Democracy in Divided Societies: Electoral Engineering for Conflict Management (Theories of Institutional Design)

Reilly analyzes the design of electoral systems for divided societies, examining various divided societies which utilize "vote-pooling" electoral systems--including Papua New Guinea, Sri Lanka, Northern Ireland and Fiji. He shows that political institutions which encourage the development of broad-based, aggregative political parties and where campaigning politicians have incentives to attract votes from a range of ethnic groups can, under certain conditions, encourage a moderate, accommodatory political competition and thus influence the trajectory of democratization in transitional states. This is a challenge to orthodox approaches to democracy and conflict management. Cover 1 Half-title 3 Series-title 4 Title 5 Copyright 6 Dedication 7 Contents 9 Illustrations 11 Figure 11 Maps 11 Tables 12 Acknowledgements 13 Abbreviations 15 1 Introduction: democracy in divided societies 17 Democracy, ethnicity and ethnic conflict 19 Institutional engineering 21 The theory of centripetalism 23 Electoral engineering for conflict management 28 Types of electoral systems 30 Plurality-majority systems 31 Semi-proportional systems 33 Proportional representation (PR) systems 33 Preferential voting 34 Centripetalism versus consociationalism 36 Conclusion 40 2 The historical development of preferential voting 43 From run-off elections to preference voting 44 The single transferable vote 47 The alternative vote 49 The supplementary vote 52 Conclusion 57 3 Centripetal incentives and political engineering in Australia 58 Introduction 58 Centripetal politics in Australia 60 Preferential elections in Australia 62 Case study: the 1990 federal election 66 Wider impacts 69 Conclusion 72 4 The rise and fall of centripetalism in Papua New Guinea 74 Democracy and ethnic conflict in Papua New Guinea 76 Ethnicity and elections 78 The alternative vote in Papua New Guinea 84 Case study: Kaindi and Henganofi electorates, 1964 87 Case study: Dei electorate, 1972 91 First-past-the-post in Papua New Guinea 97 Electoral violence 102 The reintroduction of preferential voting? 105 Conclusion 108 5 Electoral engineering and conflict management in divided societies (I): Fiji and Sri Lanka compared 111 The Fiji Islands 112 Case study: the 1999 Fijian elections 121 Lessons from Fiji 124 Sri Lanka 128 Sri Lanka's electoral arrangements 131 Case study: four presidential elections in Sri Lanka 136 Conclusion 140 6 Electoral engineering and conflict management in divided societies (II): Northern Ireland, Estonia and beyond 145 STV and the politics of accommodation in Estonia and Northern Ireland 146 Estonia 148 Northern Ireland 150 Case study: The 1998 Northern Ireland Assembly election 153 Other experiments: preferential voting in Europe and North America 157 AV in the United Kingdom 157 AV in Bosnia-Herzegovina 159 Preferential voting in the United States and Canada 160 Conclusion 162 7 Technical variations and the theory of preference voting 165 Introduction 165 Single-member versus multi-member districts 167 Optional versus compulsory preference marking 171 Theoretical pros and cons 174 Conclusion 180 8 Conclusion: assessing the evidence 183 Incentives for moderation 187 Iteration 188 Fairness 190 Mass versus elite activity 192 Party system development 194 Coalitions 195 Centralism 196 Bargaining and cooperation 198 Psychological versus mechanical factors 199 The significance of group demographics 201 Conclusion 208 References 210 Newspapers 230 Index 231 Democracy is inherently difficult in societies divided along deep ethnic cleavages. Elections in such societies will often encourage 'centrifugal' politics which reward extremist ethnic appeals, zero-sum political behaviour and ethnic conflict, and which consequently often lead to the breakdown of democracy. Reilly examines the potential of 'electoral engineering' as a mechanism of conflict management in divided societies. He focuses on the little-known experience of a number of divided societies which have used preferential, vote-pooling electoral systems - such as Papua New Guinea, Sri Lanka, Northern Ireland and Fiji. Examination of these cases shows that electoral systems which encourage bargaining between rival political actors, which promote the development of broad-based, aggregative political parties and which present campaigning politicians with incentives to attract votes from a range of ethnic groups can, under certain conditions, encourage the development of moderate, accommodatory political competition in divided societies This Text Examines The Potential Of Electoral Engineering As A Mechanism Of Conflict Management In Divided Societies. It Focuses On The Little-known Experience Of A Number Of Divided Societies Which Have Used Vote-pooling Electoral Systems. Introduction: Democracy In Divided Societies -- The Historical Development Of Preferential Voting -- Centripetal Incentives And Political Engineering In Australia -- The Rise And Fall Of Centripetalism In Papua New Guinea -- Electoral Engineering And Conflict Management In Divided Societies 1: Fiji And Sri Lanka Compared -- Electoral Engineering And Conflict Management In Divided Societies 2: Northern Ireland, Estonia And Beyond -- Technical Variations And The Theory Of Preference Voting -- Conclusion: Assessing The Evidence. Benjamin Reilly. Originally Presented As The Author's Thesis (doctoral--australian National University). Includes Bibliographical References (p. 194-214) And Index. Reilly analyses the design of electoral systems for divided societies, examining various divided societies which utilise 'vote-pooling' systems - including Sri Lanka, Northern Ireland and Fiji. Political institutions which encourage broad-based, aggregative political parties can, under certain conditions, encourage moderate, accommodatory political competition - influencing the trajectory of democratization in transitional states. The question of whether, and how, democracy can survive in divided societies has long been a source of controversy in political science.