Direct Democracy: A Double-Edged Sword 9781626377189

Citation preview

DIRECT DEMOCRACY

DIRECT DEMOCRACY A DOUBLE-EDGED SWORD

S HAUNA R EILLY

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 Names: Reilly, Shauna, author. 1980– Title: Direct democracy : a double-edged sword / by Shauna Reilly. Description: Boulder, Colorado : Lynne Rienner Publishers, Inc., [2018] | Includes bibliographical references. Identifiers: LCCN 2017043600 | ISBN 978-1-62637-714-1 (hc : alk. paper) Subjects: LCSH: Direct democracy. | Referendum. | Political participation. Classification: LCC JC423 .R333 2018 | DDC 321.8—dc23 LC record available at https://lccn.loc.gov/2017043600

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

For my father— Thanks for giving me the world

The full and complete definition of a citizen is confined to those who participate in the governing power, either by themselves or their representatives. This privilege is not attached to mere residence in a place or country, or derived from descent. . . . It is a privilege conferred in a legal manner by the act of the state. Sometimes it has been greatly extended by a revolution in the government . . . and those on whom power was then conferred retained it. He who is entrusted with this privilege in a democratic government is, more than others, invested with the power of a citizen. —Aristotle, A Treatise on Citizenship

Contents

List of Tables Acknowledgments

xi xiii

1 Power and Manipulation in Direct Democracy 2 Referenda: Rare, but Powerful

1

11

3 Recall Elections: A Mixed Power

37

5 Federalism: Power Struggles

91

4 Initiatives: True Citizen Power or Facades?

63

6 Authoritarian Governments: Limiting Citizen Power

123

8 The Power of Voters: Creating and Rejecting Peace

165

Bibliography Index About the Book

185 203 209

7 The Power of Words: Secession Referenda 9 What’s Next?

ix

141 181

Tables

2.1

2.2 2.3 2.4 2.5 2.6 3.1 3.2 3.3 3.4 4.1

4.2

4.3 4.4 4.5 5.1 5.2 5.3 6.1

Countries that Use Referenda at the National Level by Region Uruguay Algeria Hungary Bangladesh Uganda Countries that Use Recalls Countries that Allow Recalls US States Using Recalls US State Requirements Usage of Citizens’ Initiatives at the National Level Countries that Allow Initiatives at Regional and Local Levels Italian Popular Referenda New Zealand Initiatives German State-Level Laws on Initiatives Provincial Referenda National Referenda in the Philippines The European Union and National Referenda Egyptian Referenda

xi

13 18 22 27 29 33 40 42 46 47

64

65 70 76 80 102 107 116 128

xii

6.2 6.3 7.1 8.1

Tables

Venezuelan Referenda Kyrgyz Referenda Secession Referenda, 1846−2014 The Cyprus Vote

132 135 143 175

Acknowledgments

support and encouragement from family, friends, and colleagues enabled me to complete the project. My father talked me off the ledge whenever things were not going well. Shamima Ahmed, my former department chair, provided support and motivation to finish. Samra Pilav and Caroline Braden were epic cheerleaders every morning in the office and kept me sane all summer long. The breadth, completion, and quality of the book were made better by a number of colleagues. Jeffrey Zimmerman assisted with the theoretical development of the proposal and provided several title ideas. Tonya Krouse participated in biweekly writing check-ins to make sure that I was on track. Richard N. Engstrom supported the initial idea and assisted with case study selections. Discussants at the annual meetings of the Ohio Association of Economists and Political Scientists, Kentucky Political Science Association, and Southwest Political Science Association provided valuable feedback on individual chapters to ensure quality and thoroughness. Finally, the professional development fund of the College of Arts and Sciences at Northern Kentucky University provided financial assistance for me to travel to those conferences.

The cliché “it takes a village” certainly applies to this book. The

xiii

1 Power and Manipulation in Direct Democracy

create laws and make the government accountable, but they often do so with limited information and by being manipulated by various stakeholders. This leads to some important questions about how direct democracy is utilized around the world. Is direct democracy as democratic as it seems? Do citizens really have power in the process, or are they just being manipulated? Several notable examples of direct democracy in 2016 and 2017 have illustrated the significance of these mechanisms in terms of independence (Scotland and Great Britain), constitutions (Thailand), and peace agreements (Colombia). However, is the power actually in the hands of the people, or is this a facade? In this book, I evaluate the power and manipulation of citizens in the direct democracy process. By exploring the power relationships involved, I demonstrate the advantages and disadvantages of this process and its impact on citizens. Driven by case studies, I examine the complexity of direct democracy, its motivations, and the repercussions of how it is used around the world. Direct democracy is a ubiquitous term in today’s governance. It promises to make the voters extraordinarily powerful and the government impotent. This promise is not uniformly enacted, as each country has ways of empowering and limiting citizens through this Direct democracy is a double-edged sword: voters can directly

1

2

Direct Democracy: A Double-Edged Sword

process—sometimes simultaneously. There are safeguards in the mechanisms that prevent voters from having too much power (e.g., signature requirements, timing issues, and financing), but those can come at the price of true direct democracy. Questions that develop through the study of direct democracy across the globe include how much power citizens should have through this process and whether governments are providing safeguards or are actually manipulating voters. While this book does not completely address a normative perspective, it sheds light on the second question. The use of direct democracy has declined in some ways since the 1990s when it was at its peak. Governments and the people saw the challenges that came with the high numbers of referenda (both citizen and government initiated). As such, there has been a small decline but steady use of direct democracy around the globe. Some countries are still moving toward embracing this type of election while others are continuing to limit these opportunities to influence policy (Qvortrup 2017). Certain aspects have contributed to an increased use of direct democracy in some countries—political developments have increased pressure on governments and, as a result, they are more willing to hold constitutional referenda. What do I mean by direct democracy? Direct democracy, in its most primitive forms, includes referenda (legislative and popular), initiatives (agenda and citizen), and recall elections. These are all designed to give power to voters in different ways and to different extents. Voters have the ability to vote on legislative proposals, propose new laws or ideas, and remove officials from office prior to the end of their terms. These are immensely powerful tools and, as I demonstrate in the coming chapters, can be used to empower citizens or manipulate them, creating a facade of democracy. Arguments Surrounding Direct Democracy Advocates of direct democracy, like myself, believe that it is an important part of the democratic process. They discuss issues surrounding how voters are educated through the process (D. A. Smith and Tolbert 2009), accountability, and protection against tyranny (Matsusaka 2004). Furthermore, there are cues to assist voters with making informed choices in the voting booth and extensive campaigns to provide additional details (Lupia 1994a). There are also

Power and Manipulation in Direct Democracy

3

those who talk about saliency and how ballot measures can increase turnout for other races1 (Childers and Binder 2012; Donovan et al. 2005; Tolbert, Grummel, and Smith 2001; Dubois 1979). Another benefit of direct democracy is the increase in civic engagement (Smith 2002) that comes from using these mechanisms, as voters have the ability to engage directly in changing policies as well as to participate in campaigns surrounding these elections. Additionally, citizens who are frequently exposed to direct democracy mechanisms have higher levels of efficacy and are more positive about their government (Bowler and Donovan 2002a). Despite the benefits of direct democracy, there are also perceived drawbacks to these processes. One of the most prevalent discussions is about uninformed voters (Lupia 1994a, 2001; Nicholson 2003, 2005). Voters are often unaware of larger up-ticket races, so how can they be knowledgeable enough about policies to make laws? Another issue is that many of these issues are large in scale or extremely technical; are voters capable of making these choices? One only needs to look at tax implications from initiatives in the United States or the most recent Brexit referendum in Britain to see that the impact of these decisions can reach far beyond the understanding of typical voters (and maybe even politicians). Likewise, many referenda (popular or legislative) deal with election designs— these are complicated issues that voters may be unable to grasp, even with a detailed campaign and voter information distributed. The media reports following the British vote show voter remorse. So, we must question the wisdom of putting these types of decisions into the hands of voters. Voters can be manipulated in elections (particularly in direct democracy elections) by campaigns, spending, or even the questions posed to them. Furthermore, these drawbacks can be exacerbated by fear campaigns designed to play on the weaknesses of voters (Gastil, Reedy, and Wells 2007; Gerber 2001). Many ballot questions deal with issues of discrimination where the question is designed to advantage one group over another; there are also issues of morality and civil rights that adversely effect minority citizens (Haider-Markel, Querze, and Lindaman 2007; Smith, DeSantis, and Kassel 2006; Donovan et al. 2005; Jackman 2004; Hajnal, Gerber, and Louch 2002). Voters are also limited by a number of procedural issues or regulations and government agents. Moreover, with judicial review of ballot measures and laws proposed through direct democracy,

4

Direct Democracy: A Double-Edged Sword

many are limited or overturned by the courts (Manweller 2005; Miller 2009). Language complexity issues (Reilly 2010; Reilly and Richey 2009; Magleby 1984) further complicate these elections. Ballot language is not always clear, which further obfuscates the meaning of the questions posed. Ideally, the campaign will clarify issues for voters, but there have been several instances in which ballot language was intentionally complex and made it challenging for voters to make good decisions. The prevalence of these elections can result in voter fatigue and roll-off (Bowler, Donovan, and Happ 1992). The Swiss vote multiple times a year with several questions on each ballot. Long ballots and substantial numbers of campaigns can distract voters and politicians alike. Voters can be overwhelmed by multiple elections, and they are likely to vote less often or not vote down ballot when there are frequent elections.2 Politicians are also affected by frequent direct democracy elections because they not only have to adjust the laws proposed by direct democracy measures, they must also actively participate in campaigns for and against these measures. In the end, recall elections can disrupt the governing process by interrupting a politician’s term with additional elections. Recent news coverage indicates that some voters question the truly democratic nature of direct democracy (Taub and Fisher 2016). There is strong evidence that money plays a role in the direct democracy process (see Broder 2000; Zisk 1987; Lowenstein 1982). Campaigns are expensive and require funding sources, which often have a stake in the outcomes of these elections. For example, a referendum on an environmental issue may have proponents who want to save the environment, so they will spend money to assist in passing that referendum. But there may also be industries that would be hurt by the environmental measure, so they will spend money to defeat it. This influx of money muddies the issues surrounding the measure—it creates an information environment, but savvy voters will need to spend time determining the true impact and need for such a measure. Money influences voters and complicates the direct democracy process. One of the largest issues facing direct democracy is the role of special interests. Special interests can encapsulate different groups, including the military, businesses, opposition groups, and moneyed interests. Throughout this book, I illustrate where special interests have pervaded every

Power and Manipulation in Direct Democracy

5

aspect of direct democracy. Whether it is to oust leaders in countries like Venezuela or to directly propose ballot initiatives in New Zealand, these special interests are active in the direct democracy process. This has made people distrustful of the mechanisms and, thus, less likely to participate and engage. These special interests channel money into the process and, consequently, can be a large part of the campaign to pass or defeat different measures. Regional Variations Regional variations are important to note not only because of the forthcoming case selections, but also because of their impacts on the use and success of referenda. When looking at the motivation behind direct democracy methods, there are a number of explanations for why countries and regions explore these mechanisms. One reason for this development is the failure of traditional electoral systems to create a representative government that is responsive to the people. Here, the motivation comes from the people attempting to get more access and influence in government (Barczak 2001). Now, these neopopulist approaches can develop from excluded interests gaining power (as experienced in parts of Western Europe and Latin America) or from the people (as in Egypt 2012). Although the outcome in some elections is not always predictable, referenda in Africa are often merely symbolic. Failed referenda are rare in Africa, as these are often state sponsored, and there are clear media signals about the intent and expected vote. Additionally, referenda in Africa typically are used only for constitutional manners and are package deals that include many different issues. Thus, in emerging and decaying democracies, it is challenging for voters to participate and not pass all components, even when many are unpopular, because of the value of some of those added to the package; for example, Zimbabwe 2000 and Venezuela 2008 (Kersting 2014). The African experience with direct democracy is seen as a relatively new phenomenon, but is actually a reflection of the decolonization process of the 1960s. Here, direct democracy (mainly in the form of referenda at the national level) was a culmination of the independence process, but it was reignited in the 1990s as these governments moved toward multipartism (Tripp 2004). Internal and

6

Direct Democracy: A Double-Edged Sword

external influences led to these developments in both experiences (Kersting 2009). A more modern factor in the development of direct democracy at the state and local levels is an intent to create conflict resolution, accountability, and stability. This type of plebiscite usage is fairly rare, but holds important promise in the region (Kersting 2009). These mechanisms are not used consistently across countries or even across political leaders in a single country. Regimes

Thinking about the context of a political regime and its desire to utilize direct democracy is important to understanding the complexity of the process. In some contexts (e.g., the former Soviet Union), the referendum can be dominated by the executive branch, which weakens the already fragile democracy in place and threatens the success of the proposition and the long-term legitimacy of the use of these mechanisms (Bâişanu 2010). Thus, a system designed to further deepen democracy can actually be threatened by the use of these manipulative elections. There have also been arguments that referenda serve as constraining instruments on government (Dicey 1890, 1911; Qvortrup 1999)—specifically, that referenda serve as a check and balance on government and that they limit governments from going above their constitutional roles (Dicey 1911). Direct democracy in establishing a national identity and in state building can legitimize the new institutions within a country (Kersting 2009, 2014). Also, direct democracy is a way to create inclusive decisionmaking because it allows all factions within a state to have a say in the outcome of a particular law or amendment (Kersting 2009, 2014). Finally, the power building of authorities can be strengthened and extended through direct democracy (Rourke, Hiskes, and Zirakzadeh 1992; Kersting 2009, 2014). Those who believe in elite institutions are also more likely to use direct democracy because of a concern regarding the power structure (Bjorklund 1982). These elites have a different set of problems in newly democratic countries. In Latin America, after the Cold War ended, competitive authoritarianism developed with charismatic leaders. These leaders had strong populist ties during these countries’ development, which contributed to the growth of direct democracy mechanisms in the Latin American west coast. These referendum mechanisms created a threat to a culture of horizontal

Power and Manipulation in Direct Democracy

7

accountability between the government and elites (Levitsky and Loxton 2013). As a result, many of the weak democracies that were only paying lip service to democratic processes slid into competitive authoritarianism. Developing and Developed States

Most developing countries use referenda sparingly and focus on the basic building blocks of economic and democratic development (Frey 2003). These countries have a different perspective on the use of referenda. They are forging new democracies and can become electoral authoritarian regimes under very specific circumstances, which leads to my theory on authoritarian usage of direct democracy. Many counties limited the use of referenda after World War I—this was particularly true in the Weimar Republic, Austria, and Czechoslovakia. The political class’s fear in many of these countries was of the expression of popular will (Băişanu 2010). This, of course, led to other issues (i.e., the rise of Adolf Hitler and World War II), but it also demonstrated the unwillingness of the political class to be at the mercy of public will. There are several reasons why individual political leaders are less than excited about the use of referenda—namely, because of the personal leadership costs. For example, Charles de Gaulle took a number of political and personal risks during France’s use of referenda between 1958 and 1969. Another example is Canadian prime minister Brian Mulroney who presented a national referendum that failed in 1992, which resulted in major political losses for Mulroney and his party. Developing and developed states have different experiences with direct democracy, and they utilize these mechanisms to different ends. Developing and decaying democracies have increasingly utilized these mechanisms to create legitimacy, for state building, and even to appear more democratic than they are. Direct democracy can be used to establish legitimacy for existing regimes by asking for support for the regime (e.g., the Venezuela and Bangladesh case studies discussed in the book). Furthermore, referenda can be used in state building to create a notion of a united government or an established democracy. Finally, many countries include direct democracy mechanisms in their constitutions so as to appear democratic. However, these procedures can be suspended or have so many layers of barriers that it is almost impossible to move forward with them.

8

Direct Democracy: A Double-Edged Sword

Thus, while some countries appear to have these mechanisms, they do not have them in actuality. The case studies I included demonstrate the key differences between established democracies and those that are developing or decaying. Direct democracy is not created equal, and scholars have identified different types of ethnonational referenda (Wildavsky 1973; McGarry and O’Leary 1993; O’Leary 2001; Qvortrup 2014b). Under the typology of ethnonational referenda (held in divided societies), those involved in the separation or division of a country take significant attention, referred to as secession referenda (e.g., Scotland 2015 and Quebec 1995). During these types of referenda, there is a movement toward homogenizing referenda (Sheehy 1991). There are also right-sizing referenda that occur after a major conflict or regime change (this was common after World War I to determine where new borders would be established).3 Additional differenceeliminating referenda occur in authoritarian regimes to lend legitimacy to policies of ethnic or national homogenization (e.g., the Soviet Union 1990). Difference-managing referenda occur in countries where democratic procedures are established following longterm ethnic disagreements (e.g., Canada 1992 and Northern Ireland 1998) (Qvortrup 2014b).4 Challenges of Direct Democracy Around the World As explored in later chapters, direct democracy is susceptible to being used by politicians to serve their own interests and not just those of the people. Some have argued for protecting the state by limiting the reach of direct democracy through restricting the number of topics or making referenda difficult to initiate. There can be fiscal or long-term policy implications through referenda that have real effects on a state. Hence, making these mechanisms more insulated from the public does have merits, not only by changing their ease of use but also by changing the signature quotas or participation thresholds. My goal is to provide a snapshot of how direct democracy is utilized globally to demonstrate the power and manipulation within the process. There are a number of variations to the usage of direct democracy. In each chapter, I highlight countries that utilize the process frequently and those that have limited experience with the mechanism. There also are variations in how much power these mechanisms have in

Power and Manipulation in Direct Democracy

9

each country—some are simply advisory, others are binding. And there are differences in how easy these mechanisms are to use as well as in terms of signature requirements, time requirements, intimidation, government involvement, and other barriers. Looking at direct democracy from a comparative perspective is not a new area of research in political science. However, looking at it from an authoritarian perspective and evaluating this type of direct democracy from a global perspective is a rarity. The literature frequently focuses on individual countries or regions to address problems and issues with direct democracy. This book is different than existing literature due to its more global approach that examines regimes and their use of direct democracy. Eighty-four countries worldwide utilize some form of direct democracy in their constitutions, and 74 explicitly prevent the use of such measures, demonstrating that this is not a phenomenon with equal use across the globe. Thus, states that use these measures do so for specific reasons. Plan of the Book In this book, I evaluate the power of citizens to truly implement change and influence the governing of their respective jurisdictions through this process. In the first part, I look at the different types of direct democracy mechanisms—referenda, recalls, and initiatives—through specific case studies around the world to illustrate the variations in the utilization of these mechanisms. In Chapter 2, I focus on referendum usages and whether usage rates influence the value and power of these mechanisms. In Chapter 3, I discuss recalls and how they are used as a check and balance on political leaders during their terms; however, I also look into the dark side of recalls, where tyrannical leaders are buoyed by their survival of a recall and increase their authoritarian activities as a result. In Chapter 4, I examine initiatives and how citizens work together to propose and pass legislation through direct democracy mechanisms. This valuable power of citizens is often seemingly co-opted by interests, or the power is limited by the state to the point where this mechanism is out of reach for most citizens. In Chapter 5, I describe how direct democracy is used at different levels of government and in combination. In the last section, I discuss these mechanisms in different contexts to provide a fascinating depiction of direct democracy in authoritarian regimes

10

Direct Democracy: A Double-Edged Sword

(Chapter 6), state breakdowns (Chapter 7), and peacebuilding (Chapter 8). I conclude by making theoretical ties between the chapters and proposing suggestions for future work. Notes 1. Although this is clearly manipulated in some struggling democracies. 2. “Frequent” is also in the eye of the beholder—for some voters, elections every year can seem overwhelming. Others may be overwhelmed by elections more often than every 4 years. 3. Some influences on right-sizing referenda include ethnic fractionalization, the end of war or conflict, gross domestic product (GDP), and changes to the international system. 4. Difference-managing referenda have two forms: devolution referenda, in which the power is delegated to smaller territories, and power-sharing referenda, in which agreements are made on consociational power sharing (e.g., Northern Ireland 1998).

2 Referenda: Rare, but Powerful

influence for citizens in a democracy. Their use provides an important role for citizens to give feedback to and get accountability from their governments. By using referenda on a national scale to determine policies, states have the opportunity to involve citizens in national decisionmaking. When used, these referenda have significant repercussions on government structures, electoral systems, and government ownership. Nationwide referenda are rare, and there is variance in how they are used: regularly, sporadically, or not yet but with the ability to be used in the future (Butler and Ranney 1978). When used, these referenda have significant repercussions on government structure, electoral systems, and government ownership. Referenda are a pervasive part of political systems at the local and state levels, particularly in Switzerland and the United States.1 As of this writing, 152 countries have utilized national referenda for varying purposes and to varying degrees. I focus on five case studies (Uruguay, Algeria, Hungary, Bangladesh, and Uganda) in the context of global usage to demonstrate the power of these referenda. As a result of their infrequent use, there remains a gap in the literature on the power of national referenda and whether governments are truly seeking voter input or manipulating voters to assent to the government’s decisions. Some studies focus on specific countries The use of referenda on a national scale is a primary form of

11

12

Direct Democracy: A Double-Edged Sword

or even state and local uses of referenda (Bowler and Donovan 1998; Bowler, Donovan, and Tolbert 1998). There have also been notable uses of national referenda among countries joining the European Union (EU) (De Vreese and Semetko 2004; Hug and Sciarini 2000). The role that national referenda have played during integration into the EU is particularly notable and dates back to the 1990s in some countries (Denmark, France, and Ireland), as examined in Chapter 5. However, as mentioned above, 152 countries have utilized national referenda for varying purposes and to varying degrees (see Table 2.1). Why Use Referenda Nationally? Referenda are items called by political authorities (at either the executive or legislative level) to have a vote on a specific issue. This authority is traditionally granted by the constitution, but can also be a function of a specific constitutional authority (Beramendi 2008). There may also be required or mandatory referenda on specific issues such as a policy decision that raises a constitutional question. Optional referenda to get voter feedback may also be used for questions not involving the constitution when it might be politically beneficial, such as when the president’s party is a minority in the legislature or when the topic is one that deeply divides the country. The binding nature of these referenda varies by state. While considered a modern form of governing, national referenda date back to the French Revolution and have been used regularly in Switzerland since 1848. A number of countries have the constitutional ability to conduct referenda on a national level (see Table 2.1); however, far fewer use this power regularly. One of the challenges with national referenda is that, because they are used so infrequently, it is almost impossible to make universal comments or theories about their nature and impact (Butler and Ranney 1994). However, there are shared experiences and categories of referenda that can assist our understanding of ballot measures, and there is something unique about how and why countries use this mechanism and the power granted to voters through referenda. Each country has its own procedures, expectations, and requirements. There are also varying types of national referenda—on constitutional, territorial, moral, and other topics too varied to categorize

Table 2.1 Countries that Use Referenda at the National Level by Region Africa

Algeria Morocco Comoros Madagascar Togo Congo Equatorial Guinea Niger Benin Burkina Faso Burundi Central African Republic Ghana Liberia Sudan Botswana Cameroon Chad Djibouti Gambia Guinea Senegal Somalia South Africa

Americas

Uruguay Guatemala Ecuador Panama Chile Colombia Bolivia Brazil Peru Canada Costa Rica Bermuda Cuba Paraguay Saint Kitts and Nevis Argentina Bahamas Honduras Suriname Antigua and Aruba Dominica El Salvador Grenada Nicaragua

Asia

Philippines Maldives Kyrgyzstan Korea Azerbaijan Iran Uzbekistan Bangladesh Tajikistan Kazakhstan Pakistan Taiwan Turkmenistan Mongolia Singapore Sri Lanka Thailand Vietnam Afghanistan Japan East Timor

Europe

Switzerland Italy Liechtenstein France Ireland Denmark Slovenia Romania Estonia Greece Latvia Lithuania Sweden Norway Slovakia Spain Andorra Hungary Iceland Poland Armenia Malta Portugal Turkey

Middle East

Egypt Syria Iraq Yemen Bahrain Qatar

Oceania

New Zealand Australia Cook Islands Marshall Islands Niue Samoa Kiribati East Timor Nauru Vanuatu

13

continues

Africa

Côte d’Ivoire Mali Mauritania Namibia Rwanda Seychelles Sierra Leone Swaziland Uganda Ethiopia Gabon Malawi São Tomé and Príncipe Tunisia Zambia Angola Cape Verde Lesotho Mauritius Mozambique Niger

Americas

Saint Lucia Saint Vincent and the Grenadines

Asia

Europe

Albania Austria Belarus Bulgaria Cyprus (North) Georgia Gibraltar Luxembourg Ukraine Finland Macedonia Moldova Russian Federation Croatia Czech Republic Montenegro Netherlands Serbia United Kingdom Jersey

Middle East

Oceania

Source: Beramendi 2008. Note: Countries listed in descending order according to usage of national referenda. Classifications of regions and usage determined by source.

14

Table 2.1 Continued

Referenda: Rare, but Powerful

15

together—that make it challenging to study these elections. For constitutional changes, these may reflect an ever evolving experience with globalization and changes that must be made to reflect new international standards or needs from within the state. Territorial referenda involve changes to borders or joining larger unions; in previous experiences, these have looked at European integration or at joining larger alliances such as the North Atlantic Treaty Organization (NATO). Another surprising area of national referenda that has not been considered in the literature is referenda on the future of state-owned enterprises. As discussed in the case studies below, a remarkable number of referenda focus on government ownership. The constitutional referendum, in particular, holds an important role in the political process. This is where voters have the opportunity to directly influence the constitution as well as create legal frameworks for the country. This subset of referenda is paramount to nation building and reflective of the state’s plan and creates a second level of representation in the decision making process for voters. They come with additional challenges regarding the role of citizens in establishing constitutional law, and the stakes of these referenda are high for governments. These types of referenda are more susceptible to elite manipulation and are less about the will of the people. They also address issues of sovereignty (Tierney 2009) (discussed further in Chapter 7). Demand for Referenda One of the issues about referenda is when they are used. There are two reasons why referenda are traditionally used: as a minority tool and by those with similar ideologies (Bjorklund 1982). Referenda as a minority tool only work in specific contexts: when political parties are split on the issue and when there is a strong sentiment among voters on a particular issue. Thus, referenda can either be used to mediate an issue or as a precipitous moment to make changes (Bjorklund 1982; Butler and Ranney 1978). Each referendum has its own special history, context, and motivation (Butler and Ranney 1978). However, there are some shared features among the use of referenda, and researchers have common focuses when looking for patterns in the demand for referenda. Tor

16

Direct Democracy: A Double-Edged Sword

Bjorklund (1982) investigates this by looking at two types of referenda (specifically in Norway). First, general referenda put forward vaguer issues such as political rights and formalizing these issues into the constitution. There are also specific issue referenda that are submitted to the public to determine the outcome, which transfers the responsibility from the legislature or executive to the people. This vagueness may also be a function of political divisions in those institutions and an inability to pass legislation on specific issues (Bjorklund 1982). Although these may be limited to Norway, they present an opportunity to investigate national referenda. Referenda on nationalism and separatism are much more common than one might think. In American history, there have been notable examples of these types of movements and referenda in Massachusetts (1788), Texas (1861), Virginia (1861), and Tennessee (1861) (Mattern 1921; Lee 1981; Fisch 2006). Referenda were also held in Canada (Nova Scotia), South Africa (Natal), and Australia about their future in the British Empire (Qvortrup 2014b). However, the number of ethnonational referenda decreased after World War I (Qvortrup 2014b), save for some particular referenda such as the Soviet Union’s referendum on the future of the union and whether to dissolve in 1991. These are examined further in Chapter 7. Voting Beyond the activities of the referenda themselves, these referenda have impacts on voting behaviors, political parties, and even particular leaders. When it comes to voting, issues of party identification and ideology transfer over to referendum voting (De Vreese and Semetko 2004). The issues facing referenda are rarely as clear as the question posed, but issues of economics, government popularity, regional contexts, political knowledge, and fear plague the process (De Vreese and Semetko 2004; De Vreese 2004). These elections are rarely held in a vacuum (Nicholson 2005), and many other issues can influence voting choices and election outcomes. The referenda vote often calls on issues beyond those on the ballot; for example, they can be issues of economics (Clarke and Kornberg 1994), issues of political leadership (e.g., those in Venezuela in 2008 and 2009), and other national issues (De Vreese

Referenda: Rare, but Powerful

17

2004). Influencing factors can also change during the campaign, including cynicism, campaign strategy, and media impact (De Vreese and Semetko 2002; De Vreese 2004). Case Studies I selected five case studies for this chapter: Uruguay, Algeria, Hungary, Bangladesh, and Uganda. These five represent different experiences with referenda as well as different regions of the world. These cases have had different historical experiences (be they colonial histories or civil war). They have different rules, regulations, and thresholds guiding their direct democracy process. Some have used national referenda multiple times, and others have used them infrequently. These cases also differ dramatically on the topic of the referenda—from colonial constitutions to limiting government monopolies. Although not chosen at random, they provide a thorough depiction of the larger experience of national referenda around the world. Uruguay

Uruguay is open to referenda and poses no form of regulation, providing substantial freedom for its residents (Beramendi 2008). The constitution of Uruguay provides for the use of referenda at the national and subnational levels and may be used to repeal or abrogate laws. The most common terms include popular legislative initiative (iniciativa popular legislativa); referendum on constitutional matters (plebiscito), or the more direct translation, popular consultation (consulta popular); recall (revocatoria de mandato); and open town meeting (cabildo abierto) (Beramendi 2008). A referendum in Uruguay allows voters to ratify or reject a duly approved law within one year of its coming into force. A referendum can also be used to oppose a decree issued by a subnational assembly such as the Junta Departamental. The plebiscito (as mentioned above) is the procedure by which citizens approve or reject a proposed constitutional reform (Altman 2011). Direct democracy has been used in Uruguay since 1917 and employed 21 times, with multiple elections on the same ballot (see Table 2.2). Direct democracy mechanisms were in place prior to the

18

Direct Democracy: A Double-Edged Sword

full democratic consolidation in Uruguay. In 1980, the military regime drew up a charter that would have provided for a strong continuing military role along the lines of the 1976 constitutional decrees, including by legitimizing the new role of the National Security Council (Cosena, or the Consejo de Seguridad Nacional). The document would have also greatly reduced the roles of the General Assembly and the political parties. A plebiscito held on November 30, 1980, however, rejected the new military-drafted constitution (Altman 2011). Table 2.2 Uruguay Ballot Name

Uruguayan Constitutional Referendum Uruguayan Constitutional Referendum Uruguayan Constitutional Referendum

Uruguayan Constitutional Referendum Uruguayan Constitutional Referendum Uruguayan Constitutional Referendum Uruguayan Constitutional Referendum Uruguayan Constitutional Referendum

Summary of Language

Constitution and establishment of electoral procedures and laws. Redesign of the constitution, adding judiciary, social rights, and initiatives. Reform 1: Recognition of political parties and reorganization of the senate. Reform 2: One candidate per political party and reorganize local government. Constitutional redesign, eliminate political parties, and electoral changes. Initiative 1: Unbundling of election dates. Initiative 2: Separate election of president and vice president. Constitutional reforms and election law changes.

Creation of a presidential system, with separate election dates for president and parliament, and no political parties. Unión Cívica initiative for a presidential system. Authorizes initiative to enact a presidential system.

Results

Date

For

1934

228,145 (95.75%)

1917

1938

84,992 (95.15%)

338,802 (93.45%)

333,802 (97.99%)

1942 1946

1951 1958

1962

443,414 (77.17%)

289,101 (43.14%) 252,353 (37.65%)

232,076 (54.0%)

Against

Turnout

10,124 (4.25%)

56.3%

4,330 (4.85%)

38.2%

23,385 (6.55%)

6,847 (2.01%) 131,163 (22.83%)

66.9%

128,775 (19.21%)

67.4%

197,684 (46.0%)

37.1%

235,941 (23.04%)

634,080 (61.95%)

37.1%

153,662 (15.01%) 195,623 (16.7%)

975,397 (83.3%)

76.7% continues

19 Table 2.2 Continued Ballot Name

Uruguayan Constitutional Referendum

Uruguayan Referendum

Uruguayan Constitutional Referendum Uruguayan Amnesty Referendum Uruguayan Constitutional Referendum Uruguayan Privatization Referendum Uruguayan Constitutional Referendum Uruguayan Referendum

Summary of Language

Date

Reforma Naranja (Orange): 1966 Change to the government structure and electoral changes. Reforma Amarilla (Yellow): Indirect election of president, no political parties, and pension amounts for legislators. Reforma Gris (Gray): Creation of presidential system, where the president can be reelected, may dissolve parliament and separate election time from parliament. Reforma Rosada (Pink): Presidential term limited, but can dissolve parliament. Held in conjunction with the 1971 general election.

Single immediate reelection of president and vice president (Article 152). Constitutional Council established 1980 by the military government, which would have given the military the Chilean model of much power and impunity. Attempt to repeal amnesty for April former military juntas. 1989 Use of social security funds.

The referendum committees decide on the fast-track procedure to provide for the privatization of certain state-owned enterprises. Splitting voting procedures and times for president, parliament, and governors. Constitutional amendment to provide ages for local deputies and instituted payment mode of pensions. Pensions

Education fees

Nov. 1989

For

Results

786,987 (64.89%) 175,095 (14.2%)

Against

182,345 (14.8%)

Turnout 74.3%

86,315 (7.0%)

1,020 (0.08%)

Immediate reelection 491,680 1,248,439 (28.25%) (71.75%) Presidential resignation 1,870 1,738,249 (0.10%) (99.9%) 707,118 945,176 (42.8%) (57.2%)

Retain 1,082,508 (55.95%) 1,681,592 (81.78%)

Repeal 799,109 (41.3%) 374,763 (18.22%)

92.7%

86.9%

87.9%

88.7%

1992

1,293,016 (71.55%)

489,302 (27.45%)

82.8%

Aug. 1994

559,321 (31.08%)

1,240,059 (68.92%)

86.2%

1994 Nov

1,540,462 (72.3%) 694,351 (32.6%)

590,156 (27.7%) 1,436,267 (67.4%)

91.5%

continues

20

Direct Democracy: A Double-Edged Sword

Table 2.2 Continued Ballot Name

Uruguayan Constitutional Referendum Uruguayan Referendum Uruguayan ANCAP Referendum Uruguayan Constitutional Referendum Uruguayan Amnesty Revocation Referendum Uruguayan Voting from Abroad Referendum Uruguayan Constitutional Referendum

Summary of Language

Date

Election reforms

1996

Judiciary

1999

Directors of state-owned companies. Repeal of petroleum monopoly.

2003

The public service of sewerage will be served exclusively and directly by state legal persons. Removed the Law on the Expiration of the Punitive Claims of the State, which granted amnesty for human rights abuses under the 1973–1985 dictatorship. Enable mail-in votes by citizens living outside Uruguay.

2004

Lowered the age of criminal responsibility from age 18 to age 16.

2014

2009

For

Results

1,015,828 (52.2%)

949,935 (43%) 839,004 (38.1%) 1,201,626 (63.72%)

1,440,006 (64.6%)

1,090,859 (47.36%)

Against

Turnout

1,254,746 (56.9%) 1,365,677 (61.9%) 684,129 (36.28%)

91.8%

1,212,477 (52.64%)

89.9%

930,288 (47.8%)

788,924 (35.4%)

850,691 (36.93%)

1,452,645 (63.07%)

1,110,283 (46.81%)

1,261,834 (53.19%)

86.2%

83.4% 89.6%

90.5%

Source: Altman 2014. Note: ANCAP is Administración Nacional de Combustibles, Alcoholes y Portland.

The vast majority of Uruguay’s referenda have focused on government redesign and election processes (Altman 2014). Uruguay has been prolific in its use of referenda, having started the process as early as 1917 and conducted 21 national referenda, with multiple referenda elections on some ballots. Part of this proliferation is the result of Uruguay’s military coup and government challenges. Uruguay has a long and active history with referenda. While each of the referenda has changed (or proposed) the government’s composition—everything from the presidential system to the senate structure, electoral processes, the role of political parties, and the creation and use of courts—they have also focused on other major issues facing the country.

Referenda: Rare, but Powerful

21

Uruguay held a referendum in 1989 on amnesty for the military and nongovernmental actors who had participated in the civil-military dictatorship between 1973 and 1985. A number of crimes were committed under this dictatorship, but the voters proposed and supported amnesty to forge peace and a new government. In 2009, a referendum granting amnesty for human rights abuses under the 1973– 1985 dictatorship failed at the ballot box, demonstrating a change in how the people viewed these issues (Altman 2011). Another notable referendum was the repeal of laws on public enterprises in 1992. Referenda in 1994 also looked at state spending on education and pensions. Uruguay revisited the issue again in 1996 and 2004. In 1996, Uruguay held a referendum to end the petroleum monopoly by the state-owned petroleum company Administración Nacional de Combustibles, Alcoholes y Portland (ANCAP), which the majority of voters approved. Furthermore, in 2004, a referendum was held about the ownership of the water supply and whether it should be owned by the state (Altman 2011). Uruguay offers a unique contribution to this study in that it is the most prolific user of national referenda among these cases. Uruguay has used referenda for three major issues: government (and election) redesign, amnesty, and public ownership and spending. This correlates with the uses of the three previous cases, but builds a solid foundation for deepening democracy in Uruguay. Algeria

Algeria first adopted direct democracy measures in 1961 when drafting a new constitution (Beramendi 2008). Direct democracy is used only at the national level through legal provisions for optional referenda; it is not used at the regional or local levels. Direct democracy results are always binding to the government, and both constitutional and other issues can be brought to a referendum (Beramendi 2008). The country uses referenda on a mandatory and optional basis.2 Referenda in Algeria have principally been used to deal with constitutional issues (see Table 2.3). Constitutional amendments are subject to mandatory referenda, but any amendment not considered by the Constitutional Council as infringing on rights may be enacted by the president without a referendum (Algerian Constitution, Article 176; Beramendi 2008). The president can submit to referendum amendments proposed by three-quarters of the members of both houses of Parliament (Algerian Constitution, Article 177; Beramendi 2008).

22 Table 2.3 Algeria Ballot Name

French Constitutional Referendum French Constitutional Referendum

Topic/Summary

Date

“Do you agree that the assembly 1945 now elected will serve as a constituent assembly?” Voters were asked whether they May approved of a new constitution 1946 proposed by the National Assembly elected in 1945. French Voters were asked whether they Oct. Constitutional approved of a new constitution 1946 Referendum proposed by the constituent assembly elected in June. French “Do you approve of the 1958 Constitutional Constitution offered by the Referendum Government of the Republic?” French “Do you approve the bill 1961 Referendum on submitted to the French people Algerian Selfby the President of the Republic Determination and concerning the selfdetermination of the populations of Algeria and the organization of the public authorities in Algeria prior to self-determination?” Algerian “Do you want Algeria to become 1962 Independence an independent state, co-operating Referendum with France under the conditions defined in the declarations of 19 March 1962?” Algerian New constitutional design. 1963 Constitutional Referendum Algerian National Charter committed the country June Charter to socialism. 1976 Referendum Algerian Constitutional Restoring direct election of Nov. Referendum the president. 1976 Algerian National Changes to the 1976 charter. 1986 Charter Referendum Algerian Changes to limit power of 1988 Constitutional ruling party. Referendum Algerian Removes references to socialism 1989 Constitutional from the constitution, and allows Referendum for a multiparty democracy.

For

Results

Against

313,425 (96.1%)

12,741 (3.9%)

117,378 (38.4%)

188,524 (61.6%)

175,049 (48.5%)

Turnout

185,992 (51.5%)

17,668,790 4,624,511 (79.3%) (20.7%) 1,749,969 (69.5%)

767,546 (60.5%)

5,795,581 (99.7%)

16,534 (0.28%)

91.9%

5,166,185 (98.01%)

82.7%

7,130,033 (98.4%)

105,047 (1.99%)

118,508 (1.6%)

7,407,626 (99.0%) 10,508,863 (98.3%)

67,683 (1.0%) 181,063 (1.7%)

92.9%

9,341,429 (92.3%)

7,290,760 (73.4%)

780,677 (7.7%)

2,637,678 (26.6%)

83.0% 79.0% continues

Referenda: Rare, but Powerful

23

Table 2.3 Continued Ballot Name

Algerian Constitutional Referendum Algerian Civil Concord Referendum

Topic/Summary

Amendments prohibit use of Islam and ethnic identity in domestic politics. Aimed at ending the civil war, and allowed Islamist fighters not guilty of rape to avoid prosecution if they turned themselves in. Algerian National “Are you for or against the Reconciliation Charter of Peace and National Referendum Reconciliation that the government has proposed?”

Date

1996 1999

2005

For

Results

Against

10,945,321 1,809,793 (85.8%) (14.2%)

Turnout 79.8%

14,583,075 202,496 (98.6%) (1.3%)

14,054,164 381,127 (97.36%) (2.64%)

Sources: Compiled by the author from Joffé 2008; Roussellier 2005; Van Walraven 2009.

As demonstrated in Table 2.3, there have been 15 referenda in Algeria. Some date back to 1945 when it was still a French colony. The first referenda in Algeria focused on wider French reforms in 1945, 1946 (twice), and 1958. Algeria, like other French colonies, participated in national referenda to confirm or reject French constitutions. 1961 marked a major change for Algeria, when the country held a referendum on Algerian self-determination (Nohlen and Stöver 2010). The 1961 referendum included voters in both France and Algeria who approved self-determination. There were varied reports of the actual turnout, with some government reports indicating a turnout of 92 percent while others suggested less than 60 percent. This was followed by a vote on independence in 1962. Algerians voted overwhelmingly in favor of independence (99.7 percent). In 1963, Algeria held its first independent referendum on a new constitution. It was approved with high turnout and approval. A referendum was used again in June 1976 to approve a national charter that committed the country to socialism—a response to many issues that plagued Algeria, such as severe poverty and crime (Nohlen, Krennerich, and Thibaut 1999). Later, in November 1976, Algeria held another national referendum to reestablish the People’s National Assembly, which had been suspended in 1965 during a coup. These, too, saw high turnouts and a large degree of support for the measures.

24

Direct Democracy: A Double-Edged Sword

Algeria made revisions to its constitution and charter through referenda in 1986 and 1988, which each had high approval ratings and high turnout. In February 1989, Algeria’s use of referenda changed as it moved toward multiparty democracy in a new constitutional referendum (Nohlen, Krennerich, and Thibaut 1999). The new constitution removed references to socialism and received heavy criticism from radical Islamists and opposition groups. This would change the power structure within the country in favor of the government vis-à-vis the people. Although approved, this marked a change in Algerian referenda because it was approved by only 73 percent of voters—much lower than previous referenda—and had a lower turnout than other national referendum elections. During the Algerian Civil War (beginning in 1991), 3 another referendum was held in 1996 to reduce the use of Islam and other ethnic identifications in domestic politics. Despite major opposition, the amendment was approved by a high margin of voters (86 percent) and had a higher level of turnout (nearly 80 percent) (Nohlen, Krennerich, and Thibaut 1999). This was followed by the Civil Concord Law referendum in 1999, which was held to end the Algerian Civil War. It included a provision that allowed Islamist fighters to surrender and avoid further prosecution (much like the amnesty referenda in Uruguay). This was supported by 98 percent of voters and had a turnout of 85 percent. In 2005, the Algerian government attempted to bring closure to its civil war by having a referendum on its Charter of Peace and National Reconciliation. This charter provided amnesty to those who participated in the civil war (unless they had committed rapes, mass murder, or public bombings). Furthermore, the charter removed any blame on the state for the events of the war and prevented leaders of the Islamist rebels from participating in Algerian politics. It also offered reconciliation to the families of the dead and disappeared by providing some compensation. This faced harsh opposition from the international community (and many Algerians) because it glossed over the atrocities performed during the civil war. The idea of the referendum was to give voters a voice and power in the process; however, many felt that their voice and power had been stripped because of the conditions of the referendum. There was widespread violence on the day of the referendum that reflected the widespread animosity toward the process. Some places saw turnout as low as 11 percent while others claimed 99 percent turnout. Voters overwhelm-

Referenda: Rare, but Powerful

25

ingly supported the referendum, leading to further separation from the civil war and a movement toward a more careful peace (Wallis 2005a, 2005b). Algeria provides an interesting case for this analysis because of its history as a French colony and its experience with civil war. One striking feature of the Algerian referenda was the high percentage of voter approval and the high level of turnout for these measures. Given Algeria’s authoritarian past, this links well with Chapter 6 on authoritarian regimes and direct democracy. Algeria also used its referenda for notable issues: independence, constitutions, charters, and peacebuilding. These connected ideals make Algeria’s use of national referenda striking and powerful. Hungary

Since 1989, the People’s Republic of Hungary has used referenda six times. The citizens voted to pass the referenda all six times, although one of the six was deemed invalid due to not meeting the 50 percent turnout threshold required by the constitution. According to the constitution, there are two main types of national referenda: mandatory and optional. In mandatory referenda (i.e., citizens’ initiatives for the introduction or repeal of legislation), parliament (the National Assembly) must call a referendum at the initiative of a minimum of 200,000 voters whether or not it approves. In optional referenda, the law states that parliament shall consider issuing a writ of referendum, but is not obliged to do so. An optional national referendum may be held at the initiative of voters (between 100,000 and 200,000), the president, the government, or at least one-third of parliament. In both cases, the calling of a referendum is the exclusive competence of parliament while the date of the referendum is set by the president of the republic (Pállinger 2012). Referenda are held when at least 200,000 electors petition, or on the motion of the president of the republic, the government, or 100,000 electors. There are also limitations on the subjects permitted in a referendum.4 One important role in Hungary’s referenda is that of its voter turnout threshold (Pállinger 2012). A nationwide referendum is valid only if: (1) more than half of the voters cast valid votes; and (2) at least one-quarter of all registered electors give the same answer to the referendum question (Beramendi 2008; Pállinger 2012).5 Without meeting the voter turnout threshold, the referendum is not made into law.

26

Direct Democracy: A Double-Edged Sword

As for the referendum’s outcome, the constitution distinguishes two types of referenda: binding and indicative. The results of mandatory referenda are always binding. The results of optional referenda may be binding or indicative, depending on the decision of parliament when it calls the referendum. The results of a referendum called with regard to a law already enacted by parliament but not yet signed by the president of the republic (a rejective referendum) is binding (Beramendi 2008). Hungary has used referenda on six occasions, sometimes asking more than one question at a time (see Table 2.4). A prime example of the multiple uses of referenda includes the four-part referenda held in 1989 (Nohlen and Stöver 2010). The four questions were loosely connected to the idea of election reform. The first question was about the timing of the presidential elections, the second was on the role of the Socialist Workers’ Party in workplaces, the third was on whether the party should have to provide a list of its properties, and the last asked whether the Workers’ Militia should be dissolved. Although plagued by low turnout, all of the proposals met the participation threshold and passed. The referendum that followed in 1990 also dealt with elections, this time regarding if the president should be elected directly. However, the referendum was plagued by low turnout and invalidated. In 1997 and 2003, Hungary voted on referenda to join larger alliances. Voters supported the idea of joining NATO in 1997 and the EU in 2003 (Nohlen and Stöver 2010). These larger alliances demonstrated Hungary’s role in the larger international community. Hungary, like many of the other countries joining the EU, faced criticism of its campaign to join (E. Butler 2007). The media was mostly positive and dispelled the myths of repercussions from joining the EU. The referendum passed; however, like many referenda before it, it was plagued by low turnout. Only 38 percent of voters participated—far short of the expected 70 percent. While legally binding, there were some concerns about the implications and outcomes (Partos 2003). Another referendum was held in 2004; this time, it had two parts and was a reflection of the political climate and the increased costs of providing services. The first dealt with public health services, and the second enabled ethnic Hungarians to apply for citizenship. Both parts were supported by voters, but one referendum had low turnout. By 2008, Hungary had undergone immense polarization

27 Table 2.4 Hungary Ballot Name

Presidential Election Referendum

Topic/Summary

Date

“Should the president be elected after 1989 parliamentary elections?” “Should organizations related to the Hungarian Socialist Workers’ Party be banned from workplaces?” “Should the Hungarian Socialist Workers’ Party account its properties owned or managed by it?” “Should the Workers’ Militia be dissolved?” “Should there be direct elections for 1990 the presidency?”

Hungarian Presidential Election Referendum Hungarian NATO “Do citizens want to become a member 1997 Membership of NATO?” Referendum Hungarian “Do you agree that Hungary should 2003 European Union become a member of the European Membership Union?” Referendum Hungarian Dual “Do you agree with the notion that public 2004 Citizenship health service institutions and hospitals Referendum should remain state or local government property, and, in accordance with that, the Parliament should repeal the contradictory law?” “Do you want the Parliament to pass a law that enables ethnic Hungarians with nonHungarian citizenship and residence, who affirm their Hungarian nationality, either with a Hungarian identity card described in Par. 19 of Act LXII/2001, or in a way specified in the forthcoming law, to apply for and be granted Hungarian citizenship?” Hungarian Fees “Do you agree that inpatient care should be 2008 Abolishment exempt from daily hospital fees with Referendum effect from 1 January in the year after the referendum is held on the present issue?” “Do you agree that family doctor care, dentistry care and special outpatient care should be exempt from consultation fees with effect from 1 January in the year after the referendum is held on the present issue?” “Do you agree that students in statesubsidized higher education should be exempt from tuition fees?” Source: Material from Pállinger 2012.

For

Results

50.1%

95.1%

Against 49.9% 4.9%

Turnout 58.0%

95.4%

4.6%

94.9% 85.9%

5.1% 14.1%

14.0%

85.3%

14.7%

49.2%

83.8%

16.2%

45.6%

65.0%

35.0%

37.5%

51.6%

48.4%

84.1%

15.9%

82.4%

17.6%

82.2%

17.8%

50.5%

28

Direct Democracy: A Double-Edged Sword

over the prior decade; of the many campaign promises made, most were unrealistic or unachievable given the political climate. Because of the government’s willingness to acknowledge that it would be unable to address its campaign promises and had to focus on austerity measures, the opposition proposed a referendum to put pressure on the government to bow to the people’s will (Pállinger 2012). As such, Hungary held a three-part referendum in 2008 to revoke government reforms regarding doctor visit fees, medical fees for days in a hospital, and tuition for higher education (Nohlen and Stöver 2010; Baji et al. 2011). All three parts were supported by voters, leading the prime minister to abolish these fees following the referendum. Hungary provides unique insight into the use of referenda in an Eastern European country. Hungary uses referenda for two major issues: for joining larger entities and for determining services and fees. The first is not so rare, but the use of referenda to determine services and fees is somewhat unique. An additional issue that makes Hungary an excellent case study for national referenda is the role that voter turnout thresholds play in the acceptance of referenda. Despite changing the threshold to accommodate different votes, Hungary is still plagued with low voter turnout in referendum elections—in a direct contrast to the Algerian model. Also, Hungary has repeatedly placed several measures on the same ballot—from two to four on the ballot—which makes these elections more complex and potentially leads to countercampaigns and confusion. A number of referenda have been prevented in Hungary, particularly after 2006. These were stopped by parliamentary actions to adopt some of the proposed changes prior to a vote, but parliament has also outright cancelled referenda (e.g., in 2009) because the subject matter changed during the course of the proposal (Pállinger 2012). In the Hungarian political system, direct democracy was first enacted during Hungary’s democratic transition and gave more power to the people to directly influence their government. However, continued elitism and a failure to involve citizens in decisionmaking have eroded faith in the Hungarian political system, and the use of direct democracy to fully engage the public has not lived up to its promise. Though referenda can be used as a threat in the political system when it is not supported by the legislature, Hungary does not provide support for citizen-initiated referenda (Pállinger 2012). Regardless,

Referenda: Rare, but Powerful

29

power for citizens through these mechanisms in Hungary seems limited, yet there continues to be a push for the government to accept citizen power through these mechanisms. Bangladesh

In Bangladesh, the Referendum Act of 1991 provides for referenda on particular articles of the Constitution of Bangladesh. These referenda are proposed by Parliament and receive presidential assent. At that point, they are presented to the public for a vote. There have been three referenda in Bangladesh. Ironically, none were held after the passage of the Referendum Act; rather, these three instances (see Table 2.5) led to the addition of the provisions for referenda to the constitution to ensure proper implementation of the elections. Bangladesh received independence in 1971 and had a rocky first few years. The country worked on establishing a constitution and national identity. However, these goals were undermined by a famine in 1974, the emergence of a one-party state, the assassination of the prime minister, a military coup, and the assassination of the president installed after the coup in 1975. These events created challenges for the new independent country. The first referendum was held in 1977, which was a vote of confidence (like many other referenda around the world) on President Ziaur Rahman (hereafter Zia) who became president through his military leadership during the assassinations of Bangladeshi leaders and the military coup. Voters

Table 2.5 Bangladesh Topic/Question

“Do you have confidence in President Major General Ziaur Rahman and the policies and programs adopted by him?” “Do you support the policies of President Ershad, and do you want him to continue to run this administration until a civilian government is formed through elections?” Creation of a parliamentary government. Source: Compiled from Ākhatāra 2001.

Date

For

Results

Against

Turnout

1977

98.9%

1.1%

88.1%

1985

94.5%

5.5%

72.2%

1991

83.6%

16.4%

35.2%

30

Direct Democracy: A Double-Edged Sword

were asked, “Do you have confidence in President Major General Ziaur Rahman and the policies and programs adopted by him?” There was a turnout of 88.1 percent, and 98.9 percent voted in support of the president. The confidence question was not unique in its approach, as many countries use referenda in the same way; however, the policies and programs portion is new. The referendum asked not only about confidence in a person, but also about approval of his work. Given the turmoil in Bangladesh, this was a loaded question that gave power to voters in theory, but its appearance to external actors was that of a continued form of citizen manipulation. Bangladesh went through continued turmoil after the 1977 referendum. Zia created the Bangladesh National Party and was elected president in 1979 with 76.6 percent of the vote—a huge margin over his nearest rival (T. B. Smith 1986). Zia was assassinated in May 1981 in an attempted coup (Zafarullah and Akhter 2001), but constitutional governance continued. There was a power vacuum in Bangladesh after Zia was assassinated. Justice Abdus Sattar became acting president, and General Hussain Ershad maintained the power structure while preventing a coup from other sources, namely Major General Abul Manzoor. Sattar made several concessions to the military’s demands and to Ershad himself; however, he struggled to create institutional power. Sattar was elected in a popular vote in December 1981. General Ershad, who publicly supported the constitutional government and created order in the military after the coup, began to argue that the military had a role in national decisionmaking. He stated that political stability and government accountability were conditional on the military’s support and cooperation (Ershad 1981; T. B. Smith 1986). While the government did not embrace this view, Ershad continued to exert pressure on the president to include the military in the government’s structure. Having developed some military presence in leadership and taking advantage of strife within the ruling party, Ershad staged a coup in March 1982 (Zafarullah and Akhter 2001). After the coup, Ershad suspended the constitution, proclaimed martial law, dissolved Parliament, and banned political activity (Rahmān 1983). These actions and Ershad’s lack of personal connection to the public created doubt among citizens, and he was highly criticized. As such, Ershad went to the public in a referendum to gain support for his regime and his performance and policies in 1985 (T. B. Smith 1986). The 1985 referendum on military rule and to confirm public support for Ershad was Bangladesh’s second referendum. The ref-

Referenda: Rare, but Powerful

31

erendum asked voters, “Do you support the policies of President Ershad, and do you want him to continue to run this administration until a civilian government is formed through elections?” The referendum was not without some conflict, as the opposition to Ershad organized a strike on the day of the election and argued that the results were fraudulent. Civil society urged a boycott of the referendum due to Ershad’s actions. Many of the opposition leaders were arrested (including the wife of the assassinated president Zia and head of the Bangladesh National Party). Additionally, there were issues with the election procedures. Voters did not have a secret ballot; rather, they placed their ballot in one of two boxes: one that had General Ershad’s picture, and the other that was black to mark disapproval. This led to discussions about voter coercion and how the results could not be trusted. While the turnout was lower in 1985 (72.2 percent), the results provided support for Ershad’s regime with 94.5 percent in favor of his policies (T. B. Smith 1986). This gave Ershad a little bit of support to move forward with the creation of his own political party (Bangladesh Awami League) and the reinstatement of Parliament through elections in 1986. The 1986 elections did little to change governance in Bangladesh or deepen democracy. Ershad did not gain the legitimacy that he wanted. The Jatiya Party won the most seats in Parliament, the Bangladesh National Party boycotted the election, and turnout was only 61 percent. Thus, the 1988 elections that were held just 2 years later reaffirmed the Jatiya Party’s leadership (despite being boycotted by many other parties) and demonstrated that Ershad did not have a popular mandate, undoing much of his democratic delusions. The most recent referendum in Bangladesh was in 1991, regarding the new constitutional amendment to reinstitute parliamentary government and create a prime minister position as head of the executive branch. This referendum had a low turnout of only 35.2 percent, but received support from 83.6 percent of those voters. Referenda in Bangladesh were utilized to create legitimacy for military rules and even to further consolidate the military’s power (Zafarullah and Akhter 2001). The military regime utilized these mechanisms to further its own agenda and demonstrated substantial manipulation of the process and the public. After 1991, referenda were incorporated into the constitution as a power of the people rather than to prop up illegitimate regimes.

32

Direct Democracy: A Double-Edged Sword

Uganda

Direct democracy has been used only at the national level for referenda in Uganda. Under Article 255 of the Ugandan Constitution, Parliament shall by law make provisions: (1) for the citizens’ right to demand the holding by the electoral commission of a referendum (nationally or subnationally) on any issue; and (2) for the holding of a referendum by the electoral commission, if the government refers any contentious matter to a referendum (Beramendi 2008). Legislation and guidelines for citizens’ initiatives and recalls also exist, but have never been used. Although Uganda does not face much criticism regarding direct democracy, the country does receive a lot of criticism regarding its political divisions, violations of human rights, and extreme poverty. Direct democracy has been used twice in Uganda to explore multiparty democracy (see Table 2.6). Uganda banned political parties in 1986 in the nonparty movement system under President Yoweri Museveni.6 Museveni had been involved in the movements to remove previous presidents Idi Amin and Milton Obote. His presidency was supported by Western countries, which felt that he represented change in Uganda and would lead to democratic openings. The first referendum was held in 2000 to determine what type of political system to adopt. Voters were asked, “Which political system do you wish to adopt, Movement or Multi-Party?” With a voter turnout of around 50 percent, there was high support (over 90 percent) for the movement system—91 percent of voters favored a movement political system7 while only 9 percent favored a multi-party system (Uganda Election Commission 2015). Museveni’s political future rested on the outcome of this election, and there was substantial pressure on voters to support this referendum and continue to invest power in the president and his agenda. In 2005, voters were posed the question, “Do you agree to open up the political space to allow those who wish to join different organizations/parties to do so to compete for political power?” Ugandans voted to return to multiparty politics (with more than 90 percent support). The vote supported the change, but there were doubts about the turnout numbers (Uganda Election Commission 2015). The initial reports of the 2005 referenda indicated that turnout was nearly 17 percent lower than the figure released officially. This led to some concerns about the election procedures, as low turnout would have been politically embarrassing for Museveni. Although this was Musev-

Referenda: Rare, but Powerful

33

Table 2.6 Uganda Ballot Name

Ugandan Multi-Party Referendum Ugandan Multi-Party Referendum

Language

“Which political system do you wish to adopt, Movement or Multi-Party?” “Do you agree to open up the political space to allow those who wish to join different organizations/ parties to do so to compete for political power?”

Results

Date

For

July 28, 2005

3,643,223 (92.44%)

Against

Turnout

297,865 (7.56%)

47.33%

June 29, Movement Multi-Party 2000 90.7% 9.3%

51.1%

Source: Compiled from Muriaas 2009.

eni’s second referendum in 5 years, he was moving away from his ardent support for the movement system. The country voted to return to a multiparty system, but Museveni oppressed political opposition, passed a constitutional amendment that ended presidential term limits, and enabled the continuation of his rule. Uganda’s experience with referenda is somewhat limited, with only two proposals directly reversing each other. However, the country’s history with authoritarian regimes using referenda demonstrates that there is a potential for democratic opening. Despite these votes, Museveni used executive power to continue his rule and oppose opposition parties. Museveni was elected to his fifth term as president in 2016 and continues to be a powerful leader in the region. There continues to be controversy over his elections, the oppression of opposition through intimidation at polling locations, vote rigging, and voter intimidation; however, Museveni has embraced the electoral process and continues to hold elections every 5 years. Analysis Looking at these five cases and their use of referenda, it is clear that the circumstances of each referendum are unique, but the context of the country influences its use of referenda. Whether it is to approve a colonial constitution or to change the electoral procedures within a country, these referenda have varying circumstances. These cases demonstrate

34

Direct Democracy: A Double-Edged Sword

that there are issues related to national referenda that have important ramifications within each country. While national referenda are quite rare, these countries’ various uses of them demonstrate the importance of these types of mechanisms in the democratic process. The power of the people is inconsistent in these cases. Uruguay and Algeria have both utilized referenda for constitutional changes and moving toward peace through amnesty programs for rebels. Turnout in both countries is high, with significant citizen buy-in to the process. The use of referenda in these two countries increases the power of citizens to make and endorse changes in their government. This is followed by Hungary, which has used referenda as a function of its democratic transition and continues to deepen its democracy through these mechanisms and giving voters power. This power is used circumspectly in Bangladesh and Uganda. Bangladesh’s and Uganda’s uses of referenda to support the regimes have had mixed results. While it appears that voters have more of a voice, they are used to prop up the regime proposing the referendum. Thus, referenda are not a function of deepening democracy, but rather a way to pay lip service to the democratic process. Uganda changed its electoral system twice through referenda, first to outlaw political parties, then to allow them. Despite these changes, the executive continued to oppress political opposition (and even imprison opposition party leaders) and suspended presidential term limits. Thus, referenda are designed to give power to voters, but that power can be limited, be manipulated, or even be just a show of democracy to appease international communities. Notes 1. The United States was one of the first countries to embrace direct democracy. However, the country is also concerned about the power of these measures. The use of referenda is regularly and solidly rejected when it threatens the union (Qvortrup 2014b). An example is the Confederate states’ desire to leave during the American Civil War. Some states actually did vote to leave. There have also been calls for a secession vote within Texas but, as yet, these continue to be rejected. 2. Mandatory referendum: any constitutional amendment that may infringe on rights of citizens. Optional referendum: any constitutional amendment not considered by the Constitutional Council as rights infringement, or any amendment proposed by three-fourths of each house of Parliament.

Referenda: Rare, but Powerful

35

3. The Algerian Civil War began in 1991 after the military prevented the 1991 National Assembly elections in order to prevent the Islamist party from gaining control of the Assembly. It was a particularly bloody war in which 150,000 Algerians died over 14 years. 4. Referenda are not permitted on the following issues: • Acts on the budget, execution of the budget, types of central taxes and stamp duties, custom duties, and the regulation of local taxes; • Obligations from international agreements in force, or the contents of the acts containing these obligations; • The provisions of the constitution regarding referenda and popular initiatives; • Any issues of personnel and organizational change that fall under the authority of parliament; • The dissolution of parliament; • The government’s program; • The declaration of a state of war or a state of emergency; • The deployment of the armed forces abroad or within the country; • The dissolution of the delegates’ assembly of the local government; and • A declaration of amnesty. 5. The 25 percent quorum is lower than was initially proposed in Hungary. Until 1997, referenda were required to have a 50 percent turnout to be approved. 6. Using referenda to legitimize party systems is common in Africa— they are often used to provide support to authoritarian rulers (Kersting 2014; Bratton and Lambright 2001). 7. This refers to the Ugandan idea of a non-party system.

3 Recall Elections: A Mixed Power

to limit an elected official’s term and remove him or her from office rather than implementing procedures such as impeachment. Recall elections allow voters (or another authorized body) to demand a vote on the possible removal of elected officials prior to the end of their term. However, when an elected official wins in a recall, the impact can be devastating for citizens, making this a challenging struggle for power and a balancing act between voters and elected officials. Twenty-three countries currently have laws giving citizens the ability to recall elected leaders (to varying degrees). Recall elections add to elected officials’ accountability; however, they can be extraordinarily disruptive as well as an added expense to the state for election administration. For those in support of recall elections, this serves as yet another check and balance in the system that keeps elected officials culpable during their terms. It also provides an opportunity for citizens to push for accountability from leaders outside the official chains of impeachments. Impeachments can be started only through governmental means and by government officials, leaving citizens outside of the process. Thus, supporters feel that recalls enable citizens to have a voice in the accountability process. Recall elections, because of their criticism, complexity, and highly involved processes, are used more commonly at the state and Recall elections are a form of direct democracy that allows voters

37

38

Direct Democracy: A Double-Edged Sword

local level and rarely at the national scale. They are the least used mechanism of direct democracy, yet their use is a fascinating expression of public power, and failed recalls can have significant repercussions that possibly result in tyrannical leadership. By using case studies (Nigeria, Venezuela, Taiwan, and the United States), I illustrate the power of recalls, their successes, and the backlash that occurs when they fail. There are pro and con arguments about the use of recalls in any democracy. On the affirmative side, they add a layer of accountability and permit voters to end the term of an official early. They encourage civic engagement and knowledge between elections. They also provide an alternative to impeachment. But there are many drawbacks to the process. They have a disruptive effect on a government, leading to additional elections and governing delays. Furthermore, having the threat of a recall can limit the work of elected officials, as they may not make necessary difficult decisions because they are afraid of a recall attempt. This undermines their authority and ability to act during their term in office. Additionally, much of these processes are not really a function of individuals but rather the function of special interest groups or opposition groups, thus moving away from the underlying motivations of citizen power in a direct democracy. Finally, recalls are costly and undermine the idea of republican government (Welp 2016). Different governments develop their own criteria for initiating and executing a recall. The theoretical underpinning of recall elections is that representatives should be accountable to their constituents throughout their term, not only during elections. Thus, if voters want to end their term prematurely because they are unhappy with the performance of the leader, they have the ability to send the leader that message. That message is conveyed, regardless of the outcome of a recall. While disappointment in a leader is certainly a major component of this type of mechanism, it can also be a result of specific policy initiatives or economic conditions. The literature tells us that incumbents are punished for a variety of reasons (for a discussion of this in the context of different democratic regimes, see Helwig and Samuels 2008); this is even more pronounced during a recall election. The results of a recall election can be highly polarizing and disruptive to the governing process. Although elections are often frequent,

Recall Elections: A Mixed Power

39

adding additional layers of elections and accountability can disrupt the entire process and be expensive for the state to administer. That said, recall elections serve as an additional mechanism of oversight and encourage accountability between representatives and their electorates (Beramendi 2008; Welp 2016). Recall elections have different legal and structural aspects that influence how successfully and how frequently citizens use them to remove elected representatives. There are three important components of the recall mechanism. First, only the legal framework that outlines the use and manner in which elected officials are subject to recall elections. This includes where these rules are developed and incorporated (e.g., constitutionally, statewide, or nationally). Second, requirements must be met to enact a recall. Each jurisdiction has different requirements; these can be a specific number of legislative votes or the signature requirements for a popular petition. These are not minimal requirements (and I am not arguing that they should be); these requirements often mean working across party lines and other divisions in society to remove someone from office (e.g., race, class, ethnicity, and language). Finally, the quorum required either in the legislature or by turnout could affect the success of recall elections. Typically, the higher the office being recalled, the higher the requirements to recall that official (Beramendi 2008). Types of Recalls The most straightforward form of recall allows for a specified number of citizens to petition for a recall election. However, there is also a mixed form where government bodies can start the process and pose the question to voters about removing an elected official from office (e.g., in Austria, Iceland, Palau, and Romania). Additionally, there may be governmental oversight to the process where, once voted on by the voters, it must be processed and approved by the appropriate body such as in Uganda (Beramendi 2008). A major variation in recall elections is which offices are open to recall elections. There are differences not only in what offices can be recalled, but also in what mechanisms of recall are used at each level. Some jurisdictions provide for a mixed recall for higher-level executive officials and a full recall for members of national legislative bodies.

40

Direct Democracy: A Double-Edged Sword

The Development of Recalls The development of recalls first began at local and state levels in Switzerland and the United States as a result of increased populist ideals. They spread throughout Latin American in the 1990s as countries developed new constitutions with mechanisms for participatory democracy. The vast majority of these countries continue to use recalls at the state and local levels and rarely for nationally elected officials (see Table 3.1). The constitutional provisions that include provisions for recall can be outlined specifically or more generically with “statements about the possibility of revoking the mandate of elected officials as a right of the people in those countries” (Beramendi 2008: 111) (e.g., Ecuador, Ethiopia, Peru, and Taiwan). In each instance, there are specific legal rules with differing levels of citizen involvement. Who Can Be Recalled?

Who can be recalled depends on the jurisdiction. Some national governments have included mechanisms to recall their highest levels of executive offices while others have reserved this mechanism only for state and local officials (see Table 3.2). There are similarities among countries that have used recall elections that amplified the use of recall and other direct democracy mechanisms such as the influence of populism, as explained in previous chapters. Moreover, there seem to be distinctions in the types of governments that have recall elections. Recalls are more consistent with presidential systems of government (at least on the executive level) than with a parliamentary system (Beramendi 2008). When looking Table 3.1 Countries that Use Recalls Africa

Ethiopia Nigeria Uganda

Americas

Argentina Columbia Cuba Ecuador Peru United States Venezuela

Source: Beramendi 2008.

Asia

Kyrgyzstan Taiwan Turkmenistan

Europe

Austria Belarus Liechtenstein Germany Iceland Romania Serbia

Oceania

Kiribati Micronesia Palau

Recall Elections: A Mixed Power

41

at recalling members of the legislature, this is more consistent with the use of single-member districts rather than multimember districts. While there is a variation in who can be recalled, there are often are differences in the mechanisms used for recall elections. Latin America has experienced a surge in the use of recall elections, with 21,000 attempts at recalls since the 1990s (Welp 2016). Recalls have been used to ingrain participatory democracy in much of South America, particularly after the political crisis in Colombia in 1991 (Welp 2016). What Is the Process?

A unique aspect of the recall process is the different grounds for recalling an official. No legal or constitutional requirement is needed to petition for a recall of elected officials. This is a function of the will of the people (or will of the authoritative body) to determine whether they want to recall an elected official. In this characteristic, the mechanism differs very much from impeachment, which has concrete constitutional requirements to start the process. We have seen recalls focused on corruption, incompetence, or even unpopular policy proposals. The process of recall elections requires five phases (Beramendi 2008): (1) an initiation of the recall; (2) the processing and validation of the initiative for a recall (either collection of signatures through a petition or votes by the authoritative body); (3) a recall campaign to encourage voters to participate and cast their votes in one direction or the other; (4) the vote itself; and (5) the results and potential replacement of the recalled official. Each of these stages is specific to the country, state, or municipality where this occurs, indicating challenges to making broad assertions about recall elections. Once signatures have been obtained, there are different criteria for whether the recall is approved or rejected. Table 3.2 illustrates these criteria. These include who can vote in the referendum, the percentage of electors who are required to participate in the elections, and the percentage of votes required in favor of or against the recall. These requirements in their simplest form can be found in Belarus where only a simple majority of voters is required. In the most strenuous form, Pohnpei in the Federated States of Micronesia requires 60 percent of registered electors to approve a recall. The outcomes of recalls can be disruptive forces and interrupt the operation of government. Officials are required to take time out

Table 3.2 Countries that Allow for Recalls Country or State

Argentina

Austria

Belarus

Colombia

Cuba

Ecuador Ethiopia

Germany Kiribati

Kyrgyzstan Liechtenstein Iceland

Micronesia (Chuuk)

Elected Official Subject to Recall

Local elected officials

President

Deputies

State governors, mayors

Delegates to municipal assemblies President, governors, mayors, prefectos, and deputies Members of the legislature

Entire legislature (of selected states) Mayors Members of the legislature

Deputy of Legislative Assembly or Assembly of People’s Representatives Entire legislature President

Governor or lieutenant governor Senator or representative

Type of Recall

Initiated and approved by citizens at the regional and local levels. Initiated by authorities and approved by citizens.

Initiated and approved by citizens at the national level. Initiated and approved by citizens at the regional and local levels. Initiated and approved by citizens at the regional and local levels. Initiated and approved by citizens at the national level.

Initiated and approved by citizens at the national level. Initiated and approved by citizens at the regional and local levels. Initiated by authorities and approved by citizens. Initiated and approved by citizens at the national level. Initiated and approved by citizens at the national level. Initiated and approved by citizens at the national level. Initiated by authorities and approved by citizens. Initiated and approved by citizens at the national level.

Full or Mixed

Full

Mixed

Full

Full

Full

Requirements

Varies by locality.

50% of the House of Representatives and twothirds of the votes cast. Not less than 20% of the citizens eligible to vote and who reside in the corresponding area. Varies by locality.

Varies by locality.

Full

At least 15% of the registered electors for president and governors, and at least 10% of the registered electors in the electoral district for other officials.

Full

Varies by locality.

Full

Varies by locality.

Full

Varies by locality.

Full

Registered electors equal in number to at least 15% of those who voted in the last general election. Registered electors for the district or region equal to at least 20% of those who voted in the last general election in the same district or region.

Full

Mixed

Varies by locality.

75% of the legislature.

Country or State

Micronesia (Kosrae) Micronesia (Yap) Nigeria

Palau

Elected Official Subject to Recall

Governor, lieutenant governor, state court justice, and senators

Full

Governor, lieutenant governor, Initiated and approved by citizens at and legislative representative the national level. Members of the Senate or the Initiated and approved by citizens at House of Representatives the national level. Member of the legislature Initiated and approved by citizens at the national level. President or vice president Initiated by authorities and approved by citizens.

Full

Romania

Mayors and regional elected authorities President

Serbia

President

Turkmenistan Uganda United States Venezuela

Full or Mixed

Initiated and approved by citizens at the national level.

Peru

Taiwan

Type of Recall

Initiated and approved by citizens at the regional and local levels. Initiated by authorities and approved by citizens.

Initiated by authorities and approved by citizens. President or vicepresident Initiated by authorities and approved by citizens. Local and regional leaders Initiated and approved by citizens at (legislators and city officials) the regional and local levels. President Initiated by authorities and approved by citizens. Members of Parliament Initiated by citizens, approved by authorities. State and local officials Initiated and approved by citizens at the regional and local levels. All elected officials Initiated and approved by citizens at (including president) the national level.

Source: Data primarily from the Beramendi 2008.

Full

Full

Mixed Full

Mixed Mixed

Mixed Mixed Full Full

Full

Requirements

At least 25% of the persons qualified to vote for the office under recall, except for the justice— who requires the same number of signatures as statewide elective office in Kosrae. At least 25% of the voters qualified to vote for the office being recalled. More than 50% of the voters registered to vote in the member’s electoral district. No fewer than 25% of the number of voters who participated in the most recent election. Resolution adopted by two-thirds of state legislature representatives in not less than three-fourths of the states. Varies by localities.

Chamber of Deputies and the Senate (during a joint session), by a majority of both bodies and after consultation with the Constitutional Court. Two-thirds of total representatives.

Initiated by one-fourth of all members of the legislature and passed by two-thirds of the members. The People’s Council may express a lack of confidence in the president. Varies by state.

At least 20% of registered voters in the relevant electoral district.

44

Direct Democracy: A Double-Edged Sword

of governing to fight against these recalls. The aftermath can require yet another election to fill the vacated seat, or it can buoy a political leader. Thus, the challenges of a recall permeate the entire process. Why Use Recalls?

When thinking of an overarching theory for the use of recalls, some common themes emerge. Recalls may seem to exist to hold leaders accountable, but that is not how they are really used. In some regimes, should recall elections occur, the process is not necessarily transparent and can be subject to serious corruption. Furthermore, should the elected official survive the recall attempt, there can be an increase in his or her sense of power and authority. Thus, the recalled official is more likely to take advantage of the failure of the recall and pursue his or her agenda with increased fervor. A second consideration is that, although many democracies include aspects of direct democracy, many do not use them. This desire to appear democratic and open is an influential part of how and why these mechanisms are included. However, the ease of use and the level of usage are important factors when evaluating the quality of recall mechanisms. An issue that may influence the mechanism’s openness is the regime’s willingness to allow for challenges to the leadership. Because of the intricacies of the recall process, recalls can be impossible to trigger if there are high petition or legislative thresholds. While there are theoretical perspectives of the recall process, Yanina Welp (2016) studied reasons that voters gave for activating a recall. These include, but are not limited to, a lack of fulfillment of election plans, misuse of public resources, bad management, nepotism, lack of transparency and accountability, and negligence. There are some notable impacts on elections when recall elections occur (much like any use of direct democracy). Turnout is generally higher when states allow representatives to be recalled (at least in the United States and Canada), albeit long-term effects have been harder to establish in other democracies. Recall mechanisms can also encourage the resignation of unpopular leaders to prevent a vote. Those who survive recall elections can be strengthened by their opponents’ failure to recall them (example: California governor Ronald Reagan who faced failed recalls in 1967, 1968, and 1972) or face defeat during the next election (Qvortrup 2011).

Recall Elections: A Mixed Power

45

Case Studies This analysis looks at the four cases of the United States, Venezuela, Taiwan, and Nigeria. I chose these cases because they represent different levels of development, economic growth, and use of the recall mechanism. These cases are regionally diverse and their uses of recalls vary on a multitude of levels (see Table 3.2). The United States

The recall mechanism, like other direct democracy measures, exists solely on the state and local level in the United States. The use of recall methods vary state to state as a result. However, there are some general procedures for recall elections. As a matter of federalism, recall elections do not apply to members of the national legislature regardless of state laws. The recall process varies in its details from one state to another but, in general, it follows these steps: 1. File an application to circulate a recall petition (some states allow petitions only if they meet certain grounds for recall). 2. Circulate a recall petition, gathering a specified number of signatures in a limited period of time (these vary by state). 3. Submit petitions to election officials for verification of signatures. 4. If sufficient valid signatures are presented, a recall election is held (National Council of State Legislators 2016). Recall elections are used far less often than other mechanisms of direct democracy in the United States, but they have been used as a check and balance on many state and local officials. As indicated in Table 3.3, 18 states allow recalls at the state level, and a total of 36 allow the recall of local officials. These recall devices, used since 1903, were adopted in many states and are used at both the local and state levels. Starting in 1908, some states began incorporating the use of recalls statewide for local and municipal elections to make them uniform in application. The vast majority of recall elections in the United States are at the city council or school board level (roughly 75 percent).

46

Direct Democracy: A Double-Edged Sword

Table 3.3 US States Using Recalls Recall Elections Held with Election for Successor Arizona California Colorado Illinois Nevada North Dakota Rhode Island Wisconsin

Recall Election Followed by Special Election for Successor Georgia Louisiana Michigan Minnesota Montana New Jersey

Recall Election Where Successor Is Appointed Alaska Idaho Kansas Oregon Washington

Source: National Council of State Legislators 2016.

Recall efforts against state legislators are more common than at the national level, but are still rare. Recall attempts against legislators have gathered sufficient signatures to trigger an election only 38 times since 1913 (see Table 3.4). 1 Of those recall attempts, 66 percent were successful. A majority of the recall elections were recent (45 percent took place between 2011–2013), 2 marking an increase in the use of recall elections against state legislators. Recall elections have been used in the United States at the gubernatorial level three times. In 1921, North Dakota voters recalled Governor Lynn J. Frazier as well as the attorney general and the commissioner of agriculture (Zacks 2012). In 2003, California voters recalled Governor Gray Davis in a much-publicized campaign involving porn stars, actors, and politicians. This was not the first attempt to recall leaders in California, as there were 32 previous attempts to recall California governors since 1911. The 2003 recall, however, was the first to make it to the ballot. In 2012, Wisconsin governor Scott Walker faced a recall election, but survived and went on to win subsequent gubernatorial elections. Arizona voters triggered a recall election in 1988 for Governor Evan Mecham; however, he was impeached by the legislature prior to the election. California 2003. The recall of Governor Gray Davis in 2003 is an

example of state-level use of this mechanism. Davis, a Democrat, was elected in 1998 and reelected in 2002 with a 5 percent margin over his Republican opponent. He was governor during a challenging

Table 3.4 US State Requirements

Alaska

Arizona

California

Colorado Georgia

Idaho Illinois

Kansas

Louisiana

Signature Requirement

All elected public officers of the state 25% of the votes cast in the last election for the official being recalled. except judicial officers Every public officer in the state holding 25% of the votes cast in the last election for the official being recalled. an elective office State officers, members of the For statewide officers: legislature, judges of courts of appeal 12% of the votes cast in the last election for the official being recalled, 1% from each of 5 counties. State senators, members of the assembly, members of the Board of Equalization, judges of courts of appeal: 20% of the votes cast in the last election for the official being recalled. Every elective officer of the state 25% of the votes cast in the last election for the official being recalled. Public officials who hold elective office For statewide officers: 15% of eligible voters for office at time of last election, one-fifth from each congressional district. Others: 30% of eligible voters for office at time of last election. Every public officer in the state except 20% of eligible voters for office at time of last election. judicial officersr 15% of the votes cast for governor in the preceding general election Governor from each of at least 25 counties. Also required are the signatures from at least 20 members of the House of Representatives and 10 members of the Senate, with no more than half the signatures of members of each chamber from the same political party. All elected public officers in the state 40% of the votes cast in the last election for the official being recalled. except judicial officers Any state official except judges of the If over 1,000 eligible voters: courts of record 33.3% of eligible voters for office at time of last election. If fewer than 1,000 eligible voters: 40% of eligible voters for office at time of last election. All elective officers except judges 25% of total votes cast for position at last election. of the courts of record

Circulation Time

Not specified

120 days

160 days

60 days 90 days

60 days 150 days

90 days

180 days 60 days continues

47

Michigan

Who Can Be Recalled

Minnesota Montana Nevada New Jersey

Who Can Be Recalled

State executive officers, legislators, and judges of the Supreme Court, Court of Appeals, or a district court Any person holding a public office of the state

Every public officer in the state Any elected official in the state or representing the state in the US Congress North Dakota Any elected official of the state or legislative district Oregon Every public officer in the state

Rhode Island

Washington Wisconsin

48

Table 3.4 Continued Signature Requirement

Circulation Time

25% of total votes cast for position at last election.

90 days

For statewide officers: 10% of eligible voters for office at time of last election. For district officers: 15% of eligible voters for office at time of last election. 25% of the votes cast in the last election for the official being recalled. 25% of the registered voters in the electoral district of the official sought to be recalled.

3 months

25% of the votes cast in the last election for the official being recalled.

15% of total votes cast in officer’s district for all candidates for governor in the last election. Governor, lieutenant governor, secretary 15% of total votes cast for said office in last general election. of state, treasurer, attorney general Every elective public officer of the For statewide officers: state except judges of courts of record 25% of the votes cast in the last election for the official being recalled. Others: 35% of the votes cast in the last election for the official being recalled. Any state, judicial, congressional, or 25% of total votes cast for the office of governor at the last election legislative official within the same district or territory of that officer being recalled.

Sources: Compiled from National Conference of State Legislatures 2016; Beramendi 2008.

60 days Governor or US senator: 320 days All others: 160 days Not specified

90 days

90 days

Statewide officers: 270 days Others: 180 days 60 days

Recall Elections: A Mixed Power

49

time in California, and the economic situation changed dramatically during his tenure. As a result of the California economic contraction, Davis and the democratic legislature raised fees and reduced services across the state. Combined with energy issues, this created an atmosphere ripe for a recall (Garrett 2004; Alvarez et al. 2004). The recall effort was organized by a number of groups and benefited from a $1.6 million donation from Republican congressman Darrell Issa to pay signature gatherers to recall Davis. Under California law, 900,000 valid signatures (or 12 percent of the number of voters in the most recent gubernatorial election) were needed to recall the election. California, similar to other states, elects the replacement candidate at the same time as the recall election (see Table 3.4). Replacement candidates had the choice of collecting signatures to be placed on the ballot or paying a $3,500 filing fee (Garrett 2004; Alvarez et al. 2004). On the recall ballot, there were 135 candidates to replace Davis, ranging from politicians to movie stars. During the campaign, two candidates significantly emerged as alternatives: Arnold Schwarzenegger for the Republicans and Cruz Bustamante (the lieutenant governor) for the Democrats. The election faced obstacles with court challenges and the huge number of replacement candidates and, in the end, the results were decisive. Governor Davis’s recall received 55 percent of the recall votes; Schwarzenegger received 49 percent of the election vote, and Bustamante received 32 percent. The turnout for this election was staggeringly high at 61 percent of registered voters (as compared to other direct democracy votes). This high turnout illustrated the saliency of the election in the citizenry; voters had the extraordinary power of being able to remove a sitting governor and replace him with an alternative candidate. This provides substantial power to the people to make decisions about the future of their state. The choice of a celebrity to replace the sitting governor fed criticism of direct democracy elections, as voters may not have chosen the best policy choice but rather the candidate with the highest name recognition. Schwarzenegger went on to serve as governor of California until November 2011 (Alvarez et al. 2004).3

Wisconsin 2012. Wisconsin allows the recall of any incumbent elective officer once during the term, after the first year in office (Wisconsin Constitution 2016). It requires a petition of at least 25 percent of the votes cast for the office in the preceding election.

50

Direct Democracy: A Double-Edged Sword

Wisconsin governor Scott Walker, a Republican, took office in early 2011. Walker campaigned on fiscal conservativism and was an early favorite in the field. He promised to make budget cuts and tax cuts and to refuse aspects of federal funding. He suggested that he would make up budget shortfalls by cutting state employee salaries and jobs. Once elected, Walker had approval ratings around 50 percent (Blumenthal 2012). The biggest challenge to the Walker governorship was a dispute over collective bargaining (Kocieniewski 2012; Zacks 2012). Walker was accused of union breaking, and his favorability declined. In 2012, a recall election was called for the governor and lieutenant governor of Wisconsin. In addition to the recall election of the governor, there were four state senator recall elections on the same day. This was only a year after the monumental recall of state legislators in Wisconsin in which nine state senators faced recall (eight survived and one was recalled). A unique component of the Wisconsin recall election is that there is a primary to narrow down the field to one representative from each party prior to the recall election (Government Accountability Board, n.d.). There was a clear partisan motivation in this recall since the governor, lieutenant governor, and all four senators were Republican.4 Signature collection was easily obtained in the 60 days allotted by Wisconsin law. In January 2012, members of United Wisconsin, the organization that spearheaded the recall effort, announced that it had collected 1 million signatures (more than the 540,000 required). Despite their efforts, early polls indicated that Walker would survive the recall. Many signatures were challenged; however, given that United Wisconsin had nearly double the required amount of signatures, this challenge was seen as an empty threat. Several notable signatures were collected, including 29 Wisconsin circuit court judges, some of whom had ruled on cases involving the Walker administration. Journalists and union leaders also signed the petition. Another notable feature of the Wisconsin recall election was the cost. Conducting the primaries and the election cost the state election board almost $27 million. Furthermore, the money spent by candidates and outside groups was staggering—more than any other election in Wisconsin history. The advocacy group Wisconsin Democracy Campaign reports that candidates and outside groups spent more than $80 million during the recall race (Wisconsin Democracy Campaign

Recall Elections: A Mixed Power

51

2016). This was a dramatic increase from the previous election; during the 2010 gubernatorial election, only $37.4 million was spent (Moyers and Winship 2012). The results of the recall were mixed. Governor Walker became the first governor to survive a recall election and he received more votes to retain him than those of his previous election. The lieutenant governor also retained her seat. In the senate recall elections, two Republican incumbents retained their seats, the open seat was won by a Republican and one senator was recalled and replaced with a Democrat, changing the balance of power in the State Senate to a Democrat majority. Turnout (57.8 percent) for the recall was higher than every previous Wisconsin gubernatorial election that had not coincided with a presidential election (Montopoli 2012). The recall did little to tarnish Walker’s political career and seemingly boosted his efforts to move forward with his mandate. Walker was popularly elected again as governor in 2014 and made a bid for the Republican presidential nomination in 2016. The use of recalls for executive positions at the state level as well as a multitude of offices in local government demonstrates the connection to democratic accountability to voters. In a comparison between Davis and Walker, both faced a recall due to economic issues and policy choices made during their term. However, the success of the recall in California ended Davis’s career while Walker’s career flourished after he survived the recall election. These cases illustrate that the aftermath is almost as important as the recall process. What is clear is the power of voters in these elections. Voters had the ability to recall a governor and challenge another. Their power to provide another check and balance in the system demonstrates the value of this process. However, there were also key issues of manipulation that plagued voters, which they had to address throughout the process. Given that members of the US House of Representatives are elected every 2 years, it seems redundant to include recall measures for these offices. The checks and balances instituted in the federal government make recall elections less necessary than those in lessstructured systems. Furthermore, the disruption to the governing process (particularly during a president’s 4-year term) and economic stability of the country could be another incentive to limit recall elections to state and local races.

52

Direct Democracy: A Double-Edged Sword

Venezuela

Venezuelan law permits two direct democracy mechanisms: referenda and recalls. The Venezuelan voters have an infamous recall election history. Venezuela is one of the more open countries to recall elections, as they are provided for in two articles of the 1999 Constitution: Article 72: All . . . offices filled by popular vote are subject to revocation. Once one-half of the term of office to which an official has been elected has elapsed, a number of voters representing at least 20 percent of the registered voters in the affected constituency may petition for the calling of a referendum to revoke that official’s mandate. When a number of voters equal to or greater than the number of those who elected the official vote in favor of the recall, provided that a number of voters equal to or greater than 25 percent of the total number of registered voters vote in the recall referendum, the official’s mandate shall be deemed revoked and immediate action shall be taken to fill the permanent vacancy as provided for by this Constitution and by law.

Article 233: The President of the Republic shall become permanently unavailable to serve by reason of any of the following events: death; resignation; . . . recall by popular vote. . . . When the President of the Republic becomes permanently unavailable to serve during the first four years of his constitutional term of office, a new election by universal suffrage and direct ballot shall be held within 30 calendar days. Pending the election and inauguration of the new President, the Executive Vice President shall take charge of the Presidency of the Republic. In the cases described above, the new President shall complete the current constitutional term of office. If the President becomes permanently unavailable to serve during the last two years of his constitutional term of office, the Executive Vice President shall take over the Presidency of the Republic until the term is completed.

One of the reasons that the Venezuelan people allow recall elections at the executive level is that the presidential term is 6 years and, therefore, a recall is more appropriate than in a system where the presidential term is shorter (Maya and Lander 2011). Venezuela embraces the use of recall elections and these elections are a substantive feature of the country’s style of government. This is unique given the challenges Venezuela has with democracy. There was an attempted recall of President Hugo Chávez in 2004. It was initiated by 2.4 million Venezuelan citizens and remains

Recall Elections: A Mixed Power

53

one of the most prominent examples of the use of the recall mechanism at the highest level worldwide. In August 2003, a civil association in Venezuela called Súmate collected 3.2 million signatures to recall the president. The signatures were rejected by the National Electoral Council (CNE) because they were collected prior to the midpoint of the presidential term and, thus, before they could be constitutionally acceptable for a recall. These signatures were stored by the CNE. The government raided the CNE offices in September 2003, jeopardizing the confidentiality of these signatures (McCoy 2005; Montoya 2008). When signatures were collected again by Súmate in November 2003, they totaled 3.6 million names (Madde and Kaltwasser 2012). This petition was rejected by CNE because of the high number of invalid names. This second failed attempt resulted in riots and civil unrest, during which 9 were killed, 339 were arrested, and 1,200 were injured. Súmate appealed to the Venezuelan Supreme Court to validate some of the rejected signatures (Hellinger 2005). The Electoral Chamber of the court reinstated many of the signatures, bringing the total valid signatures to 2.7 million. This was higher than the 2.4 million signatures required, forcing a recall election. However, this was overturned by the Constitutional Chamber of the court a week later. Subsequently, the government collected the signatures to ensure accuracy. With the government involved, this dramatically changed the tenor of the conversation about the recall. The names of those who signed the petition were posted online for the public to view, and many of these signatories faced personal repercussions for signing (Corrales and Penfold-Becerra 2007). Some lost their jobs (governmental or private), others were harassed, and some retracted their signatures to preserve their way of life. Despite the personal challenges for many voters, the CNE certified in 2004 that it had over 2.4 million signatures on the petition. This was a highly political process, with exchanges back and forth about the legality of signatures and forging of identifications. One important issue that influenced the recall election was the timing of the election. Timing influenced not only the petition’s acceptance, but also the results of the election. If Chávez had been recalled during his term his vice president would have served the rest of his term. If a recall had been successful prior to the fifth year, there would have been a new presidential election within 30 days. Thus, when the vote was called on August 15, 2004 (before Chávez’s

54

Direct Democracy: A Double-Edged Sword

fifth year), both opposition forces, who wanted a new election, and Chávez, who believed he would stand for reelection if recalled, were happy (Roberts 2012; Kornblith and Jawahar 2005). Once the petitions were certified and the election day set, a campaign ensued. There were arguments that foreign agents were involved in the election. Súmate received a grant from the United States to aid in voter education, but many Venezuelan officials argued that these funds were to assist in the recall election to oust Chávez. Despite arguments to the contrary, this tainted the campaign and gave increased support for Chávez. Despite low approval levels for Chávez, the polls showed consistently low support for the recall election throughout the campaign. The referendum question posed to the people of Venezuela was: “¿Está usted de acuerdo con dejar sin efecto el mandato popular otorgado mediante elecciones democráticas legítimas al ciudadano Hugo Rafael Chávez Frías como presidente de la República Bolivariana de Venezuela para el actual período presidencial? ¿NO o SÍ?” Translated into English this is: “Do you agree to revoke, for the current term, the popular mandate as President of the Bolivarian Republic of Venezuela conferred on citizen Hugo Rafael Chávez Frías through democratic and legitimate elections? NO or YES?” The question had some clear messages. The first was that this affected only the current term of Chávez and that he would be free to run again. Another message was that this was contrary to the democratic and legitimate elections conducted in December 1999. This implied that the recall election was not democratic (consistent with the theme of the campaign about US involvement and even the collection of signatures). Thus, the question itself conveyed a message about how the voters should participate. For the president to be recalled, three conditions had to be met. First, there needed to be a turnout of a minimum of 25 percent (or in this case, 3.56 million) of registered voters. Second, there had to be more votes against Chávez than the number that had voted for him in the previous election, which was 3.76 million (Jiménez 2011; Prado and Sansó 2011). Moreover, there had to be more yes votes than no votes cast. There was a high level of turnout (70 percent), much higher than expected based on turnout levels in presidential election years. However, there were disputes about some of the election day activities: there were delays at the polls, new technology (including fingerprint

Recall Elections: A Mixed Power

55

scanners), and multiple extensions of the closing time of the polls. The exit polls indicated that there was a huge margin for the anti-Chávez forces, but there were questions about the methodology and lack of political influence at these polls (Jiménez 2011; Prado and Sansó 2011). Just before 4 P.M. on election day, the CNE announced that someone had found an audio recording that declared the opposition forces had won and Chávez’s mandate was revoked. Because the polls had not yet closed (they did not close until 8 P.M. that night), many argued that this was an attempt to sabotage the referendum by keeping opposition voters home. There were calls for an inquiry into the incident. However, given the outcome of the election, little was done to address this (Prado and Sansó 2011; Montoya 2008; Taylor 2005). There were other arguments that there had been electoral fraud, yet the election observers, led by former US president Jimmy Carter, argued that no fraud had taken place (McCoy 2005; Montoya 2008). When looking at the requirements to remove Chávez, the election failed to meet all three. Only 5.8 million (59 percent) voters cast their ballots against the recall. Therefore, although the election met the first two conditions (turnout and more votes than those who voted for Chávez in 1999), it failed the third as more voters voted no than yes. While this clearly did not go the way that the petitioners had expected, it was a good exercise of the mechanism of recall elections. Certainly, there were issues with the election (from the petitioning process to the execution), and we now know the subsequent decay of democracy that occurred in Venezuela after this election. Nevertheless, the exercise of the recall election in 2004 became the most famous national executive recall internationally. Chávez continued to be president in Venezuela until his death in 2013. During that time, he won his reelection in 2007, continued to erode aspects of democracy in Venezuela, and extended presidential term limits so he could continue his presidency. Chávez, like Walker in Wisconsin, seemed buoyed by the recall election and saw the survival of his presidency as a way to continue his policy interests and increasingly authoritative reign. Chávez’s vice president Nicolàs Maduro Moros became president in 2013 after Chávez died. Maduro governed in the same manner as Chávez, but with less popularity. In October 2016, the recall process was initiated in an attempt to peacefully change the government. On October 20, however, five separate lower courts approved injunctions to suspend the recall (Toro 2016). Venezuela’s claim as a democracy has been tenuous at best, but this

56

Direct Democracy: A Double-Edged Sword

suspension virtually guaranteed conflict in the country. On October 24, the legislature called for the impeachment of Maduro.5 Taiwan

Taiwan instituted direct democracy at the national level in the 1989 Election and Recall Law in an effort to make the country more democratic (Copper 1990). This allowed for referenda by authorities and agenda as well as recall elections (Copper 1990). The Taiwanese Referendum Act was added in 2003 to provide for initiatives, assuming that signatures in support of the initiative could be collected from half of the voters who took part in the previous presidential election (Kobori 2014). Furthermore, there have been referenda in Taiwan of pivotal importance—for example, the 2008 referenda for joining the United Nations (both referenda failed due to low turnout). The 1989 Election and Recall Law provides that “a person duly elected may be recalled by his constituency in accordance with law,” including members of the central government (the National Assembly of the Legislative Yuan and of the Control Yuan) as well as members of the local governments. At the county and local level, Taiwan has a more open process that allows for the recall of local legislators, municipal mayors, council members, township chiefs, village leaders, and li leaders. The recall process at the national level for offices such as the president or vice president must be initiated by the legislative body (Copper 1990). This requires two-thirds of all members of the legislature to agree to recall the executive member, which is then followed by a popular vote of the eligible voters. The citizens and civil servants of Taiwan, while open to direct democracy procedures, have used them limitedly. Referenda have been used only three times, although there were multiple questions on the same ballot, and Taiwan has yet to successfully exercise recall elections at the national level. 2006 presidential recall attempt. On June 27, 2006, the Taiwan legislature called for a recall vote for embattled President Chen Shuibian. The recall motion proposed by the opposition party was supported by 119 of the 221 members, far short of the two-thirds majority needed. Chen battled issues of corruption in his administration and had tried to push for separation from China. Following this fail-

Recall Elections: A Mixed Power

57

ure in the legislature, there were many grassroots movements to remove Chen and, in the fall of 2006, more than 1 million signatures were collected, each accompanied with a $3 donation for the campaign against Chen. There were protests in the street by an estimated 250,000 people, but the legislature did not call for a further recall vote. Chen served as president from 2000 to 2008. After he was no longer president, Chen lost his immunity and was investigated for corruption and abuse of authority. He was found guilty and given a life sentence and fined $6.1 million, but was granted medical parole in 2015 (Hsu 2015). 2015 legislative recalls. There were early attempts at using recalls

in Taiwan in 1994 and 1995 over the issue of a nuclear power plant. The threshold for the turnout was increased to 50 percent just prior to the election, which prevented the Chinese Nationalist Party (KMT) legislators from losing their seats. Taiwan’s major test of its ability to apply direct democracy—specifically recall elections—came in 2015 with the attempted recall of three KMT legislators: Alex Tsai, Wu Yu-sheng, and Lin Hung-chih (Hsu 2015; Wen 2014). On February 14, 2015, there was an attempt to vote Tsai out of office. The explanations given for the recall were that Tsai failed to keep his campaign promises to voters, his conduct in the Legislative Yuan was unethical and unprofessional, and his management of KMT candidate Sean Lien’s mayoral campaign had been bungled. There had been low attendance in the Yuan since Tsai’s election (lower than those of the legislators in remote districts), and only two new laws and amendments had been sponsored (Wen 2014; Chen and Chung 2015). Thus, the recall moved forward. The recall campaign focused on the democratic means of direct democracy and Tsai’s record. Regardless of how successful the campaign was in convincing voters, it failed to get the voters to the polls. An overwhelming majority of the 79,303 votes cast supported the recall. Given that there were 317,434 eligible voters in Tsai’s constituency, 158,717 voters were required to vote for the results to be valid; this meant that turnout was only 24.98 percent (Wen 2015). The recall failed to reach the 50 percent voter turnout required by law. As a result, the Democratic Progressive Party called for changes to the Referendum Act to make the turnout threshold less restrictive in its usage. Tsai argued that the recall campaign was an epic failure, as it did not even reach close to the number of votes cast for him in

58

Direct Democracy: A Double-Edged Sword

his previous elections, and also argued that the recall was initiated by “professional students and political scam gang” and was not supported by the general public (Wen 2015). Remarkably, this was the first recall election in Taiwan demonstrating the ability to preserve the accountability of elected leaders. Though unsuccessful, it does demonstrate some support for the democratic structures of the government. The grassroots nature of the recall exhibited that this was an action of the people, but Tsai’s response continued to illustrate the problem with leaders who survive a recall election. These leaders appear to become more secure in their positions and more insulated from political accountability. Nigeria

Nigerian politics are plagued with corruption at every level (Adegbite 2015; Adedeji and Otite 1997; Tignor 1993), even though there have been serious efforts to combat this in recent years. The Nigerian government has to abide by specific regulations when it comes to recall elections (Adegbite 2016). There also have been efforts to change the behavior of elected officials to keep them in line with government regulations of official conduct. In fact, there were specific efforts to curtail the spending of leaders and the lavishness of elected officials when they travel overseas. However, in an effort to make the country more democratic, in 1999 the constitution was amended to provide for recalls (Lawal and Oladunjoye 2010). Sections 69 and 110 of the constitution provide for recalls in the federal and state legislative branches: A member of the Senate or of the House of Representatives may be recalled as such a member if: (a) there is presented to the Chairman of the Independent National Electoral Commission a petition in that behalf signed by more than one-half of the persons registered to vote in that member’s constituency alleging their loss of confidence in that member; and which signatures are duly verified by the Independent National Electoral Commission; (b) the petition is thereafter, in a referendum conducted by the Independent National Electoral Commission within ninety days of the date of receipt of the petition, approved by a simple majority of the votes of the persons registered to vote that in member’s constituency.

Recall Elections: A Mixed Power

59

Nigeria has a high signature threshold for a recall, requiring 50 percent of registered voters in the relevant jurisdiction to sign the petition. Once that threshold has been reached, there must be an election within 90 days, requiring only a simple majority of the voters in that electoral district (Adegbite 2016). Many countries constitutionally provide mechanisms of direct democracy but do not use them. So although Nigeria has recall mechanisms, there is no indication that Nigeria has ever successfully executed a recall election. The process was initiated against Hembadoon Amena, a State House of Assembly member for the People’s Democratic Party (PDP), in 2013 in the state of Benue. This process was stalled and never moved forward (Adegbite 2015). Amena did not run again in 2015 for the Benue assembly. Amena’s constituents alleged poor performance and ineffective representation (Abah 2014a). Signatures were collected and the referendum was scheduled for July 2014. It was postponed indefinitely, but not “canceled” (Abah 2014b). One of the reasons for postponement was the civil unrest in Amena’s Katsina-Ala West constituency, which resulted in property damage and arson in the district (Adegbite 2015). A second reason was a discussion over the collection of signatures and whether the threshold had actually been met (Abah 2014b). This is a clear example of wanting to make the mechanism available while not being committed to the process. By canceling or postponing a recall for reasons unrelated to the issue and limiting the power of citizens, the government decreases voter efficacy for these types of populist activities. Compromising the institution demonstrates a lack of commitment to democratic activities and a desire to squash civil unrest as well as to continue the governing process without interruption. Unrest is a form of civil disobedience, and there was potential for continued unrest with the government being unresponsive to the public. Amena’s party was in the majority at the time, and maintenance of every seat in the assembly was needed to continue their control. The PDP had been the majority party in Benue for 15 years and had seen a gradual decline in power. The election in 2015 changed the balance of power in the legislature; thus, it was paramount to postpone the recall and maintain the PDP’s power base.

60

Direct Democracy: A Double-Edged Sword

Analysis The relatively limited international diffusion and use of recalls suggests that this procedure is problematic in a representative democracy. These cases provide an intensive look at the use of recall elections across the globe and the different contexts that emerge. The comparisons of levels of government (national vs. state), the levels of accessibility, and the process and successfulness demonstrates that there is much to be learned from analyzing recall elections. Conducting recall elections requires many state resources, and petitioners need to be aware of the positive and negative consequences of signing a petition to recall an elected leader. While there is real power for voters, as seen in Venezuela and Wisconsin, signatories may be publicly exposed and this could have a significant impact on their employment and even their freedom. Thus, voters and petitioners alike need to consider the process seriously before moving forward. When looking at the differences in political success, comparisons can be made between Chen and Chávez, both nationally elected executive officials. However, the recall process went differently for each of them. Chen left politics in disgrace and ended up in prison even though the recall did not make it to the ballot, whereas Chávez became an even stronger leader after his political survival of a recall election. Tsai and Walker, although holders of lower offices, also became emboldened by their survival. It seems that the use of recalls can have the opposite of the expected effect. The level of usage, referring to the level of government and the frequency, indicates a level of confidence in the citizenry. When recalls are used frequently, there is more support for democratic activities. It makes sense that when the process is more open, it would be more democratic. However, there are also reasons not to use this mechanism and to allow govenrment officials the ability to govern. Governing is not always popular, resulting in discontent and a need to insulate from continuous recall attempts. Manipulation is clear in the use of recalls through campaigns and even the suspension of the process for political motives. The manipulation of the process by threatening those who participate makes voters fearful to use these types of mechanisms. If we possess these mechanisms, it is critical that citizens can use them without fear of reprisal. The recall is a powerful tool and should be used sparingly

Recall Elections: A Mixed Power

61

and cautiously to ensure that the process is respected and used for the correct reasons. Recall elections need to balance the principles of participation as well as effective governance. Wisconsin faced an expensive recall election 12 months after popularly electing a governor, and yet had no change in the leadership of the state. Balance is a key component to this and, while the vast majority of recalls are politically motivated rather than a result of maladministration (though I suppose that can be one and the same), making sure that a recall serves the interest of the state is paramount to its use. It is important to make these mechanisms available to voters, but also to insulate them so they are not conducted on a political whim. Notes 1. Eleven of those occurred in a single year in Wisconsin (2011). 2. It should be noted that two of these were for state senate presidents, indicating that there is no safe level of power in the legislature. 3. Governor Arnold Schwarzenegger served the term-limit maximum in California. During his tenure, he had tumultuous approval levels (both very high and very low)—humorously, only one point above Gray Davis’s approval rating at the time of the recall. Schwarzenegger turned to other methods of direct democracy to circumvent the State Legislature, which became gridlocked during his term. All four of the initiatives that he sponsored in a special election in 2005 failed. 4. One of the senate seats ended up being open due to the resignation of the incumbent before the vote. 5. At the time of this writing, Nicolàs Maduro Moros had not yet been impeached.

4 Initiatives: True Citizen Power or Facades?

to petition changes to the law through elections. The process varies in complexity and feasibility by country. Citizens in 36 countries are able to petition the government in the form of an initiative at the national level, although only 15 countries have utilized this mechanism (see Table 4.1), making initiatives slightly more used than the recall elections examined in the previous chapter. Furthermore, initiatives are also used at the regional and local level in 30 countries (see Table 4.2). Initiatives are often more of a bottom-up process that puts additional powers in the hands of citizens to determine laws and legislation in their countries. There are two general types of initiative processes: citizen (or popular) and agenda initiatives. 1 Popular initiatives are when citizens propose legislation or constitutional amendments and collect signatures for them to be placed on the ballot and voted on by voters. Agenda initiatives are when citizens propose laws or amendments and, once they receive enough signatures, they are submitted to the legislature for consideration rather than for the constituency to vote on. The focus of this chapter is on the popular or citizen initiative. The history of initiatives can be traced through populist movements around the world. The first iteration of initiatives started in Switzerland in the mid-1800s at the canton level. This was followed Initiatives, or popular referenda, are processes that allow voters

63

64

Direct Democracy: A Double-Edged Sword

Table 4.1 Usage of Citizens’ Initiatives at the National Level Frequent Users

Italy Liechtenstein Switzerland Uruguay

Source: Beramendi 2008.

Occasional Users

Bolivia Colombia Ecuador Federated States of Micronesia Hungary Latvia Lithuania Macedonia New Zealand Palau Peru Philippines Russian Federation Serbia Slovakia Slovenia Taiwan Turkmenistan Togo Uganda

Not Yet Utilized but Have the Mechanism Albania Belarus Cape Verde Costa Rica Croatia Georgia Liberia Malta Marshall Islands Nicaragua Republic of Moldova

in the United States in the late 1800s at the city and county level, then at the state level. More recently, there have been expansions into other regions. Tables 4.1 and 4.2 provide examples of the initiative process and how it is used around the world. Initiatives are a way for the citizenry to drive the legislative agenda, usually as a response to an unresponsive legislature (Phillips 1993; Dodd 1995). This is a way to circumvent the legislature or perceived issues of self-interest, corruption, or ineffectiveness in government. However, in recent years, the system has been pervaded by professionals and interest groups using this mechanism to further their agendas. Although a number of positive attributes associated with direct democracy apply to initiatives, there also are drawbacks specific to initiatives. There are opponents to the citizen-initiated referenda who have the same concerns as addressed in Chapter 1—that petitioners would be used to target minorities and strip them of valuable

Initiatives: True Citizen Power or Facades?

65

Table 4.2 Countries that Allow Initiatives at Regional and Local Levels Regional Level Brazil Colombia Ecuador Germany Hungary Italy Macedonia Mexico Norway Poland Peru Slovenia Sweden Switzerland United States

Local Level

Austria Bolivia Brazil Canada Colombia Croatia Czech Republic Ecuador Finland

Source: Beramendi 2008.

rights and freedoms (as seen in Switzerland for illegal immigrants and in the United States for same-sex marriage). Among the concerns about initiatives is the idea that citizens, who are unaware of the larger picture of government, propose and make laws binding the hands of government officials. Others are concerned that initiatives are often nonbinding, making them more advisory than effective. These are seemingly opposing arguments as some are concerned with the power of these referenda and others are concerned that they are not powerful enough. Beyond the proposal issues of initiatives, there are questions about the role of voters in the process. Because these elections are citizen initiated, the attention and media coverage can be limited. Thus, there is a concern about not only citizens proposing laws, but also how they will then vote on those laws. However, the literature indicates that even in this environment, voters are able to make informed choices consistent with their political preferences (Bowler and Donovan 2002b, 2004; Lupia 1994a, 1994b, 2001), and they are generally not reliant on spin and campaigns (Qvortrup 2015). This adds support to arguments for the use of initiatives to give voice to the public’s policy interests.

66

Direct Democracy: A Double-Edged Sword

Despite the unique opportunity for voters to propose laws, there are those who argue this mechanism can be abused by the electorate to hurt minorities. However, there are also examples where this is clearly not the case. The gay marriage initiatives in the United States in 2004 show how this mechanism was used to restrict minority groups. Alternatively, an example of when voters chose not to abuse the mechanism is the Swiss refusal to back anti-immigrant initiatives (Parkinson 2001). In recent years, there has been evidence that the initiative system has been pervaded by professionals and interest groups who use this mechanism to further their agendas (Manweller 2005). Initiatives have puzzled scholars, and few have focused their research solely on initiatives. One of the areas in which there is discord is turnout. Early research indicated that the inclusion of initiative mechanisms did not result in higher voter turnout (Everson 1981); however, more recent literature indicates that turnout is higher in places where initiatives are permitted on the ballot (Smith and Tolbert 2004; Smith 2002; Tolbert, Grummel, and Smith 2001). Another area of discord is the role of political parties in the initiative process. Early research indicated that political parties were not important to the process because they rarely received any benefits from these elections (Magleby 1989). In recent years, there has been evidence that political parties are critical to the initiative process— they provide heuristic cues, and they often provide opinions on the initiative and even sponsor them indirectly (Smith and Tolbert 2004; Buček and Smith 2000; Scarrow 1999). Initiatives provide scholars with the opportunity to investigate who proposes these measures. Mathew Manweller’s American study (2005, 2012) provides a typology to classify these petitioners: amateurs, zealots, professionals, lawyers, politicians, and victims. Typically, when we think of petitioners of initiatives, we expect that they lack experience or are amateurs. Manweller indicates that there are actually many types of petitioners of initiatives, and different motivations drive the use of these petitions. These types of petitioners are reflected in the types of measures that are presented. Throughout this chapter, I show that the type of petitioner is reflective of the type of success and the ability to overcome the barriers presented. Furthermore, there is evidence that politicians often turn to the initiative process for political motivation and to achieve goals that they are unable to realize in the leg-

Initiatives: True Citizen Power or Facades?

67

islature. These findings further demonstrate that Manweller’s typology has global applications. Case Studies In this chapter, I present five case studies, representing different usage as well as different regions: Italy, New Zealand, Germany, Colombia, and Kenya. The most frequent users of initiatives in this analysis are from Italy, where they frequently use two types of initiatives. Citizens of New Zealand are occasional users of initiatives but, because the initiatives are advisory, they are often ignored by the government. German citizens use initiatives quite frequently at the regional and local level. Finally, citizens of Colombia and Kenya have limited forays into the process with mixed results. These five cases provide insight into the initiative process and what interests and stakeholders drive the process in each instance. These cases depict a broad illustration of the initiative process in several contexts to fully illuminate the power and limitations of citizens through this mechanism. Italy

Italy has legal provisions for using initiatives at the national and regional level. Several articles of the Italian constitution from 1947 provide the foundation for initiatives, although Italy has had the ability to use direct democracy mechanisms since 1929 . The actual use of these types of elections dates back to 1974. There are two types of initiatives in Italy: citizen and abrogative. A citizen initiative provides for the submission of a new law or article, whereas an abrogative initiative permits a popular vote on a statute or law in part or in whole that is already in action (Uleri 2012). To call for a citizen initiative in Italy, a petitioner must collect valid signatures of 500,000 registered voters or be requested by five regional councils. 2 There is 6-month time limit to collect signatures. Often the popular referendum process (from collecting signatures until a vote) can take almost 2 years. Signatures in support of the petition must be accompanied by documentation and photo identification to prove that the people signing are registered voters and who they claim to be. This is an added burden that few other

68

Direct Democracy: A Double-Edged Sword

countries require. The petition is then filed with the Court of Cassation to determine the validity of the signatures. After the signatures are validated, the petition is presented to the Constitutional Court of Italy, which has the ability to reject the petition for cause.3 Assuming the Constitutional Court approves the subject matter, the petition would then be submitted to the president of the republic for a vote. There is a quorum requirement of 50 percent plus one of all eligible electors to vote on the referendum. If this threshold is not met, the referendum will fail. Most recent popular referenda have been unsuccessful because they have failed to meet the threshold requirement. The high threshold requirement has been a criticism of the process. Furthermore, initiatives are limited in which topics can be proposed; they cannot address issues of international treaties, taxes, public expenditures, or amnesty and pardon issues. For an abrogative initiative, a minimum of 10 voters must demand a vote. Abrogative initiatives are a powerful tool of the people to control the decisions made by the legislature, putting the power in their hands to essentially veto legislation. However, it can also have the opposite effect and instead strengthen support for legislation. These are amazingly common; there were 30 abrogative initiatives in a 1995 alone (Uleri 2011).4 An abrogative initiative requires substantial constitutional knowledge to create a justification for the Constitutional Court. Once presented, the court has to adjudicate on whether it is admissible. If approved, it is then set for a vote (Uleri 2012). There is no time line for when these laws are enacted to when an abrogative vote or petition must occur, so these can be recent or old laws. For recent laws, this gives the people specific checks and balances on the system; with older laws, it allows citizens to drive the agenda of government to focus on the amendments to the legislation that they want. Regardless of the number of abrogative initiatives, they are all voted on at the same time, sometime between April 15 and June 15. Generally, there are several (as many as 12) issues voted on (Uleri 2012; Uleri and Fideli 1996). Between 1948 and 2010, there were 62 abrogative initiatives voted on by the public. Of these, only 19 were approved by voters (Uleri 2012).5 Oddly, these abrogative initiatives can be broken into periods where, as a whole, these measures are accepted or rejected.6 The vast majority of initiatives (particularly abrogative initiatives) are a function of larger groups rather than specific individu-

Initiatives: True Citizen Power or Facades?

69

als (see Table 4.3). For example, Catholic organizations have played a large role in the petitioning process to change laws to better reflect civil society’s interests (Uleri 2011). Additionally, political parties play a large role in the process. While citizen driven, the vast majority of initiatives are actually sponsored by a political party as a way to circumvent the legislative process. The Radical Party has presented the most petitions and receives less than 3 percent of the popular vote each election, challenging the notion that this is a power of the people. It has also worked in concert with other small political parties to sponsor these referenda. Unsurprisingly, the focus of many of these popular referenda has been electoral reform or other criticisms or attempts to restrict the power and business interests of former prime minister Silvio Berlusconi. However, through these mechanisms, groups have also championed issues surrounding divorce, abortion, and embryonic research. This is not an unusual use of popular referenda, but it does demonstrate a consistent issue within Italian government that political parties can propose these measures and that these parties feel the need use these mechanisms to enact change. These politician-type petitioners are using the system for their own political gain rather than working within the traditional norms of governing. Even though politicians are some of the more prolific users of this mechanism, initiatives play a minor role in Italian politics. Despite this role, the use of the system by political parties and civil society groups demonstrates that this is not just a function of amateur voters, but rather a way to manipulate the system to focus on issues that these groups are unable to affect through legislative means. This illustrates that it is not a function of the people, as the populists intended, but a function of professional groups to pursue alternative policy goals. New Zealand

New Zealand has a varied experience with direct democracy, including referenda and initiatives (Morris 2014).7 New Zealand has utilized direct democracy since 1911, but it was not added to the electoral procedures until 1993. Much of the early pressure to utilize these mechanisms came from the Women’s Christian Temperance Union, which had also influenced Australia. The nine referenda held between 1911 and 1993 focused on two areas: morality (alcohol and

70 Table 4.3 Italian Popular Referenda Title

Date

Referendum on Divorce

1974

Police Powers Referendum

1978

Party Funding Referendum

1978

Liberalizing Abortion Laws Prohibiting Abortion Police Powers Referendum Life Imprisonment Referendum Gun License Referendum

1981

Referendum on Abolishing the Wage Escalator Referendum on Location of Nuclear Plants Referendum for Nuclear Plants

Referendum on Nuclear Plants Abroad Referendum on Judges’ Civil Responsibility Referendum on Ministers’ Board of Inquiry Referendum on Law Restricting Hunting Referendum on Hunting on Private Property

1981

1981

1981 1981 1985 1987 1987

1987 1987 1987 1990 1990

Question

“Do you want the Law of 1 December 1970, No. 898, on the regulation of cases of dissolution of marriage, to be abrogated?” Referendum on the repeal of the Reale Law, which defined the powers and engagement of the Italian police during riots or simple street protests. Referendum on the repeal of a law establishing public funding of the parliamentary parties in an attempt to stop the funding of parties by big industrial groups and interests. Referendum on abolishing the remaining limits on abortion. Referendum on restoring the ban on abortion.

Second referendum on the repeal of the Reale Law. Referendum asking voters to reject life imprisonment as the highest level of punishment for crimes. Referendum to repeal the law on licenses to guns, allowing the police to give weapons to some citizens. Referendum on repealing the wage escalator, which automatically increases the salaries of Italian workers at the same rate as inflation. Referendum asked to abolish the power of the state to ask local administrators to accept new nuclear plants in their territories. Referendum to abolish rewards for local administrations with nuclear and coal plants in their territory. Referendum to prevent the National Board for Electricity from building nuclear power plants outside of Italy. Referendum to abolish law that gave judges immunity from responsibility for judicial errors. Referendum to abolish the law of immunity for parliamentary ministers.

Referendum on repeal of law that restricted hunting.

Referendum on repealing the law allowing hunting on private properties.

Result

No (59.3%) Turnout 87.7%

No—keep the law (76.5%) Turnout 81.2%

No—keep the law (56.4%) Turnout 81.2%

No (88.4%) Turnout 79.4% No (68.0%) Turnout 79.4% No (85.1%) Turnout 79.4% No (77.4%) Turnout 79.4%

No (85.9%) Turnout 79.6%

No (54.3%) Turnout 77.9%

No (80.6%) Turnout 65.1%

No (79.7%) Turnout 65.1%

No (71.9%) Turnout 65.1%

No (80.2%) Turnout 65.1%

No (85.0%) Turnout 65.1%

Yes (92.2%) Turnout 43.3%

Yes (92.3%) Turnout 43.3%a continues

71 Table 4.3 Continued Title

Referendum on Health

Referendum on Electoral Law

Referendum on Local Health Units Referendum on Medicinal Drugs Referendum on Public Banks Referendum on Senate Elections Referendum on the Ministry of Agriculture Referendum on the Ministry of State Holdings Referendum on the Ministry of Tourism Referendum on Union Representation Referendum on Union Representation (Partial Repeal) Referendum on Union Dues Referendum on Collective Contracts Referendum on Exile of Mafia Members Referendum on RAI (Radiotelevisione Italiana) Referendum on Commercial Licenses

Date

1990 1991 1993 1993

1993

1993

1993 1993 1993 1995 1995

1995

1995 1995 1995 1995

Question

Referendum on repealing the right of the Italian Ministry of Health to establish limits on pesticides. Referendum on abolishing the preference votes in the Proportional Representation election system. Referendum to repeal the law limiting the intervention of local heath units on environmental problems. Referendum to repeal the law limiting the use of medicinal drugs. Referendum to repeal the law regulating the administration of public banks. Referendum to repeal the law on using proportional representation in senate elections. Referendum on the abolishment of the Ministry of Agriculture. Referendum on the abolishment of the Ministry of State Holdings.

Referendum on the abolishment of the Ministry of Tourism. Referendum repealing the law on union representation.

Result

Yes (93.5%) Turnout 43.5%a

Yes (95.6%) Turnout 62.5%

Yes (82.5%) Turnout 76.9%

Yes (55.3%) Turnout 77.0% Yes (89.9%) Turnout 77.0% Yes (82.7%) Turnout 77.1% Yes (70.1%) Turnout 77.0%

Yes (90.1%) Turnout 76.9%

Yes (82.2%) Turnout 76.9%

Yes (50.0%) Turnout 56.9%

Referendum on a partial repeal of the laws on union representation.

Yes (62.1%) Turnout 56.9%

Referendum on repealing the law that allows for the direct deduction of union dues from salaries. Referendum on repealing the law on collective contracts for public sector workers.

Yes (56.2%) Turnout 56.9% Yes (64.7%) Turnout 56.9%

Referendum on repealing the law on the internal exile of mafia members.

Referendum to repeal the public ownership of RAI and allow for partial privatization.

Referendum repealing the laws that allow one person to own multiple television channels.

Yes (63.7%) Turnout 57.0%

Yes (54.9%) Turnout 57.2%

No (64.4%) Turnout 57.0%

continues

72 Table 4.3 Continued Title

Referendum on Television Advertising Referendum on Film Advertising Referendum on Publicity Grouping Referendum on Shop Hours Referendum on Local Council Elections Referendum on Minister of the Treasury Referendum on Conscientious Objectors Referendum on Hunters Referendum on Judges Referendum on the Order of Journalists Referendum on Judicial Work

Date

1995 1995 1995 1995

1995

1997 1997 1997

1997 1997

1997

Referendum on 1997 Ministry of Agrarian Politics Referendum on 1999 Electoral Law

Referendum on 2000 Election Spending Referendum on 2000 Proportional Representation Referendum on 2000 Superior Council Judges Elections Referendum on 2000 Magistrates

Question

Result

Referendum repealing the law that allows for advertisement breaks during films.

No (55.7%) Turnout 57.9%

Referendum repealing the laws that allow advertisers to sell ads to multiple stations.

Referendum repealing the law allowing television publicity monopolies.

Referendum repealing the law on shop opening hours. Referendum repealing laws dealing with local council elections.

Referendum repealing law on how much the minister of the treasury receives during privatization. Referendum on repealing the law that restricts conscientious objectors.

Referendum on repealing law that allows hunters to hunt on private property. Referendum to repeal the law that restricts judges’ careers. Referendum repealing the law on who can be admitted to the Order of Journalists.

Referendum repealing the law that allows judges to have other employment outside of their role as judges. Referendum to abolish the Ministry of Agrarian Politics.

No (57.0%) Turnout 57.9%

Turnout 57.8% No (56.4%)

No (62.5%) Turnout 57.8% No (50.6%) Turnout 57.1%

Yes (74.1%) Turnout 30.1%a

Yes (71.7%) Turnout 30.1%a

Yes (80.9%) Turnout 30.2%a Yes (83.6%) Turnout 30.2%a Yes (65.5%) Turnout 30.0%a Yes (85.6%) Turnout 30.2%a

Yes (66.9%) Turnout 30.0%a

Referendum to change the electoral system from mixed-member proportional to single-member districts. Referendum to repeal law that allows for reimbursement of election spending. Referendum to change the electoral law away from proportional representation.

Yes (71.1%) Turnout 32.2%a Yes (82.0%) Turnout 32.4%a

Referendum to separate magistrates between judicial positions and government ministries.

Yes (69.0%) Turnout 32.0%a

Referendum to repeal the proportional election of Superior Council judges.

Yes (91.5%) Turnout 49.5%

Yes (70.6%) Turnout 31.9%a

continues

73 Table 4.3 Continued Title

Referendum on Judges Referendum on Termination of Employment Referendum on Trade Union Dues Referendum on Termination of Employment Referendum on Electric Cables Referendum on Embryos Referendum on Embryo Research Referendum on Personhood Referendum on In Vitro Fertilization Referendum on Chamber of Deputies Referendum on Senate Referendum on Chamber of Deputies Election Referendum on Privatization of Water Services Referendum on Water Services Referendum on Nuclear Power Referendum on Legal Impediments

Date

2000

2000

2000 2003 2003

2005 2005

2005

2005

2009

Question

Referendum to repeal law that allows judges to hold other positions. Referendum to repeal the law that requires employers to rehire employees fired without sufficient cause. Referendum to prevent trade unions from automatically collecting dues.

Result

Yes (75.2%) Turnout 32.0%a No (66.6%) Turnout 32.5%a Yes (61.6%) Turnout 32.2%a

Referendum to require small companies to rehire workers fired without sufficient cause.

Yes (86.7%) Turnout 25.5%a

Referendum on repealing the law that defines embryos as people. Referendum to allow in vitro fertilization treatment with donated eggs or sperm.

Yes (87.7%) Turnout 25.7%a Yes (78.2%) Turnout 25.6%a

Referendum to prevent the installation of electric cables on private property. Referendum to repeal the limitations on clinical and experimental research on embryos. Referendum to repeal the limitations on access to embryos for research.

Referendum on election matters.

Yes (85.6%) Turnout 25.6%a Yes (88.0%) Turnout 25.7%a Yes (88.8%) Turnout 25.7%a

Yes (77.6%) Turnout 23.3%a

2009

Referendum on election matters.

Referendum to allow a candidate to run in more than one constituency.

Yes (77.7%) Turnout 23.3%a Yes (87.0%) Turnout 23.8%a

2011

Referendum to repeal the law that allows for privatization of water services.

Yes (95.4%) Turnout 54.8%

2009

2011

2011

2011

Referendum to repeal the regulations on tariffs for water supply services. Referendum to repeal laws that allow nuclear power plants on Italian territory. Referendum to repeal the law that allows for government officials to be excused from court appearances if they were prevented from doing so because of government commitments.

Yes (95.8%) Turnout 54.8% Yes (94.1%) Turnout 54.8% Yes (94.6%) Turnout 54.8%

Sources: Data compiled from Nohlen and Stöver 2010. Note: a. Invalidated by the courts because they did not have a high enough turnout to meet the threshold, and so failed regardless of vote.

74

Direct Democracy: A Double-Edged Sword

gambling) and election reforms (Parkinson 2001). The focus on election reforms put pressure on the government. The Royal Commission on the Electoral System began investigations into different forms of direct democracy, namely referenda and initiatives. Though they made significant changes to the electoral system in 1996, the Royal Commission continued to reject moves toward a greater use of direct democracy (Parkinson 2001; Royal Commission on the Electoral System 1986). The seminal event that led to the use of initiatives in New Zealand was the introduction of a bill in 1986 by Fran Wilde, a Labour member of Parliament, to decriminalize homosexuality and eliminate discrimination based on sexual orientation (Royal Commission on the Electoral System 1986). The debate that followed called on the moral issues that had instigated the Women’s Christian Temperance Union’s work in the early 1900s. A petition was circulated and 835,000 signatures were collected in opposition to the bill, calling into question whether the government should act against the preferences of the people (Parkinson 2001). Opponents to the legislation began promoting initiatives and referenda as a way to stop legislation and agendas that did not reflect the wishes of New Zealanders. Albeit, the motivation of these petitioners was not more democratic openings, but rather to continue their agenda against gay rights and other social liberalism. These groups— namely the Coalition of Concerned Citizens and One New Zealand Foundation—made several submissions to the Royal Commission, forcing it to consider the risk of tyranny of the majority. Thus, the Royal Commission continued to reject these mechanisms, as they would undermine government actions and could adversely affect minorities (Parkinson 2001; Royal Commission on the Electoral System 1986). Despite this rejection, National Reform, a lobby group of the National Party, allied with these groups and made this the focus of the party’s conference in 1989. The National Party included the creation of citizen-initiated referenda in its election platform for 1990. In its success in the 1990 election, the National Party introduced the Citizens Initiated Referenda Bill in 1992, and it was passed in 1993, thus beginning the use of the mechanism in New Zealand. New Zealand’s initiatives are nonbinding and can be proposed by individuals or groups (Bowler and Donovan 2004). Given how initiatives are proposed and enacted in New Zealand, this advisory

Initiatives: True Citizen Power or Facades?

75

nature of initiatives is important to protecting the states.8 Furthermore, the government is able to (and does) ignore the results of these initiatives (Parkinson 2001). This means that initiatives are little more than a national poll, but with a bit more influence. The petition must be authorized by the clerk of the House of Representatives. The petition is distributed to the public for comments, and the clerk establishes the final wording of the petition prior to the collection of signatures (this process generally takes about 3 months). Despite the citizen-initiated process, the clerk of the House, rather than the proponents of the initiative, has the final say on the wording that will appear on the ballot. Once approved, the petitioners must collect signatures from 10 percent of registered voters within 12 months.9 The signature threshold was set high enough to ensure that only essential issues are proposed and the system is not hijacked by special interests (Catt 1996), but not so high as to prevent the use of initiatives. Therefore, although it opened the door to greater citizen involvement, there are still hurdles in place to protect democracy. The signatures are then verified by the clerk. Once qualified, the petition is submitted to Parliament, and within 30 days the governor-general sets a date for the referendum to occur within 12 months of the presentation (Parkinson 2001; Vowles 1995; Hughes 1994).10 This is substantially less restrictive than Italy’s process. Traditionally, the vote is held with a general election, although this is not mandated by law. While there have been fewer than 60 proposals submitted to the clerk since 1993, many did not make it through the signature process or were withdrawn by the petitioner. These proposals have been based on a number of topics (e.g., education, environmental issues, criminal justice, health care, and conservation). The campaign is limited based on time, but also on expenditures. Petitioners and opponents can spend only NZ$50,000 (roughly US$36,000) to promote or defeat the petition. A caveat to this is that the Royal Commission set no spending cap on informing the public of the issues; however, funding disclosures and reviews may be necessary if the spending becomes too large. New Zealand includes several barriers to the process for a number of reasons, most of them stemming from the way citizen-initiated referenda were introduced in New Zealand. The nonbinding nature, control by the House clerk over the wording, spending caps, and high signature quota insulated the process from abuse by special interests

76

Direct Democracy: A Double-Edged Sword

and limited minority populations. The cost of conducting a referendum is prohibitive (NZ$10−11 million, roughly US$7.5 million) so these hurdles also protect the system from continued reckless spending. Additionally, this provides a protection of policy in important areas that many feel are too important for citizens to control (Catt 1996). Table 4.4 provides a list of the petitions that made it to a vote— all but one were proposed by individuals rather than groups. The first initiative in 1995 sought to reduce the number of professional firefighters. Despite an overwhelming no vote of 88 percent, the number of firefighters was reduced in successive years due to budgetary constraints and policy needs. So while the public had a voice, the overwhelming needs of the government (budgetary and civil service) played a larger role. The vote on this initiative was not held in conjunction with a general election, and thus there was a low turnout of 27 percent. There were two initiatives in 1999. The first dealt with electoral reform, reducing the number of the members of Parliament from 120 to 99 (which had recently been increased by recommendation of the Royal Commission). This initiative was supported by 81.5 percent

Table 4.4 New Zealand Initiatives Date

Dec. 2, 1995 Nov. 21, 1999 Nov. 27, 1999 July/Aug. 2009 Nov./Dec. 2013

Individual or Group Sponsorship

Group: Firefighters Union

Individual

Individual

Individual

Individual

Source: Parkinson 2001.

Question

“Should the number of professional firefighters employed full-time in the New Zealand Fire Service be reduced below the number employed on 1 January 1995?” “Should the size of the House of Representatives be reduced from 120 members to 99 members?” “Should there be a reform of our justice system placing greater emphasis on the needs of the victims, providing restitution and compensation for them and imposing minimum sentences and hard labor for all serious violent offences?” “Should a smack as part of good parental correction be a criminal offence in New Zealand?” “Do you support the Government selling up to 49% of Meridian Energy, Mighty River Power, Genesis Power, Solid Energy and Air New Zealand?”

Results

No (87.8%) Turnout 27.0% Yes (81.5%) Turnout 84.8% Yes (91.8%)v Turnout 84.8% No (87.4%) Turnout 56.1% No (67.3%) Turnout 45.1%

Initiatives: True Citizen Power or Facades?

77

of the vote. Because initiatives are advisory, the government did not move to change the number of members. A few years later, in 2006, a bill was introduced in Parliament to reduce the number of members to 100; however, this was voted down because members found it incongruent with the recent change to the mixed-member proportional electoral system. The second initiative election held in 1999 looked at the reformation of the judicial system to provide restitution and compensation to victims while imposing minimum sentences and hard labor for serious offenses. This measure passed with 91.8 percent of the vote. The motivation for this initiative was twofold. First, there had been an assault on an elderly woman in Christchurch, and the media coverage focused on vengeance and punishment. Second, the recently elected Labour government’s new justice minister expressed a preference for restorative justice, not tougher sentences. The initiative sought to provide justice for victims and focused on punishment. Despite these issues, the government did not enact the changes proposed in the initiative. There was a low information environment surrounding these two initiatives, and both were clearly focused on changing or preventing government reforms that had been the platform of the 1999 election (Parkinson 2001). These measures were held in conjunction with the general election of 1999 and saw a voter turnout of 82.8 percent, but the mandate of the elected officials won out over these initiative proposals. Like the 1995 initiative, the 2009 initiative was not held in conjunction with a general election. Rather, the citizen initiative referendum was held by mail in August 2009. The question was on the criminality of parental spanking of children. Turnout in 2009 was higher than in 1995, at 56 percent. Despite an overwhelming no vote of 87.4 percent (meaning that the vote was not in favor of criminalizing spanking), there was discord regarding how the question was written, with opponents arguing that the question itself was value laden.11 The petitioners of these initiatives were focused on specific issues facing society. The 2013 initiative took place by mail from late November to early December 2013. The initiative looked at the government’s policy on privatizing four energy-related, state-owned companies and reducing the government’s ownership of Air New Zealand. There was conflict over the collection of signatures; the groups collecting signatures submitted more than the 10 percent threshold, but upon

78

Direct Democracy: A Double-Edged Sword

verification it was found that there was a shortage, and petitioners were given 2 more months to obtain extra signatures. Once verified, the referendum was set. From the beginning of the process, Prime Minister John Key indicated that the government would ignore the outcome and proceed with the sale, citing the 2011 election as having given it a mandate to reduce government ownership. The referendum had a 45 percent turnout, and 67 percent voted against the sales. Despite these outcomes, the sales went ahead as planned. A feature of this initiative, like the firefighter reduction initiative, was the fiscal constraint on government; these restraints and needs went beyond the public discourse. One unique feature of initiatives in New Zealand is that, although the initiation came from special interest groups (National Party, Coalition of Concerned Citizens, etc.), most proposals have come from amateur petitioners. Therefore, initiatives have not become a tool of the elite (even the role of the Firefighters Union in 1995 proved less professional because of the first and only attempt). New Zealand provides an excellent example of the tension between government and the people in policymaking through initiatives. This tension reflects the concerns about using initiatives to make policies, but also why they are used as advisory rather than binding. Germany

Germany is a federal state, composed of 16 states that are the primary administrators of the government. Federal states have authority on transportation, education, policing, and the environment. Furthermore, there are 13,854 local authorities who make decisions at the local level on issues such as taxation, energy, infrastructure, and sanitation. None of the Bundesländer states originally allowed for binding referenda, but most allow for nonbinding ballot questions. As direct democracy was introduced in the communities, nonbinding public initiatives (Bürgerbegehren) were expanded. Some states created the general right to state popular initiatives (Volksbegehren). Germany’s experience with direct democracy has been on the incline since reunification (Kampwirth 2004; Schiller 2011). The vast majority of these experiences have been at the state level. The year 1990 was important to German politics and a turning point for the use of the direct democracy mechanism because these instruments became adopted or improved at the state and local levels, and all states

Initiatives: True Citizen Power or Facades?

79

introduced direct democracy at the municipal level (Schiller 2011). Between 1990 and 2004, Germany held 145 statewide popular initiatives and 10 citizen-initiated referenda. One issue that has plagued the use of initiatives and referenda in Germany is that they are not particularly citizen friendly, which limits the power of citizens in the process. Looking at the use of initiatives and referenda at the regional and local levels, 6 of the 11 states of the Federal Republic had initiatives and referenda in their constitutions starting in 1945, and all 16 states introduced some form of direct democracy after reunification. Four of the 16 federal states have provisions for full-scale citizen initiatives (Bavaria, Baden-Wuerttemberg, Hesse, and Sarre), but do not have provisions for agenda initiatives. Four others require an agenda initiative as a precursor to the citizen initiative process (Brandenburg, Hamburg, Saxony, and SchleswigHolstein). Additionally, six states (Berlin, Bremen, Lower Saxony, Mecklenburg, West-Pomerania, North Rhine-Westphalia, and Thuringia) provide for only agenda initiatives. The last two states (RhinelandPalatinate and Saxony-Anhalt) provide for the option of an agenda initiative as a precursor to a citizens’ initiative. The requirements for initiatives in these states are included in Table 4.5. Germany allows for indirect initiatives where voters collect signatures to request action by the government, which are nonbinding at the state level and, in some places, the municipal level. At the state level, there are three types of direct democracy (Schiller 2012): • Volksbegehren (people’s request) is a citizens’ initiative—if the government ignores the request, it could directly lead into a Volksentscheid. • Volksbefragung (people’s inquiry) is nonbinding facultative ballot question and the most common type of referendum. • Volksentscheid (people’s decision) is a binding plebiscite, but this type is used only if the constitution requires it.

Focusing on the people’s initiative (Volksbegehren), there are three variations at the municipal level (Schiller 2012): • Bürgerbegehren (citizens’ request) is a local citizens’ initiative, • Bürgerbefragung (citizens’ inquiry) is a local ballot question, and • Bürgerentscheid (citizens’ decision) is a local plebiscite.

80

Direct Democracy: A Double-Edged Sword

For the specific regulations of each state, see Table 4.5. Full-scale initiatives have been used since 1945, while agenda initiatives have been more frequent since the 1990s. City-states also have the opportunity to focus on citizen initiatives. Table 4.5 German State-Level Laws on Initiatives Legislative Body BadenWürttemberg Bavaria Berlin

Brandenburg Bremen

Hamburg Hessen

State Signatory Quorum

16.67% of registered voting population must sign petition. 10% of registered voting population must sign petition. 7% of registered voting populationmust sign petition (20% for constitutional changes). 80,000 signatures (200,000 when demanding reelections). 10% of registered voting population must sign petition (20% when demanding constitution changes or reelections). 5% of registered voting population must sign petition. 20% of registered voting population must sign petition. 120,000 registered voters.

Mecklenburg West-Pomerania Lower Saxony 10% of registered voting population must sign petition. North Rhine8% of registered voting population Westphalia must sign petition within 8 weeks. Rhineland300,000 signatures. Palatinate Sarre 20% of registered voting population must sign petition. Saxony 450,000 signatures. Saxony-Anhalt 11% of registered voting population must sign petition. Schleswig5% of registered voting population Holstein must sign petition. Thüringen 10% of registered voting population must sign petition. Federal Republic of Germany

10% of the voters in the affected area.

Local-Level Signatory Requirement

5−10% of registered voters 3−10% of registered voters 3% of registered voters

Time Limit

Voter Turnout Threshold

14 days

10−20%

14 days

4 months

25%

15%

10% of registered voters 5% of registered voters

4 months

25%

2−3% of registered voters 10% of registered voters 4.2−10% of registered voters 10% of registered voters 3−10% of registered voters 6−15% of registered voters 5−15% of registered voters 5−15% 6−15% of registered voters 10% of registered voters 6−7% of registered voters

21 days

Majority only 25%

3 months

14 days

No time limit 6 months 2 months

20%

25%

25%

20%

30%

14 days

30%

6 months

20%

6–8 months 25% 6 months 25%

4 months

10−20%

Sources: Compiled from Shiller 2011b; Kampwirth 2004. Note: For constitutional referenda, all states have a minimum approval quorum of 50% except Bavaria which requires 25%.

Initiatives: True Citizen Power or Facades?

81

On average, 200 local direct democracy elections are held in Germany each year with the majority in Bavaria (Kampwirth 2004; Schiller 2011).12 The process has been blocked continually at the national level out of fear of the people’s power to change legislation outside the traditional legislative process.13 The extensive use of direct democracy at the local level provides citizens the opportunity to influence policy from the grassroots. While there are variations in how each state and locality conducts an initiative, some general processes are followed. To start, citizens must present a formal application with the proposed change. There are limitations on the topic—constitutional issues are excluded in Berlin, Hesse, and the Saarland, and economic issues are excluded in all jurisdictions (Schiller 2011). Once the subject matter has been approved, the proposal must include a financial statement, which encompasses the cost of the measure and recommendations on funding. The petition then needs to be accompanied by signatures. The next step is to collect signatures—there is a signature quorum, and there are timing issues that vary by state. Some regulations are as restrictive as 2 weeks, and some are more permissible, up to 12 months. Once the signatures are collected, the applications are checked to ensure that a quorum of signatures has been collected (which varies from 3,000 to 120,000 signatures depending on the jurisdiction—see Table 4.4 for details). Once approved, an election is scheduled with a legally binding result. Some states require a threshold of participation while others do not. Overall, 145 popular initiative petitions have been started, but only 10 have resulted in an election at the state level. The vast majority of these occurred in Bavaria, making it the only state in Germany to regularly use this type of direct democracy. After the initial successes in Bavaria and Hamburg, state governments and constitutional courts blocked future initiatives. The argument for limiting initiatives was that having citizens involved in the lawmaking process directly violated the norms of German democracy. This illustrates not only the lack of support that the mechanisms have in the government, but also the lack of trust in the citizenry to make good decisions in the initiative process. Despite this overarching poor opinion of the citizenry, state parliaments in Bremen, Hamburg, Nordrhein-Westfalen, and RheinlandPfalz decided to lower the barriers to direct democracy at the state and local levels, albeit these attempts to make it more accessible were fairly limited (Kampwirth 2004; Schiller 2011).

82

Direct Democracy: A Double-Edged Sword

At the municipal level, the most common use of direct democracy has been in the form of citizen initiatives or popular referenda. Local councils also have the power to call for referenda, but mandatory referenda are not available at the local level. Recall elections are not initiated by the public; they must be generated by the local council. Therefore, while different direct democracy measures exist at the local level, citizen initiatives are the most common and the largest area for direct influence on policies. The rules for municipal initiatives are set by the regional states. Initiatives can be adopted by the local council or put to a vote (Schiller 2011). At the local level, many authorities began widespread reform during the lead-up to reunification; all but one state has direct democracy. Like at the state level, signatures are required to propose a popular initiative, and the signature quorum depends on the size of the community, from 2 percent to 20 percent. Generally, there is no time limit for signature collection; however, if the initiative is against a specific action of the local authority, the time limit is restricted to between 4 weeks and 3 months, depending on the jurisdiction. There are some limitations on issues at the local level. Most states exclude important local issues, and only Bavaria, Hamburg, Hesse, and Saxony allow initiatives on any topic. Should the local council decide to accept the initiative, there will not be an election on the issue. If not, a vote is scheduled. For an initiative referendum to be accepted, it needs to reach a turnout threshold of 10−20 percent, again dependent on the size of the community.14 Most states have imposed an exclusion period of 1 to 3 years to prevent repeal of a successful local referendum by the council, and the referendum can be changed only by a new referendum. Despite the relative accessibility and use of initiatives at the state level, they are far more frequent in smaller municipalities. Between 1956 and 2009, there were a total of 5,534 initiatives or referenda at the municipal level. 54 percent of those were held in municipalities with fewer than 10,000 people, and another 23 percent in municipalities with 10,000−50,000 people (Schiller 2011). The Germany case study provides insight into how initiatives can be used at the state and local level with vigor. Their exceptional use at these levels demonstrates the role that citizens can have in the process of governing through initiatives. Using initiatives at the local level provides an opportunity for citizens to directly impact policy when the stakes are lower and there is less fear about citizen

Initiatives: True Citizen Power or Facades?

83

involvement. As such, it seems that lower stakes and a more permissible system result in higher usage of initiatives. Colombia

Colombia has had direct democracy mechanisms since 1957 and has utilized national referenda six times, one of which was a private initiative. Initiatives are allowed at the national, regional, and local levels in Colombia along with referenda and recalls. Individuals and civil society organizations (e.g., civic, indigenous, professional, or labor groups) can propose an initiative. However, petitioners must create and register an initiative committee of nine citizens to move forward in the process. Initiatives require 0.05 percent of registered voters, which is a very low requirement for signatures. To be submitted, an application requires a proposed title, a general explanation, and a full text describing the proposed changes. These are determined by the proponents of the initiative but require oversight from the legislature. The legislature is responsible for determining the legality of the proposal. Initiatives are limited to one topic and cannot be commercial in nature; that qualifies as an advertisement (Cardenas, Junguito, and Pachón 2006). Furthermore, a number of topics cannot be addressed in an initiative: international treaties, taxes and expenditures, public order, amnesty, and pardons. However, it should be noted that initiatives in Colombia are advisory to the legislature. Given the ease of proposing a referendum, it is surprising that more are not initiated in Colombia. The lack of usage is usually blamed on how difficult it is to propose these measures. However, the stress of civil conflict and the outcome of the 1997 initiative may contribute to the lack of desire to utilize this type of measure to achieve goals. Countries that are immersed in civil conflict may turn to these measures, but it is difficult to continue to advocate for their use when they are unsuccessful. The October 26, 1997, Colombian referendum was a private initiative. It was sponsored by members of civil society, religious groups, and others tired of the violence and kidnappings plaguing their country (Gomez 2001). Colombia had been experiencing a spiral of violence and an uptick in coca production, leading to even greater civil conflict between the guerrillas, the government, and the people. The 1997 initiative was proposed to stop the conflict and

84

Direct Democracy: A Double-Edged Sword

begin the peacebuilding process. It was unofficially titled “Peace, Life, and Liberty”: I AM VOTING FOR PEACE, LIFE AND FREEDOM. —I promise to be a builder of peace and Social Justice, to protect life and reject all violent action and welcome children mandate for peace. I demand the actors in the armed conflict: —NO more war: peacefully to resolve the armed conflict. —NO more atrocities: respect for international humanitarian law. YOUR VOTE − THE VICTORY OF PEACE

The referendum was held in conjunction with other elections, and there were high levels of turnout in urban areas and less in rural areas due to pressure from guerrilla groups (Hoskin and Murillo-Castaño 1999). Though the language of this measure was incendiary, there is little evidence that the results did not reflect the will of the people. Overall, 91.5 percent voted in favor of the referendum, with a total turnout of 47.9 percent. This monumental vote of over 10 million Colombians demonstrating their desire to end armed conflict in the country influenced the reopening of peace negotiations (Tickner 1998; Alape 1998). There was significant interference in the election from guerrilla groups, who imposed a boycott. Additionally, there were divisions in participation rates, with city residents participating in the referendum while rural voters were prevented from voting by guerilla groups (Murphy and Cobb 2016; Idler 2016). The hope was that the overwhelming support from urban voters would force guerrilla groups to the negotiation table. The result of this vote increased pressure on the guerrilla groups, particularly the Revolutionary Armed Forces of Columbia (FARC), resulting in eventual peace talks and a later referendum on the peace deal in 2016 (see Chapter 8 for additional discussion). Initiatives are certainly an avenue for political dealmaking, but the use of direct democracy in Colombia demonstrates that many contextual issues can influence the success or failure of these measures and mechanisms (the 1997 initiative and the 2016 referendum). Although civil society did buy into the process for the 1997 initiative and was successful in its passage, it was unable to overcome deep-seated fears and divisions within the country over the issues surrounding the civil war so as to enact change. However, it was able to build public pressure against the guerrillas. Thus, the later

Initiatives: True Citizen Power or Facades?

85

referendum, while making huge overtures to peace, made too many concessions to FARC and other rebels to be acceptable to the public. Kenya

Kenya’s democratic struggle pervades the entire political system as well as the direct democracy process. The Kenyan government has been besieged by corruption and has little credibility among its citizens. As the democracy struggles, Kenya has not been able to deal with major issues plaguing the country, from poverty to terrorism to HIV/AIDS. There have been widespread changes in government, but these have failed to have any long-term effect (Masakhalia 2011). The 2010 Constitution provides for deepening democracy in Kenya, including recalls for members of Parliament and initiatives. Referenda have been used in Kenya since 1991 to consult the public. However, these are not backed by law, but rather are a decision of the president. Though these actions have deepened democracy, Kenyans still have limited ability to directly elect many of their officials (e.g., mayors, commissioners, chiefs, and ministers). Many of these indirect practices were established during British independence in 1963. However, Kenya has gone through significant changes since then, with growth in education opportunities and changes in lifestyles, and citizens are capable of making decisions about officials as well as taking part in meaningful bottom-up initiatives. Because corruption is so rampant in Kenya, it is possible that this corruption could influence elections—even initiatives. However, the size of the democracy and the changes that continue to develop in Kenya allow for the potential for initiatives to make important changes (Masakhalia 2011). Kenyan citizens have made limited use of direct democracy mechanisms. They have used referenda twice on a national level for constitutional changes, most recently in 2010. This new 2010 Constitution provided the Kenyan public the ability to propose initiatives. Popular initiatives can either be a general suggestion15 or concrete legislation. Proposers need to collect the signatures of at least 1 million registered voters (roughly 7 percent of the population) and provide them to the Independent Electoral and Boundaries Commission (IEBC) for verification (Masakhalia 2011). Once the signatures are verified, the IEBC submits a draft bill to each of the 47 county assemblies for consideration within 90 days (Ford and Kemokai 2014). If

86

Direct Democracy: A Double-Edged Sword

the county assemblies approve the draft legislation, it is introduced into Parliament. There are two possible outcomes of this process. If the bill is passed by Parliament, it goes to the president for assent. However, if the bill does not pass, it is submitted to the people in a referendum. If the bill clears this stage, Kenyans will vote on the referendum. For it to become law, it must be passed by a simple majority of total votes cast, and a quorum of at least 25 percent of registered voters in a minimum of half of the 47 counties must participate in the referendum (Ford and Kemokai 2014). It is important to note that Kenya allows initiatives only at the national level, not the regional or local levels. Kenya’s experience with initiatives is not without controversy. The first and so far only attempt to utilize an initiative was a petition submitted in 2015. However, a myriad of factors contributed to the limitation of voter power in this instance. The Constitution of Kenya Amendment Bill (2015), hereafter referred to as the Amendment Bill, was proposed by the Coalition for Reforms and Democracy (CORD). CORD collected 1.6 million signatures to support the bill and presented these to the IEBC (Karanja, 2015).16 CORD is an organization deeply connected to the opposition party, and the Amendment Bill sought to make key changes to the electoral management process. Additionally, the amendment gave the judiciary greater autonomy, provided the perception of neutrality in electoral decisionmaking, entrenched the Ethics and Anti-Corruption Commission, and included discussion of allocation of funds and public sector jobs (Menya 2015). The purpose of the bill was to increase accountability and transparency in the election process, focusing substantially on the IEBC. The challenge, of course, to this initiative moving forward was that the IEBC was responsible for the verification of signatures. IEBC checked the signatures and found only 891,598 of the signatures to be valid (Gibendi 2016). In accordance with the law, IEBC rejected the petition, saying it did not have sufficient signatures. This supported CORD’s assertions that the IEBC colluded with the Jubilee Coalition and was in need of reform (Karanja, 2015; Otieno 2016). It was a double-edged sword for the IEBC: whether it did or did not do its job, it faced criticism. Therefore, despite any discussion of corruption, the IEBC could not win in this circumstance. This initiative provides an excellent example of how the powers of voters are limited by the system. The Amendment Bill was specifically designed to trigger a popular vote and provided a carrot of additional funding for the 47 assemblies, making it unlikely that they would

Initiatives: True Citizen Power or Facades?

87

reject the bill. However, the bulk of the Amendment Bill addressed the IEBC, the independence of the judiciary (by altering the appointment process to include members representing each party), and devolution of the national government (by increasing the funds allocated to the counties). Thus, the entire Amendment Bill was politically encumbered. CORD was supported by the opposition party, which had performed well in the 2007 general election and was active in the defeat of the 2005 referendum. By proposing the initiative at this time, the election would provide momentum into the 2017 elections and help defeat the incumbent Uhuru Kenyatta and the Jubilee Coalition. CORD argued that the 2017 elections would be tainted without a complete overhaul of the IEBC due to the irregularities of the 2013 elections (i.e., equipment failure).17 The Jubilee Coalition argued that this was a tactical strategy and CORD was trying to discredit the outcomes of the 2017 elections even before they were held. They felt this would lead to violence and continued undermining democracy in Kenya. The Jubilee Coalition focused on the apolitical nature of the IEBC and believed that the CORD changes would make it more political. The Jubilee Coalition also faced conflicts in the months leading up to the discussion of the initiative—it would lose valuable political capital if the initiative went forward and it lost. Various scandals and corruption issues had forced the president to fire several cabinet members in 2015. Polling indicated that 25 percent of Kenyans felt the government itself was capable of fighting corruption, and 67 percent indicated they felt that none of the officials suspected of corruption would be convicted. Beyond corruption, the Jubilee Coalition was also aware of public dissatisfaction with how it handled terrorism threats posed by the al-Shabaab militants. Therefore, the initiative focused on the performance issue of the Jubilee Coalition rather than the issues surrounding IEBC. In addition to the political implications of the Amendment Bill, there were financial issues that needed to be addressed. Because the IEBC received its financial allotment prior to the initiation of the Amendment Bill, there were no finances to support an initiative election. It requested and received supplemental budgetary allotments from the state treasury to support this process. However, there was controversy about this additional funding being provided to IEBC because it did not communicate the financial issues to CORD prior to the request (Otieno 2016).

88

Direct Democracy: A Double-Edged Sword

The 2016 initiative attempt in Kenya demonstrates the power of the people to request changes to their government. However, it proved to be a challenge not only to the political landscape, but to the amendment process itself. A proposal to change the very body that was responsible for certifying the signatures proved too much for Kenya’s emerging democracy. The atmosphere of corruption that plagues the Kenyan government made the previous election results circumspect and caused deep-seated distrust in the IEBC and its verification of the signatures. Regardless of whether there were issues of political corruption, the initiative never made it to the voters, and the 2017 election proved to be the real test of the people’s will. The first election on August 8th indicated that the incumbent Uhuru Kenyatta from the Jubilee Party was reelected but these were contested and annulled by the court. The parliamentary elections upheld by the court saw a majority for the Jubilee Party in both the senate and National Assembly, although they did lose several seats. CORD was not as successful electorally and continue to work outside the system to change the process and composition of the IEBC. One of the ironies of the Kenyan attempt at an initiative is that it was not really a reflection of popular will. The process was hijacked by CORD to work outside the legislature, where it was a minority, and change the rules to gain favor in the upcoming election, not unlike other times initiatives have been used (see the Italy example earlier in this chapter). The political elite manipulated the process to advance their agendas and weaken the power of the people as well as the Jubilee Coalition. The 2015−2016 test of the amendment process provides insight into the nature of Kenyan democracy, but it also shows the limitations of the institutions and how they can be manipulated. There remains potential for a deepening of democracy, but it appears that this first foray may have weakened the institutiofns of government and of direct democracy in Kenya rather than strengthening them. Analysis The use of initiatives at different levels of government demonstrates the power and limitations that these mechanisms have in an electoral system. National initiatives are fairly rare given the signature require-

Initiatives: True Citizen Power or Facades?

89

ments and lack of policy areas available to initiative processes. However, local government policymaking comes with lower stakes than at the national level. Thus, local governments are more open to initiative processes. Italy, which allows for two different forms of initiative, faces the challenge of having the system manipulated by civil society in attempts to oppose political parties. Catholic organizations’ and political parties’ reach through the initiative system puts a stranglehold on the process. New Zealand’s mechanisms are utilized primarily by amateur petitioners, and the entire process grew out of a desire to keep moral issues in the hands of the people rather than the government. Moreover, because New Zealand’s process is advisory, government policy almost always supersedes the initiative vote itself, rendering the process impotent. The German model that uses initiatives at only the state and local levels illustrates the connection to federalism that drives the process. While the states heartily use initiatives, the local use of initiatives is surprisingly robust. Colombia’s one-time usage of an initiative had the potential to lead to widespread peace in a country that had experienced civil war for decades; the 2016 referendum on a similar issue reveals that the terms of the agreement proved too much for voters. Finally, Kenya’s foray into initiatives illustrates how attempts to deepen democracy can be manipulated by opposition parties and those wanting to maintain their power base. Initiatives are not without their flaws. The process of initiatives gives additional power to the citizenry, but that opens the door to several issues such as the potential for manipulation and the differences between government and citizen priorities. Additionally, conflict and corruption can pervade this system. Government attempts to protect against these issues are seen as additional barriers to citizen involvement. There is indeed a narrow line between preventing citizen involvement and protecting citizens from themselves. Notes 1. Different countries may have further variations. For example, Italy has abrogative initiatives to review existing legislation. 2. Some have argued that this is a large number of signatures. However, the population of Italy is nearly 60 million people, with roughly 47 million registered voters.

90

Direct Democracy: A Double-Edged Sword

3. The court does not have disclose why it rejects the petition, but has the power to reject it completely. 4. However, that is very rare, as there were only 144 during the 38-year period between 1970–2008. 5. That said, not all reached the quorum to change the law. 6. The periods between 1970−1985 and 1997−2009 are classified as no periods whereas from 1987–1995 was a period of yes votes. 7. In the late 1800s, New Zealand flirted with a referendum bill that would provide for nonbinding government initiatives. However, this was passed by the legislature and fell from the agenda in the early 1900s. While not codified, referenda were still held, but not with any frequency. Creative interpretations would allow for a simple majority to pass legislation (Electoral Act 1993, section 68), but these would not be binding on the government (Morris 2014). Commonly (albeit without constitutional support) the government of New Zealand conducted a referendum on alcohol licensing in conjunction with the general elections from 1894 to 1987 (Morris 2014). There were additional referenda held in 1949, 1967, 1990, 1992, 1993, 1997, 2011, and 2015−2016. 8. At the time of adoption, the government indicated that it would hold a review of the initiative process in 5 years (roughly 1998) and, if it determined that the process was working well, there was the possibility that the process would become a binding mechanism. However, this review did not occur, and the process remains nonbinding (Parkinson 2001). Given the government’s response to the citizen-initiated referenda both at the start and throughout their usage, it seems unlikely that these will ever become binding. 9. Roughly 300,000 signatures, although more are needed to ensure they pass the verification process. 10. Members of Parliament can vote to delay the poll for up to 1 year. However, this requires a majority vote of 75 percent. 11. It seems that even when the question is written by a member of the government, in this case the House clerk, it can still have attached bias that influences the outcome. 12. Bavaria has utilized these the most. There were more than 1,260 initiatives and 578 referenda in the first 6 years after direct democracy mechanisms were instituted. 13. In the Weimar Republic, three initiatives and two national referenda were held in 1926 and 1929; in the National Socialist period, three plebiscites were held, but there were issues with the question designs (many felt they were biased) and there was evidence that the results were manipulated. 14. Hamburg requires only a simple majority of votes cast. 15. If the initiative is a general suggestion, the proposers of the initiative will be responsible for formulating it into a draft bill. 16. CORD held about 40 percent of the legislative seats at the time of this initiative. 17. There was some discussion of corrupt procurement practices where specific officials (those currently on the IEBC) were named.

5 Federalism: Power Struggles

ferences in government. These elections occur at different levels of government to varying degrees. How these mechanisms interact with types of government systems at different levels reveals the power struggle of direct democracy in a unique way. Voters have the ability to use direct democracy at varying levels of government that are sometimes in conflict with one another. Direct democracy can also be used within a country to provide strength and sovereignty to multiple levels of government as well as to test the boundaries of which level makes the most decisions. Countries select structures that serve their people and choose how to divide power between multiple levels of government. Each level provides different services to the people and requires different types of citizen involvement. Decisions on where to invest power between levels of government have both costs and rewards. When power is centralized, the national government is responsible for providing services and protecting groups. Governments that limit power in lower levels of government often do so to protect minorities and the poor. Furthermore, the national government has more resources to provide for programs than lower-level governments, so when power is invested more in the national government, it can provide more services overall. However, that comes at the expense of larger

Direct democracy elections are not impervious to structural dif-

91

92

Direct Democracy: A Double-Edged Sword

government and/or additional services.When power is decentralized, it provides opportunities for various inputs from the community and other stakeholders. Lower levels of government may also be more aware of the issues afflicting their regions and be able to address these issues with more flexibility. But some local governments are unable to provide services because of limited incomes and may flounder when required to provide multiple services beyond their means. Theory of Use by Government Systems

The type of government system affects the power structure of direct democracy usage. Unitary systems in which power is concentrated will have a different experience than confederal systems in which power is decentralized. Given the decentralized nature of a confederal system, it is more likely to utilize direct democracy because it continues to disperse power and allows for power sharing at multiple levels. More centralized systems may not be as willing to allow voters to participate in the power-sharing structure, particularly when other levels of government are prevented from having a voice. This may lead to variations in the use of different mechanisms as well. It follows that unitary systems may be less open to initiatives than confederal systems. Openness hypothesis: if a country utilizes a confederal system, then it will have more open direct democracy procedures.

Consistent with the openness hypothesis, the frequency hypothesis offers an explanation for usage. Systems that have more permissible procedures will likely use more direct democracy mechanisms because the procedures are more open and petitioners have fewer obstacles to propose measures. Systems that have smaller governmental units, as they do in confederal systems versus unitary systems, will be more open to direct democracy because the stakes are lower. Given the power structure of direct democracy, smaller governmental units provide citizens with the opportunity to have more power than they do at the national level. The use of direct democracy at these levels demonstrates that citizens really do have a voice in changing their local institutions and governance (Frey and Stutzer 2004). However, municipal- or communal- or barangay-level use of direct democracy mechanisms may not be as well known and usage does not appear to

Federalism: Power Struggles

93

be different in the size of that municipality (Ladner 2002). In fact, there also appears to be a difference rather than a size issue in how these mechanisms are used on a developmental level, meaning that the level of economic development plays a larger role than the size of a municipality (Attahi 1996). The more developed the country, the more likely it is to utilize these mechanisms at the lowest level of government—at least in some regions (Attahi 1996). Frequency hypothesis: if a country utilizes a confederal system, then it will utilize direct democracy more often.

Different levels of government have vested interests in limiting the power of direct democracy. For example, the stakes are much lower in elections at the local level than they are at the national level. Given the dichotomy of government’s power structure, it seems that the more power is vested in government, the less likely it is to provide these direct democracy mechanisms and vice versa (the less power the government has, the less it has to lose by providing direct democracy opportunities to the public). As such, direct democracy is more likely to be used at lower levels of government and can be delegated to those levels to provide democratic openings. Further, it is more accessible and feasible at lower levels because of the smaller units of government. Indeed, the ideal version of Greek democracy that fostered direct democracy mechanisms depended on a smaller-sized population to operate. Thus, it is not surprising if there is a difference in usage by smaller (and closer to the public) units of government. An argument could be made that smaller governments have less power and thus are less likely to share that power with the people, whereas larger governments have more power and therefore can allow power sharing with the people because there is more power all around. Decentralization hypothesis: if a country uses direct democracy on multiple levels, then it will be more likely to utilize it at the lower level than at the national level.

Case Studies

I use case studies from Switzerland, Canada, the Philippines, and Uzbekistan to examine the power that direct democracy mechanisms

94

Direct Democracy: A Double-Edged Sword

have at the national, state, and local levels. These cases involve different government systems and provide insight into the uses of these mechanisms at varying levels. Furthermore, these cases have different experiences with direct democracy and different evolving histories of usage, which makes them valuable to the study of multilevel use of direct democracy. The rich history of Switzerland’s direct democracy usage in its confederal system compared to Canada’s federalist approach and recent forays in direct democracy at all levels demonstrate differences in practices and entrenchment. The Philippines and Uzbekistan, both unitary systems, provide a non-Western perspective of use at multiple levels to further elaborate the analysis. The chapter concludes with discussion of the referenda to unify and secede from the European Union (EU). Switzerland

Switzerland is a democratic federal republic. It has three levels of government: confederation, canton, and communal. Power is divided between these levels of government, and each allows for widespread participation by the people through direct democracy. The source of power in the Swiss system tends to be bottom up rather than top down. Higher levels exercise power only when it is not in the purview of the lower levels of government. This makes the power structure design much more like that of a confederacy. Deep cleavages exist in Switzerland that impact its political culture and democracy. These cleavages developed along religious, language, and class lines and have had long-term effects on Swiss voting behavior, particularly at the canton level (Goldberg 2017; Kobach 1993). Thus, the bottom-up power structure of the Swiss system allows for incorporation of these groups and a voice in the political process through democratic elections and direct democracy mechanisms. Direct democracy is often associated with consensus forms of government that are necessary in systems with deep cleavages (Lutz 2012). Direct democracy is at the core of the Swiss democratic experience (Trechsel and Sciarini 1998; Treschsel and Kriesi 1996; Kobach 1993). The Swiss experience may also be the most renowned case of direct democracy. Dating back to 1291, the Swiss have used the process with unmatched frequency and fervor. Switzerland is the most prolific user of referenda, accounting for more than 50 percent

Federalism: Power Struggles

95

of all referendum votes globally (Serdült 2014). It was also the first European country to permanently adopt these types of laws. The direct democracy process in Switzerland developed from the grassroots, starting at the canton level before moving to the federal level. Much of this development was because the party system in Switzerland was so weak (Lutz 2012; Treschsel and Kriesi 1996). The first canton to introduce veto rights was St. Gallen in 1831 as a precursor to the optional referendum, and other cantons quickly followed; it was adopted into the modern Swiss constitution in 1848 (Gruner 1969; Möckli 1996; Lutz 2012). The modern history of direct democracy demonstrates that its use is more prevalent since the nineteenth century. The history of these Swiss experiences shapes its current iteration. The power structure in Switzerland is open to the possibility of voter participation. The decentralized governmental process allows for power sharing with cantons and communal governments; thus, its openness to all forms of direct democracy mechanisms demonstrates support for citizen power sharing in government. National level. At the national level, Switzerland is a prolific user

of direct democracy. Swiss voters have been asked to vote on direct democracy measures two to four times every year since 1968 (Lutz 2012). A fixed calendar with four dates is used to identify each year for these elections. There are several variations of the direct democracy process at the national level in Switzerland. The first mechanism utilized is a legislative referendum, which is a referendum driven by the legislature. Adopted in 1848, constitutional modifications require a mandatory referendum and not only a majority of the popular vote, but also a majority of cantons—each full canton gets one vote, half cantons get a half vote (Serdült 2014; Kobach 1993, 1994).1 Additionally, since 1977, all international treaties to join a collective security organization or supranational community must be ratified by referendum with a majority of the country as well as a majority of the cantons (Eschet-Schwarz 1989; Treschsel and Kriesi 1996). The second type of mechanisms is optional referenda, which were adopted in 1874. Optional referenda are citizen driven, whereby legislation is challenged. Once a new law is published, citizens must collect 50,000 signatures within 90 days to challenge the law (Eschet-Schwarz 1989). If that threshold is met, a vote will be held and, if a simple

96

Direct Democracy: A Double-Edged Sword

majority of the popular vote rejects the new law, it is removed. These types of legislative referenda do not require a cantonal majority (Eschet-Schwarz 1989). It is not unusual to have 10 or more optional referenda per year; more than 550 national referenda have been held in Switzerland2 (Serdült 2014; Papadopoulos 2001; Kobach 1993, 1994). Although the use of mandatory and optional referenda has decreased in recent years (Lutz 2012), the use of a third type of direct democracy in Switzerland has increased. The third mechanism, federal popular initiatives, was adopted in 1891 to allow requests for constitutional amendments. Typically, citizens submit a full draft of the text they want amended. For a popular initiative, 100,000 signatures must be collected within 18 months and then a vote on the measure is called3 (Trechsel 2007; Lutz 2012). Typically, 2 to 3 years between the signature collection and the vote are used to cool popular moods. Parliament can propose an alternative amendment, leading to two competing amendments on the ballot (Eschet-Schwarz 1989). An initiative or an alternative amendment proposed by parliament must receive a majority of votes as well as a majority of canton support (Eschet-Schwarz 1989). Popular initiatives on constitutional amendments are designed to bring new ideas into the foray, whereas optional referenda prevent the enforcement of laws approved by the legislature. Because Switzerland has such serious cleavages in its population, the citizen initiative allows for voters to propose changes to government, and the double majority required to pass referenda prevents total manipulation of the system by specific fragments. Less than 50 percent of initiative petitions come to a vote, and roughly 5 percent are approved (Lutz 2012). Initiatives have been more popular post-1970, but are still far less utilized than referenda. Referenda provide the opportunity for the preferences of citizens and cantons to be included in federal policy. In doing so, these mechanisms provide consolidation as a nation and legitimacy for the federal system (Eschet-Schwarz 1989; Treschsel and Kriesi 1996; Kobach 1993, 1994). The outcome of direct democracy has often created more unity on political, economic, and social levels between the cantons and the federal government. That said, it is not uncommon for cantonal parties to take positions that differ from those of the federal government on referenda, which places the cantonal interests at odds with the national government (Eschet-Schwarz 1989). The frequency of the use of direct democracy in Switzerland reveals several nuances in the power division within the country.

Federalism: Power Struggles

97

First, the balance of power between political elites and citizens is a critical component of the optional referendum. Optional referenda allow citizens to essentially stop laws from being enacted. Second, because direct democracy mechanisms are used on multiple levels, the usage at each level is critical to our understanding of the mechanism. There is frequent usage at the national level; however, it is important to investigate the other levels as well. Canton and communal levels. Cantons play an important role in

Swiss democracy and are central to its political culture. Most political issues are resolved within the cantons and even the federal assembly elections are organized along cantonal lines (Eschet-Schwarz 1989). When campaigning for the federal assembly, much of the discussion focuses on cantonal issues rather than national issues. Referenda are also held at the cantonal and community levels. Each of the 26 cantonal constitutions detail the requirements for direct democracy mechanisms, which were introduced after the foundation of the Swiss federation in 1848 (Serdült 2014; Gruner 1969). Although the requirements vary broadly between cantons, many Swiss cantons ended up with relatively the same set of direct democracy mechanisms without their imposition by a strong central force (Serdült 2014). This provides strong evidence of policy diffusion across the cantons. Much of the use of these mechanisms at the canton level has been tracked to the communal traditions and the ideas of the French Revolution (Serdült 2014; Hangartner and Kley 2000; Roca and Auer 2011). The strength of the direct democracy process at the canton level is a response to the structure of Swiss government. The cantons provided the foundation of modern Swiss political institutions (Serdült 2014; Roca and Auer 2011). Citizens’ initiatives provide a grassroots option to solve problems in the Swiss system. The structure of citizens’ initiatives allows no opportunity for national leaders to influence these initiatives. The primary structure allows for the cantons and the people to have the largest voice in the success or failure of an initiative. Certainly, the federal government can determine the will of the people by the topics and successes of these initiatives, and they give it insight into what is happening in different cantons in the country. Although an initiative may not pass, it does provide awareness into ways to deal with problems in the future (Eschet-Schwarz 1989). There was some debate in the late-1880s about the use of direct democracy mechanisms at the national level to promote political

98

Direct Democracy: A Double-Edged Sword

unification because of the implications they would have on the independence of lower levels of government and the creation of a more centralized government. Federalism in Switzerland has been strengthened due to the use of direct democracy because it allows for autonomy at the canton level (Eschet-Schwarz 1989). The use of direct democracy at the canton level provides an illustration of the support of the national government for the sovereignty of these smaller units. Direct democracy is used at the canton level to give voters more influence on government policymaking. Despite that openness, there are limitations on how referenda can be used. The primary restriction is on budget issues, although some cantons have added restrictions (Bâişanu 2010). Additional notable variations when it comes to referenda are that constitutional referenda are obligatory in the cantons and legislative referenda are optional in Roman Switzerland but obligatory in German Switzerland (Bâişanu 2010). Cantons vary in how they support constitutional amendments. A core group of cantons is more supportive of governmental proposals, but others are less supportive. Often this has to do with the composition of a canton itself; those cantons with a large minority population and culture are less supportive of government proposals. These minority factors include religion (most cantons are Protestant), language (i.e., Swiss German), and industry (service or agriculture). Additional issues such as political culture (being traditional or a modernizer) are also factors in the support for amendments. Traditional cantons tend to be less supportive of amendments than modernizers (Eschet-Schwarz 1989). Swiss use of referenda has increased post−World War II. Roughly 75 percent of these constitutional referendum proposals passed and were included in the constitution (Eschet-Schwarz 1989), whereas optional legislative referenda received significantly less support (roughly 45 percent) (Eschet-Schwarz 1989). Interestingly, the behavior of cantons on initiatives often depends on who is initiating the measure. Left-leaning groups, which propose roughly 40 percent of the initiatives, usually focus on centralization of power, which tends be unsupported among the cantons (Eschet-Schwarz 1989). However, the vast majority of initiatives without a counterproposal fail (90 percent) regardless of who proposes them (Eschet-Schwarz 1989). As a whole, at the canton and city levels, 81 percent of direct democracy measures are passed by the public (Buetzer 2011). There is also a large difference in the number of referenda by canton; for

Federalism: Power Struggles

99

example, Zurich has held nearly 500 referenda since 1970 while Ticinio has had less than 60 in the same time period. There are also similarities between the cantons, as all cantons allow for legislative referenda (though they differ on whether they are mandatory or optional). There are differences in how citizens can demand a referendum, and many cantons have changed these requirements over time (Barankay, Sciarini, and Trechsel 2003).4 However, signatures from about 1 percent of the electorate are generally required for a citizen-initiated referendum. Differences in time constraints also exist—some cantons have no time restrictions while others have expanded or introduced time constraints (Barankay, Sciarini, and Trechsel 2003). Cantons also provide for administrative initiatives that can demand that specific work be done by the administration (e.g., building or repairing infrastructure), but these are rarely held on fiscal issues. At the communal levels, legislative referenda and popular initiatives are common. Given the proximity between the communal government and voters, fewer limitations on topics exist and even budgetary issues are open to the direct democracy process. Swiss municipalities have the highest number of direct democracy decisions of any governmental level around the world (Marxer and Pállinger 2011). Between 1990 and 2000, the 111 cities that utilize direct democracy voted on 3,442 different measures, compared to 1,142 at the canton level and 116 at the state level during the same time (Buetzer 2011). Even at the local level, Swiss elites have focused on consensus leadership and involved voters on a number of issues. Variation exists in how much decisionmaking power lies at the commune level; much of the direct democracy mechanisms depend on canton laws, but a few have more far-reaching laws to extend their mechanisms (Buetzer 2011). The Swiss experience provides support for all three hypotheses. First, the Swiss system is quite open to direct democracy measures. All three common types of mechanisms are available at each level of government. Second, the confederal system utilized in Switzerland encourages citizen involvement and helps to overcome the cleavages in the population and build consensus. Finally, there is support for the decentralization hypothesis in just the sheer number of direct democracy mechanisms used at the local level vis-à-vis the national level. The Swiss are the most prolific users of direct democracy, but there is certainly more usage at lower levels than at the national level.

100

Direct Democracy: A Double-Edged Sword

Canada

Canada has a federalist system with strong regional fragmentation. The Canadian government is divided into three levels: federal, provincial or territorial, and municipal. Direct democracy is used to a limited extent in Canada. At the national level, no law governs the use of referenda, but there have been a number attempts, albeit unsuccessful, to create this legislation: Bill C-40 and Bill C-9 in 1978, Bill C-311 in 1988, and Bill C-287 in 1991 (Boyer 1982; Côté 1992; Marquis 1993). In June 1992, the federal government passed An Act to Provide for Referendums on the Constitution of Canada (Bill C-81) that allows the federal government to hold referenda and provides the administrative and legal setup should the government hold a referendum (Côté 1992). National referenda. National referenda in Canada are relatively

rare, but they typically are significant. Although only three national referenda have been held in Canada since 1898, it conducted North America’s first national referendum and many provinces adopted direct democracy mechanisms in the early 1900s (Donovan 2014). Canada dealt with prohibition in its first national referendum in 1898. However, the vote was split between the provinces, with all but one province (Quebec) voting for prohibition, but the overall vote was 51 percent in favor of adopting prohibition. Furthermore, the turnout was low with only 543,073 votes cast and a total population of 5.17 million, meaning participation was around 10 percent (Boyer 1982). As a result of the split vote and low turnout, the Canadian government decided not to adopt the measure. The second national referendum was on the issue of conscription in 1944. Voters were asked: “Are you in favor of releasing the Government from any obligations arising out of any past commitments restricting the methods of raising men for military service?” With Quebec being the only province to vote against the measure again, 64.5 percent of voters voted in support of the measure. Turnout was higher with 4.5 million voters participating (nearly 40 percent of the total population). The measure was adopted by the government of Canada and allowed conscription for World War II, although those conscripted were not immediately sent overseas (Boyer 1982). The third referendum has had perhaps the most lasting political impact. In 1992, Canadians were asked to vote on the Charlottetown Accord developed in wake of the Quebec separatist movement and the failure of the Meech

Federalism: Power Struggles

101

Lake Accord in Manitoba and Newfoundland (Winn 1992; Côté 1992; Johnston 1993). Voters were asked: “Do you agree that the Constitution of Canada should be renewed on the basis of the agreement reached on August 28, 1992?” A number of issues surrounded the Charlottetown Accord, including aboriginal self-government, senate reform, bringing Quebec into the constitutional framework, and acknowledging Quebec’s distinctiveness within Canada.5 Although initially popular among the premiers, some hesitancy still existed in Quebec. During the campaign, many strong opposition forces were in effect—former prime minister Pierre Trudeau, Preston Manning and the Reform Party, Lucien Bouchard and the Bloc Quebecois, and Jacques Parizeau and the Parti Quebecois (Côté 1992; Johnston 1993). The results of the referendum were split, with Alberta, British Columbia, Manitoba, Nova Scotia, Quebec, Saskatchewan, and Yukon voting against the referendum (7−12) and a total of 55 percent of voters nationwide voting against the agreement. In total, 13.6 million voters participated in the election. The federalist nature of the Canadian confederation is such that it shares power between the provinces and the national government. The use of national direct democracy three times demonstrates less use than in some countries, yet the desire to utilize direct democracy to make big decisions about the country’s future demonstrates the value that Canadians place on the referendum process. Provincial and municipal referenda. Provincial referenda have

roots in twentieth-century populism. However, many of the initial direct democracy provisions were abandoned until they were formally introduced in western provinces in the mid-1910s (Boyer 1988, 82). Although these were initiated in the 1910s, many of the reasons for utilizing direct democracy were abandoned when Canada moved toward a multiparty system and allowed for different voices within the political process (Donovan 2014). Provincial referenda remain a tool used quite sparingly. A resurgence of direct democracy ideals happened in the 1970s—albeit the Canadian version of direct democracy is far more limited than that in the United States, its southern neighbor. One example of this is that there are “few rules mandating that Canadian legislatures (provincial or federal) refer questions to voters” (Donovan 2014: 131). There are some exceptions: Alberta and British Columbia require a provincial referendum when there is a national constitutional amendment, and

102

Direct Democracy: A Double-Edged Sword

Manitoba requires referenda on provincial tax rate increases (Mowery and Pelletier 2001; Donovan 2014). Generally, provincial referenda are not binding, but these rules are set by provincial governments and can depend on the referendum. Additionally, thresholds (vote and turnout) are left up to provincial governments and can depend on the topic of the legislation. There have been general topics of consistency across the provincial referenda (see Table 5.1): prohibition/alcohol sales (35), federalism (10), electoral rules (9), and daylight saving time (5) (Donovan 2014). Some other topics that provinces have addressed are health care, education, lottery, abortion, and budgets. All referenda on the provincial level are consultative rather than binding (Royal Commission on Electoral Reform 1991). The vast majority of these are held in western Canada, which is more open to the populist ideas that drive direct democracy (Marquis 1993). Provincial use of direct democracy is higher than national use. Almost all provinces have developed direct democracy mechanisms, and many use them frequently. The important difference is that national referenda are binding and provincial referenda are advisory. Thus, although the people have more frequent voices in provincial mechanisms, they do not necessarily give more power to the voters. Initiatives exist at the municipal level and were available at the provincial level in western Canada for a short time in the early 1900s.6 These are the weakest units of the political system and have

Table 5.1 Provincial Referenda Alberta British Columbia Manitoba New Brunswick Newfoundland Northwest Territories Nova Scotia Nunavut Ontario Prince Edward Island Quebec Saskatchewan Total

7 13 7 4 5 2 4 2 5 7 5 8 72

Source: Donovan 2014. Note: Yukon and Nunavut did not have any referenda at the time of this writing.

Federalism: Power Struggles

103

limited authority to make laws; as such, there was initially some resistance to the use of direct democracy at the municipal level because it would turn the limited power over to the electors rather than the local governments. However, prohibition and the Canada Temperance Act of 1878 changed the minds of lawmakers to allow voters to have a larger voice in lawmaking. The use of initiatives at the local level is mandated by local councils, but there are generally provisions in the provincial government to overrule any law that extends beyond the powers of a municipality. Their use is so prevalent at the municipal level that there are often close to 100 referenda each year between cities, districts, towns, and villages. Because these referenda are used so often, they are considered part of the local government lawmaking process by many provinces (Boyer 1992). The Canadian case provides a perplexing view of direct democracy. It is used more frequently at the provincial and local level, but these initiatives are rarely as powerful as they are at the national level. Additionally, the quality of a question asked at the national level has substantially more importance than at the provincial and local levels (with the exception of the Quebec separatist movement examined in Chapter 7). Some support for the decentralization hypothesis exists because direct democracy is used more frequently at lower levels. There is also support for the openness and frequency hypotheses, as there is less use of direct democracy in Canada’s federalist system than in the Swiss confederal system. The Philippines

Direct democracy has been utilized frequently in the Philippines to various ends.7 But before examining the use of direct democracy, a brief overview of the government structure and system is critical. The Philippines government has a unitary structure, in which most of the power resides in the president. There are four secondary levels of government (referred to as local government units): autonomous regions or provinces and independent cities, municipalities, and barangays. Although sometimes achieved by referendum, the president has the authority to change local government units. The Philippines provides for referenda and initiatives at the national, regional, and local levels. Lower levels of government are more open to direct democracy mechanisms. Referenda have been

104

Direct Democracy: A Double-Edged Sword

held nationally since 1935. The Philippine Constitution of 1946 enables amendments when approved by a majority of votes within 3 months of approval (Kobori 2014).8 The Philippine constitution was updated in 1987 to include the people’s initiative. Article XVII Section 2 of the Philippine constitution states: Amendments to this Constitution may likewise be directly proposed by the people through initiative upon a petition of at least twelve per centum of the total number of registered voters, of which every legislative district must be represented by at least three per centum of the registered voters therein. No amendment under this section shall be authorized within five years following the ratification of this Constitution nor oftener than once every five years thereafter.

This provision was further supported by the Initiative and Referendum Act (Republic Act 6735), which defines an initiative as: 1. A petition to propose amendments to the constitution. 2. A petition to propose enactment of national legislation. 3. A petition to propose enactment of a local resolution or ordinance at the regional, provincial, city, municipal, or barangay levels.

At the local level, the law also provides for an indirect initiative, where a people’s initiative is advisory and is sent to the local or national legislature. The Philippine constitution codifies three forms of direct democracy: direct initiatives (giving voters the power to propose constitutional amendments or legislation through elections), agenda initiatives (allowing voters to send a proposition to legislative bodies for action), and optional referenda (allowing voters to approve or reject legislation proposed by the legislature) (see Section 3 of the Initiative and Referendum Act in the Philippine Constitution of 1987).

National level. At the national level, all referenda are binding. Constitutional amendments and issues dealing with foreign militaries are subject to mandatory referendum, but constitutional amendments and other issues of national importance can be subjected to an optional referendum. When it comes to limitations on the topics of optional referenda, they cannot include more than one subject or involve statutes on emergency measures. The legislature can drive referenda, and there is no specific quorum of turnout or voting threshold (only a simple majority) required by law for a ref-

Federalism: Power Struggles

105

erendum to pass. Furthermore, voters can drive citizens’ initiatives and agenda initiatives on constitutional amendments and legislative proposals. To propose a citizens’ initiative, voters are required to collect signatures equivalent to 10 percent of the total number of registered voters and must include voters from each legislative district (3 percent of each district’s voters). For an agenda initiative, the threshold of signatures is higher, requiring 12 percent of registered voters (including 3 percent of registered voters in each district). The petitions for collecting signatures for both citizens’ and agenda initiatives require the full proposed text and general explanation of what the initiative is attempting to do. When submitted to the Independent Electoral Commission, the petition must include the text of the law, the proposition and its reasoning, signatures of voters, and an abstract or summary of 100 words. After this is submitted, the commission checks the signatures and ensures that the process was followed. Once qualified, a vote is scheduled between 45 and 90 days later. The Philippines has experienced three periods of national referenda. The first was during the Commonwealth period (1935–1946). The second was during the martial law period (1965–1986). The third established the post-Marcos era (1998–present). During this period, there have been significant limitations on direct democracy, particularly initiatives after the 1987 constitution. Initiatives relating to the constitution can be called only once every 5 years. The first referendum in the Philippines was held in 1947 to approve rights for American companies to operate within the new country (Smith 1986). President Ferdinand Marcos used referenda frequently during his terms. The first of these referenda, held in 1972, was a national referendum where people voted in barangay meetings. Any citizen aged 15 years and older was able to participate. Three issues were presented to the public and 15 million people participated. The first issue focused on the new constitution; 97 percent of the barangays approved. The second focused on the need for a constitutional referendum; 93 percent voted against it. 89 percent voted against the third issue of having national elections later that year. This buoyed Marcos’s agenda, so he continued to encourage voters to participate in barangay assemblies. An additional referendum was held on July 27, 1973, on Marcos’s continuation as president. Voters were given ballots where they could

106

Direct Democracy: A Double-Edged Sword

provide comments about government as well as an answer to the referendum. A total of 90.7 percent of the people voted to extend Marcos’s term (Smith 1986). In February 1975, voters were asked three questions, which were conducted as traditional elections in polling booths rather than the barangay meetings. The first question focused on the restructuring of local government, which 69 percent approved. The second question asked about the way that Marcos was administering martial law, which 87.6 percent approved. The third question asked for approval of continuing martial law, which received 86.7 percent support. The 1976 referendum asked voters again about martial law and Marcos’s development of a new legislature. Voters overwhelming supported the continuation of martial law with 97.9 percent of the vote. Also, 90.6 percent voted to allow Marcos to establish a new legislature with some members appointed by the president. Government sources also recorded a turnout of 97 percent (Smith 1986). Rather than hold traditional elections, Marcos held additional referenda in December 1977 to continue as the president and prime minister (Smith 1986). Additionally, a referendum was held on the autonomy of Southern Mindanao, in which 97.9 percent rejected the proposal of autonomy (He 2002). In 1981, voters participated in two more referenda. The first focused on a new structure for the executive branch. The second focused on new procedures for sanctioning a political party (Smith 1986). Serious issues arose with preventing opposition groups from participating in the campaign process; the government dominated the media and information surrounding the referendum. This continued to decay the facade of democracy that was promoted through these referenda. The opposition to Marcos had become more organized and the lower level of support for these referenda demonstrated that a break was forming in the powerhold that the Marcos regime had on the people.9 Marcos’s use of referenda boils down to his own desire for legitimacy—not to give the people a voice, but rather as a political strategy (Smith 1986). He frequently prevented opposition groups from having a voice in the referendum campaigns. The high levels of support could be indicative of real support for his regime; however, the police and other officials were active in the get-outthe-vote process (and vote choice), and a presidential decree (No. 229) was even issued to make voting without a cause a criminal offense with the possibility of 6 months in jail (Smith 1986). Often

Federalism: Power Struggles

107

charges of tampering with the counting of ballots were made and the early use of open meetings for votes shows that coercion was rampant throughout the referenda. Marcos’s decision to utilize referenda to continue his leadership outside of the usual channels of elections demonstrates his lack of support for basic electoral structures. Additionally, he used referenda to continue and to gain support for his martial law actions and constitutional restructuring (Smith 1986). These referenda also enabled Marcos to attempt some international legitimacy and garner support for his actions from international actors such as the United States. The gradual decline of participation and support for Marcos’s referenda demonstrates that his legitimacy continued to erode. Despite the use of democratic procedures, it is clear that Marcos did not support democracy and freedom of choice or will. The remaining referenda in the Philippines have focused on the creation of autonomous regions (see Table 5.2). These autonomous region referenda have demonstrated the clear division between Muslims and Christians in the Philippines (He 2002).

Table 5.2 National Referenda in the Philippines Name

Philippine Constitutional Plebiscite Philippine Women’s Suffrage Plebiscite Philippine Constitutional Plebiscite Philippine Constitutional Plebiscite

Topic/Summary

Date

Granting women the right to vote.

1937

Ratify the constitution.

Export tariffs.

Establishment of a bicameral legislature. Tenure and term limits for the president and vice president. Creation of a commission on elections. Philippine Constitutional Enlargement of House of Plebiscite Representatives. Allow members of Congress to hold multiple positions. Philippine Constitutional Ratification of the constitution. Plebiscite

1935

1939

1940

1967 Jan. 1973

Vote

For 96.4% Against 3.6% For 90.9% Against 9.1% For 96.6% Against 3.4% For 79.1% Against 20.9% For 81.7% Against 18.3% For 78.0% Against 22.0% For 18.0% Against 82.0% For 16.5% Against 83.5% For 95.3% Against 4.7%

Turnout 65.0%

86.0%

50.0%

81.6% 87.0% continues

108 Table 5.2 Continued Name

Topic/Summary

Date

Vote

Philippine Executive and Legislative Powers Referendum

President has power to restructure local government and appoint local officials; allows martial law to continue. Martial law should be continued; ratify amendment to the constitution substituting the Interim National Assembly for the Regular National Assembly. President Ferdinand Marcos should continue as president and prime minister. Provide more powers to the president and create a new executive committee on succession and accreditation of political parties. Hold barangay elections pursuant to Proclamation No. 2088. Approval of constitutional amendments (creating office of vice president; abolishing the executive committee; establishing elections for regular members of National Assembly). Abolishing the office of the prime minister and the Regular National Assembly. Creation of autonomous region including six provinces.

1975

For 87.8% Against 12.3%

1976

For 86.7% Against 13.3%

1977

For 89.5% Against 10.5%

Martial Law Referendum

Philippine Constitutional Plebiscite Philippine Constitutional Referendum

Philippine Constitutional Referendum Referendum

Philippine Constitutional Plebiscite

Philippine Constitutional Plebiscite

Cordillera Autonomous Region Creation Plebiscite Autonomous Region in Muslim Mindanao Plebiscite Shariff Kabunsuan Creation Plebiscite Dinagat Islands Creation Plebiscite Quezon del Sur Creation Plebiscite Davao Occidental Creation Plebiscite

Continuation of martial law.

Expansion of the Autonomous Region in Muslim Mindanao (ARMM). Inclusion of specific regions in the ARMM. Creation of province of Shariff Kabunsuan. Creation of province of Dinagat Islands. Creation of province of Quezon del Sur. Creation of Davao Occidental province.

Source: Compiled by the author from Quezon n.d.. Note: a. Turnout a function of the individual region.

July 1973

April 1981

For 90.8% Against 9.2%

For 68.0% Against 32.0%

Turnout

Held in barangays, vote by show of hands. Held in barangays rather than traditional vote.

78.0%

June 1981 1984

For 69.0% 80.9% Against 31.0% Majority yes (exact figures unavailable).

1987

For 76.4% Against 23.6%

1998 2001

2006

2006

2008

2013

For 30.9% Against 69.1%

For 81.2% Against 18.8%

For 14.7% Against 85.3% For 97.0% Against 3.0% For 52.6% Against 47.4% For 43.4% Against 56.6% For 76.6% Against 23.4%

83.2% a

a

a a

37.9% a

Federalism: Power Struggles

109

Although the use of referenda at the national level in the Philippines is higher than in Canada and far exceeds the expectations based on theory, direct democracy was used at the national level as a way to create a mandate for the president and exacerbate the manipulation of voters. However, it also demonstrated some democratic principles and an opportunity for voters to have a voice (albeit minimally). Provincial and local levels. The desire to utilize lower levels of

government in the direct democracy process is evident from the first attempts at public consultation by Marcos in 1972. Local-level referenda or consultation encourage bottom-up democracy in many localities. To start an initiative at the local government level, petitions can be filed with the Regional Assembly or local legislative body. If no action is taken on a petition within 30 days, proponents can invoke an initiative and move toward a public vote. Then signatures must be collected. Proponents have time limits to collect the appropriate number of signatures: 120 days in autonomous regions,10 90 days in provinces and cities, 60 days in municipalities, and 30 days in barangays.11 The number of signatures varies by the local government unit: 2,000 registered voters in autonomous regions, 1,000 in provinces and cities, 100 in municipalities, and 50 in barangays. The signatures must be witnessed by a representative of the Commission on Elections as well as a petitioner. These are then verified in the appropriate government unit to ensure the correct number has been collected. Upon verification, a date is set for an initiative. For a proposition to be approved, it must receive the majority of votes cast; a proposition then goes to the local government body for enactment within 15 days. It is possible for a local government to enact a proposition anytime within the process, thereby canceling the initiative. Local initiatives can be exercised only once a year and are restricted to issues within the power of the local government body to enact. A local government cannot repeal, change, or amend an initiative within 6 months of passage, and an initiative can be repealed, changed, or amended only within 3 years with a vote of three-quarters of all members of that government unit (barangays are subject to a 1-year limitation). Any local government unit may introduce a referendum. It must be held by the Commission on Elections within 60 days in the provinces and cities, 45 days in municipalities, and 30 days in barangays. All initiatives and referenda are subject to court

110

Direct Democracy: A Double-Edged Sword

review and there is no limitation within the law on the court’s ability to void any proposition. Many referenda at the provincial level deal with the creation of new provinces or reclassifying municipalities as cities. The most recent referendum was in 2013 with the creation of the new Davao Occidental province. The creation of Davao Occidental province was initially proposed by Marc Douglas Cagas IV, a Davao del Sur congressman. Cagas later withdrew his proposal in light of the looming 2013 election. The proposal was reinvigorated by Franklin Bautista, a congressman for the second district of Davao del Sur. The idea was to create a new province out of the southern part of Davao del Sur. Once proposed, the House of Representatives and the senate passed Republic Act 10360 to create the new province and this was set for a plebiscite for ratification. Registered voters of Davao del Sur were able to vote on this split into two provinces; 76 percent voted in favor of its creation. Recall elections are permitted under the Republic Act 7160 in the Philippines. To initiate a recall, a petition must be signed by a minimum of 25 percent of registered voters in a local government unit. A recall can be called only once during an official’s term, and after 1 year from when he or she took office but not within 1 year of the next election. At the local level, elected officials can be recalled. These elections are triggered when all elected officials within a local government unit endorse a recall or when 25 percent of registered voters within that the same local government unit sign a petition. The recall elections in the Philippines call for recalled officials to run for their position again against opposition, rather than a simple removal process (Kobori 2014). It is important to note that recalled officials cannot resign when facing a recall election, but may participate in the election. The candidate with the most votes is elected to the position (recalled candidate or not). At the local level, initiatives and referenda can be held only once a year. For an initiative at the local level, minimum signature requirements differ depending on the local government unit. In cities and provinces, petitioners are required to collect signatures of 1,000 registered voters; in autonomous regions, they are required to collect the signatures of 2,000 registered voters. Recall elections have been somewhat limited to the local and barangay level. The last recall election that occurred above the barangay level was the 2015 Puerto Princesa mayoral recall. On March 17, 2014, Al Babo and Joey Mirasol filed a recall petition with the

Federalism: Power Struggles

111

Commission on Elections citing “‘breach of trust,’ declining tourism activities and a deteriorating peace and order situation” as reasons for recalling current Puerto Princesa mayor Lucilo Bayron (Anda 2014:1). Bayron argued that this was a political move by the previous mayor, Edward Hagedorn, contending that Hagedorn just did not accept the outcome of the election. A petition including over 40,000 signatures (exceeding the approximately 19,000 signatures needed) was submitted. The commission approved the recall petition in April 2014 despite not having the funds to verify the signatures submitted. Financial woes plagued Puerto Princesa, which influenced not only this recall but other petitions that were delayed. Petitioners asked the supreme court to mandate the certification of the signatures. The court, in a unanimous opinion, ordered the commission to move forward with the election in November 2014, and indicated that there was an existing appropriation in the budget for funding recall elections. Despite the court’s involvement, Bayron sought to stop the proceeding by arguing election irregularities because of the time gap between collection and verification of the signatures, but these arguments were dismissed by the court. Another issue that plagued the election process was a suspension of voter registration between April and May 2015 because the Voters’ Registration Act prohibits voter registration prior to a recall election, so voters could not register during the campaign. Initially, 45 candidates registered to run, but this was reduced to 29 by election day, with Bayron and Hagedorn as the front-runners. The election started on 7:00 A.M. and ended at 3:00 P.M., and the ballots were manually counted at the precinct level before being transported to the provincial capital where counting began around 8 P.M. Although there were discussions of tampered ballot boxes and accusations of irregularities from the Hagedorn campaign, there was no evidence to support these allegations. The next day, after hours of counting and delays due to court filings by Hagedorn, the commission on elections proclaimed Bayron as the winner with 53 percent of the vote. Allegations of vote buying were made by the Hagedorn campaign; however, election watchdogs (the Legal Network for Truthful Elections) were generally happy with the procedures and execution of the election. Some of the discrepancies noted at the ballot box were explained by the changes in the list of voters (deactivation after not voting for two consecutive elections) rather than voter fraud. Amusingly, this was not the first mayoral recall in Puerto Princesa. Hagedorn, Bayron’s challenger, was also involved in the recall election of Victorino Dennis

112

Direct Democracy: A Double-Edged Sword

Socrates. Socrates had replaced Hagedorn in 2001, when Hagedorn had been term limited and unable to run again. Hagedorn participated in the recall election of Socrates and won, setting his new term limit to 2013, which coincided with Bayron’s initial election.12 Thus, although the democratic nature of a recall election is to keep leaders accountable, it became a tool of manipulation by specific interests to maintain their power in the mayor’s office. The Philippine use of direct democracy is a fantastic opportunity to test the hypotheses proposed earlier. Its unitary system ensures that power is centralized and its use of direct democracy to create a mandate for the government’s regime demonstrates the government’s support for the desire to maintain a centralized power, although it does use direct democracy more often than expected. The system is focused on government control, with limited citizen ability to enact change through the system, illustrating support for the openness hypothesis. The frequency hypothesis, although supported in numbers, is not supported in substance. Finally, for the decentralization hypoth-esis, direct democracy is used more frequently at the provincial, local, and barangay levels than at the national level (there are also more types of direct democracy available to these levels). Uzbekistan

The Central Asian country of Uzbekistan provides an opportunity to explore direct democracy in the former Soviet Union. Its constitution provides for referenda at the national level and regional levels. However, regional referenda are limited to the Republic of Karakalpakstan. There is no law permitting referenda at the local level (Bektemirov and Rahimov 2001). Its constitution does not provide for other mechanisms of direct democracy. Despite the use of direct democracy, many authoritarian regimes in Central Asia have entrenched their power through these mechanisms (Hill and White 2014).

National level. At the national level, an Independent Electoral Commission is responsible for administering direct democracy mechanisms. They can be used on any issue of national importance when it comes to national referenda and initiatives, with the exception of civil service, taxes, amnesty and pardons, and issues of territorial borders. These can be initiated by voters, the president, and the legislature. Although anyone can propose a referendum question,

Federalism: Power Struggles

113

only the president and the legislature can finalize the ballot text. When it comes to voting, a turnout quorum of 50 percent of the electorate is required for optional referenda, whereas for a mandatory referendum a quorum is not specified. Citizens must register a committee of 100 citizens to propose an initiative. Once an application is submitted and checked by the commission, a vote will occur between 45 and 90 days. Four national referenda have been held in Uzbekistan. The first was the Soviet Union referendum in March 1991. This referendum asked voters: “Do you consider necessary the preservation of the Union of Soviet Socialist Republics as a renewed federation of equal sovereign republics in which the rights and freedom of an individual of any nationality will be fully guaranteed?” In Uzbekistan, 94.7 percent of voters voted to support the measure—higher than the full Soviet Union vote of 77.9 percent in favor (Treacher 1996; Polat 2002).13 This was followed by the Uzbekistani independence referendum in December 1991. In this election, voters were asked if Uzbekistan “should remain part of a renewed Union (federation) as a sovereign republic with equal rights?” The referendum had a high turnout of 94.1 percent, and 98.3 percent of voters voted in favor of it. As a result, Uzbekistan became an independent republic within the larger Soviet federation (Polat 2002; Treacher 1996).14 In 1995, Uzbekistan held a referendum on extending presidential terms; it would extend the presidential term of Islam Karimov until 2000, when the next presidential election would occur (Hill and White 2014). This referendum was approved by 99.6 percent of voters, with an extraordinarily high turnout of 99.3 percent (Polat 2002; Abdurasulov 2007; Abazov 2008). In 2002, voters were back at the polls for two referenda (Hill and White 2014). The first referendum was on extending the presidential term from 5 to 7 years; it was approved by 91.8 percent of voters. The second referendum was about introducing a bicameral parliament (Khegai 2004; Fedorov 2012); it was approved by 94.7 percent of voters. Voter turnout was 92 percent.

Regional referendum. The only mandatory referendum available at the regional level is for the purpose of secession. And the only region that has the ability to hold referenda is the Republic of Karakalpakstan (Ilkhamov 2002). The Republic of Karakalpakstan is also the only autonomous region of Uzbekistan, and the mechanism exists in case it should eventually decide to leave Uzbekistan. The mechanism for

114

Direct Democracy: A Double-Edged Sword

this referendum is outlined in the Uzbek constitution as: “Article 74. Chapter XVII. The Republic of Karakalpakstan shall have the right to secede from the Republic of Uzbekistan on the basis of a nationwide referendum held by the people of Karakalpakstan.” The specifics of the referendum would be governed by Karakalpakstan at the time of the referendum and would outline voting and campaign procedures. Uzbekistan’s limited foray into democracy since its independence has flavored its use of direct democracy. The limited use at the subnational level means that voters’ only opportunity to exert power over policy through these mechanisms must come at the national level. As a unitary system, Uzbekistan’s government is far more closed to direct democracy measures than some of the other cases explored in this chapter. Further, its use of direct democracy measures only four times has limited voters’ ability to exert influence in this manner. This supports the openness hypothesis and frequency hypothesis, but does not support the decentralization hypothesis as these mechanisms are even further limited at the subnational levels. Analysis The openness of the direct democracy system is related to usage. The Swiss system has the most open and permissible processes of all the cases analyzed here. The Canadian case has more permissible systems at the provincial and local levels than at the national level. The Philippine and Uzbek systems are fairly limited in public ability to initiate the process, and their direct democracy systems illustrate the opportunities to manipulate the process and the people. The frequency hypothesis has support in the Swiss and Canadian cases; however, the Philippines and Uzbekistan test the validity of this hypothesis. One could argue that the cases that use direct democracy at multiple levels will necessarily provide evidence of use at each level. However, looking at the quality of usage, the Swiss system in particular and the Canadian system use less manipulation than the Philippine and Uzbek cases. Additionally, the severe limitations of the Uzbek case prevent much usage beyond the national level. Thus, although these cases do use direct democracy frequently, the Philippine and Uzbek case studies demonstrate that their usage is suspect.

Federalism: Power Struggles

115

There is support for the decentralization hypothesis in that the Swiss, Canadian, and Philippine systems utilize direct democracy more at the lower levels than at the higher levels of government. As noted in the argument of the decentralization hypothesis, there are several possible reasons for this, including lower stakes in citizen involvement at these levels and these levels of government being closer to the people. Regardless of the type of governmental system, there appears to be more usage at the local and state levels than at the national level. The European Union The next section is an illustration of the use of direct democracy at the supranational level, providing the ultimate confederal system to investigate direct democracy usage. The European Union provides a unique opportunity to explore the use of direct democracy on a mixed level because of its supranational characteristics, combined with the national-level referenda used to ratify issues. EU member states have utilized these mechanisms at the country level as well as at the supranational level, which are detailed in Table 5.3. There have been notable uses of national referenda among countries joining the EU (Qvortup 2014a; De Vreese and Semetko 2004; Hug and Sciarini 2000; Bogdanor 1981). The role that national referenda played during integration into the EU is particularly notable and dates back to the 1990s in some countries (Denmark, France, and Ireland). One key feature is that direct democracy has been important in garnering support for different expansions and changes within the EU. European Integration

Western democracies used national referenda rarely until European integration. The issue of European integration was framed in the context of each individual country, resulting in different issues, political elite cues, and challenges. The main reason that states have utilized referenda beyond constitutional requirements is that integration cut across party lines and even socioeconomic cleavages (Bogdanor 1994). One key issue was the order in which the integration votes occurred. European integration advocates used the domino-strategy (Bjorklund 1996) to synchronize votes, starting

116 Table 5.3 The European Union and National Referenda Name

Explanation

Countries and Dates

Maastricht Treaty

Initial treaty to integrate Europe.

Italya 1989; France, Ireland, and Denmarkb 1992c

Single European Act

Set the European Community an objective of Denmark 1986; Ireland 1987 establishing a single market by Dec. 31, 1992.

Treaty of Amsterdam

Treaty to give powers reserved for national governments to the European Parliament.

Ireland and Denmark 1998

Euro

Whether or not to accept the terms of the Maastricht Treaty.

Two referenda were held, both rejected accession: Denmark (rejected) 2000; Sweden (rejected) 2003

Treaty of Nice

Reformed the institutional structure of the European Union.

Treaty Establishing Ratify the treaty establishing a constitution. a Constitution for Europe Treaty of Lisbon European Fiscal Compact Unified Patent Court

Greek Bailout Referendum

Danish European Union Opt-Out Referendum

Ireland (rejected) 2001; (accepted with the removal of common defense policy) 2002

Spain (favor); France (against); the Netherlands (against); Luxembourg (favor) 2005

To continue to create a coherent European Union Ireland (rejected) 2008; with efficiency and democratic legitimacy. (ratified) 2009 Requiring member states to adhere to stricter fiscal policies.

Ireland 2012

A proposed court between member states for litigation related to European Union patents.

Denmark 2014

Whether to change to opting into participating in the area of justice and home affairs.

Denmark 2015

A Greek referendum to decide whether it would Greece 2015 accept the bailout conditions to its debt crisis proposed by the European Commission, the International Monetary Fund, and the European Central Bank.

Dutch Ukraine– A consultative referendum initiated by over European Union 300,000 Dutch citizens. Association Agreement Referendum

Denmark (rejected) 2016

Sources: Compiled by the author from European Union 2017; Bogdanor 1981. Notes: a. A consultative referendum to create a mandate for the European Parliament to elaborate the European Constitution. b. Denmark held two referenda on the Maastricht Treaty; the first was defeated in 1992 but, after negotiations and the creation of four opt-outs for Denmark, a new referendum was held in 1993 and voters approved the treaty with the opt-outs. c. Ratification after treaty was signed.

Federalism: Power Struggles

117

with safe countries that would support integration and then using their support to convince others of the risk of potential isolation (Jahn and Storsved 2016). In Chapter 2, I discussed Hungary’s use of referenda to join the European Union to show how these referenda were specifically used. The countries of the European Union have utilized referenda to determine both their willingness to join and their desire to leave. There have been notable uses of national referenda among countries joining the EU (Qvortrup 2014a; De Vreese and Semetko 2004; Hug and Sciarini 2000). The role that national referenda played during integration, and in the recent referendum in Great Britain to leave the EU, demonstrates the power of these referenda within the EU. From timing the referenda in sequence to manipulating voters to campaigns that played on the fears of voters—the power of these referenda have long-term economic consequences. Most recently with the referendum in Great Britain on the future of its connection with the EU, secession through referendum seems plausible. Key referenda in the European Union have focused mainly on expansion. Member states that have held referenda to join the EU include: France, Ireland, Norway (voted against joining), and Denmark in 1972; United Kingdom in 197515; Greenland (rejected joining) in 1982; Austria, Finland,16 Sweden, and Norway (failed) in 1994; Malta, Slovenia, Hungary, Lithuania, Slovakia, Poland, Czech Republic, Estonia, and Latvia in 2003; and Croatia in 2012. Although referenda are an option for member states who seek to join the EU, they are not required by the European Union. However, some member states may be required by their own constitutions to use referenda. Voting in these expansions has been the high point in literature on the European Union. Voting is seen to be a result of external factors such as the economy, immigration, and even the length of the incumbency in government more so than the EU itself (Hobolt 2009; Qvortrup 2001, 2014a). As such, convincing voters to support further integration in the EU is a reflection of greater political circumstances. Secession

There have been two referenda in the United Kingdom to leave the European Union/European Community. The first was in June 1975. Political parties and most stakeholders were in favor of maintaining the membership, but the Labour Party initiated (and voted 2−1 in

118

Direct Democracy: A Double-Edged Sword

support of withdrawing) and drove the movement for withdrawal. In an effort to unify the country and to deal with the division of European Community supporters and dissenters in the cabinet, a referendum was called to address the issue. Voters were asked: “Do you think the UK should stay in the European Community (Common Market)?” There was a 64 percent turnout, and 67.2 percent voted to remain in the European Community. In 2016, the United Kingdom held a second referendum on the EU (see Figure 5.1). This came after 3 years of intense pressure on Prime Minister David Cameron from both the public and members of his party. Although Cameron won reelection in 2015 with a clear majority, the pressure from the public for this referendum was clear. The European Union Referendum Act 2015 was presented to Parliament to permit a referendum. Although Cameron supported remaining a member of the European Union, there were a number of areas that he wanted to renegotiate.17 Alas, negotiations between groups about the future of membership in the EU yielded a scheduled referendum in 2016. One key facet to the referendum was who would be eligible to vote, as the United Kingdom has quite permissible voting rights. Eligible voters included UK residents who were British or Commonwealth citizens.18 Voters had to register by June 7, 2016, and 46.5 million people were eligible to vote. The focus of the referendum was to leave or to remain in the European Union. Several groups were associated with each side of the campaign. In fact, the referendum had many facets, particularly divisions within parties supportive of each side. Members of the government were supportive of different sides of the issue. The government itself launched a campaign to remain a member of the European Union. Polling indicated that the public wanted more information from the government about the issues, so the government distributed leaflets to every household. However, it was highly criticized for spending money on the campaign—estimated at 9.3 million pounds (roughly US$13.8 million). Further, a voting guide was also sent to every household to raise awareness of the upcoming referendum in May 2016; both sides were given full pages to explain their positions. Polls were open from 7 A.M. until 10 P.M. on election day, and more than 41,000 polling locations were offered. Ballots could also be cast by mail. The ballot question was researched by the Electoral Commission, and consultants felt that it was written in a neutral tone and was clear and straightforward. The results of the referendum were clear, with 51.9 percent voting to

Federalism: Power Struggles

119

Figure 5.1 UK Ballot on Membership in the EU

Referendum on the United Kingdom’s membership of the European Union

Vote only once by putting a cross to your choice

X

in the box next

Should the United Kingdom remain a member of the European Union or leave the European Union?

Remain a member of the European Union

Leave the European Union

leave the EU and 48.1 percent in favor of remaining. A total of 33.6 million votes were cast, a turnout of 72.2 percent of registered voters. A petition was circulated in the aftermath to hold another referendum, but it was dismissed by the government. Although the 2016 referendum is only advisory, the government did announce that it planned to follow the results and start the withdrawal process. Future Enlargements

The expansion of the EU and North Atlantic Treaty Organization (NATO) membership has extended direct democracy in a number of Eastern European countries. This connection with the West and additional democratic procedures have changed and challenged traditional governing in this region. Many of these countries have semiauthoritarian regimes that have histories of human rights violations. The success and failure of EU expansion in this region points to not only the economic and social repercussions, but also the democratic institutions that need to be utilized to conform to this organization. The literature demonstrates a painful process and substantial complications for countries such as Slovakia (Rybár 2005; Bucek 1999) and Belarus (Eke and Kuzio 2000). Countries that seek to join the EU may hold a referendum

120

Direct Democracy: A Double-Edged Sword

as part of that process. Sometimes these referenda are required by the state (e.g., France and Turkey); in other cases, referenda are seen as consultative to make sure that the government is entering into an agreement with a larger organization that the public favors joining. Supranational Level

The treaty establishing a constitution for Europe, which was ultimately rejected, included the ability to propose initiatives. These initiative proposals would have required 1 million signatures from a minimum number of member states and could have proposed different issues for the European Commission to consider. This would have applied to only the EU and would not have been a part of national-level governing.19 Because of the proposal’s vagueness, which failed to indicate the exact mechanism of direct democracy, it did not move forward. The Lisbon Treaty, ratified in 2009, enabled limited initiative rights. Proposers of a European Citizens’ Initiative must consist of at least seven European citizens from at least seven member states. These initiatives must deal with substantial issues within the EU. A proposed initiative must include a title, a brief description, a list of objectives, and references to treaty provisions that are relevant to the proposal. This must be posted in one of the official languages of the EU online (translations into other languages may also be posted online, but these translations need to be approved by the European Commission for inconsistencies in the translations).20 The proposers are required to have a minimum of 1 million European nationals sign the petition. Proposers can also present legislative proposals to the European Commission. 21 Signatures can be gathered in electronic or paper form, within 12 months of registration with the European Commission. After the signatures are collected, they are presented to the European Parliament, which provides an official hearing. Although no vote is held (at this point) by the citizens of the EU, it does provide the first transnational/supranational direct democracy mechanism (Kaufman 2012). Opponents have argued that this places an undue burden on proposers given the diversity of languages spoken in the European Union. Further, initiatives (seen as a grassroots way to offer new legislation) are considered different from the right to petition (which is used to protest perceived infringements on European law). Although

Federalism: Power Struggles

121

there have not yet been any full initiatives presented to the European Commission, attempts have been made in the areas of water rights, traffic, and environmental protections. The regulations for the European Citizens’ Initiative went into effect in April 2011, with the first initiatives launched 1 year later. So far, there have been three successful initiatives, and six are currently open for signature collection. The process is slow, but the option is there for citizens to engage in and influence legislation in the EU. The EU adds to the literature on direct democracy in multilevel governance. Although the EU has not had the same experiences as the other cases I investigated in this chapter, this case does demonstrate the challenges of multilevel uses of direct democracy. The national use of referenda to determine integration or secession demonstrates the challenges of direct democracy at multiple levels. Although my hypotheses do not specifically address the unique circumstances of the European Union, as the EU moves forward with direct democracy on the supranational level this will become an area of future study. Notes 1. Canton votes are determined by the majority vote within the canton so, if 51 percent of a canton votes in favor of a referendum, the canton’s one vote would be cast in favor of the referendum. 2. For a complete listing of these, please see Serdült (2014). 3. Prior to 1978, only 50,000 signatures were required. 4. Some cantons have increased the number of required signatures, whereas others have reduced them (e.g., Valais, which halved its required number of signatures). 5. In 1982, when Canada patriated its constitution, the National Assembly of Quebec did not assent to the new constitution, leaving it outside the framework. 6. British Columbia included a question on popular initiatives in its 1991 referenda that passed with 83 percent of the vote, yet it has not yet been enacted at the provincial level. 7. Referenda in the Philippines have had varied uses. In 1973, they were used to extend Ferdinand Marcos’s presidential terms (Kobori 2014). 8. Referenda were held in the years following this revision; however, when Marcos took power in the 1970s he initially used referenda, but there is little evidence to suggest that these were “free or fair in any meaningful sense of the term” (Kobori 2014: 212). Furthermore, Marcos’s use of referenda suggests that it was only a facade of complying with constitutional law (Smith 1986).

122

Direct Democracy: A Double-Edged Sword

9. It should be noted that there was a majority no vote in 9 provinces and 11 cities (Smith 1986). 10. There are two autonomous regions in the Philippines—Cordilleras and Mindanao. These are the primary administrative levels responsible for the regions’ political, social, and economic well-being. There have been several national referenda on autonomous regions. As of 2012, there is an agreement about establishing a third autonomous region in Bangsamoro. 11. Barangays are the smallest administrative unit of the Philippine government. The term barangay is the Filipino word for a village, district, or ward, which is composed of a small number of families (generally 50–100). Direct democracy is common at the Barangay level. 12. Incidentally, when Lucilo Bayron ran for mayor, his opponent was Edward Hagedorn’s wife. 13. The 1991 referendum followed the perestroika period in the Soviet Union, and the referendum was used to resolve questions about the future of the Union of Soviet Socialist Republics (USSR). Although several elections and separatist movements followed (Hill and White 2014), the referendum was designed to determine national unity. 14. The mixed experience of referenda under communist and postcommunist rule in the Soviet Republic and post-Soviet Russia illustrates a conundrum of being both democratic and authoritarian. This mechanism is used by those regimes that are openly undemocratic and those that practice democracy (Hill and White 2014). 15. The United Kingdom did not initially have a referendum to join; rather, it joined in 1973. Given political pressures, it held a referendum to remain in 1975 and one to leave in 2016. 16. The Aland Islands, an autonomous region of Finland, also voted in support of its entrance to the EU in 1994. 17. Such as protection of the single market for noneurozone countries, the reduction of bureaucratic measures, exempting Britain from additional merging with the EU, and restrictions on immigration from the EU. 18. Citizens of Ireland, Malta, or Cyprus who were residents of the United Kingdom were also able to vote. British citizens who had lived in the United Kingdom within the past 15 years but currently lived abroad were also able to vote. 19. Albeit there may be repercussions at the national level. 20. Translations are a responsibility of the proposers, but checking them must be a function of European Commission. 21. Signatories have to provide verification of citizenship, but that verification depends on the member state. So in Finland, signatories provide their name, address, and date of birth, whereas Italian signatories have to provide additional information such as place of birth and passport number (Kaufman 2012). Signatures need to be verified by the member state from which signatories claim citizenship. These various requirements can also lead to different levels of support.

6 Authoritarian Governments: Limiting Citizen Power

into reinforcing authoritarian regimes. More authoritarian regimes have multiple reasons for developing these mechanisms, but they are not always for the same types of democratic underpinnings that exist in nonauthoritarian regimes. In this chapter, I look at the development of direct democracy in Egypt, Thailand, Venezuela, and Kyrgyzstan and how these mechanisms are utilized to create a more powerful regime and advance authoritarianism. Direct democracy is touted as a way to involve people in the governing process, but these mechanisms can also be used to achieve alternative policy ends. At times, referenda that serve the populist notion of reigning in executive power are used as a weapon by illiberal governments (Qvortrup 2017). Notably, governments support governing decisions under the guise of democracy, but do so as a way of reinforcing the authoritarian regime. These cases studies show the development of direct democracy in each country, their usage and success in determining citizen involvement, and support for these regimes. In more authoritarian regimes, referenda are used to quell domestic conflict and international pressures required for international aid (Tripp 2004). The “general tendency . . . to enhance the power of the presidency and reduce democratic provisions” means that unconstitutional measures are frequently used when introducing Citizens can be manipulated under the guise of direct democracy

123

124

Direct Democracy: A Double-Edged Sword

these referenda (Hill and White 2014: 35). Often, a single referendum is held in these regimes to provide an opportunity for the people to demonstrate support for and, thereby, legitimize their leaders (Smith 1986; Qvortrup 2017). Beyond the examples provided in this chapter, there are illustrations throughout the book of leaders who utilize referenda to legitimize regimes, such as Ferdinand Marcos in the Philippines, Hussain Ershad in Bangladesh, Vladimir Putin in Russia, and Yoweri Museveni in Uganda (as well as others not discussed in length in this book: Evo Morales in Bolivia, Rafael Correa in Ecuador). These regimes have used referenda as a strategy to maintain their power base and change the political system. In every case, the opposition is kept divided and weak. Many times, these elections are fraught with intimidation, lopsided media coverage, and outright fraud to ensure specific results (Qvortrup 2017). Not all direct democracy processes are used long term—often, they develop due to only one issue or, alternatively, the long-term desire for additional democratic institutions. This second type of direct democracy is actually more of a development of antiparty sentiment, a challenge to the existing democratic or authoritarian regime, a product of historical circumstances, or a response when the government is regarded as too strong or too weak (Bjorklund 1982). Romania’s skirting of the authoritarian democracy terminology caused it to change its laws in 2003 to create a limited way for referenda to be used—for the revision of the constitution and relegation of the state chief. This allowed the country to build legitimacy through this mechanism and to create a more authoritarian state leadership by giving the president greater authority when dealing with the prime minister and Parliament. Thus, as the state opens the door to these mechanisms, it also invests more power in executive leadership, which could have a dramatic impact on its quality of democracy (Băişanu 2010). Why Do Authoritarian Regimes Use Direct Democracy? I propose additional interrelated theoretical arguments for why authoritarian regimes utilize direct democracy. Due to the challenging nature of authoritarian governance, many regimes face both internal and external threats to their leadership. I propose that authoritar-

Authoritarian Governments: Limiting Citizen Power

125

ian governments utilize direct democracy (typically referenda) to build their national unity,1 find support for the regime, manipulate the public, and provide a semblance of democracy in their state. Nation-building hypothesis: authoritarian regimes will utilize referenda to build a national identity.

Authoritarian regimes utilize direct democracy as a way to build the solidarity of the state. For example, the borders of many Central Asian republics were designed by Joseph Stalin not necessarily due to commonalities or unity but rather to split the opposition to his regime, resulting in a form of gerrymandering to protect the regime’s authority. African states have similar roots, with borders designed due to colonial interests and road development rather than community understanding (Kersting 2009). This has led several fragmented states to turn to authoritarian regimes. These regimes then need to build solidarity among their citizens to solidify their power base. Many of these regimes use direct democracy to do this. The topics of many of these measures are connected to this idea of nation building and forging a unified state. The nation-building hypothesis is directly connected to the literature presented above. The independent process of many colonial countries (Tripp 2004) shows the need for forging a new unity; this can also been seen in the development of Latin American democracies (Levitsky and Loxton 2013) and expansion of the European Union. Furthermore, it may be used to fortify the leader of the regime (Schmitt 1926). Instability of regime hypothesis: authoritarian regimes will utilize referenda where there is instability in the regime to increase support for the leadership.

Although the use of direct democracy in authoritarian regimes may be surprising, it is particularly concerning given the authoritarian regimes that utilize these mechanisms. Direct democracy tends to be used in more unstable authoritarian regimes; this can be explained in conjunction with the two previous hypotheses in that this is a way to build and maintain support for the state. Regionally, specific countries in the Middle East that have cracked down on dissidents have the highest use of direct democracy in the region. Those who threaten the regime are often dismissed through the development of these types of neopopulist activities (Barczak 2001).

126

Direct Democracy: A Double-Edged Sword

Regime stability is of utmost importance to authoritarian governments to maintain their power bases (Kersting 2009; Qvortrup 2017). Using these measures to garner support for the regime leads to increased support from the people. Manipulation hypothesis: authoritarian regimes will use referenda to manipulate voters into supporting the regime’s mandate.

Connected to the instability hypothesis is manipulation. Direct democracy is used in these regimes as a way of manipulating citizens of these countries into a feeling of efficacy, but they do not really have any influence on the larger nature of the country. This manipulation serves as a way of continuing to obtain support for a regime that does not always respect citizen or human rights. This is also connected to the theoretical underpinnings of the next hypothesis. Efficacy hypothesis: authoritarian regimes will use direct democracy to build efficacy among voters.

The turnout in these elections varies greatly. I propose that direct democracy is used to provide a feeling of efficacy in the government due to the democratic openings, as voters who are allowed to participate in the process are more likely to feel like their voice matters. The higher the use of direct democracy in authoritarian regimes, the more voters will feel they are part of the process, even if they are being manipulated or tricked. Case Studies The case studies in this chapter provide an illustration of the motivation, success, and connection between the state and authoritarian regime when direct democracy mechanisms are employed. The global use of direct democracy demonstrates who buys into these national-level mechanisms (for details, see Beramendi 2008). Authoritarian regimes are more likely to utilize these measures than nonauthoritarian regimes.This is particularly true in Asia and the Middle East. I selected these cases because of the varied experiences with the direct democracy process, different regional locations, and differing historical experiences with authoritarian regimes. This

Authoritarian Governments: Limiting Citizen Power

127

chapter focuses on four case studies: Egypt, Thailand, Venezuela, and Kyrgyzstan. Egypt

Although the spring of 2011 influenced Egyptian democracy, the country had made inroads into direct democracy activities several years earlier. Egypt has utilized national referenda more frequently than any other Middle Eastern country, though followed closely by Syria2 (Beramendi 2008). Egypt has held referenda for a variety of activities—to dissolve Parliament (1987, 1990), confirm the president (1981, 1987, 1993, and 1999),3 and make constitutional reforms (most notably in 2007, 2011, 2012, and 2014). Table 6.1 provides an illustration of the use of direct democracy in Egypt, topics, and vote totals. Egypt is an excellent case study to compare with other authoritarian regimes because of their controversial use of direct democracy and turnout rates. Evidence exists that the referenda have also been rigged (Marques and Smith 1984). In 1980, authorities claimed to have counted 10 million votes in 6 hours (Marques and Smith 1984). Evidence also exists of intimidation regarding voter turnout; the regimes introduced fines for nonvoting and continued to increase the amount of the fines over time (Blaydes 2006). In 2007, the proposed referendum involved extensive constitutional reforms: banning religious political parties, adopting new election laws and procedures, granting the president the ability to unilaterally dissolve Parliament, and new terrorism laws (Beramendi 2008). This vastly changed the power of the government with the potential to shape the country’s political future. It is concerning that, although 75 percent of voters approved this referendum, there was only a 27 percent voter turnout. Given the political history over the next few years, it would seem that these changes did the exact opposite of what President Hosni Mubarak intended with this referendum: the coercion and lack of transparent substantive democracy limited the quality of this type of direct democracy. The power of citizens in Egypt was lauded in 2011 with the emergence of the Arab Spring, and there have been a notable number of referenda on the constitution in recent years. These referenda had low levels of participation rates (hovering around 30–35 percent), illustrating that referenda are not necessarily a vehicle for change by mass populations. It would seem that the desire to hold

Table 6.1 Egyptian Referenda Title

1. Nassar for President 2. New Constitution

1. Formation of the United Arab Republic (UAR) 2. Selection of President UAR 30 March Program Federation of Arab Republics

Date

1956 1958

Referenda to select Gamal Nassar for president and ratification of a new constitution. Formation and selection of president for the UAR.

Statement of electoral reform. Change to develop a new Federation of Arab Republics with Libya and Syria. Constitutional Amendment Sept. 11, Changes to the constitution of Egypt. 1971 October Paper 1974 Four-phased approach to a multiparty system. Protection of National Unity 1977 Protection of National Unity 1978 and Social Peace 1. Egypt-Israel 1979 1. This treaty would create peace Peace Treaty between Egypt and Israel. 2. Change to Political 2. Changes to the system would include System multiparty politics and bicameral parliament through the creation of the Shura Council. Constitutional Referendum 1980 Creation of the Shura Council as the upper house in Parliament. Protection of National Unity 1981 Electoral Reform 1987 Reform set aside 48 seats for independent candidates in the legislature. Early Dissolution 1990 Referendum after the Supreme of Parliament Constitutional Court ruled that the 1987 elections were unconstitutional because the 1986 electoral law discriminated against independent candidates. Constitutional Referendum 2005 Established direct elections for president. Constitutional Referendum 2007 Changes to the electoral law that were passed by Parliament also in 2007. Constitutional Referendum 2011 Presidential term limits, appointments of vice presidents, commission a new constitution, and easier qualifications to the ballot for candidates. Constitutional Referendum 2012 Approval of new constitution.

Constitutional Referendum

1968 Sept. 1, 1971

Topic

2014

Approval of new constitution after the 2013 coup. Revisions to the 2012 Constitution.

Source: Nohlen, Krennerich, and Thibaut 1999. Notes: a. Unless otherwise stated, turnout in these measures is over 90%. b. Government critics suggest that it was actually closer to 5% turnout.

Resulta

1. Approved by 99.9% 2. Approved by 99.8%

1. Approved by 100.0%

2. Approved by 100.0% Approved by 100.0% Approved by 99.9% Approved by 99.9%

Approved by 99.9%

Approved by 99.4% Approved by 98.3%

1. Approved by 99.9%

2. Approved by 99.7% Approved by 99.9%

Approved by 99.5% Approved by 88.9%

Approved by 94.3% (markedly lower turnout at 58.6%)

Approved by 82.9% (turnout 53.6%) Approved by 75.9% (turnout 27.1%)b Approved by 77.3% (turnout 41.9%) Approved by 63.8% (turnout 32.9%) Approval of 98.1% (turnout 38.6%)

Authoritarian Governments: Limiting Citizen Power

129

referendum elections and gain legitimacy for the changes in the constitution would serve government stability more so than the protests that sparked the desire for change in government. Power may reside in the people, but they are not exercising it through direct democracy. Thailand

The 1992 protests in Bangkok gave hope to a new stable democracy in Thailand. Thais developed a strong civil society, held several free elections, and were even rated by Freedom House as Free in 1999. However, the 2006 military coup shattered that hope for democracy; Thailand moved toward semiauthoritarianism, with the military driving many of the policy decisions. Many of the reformers who had been so important in the 1990s grew complacent, partly as a result of the economic crisis that plagued Asian economies, and stopped working to forge and protect the democracy (this meant closing down nongovernmental organizations, media watchdogs, and even monitors). One of the reasons that Thailand’s democracy was so susceptible to decay was the rise of Thaksin Shinawatra (Baker and Phongpaichit 2014). Thaksin, a powerful businessman, used his personal finances to build a political party and run for prime minister. He expanded his connections and resources to recruit supporters and insulate himself from the judicial system. His policy proposals for health care and business loans were attractive to Thailand’s large poor population (Hewison 2014). Thaksin was elected in 2001 with 40.6 percent of the vote. But instead of supporting democratic institutions, he used his power to influence the media, civil service, and Muslim insurgents. Thaksin was also quick to reward those who supported his leadership with contracts or plum positions and to punish those who acted against him (Pathmananad 2016). He was reelected again in 2005 with 60.7 percent of the vote, drawing on the political power of the marginalized poor. Those who were disillusioned with his leadership did not use the election to contest Thaksin, but rather used a process similar to 1992 and went to the streets to protest. These protests led to a military coup in September 2006, and Thaksin went into exile (Fry 2014). Although the protests achieved their goal of removing Thaksin, the military eviscerated the democracy. The protest that followed in 2010 involved massive bloodshed, with 80 dead and much of the business district in flames.

130

Direct Democracy: A Double-Edged Sword

Two attempts have been made at referenda in post-Thaksin Thailand. In 2007, there was a referendum on the new constitution (Ockey 2008; Nelson 2011). The changes made to the constitution in this referendum allowed half of the senators to be appointed rather than elected, limited the prime minister to two 4-year terms, banned the prime minister from owning private companies, and created an accessible impeachment process for the members of government (Ockey 2008; Nelson 2011).4 Prior to the vote, each household received a copy of the new constitution. Though there appeared to be support from across the country (57.8 percent), substantial differences existed by region (Ockey 2008; Nelson 2011). In 6 provinces, approval of the new constitution was more than 90 percent, whereas in 24 provinces the majority was against the draft, and 2 provinces (Nakhon Phanom and Roi Et) rejected the referendum with over 75 percent of the vote (Ockey 2008; Nelson 2011). Voter turnout was at 57.6 percent. The 2016 referendum in Thailand was also constitutional (Jory 2016; McCargo 2017). Important changes to the new constitution included having the Senate appointed rather than elected, allowing the Senate veto power over the House of Representatives in amending the constitution, and permitting a prime minister to be appointed from outside either House. These changes significantly altered the democratic balance in Thailand, shifting it from a democratic power to a more authoritarian regime with vested power in the military. Despite these changes, the new constitution was approved by 61 percent of voters, with a 59 percent turnout (Jory 2016; McCargo 2017). The questions posed to the voters included: 1. Do you approve or disapprove of the draft constitution? 2. Do you approve that for contributing continuity of the country reform according to the national strategic plan, it should be stipulated in the Transitory Provisions of the Constitution of the Kingdom of Thailand that for the duration of 5 years from the first sitting of the National Assembly under this constitution; the joint sitting of the two chambers of the National Assembly shall convene to consider approving a person to be appointed as the Prime Minister?

This referendum was not without controversy. The National Council on Peace and Order banned criticism of the draft constitution and did not permit monitoring the referendum (Jory 2016; McCargo 2017).

Authoritarian Governments: Limiting Citizen Power

131

Activists against the new constitution were arrested, detained, and prosecuted in military courts, as were those who said they intended to vote against the referendum. Support for the referendum was also regionally divided (Jory 2016; McCargo 2017). The Isan region and three Muslim majority southern provinces rejected the referendum. A second proposal was held at the same time that allowed for the next prime minister to be jointly elected by the Senate and House of Representatives (Jory 2016; McCargo 2017). This second proposal was also approved. As democracy decays in Thailand, the use of referenda as a visage to democracy demonstrates that Thailand is not developing democratic procedures. Rather, the country continues to manipulate voters, and the democratic activities of a referendum actually restrict voters’ power. These most recent referenda have seen the prosecution of opposition forces and a move toward appointed rather than elected leadership. Thus, though the facade of procedural democracy is used through referenda, substantive democracy is eroding in authoritarian Thailand. Venezuela

Latin American countries adopted direct democracy mechanisms in the 1990s. The reason for this was twofold: first, issues in political representation issues aggravated the neopopulists (see Peru and Venezuela); and, second, internal crises forced the elites to pander to the people to maintain their system (Beramendi 2008). The initial conception of direct democracy was added to the Venezuelan constitution in 1999 as a way to increase participatory democracy. The constitution was amended to include consultative referenda, referenda to repeal and approve national and international laws and the constitution, and recalls. Other Latin American countries utilize direct democracy, but Venezuela is unique in that it has the fourth-highest frequency of using these mechanisms and is uniquely authoritarian in its democracy. The Venezuelan use of direct democracy is a more recent inception and is partially a result of the Hugo Chávez presidency. Table 6.2 provides an illustration of the use of direct democracy in Venezuela, topics, and vote totals. Skepticism exists regarding the use of direct democracy in Latin America due to the instability of the political system (Bolivia); conditions of the state (Colombia); and, more specifically, the context of the Bolivarian Revolution in Venezuela (Beramendi 2008). Due to this history, the use of citizen initiatives is

132

Direct Democracy: A Double-Edged Sword

less likely, and the government is more prone to use referenda to bolster the regime. Table 6.2 illustrates these topics and forums on a national scale. Under President Chávez, Venezuela utilized direct democracy in notable ways. Chávez proposed including additional forms of direct democracy in the constitution and made these changes through a 1999 referendum. When Chávez first asked Venezuelans to eliminate the presidential term limits in December 2007, they told him no. However, as if resigned to the idea that he would keep on asking until he received the answer he wanted, voters eventually said yes (Hellinger 2005). Venezuela’s second constitutional referendum in 14 months was approved by a resounding 54 percent to 45 percent margin, allowing Chávez to run for a third 6-year term in 2012.

Table 6.2 Venezuelan Referenda Title

Date

Presidential Terms

1957

Convening a Constitutional Assembly

April 1999

Approve the New Constitution

Dec. 1999

Topic

Voters were asked if they approved of the president staying in power without a new election and appointing a national parliament. Voters were asked two questions: first about whether to convene a constitutional assembly; and second on how to elect the constitutional assembly. Voters were asked if they endorsed the new constitution designed by the constitutional assembly. Suspended leaders of the unions until a trade union leadership election took place, consistent with the Venezuelan constitution. Recall referendum of President Hugo Chávez.

Trade Union Leadership

2000

Recall

2004

Constitutional Referendum

2007

Two proposals to amend the 69 articles of the 1999 constitution.

Amendment 1

2009

Abolished term limits for the president and other elected officials.

Source: Nohlen 2005.

Result

Approved by 86.7% of voters Question 1: Approved by 92.4% Question 2: Approved by 86.5% Approved by 71.8% (turnout 44.4%)

Approved by 69.4% (turnout 23.5%)

Defeated by 58.0% (international concern of election fraud) Proposal A—Defeated by 50.7% Proposal B—Defeated by 51.1% (turnout 55.9%) Approved by 54% (turnout 70%)

Authoritarian Governments: Limiting Citizen Power

133

The 1999 Venezuelan constitution allows for a recall when half the term of office of an elected official has elapsed and when at least 20 percent of the registered electors of the corresponding electoral district demand a recall referendum (Beramendi 2008). An official can be recalled and replaced when (1) the number of voters to approve the recall is equal or greater to the number who elected the official originally; (2) 25 percent or more of registered electors participate in the recall referendum; and (3) more voters vote to recall than not to recall. In the 2004 recall election (as discussed in Chapter 3), the opposition unsuccessfully tried to remove Chávez from power.5 This recall election solidified Chávez’s commitment to using direct democracy mechanisms. This led to the third notable use of direct democracy with the 2007–2009 referenda in Venezuela. Although there was an earlier attempt at a recall election in August 2003, it was thwarted by the ruling coalition, which declined the petition due to failure to comply with essential formalities. Thus, in September 2003, the regulating body—the National Electoral Council (CNE)— approved the rules to regulate a recall. These additional regulations prolonged the process to almost 11 months. Venezuelan direct democracy is employed in multiple fashions, first through the clear manipulation of voters to support the regime. If voters are unsupportive of a measure, they are likely to see it again, as demonstrated in the 1999 referendum. Campaigns are clearly influenced by the government and tainted by government involvement. The use of direct democracy, although designed to build legitimacy for the regime and create efficacy among voters, actually leads to the increased manipulation of voters in a less-thancovert manner. Kyrgyzstan

Kyrgyzstan touts itself as the most democratic of the former Soviet republics in Central Asia and uses these direct democracy mechanisms at the highest frequency in Central Asia. Legal provisions allow for referenda and initiatives at the national level, and 12 referenda have been held since 1991. Legal provisions for local-level referenda have been made as well. These local-level referenda are under the direction of the local council, which can determine the means and mechanisms for having these types of elections. The Law on Referendums (adopted

134

Direct Democracy: A Double-Edged Sword

on June 28, 1991) outlines these procedures only for national elections and has no provisions for their larger impact. Few subject limitations exist on the use of direct democracy in Kyrgyzstan, and they regard only referenda on civil service, taxes, public expenditures, and amnesty and pardon issues (International IDEA 2014). Furthermore, the president, legislature, and registered voters can initiate referenda. The president must approve the text for initiatives and referenda, indicating that, although there is flexibility in the process, the president makes all final decisions (International IDEA 2014). However, the system demonstrates some of the key issues with direct democracy. In Kyrgyzstan, to present an initiative, petitioners need to gather 30,000 signatures for a proposal for legislation, whereas a proposal for an amendment to the constitution requires 10 times as many (Beramendi 2008; International IDEA 2014). Table 6.3 provides an illustration of the use of direct democracy in Kyrgyzstan, topics, and vote totals. The 1991 referenda on the Soviet Union allowed republics to vote on the future of the union.6 In total, nine republics took part in the election. Among these republics, 98 percent voted in favor of a renewed federation, with a 70 percent turnout. In Kyrgyzstan, 95.9 percent voted in favor, with a turnout of 92.3 percent. The country also held a specific vote for its voters about only the Kyrgyz Republic and whether it should be a sovereign republic with equal rights. This had a much lower approval rate (62.2 percent) and 10 percent lower turnout than the unionwide referenda (Democracy Assistance and Elections News, n.d.). In January 1994, Kyrgyzstan was back at the polls. This time, the purpose was to extend the powers of the president (something that is also seen in subsequent referenda). The 1994 referenda expanded the powers to allow the president to act as head of state. This referendum, consistent with those preceding it, had high approval and turnout rates. In October 1994, the Kyrgyz had the opportunity to vote on two additional referenda: the use of referenda and the introduction of a bicameral legislature. Both of these were approved by almost 90 percent of voters, again with a high turnout of 86 percent (Democracy Assistance and Elections News, n.d.). Although this was a dip in turnout for Kyrgyzstan referenda, it was the second referendum in 1994, so voter fatigue may have played a role. In February 1996, Kyrgyzstan voted on approving a new constitution designed by the president (Fumagalli 2016). Here, we see a

135 Table 6.3 Kyrgyz Referenda Title

Date

Future of the Soviet Union

1991

Kyrgyzstani Sovereignty

1991

Presidential Referendum

Jan. 1994

Constitutional Referendum

Oct. 1994

Constitutional Referendum

1996

Constitutional Referendum

1998

Constitutional Amendment

2003

Presidential Referendum

2003

Constitutional Referendum

2007

Constitutional Referendum

2010

Question

“Do you consider necessary the preservation of the Union of Soviet Socialist Republics as a renewed federation of equal sovereign republics in which the rights and freedom of an individual of any nationality will be fully guaranteed?” “Do you agree that the Republic of Kirghizia should be in the renewed Union as a sovereign republic with equal rights?” “Do you confirm that the President of Kyrgyzstan who was democratically elected on 12 October 1991 for 5 years is the President of the Kyrgyz Republic with the right to act as head of state during his term in office?” Whether voters approved of using referenda to decide on amendments to the constitution, laws, and other important questions of political life. Whether voters approved of the introduction of a bicameral parliament. “Do you approve the law of the Kyrgyz Republic ‘On amendments and additions of the Constitution of the Kyrgyz Republic,’ a draft which was published in the Decree of the President of the Kyrgyz Republic on 3 January 1996?” Several amendments to the constitution including: • private ownership of land, • restructuring Kyrgyzstan’s bicameral parliament, • greater freedom for the media, and • limiting the immunity of deputies in parliament. Constitutional amendments, including the following: • concentration of power in the Supreme Council, • a unicameral assembly, and • more concentrated power in the Supreme Court. Extension of term and immunity for President Askar Akayev: • a provision asking if Akayev should stay in office to the end of his term, slated for December 2005, to implement the new constitutional amendments, • provide absolute and permanent immunity for the president and his family. “Do you agree with the new version of the constitution? (Revised after the Tulip Revolution)” “Do you agree with the new electoral law, as proposed by the President for the referendum?” Adoption of a new constitution that redefines the president-parliament relationship and places limits on dominance by a single strong political party or leader.

Source: Compiled by author from Kyrgyzstan Election Guide 2017.

Result

Kyrgyzstani results: 95.9% in support (turnout 92.3%) For 62.2% (turnout 81.7%)

For 97.0% (turnout 96.0%) Both approved by 88.0% of voters (turnout 86.0%) For 98.6% (turnout 96.6%) For 95.4% (turnout 96.4%) For 89.2% (turnout 86.7%)

For 91.8% (turnout 86.7%) For 95.4% (turnout 80.6%) For 91.8% (turnout 72.3%)

136

Direct Democracy: A Double-Edged Sword

similar pattern of turnout (96.6 percent) and approval (98.6 percent). In 1998, there was a constitutional referendum on a number of provisions. These provisions incorporated a variety of topics: private ownership of land, changing the number of seats in parliament (increasing in one and decreasing in the other)7, removing immunity for members of the legislature and their impact on the national budget, and increasing the freedom of the media (Fumagalli 2016). This referendum altered the balance of power between the legislature and the executive and gave freedoms to the people of the republic. The changes were approved by 95.4 percent of voters, with a turnout of 96.4 percent (Democracy Assistance and Elections News, n.d.). In 2003, Kyrgyz voters were presented with two referenda. The first referenda extended the term of the current president Askar Akayev and granted him and his family immunity. The second referendum made constitutional revisions that would give more power to the president at the expense of the Supreme Council and Constitutional Court. The presidential referenda passed with 91.8 percent approval, and the constitutional referenda with 89.2 percent approval. Turnout for these referenda was 86.7 percent, demonstrating the continued high support for the mechanism (Democracy Assistance and Elections News, n.d.). In 2005, Kyrgyzstan faced an internal conflict called the Tulip Revolution. During this revolution, President Akayev and his government were overthrown due to his perceived corrupt practices and authoritarian actions (see discussion of referenda above for elaboration) (Cummings 2013; Ulu 2008).8 The Tulip Revolution was mostly nonviolent and initially began as a response to the February 2005 elections, during which Akayev’s candidates received several victories; however, there were issues of election fraud and international observers were critical of the election process (Radnitz 2006; Cummings 2013). This was followed by a vote of no confidence in the legislature as a symbol of distrust in Akayev. Opposition within the legislature continued to build, and protests occurred throughout the country.9 Due to substantial protests on March 19 and 20, Akayev asked the Central Elections Committee to investigate claims of election fraud, focusing particularly on the districts that had caused concern for the protestors. Though a step in the right direction, it was insufficient for opposition forces, which increased their pressure on Akayev. On March 23, Akayev and his family fled to Kazakhstan and then Russia. In the following weeks, stabilization

Authoritarian Governments: Limiting Citizen Power

137

returned to Kyrgyzstan, Akayev resigned as president, and a new parliament was installed. The new regime held elections later in 2005 and worked on exposing corruption in the government (Altaweel, Sallach, and Macal 2012). The government installed new amendments to the constitution in 2006; however, the Constitutional Court of Kyrgyzstan invalidated these amendments, resulting in constitutional referenda in 2007. The referenda focused on voter approval of the new constitution and changes to the electoral law. The new constitution continued to consolidate power in the hands of the president by giving him the power to appoint or dismiss government, civil services, judges, bank directors, members of the electoral commission, and members of the National Security Council. However, the president would have no authority to dissolve the Supreme Council. Despite international concerns about the referenda (a low information environment and a number of irregularities), both passed with 95 percent approval and 80 percent turnout (Democracy Assistance and Elections News, n.d.; Juraev 2008). Political turmoil continued in Kyrgyzstan after the 2007 referenda and precipitated a revolution in April 2010 during which President Kurmanbek Bakiyev, who served after President Akayev, was ousted. This revolution was more violent and culminated from the increasing ethnic tension between the Kyrgyz people and the Uzbeks in southern Kyrgyzstan. The conflict lasted for 8 months and resulted in the deaths of 2,000 people and the displacement of 400,000 (Altaweel, Sallach, and Macal 2012; Cummings 2013). The 2010 constitutional referendum in June 2010 was designed to limit the president’s powers and return more power to the people and the legislature. The focus was on creating a parliamentary democracy and encouraging political parties to focus on political issues rather than those of religion or ethnicity. The voters supported the referenda, with more than 90 percent in favor (Democracy Assistance and Elections News, n.d.). Numerous observers felt that the referendum would empower political parties that wield power in an extremist manner. A parliamentary election occurred at the same time as the referendum. The pro-Bakiyev political party (Ata-Zhurt) campaigned on rolling back some of the initiatives in the new constitution and bringing Bakiyev back from exile. It won with a plurality of the vote. The violence that erupted after the referendum and election results left 500 dead and caused 100,000 Uzbek refugees to flee (Huskey

138

Direct Democracy: A Double-Edged Sword

and Hill 2011). Many international actors were concerned with the growing humanitarian crisis in the southern region and asked the government to act. The referendum had opened the door for a new presidential election in 2011, and voters went to the polls in October 2011. There were 16 candidates, including a representative of the Ata-Zhurt party. Ultimately, former prime minister Almazbek Atambayev, leader of the Social Democratic Party of Kyrgyzstan which had championed the new constitution, won the presidential election by a 50 percent margin. Analysis The four authoritarian democracies in these case studies have experienced varied successes and failures with the direct democracy process. Each has had its own experiences and reasons for development. However, common themes are apparent. All four of the cases studied used referenda to develop a national identity by amending the constitution to make it their own. Egypt used referenda to dissolve Parliament and confirm presidents as a way of providing a national identity. Thailand utilized referenda to create new constitutions and reorganize government structure. Venezuela used referenda to repeal and approve national and international law. Kyrgyzstan used referenda to determine its future and design a new constitution as it became independent from the Soviet Union, thus moving its national identity forward through citizen participation in referenda. Although each state has its own experience with instability, the Arab Spring in Egypt in 2011 created instability that led to a referendum on a new constitution and government, whereas in Venezuela the instability around Chávez’s leadership in 2004 led to the increased use of referenda mostly as a way to garner more political support for the existing leader. Thailand’s leadership change and resulting instability led to the use of referenda post-Thaksin. Furthermore, the overthrow of two different governments and its emergence as a new state created instability for Kyrgyzstan in its experience with referenda and led to several hundred deaths of Uzbeks in Kyrgyzstan. Each of regimes mentioned above made efforts to seem more transparent and engaged in democracy; however, the resulting elections and actions demonstrate that all regimes were trying to encour-

Authoritarian Governments: Limiting Citizen Power

139

age democracy to manipulate voters into believing they had efficacy, which in reality was a fallacious concept. All four cases saw the transition of a democratic regime into a more hybrid or authoritarian democracy during these experiences with direct democracy. The first instances of direct democracy in most of these regimes had high turnout at a rate of more than 90 percent. However, more recent usage of these mechanisms tells a different story of far less turnout. There are multiple reasons for this. First, accounting for turnout numbers is more precise in today’s electoral systems. Second, information diffusion through the Internet has increased awareness about democratic procedures and the idea of substantive democracy. This may have led to more disillusionment with the political process and, consequently, decreased turnout. However, it also may be that the novelty of these elections eventually wears away and voters become weary of participation (much like the ballot fatigue experienced in countries that had several elections on the same ballot or multiple elections in close proximity to each other).

Notes 1. The Third Reich used referenda in 1933, 1934, 1936, and 1938. These referenda were “difference-eliminating” referenda that sought to build national unity and support for the regime (Qvortrup 2014b; Hitler 1934). 2. Syria’s troubled political atmosphere in 2013 and 2014 indicates that its regime is also substantially authoritarian. However, referenda have been used in Syria dating back to at least 1971. The majority of these referenda were to replace traditional elections, in that they confirmed presidents rather than electing them. In 2007, President Bashar al-Assad was confirmed through a referendum for a second 7-year term. A notable difference from Syria’s Egyptian counterpart, the 2007 referendum had a reported turnout of 95 percent of the population with a yes vote of 97 percent. Given the political turmoil that occurred post-2007, it seems that this was not a true democratic activity, but rather a mode of manipulation of the people in which voters seemed to have a voice but the results were not truly democratic. 3. The number of referenda increased in the 1970s and 1980s in North Africa, specifically in Egypt. Political leaders chose direct democracy as a way to legitimize and strengthen their positions, particularly in Egypt and Libya (Kersting 2014). President Mohamed Morsi (from 2012 to 2013) ran his campaign against this type of fake democracy, but once in office attempted to utilize these mechanisms to further his authoritarian rule, enraging voters (Kirkpatrick 2012). However, the military government that followed him proposed additional referenda (Kersting 2014).

140

Direct Democracy: A Double-Edged Sword

4. Had the constitution been rejected, the military government could have adopted any previous constitution. 5. The recall was initiated by 2.4 million Venezuelan citizens. A total of 40.6 percent of the voters supported the recall, and the president remained in office (Beramendi 2008). 6. The newspapers covering the 1991 referendum referred to it as “Soviet Vote Becomes a Test of Loyalties” (Schmemann 1991). Although Mikhail Gorbachev was supportive of more openness, he was not in favor of dismantling the Soviet state. Thus, the referendum was designed to reaffirm allegiance to the Soviet Republic and, ultimately, had a substantially negative impact on the union’s future. Additionally, Gorbachev was also eager to secure his own position of leadership and prevent power sharing with political parties or subnational governments (Qvortrup 2014b). 7. In 2007, Kyrgzstan changed to a unicameral legislature. 8. Color Revolutions (nonviolent resistance in response to authoritarianism) occurred in several former Soviet states in the 2000s. 9. Solidarity protests also occurred in international locations in support of the Kyrgyz people.

7 The Power of Words: Secession Referenda

democratic way of declaring separation from a larger state.1 Countries are composed of different groups of people who from time to time want sovereignty and self-governance; as such, secession or nationalist movements will continue to evolve (Pavkovic and Radan 2007). In this chapter, I analyze notable cases to provide an in-depth understanding of the use and varying results of and responses to these referenda. Furthermore, I examine the language used in secession referenda to demonstrate how voters are manipulated at the polls over serious issues. When part of a country desires secession from the larger state, it often inspires conflict and challenges. Both peaceful and violent approaches to this process have been taken; in this chapter, I focus on the somewhat democratic and peaceful process of a referendum posed to a population to separate from the larger country. Referenda have been used successfully and unsuccessfully in the secession process. Unquestionably, citizen power is part of these secession referenda, but is that power tempered when campaigns and question options manipulate voters? I discuss in a big picture analysis why states use secession referenda and focus on four cases: Quebec, Scotland, Crimea, and Sudan. Secession referenda have gained notoriety in recent years as a

141

142

Direct Democracy: A Double-Edged Sword

Referenda have been used since 1846 to forge secession and create new states. There are currently 117 known secession referenda, and 4 are scheduled in the next few years.2 Table 7.1 provides details on these referenda. Most notably in the past was the use of referenda to decolonize 3 and the use of referenda to create new republics during the collapse of the Soviet Union.4 Separatist movements within the state usually drive secession referenda, and these referenda take place after negotiation or conflict. One potential benefit of a referendum (particularly when both the separatist movement and the state sponsor it) is acknowledgment from the international community and the recognition of its outcome—be it for secession or not. When the state does not support the referendum, the international community rarely accepts the outcome or the new government. Additionally, succession referenda can backfire spectacularly. Voters can punish the proposers who intend to utilize referenda for political gains (Jamaica 1961) and can also vote differently than expected in the polls (Britain 2016) (Qvortrup 2017). Thus, the proposers must expect that there is potential for the result to be different than they planned. Typically, secession referenda require the two quorums for turnout and approval (He 2002; Qvortrup 2014b). These quorums are put into place to ensure the legitimacy of the outcome and the will of the people. Requiring a turnout threshold as well as the requirements for a majority (simple or special) can make the difference in the passage or failure of a secession referendum. Special majorities often refer to the requirement of support from two-thirds of voters to ensure that this is the will of the people; these high requirements can also be unreasonable given the challenges of identifying voting populations or deep divisions within a country that initially caused the need for a referendum. In the preliminary aspects of a secession referendum, negotiations occur to determine the date of the election, the ballot question, and even voter eligibility. Typically, the current electoral rules of the state are used to conduct this type of referendum. However, issues such as who can vote, when the vote will be held, and who will monitor the election are topics of discussion prior to the referendum. Another issue that is typically negotiated prior to the referendum is whether the vote will be binding and, if this is the case, stakeholders must determine and agree on what will occur after the vote. Even with failed votes of secession (twice in Quebec and most recently with

143 Table 7.1 Secession Referenda, 1846−2014 State

Date

Switzerland Liberia

1802 1846

Texas Virginia Tennessee Norway Iceland Western Australia Iceland Cambodia Mongolia Faroe Islands Nagaland Saar Malta Togo Cameroon Central African Republic Chad Comoros Republic of the Congo Dahomey Djibouti French Polynesia Gabon Guinea Côte d’Ivoire Madagascar Mali Mauritania New Caledonia Niger Saint Pierre and Miquelon Senegal Upper Volta Jamaica Samoa

1861 1861 1861 1905 1918 1933

Maryland

1853

Former Colonial Power

France American Colonization Society Maryland State Colonization Society United States United States United States Sweden and Norway Denmark Australia

1944 1945 1945 1946 1951 1955 1956 1956 1958 1958

Denmark France China Denmark India Francea United Kingdom France Francea Francea

1958 1958 1958 1958 1958 1958 1958 1958 1958 1958 1958 1958

Francea Francea Francea Francea Francea Francea Francea Francea Francea Francea Francea Francea

1958 1958 1958

1958 1958 1961 1961

Majority Vote for Secession x x x

x x x x x x

x x x x x x x x

Yes Yes Yes

Yes Yes Yes Yes Yes No

x

x

Internationally Recognized Yes Yes Yes

No No No Yes Yes No

Yes Yes Yes No No No No Yes No No

Yes Yes Yes No No Yes Yes No Yes Yes

No No No No No No No No No No No No

Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes

No No No

Francea Francea Francea

Francea Francea West Indies New Zealand

Gained Secession

No No No Yes

Yes Yes Yes

Yes Yes Yes Yes

continues

144 Table 7.1 Continued State

Algeria Malta Rhodesia Djibouti Puerto Rico West Papua Northern Mariana Islandsb Bahrain Niuec Comoros Trust Territory of the Pacific Island Guam Aruba Djibouti Nevis Quebec Cickei Guam Micronesiad Marshall Islands Palaud Palaud Cocos Island Falkland Islands New Caledonia Slovenia Armenia Azerbaijan Croatia Estonia Georgia Kosovo Latvia Lithuanian Macedonia NagornoKarabakh Ukraine Transnistria Turkmenistan Uzbekistan

Date

Former Colonial Power

1962 1964 1964 1967 1967 1969 1969

France United Kingdom United Kingdom France United States Indonesia United States

1976 1977 1977 1977 1980 1980 1982 1983 1983 1983 1984 1984 1986 1987 1990 1991 1991 1991 1991 1991 1991 1991 1991 1991 1991

United States Netherlands France Saint Kitts and Nevis Canada South Africa United States United States United States United States United States Australia United Kingdom France Yugoslavia Soviet Union Soviet Union Yugoslavia Soviet Union Soviet Union Yugoslavia Soviet Union Soviet Union Yugoslavia Azerbaijan

1970 1974 1974 1975

1991 1991 1991 1991

United Kingdom New Zealand France United States

Soviet Union Moldova Soviet Union Soviet Union

Majority Vote for Secession x x x

x x x

x x x x

x

x x x x x x x x x x x x x x x

Gained Secession Yes Yes e

No No No No

Internationally Recognized Yes Yes Yes Yes Yes Yes Yes

Yes Yes Yes No

Yes Yes Yes Yes

No No Yes No No

Yes Yes Yes No Yes No Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes No Yes Yes Yes Yes

e

No Yes No No No No No No Yes Yes Yes Yes Yes Yes No Yes Yes Yes e

Yes e

Yes Yes

Yes No Yes Yes

continues

145 Table 7.1 Continued State

Bosnia and Herzegovina Montenegro South Ossetia Eritrea US Virgin Islands Puerto Rico Curacao Moldova Bonaire Saint Maarten Saba Saint Eustatius Bermuda Quebec Anjouan Nevis Puerto Rico East Timor Saint Maarten Bonaire Saba Kurdistan Curacao Saint Eustatius Montenegro South Ossetia Transnistria Tokelau Tokelau South Sudan Puerto Rico Crimea Donetsk Lugansk Veneto Scotland Catalonia Saint Eustatius

Date

1992

1992 1992 1993 1993 1993 1993 1994 1994 1994 1994 1994 1995 1995 1997 1998 1998 1999 2000 2004 2004 2005 2005 2005 2006 2006 2006 2006 2006 2011 2012 2014 2014 2014 2014 2014 2014 2014

Former Colonial Power

Yugoslavia

Yugoslavia Georgia Ethiopia United States United States Netherlands

Netherlands Netherlands Netherlands Netherlands United Kingdom Canada Comoros Saint Kitts and Nevis United States Indonesia Netherlands Netherlands Netherlands Iraq Netherlands Netherlands Serbia and Montenegro Georgia Moldova New Zealand New Zealand Sudan United States Ukraine Ukraine Ukraine Italy United Kingdom Spain Netherlands

Majority Vote for Secession x

x x x

x x x x x x x x x

x x x x

Gained Secession

Internationally Recognized

No

Yes No Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes No Yes Yes Yes Yes Yes Yes No Yes Yes Yes No No No Yes Yes Yes No No No No Yes Yes Yes

Yes e

Yes No No No Yes No No No No No No Yes No No Yes No No No No No No Yes e

No Yes No e e

No No No No

Yes

Sources: Compiled by author from Laponce 2010; Pavkovic and Radan 2007; LeDuc 2015; Qvortrup 2014. Notes: a. Referendum on French constitution. A no vote led to secession for the colony. b. Integration with Guam. c. Integration with New Zealand. d. Vote to become associated with the United States. e. De facto secession.

146

Direct Democracy: A Double-Edged Sword

Scotland in 2014), the separatist movement does not dissolve, but rather can reemerge in terms of political parties or as a future separatist movement. Finally, a large part of the decision to hold a referendum on secession is the ability to logistically hold a referendum. The issues of who should vote, who is eligible, where and when the vote should be held, and even the question itself create more problems than the referendum might solve (Qvortrup 2014b).5 The literature on secession referenda is a bastion of case studies with articles focusing on single cases (see, e.g., Darmonovic 2007 on Montenegro; Vidmar 2014 on Scotland). Little literature builds on a theory of why referenda are used for secession, and I hope to provide insight into this issue. Several related theories can be proposed about the use of referenda for secession. A number of arguments have been made about why proposed independent states choose referenda for secession. In this chapter, I analyze three hypotheses with respect to referenda: referenda that result from the decolonization process, state breakdown, and distinct identities. These three types of secession referenda use different forms of manipulation to challenge the true power of voters in this process. More specifically, I also examine a fourth hypothesis on how ballot language contributes to such manipulation. Decolonization hypothesis: secession referenda are held to create new states out of former colonies.

The decolonization hypothesis asserts that states use secession referenda to sever ties with their colonizers and forge new identities with more homogeneous states. When looking at the role that decolonization plays in secession referenda, it is clear that several states gained secession through this mechanism. As colonizers sought to focus their resources on their main countries, and as colonies sought to make more of their own decisions, secession referenda were a seemingly peaceful way of changing constitutions and creating new countries. Furthermore, this was a solution for both colonies and colonizers as it allowed colonizers to extricate themselves from their colonies and colonies to develop more self-determination. Constitutional referenda of these forms are primarily at the direction of the colonizer, with a few modern exceptions. When testing the decolonization hypothesis, it is vital to look at the early uses of secession referenda. Table 7.1 provides a time line of

The Power of Words: Secession Referenda

147

secession referenda and how they were used around the globe. In 1958, French colonies throughout Africa achieved secession through referenda on the French constitution. A no vote on the constitution enabled 19 states to gain secession. The countries forged new identities and developed new states as a result of their secession. Although African states were the most prolific in their secession, other secession referenda were used to form states from other colonial powers: Cambodia from France and Mongolia from China in 1945, Samoa from New Zealand in 1961, Bahrain from the United Kingdom in 1970, and East Timor from Indonesia in 1999, to name a few examples of successful decolonization referenda. There have also been failed referenda: Puerto Rico has made several attempts at secession from the United States through referenda and has failed each time. However, the vast majority of secession referenda are successful. State breakdown hypothesis: secession referenda are held during a period of state breakdown to forge new states.

Similarly, the state breakdown hypothesis explores states that broke down due to internal conflict and international pressure. Internal conflict creates the need for peaceful solutions, and states see secession referenda as a way of forging new countries. A secession referendum provides the legitimacy needed to achieve international recognition, involves the people in a democratic process, and allows states to forge new identities. To test the state breakdown hypothesis, it is important to look at two different, but similarly timed, state breakdowns. The Soviet Union and Yugoslavia conducted a number of secession referenda between 1990 and 1992. In the Soviet Union, nine states declared secession through peaceful democratic processes that the international community recognized. During the same time period, Yugoslavia, in the midst of serious internal conflict and civil war, experienced six secession referenda to find a solution to international pressure to bring peace to the region. Distinct identity hypothesis: secession referenda are held to recognize the distinctiveness of a specific area.

Factions within a country often argue that they have distinct identities and need special recognition such as statehood. This idea of the distinct identity hypothesis proposes that some portions of

148

Direct Democracy: A Double-Edged Sword

countries have different characteristics than other portions of the state. A distinct status leads to the desire for additional recognition or separation to celebrate the identity. Distinctiveness could be, for example, that the demographics or historical origins of a specific region of a country are different from the rest. In addition, distinctive identity referenda may be more about recognition than true secession from a country. These referenda are more likely to fail due to the desire for recognition rather than true secession as well as the desire to continue ties with the parent state but in a different form. For the distinct identify assertion, it is important to look at the efforts that have been made to recognize distinctiveness. This can be complementary to the state breakdown and decolonization hypotheses, as the call for referenda also come from factions within the state. Regarding the distinct identity hypothesis, a number of countries have areas that have attempted distinct identity secession referenda. Scotland and Quebec are good examples. Scotland has argued for the need to separate from the United Kingdom in an effort to forge a distinct identity and a unified state. Likewise, Quebec has desired distinctiveness from Canada since the British and French colonized Canada. Scotland’s and Quebec’s referenda were not borne out of conflict or civil war, but rather via more peaceful and legislative procedures. Ballot language hypothesis: the clarity of ballot language will impact vote choice.

Research has been conducted on the impact of ballot language on participation (Reilly 2010; Reilly and Richey 2011; LeDuc 2015). The more complex the ballot language, the less likely voters are to participate in voting on a measure (Reilly 2010; Reilly and Richey 2011). Furthermore, if questions remain about an issue, or if voters are confused about the consequences of a vote, they are more likely to vote against a given measure (Reilly 2010; LeDuc 2015). When it comes to secession votes, even more complex issues that are unknown at the time of the vote may exist. However, it is clear that if there have been years of negotiations regarding secession, and if all issues surrounding secession have been determined and publicized, the measure will pass because voters understand the complete consequences of their votes. However, if there is controversy or unknown issues, the referendum will more than likely fail, as voters

The Power of Words: Secession Referenda

149

are unsure of the consequences of their votes. Studies on secession referenda refer to the impact of these questions. For example, when polling about the Scottish independence vote, voters indicated that the wording of the question had an impact on the outcome.6 Question wording in Ireland, Chile, Andalusia, and East Timor have driven some of the results especially in the context of ethnonational issues (Qvortrup 2014b). Although all four of these hypotheses have evidence to support them, it is also necessary to look more closely at specific referenda, their campaigns, their ballot language, and their historical development. These are examined in the case studies below. Case Studies This chapter focuses on four notable cases from different regions involving secession referenda to varying levels of success. I selected Quebec, Scotland, Crimea, and Sudan for further examination because of their diverse experiences and international reception. These cases represent different regions, different results, and varied circumstances. Quebec and Scotland represent Western democratic processes that ultimately failed, whereas Crimea and Sudan each used a mechanism for successfully separating in the midst of conflict and international pressure. Furthermore, the language used in each case study represents different approaches and levels of voter manipulation. Quebec

Quebec’s secession movement has links to colonial powers and the conflict between the British and the French over who settled parts of Canada. Quebec’s secession movement gained momentum during the 1960s. Quebec has conducted two referenda on its future as part of Canada. Both of these referenda, conducted in 1980 and 1995, failed to garner enough support to facilitate a new nation-state, and the discussion of separation continues today. The 1980 referendum was the first on Quebec’s status within Canada, and it was driven by the provincial political party Parti Quebecois (PQ). Elected to a majority in the provincial legislation in 1976, the focus of the PQ’s platform was to create an independent nation-state for Quebec with economic partnerships with Canada.

150

Direct Democracy: A Double-Edged Sword

The PQ’s agenda was directly in conflict with that of Prime Minister Pierre Trudeau from Quebec, who was reelected in 1979. Trudeau participated heavily in the referendum campaign, focusing on the unnecessarily complex wording of the question as well as on what it meant to be a Quebecker. On May 20, 1980, voters were asked: The Government of Quebec has made public its proposal to negotiate a new agreement with the rest of Canada, based on the equality of nations; this agreement would enable Quebec to acquire the exclusive power to make its laws, levy its taxes and establish relations abroad—in other words, sovereignty—and at the same time to maintain with Canada an economic association including a common currency; any change in political status resulting from these negotiations will only be implemented with popular approval through another referendum; on these terms, do you give the Government of Quebec the mandate to negotiate the proposed agreement between Quebec and Canada?

Voters rejected the referendum, with 40.4 percent voting yes and 59.6 percent voting no. The overall electoral turnout was high at 85.6 percent. Despite the referendum’s failure, the Quebec separatist movement continued its quest. The 1995 referendum received support from not only the provincial political party, but also the new national separatist party, the Bloc Quebecois (BQ).7 Polling leading up to the referendum indicated that there would be a narrow margin between the yes and no votes (Cardinal 2005; Fox, Anderson, and Dubonnet 1999). Negotiations surrounding the 1995 referendum focused on creating an economic and social partnership with Canada should Quebec secede. In addition, the focus of the referendum was to seek sovereignty and then negotiate the partnerships with Canada after the election. However, a major stipulation of the provincial leadership was that, if negotiations broke down after a successful vote, the provincial government would be able to declare immediate secession. Another issue of negotiation was the question posed to voters. In the 1980 referendum, the question had sought only to provide authority to the provincial government to negotiate sovereignty, with another referendum to follow to ratify any decisions. But the 1995 referendum was on sovereignty specifically, with negotiations to follow (Cardinal 2005). The question used obfuscated the pure issue of separation and did not actually mention the word “separation”: “Do you agree that Quebec should become sovereign after

The Power of Words: Secession Referenda

151

having made a formal offer to Canada for a new economic and political partnership within the scope of the bill respecting the future of Quebec and of the agreement signed on June 12, 1995?” Voters were confronted with a long campaign arguing different sides of the issue; however, the language on the ballot made the decision less clear to voters and left open the possibility of discussion and cooperation between Canada and Quebec. This obfuscation is consistent with the literature (LeDuc 2015) which demonstrates that when voters are unclear about the circumstances or outcomes, they are less likely to support the measures. Between June and October 1995, each side campaigned to declare sovereignty (oui) or to remain part of Canada (non). Canadian prime minister Jean Chrétien was not popular among moderates, so he played a limited role in the campaign to allow the non side to be more successful (Chrétien 2007). Progressive conservative leader Jean Charest, who was also from Quebec and would go on to become the provincial premier from 2003–2012, also played a part in the campaign. The Oui campaign included PQ premier Jacques Parizeau and BQ leader Lucien Bouchard. The campaigns were divided in focus: the “oui” campaign focused on the need for Quebec’s sovereignty while the ‘non’ campaign focused on the uncertainty a sovereignty vote would create (Argyle 2011). The campaign became more intense when French president Jacques Chirac announced that France would recognize Quebec in the event of a successful sovereignty vote (Cardinal 2005; Philpot 2005). A complication that developed during the campaign was that of the aboriginal people and their inclusion in a sovereign Quebec or additional self-determination in the wake of a successful vote (Haljan 2014). How would their distinct identity be recognized in the new Quebec? This uncertainty, plus discussions about what the future of Quebec would look like (whether the country would join France, gain independence, or have a loose connection with Canada), made voters uneasy about the election’s outcome. In mid-October, a huge shift in polling numbers occurred; the yes side saw a surge and took the lead, forecasting the rest of the campaign. Finally, on October 30, 1995, with 4.8 million votes cast, 49.4 percent voted in favor of sovereignty and 50.6 percent in favor of staying unified with Canada. This was the highest turnout in Canadian history, with 93.5 percent of registered Quebec voters participating (Cardinal 2005). The unexpected results of the Oui campaign

152

Direct Democracy: A Double-Edged Sword

left the leaders unprepared to give concession speeches. Premier Parizeau called on French-speaking voters to maintain their support for the separatist movement and blamed the loss of the sovereignty vote on money and ethnic voters. The following morning, the PQ asked for Parizeau’s resignation based on these statements and on the failure of the referendum, and he complied (Hébert 2014). Following the loss of the referendum, Bouchard became the leader of the PQ and the premier of Quebec. Charest also left federal politics to become the leader of the Liberal Party in Quebec. The Quebec separatist movement has waxed and waned in the years since the referendum, but the PQ and the BQ continue to be present in Canadian politics to celebrate and recognize Quebec’s uniqueness. Quebec is an excellent example of the distinct identity hypothesis. Not only does the continued desire for recognition permeate Canadian politics, but the question is also clearly framed to continue ties with Canada in various matters. The future of aboriginal populations and their distinct identity were also left unsettled. Furthermore, the ballot language exacerbated the uncertainty of the election’s outcome. Scotland

There is a long history in the United Kingdom of discussions of devolution, independence, and special recognition of the different countries. There have been referenda on the future of Wales, Scotland, and even Northern Ireland (Bogdanor 1981). The most recent secession referendum in Great Britain focused on Scotland. During the 2011 election campaign, the Scottish National Party (SNP) committed to holding a referendum on Scottish secession (Stuart 2011; Qvortrup 2014a). The SNP won 69 of the 129 seats in the Scottish Parliament, giving the party a clear mandate to hold a referendum. In January 2012, the British government offered to legislate powers for the Scottish Parliament to hold a referendum and to set the terms of the referendum, including the question posed, voter eligibility, and who would organize the referendum (Carrell and Watt 2012).8 Negotiations between the two governments continued for 10 months before reaching the Edinburgh Agreement—a document that outlined the referendum process and its legitimacy (BBC 2012a). In early 2013, the Scottish government determined that the vote would occur on September 18, 2014.

The Power of Words: Secession Referenda

153

One key issue in the referendum negotiations was who would be eligible to vote: residents of Scotland (including British Commonwealth and European Union citizens) aged 16 years and older. Voters’ lists included members of the British Armed Forces or Crown personnel who were registered to vote in Scotland, but convicted prisoners were banned from participating. Scottish citizens living abroad were unable to participate, as were other citizens of the United Kingdom who did not reside in Scotland (“Scotland’s Future” 2010; MacDonnell 2011; BBC 2012b). Some discussions took place about the legality of the Scottish Parliament’s referendum (Clegg 2012; Whitaker 2012; Black 2009). When the initial plan for the referendum was developed, it was supposed to be an advisory referendum that would have no legal effect. However, the Edinburgh Agreement allowed for the temporary transfer of power to the Scottish Parliament to hold a secession referendum. The British government stated that, if a simple majority of votes were cast in favor of secession, Scotland would become an independent country after a process of negotiations. Much discussion took place on the wording of the referendum, with the question of “Should Scotland be an independent country?” ultimately being posed to voters (Black 2013). The question was simple and straightforward, leading to no confusion about the purpose and results of the referendum. The issues surrounding the separation of Scotland were far more complex than the question posed, and perhaps further development of the question would have provided clarification of those issues. Other issues that plagued this process included the fate of Scotland’s relationship with Great Britain and the European Union should Scotland secede. These questions ultimately affected the outcome of the election (LeDuc 2015). The referendum was organized and monitored by the Electoral Commission, which regulated the 16-week period preceding the election (Black 2013). The Electoral Commission allowed the campaign organizations to spend 1.5 million pounds each (roughly US $2 million), along with specific amounts from each party and other organizations that registered with the commission (Adams 2014). Individual donations of various amounts were contributed to both sides. Polling indicated that the race became closer as the vote became more imminent. The start of polling demonstrated that the yes campaign had little chance of winning, with more than 60 percent against secession. By election day, this had radically changed: polling showed

154

Direct Democracy: A Double-Edged Sword

46 percent yes and 54 percent no (Gardham, 2014). The voter registration deadline of September 2, 2014, revealed there had been an unprecedented number of new registrations, with some councils receiving tens of thousands of applications in the final weeks of registration. In addition, more than 680,000 voters had registered to vote by mail—20 percent higher than in other elections (Adams 2014). The turnout was extremely high for Scotland, with the overall turnout being 84.6 percent (roughly 20−30 percent higher than most elections in Scotland during the twenty-first century). A total of 55.3 percent of voters rejected the secession referendum, and the four council areas where the majority voted yes contained 20 percent of the Scottish voting population (BBC 2014a). A high turnout in the Scottish referendum added to voter efficacy, as voters believed they had an actual impact on the outcome of the election and the future of Scotland. In the election’s aftermath, many elected officials in the United Kingdom and abroad voiced support for the outcome. Discussion took place on voting irregularities, but the chief counting officer quickly dismissed these concerns. In addition, some violence occurred on September 19, 2014, as the no supporters gathered to celebrate, and this reportedly led to conflict with the secession supporters. Police were able to separate the groups and made several arrests. Both sides condemned further violence in response to the vote. Although the vote was unsuccessful, it has forged greater unity within Scotland and created a higher level of support for parties who wish to secede (Fraser 2015). Although secession was not achieved in the 2014 referendum, it set the stage for future separation with the United Kingdom as well as the notable recognition of Scottish identity. It is too early to determine if there will be a continued separatist movement or if this will lead to special recognition from the United Kingdom, but Britain’s willingness to work with Scotland on this secession referendum demonstrates that there is an opportunity for the recognition of states from the larger United Kingdom. The United Kingdom voted to leave the EU in June 2016. Scotland voted to remain. As a result, there have been renewed calls for a second referendum in Scotland. Officials indicated that this vote was a significant change for Scotland and how it viewed its future in the United Kingdom. Scotland has a distinct identity within the United Kingdom that is often unrecognized. This distinctness was one of the driving forces behind the referendum for Scotland to

The Power of Words: Secession Referenda

155

gain autonomy and recognition. In hindsight of the Brexit vote, Scotland will have to face its future in either the United Kingdom or the European Union. Crimea

Crimea’s secession referendum is one of the most controversial. The 2014 Ukrainian revolution served as the catalyst for this particular referendum. One of the ongoing issues in Crimea was its connection to Russia, with 60.4 percent of the population reported to be ethnic Russians and 24 percent ethnic Ukrainians (Hashem 2014). This ethnic breakdown created issues with Crimea’s relationship with Ukraine as many Crimeans wanted a closer relationship with Russia. 77 percent of the population spoke Russian and identified with the Russian population. Additionally, Crimea has a long history with the Soviet Union. Crimea was transferred to Ukraine in 1954 (Sasse 2014) but became part of Ukraine only with the dissolution of the Soviet Union. Crimea held a referendum in 1991 to gain its autonomy, but the Ukrainian parliament abolished the Crimean constitution in 1992. The new constitution adopted in 1998 gave Crimea far less autonomy, where any legislation could be vetoed by the Ukrainian parliament (Sasse 2014; Catala 2015). In February 2014, after the removal of Ukrainian president Viktor Yanukovych, who was considered pro-Russian, Russian forces invaded major Ukrainian holdings (government buildings, military bases, and telecom facilities) in Crimea. The governing body of Crimea (the Supreme Council of Crimea) held that the removal of President Yanukovych was a coup and that the new government that replaced him was illegitimate. The referendum was seen as a response to this illegitimate government; however, this was compounded by the occupation of Russian troops in Crimea as well as by the resource richness of the Crimean territory. The Supreme Council set a date and ballot question only 10 days prior to the election, and the hurried nature of the referendum raised flags with members of the European Union and the United Nations. Russia vetoed the UN Security Council’s attempt to declare the referendum invalid (Urquhart, Williamson, and Neild 2014). Regardless, the international community did not recognize the referendum as legitimate or the voting options to be appropriate. However, Russia and the Crimean Supreme Council argued that the referendum was legal, citing a precedent in

156

Direct Democracy: A Double-Edged Sword

Kosovo and drawing parallels with the International Court of Justice’s declaration that there was no prohibition against the right to selfdetermination (Putin 2014). Western legal scholars have disputed this argument (Weller 2014). The referendum was a by-product of this invasion, where voters were asked to vote on the future of Crimea (Oliphant 2014; Saideman 2014). It should be noted that the international community and most governments regard Crimea as still being a part of Ukraine and the Russian invasion and occupation as illegal. As such, the referendum held on March 16, 2014, was to determine if Crimea should join Russia or if it should be restored as a part of Ukraine.9 The referendum asked voters about the future of Crimea, but did not allow for the current system to continue. Two choices were available on the ballot, and voters were able to choose only one of these. The choices reflected the following stances (“Official Voting Bulletin” 2014): Choice 1: Do you support the reunification of Crimea with Russia with all of the rights of the federal subject of the Russian Federation?

Choice 2: Do you support the restoration of the Constitution of the Republic of Crimea in 1992 and the status of the Crimea as a part of Ukraine?

The language of the two choices clearly illustrated an expectation, either reunification or restoration. There was no discussion of an independent Crimea, but rather the choice of continuing with the reunification with Russia or going back to Ukraine. These leading questions and a missing third option provided voters with limited choices. Prior to the referendum, it was agreed that only a simple majority was needed to declare a winning outcome. However, the discussion over the questions posed was concerning, as many argued that no status quo option was provided (Orenstein 2014). In addition, an argument was raised about what the second choice really meant and the legal issues associated with it (BBC 2014b). Further, some discussion took place on the legitimacy of the seceding groups’ claim (Catala 2015). The ballot was provided in three languages: Ukrainian, Russian, and Crimean Tatar. Only those voters with Ukrainian passports and Crimean residences were allowed to participate. The threshold

The Power of Words: Secession Referenda

157

number of votes for the results to be accepted was not discussed or agreed upon. There was a very short campaign (10-days), and the majority of the focus was on pro-Russia rather than on the status quo or restoration of the previous constitutions. A large amount of the focus was also placed on the “new government” of Ukraine and what that meant for Crimea. Polling conducted in the 5 years prior to the referendum demonstrated support for autonomy and freedom from Ukraine and Russia (International Republican Institute 2013; Tierney 2015). Polling much closer to the referendum in 2014 illustrated that there was a high level of support for separation from Ukraine and for unifying with Russia (Ignatova 2014). Regardless of the question asked and the options for the future, all polls indicated that the voters wanted to rejoin Russia with a more than 65 percent majority. The estimated results of the referendum was 96.77 percent support for Choice 1 and integration with Russia. In addition, the Supreme Council reported an 83.1 percent turnout (Vasoric and Croft 2014a ). These percentages are unconfirmed but, seemingly, there are consistent polling results with support for reunifying with Russia. Yet there were controversies over the actual turnout. Mejlis deputy chairman Akhtem Chiygoz stated that the turnout could not have exceeded 30−40 percent, which is lower than the reported turnout. In addition, military leaders, such as Igor Strelkov, admitted that the militia coerced Crimean deputies to vote to integrate with Russia (Christakis 2015). The allegations of fraud plagued the results of the election, with discussions about confiscating identification documentation, premarked ballots, and inconsistencies with who may have voted. The international community viewed the referendum and outcomes as illegal and nonbinding. However, following the referendum, on March 18, 2014, Russian and Crimean leaders signed a treaty on the unification of these two states, and the Russian Federal Assembly ratified it on March 21 (Vasoric and Croft. 2014b). Crimea is an obvious example of the state breakdown hypothesis. With Russian influences still evident in Crimea, and with the changing political system in Ukraine, clear evidence existed that the Ukrainian union was on shaky ground. The civil war and invasion of Russia in parts of Ukraine further exacerbated Crimea’s issues. Despite the international community’s clear denial of recognition, Crimea has joined the Russian Federation and has forged a more peaceful situation.

158

Direct Democracy: A Double-Edged Sword

Sudan

Sudan, plagued by civil war for decades, turned to peace agreements and referenda to establish peace and resolution to these conflicts.10 The first civil war in Sudan occurred from 1955 to 1972, focusing on greater autonomy for southern Sudan. The civil war ended as a result of the Addis Ababa Agreement, which gave southern Sudan more self-determination and control over its internal issues (Young 2012; Ahmed 2009). However, in 1983, President Jaafar Nimein introduced sharia law, which led to another outbreak of conflict. It was further exacerbated in 1989 with a military coup and the emergence of General Omar al-Bashir’s presidency. As such, civil war plagued Sudan over issues of resources and southern autonomy. A total of 2.5 million Sudanese died, and 4.0 million were displaced through the conflict (Young 2012). International pressure forced a peace accord in 2005 to end the civil war (Ahmed 2009). In 2011, after decades of conflict, a referendum was held on South Sudan’s secession. This was a complex vote that was part of the Naivasha Agreement, which established peace in 2005 between the central government of Khartoum and the Sudan People’s Liberation Army/Movement (Young 2012; Vidar 2011). As discussed in chapter four, direct democracy provides a way of overcoming deep divisions in society because it is a consensus form of governing. Because Sudan had such cleavages from the civil war, the use of a referendum to create a new state and to effectively end the conflict provided that type of consensus governing. In the negotiations leading up to the referendum, issues about how power and wealth would be divided between the regions plagued the discussions (Ahmed 2009; Young 2012). Additionally, there were issues with taking a census prior to the national election (in 2010) and the referendum. The mechanism of taking a census as the Naivasha Agreement required was a funding and logistical nightmare. There were 5 million nomadic Sudanese, 2 million displaced refugees in camps around Sudan as well as in neighboring Uganda and Kenya, and the areas where the civil war occurred made travel difficult due to unmapped minefields (Høigilt, Falsh, and Rolandsen 2010). There was also a lack of trust on behalf of the Sudanese regarding the motivations behind a census (it could be used for retribution later) and those who feared that peace was fleeting. Rebel groups denounced the census, and many threatened to attack census takers.

The Power of Words: Secession Referenda

159

Furthermore, intense negotiations took place between the central Sudanese government and the South Sudanese government to determine turnout thresholds (a minimum of 60 percent or approximately 2.3 million voters) and what vote share the yes side needed to declare secession (it agreed on a simple majority). There were also questions about what the outcome of citizenship and nationality would be for southern Sudanese (Assal 2011). They also negotiated what would happen if the vote share was successful, but the turnout was low. However, the turnout was above the threshold, and South Sudan declared its secession.11 Ultimately, regardless of the outcome, Sudanese officials stressed the importance of a peaceful transition throughout the campaign. Even Sudanese president al-Bashir said he would respect the outcome of the vote and supported the South’s ability to choose its future (Christopher 2011; Ahmed 2012). Al-Bashir also forewarned that secession did not necessarily mean that all of the South’s problems would be resolved, but that self-determination may assist in solving those issues. Campaigns for both sides of the referendum addressed issues of citizenship, wealth and resource sharing, and power divisions. External forces also played a role in the campaign. Egypt proposed a confederation between the North and South to quell any need for a referendum, but al-Bashir rejected this. In addition, Egyptian and Libyan leaders went to Sudan to try to assuage the conflict and encourage unity (Christopher 2011; Ahmed 2012). The United States also became involved in the election by extending sanctions against Sudan to ensure that the government complied with the referendum process and deadline. The United States suggested that it would drop Sudan from the list of state sponsors of terrorism if the referenda were held on time and the results were recognized as legitimate (Christopher 2011; Ahmed 2012; Vidmar 2011). Polling throughout the campaign demonstrated that a substantial majority of South Sudanese favored independence. All reported polls (conducted by both domestic and foreign entities) indicated more than 90 percent supported secession. This is unsurprising given the issues of southern sovereignty that had plagued the country throughout the civil wars and the peace agreement. The campaign was marred by controversy. There were accusations that government officials on both sides were trying to intimidate the media to prevent unfavorable coverage of the election. There were also discussions about whether everyone would recognize the outcomes of the election. Most of this conjecture was from political

160

Direct Democracy: A Double-Edged Sword

parties arguing that the peace agreement was not being upheld appropriately and that this was leading to malice around the campaign and voting procedures (Young 2012; Christopher 2011). Eventually, both sides of the campaign came together to affirm their commitment to the referendum process and encourage a campaign to increase awareness and turnout for the election. Although this was successful in encouraging awareness and turnout, it had a negative impact as well. Voters were concerned about their uncertain future in a new South Sudan, and many started to move north for protection and security (Ahmed 2012). President Al-Bashir indicated that dual citizenship would not be permitted, and many southern Sudanese would lose their government jobs in Sudan after secession. Registration started 2 months prior to the election, and many efforts were made to encourage the registration of those who had fled South Sudan during the civil war. Four million citizens registered by the deadline, and many others returned to South Sudan even though it was too late to register for the referendum (Christopher 2011; Young 2012; Ahmed 2012). However, the ballot language was problematic; instead of using words to discuss the impact of the independence vote, pictures were also used to illustrate the differences between unity or separation and what those votes meant. These pictures had specific overtones related to the activities of the election. The picture shown in Figure 7.1 is an example of one illustration on the ballot, and it is clear that separation was supposed to mean the opposite of working together. This is coercive in that it implies a meaning behind the vote that is not necessarily consistent with the vote. Voting for separation was indeed voting against unity in the larger Sudan, but it was more about promoting unity in South Sudan to combat some of the issues that had plagued the country during its civil war (Young 2012). Additionally, given the low literacy levels in Sudan, the pictures had a substantial impact on how voters would vote on the proposal. Voting was conducted over 3 days. Polling stations were set up in a variety of locations. Voters in Darfur were offered special voting stations. International polling stations were set up in eight countries with large South Sudanese populations (Australia, Canada, Egypt, Ethiopia, Kenya, Uganda, the United Kingdom, and the United States). The turnout threshold of 60 percent (or 2.3 million voters) was reached. In fact, the voter turnout far exceeded expectations with 83 percent of the eligible voters in the South. The vast majority of voters (98.8 percent) supported secession (Qvortrup 2014b; Southern

The Power of Words: Secession Referenda

161

Figure 7.1 Sudanese Ballot

Sudan Referendum Commission 2011). Despite the coercion on the ballot, there was a clear result from the vote. Sudan is an excellent example of the state breakdown hypothesis. Since the country was emerged in civil war, the means of establishing peace was to create autonomy for southern Sudan. As Sudan went through a period of state breakdown, it appeared that the only way out of the rubble of the civil war was to split the country into two and create a special identity for southern Sudan. Analysis The decolonization hypothesis has support from Table 7.1, where it is apparent that many states gained their independence from colonizers. These states gained their independence, in many cases in mass, during certain time periods. The referenda held under the

162

Direct Democracy: A Double-Edged Sword

decolonization process were generated by both the colonies and the colonizers and created a number of new states. Although many of the referenda in the 1950s by French colonies failed, these countries for the most went on to become independent states. As shown in the next chapter, peace can be forged through a referendum. The state breakdown hypothesis looks at the division of countries after conflicts develop through referenda. States that undergo periods of conflict and are unable to resolve them may turn to referenda to settle their conflicts and dissolve the states. The two cases that illustrate this include Sudan and Crimea. Both of these countries experienced conflict, and the only way to independence was through a referendum to gain support from the people for this division. These are not the only states to undergo conflict and use referenda to establish divisions; Yugoslavia after its civil war chose to use referenda to create a number of new states to achieve peace. There are often sections of a country that claim to be distinctly different from the rest. These distinctions lead to a desire to separate from the larger entity. Scotland and Quebec provide opportunities for evaluating the distinct identity hypothesis. Although these sections of the larger entity of which they are a part argued that they were distinct, both voted to remain part of the larger whole. Thus, their distinctness makes them unique but fails to provide voters with enough of a reason to secede. Ballot language in these four cases is illustrative of a language hypothesis that voters can be confused and manipulated by ballot language. If the information is unclear, voters may not understand the outcome of the election even after a substantial campaign. If issues are still outstanding and this is evident in the ballot question, voters may choose to vote against the measure rather than support something that is unclear, as suggested in the Quebec and Scotland cases. However, options may also be missing in the ballot language, as in the case of Crimea. States can make it easier for voters to participate by providing material on the ballot (e.g., the pictures on the Sudanese ballot). It is important to note that issues of manipulation through ballot language continue to exist, either by obfuscating the true meaning or by using incendiary language. Voters are vulnerable in the voting booth and can be manipulated easily by a question. When the stakes are as high as they are in secession referenda, this can limit the true power of citizens to make decisions about their country’s future.

The Power of Words: Secession Referenda

163

Another powerful statement about secession referenda is the high levels of turnout that all of these cases experienced; all had turnouts above 83 percent. This demonstrates widespread participation in determining the outcome, regardless of what that outcome is. Each side of these referenda had passionate opinions that incited arguments, but this elicited a large turnout to support the process of direct democracy.

Notes 1. There is some conflict in the literature about the use of the term independence or secession referenda. Although there are reasons for using both, for this chapter it seems that secession referenda is the more inclusive term. 2. As of this writing, two referenda are scheduled for 2017: South Ossetia will vote on independence from Georgia or on joining Russia, similar to the Crimean vote; Catalonia will vote on independence from Spain. New Caledonia will vote in 2018 on independence from France, and Bougainville will vote in 2019 on independence from Papua New Guinea. 3. This is particularly true for colonies. In 1958, 20 countries held secession referenda resulting in 19 new independent states. Saar, Cameroon, Central African Republic, Chad, Comoros, Republic of the Congo, Dahomey, Djibouti, French Polynesia, Gabon, Côte d’Ivoire, Madagascar, Mali, Mauritania, New Caledonia, Niger, Saint Pierre and Miquelon, Senegal, and Upper Volta became independent states, whereas Guinea’s referenda did not lead to secession (Kersting 2014). 4. In 1991, nine countries held referenda on secession from the Soviet Union (Armenia, Azerbaijan, Estonia, Georgia, Latvia, Lithuania, Ukraine, Turkmenistan, and Uzbekistan) and five from Yugoslavia. Those declaring secession from the Soviet Union were more successful that those from Yugoslavia mainly due to the circumstances surrounding the secession vote. Two states, Kosovo and Montenegro, had more difficulty in achieving secession and recognition. Many felt that the Soviet Union referendum held in 1991 was manipulative due to the wording used (Hill and White 2014). 5. These types of eligibility questions have been a deciding force in whether a referendum will occur at all in Morocco and Western Sahara (Jensen 2005). 6. Qvortrup (2014b) includes a study by YouGov from 2012 that looks at the wording of the question posed and sees an 8 percent swing in support for the referendum depending on the wording. 7. The BQ won 54 seats and 49.3 percent of the vote from Quebec in 1993. This election made the BQ the second-largest party in the House of Commons and the Official Opposition, giving support and substantial discussion time to the issue of separation in the federal legislature. 8. There are far more details on the process and legality of the Scottish referendum (see Qvortrup 2014b).

164

Direct Democracy: A Double-Edged Sword

9. This was not the first referendum in Ukraine about its sovereignty. In 1990, Ukraine voters were asked whether they should declare themselves as sovereign or maintain their connection to the Soviet Republic (Hill and White 2014). 10. Sudan’s first civil war was from 1955 to 1972; its second civil war started in 1983 and continued until the peace agreement in 2005. The international community was outraged by the atrocities that occurred in Sudan, and the “lost boys of Sudan” became a significant crisis in the late 1990s. Despite military coups and religious turmoil, the democratic nature of the referendum demonstrates a new future for the countries. 11. At the same time as the separation referendum, there was supposed to be a referendum in Abyei on whether to become a part of South Sudan. However, discussions about the legality of the process ultimately led to a postponement of this referendum. Additionally, there was a concern about moving forward with Abyei without having settled the decision to create two separate entities: Sudan and South Sudan. Many felt that, if Abyei’s determination was made at the same time or if there was an annexation of Abyei to the South, there would be an immediate return to war. A nonbinding and unrecognized referendum was held in 2013, and the Abyei people voted 99.9 percent to join South Sudan. However, this referendum remains unrecognized and the issue of Abyei unresolved. There are additional concerns about the future of popular consultations on South Kordofan and the Blue Nile.

8 The Power of Voters: Creating and Rejecting Peace

direct democracy can be made. One of the relatively recent uses of direct democracy has been in the peace process. This is the ultimate way to get voters to engage in the peace process and to create a mandate to establish peace. Despite this recent foray, few scholars have focused on the peace referendum process (Loizides 2014; Johansson 2009; Kersting 2009, 2010; Lee and MacGinty 2012). The use of referenda in the peace process offers the opportunity to create stability and legitimacy for the process; it also gives the people a voice in overcoming divisions in their society (Loizides 2014). Peace referenda have been held several times and have been debated in other instances such as the Palestinian-Israeli conflict (Loizides 2014), Sri Lanka (Edrisinha 1998), Nagorno-Karabakh (Johansson 2009), and Kashmir (Bose 2007). These referenda come with some risks. Although they may be able to forge lasting peace, they open the process to additional conflict and potential failure. Should the referendum fail, it is difficult to foster a long-term peace agreement with either side in the future. Further, a referendum’s failure can inflame the conflict, as it did in East Timor in 1999 (Paris 2004). As such, many international organizations have become involved in the process as a way to foster the peace process and promote the referendum. The United Nations has As shown in previous chapters, there are a number of ways that

165

166

Direct Democracy: A Double-Edged Sword

taken a strong role in fostering peace and was extensively involved in Cyprus’s and East Timor’s referenda. The UN’s peacekeeping nature works naturally with the referendum process in many of these countries. Supranational governments such as the EU and international organizations (e.g., the African Union, North Atlantic Treaty Organization, and Organization of American States) have the ability to influence the referendum process to establish peace (Paris 2004). These organizations can also delay and manipulate the process. The UN was blamed for the chaos surrounding the East Timor referendum because many felt that the vote was premature (Loizides 2014).1 Additionally, the EU’s admittance of Cyprus during the peace process undermined the referendum process. Thus, international organizations can also exacerbate the problems without appropriate action. Theoretical Argument What circumstances make it possible for peace referenda to forge lasting peace in a country? Various actors, situations, and histories influence the peace process in each country. Looking across these instances, it is possible to build theoretical frameworks to explain peacebuilding through direct democracy. International organizations often become involved in the peace process to facilitate and serve as mediators. When international organizations facilitate a referendum, it is more likely to pass for multifaceted reasons. First, international organizations can provide states with incentives to hold referenda and complete the peace process. These organizations may exert pressure through trade or embargoes. International organizations can also have unique relationships with both sides of the conflict to assist in reaching a compromise. Furthermore, these organizations may act as neutral third parties to encourage a referendum and campaign for its success (Paris 2004). International organizations hypothesis: if international organizations pressure states to hold a referendum, it is more likely to pass.

Negotiation is key to getting citizens and members of the conflict to consent to the process (Saunders 1999, 1985). If the different

The Power of Voters: Creating and Rejecting Peace

167

sides of the conflict are involved in the negotiation of terms, a referendum will be more likely to succeed. However, if either side feels that the process is not meeting its needs or that the settlement is one-sided, the referendum will not pass. Voters on both sides of the conflict need to believe they are gaining from the settlement, either because of the benefits to their side or because the other side receives the appropriate repercussions. As such, voters can look for specific attributes, such as land benefits or punishment of conflict participants (rather than amnesty), when evaluating the sides of the referendum rather than looking at the larger and long-term goals. Involvement hypothesis: if the two sides of the conflict are part of negotiating the referendum’s terms, the referendum is more likely to pass.

When there is high voter participation, more voters support the process. Thus, regardless of whether the referendum passes or fails, voter support of the referendum’s democratic process means there is likely to be peace, even if it is not established through the referendum’s outcome. High support for the process (as measured by turnout) leads to high support for peace, even if the specific circumstances are not agreed upon. Participation hypothesis: if there is widespread participation, the referendum is more likely to be successful in building peace (regardless of the outcome).

As voters utilize democratic means to determine a peace settlement, they are already buying into the peace process and are more likely to continue to support the referendum outcomes (Loizides 2014). The democratic process seemingly allows for multiple interests to be heard and exercised as well as power sharing between varied interests. This does not mean that there will not be some conflict in the peace process, but overall it will lead to a long-lasting peace. Stability comes from stakeholders determining that the value of the process and its outcomes are more important than their independent goals. If this is achieved through democratic means where the voters determine the outcome, it will have longer sustainability. Stability hypothesis: if the referendum passes, it is more likely to lead to stable peace.

168

Direct Democracy: A Double-Edged Sword

Case Studies I utilize four case studies from Northern Ireland, Cyprus, South Africa, and Colombia to investigate the impact and process of peace referenda. These cases represent different regions and different experiences with peace referenda. They also provide varied outcomes: two of the referenda passed (Northern Ireland and South Africa) and two failed (Cyprus and Colombia). All of these cases involved internal clashes spanning more than 30 years, and other processes were attempted prior to the referenda to obtain peace. Despite these experiences and electoral outcomes, all have achieved some form of détente in the wake of their referenda. Northern Ireland

Northern Ireland’s struggle with peace stems from its relationships with the United Kingdom and Ireland. The conflict grew in seriousness during the late 1960s with more than 3,000 people killed and more than 50,000 injured. At the heart of this conflict, often referred to as the “Troubles,” was Northern Ireland’s independence and creation of a unified Ireland. Substantial violence, bombings, and armed conflict between the government and Irish republicans occurred. Due to the increased violence and conflict between these groups, as well as political gridlock, the autonomous regional government for Northern Ireland was suspended in 1972 (Somerville and Kirby 2012; Acheson and Milofsky 2008). In 1973, Northern Ireland held a referendum regarding whether it should keep its status within the United Kingdom or create a united Ireland (Acheson and Milofsky 2008). While many of the political parties supported the option to remain in the United Kingdom, the Social Democratic and Labour Party called on its members to boycott the election and reject the decision by the British government to hold this referendum (Somerville and Kirby 2012). During the violence leading up to the vote, a British soldier was shot and the Provisional Irish Republican Army planted four car bombs in London to protest the referendum (200 were injured in these bombings). Voters were asked, “Do you want Northern Ireland to be joined with the Republic of Ireland outside the United Kingdom?” Voter turnout was marginal at 58.6 percent (1.03 million voters participated), and only 1.0 percent of Catholics participated; 98.9 percent voted to remain in the

The Power of Voters: Creating and Rejecting Peace

169

United Kingdom, and 1.1 percent voted to join the Republic of Ireland (Somerville and Kirby 2012; Hayes, McAllister, and Dowds 2005; Hayes and McAllister 2001). Despite this vote, violence continued in Northern Ireland, straining its relationship with the United Kingdom and the Republic of Ireland. The peace process had support from paramilitary organizations to reform the policy and withdraw army troops. This was forged under the Belfast Agreement (also known as the Good Friday Agreement) in 1998 (Somerville and Kirby 2012). A main consideration of this agreement was the acknowledgment that Northern Ireland will remain part of the United Kingdom until voters in Northern Ireland vote otherwise in a subsequent referendum (Hayes and McAllister 2001; Hancock 2011). Further, the Constitution of Ireland was amended to focus solely on the Irish Free State. Additionally, the British government recognized that the Republic of Ireland and Northern Ireland had the right to resolve their own issues. There were also political concessions by the unionist and nationalist parties to create power sharing among the different stakeholders (Hayes and McAllister 2001; Hancock 2011). Voters were asked to vote on the Good Friday Agreement referendum on May 22, 1998: “Do you support the agreement reached at the multiparty talks on Northern Ireland and set out in Command Paper 3883?” This agreement was supported by 71.1 percent of the vote with a turnout of 81 percent (1.175 million), much higher than the 1973 referendum. A simultaneous referendum was held in the Republic of Ireland, and 94.4 percent of Irish voters supported the measure (Hayes and McAllister 2001). It took until 2003 for all of the Belfast Agreement’s issues to be realized and for the final decommissioning of the Provisional Irish Republican Army’s arsenal (Acheson and Milofsky 2008). However, the impact of the referendum in establishing peace was critical to the success of the peace agreement. Having simultaneous referenda in Northern Ireland and the Republic of Ireland gave it strong and unified support. While there was an international outcry regarding the violence in Northern Ireland, sanctions and condemnation did not appear as in other conflicts. An Independent International Commission on Decommissioning was involved in the settlement, but it had not been part of the original process. The high participation in both Northern Ireland and Ireland led to increased support for the solution. The referendum ultimately put a final stamp on the peace settlement,

170

Direct Democracy: A Double-Edged Sword

giving voters the opportunity to approve the plan. This led to longterm peace in Northern Ireland. South Africa

Between 1948 and 1991, South Africa employed a system of racial segregation and discrimination called apartheid (Thompson 2001). Colonial roots created the separation of races and white superiority, but the system was extended into law only in 1948. The first laws of apartheid focused on mixed marriages and sexual relationships across racial lines. The system developed further to include classifying neighborhoods by race, which led to displacement of 3.5 million nonwhite South Africans from their homes. The system of apartheid garnered severe international and domestic opposition among minority populations. Other foreign governments and the UN condemned the action and enacted embargoes against South Africa. Domestic opposition resulted in brutal responses from the ruling National Party administration, resulting in death and imprisonment for thousands of nonwhite South Africans. Between 1987 and 1993, the National Party negotiated with the African National Congress (ANC), the party of Nelson Mandela, to end apartheid and introduce majority rule (Thompson 2001). The result of these negotiations was the South African apartheid referendum in 1992. President F. W. de Klerk (Frederik Willem de Klerk) called for a referendum regarding the end of apartheid (Strauss 1993; Kersting 2010). This followed 2 years of negotiations between the government and the ANC, but progress was slow. Violence increased in South African townships, and there was great dissatisfaction within the white community. The pressure to negotiate successfully was intense. De Klerk added that, should the referendum fail, he would resign and a general election would be held (Strauss 1993). The referendum was limited to white South African voters, who were asked: “Do you support continuation of the reform process which the State President (F. W. de Klerk) began on February 2, 1990, and which is aimed at a new Constitution through negotiation?” (Strauss 1993; Kersting 2010). This referendum proposed to end the apartheid system. The process that de Klerk started in February 1990 allowed for releasing Mandela (after 27 years in prison), suspending capital punishment, and lifting the state of emergency.

The Power of Voters: Creating and Rejecting Peace

171

De Klerk said that the time for negotiation had arrived, and that changes needed to be made. There was substantial criticism of de Klerk because only white voters were allowed to participate in the referendum; however, he remained steadfast in his decision. He argued that the government was not just handing power to the ANC, and that this would be a step toward power sharing with other races (Thompson 2001; Southall 1994). Further, he warned that international sanctions and civil conflict would be powerful if voters fought against the referendum. He was also clear about what a yes vote would mean—a mandate to negotiate with black leaders for constitutional changes (Strauss 1993). The issues that de Klerk addressed were echoed in the rhetoric surrounding the campaign. Those on the yes side focused on the peace aspect while those on the no side focused on communist and black rule (Kersting 2010). Since the government was clearly on the yes side, they had a substantial advantage. Media in much of the country was government controlled and discounts were given to the yes campaign, whereas the no campaign had limited funds and difficulty getting its message across (Strauss 1993). There were significant concerns that a no vote would lead to additional violence and international condemnation and sanctions. The ANC was not supportive of the measure because only whites could participate in the election, but it moderated its tone to secure a yes vote from white South Africans (D. Butler and Ranney 1994). In the end, 85.0 percent of voters (2.8 million) turned out to vote, with 68.7 percent voting in support of the referendum (Strauss 1993). Political leaders of the National Party and the ANC were thrilled with the results. Of course, those against the referendum argued electoral fraud, but no evidence was provided regarding irregularities. In April 1994, South Africa held its first multiracial elections. This resulted in a huge majority for the ANC, and Mandela became the first black president of South Africa (Southall 1994). The road to peace and ending apartheid was bumpy, but the lasting effects of the 1992 referendum on racial desegregation and multirace elections is undeniable. Support for the referendum that involved different stakeholders and the international community combined to lead to a successful referendum as well as lasting peace and cooperation in South Africa. The high turnout demonstrates a commitment by white voters to the process and the new status quo in race relations in South Africa.

172

Direct Democracy: A Double-Edged Sword

Cyprus

The Republic of Cyprus’ disputed northern territory has been occupied by Turkish troops since 1974. The island has experienced numerous conflicts with the United Kingdom, Turkey, and Greece (Broome 2002; Pericleous 2009). Initially, conflict on the island focused on Cyprus’s people, who wanted self-determination, and the British. However, this shifted from a colonial issue to an ethnic issue between the Turkish and Greek islanders. An informal referendum regarding Cyprus’s relationship with Greece was held in 1950. Only Greek Cypriots could vote, and there was substantial support from those who participated (95.7 percent support from only 224,747 voters). Unsurprisingly, Greek Cypriots supported a continued relationship with Greece. This setup increased conflict between Greeks and Turks in Cyprus. In 1974, Cyprus experienced a coup, prompting Turkey to invade and occupy the northern part (Broome 2002). This was followed by a declaration by the Turkish Cypriots to form the Turkish Republic of Northern Cyprus (TRNC) in 1983. The international community (particularly the UN) did not recognize the TRNC, but Turkey created a strong relationship with this new entity. Given the dispute over the territory, the UN created a buffer zone between the TRNC and the Republic of Cyprus to prevent further tension or violence. This division allowed Turkish Cypriots to keep 29 percent of the island and required all foreign troops outside of the UN peacekeepers to leave the island. In 1985, talks were held to finalize the settlement; however, Greek Cypriot leaders wanted more negotiations and this led to the collapse of peace talks. As a result, Greek Cypriot leaders, who had been supported by the international community, faced significant criticism for their actions. Talks began again in 1986 to create a bicommunal and bizonal state, but issues of Turkish forces, repatriation of Turkish settlers, and a confederal state ultimately led to disagreement too large to overcome (Pericleous 2009). During peace talks in 1988, the Greek Cypriots announced their intention to apply for membership to the European Economic Community. This move was highly contested by Turkish Cypriots and Turkey. Cyprus formally applied in 1990 and, later that year, European Community member states supported the country’s application and forwarded it to the European Commission for formal consideration. Turkey and the TRNC, which were upset by this move, created

The Power of Voters: Creating and Rejecting Peace

173

a union between themselves just a few weeks later. Javier Pérez de Cuéllar, who had been working on a peace deal between these two communities, was unable to make progress under these circumstances and noted a desire for the independence of both entities. The 1990s saw specific plans proposed by UN Secretary-General Boutros Boutros-Ghali as ways to forge peace in the country. However, he was criticized for exceeding his authority, and both sides felt the proposals did little to protect their interests or recognize their plights (Chadjipadelis and Andreadis 2007). In the meantime, the European Commission endorsed Cyprus’s application for membership; however, given the circumstances, the EU delayed negotiations for entrance. In 1994, the European Court of Justice created an injunction preventing the export of goods from the TRNC into the EU, damaging relations between the EU and Turkey. This was further exacerbated in December 1996,when the European Court of Human Rights declared Turkey to be an occupying force in Cyprus and required it to pay compensation to Cyprus, which had lost property to Turkey.2 It seemed that, despite years of talks, a settlement was not on the horizon. In August 1996, Greek Cypriot refugees demonstrated against the Turkish occupation. During these protests, civilians and British soldiers were shot by Turkish troops, creating an international incident and amplifying pressure from the international community on Turkey (Broome 2002; Chadjipadelis and Andreadis 2007; Pericleous 2009). In early 1997, Greek Cypriots indicated that they planned to purchase an antiaircraft missile system, escalating the militarization of this conflict. Although this plan was quashed by Greece in exchange for other military armaments, it demonstrated the country’s desire to arm itself against future aggressions. Despite these setbacks, the true impetus for peace and a settlement came from the EU’s decision to open accession negotiations with the Republic of Cyprus. The EU also announced in 1999 that Turkey would be a candidate for membership. The EU could have used its power to enforce a peace settlement; however, it chose to admit Cyprus before the referendum was held (Richmond 2006; Rumelili 2007; Tocci 2007). These two activities forced both sides to participate in talks, as the importance of EU membership was crucial to each country. Leaders from the Republic of Cyprus and the TRNC met to discuss a new peace process in 2002. Despite efforts to reach an agreement, talks soon became deadlocked. UN Secretary-General Kofi Annan visited with the leaders many times to work on a peace deal (Chadjipadelis

174

Direct Democracy: A Double-Edged Sword

and Andreadis 2007; Pericleous 2009). Though the leaders were unable to come to an agreement, the UN Security Council asked Annan to create a peace plan to present to both sides. This plan went through multiple iterations, and discussions were halted numerous times. Eventually, a plan was presented in March 2004 that laid out a loose federation of the Turkish Cypriot state (28.5 percent of the island) and Greek Cypriot state (which held the remaining 71.5 percent); additionally, each would have its own parliament. The federal parliament would have 25 percent of the seats set aside for Turkish Cypriots, the senate would be composed of equal members of each ethnic group, and executive power would be exercised in a presidential council whose chairmanship would rotate between the communities (Chadjipadelis and Andreadis 2007; Pericleous 2009). Veto rights regarding legislation were given to both communities. The most controversial component of the plan was land and how ownership of property that had been taken during the Turkish military intervention of 1974 would be distributed or compensated for. Additional issues included the reduction of Greek and Turkish troops on the island. To decide if the plan would be enacted, the two communities held simultaneous referenda on April 24, 2004. Greek Cypriot leaders urged voters to vote against the referendum while Turkish Cypriot leaders urged support for the referendum. Voters were asked, “Do you approve the Foundation Agreement with all its Annexes, as well as the constitution of the Greek Cypriot/Turkish Cypriot State and the provisions as to the laws to be in force, to bring into being a new state of affairs in which Cyprus joins the European Union united?” (Chadjipadelis and Andreadis 2007; Pericleous 2009). The results of the referendum were mixed (66 percent against; 33 percent in favor), with heavy support from the Turkish Cypriots (65 percent) and a resounding no from 75 percent of Greek Cypriots. There was a high turnout among both communities: 89 percent of Greek Cypriots and 87 percent of Turkish Cypriots (see Table 8.1).3 Because the Greek Cypriot community did not support the plan and implementation required support from both communities, the plan was declared null and void (Anastasiou 2007: Sitas, Latif and Loizou 2007). Despite this failure, the EU moved ahead with its accession plan, and Cyprus joined the EU on May 1, 2004. The Turkish Cypriot community was given observer status as part of the Cyprus delegation.

The Power of Voters: Creating and Rejecting Peace

175

Table 8.1 The Cyprus Vote Community

Turkish Cypriots Greek Cypriots

Yes Votes 64.9% 24.2%

No Votes 35.1% 75.8%

Turnout

87% (65,200) 88% (413,680)

Despite the failure of the 2004 referendum, the 2008 Cyprus elections focused on reunification (Anastasiou 2007). The election of Dimitris Christofias as president led to additional rounds of peace talks between 2008 and 2012. The UN tried to intervene to move the talks forward, but they failed yet again. In 2014, talks resumed of cordial relations between the two communities and, while no deal has as yet emerged, hope continues that a formal peace deal is on the horizon. The intervention of the UN and the EU has further complicated the process. While armed conflict does not appear to be a current part of the Cyprus discord, the entire conflict could have been resolved if the EU had waited until after the referendum to allow Cyprus to join. The high turnout demonstrates high interest in the process. Even though the vote was against the peace agreement, the conflict has deescalated. Thus, the referendum did forge a lasting armistice in the region (Coufoudakis 2004; Sitas, Latif and Loizou 2007). Colombia

Since the 1960s, Colombia has experienced conflict with the government, paramilitary groups, crime syndicates, and guerrilla groups fighting over territory and influence (Solimano 2000). Though each group had its own reasons for fighting, all were ultimately accused of human rights violations. During the conflict, more than 220,000 Colombians died, most of them civilians, and many others were displaced. Seventeen percent of the Colombian population were direct victims of the conflict (including many children), making it an important issue for Colombian politics and the country’s future (Angelo 2017). As discussed in Chapter 4, members of civil society worked together to propose an initiative in 1997 to start peace discussions. Peace negotiations began in 2012 and experienced many

176

Direct Democracy: A Double-Edged Sword

setbacks along the way. In 2016, the Colombian government and the Revolutionary Armed Forces of Colombia (FARC) rebels signed a cease-fire deal. Peace negotiations with the different groups had mixed results. From 1990 to 1991 during the 19th of April Movement (M-19), the Popular Liberation Army and Movimiento Armado Quintin Lame groups agreed to demobilize and disarm in exchange for amnesty for their members. FARC continued to resist peace talks, and a demilitarized zone was created between FARC and the Colombian government (Ainley 2016; Solimano 2000).4 Citizens objected to the government’s negotiations with FARC, but without FARC’s involvement in the peace process, talks were stymied. Changes in leadership within the Colombian government had no effect on the success of peace talks, despite political promises made during campaigns. The most successful peace talks with FARC started in the spring of 2011, when exploratory secret meetings were held between the Colombian government and FARC. Negotiators for each side were selected, and international observers from Cuba, Norway, Venezuela, and Chile were chosen to assist in the conflict resolution. The first major step was FARC announcing the end of extortive kidnappings in 2012 as a gesture of peace. While this was a major step forward, it faced substantial setbacks in 2013 when two police officers were kidnapped and four others were ambushed and killed by FARC (Gill 2017). FARC argued these were in retaliation for military bombings that had left 20 of its people dead, but the effect on the peace talks was chilling as it undermined the preliminary steps toward peace. In 2014, accusations that the government was monitoring private communications of the negotiators further jeopardized the talks, but an agreement was reached regarding illicit drugs despite these issues. Furthermore, the 2014 election focused drastically on the peace process. President Juan Manuel Santos won reelection; however, his opposition focused on the problems and terms of the peace process, discussing issues such as terrorism, drugs, and even mandatory jail time for FARC fighters and sympathizers. The margin of victory was thin, as the opposition’s message resonated with many Colombians who had suffered under years of FARC aggression. Despite this rhetoric, peace discussions continued. During negotiations in 2015, FARC revealed the location of landmines and suspended the practice of planting new ones (Angelo 2017). However, in April 2015, 11 sol-

The Power of Voters: Creating and Rejecting Peace

177

diers were killed by FARC in an ambush. FARC defended its actions, claiming that the army had been advancing on its camps despite the bilateral cease-fire. This substantially impacted public opinion regarding the peace process and hurt President Santos’s approval ratings. In May, 27 FARC rebels were killed by a military operation. FARC ended the cease-fire, and the talks lost momentum and credibility (Angelo 2017). To restart the discussion, the government replaced its negotiators and made a small peace offering by returning the bodies of the FARC rebels killed in the May 2015 raid to their families. FARC conducted several attacks in June, hoping to pressure the government into negotiations; however, this backfired and caused lower public support for FARC (Angelo 2017; Ainley 2016). The international observers increased the pressure on both sides to return to the negotiations and de-escalate the conflict. Later in June, the government and FARC agreed to a temporary cease-fire and de-escalation of military activities and FARC offensives. The Colombian peace referendum to ratify an agreement between the Colombian government and FARC guerrillas regarding the end of the Colombian conflict was held on October 2, 2016 (Angelo 2017). The ballot asked voters a single question to approve or reject the signed peace agreements: ¿Apoya el acuerdo final para terminación del conflicto y construcción de una paz estable y duradera?

Do you support the final agreement to end the conflict and build a stable and lasting peace?

The question was fairly straightforward, yet there were hints of manipulation in its phrasing. This referendum had the peaceful future of Colombia at stake, but it failed with 50.2 percent voting against and 49.8 percent voting in favor; 13 million voters participated, roughly a 37 percent turnout (Gill 2017). The outcome was not about the desire for peace, but rather the animosity of the people toward FARC and the decades of injury it had inflicted. The peace agreement called for amnesty and limited punishment of FARC members. Despite the referendum’s failure, President Santos, who was instrumental in the peace process, was awarded the Nobel Peace Prize for his efforts. The Colombian government and FARC signed

178

Direct Democracy: A Double-Edged Sword

a revised peace deal in November 2016. This revision included reparations for victims from FARC’s assets, transitional areas for the rehabilitation of FARC members, ways that members with minor offenses could apply to have their records cleared, and the ability of FARC to create a political party. Rather than going to the people for another vote to support the proposal, the government submitted it to Congress for approval (Gill 2017). It was approved in the House and the Senate and enacted later that year. The 2016 referendum demonstrates the underlying issues with direct democracy in Colombia and how it is used to manipulate voters (VidartDelgado 2016). The low participation in the referendum challenged the government’s mandate to assent to the peace agreement. It did, however, send a message to the government and led to revisions of the final agreement. This revised agreement is still fresh, but the possibility for lasting peace exists in Colombia because of it. International organizations were active in the process by hosting peace discussions and by brokering different aspects of the agreement. Analysis There are many paths to peace, and using a referendum to give the people a voice certainly adds to their power in the process. These cases demonstrate the degrees of success in how referenda impact the peace process. Looking at the Northern Ireland case, the country was able to establish long-lasting peace by holding simultaneous referenda in the Republic of Ireland and Northern Ireland to create trust and get both states to sign the agreement. South Africa used a referendum to allow white voters to endorse a plan to end apartheid and to change the course of history in their country. This referendum brought peace and unity to a country that had experienced 60 years of human rights abuses and forced segregation. Simultaneous referenda have the potential to backfire and not lead to peace. The cleavages in Cyprus were too substantial to overcome with a referendum in 2004, despite UN involvement. EU membership has created a working relationship between Turkish and Greek Cypriots to make peace a possibility in the future. However, the EU could have required support for the referendum prior to accession to incentivize Greek Cypriots to accept the deal. By including Cyprus prior to the

The Power of Voters: Creating and Rejecting Peace

179

referendum, there was no reason for Greek Cypriots to accept the deal. This demonstrates the benefit and the weakness of international actors becoming involved in the peace process, as I argued in the international organization hypothesis. The levels of participation in all peace referenda provide moderate support for the participation hypothesis. In Northern Ireland, South Africa, and Cyprus, voter turnout was greater than 80 percent. The Colombian turnout was substantially lower at 37 percent; many disagreed with the fundamental aspects of the peace agreement. Thus, while voters may have wanted peace, they disagreed with the circumstances and could not vote either way. Another aspect of turnout and participation was who was allowed to participate in the process. In the first Cyprus referendum and in the South African referendum, participation was limited to one side. In Cyprus, only Greek Cypriots could participate, and in South Africa only white voters could participate. This limitation on participation also challenges the legitimacy of the results. Finally, Colombia initiated the peace process in 1997 with a citizens’ initiative to put pressure on the guerrillas. This was a progressive step toward peace and resulted in a comprehensive deal in 2016. Though the 2016 deal had many facets that proved too much for voters, the vote was very close and the government saw it as a partial mandate to force change to the agreement. The revised peace deal went to the legislature and was approved, creating peace in the region. This support was also evident in Cyprus. Voters ultimately voted against the peace agreement proposed in the Annan plan; however, there has been limited violence in the country since the referendum. This provides support for the stability hypothesis. The violence that occurred in many of these cases was not the outcome of the referenda, but rather the secession referenda were a reflection of the deep fractions in society. As Qvortrup (2014b) points out, peace can result only when both parties agree on the legitimacy of the referendum outcome and when that outcome is accepted by the international community. The varied approaches to referenda demonstrated in these cases illustrate that referenda are not a guarantee to peace; however, even when they fail (as they did in Cyprus and Colombia), they give the people power to determine their own fate in a conflict. These referenda play an important role in forging peace and giving the government a mandate to negotiate future settlements.

180

Direct Democracy: A Double-Edged Sword

Notes 1. Research is mixed on whether referenda should be used to solve ethnonational conflicts. Bogdanor (1996) blatantly posited that there are too many issues about difference-managing and race conflicts such as those in East Timor that are beyond a simple up or down vote of a referendum. Thus, they should not be used for that type of conflict to instill peace. Others, such as Qvortrup (2014), laud the ability of voters to steer the course in ethnonational referenda to prevent additional violence. 2. See Judgement of 19 December, Loizidou v. Turkey (1996), not published, EU: 15318/89. 3. Total turnout of 88 percent. 4. This demilitarized zone ended up backfiring on the Colombian government because FARC used it to keep hostages, negotiate prisoner exchanges, train troops, and plan offensive actions.

9 What’s Next?

racy worldwide. Despite these issues, citizen power is the underpinning of direct democracy mechanisms. Balancing this power with the operation of government is a difficult task, and each state has chosen different ways to achieve this balance—either by making the process easy to attempt but difficult to execute, or by employing strategies throughout the process to prevent complete use of these mechanisms. I have provided evidence throughout the book regarding how these mechanisms interact with varying types of democracy. In cases where democracy is solidified, there appears to be less interference with the process (e.g., Canada, Switzerland, and Germany), and this is also true of places where democracy is deepening. In countries that are still trying to figure out the democratic process, there are opportunities for the government to interfere (either purposefully or inadvertently) and influence the process and outcomes. Although we can’t know those governments’ intentions, there is evidence that governments manipulate the process. There is also evidence that governments use direct democracy to intentionally control the process (e.g., in Venezuela, Uganda, and the Philippines) to ensure they get the desired results. Thus, there is potential for these mechanisms to be truly a function and power of the people; however, they Power and manipulation are evident in the use of direct democ-

181

182

Direct Democracy: A Double-Edged Sword

can also be manipulated to erode the power of the people and even decrease efficacy in the public sphere. The use of each type of direct democracy mechanism has proven challenging around the world. There are vastly different rules and regulations regarding these mechanisms, and it often depends on the country to determine how open the system is and why that country utilizes each mechanism. The chapters of this book demonstrate the elations and pitfalls of each mechanism and how it is used. Direct democracy mechanisms give voters a voice in the process, but how loud that voice is depends on the system. Some systems laud citizen involvement through these mechanisms, others notably prevent that voice from having too much influence. As noted in Chapter 1, there are anxieties regarding the impact of direct democracy mechanisms from uninformed voters. Further, depending on the process, these voices can ultimately be ignored (e.g., Colombia or New Zealand). Accordingly, this apprehension can be addressed via how the mechanisms are designed. Direct democracy mechanisms are used at different levels around the world. The manipulation and power of the people is often dependent on the level at which the mechanism is used. National referenda have higher stakes that can have substantial repercussions such as the breakup of a country, the retention of a leader, or even the reorganization of government. State- and local-level uses of direct democracy have less at stake and are often less susceptible to problems, with the exception of recall elections that face even firmer persuasion at the lower levels of government. The structure of government in a state can also influence direct democracy use. The choice of unitary, federal, and confederal systems has an impact on how these mechanisms are used and the effect of manipulation on the system. Unitary systems, where the power is concentrated, mean there is a need to protect that concentration of power vis-à-vis the direct democracy process. In fact, use of direct democracy in these systems may be a way to appear democratic or to strengthen a country’s legitimacy, whereas at the confederal level, where power is decentralized, direct democracy can be utilized to build national identity and to forge a common policy plan for government. In addition to the confederal issues, states with fragmented societies may rely on direct democracy to assert a national distinctiveness, even when that may be lacking beyond the government’s con-

What’s Next?

183

trol. The use in developing societies where there are deep divisions illustrates a way for citizens to come together on specific topics to change policy. While it can also lead to further fragmentation, the evidence I have provided in this book illustrates that direct democracy can be a source of hope. Although the use of direct democracy to develop national cohesion is not limited to those without a strong democracy, the Swiss—who are the most enthusiastic supporters of direct democracy—use direct democracy to address deep cleavages in the Swiss population. It is important to note there is evidence that direct democracy can be used both to protect minority groups (where the population votes down measures that will continue discrimination of minority groups) and to punish them. This illustrates how direct democracy is a powerful tool that can help or harm citizens, so it is critical to use these mechanisms with caution. Authoritarian regimes utilize these mechanisms to forge national unity, develop legitimacy for their regime, and insulate themselves from external pressures to reform. Direct democracy also provides the opportunity for citizens to feel included in government, even if they are not. States break down or experience conflict for a number of reasons. The role of direct democracy in the process of creating peace and stability in these countries provides citizens with a direct impact on the process and a voice in the outcome. There are ways that these processes can ultimately be manipulated (campaigns, international actors, or even fear mongering). However, giving citizens a voice in the process enables them to buy into the process. The Double-Edged Sword For new direct democracy users, these mechanisms have not lived up to their promise. In Chapter 1, I discussed several recent uses of direct democracy that failed or resulted in democratic decay—for example, the development of a military state in Thailand through direct democracy and the rejection of a peace agreement in Colombia. These mechanisms suggest the promise of a more involved public and a more accountable government, yet that often is not the case. The power of the people can be limited when the process is abused and manipulated by governmental regimes. Therefore, the promise and the reality of these mechanisms are not always in sync.

184

Direct Democracy: A Double-Edged Sword

A question that has permeated this research is: When is a referendum appropriate or necessary? Governments have a responsibility to their citizens to govern and involve the citizenry in important discussions. But there are serious difficulties in involving citizens in frequent questions of governance or in the more nuanced issues of conflict and peace. So while there is a desire to create a set of formulaic descriptions about when and how a referendum should be used, that desire is inherently problematic and untenable because there are so many caveats that apply to each use of these mechanisms. It is not only the responsibility of governments to make these mechanisms available, but also to monitor their usage without political interference or agenda. To that end, scholars cannot presume to create an answer as to when referenda should be used. In the second half of the book, I examined the realities of direct democracy usage across multiple levels of government—in authoritarian governments, in secession, and in establishing peace. Voters’ experiences in these different realities illustrate that direct democracy has real and significant impacts on the structure of governments and on the future of a given country. Sometimes, voters fall victim to the pitfalls of direct democracy and can be manipulated into thinking they have a larger voice in their government than they really do. Further, voters can make decisions on secession and peacemaking that have lasting effects on their country without understanding the true impact of their votes. In addition, personal issues—such as the desire for vengeance (Colombia) or land ownership (Cyprus)—can impact voters’ decisions that are not in the best interest of their country. These election outcomes provide evidence for the arguments for and against direct democracy, illustrating that direct democracy truly can be a double-edged sword.

Bibliography

Abah, Hope. 2014a. “Nigeria: Benue Lawmaker Faces Recall over Poor Performance.” Daily Trust, January 1. ———. 2014b. “Nigeria: INEC Postpones Referendum on Recall of Benue Lawmaker.” Daily Trust, July 22. Abazov, Rafis. 2008. “Independent Uzbekistan.” In The Palgrave Concise Historical Atlas of Central Asia. New York: Palgrave Macmillan, 100−101. Abdurasulov, Abdujalil. 2007. “Uzbekistan: The Legitimacy Thing.” Transitions Online 2, no. 20. http://www.tol.org/client/article/18308-the-legitimacy-thing .html Acheson, Nicholas, and Carl Milofsky. 2008. “Peace Building and Participation in Northern Ireland: Local Social Movements and the Policy Process Since the ‘Good Friday’ Agreement.” Ethnopolitics 7, no. 1: 63−80. Adams, Lucy. 2014. “Scottish Independence: Questions Raised over Campaign Spending Rules.” BBC, April 30, 2014. Adedeji, Adebayo and Onigu Otite. 1997. Nigeria: Renewal from the Roots? The Struggle for Democratic Development. London: Zed Books. Adegbite, Kehinde. 2015. “ABC of Recalling a Law-Maker in Nigeria.” June 22. Vanguard. http://community.vanguardngr.com/profiles/blogs/abc-of-recalling -a-law-maker-in-nigeria. ———. 2016. “The Law and Politics of Tenure Truncation of the Legislature in Nigeria.” Journal of International Law and Strategic Studies 1: 1–22. Ahmed, Einas. 2009. “The Comprehensive Peace Agreement and the Dynamics of Post-conflict Political Partnership in Sudan.” Africa Spectrum 44, no. 3: 133−147. Ainley, Kirsten. 2016. “The Great Escape? The Role of the International Criminal Court in the Colombian Peace Process.” Justice in Conflict. http://eprints .lse.ac.uk/68228/1/Ainley_The%20great%20escape.pdf.

185

186

Bibliography

Ākhatāra, Muhāmmada Iẏāhaiẏā. 2001. Electoral Corruption in Bangladesh. London: Ashgate. Alape, Arturo. 1998. “The Possibilities for Peace.” North American Congress Latin America Report on the Americas 31, no. 5: 36−37. Algerian Constitution. 1989. http://confinder.richmond.edu/admin/docs/local _algeria.pdf. Altaweel, Mark, David Sallach, and Charles Macal. 2012. “Mobilizing for Change: Simulating Political Movements in Armed Conflicts.” Social Science Computer Review 31, no 2: 1−22. Altman, David. 2011. Direct Democracy Worldwide. New York: Cambridge University Press. ———. 2014. “Referendums in Latin America.” In Referendums Around the World: The Continued Growth of Direct Democracy, ed. Matt Qvortrup. Basingstoke: Palgrave Macmillan, 162−185. Alvarez, R. Michael, Melanie Goodrich, Thad E. Hall, D. Roderick Kiewiet, and Sarah M. Sled. 2004. “The Complexity of the California Recall Election.” Political Science and Politics 37, no. 1: 23−26. Anastasiou, Harry. 2007. “Nationalism as a Deterrent to Peace and Interethnic Democracy: The Failure of Nationalist Leadership from the Hague Talks to the Cyprus Referendum.” International Studies Perspectives 8, no. 2: 190−205. Anda, Redempto. 2014. “Recall petition filed vs. Puerto Princesa mayor” Inquirer .net. http://newsinfo.inquirer.net/586537/recall-petition-filed-vs-puerto-princesa -mayor. Angelo, Paul. 2017. “The Colombian Peace Process: Trial and Error.” Survival 59, no. 1: 135−148. Argyle, Ray. 2011. Turning Points: The Campaigns that Changed Canada—2011 and Before. Fareham: Waterside Books, e-book. Aristotle. A Treatise on Citizenship. Assal, Munzoul A. 2011. “Nationality and Citizenship Questions in Sudan After the Southern Sudan Referendum Vote.” University of Khartoum. http:// khartoumspace.uofk.edu/handle/123456789/22192. Attahi, K. 1996. “Côte d’Ivoire.” In The Changing Nature of Local Government in Developing Countries, ed. P. L. McCarney. Toronto, ON: Centre for Urban and Community Studies, University of Toronto and Federation of Canadian Municipalities, 107–128. Băişanu, Ştefan Alexandru. 2010. “The Referendum in Different Political and Constitutional Systems.” In Annals of “Ştefan cel Mare” University of Suceava Philosophy, Social and Human Disciplines Series. Vol. 1, ed. Bogdan Popveniuc. Suceava, Romania: University of Suceava Philosophy, Social and Human Disciplines Press: 62–74. Baji, Petra, Milena Pavlova, László Gulácsi, and Wim Groot. 2011. “User Fees for Public Health Care Services in Hungary: Expectations, Experience, and Acceptability from the Perspectives of Different Stakeholders.” Health Policy 102, no. 2: 255−262. Baker, Chris, and Pasuk Phongpaichit. 2014. A History of Thailand. Cambridge: Cambridge University Press. Barankay, Iwan, Pascal Sciarini, and Alexander H. Trechsel. 2003. “Institutional Openness and the Use of Referendums and Popular Initiatives: Evidence from Swiss Cantons.” Swiss Political Science Review 9, no. 1: 1−33.

Bibliography

187

Barczak, Monica. 2001. “Representation by Consultation? The Rise of Direct Democracy in Latin America.” Latin American Politics and Society 43, no. 3: 37–60. BBC. 2012a. “Timeline: Scottish Independence Referendum.” BBC, October 15. ———. 2012b. BBC, October 14. ———. 2014a. “Scottish Referendum: Scotland Votes No to Independence.” BBC News, September 19. ———. 2014b. “Is Crimea’s Referendum Legal?” BBC News, March 13. Bektemirov, Kuatbay, and Eduard Rahimov. 2001. “Local Government in Uzbekistan.” In Developing New Rules in the Old Environment: Local Governments in Eastern Europe, in the Caucasus and in Central Asia. Budapest: Local Government and Public Service Reform Initiative, Open Society Institute Europe, 469−520. Beramendi, Virginia. 2008. Direct Democracy: The International IDEA Handbook. Stockholm: International Institute for Democracy and Electoral Assistance (IDEA). Bjorklund, Tor. 1982. “The Demand for Referendum: When Does It Arise and When Does It Succeed?” Scandinavian Political Studies 5, no. 3: 237−259. ———. 1996. “The Three Nordic 1994 Referenda Concerning Membership in the EU.” Cooperation and Conflict 31, no. 1: 11−36. Black, Andrew. 2009. “Q&A: Independence Referendum.” BBC News, September 10. ———. 2013. “Scottish Independence: SNP Accepts Call to Change Referendum Question.” BBC, January 30. Blaydes, Lisa. 2006. “Who Votes in Authoritarian Elections and Why? Determinants of Voter Turnout in Contemporary Egypt.” Paper presented at the meeting of the American Political Science Association, September 2006. Blumenthal, Mark. 2012. “Wisconsin Recall Polls: Scott Walker Leads, but the Margin Varies.” HuffingtonPost.com, June 6. https://www.huffingtonpost .com/2012/06/05/wisconsin-recall-polls-scott-walker-tom-barrett-margin _n_1570869.html Bogdanor, Vernon. 1981. The People and the Party System: The Referendum and Electoral Reform in British Politics. Cambridge: Cambridge University Press. ———. 1994. “Western Europe.” In Referendums Around the World: The Growing Use of Direct Democracy, eds. David Butler and Austin Ranney. Washington, DC: Macmillan: 24–97. Bose, Sumantra. 2007. Contested Lands. Boston: Harvard University Press. Bowler, Shaun, and Todd Donovan. 1998. Demanding Choices: Opinion, Voting and Direct Democracy. Ann Arbor: University of Michigan Press. ———. 2002a. “Democracy, Institutions, and Attitudes about Citizen Influence on Government.” British Journal of Political Science 32, no. 2: 371−391. ———. 2002b. “Do Voters Have a Cue? Television Advertisements as a Source of Information in Citizen-Initiated Referendum Campaigns.” European Journal of Political Research 41, no. 6: 777–793. ———. 2004. “Measuring the Effects of Direct Democracy on State Policy and Politics.” State Politics and Policy Quarterly 4, no. 3: 345−363. Bowler, Shaun, Todd Donovan, and Trudi Happ. 1992. “Ballot Propositions and Information Costs: Direct Democracy and the Fatigued Voter.” Western Political Quarterly 45, no. 2: 559−568.

188

Bibliography

Bowler, Shaun, Todd Donovan, and Caroline J. Tolbert, eds. 1998. Citizens as Legislators: Direct Democracy in the United States. Columbus: Ohio State University Press. Boyer, Patrick. 1982. Lawmaking by the People: Referendums and Plebiscites in Canada. Toronto, ON: Butterworths. ———. 1988. “Plebiscites in a Parliamentary Democracy.” Canadian Parliamentary Review 11, no 1: 2–5. Bratton, Michael, and Gina Lambright. 2001. “Uganda’s Referendum 2000: The Silent Boycott.” Working Paper 6. Cape Town: Afrobarometer. Broder, David S. 2000. Democracy Derailed: Initiative Campaigns and the Power of Money. New York: Harcourt. Broome, Benjamin J. 2002. “Participatory Planning and Design in a Protracted Conflict Situation: Applications with Citizen Peace-Building Groups in Cyprus.” Systems Research and Behavioral Science 19, no. 4: 313–21. Buček, Ján. 1999. “Land, ownership and living environment of Roma minority in Slovakia.” European Urban and Regional Studies 6, no. 4: 360–5. Buček, Ján, and Brian Smith. 2000. “New Approaches to Local Democracy: Direct Democracy, Participation and the ‘Third Sector.’” Environment and Planning C: Government and Policy 18, no. 1: 3−16. Buetzer, Michael. 2011. “Second-Order Direct Democracy in Switzerland: How Sub-national Experiences Differ from National Ballots.” In Local Direct Democracy in Europe, ed. Theo Schiller. Wiesbaden, DE: VS Verlag für Sozialwissenschaften: 138−156. Butler, David, and Austin Ranney. 1978. Referendums: A Comparative Study of Practice and Theory. Washington, DC: American Enterprise Institute for Public Policy Research. ———, eds. 1994. Referendums Around the World: The Growing Use of Direct Democracy. Washington, DC: Macmillan. Butler, Eamonn. 2007. “Hungary and the European Union: The Political Implications of Societal Security Promotion.” Europe-Asia Studies 59, no. 7: 1115−1144. Cardenas, Mauricio, Roberto Junguito, and Mónica Pachón. 2006. “Political Institutions and Policy Outcomes in Colombia: The Effects of the 1991 Constitution.” Inter-American Development Bank Working Paper No. 203. https://ssrn.com/abstract=1814757. Cardinal, Mario. 2005. Breaking Point: Quebec, Canada, the 1995 Referendum. Montreal, QC: Bayard Canada Books. Carrell, Severin, and Nicholas Watt. 2012. “Scottish Independence: Alex Salmond Sets Poll Date—and Defies London.” The Guardian. January 10, 2012. https://www.theguardian.com/politics/2012/jan/10/scottish-independence -salmond-poll-date. Catala, Amandine. 2015. “Secession and Annexation: The Case of Crimea.” German Law Journal 16, no. 1: 581−607. Catt, Helena. 1996. “The Other Democratic Experiment: New Zealand’s Experience with Citizens’ Initiated Referendum.” Political Science 48, no. 1: 29–47. Chadjipadelis, Theodore, and Ioannis Andreadis. 2007. “Analysis of the Cyprus Referendum on the Annan Plan.” Paper presented at the conference of the Political Studies Association. United Kingdom, November 27. Chen, Yen-ting, and Jake Chung. 2015. “Alex Tsai Recall Push Enters Final Stage.” Taipei Times, February 8.

Bibliography

189

Childers, Matt, and Mike Binder. 2012. “Engaged by the Initiative? How the Use of Citizen Initiatives Increases Voter Turnout.” Political Research Quarterly 65, no. 1: 93−103. Chrétien, Jean. 2007. My Years as Prime Minister. Toronto, ON: Vintage Canada. Christakis, Theodore. 2015. “Self-Determination, Territorial Integrity and Fait Accompli in the Case of Crimea.” ZaöRV/Heidelberg Journal of International Law 75, no. 1: 75−100. Christopher, Anthony J. 2011. “Secession and South Sudan: An African Precedent for the Future?” South African Geographical Journal 93, no. 2: 125−132. Clarke, Harold D., and Allan Kornberg. 1994. “The Politics and Economics of Constitutional Choice: Voting in Canada’s 1992 National Referendum.” Journal of Politics 56, no. 4: 940−962. Clegg, David. 2012. “Independence Referendum: Scotland Facing Constitutional Chaos.” The Courier, January 11. Copper, John F. 1990. Taiwan’s Recent Elections: Fulfilling the Democratic Promise. Baltimore: School of Law, University of Maryland. Corrales, Javier, and Michael Penfold-Becerra. 2007. “Venezuela: Crowding Out the Opposition.” Journal of Democracy 18, no. 2: 99−113. Côté, Pierre-F. 1992. Instruments of Direct Democracy in Canada and Quebec. Quebec City, QC: Directeur Général des Élections du Québec. Coufoudakis, Van. 2004. “Cyprus—The Referendum and Its Aftermath.” Cyprus Review 16, no. 2: 67−82. Cummings, Sally N. 2013. Domestic and International Perspectives on Kyrgyzstan’s “Tulip Revolution”: Motives, Mobilization and Meanings. New York: Routledge. Cummings, Sally N., and Maxim Ryabkov. 2008. “Situating the ‘Tulip Revolution.’” Central Asian Survey 27, nos. 3−4: 241−252. Darmonovic, Srdjan. 2007. “Montenegro: A Miracle of the Balkans?” Journal of Democracy 18, no. 2: 152−159. de Vreese, Claes H. 2004. “Primed by the Euro: The Impact of a Referendum Campaign on Public Opinion and Evaluations of Government and Political Leaders.” Scandinavian Political Studies 27, no. 1: 45−64. de Vreese, Claes H., and Holli A. Semetko. 2002. “Cynical and Engaged Strategic Campaign Coverage, Public Opinion, and Mobilization in a Referendum.” Communication Research 29, no. 6: 615−641. ———. 2004. “News Matters: Influences on the Vote in the Danish 2000 Euro Referendum Campaign.” European Journal of Political Research 43, no. 5: 699–722. Democracy Assistance and Elections News. n.d. “Election Guide: Kyrgyzstan.” Democracy Assistance and Elections News. www.electionguide.org/election .php?ID=903. Dicey, Albert Venn. 1890. “Ought the Referendum to Be Introduced into England?” Contemporary Review 57: 486−512. ———. 1911. A Leap in the Dark. 2nd ed. London: John Murray. Dodd, Lawrence C. 1995. “The New American Politics: Reflections on the Early 1990s.” In The New American Politics, ed. Byron D. Jones. Boulder: Westview Press. Donovan, Todd. 2014. “Referendums and Initiatives in North America.” In Referendums Around the World: The Continued Growth of Direct Democracy, ed. Matt Qvortrup. Basingstoke, Palgrave Macmillan, 122−161.

190

Bibliography

Donovan, Todd, Caroline Tolbert, Daniel A. Smith, and Janine Parry. 2005. “Did Gay Marriage Elect George W. Bush?” Paper prepared for the State Politics Conference, East Lansing, MI, May. Dubois, Philip L. 1979. “Voter Turnout in State Judicial Elections: An Analysis of the Tail on the Electoral Kite.” Journal of Politics 41, no. 3: 865−887. Edrisinha, Rohan. 1998. “Trying Times: Constitutional Attempts to Resolve Armed Conflict in Sri Lanka.” Conciliation Resources. Eke, Steven M., and Taras Kuzio 2000. “Sultanism in Eastern Europe: The Socio-Political Roots of Authoritarian Populism in Belarus.” Europe-Asia Studies 52, no. 3: 523−547. Election Guide. 2017. “Election Guide” Democracy Assistance and Election News. www.electionguide.org. Ershad, Hussain M. 1981. “Role of Military in Underdeveloped Countries.” Bangladesh Army Journal: 12. Eschet-Schwarz, André. 1989. “The Role of Semi-Direct Democracy in Shaping Swiss Federalism: The Behavior of Cantons Regarding Revision of the Constitution, 1866–1981.” Publius: The Journal of Federalism 19, no. 1: 79−106. European Union. 2017. “EU Treaties.” European Union. https://europa.eu /european-union/law/treaties_en. Everson, David H. 1981. “The effects of initiatives on voter turnout: A comparative state analysis.” Western Political Quarterly 34, no. 3: 415–425. Fedorov, Yury. 2012. “Uzbekistan: Clans, Succession, and Stability.” Security Index: A Russian Journal on International Security 18, no. 2: 39−54. Fisch, William B. 2006. “Constitutional Referendum in the United States of America.” American Journal of Comparative Law 54, no. 2: 485−500. Ford, Patrick, and Taingay Kemokai. 2014. “Direct Democracy: A Global Comparative Study on Electoral Initiative and Referendum Mechanisms.” Unpublished manuscript, McGeorge School of Law, University of the Pacific. Fox, John, Robert Andersen, and Joseph Dubonnet. 1999. “The Polls and the 1995 Quebec Referendum.” Canadian Journal of Sociology 24, no. 3: 411–424. Fraser, Douglas. 2015. “Study Examines Referendum Demographics.” BBC News, September 18. Frey, Bruno S. 2003. “Direct Democracy for Transition Countries.” Working Paper No. 165, Osnabrück: Institut für Empirische Wirtschaftsforschung (IEW). Frey, Bruno S., and Alois Stutzer. 2004. “The Role of Direct Democracy and Federalism in Local Power.” Working Paper No. 209, Osnabrück: Institut für Empirische Wirtschaftsforschung (IEW). Fry, Gerald W. 2014. “A Comedy of Errors: A Critical Analysis of the Political Drama in Thailand.” Harvard International Review 35, no. 4: 29–31. Fumagalli, Matteo. 2016. “Semi-Presidentialism in Kyrgyzstan.” In SemiPresidentialism in the Caucasus and Central Asia, eds. Robert Elgie and Sophia Moestrup. London: Palgrave Macmillan, 173−205. Gardham, Magnu. 2014. “Pro-independence Campaigners Protest Outside BBC Studios.” Herald Scotland, June 29. Garrett, Elizabeth. 2004. “Democracy in the Wake of the California Recall.” University of Pennsylvania Law Review 153, no. 1: 239−284. Gastil, John, Justin Reedy, and Chris Wells. 2007. “When Good Voters Make Bad Policies: Assessing and Improving the Deliberative Quality of Initiative Elections.” University of Colorado Law Review 78: 1435−1488.

Bibliography

191

Gerber, Elizabeth R. 2001. Stealing the Initiative: How State Government Responds to Direct Democracy. Upper Saddle River, NJ: Prentice Hall. Gibendi, Ramenya. 2016. “IEBC Throws out CORD’s Okoa Kenya Initiative.” Tuko, March 21. Gill, Lesley. 2017. “Another Chance for Peace in Colombia?” Journal of Latin American and Caribbean Anthropology 22, no. 1: 157−160. Goldberg, Andreas C. 2017. The Impact of Cleavages on Swiss Voting Behaviour. New York: Springer. Gomez, Juan C. 2001. “Prospects for Peace in Colombia: Objectives and Strategies of the Main Actors.” PhD diss., Naval Postgraduate School. Government Accountability Board. n.d. “State of Wisconsin.” Wisconsin Elections Commission www.gab.wi.gov/elections-voting/recall. Gruner, Erich. 1969. Die Parteien in der Scheiz. Bern: Francke Verlag. Haider-Markel, Don P., Alana Querze, and Kara Lindaman. 2007. “Lose, Win, or Draw? A Reexamination of Direct Democracy and Minority Rights.” Political Research Quarterly 60, no. 2: 304−314. Hajnal, Zoltan L., Elisabeth R. Gerber, and Hugh Louch. 2002. “Minorities and Direct Legislation: Evidence from California Ballot Proposition Elections.” Journal of Politics 64, no. 1: 154−177. Haljan, David. 2014. Constitutionalizing Secession. Portland: Bloomsbury. Hancock, Landon E. 2011. “There Is No Alternative: Prospect Theory, the Yes Campaign and Selling the Good Friday Agreement.” Irish Political Studies 26, no. 1: 95−116. Hangartner, Yvo, and Andres Kley. 2000. Die demokratischen Rechte in Bund und Kantonen der Schweizerischen Eidgenossenschaft. Zurich: Schulthess. Hashem, Said. 2014. “Map: Russian Language Dominant in Crimea.” Al Jazeera America, March 15. Hayes, Bernadette C., and Ian McAllister. 2001. “Who Voted for Peace? Public Support for the 1998 Northern Ireland Agreement.” Irish Political Studies 16, no. 1: 73−93. Hayes, Bernadette C., Ian McAllister, and Lizanne Dowds. 2005. “The Erosion of Consent: Protestant Disillusionment with the 1998 Northern Ireland Agreement.” Journal of Elections, Public Opinion and Parties 15, no. 2: 147−167. He, Baogang. 2002. “Referenda as a Solution to the National-Identity/Boundary Question: An Empirical Critique of the Theoretical Literature.” Alternatives: Global, Local, Political 27, no. 1: 67−97. Hébert, Chantal. 2014. The Morning After: The 1995 Referendum and the Day that Almost Was. Toronto, ON: Knopf Canada. Hellinger, Daniel. 2005. “When ‘No’ Means ‘Yes to Revolution’: Electoral Politics in Bolivarian Venezuela.” Latin American Perspectives 32, no. 3: 8−32. Hellwig, Timothy, and David Samuels. 2008. “Electoral Accountability and the Variety of Democratic Regimes.” British Journal of Political Science 38, no. 1: 65−90. Hewison, Kevin. 2014. “Considerations on Inequality and Politics in Thailand.” Democratization 21, no. 5: 846−866. Hill, Ronald J., and Stephen White. 2014. “Referendums in Russia, the Former Soviet Union and Eastern Europe.” In Referendums Around the World: The Continued Growth of Direct Democracy, ed. Matt Qvortrup. New York: Springer: 17–42.

192

Bibliography

Hitler, Adolph. 1934. Mon combat, trad. J. Gaudefroy-Demombynes et A. Calmettes, Paris. Hobolt, Sara B. 2009. Europe in Question: Referendums on European Integration. Oxford: Oxford University Press. Høigilt, Jacob, Ashlid Falch, and Oystein H. Rolandsen. 2010. The Sudan Referendum and Neighbouring Countries: Egypt and Uganda. Oslo: Peace Research Institute Oslo (PRIO). Hoskin, Gary, and Gabriel Murillo-Castaño. 1999. “Can Colombia Cope?” Journal of Democracy 10, no. 1: 36−50. Hsu, Chiehwen, ed. 2015. “Recall: A Demand for Accountability from Elected Officials.” Thinking Taiwan. February 5. http://thinking-taiwan.com/total -recall-a-demand-for-accountability-from-elected-officials/. Hug, Simon, and Pascal Sciarini. 2000. “Referendums on European Integration: Do Institutions Matter in the Voter’s Decision?” Comparative Political Studies 33, no. 1: 3–36. Hughes, Colin. A. 1994. “Australia and New Zealand.” In Referendums Around the World: The Growing Use of Direct Democracy, eds. David Butler and J. Austin Ranney. Basingstoke: Palgrave Macmillan, 154−173. Huskey, Eugene, and David Hill. 2011. “The 2010 Referendum and Parliamentary Elections in Kyrgyzstan.” Electoral Studies 30, no. 4: 876−879. Idler, Annette. 2015. “Colombia Just Voted No on Its Plebiscite for Peace. Here’s Why and What It Means.” Washington Post, October 3. Ignatova, Olga. 2014. “Public Opinion Survey in Crimea.” GfK Ukraine. http:// avaazpress.s3.amazonaws.com/558_Crimea.Referendum.Poll.GfK.pdf. Ilkhamov, Alisher. 2002. “Controllable Democracy in Uzbekistan.” Middle East Report, no. 222: 8−10. International Institute for Democracy and Electoral Assistance (International IDEA). 2014. “Country View: Kyrgyzstan.” IDEA. www.oldsite.idea.int/db /countryview.cfm?id=116. International Republican Institute. 2013. Public Opinion Survey Residents of the Autonomous Republic of Crimea May 16–30, 2013. USAID. http://www.iri.org /sites/default/files/2013%20October%207%20Survey%20of%20Crimean %20Public%20Opinion,%20May%2016-30,%202013.pdf. Jackman, Simon. 2004. “Same-Sex Marriage Ballot Initiatives and Conservative Mobilization in the 2004 Election.” Paper presented at Stanford University, Palo Alto, CA. http://jackman.stanford.edu/projects/RISSPresentation.pdf. Jahn, Detlef, and Ann-Sofie Storsved. 2016. “Legitimacy Through Referendum? The Nearly Successful Domino-Strategy of the EU-Referendum in Austria, Finland, Sweden and Norway.” West European Politics 18, no. 4: 18. Jensen, Erik. 2005. Western Sahara: Anatomy of a Stalemate. London: Lynne Rienner. Jiménez, Raúl. 2011. “Forensic Analysis of the Venezuelan Recall Referendum.” Statistical Science 26, no. 2: 564−583. Joffé, George. 2008. “National Reconciliation and General Amnesty in Algeria.” Mediterranean Politics 13, no. 2: 213−228. Johansson, Patrik. 2009. “Putting Peace to the Vote: Displaced Persons and a Future Referendum on Nagorno-Karabakh.” Refugee Survey Quarterly 28, no. 1: 122−139. Johnston, Richard. 1993. “An Inverted Logroll: The Charlottetown Accord and the Referendum.” PS: Political Science and Politics 26, no. 1: 43−48.

Bibliography

193

Jory, Patrick. 2016. “The Real Meaning of Thailand’s Constitutional Referendum.” Asian Currents. http://asaa.asn.au/the-real-meaning-of-thailands -constitutional-referendum. Juraev, Shairbek. 2008. “Kyrgyz Democracy? The Tulip Revolution and Beyond.” Central Asian Survey 27, nos. 3−4: 253−264. Kampwirth, Ralph. 2004. “Direct Democracy in Germany.” In Direct Democracy in Europe: A Comprehensive Reference Guide to the Initiative and Referendum Process in Europe, eds. Bruno Kaufmann and M. Dane Waters. Durham: Carolina Academic Press: 63–67. Karanja, Samuel. 2015. “Kenya: CORD Seeks Major Changes in Elections, Overhaul of IEBC.” Daily Nation, November 10. Kaufmann, Bruno. 2012. “Transnational ‘Babystep’: The European Citizens’ Initiative.” In Citizens’ Initiatives in Europe: Procedures and Consequences of Agenda-Setting by Citizens, eds. Maija Setälä and Theo Schiller. New York: Palgrave Macmillan: 228–242. Kersting, Norbert. 2009. “Direct Democracy in Southern and East Africa: Referendums and Initiatives.” Journal of African Elections 8, no. 2: 1−22. ———. 2010. “Direct Democracy in Constitutional Processes: The South African Plebiscite of 1992.” Politikon 37, nos. 2−3: 207−222. ———. 2014. “Referendums in Africa.” In Referendums Around the World: The Continued Growth of Direct Democracy, ed. Matt Qvortrup. Basingstoke: Palgrave Macmillan, 186−206. Khegai, Janna. 2004. “The Role of Clans in Post-independence State-Building in Central Asia.” In European Consortium for Political Research Joint Sessions of Workshop. Uppsala, April 13–18. Kirkpatrick, David D. 2012. “Egyptian Islamists Approve Draft Constitution Despite Objections.” New York Times, November 29. Kobach, Kris W. 1993. The Referendum: Direct Democracy in Switzerland. Aldershot: Dartmouth. ———. 1994. “Switzerland.” In Referendums Around the World: The Growing Use of Direct Democracy, eds. David Butler and Austin Raney. Washington, DC: AEI Press. Kobori, Masahiro. 2014. “Referendums in Asia.” In Referendums Around the World: The Continued Growth of Direct Democracy, ed. Matt Qvortrup. Basingstoke: Palgrave Macmillan, 207−217. Kocieniewski. David. 2012. “Unions, at Center of Wisconsin Recall Vote, Suffer a New Setback in Its Outcome.” New York Times, June 6. Kornblith, Miriam, and Vinay Jawahar. 2005. “Elections versus Democracy.” Journal of Democracy 16, no. 1: 124−137. Kyrgyzstan Election Guide. 2017. http://www.electionguide.org/countries/id/117. Ladner, Andreas, 2002. “Size and Direct Democracy at the Local Level: The Case of Switzerland.” Environment and Planning C: Government and Policy 20, no. 6: 813−828. Laponce, Jean. 2010. Le Référendum de Sourverainteté: Comparaisons, Critiques et Commentaires. Quebec, QC: Les Presses de L’Université Laval. Lawal, Tolu, and Abegunde Oladunjoye. 2010. “Local Government, Corruption and Democracy in Nigeria.” Journal of Sustainable Development 12, no. 5: 227–235. LeDuc, Lawrence. 2015. “Referendums and Deliberative Democracy.” Electoral Studies 38, no. 1: 139−148.

194

Bibliography

Lee, Eugene C. 1981. “The American Experience 1778−1978.” In The Referendum Device, ed. Austin Ranney. Washington, DC: AEI Press, 46−72. Lee, Sung Yong, and Roger MacGinty. 2012. “Context and Postconflict Referendums.” Nationalism and Ethnic Politics 18, no. 1: 43−64. Levitsky, Steven, and James Loxton. 2013. “Populism and Competitive Authoritarianism in the Andes.” Democratization 20, no. 1: 107−136. Loizides, Neophytos. 2014. “Negotiated Settlements and Peace Referendums.” European Journal of Political Research 53, no. 2: 234−249. Lowenstein, Daniel H. 1982. “Campaign Spending and Ballot Propositions: Recent Experience, Public Choice Theory and the First Amendment.” UCLA Law Review 29, no. 3: 519–547. Lupia, Arthur. 1994a. “Shortcuts Versus Encyclopedias: Information and Voting Behavior in California Insurance Reform Elections.” American Political Science Review 88, no. 1: 63−76. ———. 1994b. “The Effect of Information on Voting Behavior and Electoral Outcomes: An Experimental Study of Direct Legislation.” Public Choice 78, no. 1: 65−86. ———. 2001. “Dumber than Chimps? An Assessment of Direct Democracy Voters.” In Dangerous Democracy: The Battle over Ballot Initiatives in America, eds. Larry J. Sabato, Howard R. Ernst, and Bruce A. Larson. Lanham, MD: Rowman and Littlefield, 66–70. Lutz, Georg. 2012. “Switzerland: Citizens’ Initiatives as a Measure to Control the Political Agenda.” In Citizens’ Initiatives in Europe: Procedures and Consequences of Agenda-Setting by Citizens, eds. Maija Setälä and Theo Schiller. New York: Palgrave Macmillan, 17–36. MacDonnell, Hamish. 2011. “16-Year-Olds Likely to Get the Vote on Union Split.” The Times Scotland, September 17. Magleby, David B. 1984. Direct Legislation: Voting on Ballot Propositions in the United States. Baltimore: Johns Hopkins University Press. Manweller, Mathew. 2005. The People vs. the Courts: Judicial Review and Direct Democracy in the American Legal System. Bethesda, MD: Academia Press. ———. 2012. “Success from Amateur to Zealot: A Typology of Initiative Activists.” In Direct Democracy in the United States: Petitioners as a Reflection of Society, eds. Shauna Reilly and Ryan M. Yonk. New York: Routledge, 12–43. Marques, Alvaro, and Thomas B. Smith. 1984. “Referendums in the Third World.” Electoral Studies 3, no. 1: 85−105. Marquis, Pierre. 1993. “Referendums in Canada: The Effect of Populist DecisionMaking on Representative Democracy.” Political and Social Affairs Division. Government of Canada. http://publications.gc.ca/Collection-R/LoPBdP/BP /bp328-e.htm. Marxer, Wilfried, and Zoltán Tibor Pállinger. 2011. “Direct Democracy at the Local Level: A Comparison of Eight Municipalities from Four Countries around Lake Constance (Austria, Germany, Liechtenstein, Switzerland).” In Local Direct Democracy in Europe, ed. Theo Schiller. New York: Springer Science+Business Media, 157–169. Masakhalia, Alan. 2011. “Direct Democracy in Kenya: A Case to Be Made.” openDemocracy. Nov. 3. www.opendemocracy.net/alan-e-masakhalia/direct -democracy-in-kenya-case-to-be-made. Matsusaka, John G. 2004. For the Many or the Few: The Initiative, Public Policy, and American Democracy. Chicago: University of Chicago Press.

Bibliography

195

Mattern, Johannes. 1921. The Employment of Plebiscite in the Determination of Sovereignty. Baltimore: Johns Hopkins University Press. Maya, Margarita López, and Luis E. Lander. 2011. “Participatory Democracy in Venezuela: Origins, Ideas, and Implementation.” In Venezuela’s Bolivarian Democracy: Participation, Politics, and Culture Under Chávez, eds. David Smilde and Daniel Hellinger. Durham: Duke University Press, 58−59. McCargo, Duncan. 2017. “Thailand in 2016.” Asian Survey 57, no. 1: 150−156. McCoy, Jennifer. 2005. “One Act in an Unfinished Drama.” Journal of Democracy 16, no. 1: 109−123. McGarry, John, and Brendan O’Leary. 1993. “Introduction: The Macro-Political Regulation of Ethnic Conflict.” In The Politics of Ethnic Conflict Regulation: Case Studies of Protracted Ethnic Conflict, eds. John McGarry and Brendan O’Leary. London: Routledge, 1–40. Menya, Walter. 2015. “Cord Proposes Radical Changes in Okoa Kenya Drive.” Daily Nation, January 31. Miller, Kenneth P. 2009. Direct Democracy and the Courts. Cambridge: Cambridge University Press. Möckli, Silvano. 1996. “Das Gesetzesveto und-referendum: Ein Stopperstein wird zum Grundstein.” In Les origines de la démocratie directe en Suisse— Die Ursprünger der schweizerischen direkten Demokratie, ed. Andres Auer. Basel: Helibing and Lichtenhahn, 209–220. Montopoli, Brian. 2012. “Scott Walker Wins Wisconsin Recall Election.” CBS News, June 6. Montoya, Miguel. 2008. “Transparency as Tool and Weapon: The Case of the Venezuelan Presidential Recall Referendum.” In Transparency in a New Global Order: Unveiling Organizational Visions, eds. Christine Garsten and Monica Lindh De Montoya. Northampton, MA: Edward Elgar Publishing, Inc., 260–283. Morris, Caroline. 2014. “Referendums in Oceania.” In Referendums Around the World: The Continued Growth of Direct Democracy, ed. Matt Qvortrup. Basingstoke: Palgrave Macmillan, 218−245. Mowery, Tim, and Alain Pelletier. 2001. “Referendums in Canada: A Comparative Overview.” Electoral Insights 3, no. 1: 18–22. Moyers, Bill, and Michael Winship. 2012. “Campaign Cash: The Gift that Keeps on Giving.” Huffington Post, June 15. Muriaas, Ragnhild Louise. 2009. “Reintroducing a Local-Level Multiparty System in Uganda: Why Be in Opposition?” Government and Opposition 44, no. 1: 91−112. Murphy, Helen, and Julia Symmes Cobb. 2016. “Columbians Reject Deal to End 52-Year FARC Rebel War.” Reuters, October 3. National Council of State Legislators. 2016. “Recall of State Officials.” www.ncsl .org/research/elections-and-campaigns/recall-of-state-officials.aspx. Nelson, Michael H. 2011. “Looking Back Before the Election of 2011: Thailand’s Constitution Referendum and the Election of 2007.” European-Asian Journal of Law and Governance 1, no. 1: 49−74. Nicholson, Steven P. 2003. “The Political Environment and Ballot Proposition Awareness.” American Journal of Political Science 47, no. 3: 403−410. ———. 2005. Voting the Agenda: Candidates, Elections and Ballot Measures. Princeton: Princeton University Press. Nohlen, Dieter. 2005. Elections in the Americas: A Data Handbook. Vol. 2. Baden-Baden: Nomos.

196

Bibliography

Nohlen, Dieter, M. Krennerich, and B. Thibaut. 1999. Elections in Africa: A Data Handbook. Baden-Baden: Nomos. Nohlen, Dieter, and P. Stöver. 2010. Elections in Europe: A Data Handbook. Baden-Baden: Nomos. Ockey, James. 2008. “Thailand in 2007: The Struggle to Control Democracy.” Asian Survey 48, no. 1: 20−28. O’Leary, Brendan. 2001. “The Elements of Right—Sizing and Right-Peopling the State.” In Right-Sizing the State: The Politics of Moving Borders, eds. Brendan O’Leary, Ian S. Lustick, and Tom Callaghy. Oxford: Oxford University Press, 15–73. Oliphant, Roland. 2014. “Crimeans Vote Peacefully in Referendum, but Have Little Choice.” The Telegraph. March 16. Orenstein, Mitchell A. 2014. “Putin’s Western Allies.” Foreign Affairs 25. March 16. http://www.saintjoehigh.com/ourpages/auto/2014/3/27/36730719/14 -0326%20Putin_s%20Western%20Allies.pdf. Otieno, Rawlings. 2016. “CORD Warns Electoral Body over Okoa Kenya Initiative.” Standard Digital, January 25. Pállinger, Zoltán Tibor. 2012. “Citizens’ Initiatives in Hungary: An Additional Opportunity for Power-Sharing in an Extremely Majoritarian System.” In Citizens’ Initiatives in Europe: Procedures and Consequences of Agenda-Setting by Citizens, eds. Maija Setälä and Theo Schiller. New York: Palgrave Macmillan: 113–133. Papadopoulos, Yannis. 2001. “How Does Direct Democracy Matter? The Impact of Referendum Votes on Politics and Policy-Making.” In The Swiss Labyrinth: Institutions, Outcomes and Redesign, ed. Jan-Erik Lane. London: Frank Cass, 35−58. Paris, Roland. 2004. At War’s End: Building Peace After Civil Conflict. Cambridge, UK: Cambridge University Press. Parkinson, John. 2001. “Who Knows Best? The Creation of the Citizen-Initiated Referendum in New Zealand.” Government and Opposition 36, no. 3: 403−422. Partos, Gabriel. 2003. “Hungary’s EU Apathy.” BBC News, April 14. Pathmananad, Ukrist. 2016. “Network Thaksin: Structure, Roles and Reaction.” In Unequal Thailand: Aspects of Income, Wealth and Power, eds. Pasuk Phongpaichit and Chris Baker. Singapore: National University of Singapore University Press: 136−161. Pavkovic, Aleksandar, and Peter Radan. 2007. Creating New States: Theory and Practice of Secession. Aldershot: Ashgate. Pericleous, Chrysostomos. 2009. Cyprus Referendum: A Divided Island and the Challenge of the Annan Plan. Vol. 26. London, UK: I.B. Tauris. Philippine Constitution. 1987. Section 3 of the Initiative and Referendum Act. Philippines. Republic Act No. 6735. Phillips, Anne. 1993. Democracy and Difference. University Park: Penn State University Press. Philpot, Robin. 2005. Le Référendum volé. Montreal, QC: Les éditions des intouchables. Polat, Abdumannob. 2002. “Uzbekistan’s Referendum: The Dangers of OverDramatizing the Situation.” Eurasia Insight 26. January 25. http://www .eurasianet.org/departments/insight/articles/eav012602.html. Prado, Raquel, and Bruno Sansó. 2011. “The 2004 Venezuelan Presidential Recall Referendum: Discrepancies Between Two Exit Polls and Official Results.” Statistical Science 26, no. 4: 517−527.

Bibliography

197

Putin, Vladimir. 2014. “Address by President of the Russian Federation.” Russian Government. March 18. http://en.kremlin.ru/events/president/news/20603. Quezon, Manuel L. n.d. “Referendum Score Card.” Quezon Blog. www.quezon .ph/2009/06/09/the-referendum-scorecard-1935-1987/. Qvortrup, Matt. 1999. “A. V. Dicey: The Referendum as the People’s Veto.” History of Political Thought 20, no. 3: 531−546. ———. 2001. “How to Lose a Referendum.” Political Quarterly 72, no. 2: 90−100. ———. 2011. “Hasta La Vista: A Comparative Institutionalist Analysis of the Recall.” Representation 47, no. 2: 161−170. ———. 2014a. “Referendums in Western Europe.” In Referendums Around the World: The Continued Growth of Direct Democracy, ed. Matt Qvortrup. Basingstoke: Palgrave Macmillan, 43−64. ———. 2014b. Referendums and Ethnic Conflict. Philadelphia: University of Pennsylvania Press. ———. 2015. “Power to the People! But How? The Different Uses of Referendums Around the World.” Political Studies Review 13, no. 1: 37−45. ———. 2017. “Demystifying Direct Democracy.” Journal of Democracy 28, no. 3: 141−152. Radnitz, Scott. 2006. “What Really Happened in Kyrgyzstan?” Journal of Democracy 17, no. 2: 132−146. R̲ahmān, Eṃ. 1983. “Theory and Practice of Participatory Action Research.” Paper No. 992260383402676. International Labour Organization. Reilly, Shauna. 2010. Design Meaning and Choice: Petitioners and Voters’ Influence. New York, NY: Ashgate Publishing Limited. Reilly, Shauna and Sean Richey. 2011. “Language Complexity and Roll-off: Does Ballot Question Readability Impact Participation?” Political Research Quarterly 64(1): 59–67. Richmond, Oliver P. 2006. “Shared Sovereignty and the Politics of Peace: Evaluating the EU’s ‘Catalytic’ Framework in the Eastern Mediterranean.” International Affairs 82, no. 1: 149−176. Roberts, Kenneth. 2012. “Populism and Democracy in Venezuela Under Hugo Chávez.” In Populism in Europe and the Americas: Threat or Corrective for Democracy, eds. Cas Mudde and Cristöbal Rovira Kaltwasser. New York, NY: Cambridge University Press: 136−159. Roca, Rene, and Andreas Auer, eds. 2011. Wege zur direkten Demokratie in den schweizerishen Kanton. Zurich: Schulthess. Rourke, John T., Richard P. Hiskes, and Cyrus Ernesto Zirakzadeh. 1992. Direct Democracy and International Politics: Deciding International Issues through Referendums. Boulder, CO: Lynne Reinner. Roussellier, Jacques Eric. 2005. “Quicksand in the Western Sahara? From Referendum Stalemate to Negotiated Solution.” International Negotiation 10, no. 2: 311−336. Royal Commission on Electoral Reform. 1991. Reforming Electoral Democracy: Final Report, Volume 2. Ann Arbor: University of Michigan. Royal Commission on the Electoral System. 1986. Report of the Royal Commission on the Electoral System. Wellington, New Zealand: Royal Commission on the Electoral System. Rumelili, Bahar. 2007. “Transforming Conflicts on EU Borders: The Case of GreekTurkish Relations.” Journal of Common Market Studies 45, no. 1: 105−126.

198

Bibliography

Rybár, Marek. 2005. “From Isolation to Integration: Internal and External Factors of Democratic Change in Slovakia.” In Socializing Democratic Norms: The Role of International Organizations for the Construction of Europe. New York: Palgrave Macmillan, 169−189. Saideman, Stephen. 2014. “In Crimea’s Sham Referendum, All Questions Lead to ‘Yes.’” Globe and Mail, March 10. Sasse, Gwendolyn. 2014. “Crimean Autonomy: A Viable Alternative to War?” Washington Post, March 3. Saunders, Harold. 1999. A Public Peace Process: Sustained Dialogue to Transform Racial and Ethnic Conflicts. New York: Springer. ———. 1985. “We Need a Larger Theory of Negotiation: The Importance of Prenegotiating Phases.” Negotiation Journal 1, no. 3: 249−262. Scarrow, Susan E. 1999. “Parties and the Expansion of Direct Democracy Who Benefits?” Party Politics 5, no. 3: 341−362. Schiller, Theo. 2011. “Local Direct Democracy in Germany—Varieties in Federal States.” In Local Direct Democracy in Europe, ed. Theo Schiller. Wiesbaden, DE: VS Verlag für Sozialwissenschaften, 9−29. ———. 2012. “Initiative Instruments in Germany: Variations in Regional States.” In Citizens’ Initiatives in Europe: Procedures and Consequences of Agenda-Setting by Citizens, eds. Maija Setälä and Theo Schiller. New York: Palgrave Macmillan: 89–112. Schmitt, Carl. 1926. Die Kernfrage des Völkerbundes. F. Dümmler. Schmemann, Serge. 1991. “A Witness to the Revolution: Across East Europe to Moscow, the Trail of Freedom Reaches Tyranny’s Epicenter.’” New York Times, August 25: 11. “Scotland’s Future: Draft Referendum (Scotland) Bill Consultation Paper.” 2010. Scottish Government, Edinburgh. February 25. Serdült, Uwe. 2014. “Referendums in Switzerland.” In Referendums Around the World: The Continued Growth of Direct Democracy, ed. Matt Qvortrup. Basingstoke: Palgrave Macmillan, 65−121. Sheehy, Ann. 1991. “The All-Union and RSFSR Referendums of March 17.” Report on the USSR 3, 19−23. Sitas, Ari, Dilek Latif, and Natasa Loizou. 2007. Prospects of Reconciliation, Co-existence and Forgiveness in Cyprus in the Post-referendum Period. Report No. 4. Oslo, Norway: PRIO Cyprus Centre. Smith, Daniel A., Matthew DeSantis, and Jason Kassel. 2006. “Same-Sex Marriage Ballot Measures and the 2004 Presidential Election.” State and Local Government Review 38, no. 2: 78−91. Smith, Daniel A., and Caroline Tolbert. 2004. Educated by Initiative: The Effects of Direct Democracy on Citizens and Political Organizations in the American States. Ann Arbor: University of Michigan Press. Smith, Mark A. 2002. “Ballot Initiatives and the Democratic Citizen.” Journal of Politics 64, no. 3: 892−903. Smith, Thomas B. 1986. “Referendum Politics in Asia.” Asian Survey 26, no. 7: 793−814. Solimano, Andrés, ed. 2000. Colombia: Essays on Conflict, Peace, and Development. New York, NY: World Bank. Somerville, Ian, and Shane Kirby. 2012. “Public Relations and the Northern Ireland Peace Process: Dissemination, Reconciliation and the ‘Good Friday Agreement’ Referendum Campaign.” Public Relations Inquiry 1, no. 3: 231−255.

Bibliography

199

Southall, Roger. 1994. “The South African Elections of 1994: The Remaking of a Dominant-Party State.” Journal of Modern African Studies 32, no. 4: 629−655. Southern Sudan Referendum Commission. 2011. Southern Sudan Referendum Final Results Report. Khartoum: Southern Sudan Referendum Commission. Strauss, Annette. 1993. “The 1992 Referendum in South Africa.” Journal of Modern African Studies 31, no. 2: 339−360. Stuart, Gavin. 2011. “SNP Launch ‘Re-elect’ Manifesto with Independence Referendum Vow.” STV Group, April 14. Taub, Amanda, and Max Fisher. 2016. “Why Referendums Aren’t as Democratic as They Seem.” New York Times, October 4. Taylor, Jonathan. 2005. Too Many Ties? An Empirical Analysis of the Venezuelan Recall Referendum Counts. Unpublished. Department of Statistics, Stanford University, Stanford, CA. Thompson, Leonard Monteath. 2001. A History of South Africa. New Haven: Yale University Press. Tickner, Arele B. 1998. “Colombia: Chronicle of a Crisis Foretold.” Current History 97, no. 616: 61–65. Tierney, Stephen. 2009. “Constitutional Referendums: A Theoretical Enquiry.” Modern Law Review 72, no. 3: 360−383. ———. 2015. “Sovereignty and Crimea: How Referendum Democracy Complicates Constituent Power in Multinational Societies.” German Law Journal 16, no. 3: 523−541. Tignor, Robert L. 1993. “Political Corruption in Nigeria Before Independence.” Journal of Modern African Studies 31, no. 2: 175−202. Tocci, Nathalie. 2007. The EU and Conflict Resolution: Promoting Peace in the Backyard. New York: Routledge. Tolbert, Caroline J., John A. Grummel, and Daniel A. Smith. 2001. “The Effects of Ballot Initiatives on Voter Turnout in the American States.” American Political Research 29, no. 6: 625–648. Toro, Francisco. 2016. “It’s Official: Venezuela Is a Full-Blown Dictatorship.” Washington Post, October 21. Treacher, Adrian. 1996. “Political Evolution in Post-Soviet Central Asia.” Democratization 3, no. 3: 306−327. Trechsel, Alexander. 2007. “Popular Votes.” In Handbook of Swiss Politics. 2nd ed., eds. Ulrich. Kloti, Peter Knoepfel, Hanspeter Kriesi, Wolf Linder, Yannis Papadopoulos, and Pascal Sciarini. Zurich: Neue Zürcher Zeitung: 435–461. Trechsel, Alexander, and Hanspeter Kriesi. 1996. “Switzerland: The Referendum and Initiative as a Centerpiece of the Political System.” In The Referendum Experience in Europe, eds. Pier Vincenzo Uleri and Michael Gallagher. London: Macmillan: 185–208. Trechsel, Alexander H., and Pascal Sciarini. 1998. “Direct Democracy in Switzerland: Do Elites Matter?” European Journal of Political Research 33, no. 1: 99−124. Tripp, Aili Mari. 2004. “The Changing Face of Authoritarianism in Africa: The Case of Uganda.” Africa Today 50, no. 3: 3−26. Uganda Election Commission. 2015. “Elections Overview.” Electoral Commission. www.ec.or.ug/?q=content/elections. Uleri, Pier Vincenzo. 2011. “The Institutionalization of the Referendum in the Italian Political System: From the National to Regional and Local Levels.”

200

Bibliography

In Local Direct Democracy in Europe, ed. Theo Schiller. Wiesbaden, DE: VS Verlag für Sozialwissenschaften, 75−112. ———. 2012. “Institutions of Citizens’ Political Participation in Italy: Crooked Forms, Hindered Institutionalization.” In Citizens’ Initiatives in Europe: Procedures and Consequences of Agenda-Setting by Citizens, eds. Maija Setälä and Theo Schiller. New York: Palgrave Macmillan: 71–88. Uleri, Pier Vincenzo, and Roberto Fideli. 1996. “Referenda Don’t Fall from the Sky: The June Initiatives.” In Italian Politics—The Stalled Transition, eds. David I. Kertzer and Mario Caciagli. Boulder: Westview Press, 69−99. Ulu, Jyldyzbek Joldosnbek. 2008. “Post-Soviet Coloured Revolutions: An Analysis of Kyrgyzstan’s Tulip Revolution” PhD diss., Middle East Technical University. Urquhart, Conal, Lewis Williamson, and Barry Neild. 2014. “Russia Vetoes Crimea Motion as Kiev Claims it Repelled Russian Troops in South Ukraine.” The Guardian, March 15. Van Walraven, Klaas. 2009. “Decolonization by Referendum: The Anomaly of Niger and the Fall of Sawaba, 1958–1959.” Journal of African History 50, no. 2: 269−292. Vasoric, Aleksandar, and Adrian Croft. 2014a. “Crimea Votes to Join Russia, European Union Imposes Sanctions.” Huffington Post, March 17. ———. 2014b. “U.S., EU Set Sanctions as Putin Recognizes Crimea Sovereignty.” Reuters, March 16. Vidart-Delgado, Maria. 2016. “Elections and Crisis: The Case of Colombia’s Peace Referendum.” Cultural Anthropology website, October 27. https:// culanth.org/fieldsights/983-elections-and-cisis-the-case-of-colombia-s -peace-referendum. Vidmar, Jure. 2011. “South Sudan and the International Legal Framework Governing the Emergence and Delimitation of New States.” Texas International Law Journal 47, no. 3: 541−560. ———. 2014. “The Scottish Independence Referendum in an International Context.” Canadian Yearbook of International Law 51, no. 1: 359–288. Vowles, Jack, 1995. “The Politics of Electoral Reform in New Zealand.” International Political Science Review 16, no. 1: 95−115. Wallis, William. 2005a. “Algeria Plans Amnesty to End 13-Year Civil War.” Financial Times, March 31. ———. 2005b. “Algeria Claims Overwhelming Yes Vote in Referendum.” Financial Times, October 10. Weller, Marc. 2014. “Analysis: Why Russia’s Crimea Move Fails Legal Test.” BBC, March 7. Welp, Yanina. 2016. “Recall Referendums in Peruvian Municipalities: A Political Weapon for Bad Losers or an Instrument of Accountability?” Democratization 23, no. 7: 1162−1179. Wen, Lii. 2014. “Group Submits Petition to Recall KMT’s Alex Tsai.” Taipei Times, December 26. ———. 2015. “Alex Tsai Survives Appendectomy Project.” Taipei Times, February 15. Whitaker, Andrew. 2012. “Scottish Independence Referendum: Publish Legal Advice or Be Damned, SNP Warned over Referendum.” The Scotsman, January 18. Wildavsky, Aaron. 1973. “Government and the People.” Commentary 56, no. 2: 25.

Bibliography

201

Winn, Conrad. 1992. “Educational Failure May Have Defeated the Yes Campaign.” Ottawa Citizen, November 16. Wisconsin Constitution. 2016. “Wisconsin Annotated Constitution.” http://docs .legis.wisconsin.gov/constitution/wi. Wisconsin Democracy Campaign. 2016. “Campaign Finance Database.” www .wisdc.org/wdc.php. Young, John. 2012. The Fate of Sudan: The Origins and Consequences of a Flawed Peace Process. London: Zed Books. Zacks, Amy. 2012. “Recalling Governors: An Overview.” Rutgers University. http://governors.rutgers.edu/on-governors/us-governors/recalling-governors -an-overview/. Zafarullah, Habib, and Muhammad Yeahia Akhter. 2001. “Military Rule, Civilianization and Electoral Corruption: Pakistan and Bangladesh in Perspective.” Asian Studies Review 25, no. 1: 73−94. Zisk, Betty H. 1987. Money, Media, and the Grass Roots. Newbury Park, CA: Sage.

Index

abortion, 69–70, 102 Abyei, 164 accountability, 1–2, 6–7, 11, 30, 37, 39, 44, 50–51, 58, 86, 191–192, 201 Addis Ababa Agreement, 158 African National Congress (ANC), 170–171 Akayev, Askar, 135–136 al-Assad, Bashar, 139 al-Bashir, Omar, 158–160 al-Shabaab, 87 Algeria, ix, 11, 17, 21–25, 34, 144, 186, 192, 200 Algerian Civil War, 23–25, 35 Amena, Hembadoon, 59 American Civil War, 34 Amin, Idi, 44, Annan, Kofi, 173–174, 179, 188, 196 antiparty, 124 apartheid, 170–171, 178 Arab Spring, 127, 138 Austria, 7, 14, 39–40, 42, 65, 117 Australia, 13, 16, 74, 143–144, 160 authoritarian, 7–10, 25, 33, 35, 112, 122–127, 130–131, 136, 138–139, 183, 187, 190 autonomous regions, 103, 107–110, 113, 122, 192

assassination, 30–31

Bakiyev, Kurmanbek, 137 ballot fatigue, 139 ballot language, 4, 141, 146, 148, 149, 151, 152, 156, 160, 162 Bangladesh, 7, 11, 13, 17, 29, 30, 31, 34, 124 barangay(s), 103, 108–109; levels, 92, 104, 110, 112, 122; meetings, 105–106 Bayron, Lucilo, 111, 122 Belfast Agreement, 169 Berlusconi, Silvio, 69 Bloc Quebecois (BQ), 101, 150–152, 163 Bolivarian, 54, 131 Bouchard, Lucien, 101, 151 Brexit, 3, 155 British, 3, 16, 85, 101, 103, 119, 122– 123, 149–150, 153, 169, 172, 174, 188, 192 Bustamante, Cruz, 49 California, 44, 46, 49, 51, 186, 190– 191, 194 Cameron, David, 118 Canada, 8, 44, 65, 93, 100–103, 109,

203

204

Index

121, 144–145, 148–152, 160, 181, 186, 188–189, 191, 194–195 Carter, Jimmy, 55 cease-fire, 176–177 Charest, Jean, 151–152 Charlottetown Accord, 100–101, 193 Charter of Peace and National Reconciliation, 24 Chávez, Hugo, 52, 131–132 Chen Shui-bian, 56–57 Chinese Nationalist Party (KMT), 57 Chirac, Jacques, 151 Chrétien, Jean, 151 Christofias, Dimitris, 175 civil war, 17, 23–25, 34–35, 84, 89, 147–148, 157–158, 160, 162, 164, 200 Coalition for Reforms and Democracy (CORD), 86–88, 90 Cold War, 6 Color Revolutions, 140 Colombia, 41, 64–65, 67, 83–84, 132, 168, 175, 177–179, 182–183, 188, 191–192, 199–200 communist, 122, 171 confederal, 172, 182; system(s), 92– 94, 99, 103, 115, 182 Congo, Republic of the, 13, 143, 163 consociational, 10 Cyprus, 122, 166, 168, 172–175, 178–179, 184, 186, 188–189, 196, 198 Cypriot(s), 172–175, 178–179 Crimea, 141, 149, 155–157, 162, 188–189, 191–192, 199–201 Davis, Gray, 46, 61 de Gaulle, Charles, 7 decolonization, 146–148, 161, 162 de Klerk, F. W., 170–171 Denmark, 12, 115–117, 143 desegregation, 171 developing states, 7–8, 183, 186 developed states, 7, 93 discrimination, 3, 170, 183

Eastern Europe, 28, 119 economics, 16 Edinburgh Agreement, 152–153 education, 19, 28, 54, 75, 78, 85, 102

election day, 54–55, 111, 118, 153 Egypt, 5, 123, 127–128, 138–139, 160, 187, 192 election observers, 55 Ershad, Hussain, 29–31, 124 ethnic fractionalization, 10 ethnonational referenda, 8, 16 European Community, 116–118, 154 European Economic Community, 172 European integration, 115 European Parliament, 116, 120 exit polls, 55 federal assembly, 97, 157 federalist system, 100, 103 federalism, vii, 45, 89, 91, 98, 190 fingerprint scanner, 54 Frazier, Lynn J., 46 French Revolution, 12, 97 frequency, 61, 90, 92–94, 96, 103, 112, 114, 131, 133

Germany, 40, 65, 67, 78–79, 80–82, 181, 193, 195, 198 gerrymandering, 125 Good Friday Agreement, 169 Gorbachev, Mikhail, 140 governance, 1, 31, 61, 92, 121, 124, 141, 184 governor-general, 75 grassroots, 56, 58, 81, 95, 97, 120 Great Britain, 1 gross domestic product (GDP), 10 guerrilla, 83–84, 175, 177, 179

Hagedorn, Edward, 111, 122 Hitler, Adolf, 7 HIV/AIDS, 85 homogenizing referenda, 8 Hungary, ix, 11, 17, 25–29, 34, 65, 117, 186, 188, 196 human rights, 21, 32, 119, 173, 175, 178 hunting, 70 ideology, 16 immigrant, 65–66 impeachment, 37–38, 41, 56, 130 incumbency, 117 independence referenda, 113

Index Independent Electoral and Boundaries Commission (IEBC), 85–88, 90 Ireland, Republic of, 12–13, 115–117, 122, 149, 168–169, 178 Islam, 23–24, 113 Israel, 128 Issa, Darrell, 49 Italy, 64–68, 70, 88–89, 145, 200 Jubilee Coalition, 87–88 judicial, 3, 47–48, 70, 72, 77, 129, 190, 194 Karimov, Islam, 113 Kenya, 85–86, 88 Kyrgyzstan, 13, 40, 42, 124, 127, 133–140, 190–193, 197, 200 Kyrgyz, 134–137, 140

language, 4, 18–20, 33, 39, 84, 94, 98, 120, 141, 146, 148–149, 151– 152, 156, 160, 162, 191, 197 Legal Network for Truthful Elections, 111 Legislative Yuan, the Yuan, 56–57 lieutenant governor, 43, 48–50 Lisbon Treaty, 116, 120 lobby, 74 Loizidou v. Turkey, 180

Maduro, Nicolàs, 55–56, 61 Mandela, Nelson, 170–171 Manning, Preston, 101 Manzoor, Major General Abul, 30 Marcos, Ferdinand, 105–109, 121 martial law, 30, 105–108 member of parliament, 74 member state, 115–117, 120, 122, 172 Middle East, 13–14, 125–127, 193, 201 military coup, 20, 29, 129, 158, 164 minority, 3, 12, 15, 88; groups, 66, 183; populations, 76, 98, 170 moral, 12, 74, 89 movement political system, 32 Mubarak, Hosni, 127 Mulroney, Brian, 7 multimember districts, 41 multiparty, 22, 24, 32, 128, 169; system, 33, 101, 128 Museveni, Yoweri, 124

205

Muslim, 107–108, 129, 131

Naivasha Agreement, 158 Nassar, Gamal, 128 nation building, 15, 125 National Assembly, 23, 25, 35, 88, 108, 130 neopopulist, 5, 125, 131 New Zealand, 5, 13, 64, 67, 69, 74–78, 89–90, 143–145, 147, 182, 189, 192, 196, 198, 200 Nigeria, 38, 40, 43, 45, 58–59, 185, 194, 199 Nimein, Jaafar, 158 19th of April Movement (M-19), 176 North Atlantic Treaty Organization (NATO), 15, 26–27, 119, 166 Northern Ireland, 8, 10, 152, 168–170, 178–179, 185, 191, 199 non-party system, 35 Nobel Peace Prize, 177 Norway, 13, 16, 65, 117, 143, 176 Organization of American States, 166

Palestinian-Israeli conflict, 165 Parizeau, Jacques, 101 Parti Quebecois (PQ), 101, 149–152 Participation, 8, 26, 61, 81, 84, 94– 95, 100, 107, 127, 139, 148, 163, 167, 169, 178–179, 185, 188, 195, 197, 200 peace, 21, 23–25, 34, 84–85, 89, 111, 127, 130, 147, 158–162, 164–180, 183–186, 191–194, 196–201 peace talks, 84, 172, 176 peacebuilding, 10, 25, 84, 166, 188 peacekeeping, 166; peacekeepers, 172 Philippines, 13, 64, 93–94, 103–105, 107–109, 114, 121–122, 124, 181, 196 Popular Liberation Army, 176 populist movement, 63 proportional representation, 71–72 Provisional Irish Republican Army, 168–169 Puerto Princesa, 110–111 Putin, Vladimir, 124 Quebec, 8, 100–103, 121, 141–142,

206

Index

144–145, 148–152, 162–163, 188– 190, 194 quorum, 35, 39, 68, 80–82, 86, 90, 104, 113, 142

Radical Party, 69 Reagan, Ronald, 44 regimes, 6–10, 25, 31, 33–34, 38, 44, 112, 119, 122–127, 138–139, 183, 191 regulations, 3, 17, 58, 73, 80–81, 121, 133, 182 rejective referendum, 26 Republic of Karakalpakstan, 112–114 Revolutionary Armed Forces of Columbia (FARC), 84–85, 175–178, 180 right-sizing referenda, 8, 10 roll-off, 4 Royal Commission on the Electoral System, 74 Russian Federal Assembly, 157 Russian Federation, 14, 64, 156–157 Schwarzenegger, Arnold, 49, 61 Scotland, 1, 8, 141, 145–146, 148– 149, 152–155, 162, 187, 189–190, 194, 198 Scottish National Party (SNP), 152 self-determination, 22–23, 146, 151, 158–159, 172 self-governance, 141 semiauthoritarian, 119, 129 secession, 8, 34, 113, 117, 121, 141– 142, 146–150, 152–155, 158–163, 179, 184, 188–189, 191, 196 sharia law, 158 signature quorum, 81–82 Socrates, Victorino Dennis, 112 solidarity protests, 140 South Africa, 13, 16, 144, 168, 170, 178–179, 189, 193, 199 South Sudan, 145, 158–160, 164 socialism, 22–24 Soviet federation, 113 Soviet Republic, 122, 140, 164 stability, 6, 30, 51, 125–126, 129, 131, 138, 165, 167, 179, 183, 190 state breakdown, 10, 146, 148; hypothesis, 147, 157, 161–162 Sudan, 13, 141, 145, 149, 158–162, 164, 185–186, 189, 192, 199–201

Sudan People’s Liberation Army/Movement, 158 Súmate, 53–54 supranational, 95, 115, 120–121, 166, Supreme Council, 135–137, 155, 157 Supreme Court, 48, 53, 111, 135 Swiss, 4, 66, 94, 95, 97, 98, 99, 103, 114, 115, 183, 186, 190, 191, 196, 199 Switzerland, 11–13, 40, 63–65, 93–96, 98–99, 143, 181, 188, 194–195, 198, 200 Taiwan, 13, 38, 40, 43, 45, 56–58, 64, 189, 192 term limit, 107, 112, 132; presidential, 19, 33–34, 55, 128, 132 territorial, 12, 15, 100, 112 Thaksin Shinawatra, 129, 138, 196, Thailand, 1, 13, 123, 127, 129–131, 138, 183, 186, 190–191, 193, 195–196 Third Reich, 139 threshold, 8, 17, 25–26, 28, 44, 57, 59, 68, 73, 75, 77, 80–82, 95, 102, 104–105, 142, 156, 159–160 Trudeau, Pierre, 101, 150 Tsai, Alex, 57–58, 60, 188, 201 Tulip Revolution, 135–136 Turkey, 13, 120, 172–173, 180 Turkish Republic of Northern Cyprus (TRNC), 172–173

Uganda, 11, 14, 17, 32–35, 39–40, 43, 64, 124, 158, 160, 181, 188, 192, 195, 200 UN peacekeepers, 172 UN Secretary-General, 173 UN Security Council, 174 Unitary, 92, 94, 103, 112, 114, 182 union(s), 15–16, 34, 50, 69, 71, 73–74, 76, 78, 132 Unión Cívica initiative, 18 Union of Soviet Socialist Republics (USSR), 113, 122, 135 United Nations (UN), 56, 155, 166 United Wisconsin, 50 Uruguay, 11, 13, 17–21, 24, 34, 64 Uzbekistan, 13, 93–94, 112–114, 144, 163, 185, 187, 190, 192

Index Venezuela, 5, 7, 16, 38, 40, 43, 45, 52–55, 60, 123, 127, 131–133, 138, 140, 176, 181, 189, 191–192, 195 veto, 68, 95, 130, 174 Voters’ Registration Act, 111 voting, 2, 16, 19, 20, 80, 84, 94, 100– 101, 111, 113–114, 117–119, 127, 142, 148, 150, 154–156, 160, 162, 171, 177, 187, 189, 191, 194 Walker, Scott, 46, 50 Weimar Republic, 7, 90

Western Canada, 102 Western Europe, 5 Women’s Christian Temperance Union, 69, 74 World War II, 7, 98, 100 Wisconsin, 46, 48–51, 55, 60–61, 187, 191, 201 Yanukovych, Viktor, 155

Ziaur, Rahman (Zia), 29–31 Zimbabwe, 5 Zurich, 99

207

About the Book

hands of the people. But is it really as democratic as it seems? To what extent, and in what circumstances, is it less about citizen power and more about external influences seeking to manipulate outcomes? Addressing these issues, Shauna Reilly draws on and compares case studies of referenda, recall elections, and initiatives around the world to investigate the complex realities of direct democracy in action.

Direct democracy is typically lauded for putting power in the

Shauna Reilly is associate professor of political science at Northern

Kentucky University.

209