Democracy in Small States: Persisting Against All Odds (Oxford Studies in Democratization) 9780198796718, 0198796714

Citation preview

OUP CORRECTED PROOF – FINAL, 25/7/2018, SPi

D E M O C R A C Y IN S M A L L S T A T E S

...........

OUP CORRECTED PROOF – FINAL, 25/7/2018, SPi

OXFORD STUDIES IN DEMOCRATIZATION Series editor: Laurence Whitehead

........... Oxford Studies in Democratization is a series for scholars and students of comparative politics and related disciplines. Volumes concentrate on the comparative study of the democratization process that accompanied the decline and termination of the cold war. The geographical focus of the series is primarily Latin America, the Caribbean, Southern and Eastern Europe, and relevant experiences in Africa and Asia. OTHER BOOKS IN THE SERIES Democracy and Diversity: Political Engineering in the Asia-Pacific Benjamin Reilly Democratic Accountability in Latin America Edited by Scott Mainwaring and Christopher Welna Democratization: Theory and Experience Laurence Whitehead The Politics of Uncertainty: Sustaining and Subverting Electoral Authoritarianism Andreas Schedler The Politics of Accountability in Southeast Asia: The Dominance of Moral Ideologies Garry Rodan and Caroline Hughes Segmented Representation: Political Party Strategies in Unequal Democracies Juan Pablo Luna Europe in the New Middle East: Opportunity or Exclusion? Richard Youngs Turkey’s Difficult Journey to Democracy: Two Steps Forward, One Step Back İlter Turan Foreign Pressure and the Politics of Autocractic Survival Abel Escribà-Folch and Joseph Wright Legislative Institutions and Lawmaking in Latin America Edited by Eduardo Alemán and George Tsebelis The International Politics of Authoritarian Rule Oisín Tansey Building Trust and Democracy: Transitional Justice in Post-Communist Countries Cynthia M. Horne Coalitional Presidentialism in Comparative Perspective: Minority Presidents in Multiparty Systems Paul Chaisty, Nic Cheeseman, and Timothy J. Power

OUP CORRECTED PROOF – FINAL, 25/7/2018, SPi

Democracy in Small States Persisting Against All Odds ...........

JACK CORBETT AND WOUTER VEENENDAAL

1

OUP CORRECTED PROOF – FINAL, 25/7/2018, SPi

3

Great Clarendon Street, Oxford, OX2 6DP, United Kingdom Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide. Oxford is a registered trade mark of Oxford University Press in the UK and in certain other countries © Jack Corbett and Wouter Veenendaal 2018 The moral rights of the authors have been asserted First Edition published in 2018 Impression: 1 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, by licence or under terms agreed with the appropriate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above You must not circulate this work in any other form and you must impose this same condition on any acquirer Published in the United States of America by Oxford University Press 198 Madison Avenue, New York, NY 10016, United States of America British Library Cataloguing in Publication Data Data available Library of Congress Control Number: 2018938176 ISBN 978–0–19–879671–8 Printed and bound by CPI Group (UK) Ltd, Croydon, CR0 4YY Links to third party websites are provided by Oxford in good faith and for information only. Oxford disclaims any responsibility for the materials contained in any third party website referenced in this work.

OUP CORRECTED PROOF – FINAL, 25/7/2018, SPi

........... Acknowledgements ........... This book is the culmination of eight years’ work, individually and collectively. We have incurred many debts. The most significant are to the more than 250 interviewees from twenty-eight small states who freely and generously gave their time to explain how politics in their country works. We are continually impressed by how accessible and willing even the most senior political figures in small states are to outsider researchers. This book would not have been possible without these insights. Our work on small states has also been informed by many colleagues, friends, and acquaintances who have shared perceptions, provided feedback and helped us arrange interviews. There are too many people to list them all here—you know who you are! Many are acknowledged personally in other publications. While not formal interviews, these conversations have nevertheless greatly enriched this account. We would therefore like to take the opportunity to thank them all once again for their wisdom and encouragement. The research for this book has been assisted by three people in particular— Vanessa Newby, Daniel Devine, and Nienke van de Maat—who have helped us search through the mountains of secondary literature on small states and compile the statistics contained throughout. Vanessa in particular made a substantial contribution to Appendix 1, which includes a brief synopsis of all thirty-nine states covered in this book. We thank all three for being so diligent and meticulous. The book is much better for your contribution. A number of colleagues gave their scarce free time to read drafts of the manuscript. In particular, we would like to record our sincere thanks to Godfrey Baldacchino, John Connell, John Gerring, Lhawang Ugyel, and Viktor Valgardsson, whose comments have greatly strengthened our arguments and analysis. We are extremely fortunate to be a part of such a generous community of scholars who are passionate about what they do and encouraging of others’ research endeavours. This book has been written while we have worked at several institutions. When we started writing it, Jack was working at the Centre for Governance and Public Policy and the Griffith Asia Institute, Griffith University, and Wouter at the KITLV/Royal Netherlands Institute of Southeast Asian and Caribbean Studies. By the time we finished, Jack had moved to the University of Southampton, and Wouter to Leiden University. Jack also wrote the first draft of the introduction as a visiting fellow at the Sir Bernard Crick Centre for the Public Understanding of Politics at the University of Sheffield in 2015.

OUP CORRECTED PROOF – FINAL, 25/7/2018, SPi

vi

Acknowledgements

And we put the finishing touches on the manuscript while visiting the Islands and Small States Institute at the University of Malta in 2017. We would like to thank our colleagues at these institutions for their advice and support, especially during seminars at which we presented and discussed our work. In particular, we would like to thank Jason Sharman and Wesley Widmaier for their comments on our original book proposal. Jack also presented a version of this project at the Institute for Qualitative and Multi-Method Research, Syracuse University, in 2015 and would like to thank attendees who provided comments on the research design. Over the years, we have developed the arguments presented in this book in several papers written individually, together, and with other collaborators: Terence Wood, Lhawang Ugyel, Asenati Liki, Roannie Ng Shiu, John Connell, Jessica Vance Roitman, Gert Oostindie, Honorata Mazepus, Jan Erk, Ceridwen Spark, and Sebastian Wolf. We have cited these articles where appropriate but we want to thank each of these collaborators for informing our thinking. We would also like to thank the publishers of our previous work on small states for allowing us to reuse quotes and excerpts here. A full list of these citations is located in the Appendix 2. The research from this book would not have been possible without generous funding from the Netherlands Organisation for Scientific Research (NWO), project numbers 451.16.028 and 858.14.012, and the Australian Research Council, project number DP160100897. Since 2016, we have also been engaged as analysts for Freedom House’s annual Freedom in the World survey, compiling scores for European and Pacific small states. We found this exercise useful for understanding the datasets that we employ in the book. We extend particular thanks to Dominic Byatt, Olivia Wells, Kavya Ramu and Jenny Laing at Oxford University Press for their advice and guidance throughout the publication process. We would also like to record our gratitude to three anonymous peer reviewers for their generous and constructive comments that have helped sharpen our analysis. Paula Cowan helped compile the Index. Finally, the study of small states is a field whose merits are often lost on our nearest and dearest, Vanessa, Renée, Charlie, and Christopher, who have spent more time than they would like listening to us go on and on about what they consider to be a highly idiosyncratic topic. We hope that the fact that you have been able to visit a few of these countries with us alleviates the suffering. This book certainly would not have been written without you.

OUP CORRECTED PROOF – FINAL, 25/7/2018, SPi

........... Table of Contents ........... List of Tables and Maps

ix

1. The Small State Challenge

1

2. Democratization and Economic Development

23

3. Democratization and Cultural Diversity

46

4. Democratization and Institutional Design

67

5. Democratization and Political Parties

92

6. Democratization and Geography

119

7. Democratization and Small Size

144

8. Persisting Against All Odds

165

Appendix 1. Background on the Thirty-nine Small States

181

Appendix 2. Authors’ Previous Publications on Small States

209

References Index

213 243

OUP CORRECTED PROOF – FINAL, 25/7/2018, SPi

OUP CORRECTED PROOF – FINAL, 25/7/2018, SPi

........... List of Tables and Maps ........... Tables 1.1. Countries with populations below 1 million, clustered by region 2.1. Relationship between GDP per capita and Freedom House scores 3.1. Relationship between fractionalization scores and Freedom House scores 4.1. Relationship between colonial legacy and Freedom House scores 5.1. Relationship between electoral volatility and Freedom House scores 6.1. Relationship between islandness and Freedom House scores 6.2. Relationship between geographic location and Freedom House scores 7.1. Relationship between population and Freedom House scores

15 24 47 68 93 120 121 145

Maps A1.1. A1.2. A1.3. A1.4.

Africa and Asia The Caribbean Europe The Pacific

182 187 195 201

OUP CORRECTED PROOF – FINAL, 25/7/2018, SPi

OUP CORRECTED PROOF – FINAL, 26/7/2018, SPi

....................

1

....................

The Small State Challenge For at least half a century, scholars have grappled with the complex and dynamic process that we commonly call democratization. Over this period, numerous theories have been advanced to explain why democratic transitions occur and persist in some countries but not in others. Collectively, political science has found that democratic transition and consolidation is more likely to succeed in wealthy countries, with homogeneous populations, a British colonial legacy, stable and strongly institutionalized party systems, and those with particular geographical features (region and islandness). Despite the considerable resources and intellectual effort expended in pursuit of this scholarly and practical endeavour, small states, which account for roughly 20 per cent of the world’s countries, have been routinely overlooked. This group of thirty-nine states with populations of less than 1 million is statistically much more likely to be democratic than larger states (Diamond and Tsalik 1999; Ott 2000; Anckar 2002; 2010; Srebrnik 2004). But it includes numerous democracies that are low income countries, some of which have highly heterogeneous populations, no history of (British) colonialism, no institutionalized political parties, and are geographically located in the least democratic regions of the world. The divergence between mainstream theory and these routinely overlooked cases presents a significant challenge to the global study of democratization. Consider three examples. The African small island nation of São Tomé and Príncipe is one of the world’s most unlikely democracies. Democratization is positively correlated with economic growth, but São Tomé and Príncipe has a GDP per capita of around US$2,000, making it one of the poorest countries in Africa. Geographical location is another predictor of democratization but São Tomé and Príncipe is located in the Gulf of Guinea, and flanked by some of the most authoritarian states on the globe, including Cameroon, Equatorial Guinea, and Gabon. Colonial legacy is also often touted as an explanation of democratization but as one of Portugal’s five former African colonies, São Tomé and Príncipe’s colonial past appears to provide yet another barrier: some of the continent’s most brutal civil wars have occurred in Lusophone countries such as Angola and Mozambique; and Guinea Bissau still hosts one of Africa’s most oppressive regimes. Yet despite these economic, geographical,

OUP CORRECTED PROOF – FINAL, 26/7/2018, SPi

2

Democracy in Small States

and historical characteristics that theorists have identified as impediments to political liberalization, São Tomé and Príncipe has continuously held free and fair elections since the early 1990s, making it one of the most stable democracies in Africa. Similarly, in the Pacific region, only one of the eleven island nations (the Republic of Palau) has a GDP per capita higher than US$10,000. What’s more, most Pacific countries are dispersed across vast maritime zones (the Pacific Ocean accounts for one quarter of the earth’s surface) that present formidable challenges to economic development. In cultural terms, some Pacific island states are highly fragmented along ethnic, linguistic, and religious lines, another factor that is said to impede democratization. Vanuatu, for instance, has a population of just over 250,000 but more than 100 distinct language groups. In some other Pacific states politics functions without institutionalized political parties, something that most scholars since Schattschneider (1942) assume is virtually impossible. Against these odds, we might expect democracy to have struggled to take root. And yet, across the region democracy persists, elections are held, and the process of compromise and conciliation is largely maintained. Indeed, Pacific island states tend to receive very high scores from Freedom House; the one global ranking that includes all states at the time of writing. At the opposite end of this spectrum sits our third example, tiny Liechtenstein. Located in the heart of a continent in which democracy has thrived, it is one of the wealthiest countries in the world, with a GDP per capita of close to US$90,000. It has an extremely homogeneous population (German is the only official language and Catholicism is the main religion) and it maintains close relations with neighbouring Switzerland, a country commonly regarded as one of the purest and most exemplary democracies in the world. And yet, despite the presence of all these allegedly democracy stimulating factors, Liechtenstein has one of Europe’s most politically influential monarchies, which in recent decades extended its powers at the expense of parliament and government. These three examples are illustrative of a much broader trend that sees the world’s smallest states diverge from the established canon of democratization theories. They vary substantially along all factors commonly employed by scholars to explain why regimes rise and fall: they can be both very rich and very poor, have either the most ethnically homogeneous or diverse populations, and have extremely stable party systems, or no parties at all. Colonial legacy looms large in our story but some small states were never colonized. Many are located near powerful neighbours that influence their politics while others are among the most isolated countries on earth. As a result, despite their essential ‘smallness’ it is difficult to conceive of a more different group of countries than those we cover here.

OUP CORRECTED PROOF – FINAL, 26/7/2018, SPi

The Small State Challenge

3

The reason for the large number of exceptions is actually quite straightforward: democratization scholars have spent decades routinely overlooking the world’s smallest states (Veenendaal and Corbett 2015), as a result of which theories of democratization have been established exclusively on the basis of evidence from larger countries. We know that small states are more likely than large ones to score well in Freedom House rankings (Ott 2000; Anckar 2002; 2010; Srebrnik 2004). But, aside from Freedom House, the other major democracy datasets—Polity IV and The Economist’s Democracy Index— exclude many of them. As a result, virtually all scholars in the field of comparative politics and democratization have overlooked these cases and so almost everything that we think we know about democratic transition and consolidation suffers from an unstated gigantism. This book aims to rectify this omission by explaining how politics works the way it does in small states. The thirty-nine small states that we investigate are remarkably diverse. Freedom House ranks the majority as democratic but small states are also home to some of the world’s last remaining monarchies. Formal institutions are similarly varied. Westminster is the most common parliamentary setup but small states host the full spectrum of institutional designs, and some of them have political systems that have never been trialed elsewhere (Reilly 2002). At the same time, and regardless of each country’s individual Freedom House ranking, everyday political practices in small states can be remarkably similar. Indeed, when we look beneath the veneer of formal institutional design and the types of explanatory variables offered by democratization theory, we find that everyday politics in small states tends to be highly reliant on informal dynamics and the personal characteristics of key leaders and elites. Small states, it turns out, offer extreme examples of informal, personality-driven polities. The strong relationship between small country size and personalism is no accident. As we will show, politics could hardly be conducted otherwise in these settings. But, the way politics works in these micro cases does offer important lessons for scholars interested in the consequences of this type of politics on democratic performance. Examining these questions is important because personality-driven politics is said to be on the rise across the globe (Poguntke and Webb 2005; McAllister 2007), and so small states offer important insights for much larger questions. To bring hitherto understudied small states into the democratization canon, and to make sense of both the diversity and essential similarities common to the way politics works in these settings, this book makes two moves at once. The first is a theory-testing move. Because Freedom House designates that small states are more likely to be democratic, we start with democratization theory and examine the battery of existing explanations against the experience

OUP CORRECTED PROOF – FINAL, 26/7/2018, SPi

4

Democracy in Small States

of routinely overlooked small states. So, the question that animates our initial move is: why are small states more democratic than large ones? Contra received wisdom, we find that the standard theoretical explanations—economic growth; cultural diversity; colonial legacy and institutional design; the presence of an institutionalized party system; or geographic location—do not appear to have explanatory power in these settings. That is, they explain neither the democratic successes nor failures. By illustrating that small states are exceptional, we offer a powerful challenge to the existing precepts of democratization theory. While there have been several country or regionally specific studies of small state democracy, this is the first truly global analysis of politics and democracy in small states. The implications of this first move are of immense importance to questions about regime stability and democratic consolidation in particular. Our theorytesting exercise shows that decades of research that underpin a series of painstakingly compiled law-like generalizations about when democracy can survive fail the most basic empirical examination. This point is of immediate practical relevance to anybody interested in democracy promotion around the globe. By prescribing a series of necessary and sufficient preconditions that must be in place before democratization will succeed, existing theory has inadvertently served to limit the possibilities we assign to democracy.1 Our research shows that democratic government is actually far more resilient and adaptive than is commonly presumed. So, our first move offers a distinctly optimistic message for all of those who believe in the promise of representative government. To explain why small states appear to contradict conventional wisdom, the book then makes its second, theory-building move. We ask: how does domestic politics actually work in small states? Rather than relying on datasets such as Freedom House, which focus almost exclusively on the presence of formal institutions and rules, we look within these cases at the ways political authority is actually exercised. To do so we draw on archival material, secondary literature, and field research over an eight-year period in small states around the globe, including more than 250 interviews with political elites from twentyseven countries. We do not explicitly compare the qualitative practice of democracy in small states with larger ones—the literature offers more than enough evidence about the latter—but by looking within these unique cases we are able to tease out similarities and differences between countries and eras. In doing so

1 Transition theories, famously associated with Rustow (1970) but popular throughout the 1980s and 1990s (see O’Donnell et al. 1986; and Higley and Gunther 1992) are the important exception.

OUP CORRECTED PROOF – FINAL, 26/7/2018, SPi

The Small State Challenge

5

we make implicit comparisons between small and large states. In the conclusion, we tease out comparative lessons by considering the implications of what these hyper-personalistic cases can offer the global study of democracy. Despite the fact that democratic practices in small states make a mockery of any one set of necessary and sufficient preconditions, what emerges from our research is a description of how intensely informal, localized, and personalitydriven politics works in practice. We highlight a series of divergent outcomes. Small state politics tends to be highly fluid and fractious but at the same time personalism can create conformity and strong cultures of compliance, stifle pluralism, encourage dissenters to exile, and lead to informal forms of executive domination (Baldacchino 2012). This tendency for small-scale personalitydriven politics to have varying and seemingly contradictory effects has been extensively theorized, thus Dahl and Tufte (1973, 15), for example, highlight how personalism can paradoxically make leaders both more and less responsive. Our material provides real-world empirical examples of the dilemmas of small democracy in action. In doing so, our second move tempers the earlier theory-testing optimism; democracy in small states may demonstrate how resilient and adaptive this form of regime can be, but outcomes are far more contingent than is commonly presumed. The result is a nuanced and clear-eyed assessment of the strengths and limitations of highly personal and informal democracy.

W H Y S MA L L STATES MATTER Before we go into greater detail about the approach that has brought us to this point, it is important to specify exactly what is at stake in this discussion. Democracy in the twenty-first century appears to be confronted with an unenviable dilemma. On the one hand, over recent decades we have witnessed an explosion in the popularity of democratic norms and values around the globe (Huntington 1991; Keane 2009; Plattner 2015). Once deemed a radical and unstable form of government only suitable for small and self-contained Greek city states, representative democracy has come to embody regime legitimacy in the twenty-first century. Few are willing to challenge its stated premise—that governments should be selected from among the people and answerable to them. Autocrats hold elections in order to cloak their administrations in democratic legitimacy while revolutionary movements place principles such as equality and fairness in the distribution of power and resources at the heart of their manifestos. At present, all but two countries label themselves

OUP CORRECTED PROOF – FINAL, 26/7/2018, SPi

6

Democracy in Small States

as democracies, which if nothing else indicates how dominant this norm has become.2 On the other hand, particularly in those states where it is the longestablished mode of political decision-making, citizens appear to be deeply disaffected with how democracy is practised.3 Even putting the election of Donald Trump and the recent Brexit referendum to one side, the long-term trend across Europe and America sees voter turnout declining, political parties struggling to retain members, and professional politicians increasingly despised or deemed untrustworthy (Dalton 2004; Stoker 2006; Hay 2007; Flinders 2012; Mair 2013; Tormey 2015). Outside these regions, in new or transitioning democracies, democratization scholars fear that progress has stalled, and that in the absence of strong structural bulwarks against authoritarian rule the triumphant march of democratization may have halted (Zakaria 1997; Levitsky and Way 2002; Plattner 2015). The tendency for politics to become dominated by the personalities of its leaders rather than ideology or policy programme is a key theme in this discussion. Most scholars argue it is a rising feature of political life that produces negative effects (Poguntke and Webb 2005; McAllister 2007). In particular, personalism is assumed to negatively affect political representation, because the limited relevance of political ideologies and platforms reduces the accountability of politicians to their voters (cf. Mainwaring et al. 2006). In this view, democracy has arrived at a significant fork in the road. And, worryingly, we don’t appear to know which way to go. To forge a future for democracy we need to better understand why it has and has not worked in the past. Democratization scholars have long observed that small states—the most likely cases for personality-driven politics—are statistically much more stubbornly and disproportionately democratic than large ones (Diamond and Tsalik 1999; Ott 2000; Anckar 2002; 2010; Srebrnik 2004). But, few have gone further to investigate how politics is actually practiced in these countries. There is a variety of reasons for the marginal position of small states in the established canon, including their tiny populations, the fact that they are not considered ‘real’ states, their supposedly insignificant role in international politics, and the absence of data (Veenendaal and Corbett 2015). This book is based on the assumption that we have a lot to learn about how and why regimes survive, including democratic ones, from these cases. The rationale is twofold: 1) as a group they represent important examples of

2

These are Saudi Arabia and Vatican City. For the clearest statement of this argument see Runciman (2013) who builds on De Tocqueville’s discussion of the paradoxical nature of democratic rule: it appears messy and unpredictable but its inherent flexibility means it is much more likely to survive crisis than more organized but rigid regimes. 3

OUP CORRECTED PROOF – FINAL, 26/7/2018, SPi

The Small State Challenge

7

democratic success, at least according to Freedom House; and 2) because they have been overlooked for so long the intellectual payoff of studying them is likely to be enormous (Veenendaal and Corbett 2015; cf. Sharman 2016). For example, in Huntington’s (1991) classic The Third Wave, he famously observed that the democratic transition of some thirty countries in Latin America, Southern and Eastern Europe, Asia, and Africa between 1974 and 1990 resulted in a global increase in the number of democracies. However, in making this claim Huntington excluded countries with populations of less than one million (43). If he had included all available countries that became democratic between 1974 and 1990, the number of democratizing states would have risen from thirty-six to fifty-six, according to the Freedom House rankings. Initially, utilizing the full sample rather than just over half of the available dataset would have significantly bolstered Huntington’s claims about the historical significance of the Third Wave. The full impact of this omission is, however, best demonstrated by the way it has biased the subsequent Huntington-inspired literature on global democracy trends. At the dawn of the new millennium it became apparent that many of the states Huntington classified as Third Wave cases remained stuck in the ‘grey zone’ between democracy and authoritarianism (Carothers 2002, 9), and some even slid back to outright dictatorship (O’Donnell 1996; Zakaria 1997; Levitsky and Way 2002). The expected transition to liberal democracy did not materialize in many countries, and some scholars even suggested that a third ‘reverse wave’ might follow (Diamond 1996, 31). However, if the experience of the twenty excluded small states from that period were taken into account, this so-called ‘reversal’ is far less significant than these studies presume. In fact, with the exception of Fiji, which returned to democracy in 2014, but has experienced successive coups since the late 1980s, none of the Third Wave cases that Huntington excluded has reverted to authoritarianism. As this brief example shows, the consequence of including small states is a much more nuanced and clear-eyed assessment of the promises and limitations of different regime types. For democratization scholars, this analysis deepens our understanding of one of the field’s long-standing puzzles—why small states do so well in the Freedom House rankings. If democracy really is at the crossroads then we need to better understand its persistence in a range of settings, not just a few large and rich states. Moreover, to meet the standard conventions of case selection—representativeness and variation (Gerring 2007)—we argue that comparative scholars need to pay closer attention to small states. In doing so we gain a more holistic view of democratic practices, appreciate democracy’s great diversity and capacity to adapt, but also recognize its limitations and the reasons that underpin when and how it survives or fails.

OUP CORRECTED PROOF – FINAL, 26/7/2018, SPi

8

Democracy in Small States OUR APPROACH

To make our first theory-testing move, we examine the veracity of existing claims about transition and consolidation against the group of small states. We take Freedom House, whose definition rests on a version of proceduralism, most commonly associated with the work of Robert Dahl (1971), as our definitional starting point. According to Dahl, democracy (or what he called ‘polyarchy’4) can be conceptualized as a political system in which there is meaningful competition for political power (the dimension of ‘contestation’), and in which no major societal groups are excluded from participating in this competition (the dimension of ‘inclusiveness’). As a result, democracy is defined as a political system in which political contestation involves practically all adult citizens. In subsequent chapters, we explore whether the main theoretical explanations of democratic transition and consolidation are borne out in the thirty-nine small states we investigate. The variables we focus on include economic growth, cultural diversity, institutional design, party system institutionalization, geographic location, and population size. We find that these variables have limited explanatory power in regimes that Freedom House designates as either democratic or authoritarian. Having highlighted that small states tend to contradict established theory, we make our second move by turning our attention to explaining this divergence. To do so, we look within these cases to investigate how politics is practiced. Based on this description we build a theory of our own. What emerges from our intensive field research is the centrality of ‘personalism’ to small state politics. The literature on personalism and personalization is diverse, multifaceted, and dominated by studies of large states (Rahat and Sheafer 2007; McAllister 2007; Karvonen 2010). Historically it can be traced back to Max Weber’s (1978) description of ‘charismatic authority’ but today it encompasses work on the declining influence of mainstream parties and the parallel rise of radical right-wing populism in Europe (e.g. Albertazzi

In his 1971 book, Dahl refers to this system as ‘polyarchy’, as he wished to reserve the term ‘democracy’ for the ideal-type political system that is completely responsive to its citizens’ needs. According to Dahl, this system has not (yet) been realized anywhere. While Dahl’s definition of democracy appears rather minimal, he also listed the following eight requirements for a system to be labeled a democracy: 1) freedom to form and join organizations; 2) freedom of expression; 3) right to vote; 4) eligibility for public office; 5) right of politicians to compete for support and votes; 6) alternative sources of information; 7) free and fair elections; and 8) institutions for making government policies depend on votes and other expressions of preference. These eight conditions closely overlap with Freedom House’s conceptualizations of democracy and political freedom. 4

OUP CORRECTED PROOF – FINAL, 26/7/2018, SPi

The Small State Challenge

9

and McDonnell 2015; Moffitt and Tormey 2014; Mudde 2016), the increasing presidentialization of parliamentary systems (e.g. Heffernan 2003; Poguntke and Webb 2005; Dowding 2013), and the intensification of celebrity politics (e.g. Street 2004; 2012; Marsh et al. 2010). We posit that while personalism in small states shares many common features with large ones, it also has distinct characteristics. We therefore define small state personalism in the following terms: • Strong connections between individual leaders and constituents. Rather than being mediated by party systems, in small states voters and politicians have considerable opportunities for direct, personal contact (Dahl and Tufte 1973, 87; cf. Anckar 1999). This tendency is amplified by the overlapping private and professional roles that politicians undertake (Ott 2000; Corbett 2015a). Politicians are more than just legislators: they are family or clan members, friends, neighbours, or colleagues. • A limited private sphere. Contemporary democratic politics in large states is characterized by a distinction between public and private, with the institutions that define the former regulating conduct in the latter. In small states, the private sphere is dramatically reduced while the public sphere is expanded beyond the narrow confines of formal institutions. The result is a remarkably transparent political system but also one in which clear lines of accountability are blurred and concern with corruption is magnified. • The limited role of ideology and programmatic policy debate. Leaders are largely elected because of who they are rather than what they stand for. As a result, political contestation focuses on the qualities and characteristics of individual politicians rather than party manifestos (Richards 1982; Veenendaal 2013). Indeed, a number of Pacific Island states, such as Tuvalu or the Federated States of Micronesia, do not have political parties at all. • Strong political polarization. The absence of ideological difference should theoretically breed consensus but in fact, small state politics is often characterized by extreme polarization. Political competition between personalities is often fiercely antagonistic precisely because they have few ideological differences, and therefore politicians have to focus on personal disagreements to differentiate themselves (Richards 1982; Veenendaal 2013). In combination with the limited role of parties, this also potentially creates political instability, as political alliances are regularly broken. • The ubiquity of patronage. In small states nobody is faceless. Relatives and friends stick together in more visible and unavoidable ways. This leads to political dynasties and various other types of collusion. It also means that politicians in small states typically experience considerable pressure from

OUP CORRECTED PROOF – FINAL, 26/7/2018, SPi

10

Democracy in Small States constituents, who are often the same relatives and friends, to personally provide material largesse. Failure to do so can lead to electoral defeat. Patronage in the public sector is also common in small states, and public sector appointments are often made on the basis of political loyalties (Benedict 1967; Sutton 2007). • The capacity to dominate all aspects of public life. An expanded public sphere and the absence of specialist roles create opportunities for individuals to dominate politics in a manner that is virtually impossible in large states. Pluralism is uncomfortable and dissent is often stifled while dependent constituents can be easily bought off (Peters 1992; Baldacchino 2012).

We argue that this type of hyper-personalistic politics, which is ubiquitous in small states, underpins both the democracy-stimulating and repressive characteristics of political life. For example, the familiarity between citizens and politicians has often been regarded as democracy-enhancing, as it creates better opportunities for political representation and responsiveness (Dahl and Tufte 1973; Diamond and Tsalik 1999; Anckar 2002). In addition, this closeness can foster political awareness, efficacy, and participation among citizens of small settings because political decisions often have a direct impact on their personal lives (Anckar 2008). Moreover, the close connections between citizens and politicians provide a formidable obstacle to executive domination, as their extensive social connections prevent politicians from resorting to fullblown oppression or violence. But, it is clear that personalism simultaneously presents obstacles to liberal democratic development. The relative absence of ideology and the focus on political individuals undermines substantive representation, as politicians are not held accountable for their political actions. Patron–client linkages create social and economic dependency and unequal access to public resources (Peters 1992; Duncan and Hassell 2011) and strong polarization and personality clashes can breed political instability and turmoil, and in extreme cases undermine cooperation. Moreover, the opportunities for political leaders to accumulate untrammeled powers without the customary ‘checks and balances’ carries the risk of executive domination or dictatorial politics (Baldacchino 2012; Aiafi 2017). In such instances, the reasons why some countries take a more democratic path than others are largely the product of contingent combinations of key personalities and historical circumstances. Our empirical approach focuses on how competition for public office occurs in small states, how the authority of office is exercised, and how leaders interact with their constituents (and vice versa). So, we focus on the same types of issues and dynamics that Freedom House is interested in—free and fair

OUP CORRECTED PROOF – FINAL, 26/7/2018, SPi

The Small State Challenge

11

elections, the presence of free media, government responsiveness to the needs and preferences of citizens etc.—but, instead of simply highlighting their presence or absence, we examine their function and meaning to those who practice politics in small states. Our rationale for this in-depth qualitative approach is that, because formal institutions are often sidestepped in small states, regime categorization does not tell us a great deal about how politics actually works in these settings. Instead, we probe the informal aspects of politics in small states (Helmke and Levitsky 2004), including the importance of family networks and community allegiances, for instance, but also money politics and gift giving. We show that these types of activities, which are often considered deviant or corrupt in the political science literature, are in fact central to the way politics works in small states.

O U R C O MP A R A T I V E M E TH O D In-depth empirical investigation that both tests and builds theory across thirtynine countries is no easy task. It requires a method that allows us to combine both the depth of a single case study with the breadth of multi-country comparison. Our job is made easier by the fact that the size of these states ensures a degree of ‘family resemblance’ (Collier and Mahon 1993; Bevir and Kedar 2008; Boswell and Corbett 2017). We do not limit our analysis, either temporarily or spatially. Time matters because colonialism brought many of these states into being—especially in the Caribbean Sea and Indian Ocean—and shaped the types of institutions they adopted (Larmour 2005; Payne 1993). Similarly, geography is significant because many small states are also island nations, a detail which has important implications for the way political competition is described and understood (Anckar 2002; Congdon Fors 2014). To manage both the breadth and depth of our material, we favour a mode of comparative political studies built on the principle of abduction. The term ‘abduction’ is increasingly employed in comparative politics to describe a process of data analysis that involves moving back and forth between in-depth and context-rich material and abstract theory (Schwartz-Shea and Yanow 2013, 46–9; Boswell and Corbett 2017). That is, we take the puzzles and findings of different theories, demonstrated via large N comparisons, and contrast them with contextually detailed examples of the way politics is practised in small states. In doing so we move back and forth between the views and rationalizations of political elites in these settings about how and why they exercise authority, demonstrated via case studies and ‘thick descriptions’

OUP CORRECTED PROOF – FINAL, 26/7/2018, SPi

12

Democracy in Small States

(Geertz 1973) or ‘experience near’ (Jackson and Nexon 2013) analysis, and the explanations that predominate existing scholarship. This combination of contextually rich data and multiple theoretical lenses is a unique feature of this book. While explicitly comparative, this book does not conform to many of the established conventions of the discipline that have tended to privilege the analysis of formal constitutional powers—e.g. presidential, majoritarian parliamentary, and consensual parliamentary—and functions (e.g. Blondel 1985; Elgie 1995; Lijphart 1999). The study of formal institutions based on datasets such as that compiled by Freedom House has struggled to reach a consensus about why small states are more democratic than large ones (see for example Hadenius 1992 versus Anckar and Anckar 1995). In light of the fact that the majority of our cases have rarely received academic attention, we needed to develop an approach that allowed both theory-testing and theory-building in an exploratory manner (Gerring 2017). In addition, because it became increasingly apparent that the real story of politics in small states revolved around informal dynamics, we needed empirical tools that allowed us to look beyond the mere presence or absence of formal institutions. Like most comparative politics scholars, we are interested in patterns— commonalities and differences that ostensibly exist between the cases we are comparing—and an abductive approach allows us to probe when and why they are present or absent. Our questions were not resolved in individual interviews but rather findings emerged iteratively, by moving back and forth between general themes and individual stories and experiences (SchwartzShea and Yanow 2013, 46–9). Themes became meaningful when we found out that they occurred across different countries or varied considerably between them (Rhodes et al. 2009). This is the logic of the theory-testing approach we adopt as our first move: that theory provides us with a series of ‘plausible conjectures’ (Bourdon 1993) or heuristics (Abbott 2004) about democratic consolidation that ought to be repeated in small states. In this sense, we do not aim to provide law-like statements but rather to shed new light on prevailing theory. When we find similarities—that either support or challenge existing theory—we flesh them out and consider how much explanatory power each has in different contexts. Based on our abductively compiled description of democratic practices in small states, we then make a series of claims about how and why regimes rise, survive, and fall in these settings. For example, as outlined above, modernization theory predicts that Liechtenstein would be more democratic than São Tomé and Príncipe based on its wealth alone. But, by looking closely at political practices, we uncover a different story. Using a range of primarily qualitative data (discussed below), we then probe political practices in both settings to try and explain why this is the case. The result is a style of comparative politics that

OUP CORRECTED PROOF – FINAL, 26/7/2018, SPi

The Small State Challenge

13

is unashamedly iterative, anecdotal, and impressionistic (Rhodes et al. 2009; Rhodes 2011; Boswell and Corbett 2015; 2017). The advantage of this comparative but intimate approach is that it reveals a more holistic account than analysis that relies solely on the presence or absence of formal institutions. The book’s comparative approach is also reflected in its thematic structure and scope. There are a number of edited books on democracy in small states (Benedict 1967; Dommen and Hein 1985; Clarke and Payne 1987; Ott 2000; Patapan et al. 2005; Kisanga and Danchie 2007), but most are now dated and comprise collections of country studies with little substantive comparative analysis. In the Caribbean, scholars have concentrated on how Westminster traditions create extreme partisan division, sustain executive domination, and entrench one party rule (e.g. Ryan 1999; Barrow-Giles and Joseph 2006; Duncan and Woods 2007). In the Pacific, where modern representative institutions are newer and political parties less prominent, the emphasis has been on the absence of ideological cleavages (e.g. Rich et al. 2006), executive instability (e.g. Meller 1990; Reilly 2001) and the manipulation of ‘tradition’ by ruling elites (e.g. Lawson 1996; Fraenkel 2004). In both regions, there is a widely observed trend towards patronage, clientelism, and other forms of money politics (e.g. Duncan and Woods 2007; Duncan and Hassall 2011). The experience of the older European small states and small island nations in Africa and Asia is different again. That is why a truly global approach for studying small states is required. Three caveats are important before we go any further. First, from the outset it is important to be clear that we do not seek to replace each of the variables we test here with a single universally applicable explanatory variable. Not only would this be impossible given that we primarily focus our analysis on a subgroup of cases—small states—but, as will immediately become apparent in each of the chapters, this is a remarkably diverse group of cases where the consolidation of democratic transitions is contingent on a range of factors. Indeed, one of the arguments of this book is that in highly personalistic contexts agency matters (possibly more so than in larger states). As a result, we follow the recent call of some democratization scholars to explore combinations of causal factors and remain open to equifinality (for discussion see Haggard and Kaufman 2016). Second, while the starting point for both our quantitative and qualitative analysis is the definition of democracy contained in the Freedom House rankings, what will become immediately obvious is that this definition can only take us so far. In part, this is because the formal institutions central to the Freedom House survey are regularly bypassed in favour of informal political practices in small states. But also, as we would expect in such a diverse set of cases from five world regions, democracy means different things in different

OUP CORRECTED PROOF – FINAL, 26/7/2018, SPi

14

Democracy in Small States

contexts. In keeping with our abductive approach, rather than ignore or suppress these multiple meanings in favour of rigid institutional comparison, we seek to acknowledge them and explore how they shape political practices. At times, we therefore introduce a degree of fuzziness in our use of the concept. This is because in some instances we need to highlight the limitations of Freedom House for understanding how democracy is practiced in small states, while in others we need to highlight how the practice of politics in small states falls short of normative ideals. For scholars seeking a linear, law-like set of generalizations based on conceptual parsimony, this method is potentially disappointing. We prefer to see it as an attempt to bridge the growing divide between modes of political science based on precise statistical models that have little basis in everyday political practice, and thickly descriptive work that offers few theoretical lessons beyond the case under investigation. Third, having acknowledged that we favour a mode of comparative political science based on the principles of abduction, equifinality, and meaning ambiguity, we must also acknowledge that systemically examining each case in the type of depth more common to interpretive political science is not a feasible way to write this book. We outline the chapter structure in detail below, but it is important to note from the outset that we do not pretend to treat all countries equally nor do we test each theory against every case. Indeed, data constraints mean that we couldn’t even if we wanted too. Rather, we select ‘most’ and ‘least’ likely cases (Lijphart 1971) based on the independent variable we are investigating, and we therefore cluster our cases differently in each chapter. We also deliberately focus on the most theoretically salient. In the chapter on regime stability and cultural fragmentation, for example, we focus on those small states that are ethnically, culturally, or religiously heterogeneous on the grounds that they are ‘least likely’ cases. Similarly, in the chapter on economic growth, we group small states according to whether they are low, middle, or high income, and then probe how this has shaped political life. This purposeful selection ensures that we are able to marry both the depth and breadth of analysis we require in order to address our research questions.

C A SE S AND DATA The focus on small states is significant because their number has grown spectacularly over the last half century (Lake and O’Mahony 2004) so there are more of them than ever before. We choose to focus on all states with less than 1 million inhabitants because this is the group of cases that is commonly ignored in previous global studies of democratic transition and consolidation

OUP CORRECTED PROOF – FINAL, 26/7/2018, SPi

The Small State Challenge

15

Table 1.1. Countries with populations below 1 million, clustered by region Africa (5)

Asia (3)

Caribbean (11)

Europe (9)

Pacific (11)

Cape Verde

Bhutan

Antigua and Barbuda

Andorra

Federated States of Micronesia

Comoros

Brunei

Bahamas

Cyprus

Fiji

Djibouti

Maldives

Barbados

Iceland

Kiribati

São Tomé and Príncipe

Belize

Liechtenstein

Marshall Islands

Seychelles

Dominica Grenada Guyana St. Kitts and Nevis St. Lucia St. Vincent and the Grenadines Suriname

Luxembourg Malta Monaco Montenegro San Marino

Nauru Palau Samoa Solomon Islands Tonga Tuvalu Vanuatu

(see Table 1.1). We apply the scope condition of UN membership to determine what constitutes a ‘state’, and in doing so leave out non-sovereign jurisdictions (but see Aldrich and Connell 1998; Baldacchino 2010). The table above lists all of the thirty-nine countries we include. In Appendix 1, we have included a more detailed description of their economic, historical, geographical, and institutional characteristics. While recognizing that any population threshold is arbitrary, this group of small states is significantly more likely to be democratic; only 45 per cent of countries are classified as a democracy by Freedom House, whereas 72 per cent of the small states included here carry that designation. Since small states are largely absent from the major datasets, to answer our question we had to generate our own data. We start each chapter with a brief statistical analysis showing the (absence of) correlation between democracy and the independent variable under investigation in that chapter. In doing so, we seek to discover to what extent supposedly universal theories of democratization also apply to the subset of small states. Then, to build a picture of how politics is practised in small states, we draw on three main data sources: interviews; public documents and secondary literature; and participant observation.

Interviews We draw from the following datasets (for full citations see Appendix 2): • Veenendaal (2014): seventy-four interviews with political actors in San Marino, St. Kitts and Nevis, Seychelles, and Palau during 2010–11;

OUP CORRECTED PROOF – FINAL, 26/7/2018, SPi

16

Democracy in Small States • Corbett (2015): ninety-five interviews with political actors in Solomon Islands, Vanuatu, Fiji, Palau, Federated States of Micronesia, Marshall Islands, Kiribati, Nauru, Tuvalu, Tonga, Samoa during 2011–14; • Corbett and Ng Shiu (2014): eleven interviews with political actors in Samoa during 2013; • Veenendaal (2015): fourteen interviews with political actors in Liechtenstein during 2014; • Corbett (2018): seven interviews with political actors in Tuvalu during 2014. • Corbett, Xu, and Weller (ongoing): thirty-eight interviews with diplomats from small states including Antigua and Barbuda, Barbados, Bahamas, Brunei, Iceland, Malta, Cyprus, Maldives, Dominica, Guyana, and St. Kitts and Nevis, St Vincent and the Grenadines, Seychelles, Cape Verde, Kiribati, Solomon Islands, Tonga, and Vanuatu during 2012–17. • Veenendaal (ongoing): twenty-two interviews with political actors in Malta during 2017 and twenty-one interviews with political actors in Suriname during 2018.

Combined, this totals more than 250 interviews conducted in twenty-eight countries over the last seven years. We have spoken to the people who have the most intense day-to-day involvement in political practices: political elites (both government and opposition politicians, senior public servants, NGO (Non-Governmental Organizations) representatives, journalists and other civil society activists, donor consultants, local academics, and private sector representatives). Most interviews were recorded and we draw on quotes where relevant. In addition to formal interviews, we have had countless informal conversations that have added to our understanding of how politics is practised in these settings. The number of interviews and countries that we cover is one of the great strengths of this book. However, we acknowledge that we have not been able to conduct interviews everywhere. We overcome this by supplementing our interview material with the data sources discussed below. But, the limitation remains. In our defence, there are a lot of small states and they are often difficult and expensive fieldwork locations to access. The iterative nature of our fieldwork is reflected in our chosen method, and vice versa.

Public documents and secondary literature In addition to interviews, we have made extensive use of newspapers, social media, diaries, biographies, and autobiographies, records of parliamentary

OUP CORRECTED PROOF – FINAL, 26/7/2018, SPi

The Small State Challenge

17

debates, departmental and consultancy reports sourced online and from incountry archives. This material is complemented by an extensive survey of the area studies literature on each of the thirty-nine countries we investigate. Numerous scholars have undertaken individual country studies in small states but while this research often provides an extremely detailed documentary archive about how politics is practised in these settings, its authors tend to make few theoretical or comparative claims. Over the years we have collected an enormous amount of this material—often located in now defunct working paper series or the types of area studies journals that mainstream political scientists tend to overlook—and here we bring it to bear on the theoretical questions at hand. At the country level, this material is often patchy and idiosyncratic but when combined it provides a rich source of historical case material suitable for this study.

Participant observation While some aspects of democratic governance, such as cabinet meetings, are inaccessible, many of the ordinary day-to-day activities (parliament, political rallies, policy consultations etc.) can be observed. Doing so added to our understanding of how government is practised in small settings by providing a layer of detail not obtainable via interviews. Observing government elites is not the same as conventional anthropological ethnography (Rhodes 2011), which involves spending years in the field. Rather, we undertook ‘yo-yo ethnography’ (Wulff 2002; Rhodes 2011) that involved moving in and out of multiple field sites for intermittent periods timed to coincide with key events (elections, parliamentary sittings etc.). Again, we do not claim to have undertaken this type of work in every country we have visited let alone studied. But, when opportunities arose, we took them and our analysis has been enriched by this immersive experience.

Studying Elites Our empirical work largely focuses on the views and experience of political elites. This approach is pragmatic: • elites have the most intensive day-to-day involvement with democratic institutions and so if we want to know how they work, and the types of real-world challenges presented by small size, we have to ‘study up’ (Nader 1972) by asking them why they do what they do;

OUP CORRECTED PROOF – FINAL, 26/7/2018, SPi

18

Democracy in Small States • elites represent citizens in public forums and thus their statements focus community-wide tensions and attract media interest (e.g. newspaper editorials); • the size of small states means that there is greater proximity between ruler and ruled, and previous research on inclusiveness and representation suggests their views are deeply intertwined with citizen attitudes, and; • despite the often cited ‘attitudinal homogeneity’ (Anckar 1999) within small states, there is considerable diversity between political elites in these settings (they are not a unified group and critical opinion was forthcoming during interviews).

Relying on the views and experiences of elites has its limitations, as we would expect citizens to have a different experience of politics. But, given the number of understudied cases and the paucity of other data types—public opinion surveys for example—it was the most appropriate method in this instance. Historical case studies and observation-based research allowed us to compare elite reflections with actual instances, providing greater confidence in our findings. However, to some extent this limitation remains. We acknowledge that our empirical focus on elites has shaped the way we understand their role in the political process; the answers we get always reflect the questions we ask and of whom we ask them. But, endogenous limitations aside, we hope that the insights we have gained by doing so shed new light on significant questions, while also inspiring more scholars to undertake empirical work of their own in small states.

On Being Non-Resident Researchers The final methodological point to acknowledge is that while we have both spent a considerable amount of time in different small states over the years, for most of our lives we have been residents of much larger states (primarily Australia and the Netherlands). This has strengths and limitations for a project like this. The main strengths we see is that it informed our interest in comparison between countries and regions—many researchers from small states tend to focus on their own and neighbouring countries. It also gave us a point of implicit comparison between the countries we were visiting and those we call home. For example, Veenendaal, as a citizen of the famously consociational Dutch democracy, found the winner-takes-all approach to politics in small states markedly different to the Dutch system of compromise and conciliation. Corbett, by contrast, found the differences between the majoritarianism of

OUP CORRECTED PROOF – FINAL, 26/7/2018, SPi

The Small State Challenge

19

Westminster in Australia and small states in the Caribbean and Pacific to be subtler. This variation in our backgrounds and the implicit comparison it evoked is embedded throughout much of this book. The obvious limitation, particularly when engaging in the type of qualitative work that we rely on here, is that we were ‘outsiders’ in every country we visited. As outsiders, we will have missed or misunderstood important local cues and subtleties in language and meaning (for discussion see Corbett 2014). In our defence, any researcher undertaking a comparative study based on the number of countries covered here will be an outsider in most of them. But, we are conscious that the lived experience of the themes we discuss (i.e. patronage) might feel different to residents than visitors. Whether or not we have under- or over-compensated for this perception in our analysis is for the reader to judge. What we can say for sure is that being outsiders influenced the amount and level of our access to key decision makers in each of the datasets we draw on, although, again, not always to our detriment as many interviewees confessed to feeling able to speak more freely given that we had no personal stake in the domestic politics of the places we were visiting. Ultimately, there is little we can do to alter the endogenous strengths and limitations that we bring to this study individually, and as a team. But, by acknowledging the ways in which we believe our position within the social and political field of study has shaped our findings, we hope that the reader can more easily judge for themselves the merits of our assessments.

C H A P TE R OUTLINE This book is organized thematically with each chapter tackling one of the main theories from the democratization canon. In each, we bring different types of data—case studies and other documentary evidence, interviews, and observation—to bear on the question at hand. But, following our abductive approach, in addition to examining the veracity of existing theory, we also use each chapter to build our explanation of how democracy is practised in small states. The concluding chapter brings this story together and outlines the lessons that it offers ongoing debates about regime persistence. In Chapter 2, we tackle modernization theory and the belief that economic growth is the key factor behind why democratic regimes rise and fall. We find that many small states in Africa, the Caribbean, but most significantly the Pacific, are poor but retain high Freedom House scores. In contrast, some small states in Europe in particular, but also Brunei in Asia, tend to be both

OUP CORRECTED PROOF – FINAL, 26/7/2018, SPi

20

Democracy in Small States

richer but poorly scored. This allows us to show that while economic performance is clearly relevant to the survival of regimes, the key link is not how wealthy a country is but how elites both utilize their economic resources and narrate the story of their performance. In poorer states, elites work to keep public expectations low while personally rewarding loyal followers with largesse. In wealthy states, elites link high living standards with regime stability and centralized authority. The lesson is that personalistic politics can have unexpected benefits for democratization, especially in small, poorer countries. Chapter 3 interrogates the argument that cultural homogeneity is a prerequisite for democratic persistence. The thesis here is that the absence of diverse interests and agreement around cultural norms produces a unified citizenry. This is supported by the view that democracy is harder to sustain in ethnically and religiously diverse societies. The problem is that many small states are both ethnically, socially, and linguistically diverse and stubbornly democratic. Indeed, in some cases, such as the Melanesian region of the Pacific, it has been argued that hyper-fragmentation actually aides consolidation by ensuring that no one group can ever come to dominate the apparatus of the state (Reilly 2000). Conversely, many homogeneous small states have dominant cultural codes that stifle pluralism and dissent. So, homogeneity and the type of ‘thick’ community theory advocates is not a perquisite for democracy any more than economic growth is. In Chapter 4, we interrogate the claim that colonial legacy and institutional design shapes the success and failure of democratization. We find that 1) institutions designed for large states do not always work, or work as expected, in small ones; 2) that this has led to considerable reform and adaptation, especially in the Pacific; and 3) despite attempts to reform institutions, in many small states actors find it easier to circumvent them entirely. That is, institutional design is of limited value when seeking to understand how democracy is practised in small states. This propensity to sidestep or short circuit institutions can help us explain why politics in small states is often very similar, despite variations in institutional design and colonial legacy. The adaptation theme is also a strong feature of Chapter 5, where we examine the old idea that democracy is unworkable without political parties, and that an institutionalized party system is a prerequisite for a stable democracy. Some small states have institutionalized party systems and others do not. We find that, with few exceptions, the qualitative experience of countries without robust party systems is not decisively different in those small states that do have strong parties as, in all cases, personalistic rather than programmatic representation is the norm. We conclude therefore that political parties and party systems are not decisive to the democratization prospects of small states.

OUP CORRECTED PROOF – FINAL, 26/7/2018, SPi

The Small State Challenge

21

Geographic location and the democracy-stimulating effects of insularity are said to strongly affect democratic development. This argument is buttressed by the observation that over 75 per cent of democratic small states are also island nations. Conversely, consolidation is said to be more likely to succeed in countries with democratic neighbours by means of so-called snowball or demonstration effects. We problematize both competing arguments in Chapter 6. First, even though small states are extremely susceptible to external and regional influences, we show that this argument does not work for (partially) authoritarian small states in ‘democratic’ Europe (i.e. Liechtenstein and Monaco), but also not for the fully democratic small states in ‘undemocratic’ Africa (i.e. Cape Verde and São Tomé and Príncipe). Second, we find that insularity can pose formidable obstructions to democratic performance, especially in remote archipelagic states such as Comoros, Kiribati, Maldives, FSM (Federal States of Micronesia), Marshall Islands, and Seychelles. Relations between various islands united in one political unit are often strongly antagonistic, posing a threat to political stability and democracy. We argue that the way elites frame external influence is the key factor in this discussion, with the discourse of vulnerability acting as a powerful means of obtaining internal acquiescence. Finally, having challenged all available alternatives, in Chapter 7 we turn to the final explanation for the existence and persistence of democracy; that small states tend to be more democratic because they are small. This view has deep roots in antiquity and the famously democratic Greek city states. Our study of democratic practices challenges the perennial ‘small is beautiful’ claim. Small states are often remarkably transparent—everybody knows everybody, or knows someone who does—hence accusations of nepotism and political patronage are rife. Accountability is strong among families but is undermined in government by weak media and civil society organizations, and by the tendency for politics to be conducted outside of the formal public sphere. Corruption is said to be on the rise but social proximity makes achieving clear lines of accountability even more problematic. The World Bank and other international donors classify some small states as ‘failed’ or ‘failing’. So, small can mean democratic but, as we saw in Chapter 6, it can often mean illiberal as well. The benefits of small size are therefore much more mixed than the ‘small is beautiful’ literature presumes. In the final chapter, we pull together the examples infused throughout the body of the book to show that, despite the odds, political regimes in small states, the vast majority of which are democratic, can be both extremely adaptive and resilient. A close examination of political practices allows us to demonstrate how personalism is the key to explaining both the successes and failures. This argument has particular relevance in small states where social proximity is greater. However, it also has wider implications for scholars

OUP CORRECTED PROOF – FINAL, 26/7/2018, SPi

22

Democracy in Small States

of democratic consolidation in particular where the recent tendency has been to downplay the role of intentional agents in their explanations. Most importantly, it provides a nuanced and more clear-eyed analysis of the strengths and limitations of personality-driven politics, a phenomenon of growing importance to democracies across the globe. After all, by studying politics in small states, we learn something about large states, too.

OUP CORRECTED PROOF – FINAL, 26/7/2018, SPi

....................

2

....................

Democratization and Economic Development Economic development is the most widely cited explanatory factor for sustained democratic transitions (see Lipset 1959; Przeworski et al. 1996; Diamond 1999; Pinkney 2003). The argument has a number of component parts that we recap below but at its core is a belief that industrialization and the corresponding rise of an urban middle class is the key to both transition and consolidation as it creates the climate in which informed public debate can flourish. As a result, democracy, scholars have concluded, is predominantly a wealthy countries’ game. There is some disagreement about the causal sequence that underpins this argument: Lipset, Pinkney, and Diamond largely follow the older modernization theory view that economic growth causes democratization, whereas Przeworski and his coauthors maintain that economic development prevents reversals rather than being the primary catalyst for transitions (for discussion see Boix and Stokes 2003; Epstein et al. 2006). But, regardless of the causal sequence, the mainstream view is that the two variables are mutually dependent. The problem with this argument is that it doesn’t match the experience of the world’s smallest states. To illustrate the relationship between wealth and democracy, we ran an ordered logistic regression, the results of which have been presented in Table 2.1. Freedom House separates countries into three categories: not free, partly free and free (here, coded 1–3). GDP per capita was log-transformed as is common with tests using GDP (the substantive results are the same without using the log transformation). The results show that GDP per capita is a significant predictor of the level of democracy, except in small states, where there is no relationship at all. For non-small states, wealth increases the likelihood of being partly free or free. There is no significant relationship whatsoever between wealth and the type of political regime in the world’s smallest states, and as a result we must conclude that richer small states are by no means more likely to be democracies than very poor small states. Or, to put it another way, the economic growth thesis apparently only applies to large countries.

OUP CORRECTED PROOF – FINAL, 26/7/2018, SPi

24

Democracy in Small States Table 2.1. Relationship between GDP per capita and Freedom House scores VARIABLES GDP Capita (logged)

(All States) FH Category

(Large States) FH Category

(Small States) FH Category

0.627*** (0.121)

0.704*** (0.137)

0.174 (0.317)

193

154

39

Observations Standard errors in parentheses ***p