pp. 147–171. ———. 2014. “Parties, States, and Economic Reforms in the Soviet Union and China.” Paper presented at the conference “Governance, Adaptability and Institutional Stability Under Contemporary One-party Rule: Comparative Perspectives,” Nanchang, March. ———. 2016. “Zhengdang–guojia. Sulian he Zhongguo de dang, guojia yu jingji gaige” [Parties, states, and economic reforms in the Soviet Union and China]. In Zhonggong zhili yu shijing: Bijiao de shijiao [Governance and adaptation by the CCP: Comparative perspectives]. 2nd ed., eds. Keping Yu, Thomas Heberer, and Björn Alpermann. Beijing: Zhongyang bianyi chubanshe, pp. 49−86. Przeworski, Adam, Michael E. Alvarez, José Antonio Cheibub, and Fernando Limongi. 2000. Democracy and Development: Political Institutions and Wellbeing in the World, 1950–1990. Cambridge: Cambridge University Press. Radnitz, Scott. 2010. “The Color of Money: Privatization, Economic Dispersion, and the Post-Soviet ‘Revolutions.’” Comparative Politics 42, no. 2: 127−146. Regan, Patrick M., and Errol A. Henderson. 2002. “Democracy, Threats and Political Repression in Developing Countries: Are Democracies Internally Less Violent?” Third World Quarterly 23, no. 1: 119–136. Remmer, Karen L. 1989. Military Rule in Latin America. Boston: Unwin Hyman. Renmin Ribao [People’s Daily], Beijing. “Report by Hu Jintao at the 18th National

252

References

Congress of the CCP.” 2012. Beijing: Communist Party of China. www.china .org.cn/china/18th_cpc_congress/2012-11/16/content_27137540_12.htm. Resende-Santos, João. 2007. Neorealism, States, and the Modern Mass Army. Cambridge: Cambridge University Press. Resolution of the State Council Regarding the Implementation of the Program. 2013. Beijing: The Central People’s Government of the People’s Republic of China. www.gov.cn/zwgk/2013-02/05/content_2327531.htm. Reyntjens, Filip. 2005. “Briefing: Burundi: A Peaceful Transition After a Decade of War?” African Affairs 105, no. 418: 117–135. Rivera, David W., and Sharon Werning Rivera. 2014. “Is Russia a Militocracy? Conceptual Issues and Extant Findings Regarding Elite Militarization.” PostSoviet Affairs 30, no. 1: 27−50. Rivera, Mauricio A. 2010. ¿Por qué reprimen las democracias? Los determinantes de la violencia estatal en América Latina. Mexico City: Latin American Faculty of Social Sciences (FLACSO). Rivero, Gonzalo. 2011. “Oligopoly of Violence.” Unpublished manuscript, New York University. Robertson, Graeme B. 2010. The Politics of Protest in Hybrid Regimes: Managing Dissent in Post-Communist Russia Countries. New York: Cambridge University Press. Robinson, Neil. 2011. “Russian Patrimonial Capitalism and the International Financial Crisis.” Journal of Communist Studies and Transition Politics 27, nos. 3−4: 434−455. Roehrig, Terence. 2013. “The Roles and Influence of the North Korean Military.” In North Korea in Transition: Politics, Economy, and Society, eds. Kyung-Ae Park, Scott Snyder, and Nicholas Anderson. Lanham, MD: Rowman & Littlefield, pp. 47−66. Rouquié, Alain. 1987. The Military and the State in Latin America. Berkeley: University of California Press. Roy, Olivier. 1996. The Failure of Political Islam. Cambridge: Harvard University Press. Rubin, Barry. 2002. The Tragedy of the Middle East. Cambridge: Cambridge University Press. Rubongoya, Joshua B. 2007. Regime Hegemony in Museveni’s Uganda: Pax Musevenica. New York: Palgrave Macmillan. Rumin, Serge. 2012. “Burundi.” In Disarmament, Demobilization and Reintegration and Security Sector Reform, eds. Alan Bryden and Vincenza Scherrer. Geneva: Geneva Centre for the Democratic Control of Armed Forces, pp. 71114. RWI (Research World International). 2016. Post Nomination Opinion Poll Report. Kampala. Kampala: RWI. Sadiki, Larbi. 2010. “Wither Arab ‘Republicanism’? The Rise of Family Rule and the ‘End of Democratization’ in Egypt, Libya and Yemen.” Mediterranean Politics 15, no. 1: 99−107. Salehyan, Idean, and Cullen Hendrix. 2014. “Social Conflict in Analysis Database Version 3.1: Codebook and Coding procedures.” University of North Texas and University of Denver. https://www.strausscenter.org/scad.html. Sarotte, Mary Elise. 2014. The Collapse: The Accidental Opening of the Wall. New York: Basic Books. Sater, James N. 2013. “New Wine in Old Bottles: Political Parties Under Mohammed VI.” In Contemporary Morocco: State, Politics and Society Under

References

253

Mohammed VI, eds. Bruce Maddy-Weitzman and Daniel Zisenwine. New York: Routledge, pp. 9−23. Schäfer, Bernd. 2012. “Die DDR und die ‘chinesische Lösung.’” In 1989 und die Rolle der Gewalt, ed. Martin Sabrow. Göttingen: Wallstein, pp. 153−172. Scharpf, Fritz W. 1999. Regieren in Europa: Effektiv und demokratisch [Governing in Europe: Effective and democratic?]. Frankfurt am Main: Campus Verlag. Schedler, Andreas. 2002. “Elections without Democracy: The Menu of Manipulation.” Journal of Democracy 13, no. 2: 36-50. ———. 2006. “The Logic of Electoral Authoritarianism.” In Electoral Authoritarianism: The Dynamics of Unfree Competition, ed. Andreas Schedler. Boulder: Lynne Rienner, pp. 1−23. ———. 2013. The Politics of Uncertainty: Sustaining and Subverting Electoral Authoritarianism. Oxford: Oxford University Press. ———. 2014. “The Criminal Subversion of Mexican Democracy.” Journal of Democracy 25, no. 1: 5–18. ———. 2015. The Politics of Uncertainty. Sustaining and Subverting Electoral Authoritarianism. Oxford: Oxford University Press. Schedler, Andreas and Bert Hoffmann. 2016. “Communicating Authoritarian Elite Cohesion,” Democratization 23, no. 1: 93–117. Schelling, Thomas C. 1960. The Strategy of Conflict. Cambridge: Harvard University Press, 1980. Schlumberger, Oliver. 2008. Autoritarismus in der arabischen Welt: Ursachen, Trends und internationale Demokratieförderung. Baden-Baden: Nomos. ———. ed. 2007. Debating Arab Authoritarianism: Dynamics and Durability in Nondemocratic Regimes. Stanford: Stanford University Press. Schock, Kurt. 2005. Unarmed Insurrections: People Power Movements in Nondemocracies. Minneapolis: University of Minnesota Press. Schoenhals, Michael. 2011. “Recruiting Agents in Industry and Trade: Lifting the Veil on Early People’s Republic of China Operational Work.” Modern Asian Studies 45, no. 5: 1345−1369. ———. 2013. Spying for the People: Mao’s Secret Agents, 1949−1967. New York: Cambridge University Press. Schroeder, Klaus. 2103. Der SED-Staat: Geschichte und Strukturen der DDR, 1949−1990. Cologne: Böhlau. Schubert, Gunter. 2008. “One-party Rule and the Question of Legitimacy in Contemporary China: Preliminary Thoughts on Setting Up a New Research Agenda.” Journal of Contemporary China 17, no. 54: 191−204. ———. 2014. “Political Legitimacy in Contemporary China Revisited: Theoretical Refinement and Empirical Operationalization.” Journal of Contemporary China 23, no. 88: 593−611. Schubert, Gunter, and Thomas Heberer. 2015. “Continuity and Change in China’s ‘Local State Developmentalism.’” Issues and Studies 51, no. 2: 1−38. Schwedler, Jillian. 2015. “Comparative Politics and the Arab Uprisings.” Middle East Law and Governance 7, no. 1: 141−152. Scobell, Andrew. 1990. “The Death Penalty in Post-Mao China,” China Quarterly, no. 123: 503−520. Scott, James C. 1985. Weapons of the Weak: Everyday Forms of Peasant Resistance. New Haven: Yale University Press. ———. 1998. Seeing Like a State: How Certain Schemes to Improve the Human Condition Have Failed. New Haven: Yale University Press. Scully, Timothy R. 1995. “Reconstructing Party Politics in Chile.” In Building

254

References

Democratic Institutions: Party Systems in Latin America, eds. Scott Mainwaring and Timothy R. Scully. Cambridge: Cambridge University Press, pp. 100-137. Seeberg, Merete. 2013. “Authoritarianism and Elections During the Third Wave.” Statsvetenskaplig tidskrift 115, no. 4: 313−344. Sehring, Jenniver, and Christoph H. Stefes. 2010. “Die Stabilität von kompetitivautoritaren Regimen: Armenien, Georgien, Kasachstan und Kirgistan im Vergleich.” In Autoritarismus Reloaded. Neuere Ansätze und Erkenntnisse der Autokratieforschung, eds. Holger Albrecht and Rolf Frankenberger. BadenBaden: Nomos, pp. 275−294. Sewell, William H. 2005. Logics of History: Social Theory and Social Transformation. Chicago: University of Chicago Press. Shambaugh, David. 2008. China’s Communist Party: Atrophy and Adaptation. Berkeley: University of California Press. ———. 2015. “The Coming Chinese Crackup.” Wall Street Journal, March 6. ———. 2016. China’s Future. Cambridge: Polity Press. Shepsle, Kenneth A. 1986. “Institutional Equilibrium and Equilibrium Institutions.” In Political Science: The Science of Politics, ed. Herbert Weisberg. New York: Agathon, pp. 51–81. ———. 1989. “Studying Institutions: Some Lessons from the Rational Choice Approach.” Journal of Theoretical Politics 1, no. 1: 31–47. Shermatova, Sanobar. 2006. “Andizhan: Karimov, zapad I uzbekskaya oppozitsiya peresmatrivaiut otsenki.” Russia Today. http://ria.ru/analytics/20061020/54990942 .html. Shi, Tianjian. 2008. “Democratic Values Supporting an Authoritarian System.” In How East Asians View Democracy, eds. Yun-han Chu, Larry Diamond, Andrew Nathan, and Doh Chull Chin. New York: Columbia University Press, pp. 209−237. Shirk, Susan. 2002. “The Succession Game.” In The 16th CCP Congress and Leadership Transition in China, eds. Gang Lin and Susan Shirk. Washington, DC: Woodrow Wilson Center Asia Program, pp. 5–9. Shpil’kin, Sergey. 2011. “Statistika issledovala vybory.” Gazeta.ru. www.gazeta.ru /science/2011/12/10_a_3922390.shtml. Sigley, Gary. 2006. “Chinese Governmentalities: Government, Governance and the Social Market Economy.” Economy and Society 35, no. 4: 487−508. Silitski, Vitali. 2010. “‘Survival of the Fittest’: Domestic and International Dimensions of the Authoritarian Reaction in the Former Soviet Union Following the Colored Revolutions.” Communist and Post-Communist Studies 43, no. 4: 339−350. Simpser, Alberto. 2013. Why Governments and Parties Manipulate Elections: Theory, Practice, and Implications. Cambridge: Cambridge University Press. Singiza, Douglas K., and Jaap De Visser. 2011. “Chewing More than One Can Swallow: The Creation of New Districts in Uganda.” Law, Development and Democracy 15, no. 1: 1−20. Skocpol, Theda. 1979. States and Social Revolutions: A Comparative Analysis of France, Russia and China. Cambridge: Cambridge University Press. Slater, Daniel. 2010. “Altering Authoritarianism. Institutional Complexity and Autocratic Agency in Indonesia.” In Explaining Institutional Change: Agency, Ambiguity, and Power, eds. James Mahoney and Kathleen Thelen. New York: Cambridge University Press, pp. 132−167. Slater, Dan, and Sofia Fenner. 2011. “State Power and Staying Power: Infrastructural Mechanisms and Authoritarian Durability.” Journal of International Affairs 65, no. 1: 15−29.

References

255

Smaoui, Sélim, and Mohamed Wazif. 2012. “Étendard de lutte ou pavillon de complaisance?” In Au Coeur des Revoltes Arabes: Devenir Revolutionaires, eds. Amin Allal and Thomas Pierret. Paris: Armand Colin, pp. 55−79. Smith, Benjamin. 2005. “Life of the Party: The Origins of Regime Breakdown and Persistence Under Single-party Rule.” World Politics 57, no. 3: 421–451. Spooner, Mary Helen. 1999. Soldiers in a Narrow Land: The Pinochet Regime in Chile. Berkeley: University of California Press. Springborg, Robert. 1987. “The President and the Field Marshal: Civil-military Relations in Egypt Today.” Middle East Research and Information Project, Middle East Report, no. 147: 4−11. ———. 2011. “Economic Involvement of Militaries.” International Journal of Middle East Studies 43, no. 3: 397–399. ———. 2016. “A Shifting Role of the Military in Arab Politics? Regional Perspectives.” In Military Engagement in Mobilizing Societies, eds. Holger Albrecht, Aurel Croissant, and Fred Lawson. Philadelphia: University of Pennsylvania Press, pp. 71–94. Staadt, Jochen. 1996. Eingaben: Die institutionalisierte Meckerkultur in der DDR: Goldbrokat, Kaffee-Mix, Büttenreden, Ausreiseanträge und andere Schwierigkeiten mit den Untertanen. Berlin: Arbeitspapiere des Forschungsverbundes SED-Staat, No. 24. Stein, Elizabeth A. 2007. Mainstream Newspaper Coverage: A Barometer of Government Tolerance for Anti-regime Expression in Authoritarian Brazil. Research Paper. Cambridge: Harvard University, John F. Kennedy School of Government, Joan Shorenstein Center on the Press, Politics and Public Policy. Stepan, Alfred. 1971. The Military in Politics: Changing Patterns in Brazil. Princeton: Princeton University Press. ———. 1978. The State and Society: Peru in Comparative Perspective. Princeton: Princeton University Press. ———. 1988. Rethinking Military Politics: Brazil and the Southern Cone. Princeton: Princeton University Press. Stockmann, Daniela. 2013. Media Commercialization and Authoritarian Rule in China. New York: Cambridge University Press. Streeck, Wolfgang, and Kathleen Thelen. 2005. “Introduction: Institutional Change in Advanced Political Economies.” In Beyond Continuity: Institutional Change in Advanced Political Economies, eds. Wolfgang Streeck and Kathleen Thelen. New York: Oxford University Press, pp. 1−39. Stykow, Petra. 2013. “Wahlen in autoritären Regimen: die post-sowjetischen Länder im Vergleich.” In Autokratien im Vergleich, eds. Steffen Kailitz and Patrick Köllner. Baden-Baden: Nomos, pp. 237−271. Sun Liping. 2011. “Shehui shixu shi dangxia de yanjun tiaozhan.” Jingji Guancha Bao, February 25. Svolik, Milan W. 2009. “Power Sharing and Leadership Dynamics in Authoritarian Regimes.” American Journal of Political Science 53, no. 2: 477−494. ———. 2012. The Politics of Authoritarian Rule. New York: Cambridge University Press. Swidler, Ann. 1986. “Culture in Action: Symbols and Strategies.” American Sociological Review 51, no. 2: 273−286. Sztompka, Piotr. 1993. “Civilisational Incompetence: The Trap of Post-communist Societies.” Zeitschrift für Soziologie 2, no. 2: 85−95. Tang, Wenfang. 2005. Public Opinion and Political Change in China. Stanford: Stanford University Press.

256

References

———. 2007a. “Interpersonal Trust and Democracy in China.” Paper presented at the conference “Local Governance and Regime Legitimacy in China,” University Duisburg-Essen, February–March. ———. 2007b. “Renji xintuo yu gongmin shehui” [Interpersonal trust and civil society]. In Chengxiang gongmin canyu he zhengzhi hefaxing [Citizen participation in urban and rural areas and political legitimacy], eds. Zengke He, Thomas Heberer and Gunter Schubert. Beijing: Zhongyang bianyi chubanshe, pp. 303−329. Tangri, Roger, and Andrew Mwenda. 2001. “Corruption and Cronyism in Uganda’s Privatization in the 1990s.” African Affairs 100, no. 398: 117−133. Tanneberg, Dag, Christoph H. Stefes and Wolfgang Merkel. 2013. “Hard Times and Regime Failure: Autocratic Responses to Economic Downturns.” Contemporary Politics 19, no. 1: 115–129. Tapscott, Rebecca. 2016. “Where the Wild Things Are Not: Crime Preventers and the 2016 Ugandan Elections.” Journal of Eastern African Studies 10, no. 4: 693−712. Tarasenko, Pavel. 2013. “ODBK perekliuchitsia na tsvetnye revoliutii.” Kommersant, December 20. Tarrow, Sidney. 1991. “‘Aiming at a Moving Target’: Social Science and the Recent Rebellions in Eastern Europe.” PS: Political Science and Politics 24, no. 1: 12–20. Tarrow, Sidney, and Charles Tilly. 2009. “Contentious Politics and Social Movements.” In The Oxford Handbook of Comparative Politics, eds. Carles Boix and Susan C. Stoke. Oxford: Oxford University Press, pp. 435–460. Teorell, Jan, Stefan Dahlberg, Sören Holmberg, Bo Rothstein, Felix Hartmann, and Richard Svensson. 2015. The Quality of Government Standard Dataset. Version January 15. Gothenburg: University of Gothenburg, Quality of Government Institute. Thelen, Kathleen. 1999. “Historical Institutionalism in Comparative Politics.” Annual Review of Political Science 2: 369–404. Thelen, Kathleen, Sven Steinmo, and Frank Longstreth, eds. 1992. Structuring Politics: Historical Institutionalism in Comparative Analysis. Cambridge: Cambridge University Press. Thorp, Rosemary and Geoffrey Bertram. 1978. Peru 1890-1977: Growth and Policy in an Open Economy. New York: Columbia University Press. Tilly, Charles. 2004. Social Movements, 1968–2004. Boulder: Paradigm. Tolipov, Farkhod. 2009. ”CSTO: Collective Security or Collective Confusion?” Central Asia–Caucasus Institute Analyst. http://old.cacianalyst.org/?q=node /5168. Tolstrup, Jakob. 2009. “Studying a Negative External Actor: Russia’s Management of Stability and Instability in the ‘Near Abroad.’” Democratization 16, no. 5: 922−944. Trejo, Guillermo. 2012. Popular Movements in Autocracies: Religion, Repression, and Indigenous Collective Action in Mexico. New York: Cambridge University Press, Mobilereference e-book. Tripp, Aili Mari. 2010. Museveni’s Uganda: Paradoxes of Power in a Hybrid Regime. Boulder: Lynne Rienner. Tsai, Wen-Hsuan, and Nicola Dean. 2013. “The CCP’s Learning System: Thought Unification and Regime Adaptation.” China Journal no. 69: 87−107. Tumushabe, Godber W. 2009. “Trends in Public Administration Expenditure in Uganda.” Advocates Coalition for Development and Environment Policy Research Series. Kampala: ACODE.

References

257

Uganda Electoral Commission. 2016. Presidential Elections, 2016 - District Summary Reports, Final Results. http://www.ec.or.ug/ecresults/02-Final%20Presidential %20Results%20by%20District.pdf. Ulfelder, Jay. 2005. “Contentious Collective Action and the Breakdown of Authoritarian Regimes.” International Political Science Review 26, no. 3: 311–334. Valenzuela, Arturo. 1978. The Breakdown of Democratic Regimes: Chile. Baltimore: Johns Hopkins University Press. ———. 1995. “The Military in Power: The Consolidation of One-Man Rule.” In The Struggle for Democracy in Chile, 1982-1990, eds. Paul W. Drake and Iván Jaksić. Rev. Ed. Lincoln: University of Nebraska Press, pp. 21-72. Valenzuela, J. Samuel. 1992. “Democratic Consolidation in Post-Transitional Settings: Notion, Process, and Facilitating Conditions.” In Issues in Democratic Consolidation: The New South American Democracies in Comparative Perspective, eds. Scott P. Mainwaring, Guillermo O’Donnell and J. Samuel Valenzuela. Notre Dame: University of Notre Dame Press, pp. 57–104. Valiyev, Anar M. 2006. “Parliamentary Elections in Azerbaijan: A Failed Revolution.” Problems of Post-Communism 53, no. 3: 17−35. Vandeginste, Stef. 2009. “Power-sharing, Conflict and Transition in Burundi: Twenty Years of Trial and Error.” Africa Spectrum 44, no. 3: 63−86. Vanderhill, Rachel. 2012. “Learning to Be Bad: How Autocratic Leaders Adopt Strategies from Abroad to Maintain Power.” Paper prepared for the meeting of the American Political Science Association, New Orleans, August−September. Vandewalle, Dirk. 1998. Libya Since Independence: Oil and State-building. Ithaca: Cornell University Press. van de Walle, Nicholas. 2002. “Africa’s Range of Regimes.” Journal of Democracy 13, no. 2: 66−80. Varas, Augusto. 1995. “The Crisis of Legitimacy of Military Rule in the 1980s.” In The Struggle for Democracy in Chile, 1982-1990, eds. Paul W. Drake and Iván Jaksić. Rev. Ed. Lincoln: University of Nebraska Press, pp. 73-97. Vásquez Huamán, Enrique. 2000. Estrategias del Poder: Grupos Económicos en el Perú. Lima: Centro de Investigación, Universidad del Pacífico. Vogel, Ezra F. 2011. Deng Xiaoping and the Transformation of China. Cambridge: Harvard University Press. Vokes, Richard, and Sam Wilkins. 2016. “Party, Patronage and Coercion in the NRM’s 2016 Re-election in Uganda: Imposed or Embedded?” Journal of Eastern African Studies 10, no. 4: 581−600. Vollan, Björn, Yuxin Zhou, Andreas Landmann, Biliang Hu, and Carsten HerrmannPillath. 2013. Cooperation Under Democracy and Authoritarian Norms. University of Innsbruck Working Papers in Economics and Statistics. www.frankfurt -school.de/clicnetclm/fileDownload.do?goid=000000560653AB4. Volpi, Frédéric. 2003. Islam and Democracy: The Failure of Dialogue in Algeria. London: Pluto Press. ———. 2006. “Algeria’s Pseudo-democratic Politics: Lessons for Democratization in the Middle East.” Democratization 13, no. 3: 442−455. ———. 2010. Political Islam Observed: Disciplinary Perspectives. New York: Oxford University Press. ———. 2017. Revolution and Authoritarianism in North Africa. New York: Oxford University Press. Wahman, Michael. 2013. “Opposition Coalitions and Democratization by Election.” Government and Opposition 48, no. 1: 3−32. Wahman, Michael, Jan Teorell, and Axel Hadenius. 2013. “Authoritarian Regime

258

References

Types Revisited: Updated Data in Comparative Perspective.” Contemporary Politics 19, no. 1: 19−34. Walker, Christopher, and Alexander Cooley. 2013. “Vote of the Living Dead.” Foreign Policy, October 31. Wallander, Celeste A. 2007. “Russian Transimperialism and Its Implication.” Washington Quarterly 30, no. 2: 107−122. Ware, L. B. 1985. “The Role of the Tunisian Military in the Post-Bourguiba Era.” Middle East Journal 39, no. 1: 27−47. Waterbury, John. 1970. The Commander of the Faithful: The Moroccan Political Elite—A Study in Segmented Politics. New York: Columbia University Press. Way, Lucan A. 2015. “The Limits of Autocracy Promotion: The Case of Russia in the ‘Near Abroad.’” European Journal of Political Research 54, no. 4: 691−706. Way, Lucan A., and Steven Levitsky. 2007. “Linkage, Leverage, and the Post-communist Divide.” East European Politics and Societies 21, no. 1: 48−66. Wedeen, Lisa. 1999. Ambiguities of Domination: Politics, Rhetoric, and Symbols in Contemporary Syria. Chicago: University of Chicago Press. Wedeman, Andrew. 2009. “Enemies of the State: Mass Incidents and Subversion.” Paper presented at the meeting of the American Political Science Association, Toronto, September 2009. Weeks, Gregory. 2004. “Rethinking Historical Factors: Military and Political Transitions in South America.” Politics & Policy 32, no. 1: 158–174. Weiss, Linda. 2000. “Developmental States in Transition: Adapting, Dismantling, Innovating, Not ‘Normalizing.’” Pacific Review 13, no. 1: 21−55. Weiss Fagen, Patricia. 1992. “Repression and State Security.” In Fear at the Edge: State Terror and Resistance in Latin America, eds. Juan E. Corradi, Patricia Weiss Fagen and Manuel Antonio Garretón. Berkeley: University of California Press, pp. 39–71. Weyland, Kurt. 2008. “Toward a New Theory of Institutional Change.” World Politics 60, no. 2: 281–314. ———. 2012. “The Arab Spring: Why the Surprising Similarities with the Revolutionary Wave of 1848?” Perspectives on Politics 10, no. 4: 917−934. ———. 2014. Making Waves: Democratic Contention in Europe and Latin America Since the Revolutions of 1848. New York: Cambridge University Press. “What Does Xi Jinping’s Top-down Leadership Mean for Innovation in China?” 2017. A China File Conversation, China File. www.chinafile.com/conversation /what-does-xi-jinpings-top-down-leadership-mean-innovation-china. Wheatley, Jonathan. 2005. Georgia from National Awakening to Rose Revolution: Delayed Transition in the Former Soviet Union. London: Ashgate. White, Stephen. 2010. “Classifying Russia’s Politics.” In Developments in Russian Politics 7, eds. Stephen White, Richard Sakwa, and Henry E. Hale. Houndmills: Palgrave Macmillan; Durham: Duke University, p. 263–282. Press. Whitehead, Laurence. 2002. Democratization: Theory and Experience. Oxford: Oxford University Press. Wilkins, Sam. 2016. “Who Pays for Pakalast? The NRM’s Peripheral Patronage in Rural Uganda.” Journal of Eastern African Studies 10, no. 4: 619−638. Wintrobe, Ronald. 1998. The Political Economy of Dictatorship. New York: Cambridge University Press. Woo-Cumings, Meredith, ed. 1999. The Developmental State. Ithaca: Cornell University Press. World Bank, ed. 1997. World Development Report 1997: The State in a Changing World. Oxford: Oxford University Press.

References

259

Wright, Joseph. 2008. “Do Authoritarian Institutions Constrain? How Legislatures Affect Economic Growth and Investment.” American Journal of Political Science 52, no. 2: 322–343. Yin, Chien-Cung. 1998. “Equilibria of Collective Action in Different Distributions of Protest Thresholds.” Public Choice 97, no. 4: 535–567. Yom, Sean L., and F. Gregory Gause III. 2012. “Resilient Royals: How Arab Monarchies Hang On.” Journal of Democracy 23, no. 1: 74−88. Yu Keping. 2013. Zhongguo zhili bianqian 30 nian (1978-2008) [30 years of changes in China’s governance], Jilin Daxue Shehui Kexue Xuebao [Social Science Journal of Jilin University], http://www.gmw.cn/xueshu/2013-12/05 /content_9710161_6.htm. Zaslavsky, Victor, and Robert J. Brym. 1978. “The Functions of Elections in the USSR.” Soviet Studies 30, no. 3: 362-371. Zeghal, Malika. 2008. Islamism in Morocco: Religion, Authoritarianism, and Electoral Politics. Princeton: Markus Wiener Zhao, Suisheng. 1997. “Political Reform and Changing One-party Rule in Deng’s China.” Problems of Post-Communism 44, No. 5: 13−21. Zhao Liang. 2001. The Tiananmen Papers. New York: Public Affairs. Zhao Ziyang. 2009. Prisoner of the State: The Secret Journal of Zhao Ziyang. Trans. Bao Pu, Renee Chiang, and Adi Ignatius. New York: Simon & Schuster. Zhou Xun, ed. 2012. The Great Famine in China, 1958−1962: A Documentary History. New Haven: Yale University Press. Zimmermann, Ekkart. 1979. “Explaining Military Coups d’État: Towards the Development of a Complex Causal Model.” Quality and Quantity 13, no. 5: 431−441.

The Contributors

Felipe Agüero is professor at the Institute for Public Affairs at the University of Chile.

Max Bader is university lecturer with expertise on Russia and Eurasia (postSoviet area) at the University of Leiden.

Julian Brückner is a PhD candidate at the Berlin Graduate School of Social Sciences.

Aurel Croissant is professor of political science at the University of Heidelberg.

Martin K. Dimitrov is associate professor of political science and director of the Asian Studies Program at Tulane University.

Tanja Eschenauer is a PhD candidate at the Department of Political Science, University of Heidelberg.

Johannes Gerschewski is a lecturer of comparative politics at Humboldt University Berlin.

Thomas Heberer is senior professor of China politics and society at the University Duisburg-Essen.

Svein-Erik Helle is a PhD candidate at the Department for Comparative Politics, University of Bergen.

Jil Kamerling is a PhD candidate at the Department of Political Science, University of Heidelberg.

261

262

The Contributors

Lise Rakner is professor at the Department for Comparative Politics, University of Bergen. Andreas Schedler is professor at the Department of Political Science, Centro de Investigación y Docencia Económicas (CIDE), Mexico City.

Christoph H. Stefes is associate professor of comparative European and post-Soviet studies at the University of Colorado Denver. Milan W. Svolik is associate professor in political science at Yale University.

Frédéric Volpi is senior lecturer at the School of International Relations, University of St. Andrews.

Index

Africa, 44, 49, 54, 69, 85, 90, 92, 95– 97, 104, 111f., 115, 118, 132; North Africa, see Middle East and North Africa (MENA) Akayev, Askar, 136 Akena, Jimmy, 125 al-Qaeda, 206 Algeria, 206–210 Allende, Salvador, 1, 188, Alliance for Election Campaign Finance Monitoring (ACFIM) of Uganda, 128, 130 ambiguity, institutional, 15, 81 America; America, Central, 54, 69, 95; America, Latin, 18, 86, 95, 97, 175, 177, 179–185, 195, 201 America, United States of (US), 84, American Popular Revolutionary Alliance (APRA), 185–186, 192– 194 Andijan uprising, 137, 142, 147–148, Arab socialism, 204–205, 211 Arab Spring/Arab uprisings, 47, 55, 89, 98, 107, 135, 199–215 Argentina, 175, 183, 185–190, 195–196 Armenia, 135–136, 144 Arusha Peace Agreement, 104

Asia, 85, 90, 97, 171; Asia, Central, 145, 147; Asia, East, 158–159 authoritarianism, 14, 18, 43–45, 49, 58–60, 69, 76, 85, 87, 144, 155– 159, 159, 175, 193, 201, 206–207, 211–212, 214, authority, 75–79, 81–84, 86, 89, 123, 127, 146, 155, 164, 203–205, 210

autocracy, 1, 4, 6, 47, 54, 84, 114, 175 Azerbaijan, 17, 96, 135–137

balance of power, 75, 82–84, 220–223, 225 Belarus, 17, 135, 137, 144–150 Belaúnde, Fernando, 191, 194 Ben Ali, Zine El Abidine, 202, 206, 208, 210, 213 Ben Bella, Ahmed, 203 Besigye, Kizza, 118–120, 126, Boumedienne, Houari, 204–205 Bourguiba, Habib, 203, 205 Bouteflika, Abdelaziz, 208 Brazil, 175, 179, 182–183, 186–191, 195–196 Bulgaria, 24, 28–29, 33, 224 Burkina Faso, 119 Burundi, 103–104, 109 Buyoya, Pierre, 104–105

Caribbean, 44, 54, 69, 95, 97 Center for High Military Studies (CAEM), 192 Cerro, Sánchez, 185 change agent, 10–16, 119 charisma, 82, 203–204 Chile, 18, 78, 175, 177, 182–196 China, 22–25, 29, 31–38, 44, 47, 59, 65, 75–85, 145, 150, 155–15, 165– 173, 217–219, 225–226. Chinese Communist Party (CCP), 81, 155, 158–160, 164–172, 225, 228 citizen complaint, 16, 23, 27–37 civil-military relations, 181, 205 clientelism, 207–213

263

264

Index

co-optation, 14, 52, 77, 105, 107, 113, 117, 139, 169, 201, 210–212, 225 coalition, 11, 18, 43, 77, 89, 91–92, 98, 100, 104–105, 108, 117 cohesion/cohesiveness, 122, 124–126, 180, 182, 185, 188, 190 collective action, 10, 43, 46, 171, 181 collective leadership, 79, 81–82 Collective Security Treaty Organization (CSTO), 136, 142, 144, 147–150 Color(ed) Revolution, see revolution Commonwealth of Independent States (CIS), 18, 136, 144–150 Communist Party; 22–23, 107, Communist Party, China (CCP), see Chinese Communist Party compliance, 9, 11, 13, 16, 24, 46, 77, 83, 92, 145, 169, 171, 208, 220, 223 consolidation, 4, 31, 155, 169, 181 constitution, 75–77, 79–80, 83, 117, 119, 121, 126, 176–177, 182, 187– 191, 194 construction error, 13–15, 99, contentious politics, 49, 53, 68, 101, 211 contestation, 114, 200 corruption, 10, 14, 32, 107–108, 117, 131–132, 141, 143, 168, 170, 172 Council of Europe, 143, 150 coup(s), 1, 21, 26, 78, 90, 92, 98–109, 129, 138–140, 175–178, 183, 186– 195, 206, 210, 222 court(s), 2, 10, 76, 106, 119, 176 crisis, 1–2, 22, 24–25, 47, 49, 99, 108– 109, 113–115, 119, 130, 141, 166– 169, 190–191, 195–196, 200–204, 211–214, 228; crisis, causes of, 9– 15, 138, 142, 166, 185; crisis, definition, 2–3, 49–53, 217–219; crisis, in China, 31–38; crisis, in GDR, 26–30; crisis, locus of, 6–9; crisis, preemption, 16–19, 38–39, 162, 176–177, 224–225 critical juncture(s), 5–9, 14, 167, 183– 184, 202

decay, 5, 12–13, 25, 112, 115, 117–118, 132, 168, 200–201, 204, 209, 211– 214, 218, 221, 226–228 deinstitutionalization, 6, 9–15, 18, 112, 114, 118–119, 129, 132, 199, 201– 202, 211–215

Democratic Party (DP) of Uganda, 116, 120–121, 124–126 democratization, 4, 11, 43, 112–113, 142, 145, 150, 172, 185, 201, 205 Deng Xiaoping, 14, 34, 75, 78–84, 167, 170 Depression, Great, 185 developmental state, 14, 18, 156–160, 163, 165–166, 168–172 dictator/dictatorship, 43–49, 54, 62, 68, 70, 75–86, 89–91, 98, 101–102, 105–106, 108–109, 171–172, 175– 177, 180, 189, 195 discontent, 16, 21–38, 48, 98, 108, 151, 193, 224, 226 disequilibria, 9 Egypt, 52, 98, 103, 105–106, 222–223 elections, 5, 10, 17–18, 22, 47, 59, 76, 84, 103, 108, 135–147, 150, 185, 194, 212, 222; elections, multiparty, 12–13, 17, 92; elections, (semicompetitive), 21, 176, 191, 220, 222; elections, Ugandan, 111–133 elite, 5–7, 11, 13–17, 46, 48, 54, 59, 65, 77, 80–84, 89–91, 98–100, 104– 108, 116–117, 125, 135, 139–141, 158, 176, 178, 183, 185, 190, 199, 202, 204–226, endogenous sources, 6, 9 equilibria, 44, 49, 70, 83 erosion, 29, 112, 115, 132, 166, 200– 201, 204, 209, 211–212, 214, 221 Eurasia, 96, 135–136, 138, 141–144, 150 Europe, 183, 196, 206; Europe, Eastern, 22–24, 29–30, 33, 35, 38, 85, 135, 181; Europe, Southern, 95, 97, 181 eventful sociology, 18, 201 evolution, 15, 25, 200, 203, 209, 211 extremism, 148–150; extremism, Counter-, 148–150

factionalism, 125, 180, 184–187, 192, 196 Forum for Democratic Change (FDC) of Uganda, 120, 124–126 Franco, Francisco, 190 Fujimori, Alberto, 194 game theory, 46 Georgia, 107–109, 135–136, 144, 150

Index German Democratic Republic (GDR), 22–34, 218–219, 224 governance, 16–18, 21–22, 26, 30, 49, 65, 70, 81, 117, 120, 122, 141, 143, 165, 170–171, 199–215, 223, 225 government, local, 29, 59, 112, 115, 117, 122–123, 127–128, 132, 137, 148, 157, 161–165, 168, 170, 226 gradual change, 9, 14

Hassan II, 204 Honecker, Erich, 29–30 Hu Jintao, 79, 81–82 hybrid regimes, 86, 113, 128, 132

Idris, Muhammad, 203–204 incumbent, 3–5, 13, 76, 98, 102, 107, 111–115, 120–122, 125–128, 131– 133, 222 information, 16, 28, 30, 38, 48, 54, 91, 139, 142, 163; information, gathering, 21–27, 31–36; information, incomplete, 21, 29–30, 46, 141 innovation, 14, 18, 158–163 Institutional Revolutionary Party Mexico (PRI), 78, 83, 85 institutionalization, 10, 12, 81–82, 100, 103, 105, 158–160, 165–166, 171, 175–187, 190, 193–196, 203, 205 International Monetary Fund (IMF), 205, 208 internet (blogging, social media), 36, 38, 142–143, 149, 164 Iran, 85, 201, 222 Iraq, 98 Islam(ism), 106, 200, 205–210

Janaozen, 137, 142, 148 Jiang Zemin, 35, 79, 80–82 junta, 78, 86, 90, 175, 178–179, 182, 188–194, 221

Karimov, Islam, 142, 219 Kazakhstan, 17, 135, 137, 140, 142, 144–146, 150 Kyrgyzstan, 135–138, 144–145, 148, 150–151, 219

leadership, 22, 24, 26–35, 38, 75, 78– 79, 95, 112, 117, 119, 122, 124, 131, 204, 217–219, 222–223; lead-

265

ership, Chinese, 157–172, 225–226; leadership, collective, 82; leadership, political/civilian, 98, 155, 203, 222; leadership, military, 98, 104, 108, 176–177, 189; leadership, succession/transition, 5, 76, 81, 179; leadership, tenure/term limits, 80– 81, 83 legacy, 8, 142–143, 181–182, 184, 187–188, 192, 194 legitimacy, 7, 11, 14, 16, 35, 45–47, 89, 116, 125, 129, 132, 139, 141–142, 145–146, 151, 155–160, 163–165, 169, 172, 185, 187, 200, 202–206, 209, 211, 213, 220, 223, 225–226 legitimation, 113, 139, 226 Libya, 2, 56, 203–210, 213–214 linkage and leverage, Western, 206–207

Mao Tse-tun, 31, 78, 82, 85, 158 Marcos, Ferdinand, 13–14 mass/masses, 11, 35, 84, 100–101, 180, 185, 219, 224; mass media, 166; mass, protest/uprising/contestation, 9, 22–23, 25, 43–44, 46–53, 56, 61, 69, 76, 89, 98, 106–109, 214, 218– 219, 225; mass, support, 5–6, 22, 160, 192, Mbabazi, Amama, 120, 123–126 Mexico, 44, 47, 54, 56, 78,85, 225 Middle East and North Africa (MENA), 95, 97, 199, 200, 203, 205–207 Mielke, Erich, 30 military, 17, 51, 76, 80–81, 85, 89–109, 204–206, 209, 211, 219, 221–223; military, leaders/rulers, 18, 89, 91– 97, 99–103, 109, 221; military regime(s), 17–18, 85–86, 90–91, 96, 99, 109, 115, 175–196, 206, 221; military, supporters, 91–97, 99, 101–103, 109 mobilization, 5–7, 15, 47, 49, 53, 89, 98, 122, 125, 130, 160, 179, 191, 193, 201, 202, 205, 211, 214–215 modernization, 14, 156, 158, 165, 170– 172, 183, 192, 203 Mohammed V, 203–204 Monarchy, 175, 203–207, 209–210 monitoring, 22, 24, 26–27, 31–32, 34, 38, 80–81, 121, 130, 145, 161, 220, 224

266

Index

Morocco, 203–209, 214 Mubarak, Gamal, 106 Mubarak, Hosni, 103–107 Museveni, Yoweri, 12, 111–113, 115– 123, 125, 127–133

National Agricultural Advisory Service (NAADS) of Uganda, 122 National People’s Congress of PR China, 79, 80, 171 National Resistance Army (NRA) of Uganda, 115 National Resistance Movement (NRMS) of Uganda, 116–119, 122 National Resistance Movement Party (NRM) of Uganda, 111–113, 115– 133 National System of Social Mobilization (SINAMOS), 192–194 Ndadaye, Melchior, 104 neo-liberalism/neoliberalism/neoliberal, 205, 207–210, 223, 226 neopatrimonialism/neopatrimonial, 107, 151 Ngeze, Francois, 104 North Atlantic Treaty Organization (NATO), 147 Obote, Milton, 115, 120, 125, 129 Odría, Manuel, 185 Opposition/oppositional, 6–7, 10–11, 14, 16, 21–23, 26, 28, 45, 47, 52, 71, 76–77, 82–83, 89, 97–98, 102, 108, 111–133, 135–136, 138, 140–142, 149–150, 160, 171, 176, 183–184, 186, 191, 193, 202, 204–206, 210– 212, 214, 218–220, 222–225, 227 Orange Revolution, 136 Organisation for Security and Cooperation in Europe (OSCE), 143, 145– 147, 150 Otunnu, Olara, 120, 124

Paraguay, 183 paramilitary, 123, 222 parliament/parliamentary, 10, 108, 111, 117–119, 121, 125, 127– 129, 136– 137, 144, 148–149, 206–207, 209– 210 partial equilibrium analysis, 77 party, 4–5, 14, 17, 22–23, 25, 27, 30– 32, 35, 47, 75–82, 85–86, 104–108,

111–113, 116–132, 136, 148, 155– 157, 162–172, 176, 179–180, 184– 187, 192, 195, 205–206, 208–210, 222, 224–225, 227; multi-party, 12, 13, 17, 92, 112–115, 117–120, 122– 126, 129, 133, 171, 206, 209, 224; no-party, 12, 112, 115–117, 132, 224; one-party, 4, 122, 127, 156; party, opposition, 128; party, ruling, 128, 130, 222, 225, 228; singleparty, 25, 85–86, 115–116, 175; party, system, 115–117, 119, 132, 171, 184–185, 208–210, 224–225 path dependency, 7–9, 14 patrimonialism/patrimonial, 143, 207; neopatrimonialism/neopatrimonial, 107, 151 patron-client relations, see clientelism patronage, 10, 123, 128, 132, 139, 208, 222, 226 Peron, Juan, 186 Peru, 18, 175, 177, 180, 182–183, 185, 187–188, 191–192, 195–196 perverse incentives/effects, 10, 13–15, 17–18, 101, 184, 203, 222 Philippines, 49 Pinochet, Augusto, 78, 183, 188–191 Politburo: of Chinese Communist Party, 35, 75, 80–82; of Socialist Unity Party of Germany, 29–30 political control of military, 89, 101 political-military relation, 91, 95–96, 98 power-sharing, 5–6, 98, 176–177, 180, 220 preference falsification, 21, 23, 27, 45 Presidential Advisory Committee (COAP), 192–193 principal-agent problem, 48 protest, 6–7, 9, 11, 15–17, 22, 24–26, 35, 38, 43–49, 51–53, 55–62, 65, 68–70, 92–93, 101–103, 106–109, 123, 130–132, 136–138, 140–142, 147–148, 155,167, 199–202, 211– 215, 217–219, 224–225; protest, duration, 54–56, 58, 61–62, 69; protest, effects, 45; protest, motives, 59–61, 65–66; protest, size, 4, 62, 68; protest, street, 29, 46, 213, 218– 219, 224; protest, types of, 24, 52 punctuated equilibrium, 7, 207

Index Pussy Riot, 149 Putin, Vladimir, 84, 140

Qaddafi, Muamar, 204–205, 207–208, 210 quasi-parameter, 10, 12, 13–15, 112, 114, 129, 132, 221–224, 227–228

Radical Party (of Argentina), 186 rapid change/deinstitutionalization, 6, 8–9, 13, 18, 201, 212–214, 218, 221, 225–226 referendum, 118, 190–191 regime: regime, crisis, 2–3, 7–9, 11–12, 14–19, 24–25, 43–45, 49–50, 53, 70, 91, 98, 101–103, 109, 119, 214, 218, 227; regime, decay, 201, 218; regime, failure, 214, 219; regime, military, 17–18, 85, 90–91, 96, 99, 109, 115, 175–182, 184, 187, 189, 191–196, 206, 221–222; regime, monarchic, 203; regime, multi-party, 129; regime, single-party, 175 rent, 70, 139, 141, 207, 208; rent distribution, 70, 141 rentier state, 207 repression, 22–27, 30–38, 43–46, 49–55, 61–70, 89, 93–94, 100–107, 113, 122–123, 132, 138–140, 149, 159– 160, 168–171, 179, 190, 195, 200– 202, 211–214, 218–222, 224–226; repression, domestic, 100–101; repression, excessive, 51; repression, governmental, 46, 52; legitimacyrepression nexus, 200, 202, 213; repression, lethal, 45, 50, 52–53, 61–61. 65, 68–70, 218–219; repression, non-lethal, 52, 61, 62, 65, 68–69; repression, preemptive, 93; repression, private, 54–55; repression, proactive, 140; repression, reactive, 140; repression, selective, 23– 24, 27, 30, 38; repression, soft, 107; Resident District Commission (RDC) of Uganda, 127–128 Resistance Councils (RC) of Uganda, 116, 122 retirement, 75, 79–81, 83 revolution/color(ed) revolution, 3, 9, 17, 21, 24, 30, 33–34, 45, 48–49, 51, 78, 80–85, 101–109, 135–139,

267

141–142, 145, 167, 185, 193, 201– 204, 212–214, 220, 222; Rose Revolution, 103, 107–109, 136 Revolutionary Guard (Iran), 222 rupture, 3, 6–7, 9, 14–15 rural bias, 13, 112, 132 Russia, 17, 47–48, 84, 86, 135, 137, 140, 142–151, 182–183

Saif-al-Islam, 208, 210 scope conditions, 25, 112, 115 self-censoring, 16, 123 Shanghai Cooperation Organization (SCO), 18, 136, 144 Shevardnadze, Eduard, 103, 107–109, 136 shock, 9, 46, 185, 202, 221, 223, 226, 228 Soviet Union, 28, 137, 143–144, 155–156, 164, 181, 223, 225; Soviet Union, Former, 155, 164, 181, 223, 225 splits, 6–7, 11, 15, 65, 99, 104, 122 Ssemogere, Paul, 117–118 Stalin, Joseph, 26 Stasi, 23, 26–28, 30, 32, 34 state capacity, 157–160, 169, 225 surveillance, 24, 105, 224 survival, 3, 26, 45, 50, 53, 89, 91–92, 98, 101, 102, 106, 109, 135, 138, 140, 142, 175– 180, 195, 206, 218, 220, 223 sustainability, 14, 158, 170 Syria, 2, 51, 56, 98

Tajikistan, 17, 96, 135, 137, 144–145, 148–150 term limits, 17, 75–76, 78–84, 119, 221, 225 terrorism, 148–150 The Democratic Alliance, 125 Tiananmen (Square), 22, 24–25, 31–35, 38, 52, 225–226 transformative events, 18, 201 transition, 33, 46, 81, 99, 107, 112, 115, 119, 168, 171–172, 181, 183, 190–191, 210, 225 transparency, 14, 171 Trujillo, Rafael, 84 Tunisia, 56, 202–203, 205–210, 213–215

268

Index

Tunisian General Labor Union (UGTT), 205 Turkmenistan, 135, 149–150

Uganda, 12–13, 17, 111–113, 115–128, 130–132, 217, 219, 222–225, 227 Uganda People’s Congress (UPC), 116, 120–121, 124–125, 129 Ukraine, 2, 135–136, 142, 150 uncertainty, 43, 46, 50, 111, 114, 119– 120, 125, 127, 129, 131–132, 193 Uruguay, 175 Uzbekistan, 17, 135, 137, 140, 144–145, 148, 150, 219

Vargas, Getulio , 187 Velasco, Juan, 182, 192–194 violence, 51–55, 76–77, 83–86,, 103, 116, 119, 123–124, 132, 137, 148– 149, 192, 194, 213, 219, 221 vulnerability, 1, 14, 184

World Trade Organization (WTO), 143 Xi Jinping, 75, 79. 81, 170

Yanukovych, Viktor, 142 Yemen, 98

About the Book

What makes autocratic regimes vulnerable? Why, in times of crisis, do some of these regimes break down while others weather the storm? This is the puzzle addressed in Crisis in Autocratic Regimes. Taking a longterm perspective, the authors focus not on sudden shocks and ruptures, but instead on gradual processes of disintegration as they unfold over time. Johannes Gerschewski is lecturer in the Department of Social Sciences at Humboldt University of Berlin. Christoph H. Stefes is associate professor of comparative politics at the University of Colorado Denver.

269