Democratic Dilemmas: Why democracies ban political parties [1 ed.] 1138898015, 9781138898011, 9781315708836

Citation preview

Democratic Dilemmas

This book examines how democratic communities resolve dilemmas posed by anti-­system parties or, more specifically, the question of why democracies take the grave decision to ban political parties. On the one hand, party bans may ‘protect’ democracies, usually from groups deemed to undermine the democratic system or its core values, territorial integrity or state security. At the same time, banning parties challenges foundational democratic commitments to political pluralism, tolerance and rights to free speech and association. The book probes the deliberative processes, discursive strategies and power politics employed when democratic communities negotiate this dilemma. It examines discourses of securitization and desecuritization, preferences of veto-­players, anti-­system party orientations to violence, electoral systems and the cordon sanitaire as alternatives to party bans, and incentives for mainstream parties to cooperate, rather than ban, parties to achieve office and policy goals. It does so with reference to case studies of party bans, legalizations and failed ban cases in Spain (Herri Batasuna and successors), the United Kingdom (Sinn Féin and Republican Clubs) and Germany (Socialist Reich Party and National Democratic Party of Germany). Angela K. Bourne is an associate professor at the University of Roskilde, Denmark.

Routledge Studies in Extremism and Democracy

Series Editors: Roger Eatwell University of Bath and Matthew Goodwin, University of Kent. Founding Series Editors: Roger Eatwell, University of Bath and Cas Mudde, University of Antwerp-­UFSIA.

This new series encompasses academic studies within the broad fields of ‘extremism’ and ‘democracy’. These topics have traditionally been considered largely in isolation by academics. A key focus of the series, therefore, is the (inter-)relation between extremism and democracy. Works will seek to answer questions such as to what extent ‘extremist’ groups pose a major threat to democratic parties, or how democracy can respond to extremism without undermining its own democratic credentials. The books encompass two strands: Routledge Studies in Extremism and Democracy includes books with an introductory and broad focus which are aimed at students and teachers. These books will be available in hardback and paperback. Titles include: The Populist Radical Reader A Reader Edited by Cas Mudde The Far Right in America Cas Mudde Routledge Research in Extremism and Democracy offers a forum for innovative new research intended for a more specialist readership. These books will be in hardback only. Titles include: 38 The Darkest Sides of Politics, II State Terrorism, “Weapons of Mass Destruction,” Religious Extremism, and Organized Crime Jeffrey M. Bale 39 Democratic Dilemmas Why Democracies Ban Political Parties Angela K. Bourne 40 When Does Terrorism Work? Diego Muro www.routledge.com/politics/series/ED

Democratic Dilemmas

Why Democracies Ban Political Parties

Angela K. Bourne

First published 2018 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN and by Routledge 711 Third Avenue, New York, NY 10017 Routledge is an imprint of the Taylor & Francis Group, an informa business © 2019 Angela Bourne The right of Angela Bourne to be identified as author of this work has been asserted by her in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. British Library Cataloguing-­in-Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging-­in-Publication Data Names: Bourne, Angela K., author. Title: Democratic dilemmas : why democracies ban political parties / Angela K. Bourne. Description: Abingdon, Oxon ; New York, NY : Routledge, 2018. | Series: Routledge studies in extremism and democracy | Includes bibliographical references and index. Identifiers: LCCN 2018009398| ISBN 9781138898011 (hbk) | ISBN 9781315708836 (ebk) Subjects: LCSH: Political parties–Law and legislation. | Representative government and representation. | Democracy. Classification: LCC K3270 .B68 2018 | DDC 324.2/04–dc23 LC record available at https://lccn.loc.gov/2018009398 ISBN: 978-1-138-89801-1 (hbk) ISBN: 978-1-315-70883-6 (ebk) Typeset in Times New Roman by Wearset Ltd, Boldon, Tyne and Wear

To Anthony and Samuel

Contents



List of figures List of tables List of abbreviations



Introduction: pluralism, tolerance and the proscription of political parties

1

Theory and research design: from ‘militant democracy’ to security discourse, institutions and democratic competition

15

2

Ban regimes and banned parties in Spain, the United Kingdom and Germany

40

3

Herri Batasuna and its successors: from ‘terrorist threat’ to carrier of a ‘Copernican shift’ from violence

54

4

Sinn Féin and Republican Clubs: from terrorist pariah to republicanism’s pathway out of violence

84

5

The Socialist Reich Party and National Democratic Party of Germany: from ‘militant democracy’ to declining anti-­system party threats

114

6

Electoral systems as an alternative to party bans

143

7

Party bans and government formation

175



Conclusion

203

ix x xii

1

viii   Contents

Appendix 1: main features of electoral systems in Spain, the United Kingdom and Germany Appendix 2: government formation in Spain, the United Kingdom and Germany References Index

219 226 239 264

Figures

3.1 Mortal victims of ETA, 1968–2010 3.2 ETA attacks (other than assassinations), 1970–2010 4.1 Number of deaths in Northern Ireland, Great Britain and Ireland due to conflict in NI 4.2 IRA attacks and assassinations, 1970–2011 4.3 Official IRA attacks, 1971–79 5.1 Politically motivated right-­wing crimes and extreme-­right crimes in Germany, 1990–2016 5.2 Violent crimes attributed to extreme right in Germany, 1990–2016 5.3 Murders and attempted murders attributed to extreme right in Germany, 1990–2016

63 63 102 103 103 129 130 130

Tables

 I.1 Parties subject to ban proceedings in Europe, 1945–2015 1.1 Summary of party ban cases and case selection criteria 1.2 Alternatives to party bans, objectives pursued and indicators of effectiveness 2.1 Failed party bans, illegalization and legalization of parties in Spain 2.2 Proscription and legalization of parties in Northern Ireland and the United Kingdom 2.3 Proscription of parties and failed party bans in Germany 6.1 Average electoral system disproportionality at state and regional levels in the UK, Spain and Germany 6.2 Electoral system effects: HB and successors, Congress of Deputies, Basque and Navarrese parliaments, 1977–2015 6.3 Electoral system effects: Sinn Féin in the UK House of Commons (1950–2015), Northern Ireland Parliament (1922–72) and Northern Ireland Assemblies (1982, 1996, 1998–2017) 6.4 Electoral system effects: Republican Clubs/The Workers Party in the UK House of Commons (1974–2015) and Northern Ireland Assemblies (1982, 1996, 1998–2017) 6.5 Electoral system effects: Socialist Reich Party (1950–52) 6.6 Electoral system effects: National Democratic Party of Germany, Bundestag (1965–2017) and Länder (1990–2016) 6.7 Effectiveness of legal thresholds in marginalizing the NPD, 1990–2017 6.8 Summary of electoral system effects on marginalization of anti-­system parties  7.1 Ranking of Herri Batasuna and successors in Penrose-­Banzaf Index following elections for Spanish and Basque parliaments, 1993–2016 7.2 Minimum winning coalitions and minimal connected winning coalitions involving Herri Batasuna and successors in Spanish and Basque parliaments, 1993–2016

4 20 32 44 48 52 146 154

157 160 167 168 171 173 184 186

Tables   xi    7.3 Ranking of SRP in Penrose-­Banzaf Index following elections in Bremen and Lower Saxony, 1951    7.4 Ranking of NPD in Penrose-­Banzaf Index following elections in Saxony, 2004 and 2009, and Mecklenburg-­Vorpommern, 2006 and 2011    7.5 Electoral systems effects, cordon sanitaire and the marginalization of HB and its successors, Sinn Féin, Republican Clubs, SRP and NPD    7.6 Partisan party-­ban-veto-­players and coalition potential of banned, legalized and subjects of failed party ban proceedings   C.1 Findings of party ban cases (expectation of literal replication)   C.2 Findings on legalization of banned party cases (expected theoretical replication)   C.3 Findings on failed party ban cases (expected theoretical replication) A1.1 Electoral systems in Spain, Basque Country and Navarre A1.2 Electoral systems in the United Kingdom and Northern Ireland A1.3 Electoral system in German Bundestag A1.4 Electoral systems in Länder where the Socialist Reich Party contested elections, 1950–52 A1.5 Electoral systems at Land level where the National Democratic Party of Germany contested elections, 1990–2017 A2.1 Governments formed in Spain, Basque Country and Navarre, 1977–2017 A2.2 HB, its successors, and coalition formation A2.3 Governments formed in the UK and Northern Ireland, 1945–2017 A2.4 Governments formed in Bremen and Lower Saxony, 1951, after SRP electoral breakthrough and in Saxony, 2004–14, and Mecklenburg-­Vorpommern, 2006–16, after NPD electoral breakthrough A2.5 SRP and coalition formation A2.6 NPD and coalition formation

194 195 199 201 205 210 214 219 220 221 222 223 226 228 234

236 237 238

Abbreviations

Alternative für Deutschland (Alternative for Germany) Asociación Víctimas del Terrorismo (Association of Victims of Terrorism) BMI Bundesministerium des Innern (Federal Interior Ministry) BNG Bloque Nacionalista Galego (Galician Nationalist Bloc) CC Coalición Canaria (Canarian Coalition) CCa PCN Coalición Canaria – Partido Nacionalista Canario (Canarian Coalition – Canarian Nationalist Party) CDC Convergència Democràtica de Catalunya (Democratic Convergence of Catalonia) CDU Christlich Demokratische Union (Christian Democratic Union) CiU Convergència i Unió (Convergence and Union) CP Coalición Popular (Popular Coalition) Cs Ciudadanos (Citizens) CSU Christlich-­Soziale Union (Christian Social Union) DB Deutscher Bundestag (German Federal Parliament) DiL Democràcia i Llibertat (Democracy and Freedom) DP Deutsche Partei (German Party) DRP Deutsche Reichs-­Partei (German Reich Party) DVU Deutsche Volksunion (German People’s Union) EA Eusko Alkartasuna (Basque Solidarity) EAE/ANV Eusko Abertzale Ekintza/Acción Nacionalista Vasca (Basque National Action) EAJ/PNV Euzko Alderdi Jeltzale/Partido Nacionalista Vasco (Basque Nationalist Party) EB-­IU Ezker Batua-­Izquierda Unida (United Left) EBB-­IU Ezker Batua–Berdeak (United Left–Greens) ECHR European Convention of Human Rights ECtHR European Court of Human Rights EE Euskadiko Ezkerra (Basque Country Left) EH Euskal Herritarrok (Basque Citizens) EH Bildu Euskal Herria Bildu (Basque Country Unite) EHAK/PCTV Euskal Herrialdeetako Alderdi Komunista/Partido Comunista de las Tierras Vascas (Communist Party of the Basque Homelands) AfD AVT

Abbreviations   xiii EPA ERC

Northern Ireland (Emergency Provisions Acts) Esquerra Republicana de Catalunya (Republican Left of Catalonia) ERC CatSí Esquerra Republicana de Catalunya–Catalunya Sí (Republican Left of Catalonia–Catalonia Yes), ETA Euskadi Ta Askatasuna (Basque Homeland and Liberty) FAP Freiheitliche Deutsche ArbeiterPartei (Free German Workers Party) FCC Federal Constitutional Court FDP Freie Demokratische Partei (Free Democratic Party) Greens Bündnis 90/Die Grünen (Alliance 90/The Greens) HASI Herriko Alderdi Sozialista Iraultzailea (People’s Socialist Revolutionary Party) HB Herri Batasuna (Popular Unity) INLA Irish National Liberation Army IU Izquierda Unida (United left) IRA Irish Republican Army KAS Koordinadora Abertzale Sozialista (Patriotic Socialist Coordinator) KPD Kommunistische Partei Deutschlands (Communist Party of Germany) LOREG Ley Orgánica 5/1985, de 19 de junio, del Régimen Electoral General (Organic Law on the General Election Regime) MLNV Movimiento de Liberación Nacional Vasco (Basque National Liberation Movement) MSI Movimento Sociale Italiano (Italian Social Movement) NPD Nationaldemokratische Partei Deutschlands (National Democratic Party of Germany ) NI Northern Ireland NICRA Northern Ireland Civil Rights Association NSDAP Nationalsozialistische Deutsche Arbeiterpartei (National Socialist German Workers’ Party) NSU Nationalsozialistischer Untergrund (National Socialist Underground) PA Partido Andalucista (Andalusian Party) PAR Partido Aragonés (Aragonese Party) PDS Partei des Demokratischen Sozialismus (Party of Democratic Socialism) PEGIDA Patriotische Europäer gegen die Islamisierung des Abendlandes (Patriotic Movement against Islamicization of the West) PP Partido Popular (Popular Party) PSE EE Partido Socialista de Euskadi-­Euskadiko Ezkerra (PSOE) (Socialist   (PSOE) Party of the Basque Country – Basque Country Left (PSOE)) PSOE Partido Socialista Obrero Español (Spanish Socialist Workers Party) PTA Prevention of Terrorism Acts

xiv   Abbreviations Die Republikaner (The Republicans) Sturmabteilung (Storm Troopers) Social Democratic and Labour Party Civil Authorities (Special Powers) Act (Northern Ireland) 1922 Sozialdemokratische Partei Deutschlands (Social Democratic Party of Germany) SRP Sozialistische Reichspartei Deutschlands (Socialist Reich Party of Germany) UA Unidad Alavesa (Alavese Union) UPN Unión del Pueblo Navarro (Union of the Navarrese People) UPyD Unión, Progreso y Democracia (Union, Progress and Democracy) USHICOG US High Commissioner for Germany UUP Ulster Unionist Party UV Unió Valenciana (Valencian Union) UVF Ulster Volunteer Force REP SA SDLP SPA SPD

Introduction Pluralism, tolerance and the proscription of political parties

The democratic dilemma of party bans Banning a political party is a grave act for a democracy. It contradicts fundamental commitments to freedom of expression and association, pluralism and tolerance. It distorts the posited level playing field of democratic competition and disrupts the articulation and representation of citizen preferences. The party ban is the harshest of myriad penalties, strategies and manoeuvres that may be employed to marginalize undesired political projects. It is usually the mark of tyranny. Yet at some point in the twentieth century many democratic states in Europe and North America – long the bastions of democratic politics – have banned a political party. Party bans are justified as a means to protect democratic practices or the state itself. With very few exceptions, proscribed parties are anti-­system parties representing communists, the far right, ethnic minorities or religious movements. They are typically banned for promoting authoritarian political forms and violent regime change, undermining democratic commitments to equality and pluralism, serving the interests of a foreign power, undermining the territorial integrity of the state, or some combination of these. Oftentimes, the ban addresses the fear – instantiated by the paradigmatic errors of Weimar Germany and the Nazi’s rise to power through constitutional means – that democracy may be abused by ‘enemies’ of the status quo. Democracy, as Loewenstein put it in his 1937 appeal against fascism, may become the ‘Trojan horse by which the enemy enters the city’ (Loewenstein, 1937, 424). Avineri has argued that the end of the Cold War, successive waves of democratization and, in Europe at least, the security umbrella of closer economic and political integration, make the party ban increasingly redundant (2004, 2). It is true that fascist and communist parties are no longer the primary targets of ban proceedings as they were, for instance, in the interwar and early post-­war periods, and in the case of communist parties following the collapse of the Soviet Union (Bourne and Casals Bértoa, 2017). Nevertheless, challenges to liberal democratic institutions, values and practices are still channelled through the party system, for example, through parties appealing to religious fundamentalist, neo-­fascist or racist programmes or parties representing insurgent, paramilitary

2   Introduction or ‘terrorist’ organizations. Indeed, as Niesen (2002) suggests, decisions to ban parties in contemporary democracies tend to be justified as measures to protect civil society from harm to minorities, individual political rights or the rights of future generations, rather than as a tool to tackle subversion. Bligh similarly distinguishes between traditional Weimar-­inspired ‘militant democracy’ rationales, focusing on parties that seek to abolish democracy wholesale, and more contemporary party-­ban rationales focusing on parties that incite hatred and discrimination, that support violence and terrorism and that challenge the identity of the state (2013, 1321). The continuing importance of party proscription in democratic states is also apparent from its treatment in international law. Party bans are regulated by various international charters on human rights, including the 1948 Universal Declaration of Human Rights, the 1953 European Convention on Human Rights (ECHR), the 1966 International Covenant on Civil and Political Rights and the 1966 Convention on the Elimination of all forms of Racial Discrimination (Tomuschat 1992; Brems, 2006a). In Europe, the jurisprudence of the European Court of Human Rights (ECtHR) has been particularly influential. Its rulings have established the doctrine of ‘militant democracy’, which permits, under certain conditions, restrictions on rights of anti-­democratic actors where this is necessary for the protection of democracy itself (Harvey 2004; Brems 2006a). Anti-­system parties may thus continue to pose an acute dilemma for democracies: Banning a party may help to defend democracies from their ‘enemies’ but proscription risks undermining foundational liberal democratic commitments to free association, free speech and the representation of all citizens in the public sphere. Democratic states respond to the dilemma in different ways. A study of party bans in European democracies in the post war period conducted by Bourne and Casals Bértoa (2017) illustrates this point. The study showed that the majority – 20 out of 37 – European democracies surveyed had banned a party at some time during that period. Table I.1 lists banned parties and banning countries included in Bourne and Casals Bértoa’s survey. It shows that parties of similar types were banned in some democracies, and in some distinctive historical contexts like the Cold War, but not others. Fascist, neo-­Nazi and far-­right parties have been banned, stripped of political rights or forced to dissolve in Italy, Germany, Austria and the Netherlands (and outside Europe, in Israel). Such parties have not, for instance, been banned in Sweden, Denmark or Britain. During the Cold War, communist parties were banned, or subject to ban proceedings, in Germany and Greece (and outside Europe in Australia and the United States). After the collapse of the Soviet Union, communist parties in Russia, Latvia, Lithuania and Ukraine also faced proscription. Yet communist parties in Italy and France, at their peak both highly successful electoral parties, were not subject to ban proceedings. Nor was the successful Czech Communist Party of Bohemia and Moravia proscribed in the post-­Soviet period. Furthermore, while Sinn Féin, the political wing of the Irish Republican Army, was banned in 1956 and legalized in 1974, Herri Batasuna and various successors serving as the political wing of

Introduction   3 the terrorist group Euskadi Ta Askatasuna (ETA, Basque Homeland and Freedom) were legal in the democratic period except for the ten years between 2003 and 2012. Separatist and/or minority nationalist parties have also been banned in Turkey and Bulgaria. This variation in responses to anti-­system parties raises the principal puzzle I explore in the book: Why do some democracies respond to the dilemma posed by anti-­system parties by banning them, while other democracies do not? Answers to this question not only provide insights into the practical steps taken in democratic states to deal with anti-­system parties, and the impact of varying institutional arrangements, political cultures and actor constellations on public decisions of constitutional significance. It also provides opportunities to probe the deliberative processes, discursive strategies and political strategies employed when democratic communities negotiate inherent tensions in foundational commitments to tolerance and pluralism. In this book, I explore these issues through empirical case studies of anti-­system party bans in the United Kingdom, Spain and Germany. These are countries which have recently proscribed, or considered proscribing, anti-­system parties (Spain and Germany), where proscription has been an important part of state responses to paramilitary violence (Spain and the United Kingdom), or where ban decisions take place against a backdrop of the legacy of authoritarian rule (Spain and Germany). More specifically, the book examines the proscription of radical Basque nationalist parties, Herri Batasuna (HB, Popular Unity), Euskal Herritarrok (EH, Basque Citizens) and Batasuna (Unity) in 2003. Herri Batasuna had been legal and participated in elections at state and regional levels for around two decades prior to this. These parties were banned for integration in the terrorist group ETA, which pursues a separate Basque state incorporating provinces in France and Spain. The book also examines the legalization of the radical Basque nationalist parties in the form of Bildu (Unite) in 2011 and Sortu (Create) in 2012, following ETA’s declaration of a definitive cessation of armed struggle. The fate of the republican parties, Sinn Féin (We Ourselves) and its (temporary) successor Republican Clubs in Northern Ireland (1922–72) is also addressed. Sinn Féin was banned in 1956 and the Republican Clubs were banned in 1967 for integration in the terrorist group, the Irish Republican Army (IRA). Republican organizations pursued the incorporation of Northern Ireland into the Republic of Ireland. In 1972, following increasingly violent sectarian conflict in Northern Ireland between the (Irish) nationalist and Catholic community, on the one hand, and the (British) unionist and Protestant community on the other, the United Kingdom suspended Northern Ireland’s devolved institutions and ruled the territory from Westminster. Sinn Féin and the Republican Clubs were legalized almost immediately afterwards, eventually allowing Sinn Féin to play a major role in the process leading to the IRA’s renunciation of armed struggle. The German case study permits analysis of proscription proceedings against parties of the extreme right. The Socialist Reich Party of Germany (SRP, Sozialistische Reichspartei Deutschlands), effectively a Nazi successor party, was banned in 1952, soon after foundation of the democratic Federal Republic of

Table I.1  Parties subject to ban proceedings in Europe, 1945–2015 Countries

Banned parties (year)

Ideological orientation

Austria

German National Socialist Workers Party (1945) National Democratic Party (1988) Flemish National Union (1945) Parti Rexiste (1945) Flemish Block (2004) United Macedonian Organisation/Ilinden-Pirin (2001) Serbian Democratic Party (1995) Workers’ Party (2010) Nationalist Party (1959) Proletarian Left (1970) Revolutionary Communist League (1973) Enbata (1974) Corsican Movement for Self-determination (1987) Radical Unity (2002) Socialist Reich Party (1952) Communist Party of Germany (1956) Communist Party of Greece (1947) National Fascist Party/Republican Fascist Party (1947) Communist Party of Latvia (1991) Communist Party of Lithuania (1991) Communist Party of the Republic of Moldova (1991) National Socialist Movement in the Netherlands (1945) National European Social Movement (1955) Dutch Peoples Union (1978) Centre Party ’86 (1998) National Gathering (1945) Communist Party (Nepeceristi) (2008) Slovak Community-National Party (2006) Herri Batasuna (2003)/Euskal Herritarrok (2003)/Batasuna (2003)/Eusko Abertzale Ekintza (2008))/Communist Party of the Basque Territories (2008)/Askatasuna (2009)

Extreme right Extreme right Extreme right/substate nationalist Extreme right Extreme right/substate nationalist Substate nationalist Substate nationalist Extreme right Extreme right Extreme left Extreme left Substate nationalist Substate nationalist Extreme right Extreme right Extreme left Extreme left Extreme right Extreme left Extreme left Extreme left Extreme right Extreme right Extreme right Extreme right Extreme right Extreme left Extreme right Substate nationalist

Belgium Bulgaria Croatia Czech Republic France1

Germany2 Greece Italy Latvia Lithuania Moldova The Netherlands3

Norway Romania Slovakia Spain4

Turkey5

UK Ukraine

Turkey Comfort Party (1983) United Communist Party of Turkey (1991)/Socialist Union Party (1995) Socialist Party (1992) People´s Labour Party (1993)/Freedom and Democracy Party (1993)/Democratic Party (1994)/People´s Democracy Party (2003) Socialist Turkey Party (1993) Democracy Party (1994) Democracy and Change Party (1996) Labour Party (1997) Welfare Party (1998)/Virtue Party (2001) Democratic Mass Party (1999) Democratic Society Party (2009) Sinn Féin (1956)/Republican Clubs (1967) Fianna Uladh (1956) Communist Party of Ukraine (1991) Russian Bloc (2014) Russian Unity (2014) Communist Party of Ukraine (2015)6

Pan-Islamist Extreme left Substate nationalist Substate nationalist Extreme left/(minority) nationalist Substate nationalist Substate nationalist Extreme left Pan-Islamist Substate nationalist Substate nationalist Substate nationalist Substate nationalist Extreme left Substate nationalist Substate nationalist Extreme left

Source: data from Bourne and Casals Bértoa (2017). The 17 states in the survey that did not ban parties were Albania, Bosnia-Herzegovina, Cyprus, Denmark, Estonia, Finland, Hungary, Ireland, Kosovo, Macedonia, Montenegro, Poland, Portugal, Slovenia, Serbia, Sweden and Switzerland. Notes 1 In France many small, rather obscure organisations have been banned and it is often difficult to distinguish between banned parties and associations. The list provided here is a sample. 2 In the case of Germany, the far-right Free German Workers Party and National List were excluded because the Federal Constitutional Court ruled that these were associations. (See Table 2.3). 3 It is a grey area whether Dutch People’s Union is a party ban case. It was characterised as a prohibited association in 1978, but the Dutch Supreme Court later ruled that because it was not formally dissolved it could no longer be excluded from participating in elections. 4 This list of party bans in Spain does not include parties that were denied registration or party lists or lists of electoral coalitions banned prior to electoral contests (for more details see Table 2.1). 5 Party ban cases for Turkey only include those after 1983, the initiation of the longest period of (semi-) democratic rule in Turkish modern history. 6 The Communist Party of Ukraine was the same party banned in 1991 but permitted to re-emerge in 1993. The ban in 2015 was due to the party’s separatist goals and for purportedly undermining constitutional values.

6   Introduction Germany. In 2001, after a long period where the state took a more permissive stance on anti-­system parties, the German government, the Bundestag and Bundesrat, initiated ban proceedings against the neo-­Nazi National Democratic Party of Germany (NPD, Nationaldemokratische Partei Deutschlands). However, the German Federal Constitutional Court rejected the petition on procedural grounds. In 2013, the Bundesrat, representing Länder governments, relaunched ban proceedings against the NPD. While accepting the argument that the NPD sought to undermine the free democratic basic order, the Federal Constitutional Court nevertheless rejected the ban petition ruling that the NPD was not likely to achieve its goal. In what remains of the introduction, I define the key concepts ‘anti-­system parties’ and ‘party bans’, outline hypotheses explored and the methodological approach employed and present the main findings of the research.

Defining ‘anti-­system parties’ and ‘party bans’ Defining ‘political party’ is notoriously difficult, given the range of historical and political contexts in which they operate and the various normative assumptions about functions parties ought to perform (Duverger, 1954, xiv–xv and xxii–xxiv; White, 2006). Nevertheless, it is necessary to adopt a working definition in order to distinguish the object of study – political parties – from sometimes very closely related phenomena such as associations, interest groups or insurgent groups. One can begin with the criteria of self-­identification; a political party, at the very minimum, is an association of people that claims to be a political party. Official recognition is also relevant for many of the parties considered here; a political party is an association that has been formally recognized as such, often through an official registration procedure. Beyond this, political parties can be defined with reference to certain distinctive goals and modes of behaviour. Political parties are typically organizations which, in Alan Ware’s (1996, 1–6) formulation, ‘seek influence in a state’, often, but not always, fielding candidates in elections in order to occupy positions in government at various territorial levels. They may formulate a programme of government, sets of preferred policies, future-­orientated programmes for political change or simply serve as a vehicle for the political ambitions of individuals. They ordinarily pursue goals through non-­violent and legal means, although some may secretly employ illicit means or have close links to violent or clandestine groups. The task of defining anti-­system parties is more complex, partly due to a tendency for this term to be used as a synonym for the more derogatory term, ‘extremist parties’. To be sure, some anti-­system parties pursue ethically objectionable goals. However, to label all parties subject to party bans as extremist parties seems to prejudge questions about the justice of their exclusion or marginalization from the public sphere. A further difficulty for defining anti-­ system parties relates to the changing nature of the goals and rhetorical strategies pursed by parties outside the political mainstream. In the first few decades after

Introduction   7 World War II, defining anti-­system parties involved the identification of parties that clearly rejected the core institutions and values of liberal democracies. In 1966, for instance, Otto Kirchheimer distinguished between ‘loyal opposition’ and ‘opposition in principle’. Loyal opposition included groups which pursued goals in harmony with the constitutional requirements of the system, whereas ‘opposition in principle’ pursued goals ‘incompatible with the constitutional requirements of a given system’ (1966, 237). Soon after, Giovanni Sartori (1976) developed the best-­known conception of anti-­system party (1976, 117–118). In broad terms, an anti-­system party will ‘undermine the legitimacy of the regime it opposes’ (ibid., 117–118). More specifically, an anti-­system party ‘would not change – if it could – the government, but the very system of government’. Echoing Kirchheimer’s distinction, Sartori argues anti-­system parties are not talking about an ‘opposition on issues’, but an ‘opposition of principle’. The anti-­system party, then, ‘abides by a belief system that does not share the values of the political order in which it operates’ and is distinguished by the ideological distance between itself and mainstream parties. Anti-­system parties may operate from within the system or from without and include parties dedicated to revolutionary preparation and activity, as well as those which are willing to play by the democratic rules of the game, at least in the short term. One critique of these early conceptions of anti-­system behaviour is that they fail to problematize orientations to violence, which is clearly an important part of what can make some parties problematic in democratic states. Linz’s distinction between ‘loyal and disloyal opposition’ is more instructive insofar as it focuses less on programmatic goals and relations with mainstream parties than on attributes which might undermine democratic regimes (Linz, 1978, 27–38). More specifically, ‘disloyal oppositions’ are anti-­system parties which question the existence of a democratic regime and aim to change it. They are considered dangerous because they may take power or divide the population and cause civil war if, in a crisis, they can mobilize support, and if they cannot be repressed and isolated (ibid., 27). Among other things, disloyal parties will not unambiguously and publically reject the use of violent means to achieve or maintain power and may employ a rhetoric of violence to mobilize supporters. This last facet of Linz’s concept of ‘disloyal opposition’ – that is, a party’s orientation to violence – is particularly useful because it permits consideration of an important means by which anti-­system parties may undermine the legitimacy of the regimes they oppose. Over time, however, definitions such as those offered by Kirchheimer and Sartori have become more problematic. When formulating their definitions, such scholars mostly had in mind the fascist and communist movements in interwar Europe, the remnants of Western European fascist movements that lingered after the defeat of the Axis powers in World War II and West European communist parties, which in countries like Italy, France and Finland, were relatively successful in elections. These were parties that appeared to participate in democratic politics in order to destroy the democratic system itself (Capoccia, 2002a, 11). In more

8   Introduction recent times, however, anti-­system parties have often come to claim the mantle of democracy and sought to pursue their political goals through parliamentary means (ibid., 12). While there is room for reasonable doubt about the sincerity of some of these groups’ proclaimed commitments to democracy, it is hard to deny that the ambiguity, internal contradictions and inconsistencies in the stated goals of some parties make it difficult to determine with certainty whether they embrace a clear cut anti-­democratic ideology (Schedeler, 1996, 303). Capoccia’s (2002a) conception of ideological anti-­systemness helps to address this issue. This concept departs from the observation that it would be possible to define anti-­system actors as synonymous with anti-­democratic actors if a clear definition of democracy is established alongside an indication of how an anti-­system party violates or rejects this. To this end, Capoccia adopts Collier and Levitsky’s (1997) definition of democracy as a political system that, at a minimum, includes: (a) fully contested elections, (b) full suffrage and an absence of massive fraud, (c) effective guarantees of civil liberties, underpinned by notions of political equality for all citizens regardless of religion, race, colour, age etc. and (d) elected governments with a strong capacity to govern. Capoccia does not seem to specifically address the issue of violence, but we might reasonably add the additional criterion (e) commitments to pursue parliamentary representation and government office through exclusively non-­violent means. A final criterion, drawn from Backes’ discussion of religious/secularist extremisms, could include (f ) commitments to balance goals of religious freedom and secular foundations of the constitutional state (2010, 189). Essentially, if a party rejects one, some or all of these aspects of the liberal democratic system, Capoccia argues, it could be considered an anti-­system party. In this book, I adopt Capoccia’s (amended) conception of ‘ideological anti-­system parties’ because it acknowledges the variety of ways in which contemporary political parties may challenge the foundational institutions, values and principles of liberal democratic politics. I define the concept of ‘party bans’ as acts which subject parties to one of the following five measures. These forms of party ban vary in terms of the degree to which the targeted party is excluded from the public sphere. The most punishing form of exclusion is dissolution. It denies a party the right and means to participate in public life principally by disallowing participation in elections or holding office. The party’s assets may be seized, its offices closed and the party leadership sanctioned. Non-­registration, involves denial, by the state, of a new party’s right to formally exist as a party, acquire associated privileges or to participate in elections. Rights denial involves the withdrawal of certain rights and privileges – such as the right to stand in elections – of already existing parties, even though the party as such may not be formally dissolved. A lapsed ban is a ban that has been undermined by the failure of the state to prevent a successor from taking on the mantle of a banned party. A failed ban occurs when at least one major institution of the state – government, parliament or the courts – formally approves a party ban but this is not sufficient to complete proceedings to finalize the ban.

Introduction   9 Party bans are just one of a wide range of measures that can be employed against anti-­system parties. Downs’ (2012) and Capoccia’s (2005) classifications of measures of defensive democracy are useful for conceptualizing the range of such measures. Downs identifies two strategies governments can use to deal with electorally successful ‘pariah’ parties – a strategy of disengagement (‘clean hands’) or one of engagement. Each of these strategies may involve more or less tolerant responses. Combining the disengagement/ engagement dimension with the tolerant/militant dimension produces four types of strategy. Party bans are incorporated into a category of measures involving disengagement and intolerance, seen as a ‘more aggressive’ strategy involving ‘overt political and legal attempts to isolate, restrict and repress and even ban the offending pariah’ (2012, 31). In addition to party bans, ‘ban/ isolation’ strategies can include a wide variety of legal and political measures, including manipulation of electoral rules, collusion among mainstream parties to reduce anti-­system party success at elections or exclude their participation in government, and legal rules punishing offensive speech acts (like Holocaust denial), or displays of politically sensitive symbols (like Nazi paraphernalia). Alternative responses include ‘ignore’ (disengagement and tolerance), which is a ‘do-­nothing approach’ aiming to prevent an anti-­system party from ‘capturing the attention they crave’ (ibid., 31–32). A strategy of ‘co-­option’ (engagement and militancy) occurs when ‘the political establishment … engages the pariah directly on the issue or issues fueling its electoral success and tries to aggressively combat the threatening party by recapturing the policy space’ (ibid., 31). A final strategy is ‘collaborate’ (engagement and tolerance) and involves agreement by mainstream and pariah parties to work together either prior to elections, for example in the form of electoral pacts, or afterwards in the form of coalition agreements or collaboration, to pass legislation (ibid., 46). Capoccia’s (2005) classification of measures of defensive democracy distinguishes between measures designed with (a) short-­term goals, namely ‘stemming the development of an existing challenge, to prevent it snowballing [and with the] declared immediate goal of democratic survival’ and (b) long-­term goals, which ‘attempt to reinforce and stabilize the basic procedures and values of the democratic system’ (ibid., 48). A second dimension of the typology distinguishes between measures that are predominantly ‘exclusive-­repressive’ and those that are ‘inclusive-­educational’ in nature (ibid.). Plotting these variables into a two-­ dimensional property space produces four ‘polar strategies’ of defensive responses to political extremism: Party bans are conceptualized as part of a strategy of militancy (short-­term, repressive), a strategy which involves measures to ‘curb de jure or de facto, the political and civil rights of certain subjects on the basis of their political opinions and activities, which have been defined as harmful to the survival of the democratic system’. Other measures in this category include restrictions on use of party uniforms or party militias, and regulation of rights of demonstration (ibid., 58). A second strategy is purge (long-­ term, repressive) which is

10   Introduction often adopted after the transition to democracy from an authoritarian regime, involve[s] enhancing the real or perceived legitimacy of the government by, for example, ensuring the systematic loyalty of bureaucrats or prosecuting authors of political crimes connected with the previous regime. (Ibid., 49) Education (long-­term, accommodative) strategies ‘seek to strengthen democratic values and beliefs as well as democratic practices at different levels’ (ibid.). Incorporation (short-­term, accommodative) strategies involve measures that ‘endeavor to bring into the system parts of the extremist opposition, thereby simultaneously weakening the extremist camp and increasing the legitimacy of the regime and the support for it’ (ibid., 49). It is significant that both Downs and Capoccia conceptualize party bans as the most ‘militant’, ‘intolerant’, ‘aggressive’ or ‘repressive’ measure that can be employed against anti-­system parties (see also Bleich, 2011, 87). There is also agreement among them that party bans are the most risky in terms of possible damage to the overall quality of the democratic system (Downs, 2012, 49; Capoccia, 2005, 59). It is this – the relatively repressive nature of party bans and its heightened risks for the quality of democracy in practice – that makes banning political parties the grave act claimed in the opening paragraph of the book. Moreover, it is these qualities that make the study of party bans especially pertinent for understanding how democratic states deal with anti-­system parties and movements. For the reason that the stakes are so high when banning parties compared to alternatives, that decisions to ban parties can be considered the most extreme conditions in which democracies must negotiate their way through the dilemmas posed by the presence of anti-­system parties.

Hypotheses and case selection Given the relative frequency and political salience of party bans in democratic states, it is surprising how little systematic scholarly attention has been devoted to the task of explaining proscription. The literature on party bans mostly consists of country-­specific studies identifying rationales for the proscription of individual parties (e.g. Auerback, 1954; Dyson, 1975; Franz, 1982; Niesen, 2002; Turano, 2003; Tardi, 2004; Dyzenhaus, 2004; Esparza, 2004; Iglesias, 2008; Corcuera et al., 2008; Navot, 2008; Mareš, 2012). While providing useful insights into party ban rationales, these studies tend to take official rationales for party bans at face value, something that this research aims to problematize. Moreover, the topic has traditionally been dominated by the disciplines of law and political philosophy, which have addressed questions other than explanations for party bans. Legal scholars focus attention on the nature of statutory and constitutional constraints on political parties (Kirchheimer, 1961; Gordon, 1987; Tomuschat, 1992; Fox and Nolte, 2000; Sajó, 2004; Brems, 2006a and 2006b; Issacharoff, 2007; Rosenblum, 2007; Bligh, 2013). Political philosophers have explored the appropriateness of

Introduction   11 limiting democratic rights and liberties, with work on political tolerance and freedom of expression of particular relevance (Rawls, 1971; Waldron, 1981; Scanlon, 2003; Quong, 2004, Kirshner, 2014; Malik, 2008). Political scientists have reflected more systematically on the effects of party bans, than on explanations for them (Tilly, 2005; Minkenberg, 2006; Koopmans, 2005; Husbands, 2002; Casal Bértoa and Bourne, 2017). In order to address the scarcity of operationalizable theoretical statements about why democracies ban political parties in the existing literature, it was necessary to employ a complex research design involving two stages of hypothesis formation and testing, which I spell out in more detail in Chapter 1. The first step involved developing hypotheses from underdeveloped but insightful arguments about observed empirical regularities in single and ‘small n’ case studies about the conditions under which democracies ban parties. I then tested these hypotheses in a ‘crucial’ case study (Eckstein, 1979) of party bans in Spain, and more specifically the proscription of Herri Batasuna and its successors in 2003 (Bourne, 2015). Further empirical analysis of the case permitted formulation of additional hypotheses principally drawn from the fields of security studies and new institutionalism. The hypotheses that emerged from this study and which, in the second step of the research design, I now examine in this book are: H1. Democracies ban anti-­system parties if these parties have been ‘securitized’ as an existential threat. To paraphrase Buzan et al. (1998, 21), ‘securitization’ in this context is the process by which a party is presented and accepted as an existential threat to democratic institutions and values of a political community or the territorial integrity of the state, a threat requiring emergency measures outside the normal bounds of political practice. H2. Democracies ban anti-­system parties if veto players prefer proscription.

Veto players are defined as ‘actors whose agreement is required for a change in the status quo’ (Tsebelis, 2002, 17) and must be taken into account given that party ban decisions typically have major political and constitutional significance and usually involve the principal public authorities of the state. H3. Democracies ban anti-­system parties if they do not unambiguously reject violence.

12   Introduction Today, few parties openly espouse the violent overthrow of the democratic political order. However, some anti-­system parties seek to directly or indirectly legitimize the violent actions of others, or as in the case of the political wing of a terrorist group, exist to represent insurgents in the electoral arena. H4. Democracies ban anti-­system parties if alternative forms of marginalization are not effective. Alternatives I examine in the book are the marginalization of anti-­system parties through the electoral system and collusion by mainstream parties to keep anti-­system parties out of government office. Alternatives may also include a wide range of other measures identified in Downs’ (2012) and Capoccia’s (2005) studies, such as criminalization of offensive behaviours and speech acts, collaboration with anti-­system parties or civil education. H5. Democracies ban anti-­system parties if partisan veto-­players do not need to cooperate with them to win and maintain office and achieve policy goals. This hypothesis probes the incentives generated in the context of democratic competition on the decisions of partisan party-­ban-veto-­players to ban parties. It assumes that mainstream parties avoid banning anti-­system parties if they think it will damage their chance of obtaining or maintaining hold of political office or achieving policy goals when collaborating with other parties is necessary. The principal objective of this book is to test these hypotheses in additional case studies described earlier. These cases were selected to permit what Yin describes as ‘literal’ and ‘theoretical’ replication of the findings of existing research (2003, 47). Additional case studies selected on the expectation that they would replicate the findings of the first Herri Batasuna and successors party ban study (literal replication) were (1) the banning of Sinn Féin (1956), Republican Clubs (1967) and the Socialist Reich Party (1952). Additional case studies expected to produce contrasting results but for reasons predicted by the findings of the 2003 Herri Batasuna and successors case study (theoretical replication) were (2) the legalization of formerly banned parties Bildu (2011), Sortu (2012), Republican Clubs (1973) and Sinn Féin (1974) and (3) the two NPD failed party ban cases of 2003 and 2017. In the legalization cases, it was expected that each of the conditions under which parties are banned would no longer hold, while in the failed party ban cases it was expected that at least one of the conditions under which parties are banned would no longer hold. I employ a mixed methods approach to data analysis comprising process tracing and discourse analysis, as

Introduction   13 well as the quantification of electoral system effects and coalition predictions. The range of methods employed follows from the nature of the hypotheses examined and acknowledges strengths and weaknesses entailed by different methods (George and Bennett, 2005). I draw on secondary literature about the parties studied and state responses to anti-­system parties and movements; constitutional frameworks and legislation; election results and data on practices of government formation; newspaper articles, court rulings, government documents, parliamentary debates and, where possible, government archives.

Structure of the book and findings In Chapter 1, I present a theoretical framework for explaining why some democracies ban anti-­system parties while others do not. It begins with an analysis of the problematic legacy for the study of party bans of the paradigmatic concept of ‘militant democracy’. The chapter then discusses the two-­stage research design for hypothesis formation and testing employed for the study and criteria for case selection. This is followed by a discussion of the theoretical underpinnings of the five hypotheses, their operationalization and theoretical expectations regarding relationships between variables. As a prelude to case study analysis, Chapter 2 briefly describes for each of the three country case studies legal rules for banning parties and provides an overview of parties subject to party ban proceedings. The next three chapters present the results of party ban case studies in Spain (Chapter 3), the United Kingdom (Chapter 4) and Germany (Chapter 5). They address hypotheses related to securitization, veto-­players and anti-­system party orientation to violence for each type of party ban where relevant (bans, failed bans, legalization of banned parties). Chapter 6 addresses constraints imposed by electoral systems on anti-­system parties and more specifically whether electoral rules may provide a substitute for party bans. Chapter 7 addresses party ban decisions in light of government formation processes, in particular the effectiveness of the cordon sanitaire as an alternative to proscription and the impact of mainstream parties’ office- and policy-­seeking goals on incentives for banning anti-­system parties. On the basis of the empirical research presented in the book, I conclude that party bans are likely to occur when (1) the ambiguity of an anti-­system party about the appropriateness of political violence (2) facilitates a securitization of that party as an existential threat and identification of the party ban as a solution to that threat, and (3) this is accepted by party-­ban-veto-­players. On the other hand, a party ban initiative is likely to fail if veto players cannot agree it is appropriate to ban the party. Legalization of banned parties is likely to occur when (1) veto players accept (2) a desecuritization of anti-­system parties as an existential threat and, consequently, the inappropriateness of a party ban. In contrast, (1) the availability or otherwise of effective alternatives to party bans or (2) incentives of partisan party-­ban-veto-­players to cooperate with, rather than ban, anti-­system parties, do not emerge as important explanations for party ban or legalization decisions.

14   Introduction These conclusions suggest that democratic communities are likely to respond to the dilemmas posed by anti-­system parties and strategies of democratic defence predominantly through recourse to security discourses. Sufficient consensus must emerge that the party poses a serious threat to democratic values or institutions of the state in order to justify the grave act of banning a party. The ban, moreover, is likely to remain an exceptional measure targeting parties unwilling to embrace non-­violent political methods. On the other hand, the malleability of securitization discourses creates opportunities for unscrupulous political actors to mobilize security arguments to justify banning parties that represent unpopular minorities or which serve as political rivals. These complex challenges have emerged intermittently over the post-­World War II period and are likely to continue to do so in an era where populist parties and movements question core principles and values of the liberal democratic model, if not the legitimacy of liberal democracy itself.

1 Theory and research design From ‘militant democracy’ to security discourse, institutions and democratic competition

This chapter presents the theoretical foundations of the book’s research. I begin with a discussion of the existing literature on ‘militant democracy’ and argue that this paradigmatic concept provides only limited insights for explaining why some democracies ban parties but others do not. I then discuss the process of hypothesis formation, followed by an elaboration of theoretical underpinnings and operationalization of hypotheses explored in the book. In the conclusion, I outline expectations regarding relationships between variables.

The ‘militant democracy’ paradigm In both the study of party bans and democratic responses to political extremism more generally, ‘militant democracy’ is a paradigmatic concept. Its origins are usually traced to Karl Loewenstein’s (1937) appeal for robust responses to the rise of fascism in 1930s’ Europe. Loewenstein thought ‘fundamentalist’ commitments to democratic principles could be ‘suicidal’ because ‘under cover of fundamental rights and the rule of law, the anti-­democratic machine could be built up and set in motion legally’ (ibid., 423–424). Only timely implementation of anti-­fascist legislation – including bans on party paramilitary organizations and political uniforms, prosecution of incitement to violence or hatred and the proscription of subversive movements or parties – could provide democracies with effective defence against extremists (ibid., 429). In the post-­war period, justifications and techniques of militant democracy have become more widely used, with Germany the best-­known example of a so-­called militant democracy (McWhinney, 1957; Kirchheimer, 1961; Niesen, 2002; Rensmann, 2003). The ECtHR has also developed its own doctrine of militant democracy (Harvey, 2004; Brems, 2006a; Macklem, 2012). In recent years, however, many scholars have criticized the concept of ‘militant democracy’ for being too vague, too imprecise, or too narrowly focused on legal instruments (Mudde, 2004, 197; Pedahzur, 2004, 109–110; Thiel, 2009, 384). Others have argued that the concept tends to focus on movements that no longer constitute a serious threat to democratic politics, such as fascist or orthodox communist parties using democratic entitlements to win control of the state (Bligh, 2013, 1335–1336). More recently, scholars have tended to use the

16   Theory and research design concept of ‘militant democracy’ as shorthand for a much wider range of measures employed against all kinds of extremist threat (e.g. Sajó, 2004; Thiel, 2009). This enlarged concept of militant democracy has become the victim of ‘concept stretching’ (Sartori, 1970, 1033). At its most extreme, the problem is discernible in the argument that there is no alternative to militant democracy because ‘it is barely conceivable that a country does not have (or never has) taken legally defensive measures to fight … against political extremists or terrorist threats’ (Thiel, 2009, 384; see also Vidal, 2009, 260). In other words, if all democracies are in some sense militant democracies, we cannot be certain what a non-­militant democracy is and ‘militancy’ cannot serve as a variable for explaining differences in response to anti-­system parties or to other kinds of anti-­system collective actors (Bourne, 2012b). Some have preferred to reconceptualize democratic militancy as a matter of degree; that is, democracies may be conceived as either more or less militant (Pfersmann, 2004, 53). However, this strategy also raises the problem of ‘degreeism’ (Sartori, 1991, 243–257), or that focusing on differences of degree can hide important differences in kind which may help explain variation in democratic responses to extremism. Others have sought to reconceptualize militant democracy as one of a number of types of democratic response to extremism, including conceptualization of militant democracy as an alternative to a ‘preventive’ or ‘counter-­terror’ state (Sajó, 2012) or, in relation to former ruling party bans, as a response differing from ‘collaborator’ party bans or ‘post-­Soviet space’ party bans (Niesen, 2012). Others still have sought to develop more elaborate or comprehensive classificatory schemes and replace the rubric of militant democracy with a new vocabulary of ‘defending democracies’ (Pedahzur, 2004; Capoccia, 2001 and 2005; Rummens and Abts, 2010) or ‘intolerant’ democracies (Fox and Nolte, 2000). This work is useful for conceptualizing responses to democratic extremism as a dependent variable. However, it does not provide sufficiently precise operationalizable hypotheses that explain variation in state responses to extremism. In addition to work conceptualizing, criticizing and searching for alternatives to the concept of ‘militant democracy’, there are many case studies focusing on legal rationales for the proscription of a party or parties in individual states (e.g. Auerback, 1954; McWhinney, 1957; Dyson, 1975; Feofanov, 1993; Cohen-­ Almagor, 1997; Hanschmann, 2001; Koçak and Örücü, 2003; Turano, 2003; Rensmann, 2003; Ganez, 2004; Tardi, 2004; Dyzenhaus, 2004; Esparza, 2004; Brems, 2006b; Erk, 2005; Iglesias, 2008; Vidal, 2009; Kemmerzell, 2010; Mareš, 2012; Bourne, 2015; Celep, 2014). A smaller number of studies compare party ban cases across countries (Franz, 1982; van Donselaar, 2003; Navot, 2008; Niesen, 2002 and 2012; Bale, 2007; Bourne, 2012a). Many other studies compare legal and constitutional frameworks within which party bans are conducted, sometimes addressing both international and national regimes (Kirchheimer, 1961; Gordon, 1987; Tomuschat, 1992; Fox and Nolte, 2000; Brunner, 2002; Sajó, 2004; Brems, 2006a; Pohl, 2006; Karvonen, 2007; Issacharoff, 2007; Klamt, 2007; Rosenblum, 2007; Thiel, 2009; Hartmann and Kemmerzell, 2010; Bourne, 2012b; Macklem, 2012;

Theory and research design   17 Bligh, 2013). In this literature, studies of party bans in Europe and North America predominate, although studies of bans in the Middle East (especially Israel, e.g. Cohen-­Almagor, 1997; Pedahzur, 2004) and Africa (e.g. Kemmerzell, 2010; Hartmann and Kemmerzell, 2010) have extended the geographical range of the literature. Overall, the literature on party bans is dominated by the discipline of law, which mostly focuses on exploring and classifying the nature of legal and constitutional constraints on political parties rather than the question at issue in this research, namely, why some democracies ban parties but others do not. Many of these studies also tend to take official rationales for party bans at face value, something that this study seeks to problematize. The literature on party bans sits alongside work examining a broader range of measures democratic states employ to deal with anti-­system parties and movements (e.g. Capoccia, 2001, 2005; Downs, 2002, 2012; Bleich, 2011; Bleich and Lambert, 2013; van Spanje and van der Brug, 2007; van Spanje, 2010; Bale, 2003). As I spell out in more detail below, this literature provides many valuable insights for research on party bans. The relevance of this literature derives from the fact that many measures short of a party ban – such as erecting an exclusionary cordon sanitaire against anti-­system parties, or electoral rules serving as barriers to small (and often radical) parties – raise similar kinds of democratic dilemmas and strategic choices for mainstream parties as those raised by party bans. Nevertheless, this broader literature is not in itself sufficient for addressing questions about why some democracies ban parties but others do not. In the first place, it mostly focuses on responses of democratic states to the contemporary far right. In so doing, it tends to underestimate the importance of security issues as rationales for the proscription of political parties. In addition to dealing with the threat posed by parties of the far right to democratic institutions and principles, democratic states may ban parties for threatening the territorial integrity of the state, for threatening state security by collaborating with foreign powers, or for association with groups employing political violence. The broader literature on democratic responses to extremism is strongest where it develops theories to explain strategic choices of mainstream parties in relation to anti-­system parties (see especially Downs, 2002 and 2012; Capoccia, 2005; van Spanje and van der Brug, 2007; van Spanje 2010). However, this focus tends to lead scholars to conceive of political choices about party bans within a rational choice paradigm, albeit with particular attention to constraints posed by the structure of democratic competition and electoral institutions. As I argue below, this is unnecessarily constraining and, indeed, there are grounds for also giving analytical weight to the role of ideas, norms and discourse in the study of party bans.

Hypothesis formation and theory development In order to address the scarcity of operationalizable theoretical statements about why democracies ban political parties in the existing literature, it has been necessary to employ a complex research design involving two stages of hypothesis formation and testing.

18   Theory and research design Stage 1: hypothesis formation and testing in the crucial case of party bans in Spain I began by developing three hypotheses from underdeveloped but insightful arguments and observed empirical regularities in single and ‘small n’ case studies about the conditions under which democracies ban parties (see Bourne, 2015). The hypotheses were that: Democracies ban anti-­system parties if they do not unambiguously reject violence. Democracies ban anti-­system parties if alternatives to proscription are not effective. Democracies ban anti-­system parties if relevant office holders expect they will not be disadvantaged in their pursuit of office or votes for supporting the ban. I then tested these hypotheses in a case of party bans in Spain, and more specifically the ban on HB and its successors in 2003 (Bourne, 2015; see also Table 2.1). While often the subject of criticism, case studies can contribute significantly to theory development when they are ‘crucial cases’ (Eckstein, 1979), ‘critical cases’ (Yin, 2003, 40) and ‘deviant’ cases (Lijphart, 1971). In the first instance, party bans in Spain can be considered a ‘crucial case’, defined as ‘a case that must closely fit a theory if one is to have confidence in the theory’s validity’ (Eckstein, 1979, 118). The status of Spain as a crucial case is most evident in relation to the hypothesis about violence, probably the most important variable given foundational democratic commitments to negotiation of political differences through non-­violent means. In other words, if the hypothesis that democracies ban parties that do not unambiguously eschew violence is valid, it was expected, a priori, that HB, EH and Batasuna, known to be the political wing of the terrorist group ETA, would be banned. In relation to the hypothesis about violence, Spain can also be considered one of a small number of ‘most likely cases’, which are cases that ‘ought … to invalidate or confirm theories if any cases can be expected to do’ (Eckstein, 1979, 118). This is because Spain is one of the few democratic states where political parties have supported an armed insurrection against the state, even if party leaders as such did not bear arms. And finally, it was expect that other hypotheses about alternatives to proscription and the interests of relevant office holders would be valid given the range of policy instruments employed by the Spanish state to respond to political terrorism and the consolidation of pluralistic, democratic competition in Spain following the transition to democracy in the 1970s. Despite these expectations, HB was, in fact, legal for around two decades despite its relationship with ETA. Empirical research found that when HB and its successors were banned in 2003, alternatives had proved ineffective, that banning Batasuna was popular among those likely to vote for mainstream parties, and did not have consequence for government formation. Nevertheless, the fact that this party was legal for such a long time, despite its organic links

Theory and research design   19 with a terrorist group, made Spain a ‘deviant case’ in relation to the hypothesis that democracies ban parties that do not unambiguously reject violence. As Lijphart argues, analysis of the reasons why cases ‘known to deviate from established generalisations’ can contribute to theory development by ‘uncover[ing] relevant additional variables that were not considered previously, or to refine the (operational) definitions of some or all of the variables’ (1971, 692). Further analysis of the Spanish case permitted the formulation of further hypotheses principally drawn from the fields of security studies (Wæver, 1995; Buzan et al., 1998) and new institutionalism (especially Tsebelis, 2002) for further testing, namely: Democracies ban anti-­system parties if these parties have been ‘securitized’ as an existential threat to the state or democratic community. Democracies ban anti-­system parties only if veto players prefer them to do so. Stage 2: ‘literal’ and ‘theoretical’ replication of finding of the Spanish party ban case In this book, I undertake the second stage of the research design. I do so by testing the hypotheses listed above in cases of party bans, the legalization of formerly banned parties and failed party bans in further case studies from Spain, the United Kingdom and Germany. Case selection is determined by what Yin (2003) describes as a logic of ‘replication’, rather than a ‘sampling logic’. A ‘logic of replication’ is analogous to multiple experiments in which the researcher seeks to discover more robust findings by testing the results of a single experiment with further experiments, some of which may duplicate the exact conditions of the original experiment, while others might alter one or two experimental conditions to see whether findings will be duplicated (2003, 47). Cases are either ‘literal replications’, that predict similar results to findings of existing studies, or ‘theoretical replications’, that predict contrasting results but for reasons predicted by existing studies (ibid.). In this method, the mode of generalization is ‘analytical generalization’, in which ‘a previously developed theory is used as a template with which to compare the empirical result of the case study’ (ibid., 33). In other words, if empirical results turn out as predicted … they would provide compelling support for the initial set of propositions. If the cases are in some way contradictory, the initial propositions must be revised and retested with another set of cases. (Ibid., 47) To this end, cases of party bans, legalization of parties and failed attempts to ban parties have been selected on the expectation that they will provide either ‘literal’ or ‘theoretical’ replication of findings in the Spanish case. Cases expected to literally replicate the conditions under which HB and its successors were banned

20   Theory and research design Table 1.1  Summary of party ban cases and case selection criteria Parties

Type party ban Case selection criteria

Herri Batasuna, Euskal Herritarrok and Batasuna (2003) Sinn Féin (1956) Republican Clubs (1967) Socialist Reich Party (1952) Bildu (2011) Sortu (2012) Republican Clubs (1973) Sinn Féin (1974) NPD I (2003) NPD II (2017)

Ban

Original case study

Ban Ban Ban Legalization Legalization Legalization Legalization Failed ban Failed ban

Expect literal replication Expect literal replication Expect literal replication Expect theoretical replication Expect theoretical replication Expect theoretical replication Expect theoretical replication Expect theoretical replication Expect theoretical replication

in 2003 were the banning of Sinn Féin (1956) and Republican Clubs (1967) in the UK and of the Socialist Reich Party (1952) in Germany. A second set of additional case studies were expected to produce contrasting results but for reasons predicted by the findings of the 2003 HB and successors case study – or theoretical replication – were the legalization of Bildu (2011), Sortu (2012), Republican Clubs (1973) and Sinn Féin (1974). In the legalization cases, it is expected that each of the conditions under which parties are banned would no longer hold. A third set of additional cases, also selected on the expectation they would provide theoretical replication, were two failed attempts to ban the NPD in Germany in 2003 and 2017. In the failed party ban cases, it was expected that this contrasting outcome could be explained by the absence of one or more of the conditions identified in the 2003 HB and successor ban case. The advantage of analysing these three types of party ban cases is that they provide variation on the dependent variable and both positive (party ban) and negative (legalization and failed party ban) cases. Focusing on legalization of banned parties and failed ban cases – rather than anti-­system parties not subject to ban proceedings at all – responded to the fact that ban proceedings typically provide a context in which the appropriateness and rationales for party bans are widely discussed in the public sphere. Without such discussions, data crucial for analysis of variables relating to securitization discourses and positions of veto players is very difficult to obtain. The selection criteria for the cases are summarized in Table 1.1. I now turn to outline the theoretical underpinnings of each hypothesis and their operationalization for empirical research in the remaining chapters of the book.

Securitization, veto players and political violence Processes of securitization and desecuritization Securitization theory, developed in the field of international security studies, is premised on a conception of ‘security’ as intersubjective and socially constructed,

Theory and research design   21 rather than something that can be defined outside of politics (Wæver, 1995, 50–51; Buzan et al., 1998, 30–31; Balzacq, 2011, 1–4). In their seminal work, Buzan et al. define security as being ‘about survival’: It is when an issue is presented as posing an existential threat to a designated referent object (traditionally, but not necessarily, the state, incorporating government, territory and society). The special nature of security threats justifies the use of extraordinary measures to handle them. The invocation of security has been the key to legitimizing the use of force, but more generally it has opened the way for the state to mobilize or take special powers, to handle existential threats. Traditionally, by saying ‘security’, a state representative declares an emergency condition, thus claiming a right to use whatever means are necessary to block a threatening development. (Buzan et al, 1988, 21) Securitization occurs when an ‘issue is presented as an existential threat, requiring emergency measures and justifying actions outside the normal bounds of political procedure’ (ibid., 23). For its part, ‘desecuritization’ ‘shifts the issue out of emergency mode and into the normal bargaining process of the public sphere’ (ibid., 4). According to Buzan et al. desecuritization is the ‘optimal’ long-­range option, since it means to not have an issue phrased as ‘threats against which we have to take countermeasures’ and to move them out of the threat-­ defence sequence (ibid., 29). The proscription of political parties may be a response to securitization in three different security sectors – military, societal and political. Where a party is linked to a paramilitary group, party bans may be securitized as issues of military security, which, in this context, is ‘primarily about the ability of the ruling elite to maintain civil peace, territorial integrity and, more controversially, the machinery of government in the face of challenges by its citizens’ (Buzan et al. 1998, 50). Indeed, as Buzan et al. note, minority nationalist and secessionist groups are frequently objects of military interest and action and therefore securitization. They can easily be cast as threats to state sovereignty and, by the state-­ like activities they engage in, can motivate the existing state to use military force to secure its monopoly over legitimate violence. (Ibid.) Far-­right, Islamist or radical leftist parties that may challenge the institutions and constitutive principles of democratic states, or minority nationalist groups that may challenge identity conceptions of majority nations, may be securitized as threats to societal security. Societal security refers to threats to the ‘identity, the self-­conception of communities and of individuals identifying themselves as members of a community’, whether this be national, religious, racial or some other identity (ibid., 119). In other words, a threat to societal security exists

22   Theory and research design ‘when communities of whatever kind define a development or potentiality as a threat to their survival as a community’ (ibid., 119) and involves ‘action taken to defend such “we identities” ’ (ibid., 120). In this context, the referent object is ‘whatever larger group carry the loyalties and devotion of its subjects in a form and to a degree that can create a socially powerful argument that this “we” is threatened’ (ibid., 123). Finally, anti-­system parties as such may be securitized as a threat to the ‘organisational stability of social orders(s)’ (ibid., 141), or ‘political security’. Buzan et al. define political security as being about ‘threats to the legitimacy or recognition either of political units or of the essential patterns (structures, processes or institutions) among them’. More specifically, internal political threats are ‘made to … the internal legitimacy of the political unit, which relates primarily to ideologies and other constitutive ideas and issues defining the state’ (ibid., 144). These considerations inform operationalization of the hypothesis: H1. Democracies ban anti-­system parties if these parties have been ‘securitized’ as an existential threat. According to Buzan et al., we can recognize a securitization discourse because it contains within it a ‘grammar of security’ which ‘construct[s] a plot that includes existential threat, point of no return, and a possible way out’ (1998, 33). More specifically in relation to party ban debates we can identify two conditions for securitization. First, an anti-­system party is presented as an existential threat – to, for example, the physical security of individuals or the state, to the democratic community, social identities or the political system. Second, this conception of an anti-­system party as an existential security threat is seen to justify the use of the party ban as an emergency measure. Previous research confirms the importance of conceiving of both of these requirements as conditions for the securitization of anti-­system parties, because anti-­system parties may be conceived as threatening to a referent object, but there may nevertheless be disagreement on whether or not party bans are an effective or appropriate measure to address the security threat (Bourne, 2015). Drawing on Hansen, desecuritization can be conceived as involving ‘reconstitution of the friend-­enemy distinction’ (2012, 534) and reconceptualization of an issue as properly dealt with through mediation of difference in the public sphere (ibid., 531). In relation to party bans, it can be said to involve two conditions. The first is a reconceptualization of anti-­system parties as no longer constituting an existential threat to the physical security of individuals or the state, to the democratic community, social identities or the political system. Second, it involves an argument that the anti-­system party can no longer be effectively dealt with through proscription. Hansen identifies four ideal types of desecuritization. The first is change through stabilization, or ‘a slow move out of an explicit security discourse,

Theory and research design   23 which in turn facilitates a less militaristic, less violent and hence more genuinely political form of engagement’ (ibid., 539). It entails not the resolution of conflict, but a receding of an issue into the background, which in itself ‘requires continued recognition and accommodation on both sides’ (ibid., 540). This kind of desecuritization could include lapsed party bans, including acceptance that a banned party might, after time, re-­emerge under a new banner. Rearticulation, the second form, involves ‘desecuritizations that remove an issue from the securitized by actually offering a political solution to the threats, dangers and grievances in question’ (ibid., 543). Here an issue is ‘rearticulated rather than just replaced’. It could include legalization of a party in the context of peace processes addressing grievances of anti-­system parties. Desecuritization through replacement involves ‘one issue moving out of security while another is simultaneously securitized’ (ibid., 541). Finally, silencing occurs ‘when an issue disappears or fails to register in security discourses’ (ibid., 544). It involves a move from the securitized to the non-­politicized rather than to the politicized realm of the public sphere. Securitizing agents, audiences and institutions An intrinsic feature of the concept of ‘securitization’ and its antonym ‘desecuritization’ is that ‘security’ is conceived as a particular form of politics (Buzan et al., 1998, 29; Wæver, 1995, 55, 65). In other words, it requires that a public issue is politicized, ‘meaning the issue is part of public policy requiring government decisions and resource allocations, or more rarely some other form of communal governance’ (Buzan et al., 1998, 23). Politicized issues ‘appear to be open, a matter of choice, something that is decided upon and that therefore entails responsibility’ (ibid., 29). Securitization begins with a securitizing move, a discourse that takes the form of presenting something – here, a party – as an existential threat to one of the above-­described referent objects (ibid., 25). An issue is successfully securitized only if a relevant audience accepts it as such (ibid., 25 and 31). Securitization arguments must gain enough resonance for a platform to be made from which it is possible to legitimize emergency measures or other steps that would not be possible had the discourse not taken the form of existential threat, point of no return, necessity. (Ibid., 25) Moreover, ‘In a democracy, at some point it must be argued in the public sphere why a situation constitutes security and therefore can legitimately be handled differently’ (ibid., 28). The securitization literature – and my earlier party ban case study (Bourne, 2015) – points to the utility of an institutionalist approach for understanding these political processes. The assumption that institutions matter is widespread in securitization theory. In 2003, Williams noted a clear affinity between securitization

24   Theory and research design approaches and new institutionalism, due to their focus on ‘epistemological, normative and processual structures and to questions of the logics of “appropriate action” and “institutional legitimation” ’ (2003, 528). One of the conditions Buzan et al. claim facilitate securitization is that ‘the securitizing actor must be in a position of authority’ (1998, 33) (see also Balzacq’s arguments about ‘linguistic competence’ (2005, 191) and Stritzel’s (2007) point about the ‘positional power’ of securitizing agents). Similarly, Buzan et al., distinguish between ad hoc and institutionalized securitization (1998, 27–28), while others have noted a bias towards democratic decision-­making in securitization theory (Roe, 2012; Vuori, 2008, 68), which rests on an implicit assumption of institutionalized commitments to democratic practices. These points hint that the institutional setting in which securitizing agents engage merit analytical attention. A role for institutions is more explicit in securitization theory dealing with the relationship between securitization actors and audiences. Here Balzacq’s distinction between ‘moral’ and ‘formal’ support is instructive, with the difference being that formal support – by, for instance, a parliament, the Security Council or Congress – is needed to actually achieve goals posited in securitization processes (2005, 185). It may be that ‘securitization agents’, especially politicians, ‘always strive to convince as broad an audience as possible because they need to maintain social relations with the target individual group’ (ibid., 185). However, Balzacq’s distinction between ‘moral’ and ‘formal’ support usefully distinguishes between audiences whose support is ‘crucial’ and those whose support is merely ‘relevant’, with crucial support resting on institutionalized rules distributing power among various kinds of political actors, and relevant support being from actors whose moral support is desired. Importantly, institutional rules need not be formal rules. As Vuori argues, the audience is ‘anyone who has to be convinced of the necessity of security action changes within the cultural and political systems in which securitization is taking place’ (2008, 72). Institutionalized rules defining crucial actors in securitization processes can thus be formal rules enacted in laws and constitutions, or less formalized cultural and social norms. Despite acknowledging the relevance of institutional settings, the role of institutions remains underspecified in securitization theory and clear statements about how variation in institutional norms can shape securitization outcomes are lacking. For an account of party bans this is an important shortcoming. Party bans are politically sensitive, constitutionally important decisions affecting a range of fundamental rights and principles in democratic systems, and as such are typically tightly regulated in accordance with explicit legal procedures (Bourne and Casal Bértoa, 2017). New institutionalist approaches can be mobilized to address this shortcoming. They provide many insights and conceptual tools for examining how formal and informal institutional rules may affect public policy outcomes, including securitization processes (March and Olsen, 1989; Hall and Taylor, 1998; Hay, 2002; Schmidt, 2008). In my previous research on party bans in Spain, the utility of veto player theory became apparent (Bourne, 2015). According to Tsebelis, veto players are ‘actors whose agreement is required for a change in the status quo’ (2002, 17). Tsebelis distinguishes

Theory and research design   25 between institutional veto players, or those empowered by formal constitutional rules (such as parliaments and presidents), and partisan veto players, or those ‘generated by the political game’ (such as parties in a coalition government) (ibid., 19). Among other things, veto player theory aims to account for patterns of policy stability and change by focusing on the number of veto players, the ideological distance between them, and the role of agenda setters. Veto players with agenda-­setting prerogatives have significant control over which of the range of possible policies may replace the status quo, providing agreement with other veto players is possible (ibid., 2). Tsebelis also predicts that policy change is more difficult in polities with multiple veto players than it is in those with just one (ibid., 5). These considerations inform operationalization of the hypothesis: H2. Democracies ban anti-­system parties if veto players prefer proscription.

Violence as a context for securitization of anti-­system parties Important methodological differences can be observed in securitization research associated with the Copenhagen School, which has grown from Wæver and Buzan’s work cited above, and that of a ‘sociological approach’, identified with Balzacq’s work (2011). The Copenhagen School works within the poststructuralist tradition, which posits the social power of language and conceives of security as a ‘speech act’ in which security utterances ‘do things’ in the social world. The ‘sociological’ perspective views securitization as ‘a strategic (pragmatic) process that occurs within, and as part of, a configuration of circumstances, including the context, the psycho-­cultural disposition of the audience and the power that both the speaker and listener bring to the interaction’. The approach taken in this research falls within the tradition of the second, sociological approach, in large part because it develops a more elaborate conception of the role of context in securitization. Indeed, Balzacq defines securitization as: … an articulated assemblage of practices whereby heuristic artefacts (metaphors, policy tools, image repertoires, analogies, stereotypes, emotions etc) are contextually mobilized by a securitizing actor, who works to prompt an audience to build a coherent network of implications (feelings, sensations, thoughts and institutions), about the critical vulnerability of a referent object, that concurs with the securitizing actor’s reasons for choices and actions, by investing the referent subject with such an aura of unprecedented threatening complexion that a customized policy must be undertaken immediately to block its development. (Balzacq, 2010, 3) In their seminal contribution, Buzan et al. point to, but do not fully elaborate, the argument that variation in context affects securitization processes. They

26   Theory and research design argue that features of an alleged threat may serve as one of three facilitating conditions for successful securitization and comment that ‘one is more likely to be able to conjure up a security threat if certain objects can be referred to that are generally held to be threatening – such as tanks or hostile sentiments’ (1998, 33). Balzacq’s conception of context is broader in scope and more developed. He argues that the possibility of marshalling assent of an audience … rests with whether the historical conjuncture renders the audience more sensitive to its vulnerability. … Therefore, the positive outcome of securitization, whether it be strong or weak, lies in the securitizing actor’s choice of determining the appropriate times within which the recognition, including the integration of the ‘imprinting’ object – a threat – by the masses is facilitated. (2005, 182) The psycho-­cultural disposition of the audience is also important in Balzacq’s account, because ‘the success of securitization is highly contingent upon the securitizing actor’s ability to identify with the audiences’ feelings, needs and interests’ (ibid., 184). In addition to ‘tun[ing] his/her language to audiences’ experience’, ‘employ[ing] terms that resonate with the hearer’s language’ and ‘identifying the speaker’s “ways” with those of the audience’, a successful speaker of securitization appeals to the collective memories, the Zeitgeist and ‘predominant social views, trends, ideological and political attitudes that pervade the context’ in which the audience is nested (ibid., 184–186; see also Stritzel, 2007, 372). In relation to party bans, a context in which political violence becomes salient, and where an anti-­system party is perceived to contribute to such a context, is likely to facilitate securitization of that party as ‘threatening’ and thereby justifiably banned. This expectation is consistent with the observation that banned parties are often ambiguous in their commitments to peaceful negotiation of political differences. Comparative studies of legal rules on party bans show that a party’s actual promotion of violence, or a party’s potential to incite or provoke it, are among the main justifications inscribed in law for party bans (Brems, 2006a; Issacharoff, 2007; Rosenblum, 2007). Brems identifies three violence-­related legal grounds for banning parties. Parties may be banned if they actually promote violence, which in Europe, the ECtHR has established may involve explicit calls for violence as well as ambiguity about the appropriateness of violence for achieving political ends (Brems, 2006a, 166). Parties may be banned for supporting a terrorist organization. If they do not specifically use or advocate violence they may, in some countries, nevertheless be banned for fear that the parties’ activities will lead to societal unrest and ultimately violent conflict (ibid., 166–169). Furthermore, most of the banned parties listed in the book’s introduction were directly involved in acts of political violence through links with terrorist groups

Theory and research design   27 or participation in coup attempts; or were associated with groups which previously or contemporaneously committed acts of political violence: or might plausibly be seen as capable of doing so, given their ideological position. For instance, former fascist parties banned in Germany, Austria and Italy after their defeat at the end of World War II were implicated in political crimes, mostly mobilized through their control of the coercive apparatus of the state. The Communist Party of the Soviet Union/Russia was (partially) banned for being an accomplice in the 1991 August coup attempt against Soviet President Mikhail Gorbachev (Feofanov, 1993, 637; Brunner, 2002, 30). The Communist Parties of Latvia and Lithuania were involved in unsuccessful coup attempts. The Communist Party of Greece was banned after it launched a revolutionary insurrection and civil war (1946–49) (Kousoulas, 1965). In 2003, the Israeli Central Elections Committee disqualified (later overturned by Supreme Court) the Israeli Arab party, Balad, for denying the existence of the state of Israel as a Jewish and a democratic state and for ‘support for armed struggle, of an enemy state or of a terrorist organisation against the State of Israel’ (Navot, 2008, 750). From 1993, the People’s Labour Party and various successors (including Democratic Party (1994), People’s Democracy Party (2003) and the Democratic Society Party (2009)) have been banned in Turkey on grounds including support for the terrorist group, the Kurdistan Workers’ Party (Celep, 2014, 385–386; Güney and Başkan, 2008, 275). The National Democratic Party (Austria) was banned for reactivating national socialist ideas, but also conducted a terrorist campaign pursuing the return of the predominantly German-­speaking region of South Tirol to Austria (Degenhardt, 1983, 402, 455). In other cases, something less than direct implication in violent acts may contribute to justifications for bans (Rosenblum, 2007, 49; Finn, 1990, 60–61). This was the case for party bans in Germany where ‘the use of violent methods is in no way a prerequisite for bans’ (Backes, 2006, 274) and where neither of the parties banned in the 1950s were considered to present a clear and present danger (Kirchheimer, 1961, 151; Finn, 1990, 6). However, it could be argued that in such cases attitudes to violence, including revolutionary violence and glorification of violence, were integral to what made the anti-­democratic ideology of the party undesirable (see, for example, Rosenblum, 2007, 49; Tomuschat, 1992, 27). In such cases, ambiguity often takes the form of at least rhetorical commitments to democratic practices in the short term, but violent revolutionary ambitions for an unspecified future time. More contemporary examples of proscription prompted by fears that parties may provoke ‘societal unrest and perhaps ultimately violent conflict’ (Brems, 2006, 169), could include, according to Bligh, proscription procedures against the NPD (2013, 1349). In this case, the party’s apparently ‘symbolic’– rather than systematic organizational – links to violent far-­right groups created a situation in which violent actions seemed more likely (ibid., 1349). Similarly, Rosenblum argues that one of the main reasons for banning the right-­wing party Kach in Israel was fear that its leader’s anti-­ Arab statements triggered civil violence (2007, 53). In their study of party bans in Africa, Hartmann and Kemmerzell also argue that the experience of ethnic

28   Theory and research design violence was an important explanation for banning ‘particularlistic parties’ (i.e. those composed of, or acting on behalf of tribal, religious, racial or language groups) (2010, 645 and 656–657). Bligh argues that proscription of parties inciting racism and discrimination address a more generalized threat; namely the danger that such parties will contribute to the creation of a climate of violence or a climate of hate (2013, 1348). These considerations inform operationalization of the hypothesis: H3. Democracies ban anti-­system parties if they do not unambiguously reject violence. Building on Brems’s types of violence-­linked legal justifications for party bans mentioned above (2006a, 166–169), violence as a context facilitating securitization of anti-­system parties can be said to occur if a party’s failure to unambiguously reject political violence amounts to (1) the promotion of violence, (2) support for a terrorist group or (3) creation of a climate in which violent social unrest is feared, and these considerations are invoked in securitization arguments. Desecuritization may be expected in a context in which an anti-­ system party comes to unambiguously reject political violence and this leads decision-­makers to consider that it (a) no longer promotes violence, (b) no longer supports a terrorist group or (c) no longer contributes to a climate of violent social unrest. Furthermore, proposals to ban a political party could be expected to fail if a party’s orientation to violence is not deemed by decision-­makers to (i) promote violence, (ii) support a terrorist group or (iii) create a climate in which violent social unrest is feared. In accordance with the securitization theory’s conception of ‘security’ as intersubjective and socially constructed, it is not expected that the mobilization of arguments about an anti-­system party’s orientation to violence in security discourses will necessarily follow consequential changes in the broader security environment, such as an escalation of political violence or an increase in the number or seriousness or politically motivated acts of violence. Nor is it expected that securitization processes will necessarily be a reaction to consequential changes in an anti-­system party’s orientation to violence. As van Donselaar has argued in relation to parties of the extreme right: There is an element of societal evaluation of the link between violence and extreme right: If a society associates violence with the extreme right, then some form of action is generally taken against the extreme right. This is so in all five countries under discussion [Britain, France, Germany, Belgium and Netherlands]. The association does not in itself have to be factually based or even proven before punitive action takes place. Sometimes powerful action is taken in response to ‘supposed’ or ‘psychological’ involvement by the extreme right in violence. In other cases, actual

Theory and research design   29 participation in violence leads to little or no response. In other words, in serious cases a mild response sometimes takes place, whereas in non-­serious cases there have been instances of harsh responses. (2003, 279) In other words, what matters is whether an anti-­system party’s ambiguity about the appropriateness of political violence is mobilized in an argument that the party poses an existential threat. In Chapters 3, 4 and 5, I examine these propositions empirically in case studies on banned parties, the legalization of parties and failed party ban cases in Spain, the United Kingdom and Germany.

Effectiveness of alternatives to proscription Political scientists have generally been more interested in the effects and effectiveness of party bans than explanations for varying party ban practices. Views on party ban effects and effectiveness are rather mixed. Some argue illegalization may be punishing for the targeted party, not only for the reason that it is denied a chance at winning power but also because the ‘cost of claim­making increases across the board and for particular members’ (Tilly, 2005, 218; see also Bleich, 2011, 101–104). The party’s room for manoeuvre, visibility and mobilizing capacity may be severely curtailed by reductions in its organizational and financial resources, access to the media and through stigmatization (Tilly, 2005; see also Bleich, 2011, 101–104). Those who continue their association with a banned party may face criminal prosecution, limits on employment or truncated political careers (Tilly, 2005, 218). The party ban is, as Koopmans argues, an ‘act of strategic communication in the public sphere’ which, under certain conditions, may serve to deter people who consider committing a similar offence, and helps socialize citizens against extremist orientations by rewarding and satisfying those who refrain from breaking the rule (ibid., 61). As such, party bans may help clarify norms of appropriate behaviour in democratic politics (Erk, 2005, 201; Michael and Minkenberg, 2007, 1115) and thus give parties subject to ban proceedings, or those aiming to succeed a banned party, incentives to moderate ideologically (e.g. Bleich, 2011, 96, 101–103; van Donselaar, 2003, 290). Bans or even ban proceedings may negatively affect the stability of targeted parties, or provoke disarray and internal squabbles (Bleich, 2011, 104; van Donselaar, 2003, 289). In the context of democratization, party bans aiming to prevent the re-­emergence of former non-­democratic ruling parties may contribute to political stabilization in the uncertain conditions of democratic transition, or serve as a ‘backward barrier against resurgence’ (Niesen, 2002, 250; Minkenberg, 2006, 27). Bans on former ruling parties may help symbolically demarcate the authoritarian past, ‘disclose a new republic’s understanding of the paradigmatic wrongs of the old regime’ and ‘specify the new regime’s normative orientation to the future’ (Niesen, 2002, 275).

30   Theory and research design Nevertheless, most scholars working on political extremism tend to see party bans as ineffective, or even counterproductive. It has been observed that restrictions on the party are often only temporary especially if a party has deep roots in a community, or if state authorities are reluctant to prevent the party re-­emerging under a different name (Minkenberg, 2006, 37; Husbands, 2002, 64; Downs, 2012, 42; Bleich, 2011, 87–86, 96). Strategies of isolation, such as party bans, may create a context where ‘the perception of the parties of the putatively democratic “establishment” allying to deny voice to a party or parties they deem illegitimate can ultimately serve to fuel [a party’s] appeal’ (Downs, 2012, 48). Legal proceedings to ban parties may increase public exposure and opportunities to claim martyrdom or reinforce anti-­establishment critiques (Gordon, 1987, 391; Downs, 2012, 43; Bleich, 2011, 101–103). Proscription may lead to radicalization, a growth of militancy and readiness to use violence (Capoccia, 2005, 59). The party ban is not a suitable mechanism for the ‘civic re-­education’ of extremists (Husbands, 2002, 64) and may merely treat the symptoms rather than the more complex underlying causes of dissatisfaction with the status quo (Pedahzur, 2004; Backes, 2006, 281). In the context of democratic transitions, party bans can be one means by which rulers in new democracies manipulate electoral processes and thereby stall or undermine the consolidation of democratic institutions and practices (Bourne, 2012a; Niesen, 2012, 549–552). Party bans may weaken parliamentary forces vis-­à-vis other structures of power in the state, such as the coercive apparatus; misrepresent the impact of the banned party on the political system; and damage a state’s international standing (Kirchheimer, 1961, 159–161). Many further argue that party bans may damage the foundations of a democratic polity. The two most cited typologies of strategies of ‘democratic defence’ – those developed by Downs (2012) and Capoccia (2005) – conceptualize party bans as the most ‘militant’, ‘intolerant’, ‘aggressive’ or ‘repressive’ measure that can be employed against anti-­system parties. They see it as the most risky in terms of possible damage to the overall quality of the democratic system (Downs, 2012, 49; Capoccia, 2005, 59). Others observe that the party ban could be interpreted as ‘lack of faith in the democratic process’ and an ‘admission of failure’ (Gordon, 1987, 390), or that it can be counterproductive insofar as its ‘chilling effect’ may signify a silent weakening of democratic rights in the state (Niesen, 2002, 256; see also Niesen, 2012; Kirchheimer, 1961, 151–161; Wise, 1998, 302). Alongside uncertainty about the effectiveness of party bans, these purported longer-­term risks to the quality of democratic politics have led many to caution against their use and urge that the ban should at least be a measure of last resort. Downs describes the party ban as a ‘democracy’s strategic last card’, suggesting that bans are used when all other measures have proved futile (2012, 38–39; see also Avineri, 2004, 7). He also argues that mainstream parties may strategically sequence responses in the ‘search for some approach that either moderates or eliminates the offending party’ (2012, 18). Guidelines on the Proscription and Dissolution of Political Parties and Analogous Measures drawn up in 1999 by the Venice Commission, a Council of Europe advisory body, cautions

Theory and research design   31 that proscription is ‘a particularly far-­reaching measure [which] should be used with utmost restraint’ and proposes that using ‘less radical measures could prevent the said danger’ (Guideline 5). In contrast, many authors argue that important alternatives to party bans are effective at marginalizing or moderating anti-­system parties and some explicitly argue that certain alternatives to party bans are more effective. Most of these purportedly effective alternative strategies are classified as less repressive, more tolerant and more accommodative than party bans in Downs’ (2012) and Capoccia’s (2005) typologies. Of these proposed alternatives, the use of electoral rules, such as use of the plurality electoral formula in single-­member constituencies or electoral thresholds, are often singled out as more effective means for dealing with anti-­system parties than measures like party bans (Sartori, 2001, 99; Downs, 2012, 23; Backes, 2006, 281; Navot, 2008, 747; Gordon, 1987, 395; Pedahzur, 2004, 118). The so-­called cordon sanitaire, a political practice whereby mainstream parties collude to keep anti-­system parties out of office, has been seen as ‘stingily successful’ at achieving this goal in some instances (Downs, 2012, 109; see also Art, 2007, 332), although views on whether it undermines the appeal of anti-­system parties is more mixed (compare Downs, 2012, 21 and 84 and Art, 2007, 332). Others have argued that strategies favouring collaboration between mainstream and anti-­system parties in government may encourage moderation (Downs, 2002, 49, 2012, 21; van Spanje and van der Brug, 2007, 1036; de Lange, 2007, 34; Bale, 2003, 70), ‘expose’ anti-­system parties as ‘ill-­prepared to deal with the responsibilities of everyday policymaking’ (Downs, 2012, 21) or unleash debilitating internal tensions within anti-­system parties (de Lange, 2007, 27 and 23; Kestel and Godmer, 2004). Husbands suggests that ‘criminal prosecution and imprisonment of individual members of the leadership [may be] more effective or may by default amount to the same thing [as proscription]’ (2002, 61). More generally, longer-­term strategies against anti-­system parties and movements, including state support (financially or otherwise) for civil society initiatives mobilizing in opposition to anti-­system parties or the implementation of longer-­term educational and social initiatives, such as anti-­racism campaigns and ‘civics’ classes, may be more effective insofar as they address symptoms, rather than causes of anti-­system support (Husbands, 2002; Pedahzur, 2004). Where such strategies build a ‘vital “civic culture”, anchored in civic virtues and a general consensus on the fundamental values and rules’, Backes argues, this ‘ought to set narrow boundaries for extremists’ attempts to gain influence’ (2006, 281). Alongside normative appeals calling for restraint in the use of party bans, arguments pointing to the effectiveness of less repressive alternatives to party bans provide theoretical support for the hypothesis that: H4. Democracies ban anti-­system parties if alternative forms of marginalization are not effective.

32   Theory and research design Operationalizing the ‘effectiveness of alternatives to party bans’ is complex, in large part due to the wide range of measures and possible effects. Table 1.2 provides an overview of arguments in the literature about the nature of alternative measures, principal objectives pursued and suggests indicators for evaluating their effectiveness. In this research, I address the effectiveness of electoral rules and collusion among mainstream parties to exclude anti-­system parties from governing with reference to the goals and indicators in the first two rows of Table 1.2. Effectiveness of electoral system rules and a policy of cordon sanitaire are arguably the measures most directly affecting the opportunities for anti-­system parties to influence public policy. They can be examined with analytical tools of established literatures on electoral systems and government formation. The electoral system can be an effective alternative to banning parties if it prevents anti-­system parties from winning any seats, or only a small number of seats, in parliament (see Table 1.2). In such circumstances, it could be expected that decision-­makers and citizens will reason that electoral systems contain anti-­ system parties sufficiently well to make the option of banning them unnecessary. Anti-­system parties are often small parties and as such, the effects of electoral Table 1.2  Alternatives to party bans, objectives pursued and indicators of effectiveness Alternative to party bans

Principal objectives pursued

Indicators of effectiveness

Electoral rules

Prevent party winning Anti-system party does not win seats in seats or restrict a party’s electoral system where it might under a influence in parliament more favourable system. If it wins seats, the electoral system prevents anti-system party from winning seats more often than not and/ or anti-system party is underrepresented in terms of its seat share.

Cordon sanitaire

Prevent a party influencing public policy through participation in government or legislative coalitions

Governments are formed without an antisystem party subject to a policy of cordon sanitaire. Minority governments are maintained without a legislative pact with the anti-system party or without collaboration with it to pass legislation.

Restrictions on speech, criminalization, collaboration in governing

Induce ideological moderation by constraining political opportunities of antisystem parties

The public pronouncements of an antisystem party no longer reject core dimensions of liberal democratic systems, such as representative government, political equality of all citizens and the non-violent resolution of conflict.

All above measures

Reduce incentives for citizens to vote for antisystem party

Reduction in anti-system party vote share after measures implemented

Theory and research design   33 rules on the marginalization of anti-­system parties can be helpfully examined with reference to known electoral system effects on small parties (e.g. Lijphart, 1994; Gallagher and Mitchell, 2005; Norris, 2005; Carter, 2005). Characteristics of the electoral system affecting proportionality of seats obtained given vote share are particularly important in this regard, including varying electoral formulas, district magnitude and ‘legal’ electoral thresholds. In some cases, it is more straightforward to identify electoral system effects on individual anti-­system parties than in others. It is simplest in relation to legal thresholds. Where a party wins enough votes to pass an assembly-­level ‘natural’ threshold, but not a ‘legal’ threshold, the effect of the latter is apparent. In other cases, judgements about electoral system effects on individual anti-­system parties must be extrapolated from more general knowledge generated from existing studies on the effects of different electoral systems (e.g. Lijphart, 1994; Gallagher and Mitchell, 2005; Taagepera and Sugart, 1989; Norris, 2005; Carter, 2005; Farrell, 2011). The underrepresentation of anti-­system parties in terms of seats they win relative to their votes can be captured by examining an anti-­system party’s ‘advantage ratio’ (Taagepera and Shugart, 1989, 68). The cordon sanitaire can include a refusal to do all or some of the following: to form electoral cartels or collaborate in the formation of party lists prior to elections; to rely on anti-­system parties’ votes to pass legislation; to give support for any of the party’s candidates or proposals; or to permit anti-­system parties’ participation in governmental coalitions (Art, 2007, 339; Downs, 2002 and 2012; Geys et al., 2006; van Spanje and van der Brug, 2007). The cordon sanitaire, like the party ban, can be employed to prevent an anti-­system party from participating in government and enjoying the spoils of office. The cordon sanitaire does so by denying anti-­system parties the chance of mobilizing their electoral strength through governing. Neither the banned party, nor the legal but politically isolated party, can use participation in government or collaboration with other parties in parliament to pursue goals that threaten democratic institutions, practices and values. The clearest indication that a cordon sanitaire is successful is when, following agreement by mainstream parties of a policy of exclusion, and following elections in which an anti-­system party wins seats in parliament, governments are formed that consistently exclude an anti-­system party, and that anti-­system party remains excluded until the next election. In situations of minority government, a strategy of exclusion can be considered successful where the cordon sanitaire is applied systematically and mainstream parties refuse to enter into legislative pacts or more ad hoc collaboration to pass legislative measures (Strøm, 1984; Laver and Schofield, 1998). Where electoral system effects on the marginalization of anti-­system parties and practices of cordon sanitaire point to contradictory conclusions about the effectiveness of alternatives to party bans, it can be argued that effective policies of cordon sanitaire are a more important alternative to banning parties than electoral system rules. Without the ability to directly influence legislation or government policy, anti-­system parties are unlikely to threaten democratic institutions, practices and values.

34   Theory and research design As suggested above, it can be expected that judgements about the effectiveness of alternatives to party bans will facilitate the securitization of anti-­system parties as an existential threat as well as desecuritization processes leading to legalization of banned parties. Where alternatives to the party ban cannot effectively marginalize an anti-­system party, the case for using the exceptional measure of the party ban to deal with the threats that it purportedly poses are likely to be strengthened. Where alternatives effectively marginalize anti-­system parties, it is likely to be easier to make the case that a party ban is no longer needed, which may contribute to legalization of banned parties. Similarly, a plausible argument that party ban alternatives are effective may undermine attempts to securitize an anti-­system party as threatening and thereby prevent initiatives to ban parties from succeeding. In Chapter 6, I examine electoral system effects on the fortunes of HB and its successors, Sinn Féin and Republican Clubs, the SRP and the NPD at regional and national/federal levels, while in Chapter 7, I examine whether practices of government formation and parliamentary cooperation in Spain, Britain and Northern Ireland (NI), and Germany systematically exclude these parties from government.

Democratic competition and government formation In their accounts of party ban processes, various authors link party ban decisions with the rational choices of mainstream parties’ in their pursuit of votes, policy goals and political office. Capoccia’s 2005 book Defending Democracy is probably the best-­known study dealing with these themes. Aiming to explain the breakdown of democratic regimes in Germany and Italy, and its maintenance – despite the existence of serious extremist threats – in Czechoslovakia, Belgium and Finland, Capoccia examines incentives facing parties within party systems characterized by ‘polarised pluralism’. He argues that in a context of polarized pluralism, the strategies of ‘border parties’ – or those parties adjacent to anti-­ system parties occupying extreme wings of the ideological spectrum (ibid., 17) – are crucial. This is because, as a consequence of the ‘outbidding strategies’ of extremist parties, the conditions of polarized pluralism produce ‘centrifugal competition’ forcing all parties, but particularly border parties, towards extreme positions. In this context, a border party’s decision about whether or not to cooperate with extremists or embrace strategies of democratic defence may hold the key to the regime’s capacity to react to the challenge of political extremism (ibid., 19). However, the contemporary applicability of Capoccia’s analysis is limited by its reliance on Sartori’s party system theory. As Mair observes, one of the difficulties of employing Sartori’s party system types – ‘two-­party’, ‘moderate pluralism’ and ‘polarized pluralism’ – is the contemporary overcrowding of cases of moderate pluralism and a virtual emptying out of two-­party and polarized pluralism categories following the eclipse of fascist and communist anti-­system parties, particularly after the collapse of the Soviet Union (1997, 203).

Theory and research design   35 Downs’ 2012 book Political Extremism in Democracies builds on Capoccia’s approach, arguing that ‘[m]ainstream parties must contend with the consequences of different strategic responses for their own popularity with voters as well as their status as potential (or actual) parties of government’ (2012, 175). Regarding mainstream parties’ choices to opt for strategies for co-­optation (taking on extremist parties’ policy programmes) and collaboration (e.g. entering into a coalition and/or parliamentary cooperation with an extremist party) he formulates the hypothesis that: greater party system fragmentation (number and relative size of parties, as well as degree of polarization) produces increased incentives for individual parties to defect from clean hands approaches and instead pursue strategies of engagement (co-­optation and collaboration). (Ibid., 63) This is because, Downs argues, Relatively weak and divided democratic parties must either (1) adjust policy positions and thereby recoup voters lost to the extremes [co-­opt] or (2) cooperate to some extent with the pariah [collaborate]. Otherwise, such parties will suffer individually as vote and office-­seeking organisations. (Ibid., 63) Regarding strategies of isolation, which includes party bans, Downs argues that the opposite holds, or that ‘polarised, centre-­fleeing party systems therefore pose greater challenges for containing extremism through coordinated political isolation’ (ibid., 63). Presumably, in less fragmented, centre-­leaning party systems, mainstream parties would have more incentives to ignore anti-­system parties and fewer difficulties building anti-­pariah coalitions across the ideological centre of the party system. While Downs’ work presents promising hypotheses for understanding party system constraints on mainstream parties’ responses to anti-­system parties at a general level, it is a hypothesis whose analysis is better suited to ‘large n’ statistical analysis than the case study approach employed here. Indeed, levels of party system fragmentation in general and at key party ban decision-­points are similar in the cases examined in the book and compared to other political systems in Europe tend to have moderate to low levels of party-­system fragmentation.1 Nevertheless, many have argued that mainstream parties’ support for or against banning a party may be shaped by the strategic goals of winning votes, or attaining policy and office goals. Müller (2005) argues that parties located adjacent to a banned party in a policy space are more likely to attract the banned parties’ voters, although, on the other hand, these voters may conceivably reject the closest ideological substitute if it supports a ban on their preferred party (ibid., 263). Kirchheimer also remarks in the context of widespread fears of domestic communism in 1950s USA that it was ‘tempting’ for mainstream

36   Theory and research design politicians to ‘prove one’s unshakable resolve to “fight subversion” ’ and, as such, ‘endorsement of more stringent repressive [measures against the Communist Party and its members] meant protection from unfavourable public reaction’ (1961, 158). In contrast, Kemmerzell argues that, in South Africa, banning extremist fringe parties was contrary to the interests of the two main parties – the governing African National Congress and the New National Party – because this would undermine the parties’ goal to remain dominant within their respective political constituencies through ‘broad church’, inclusive electoral strategies (2010, 701). As mentioned below, Downs has made the point that strategies of isolation, such as party bans, may cost mainstream parties votes because ‘the perception of the parties of the putatively democratic “establishment” allying to deny voice to a party or parties they deem illegitimate can ultimately serve to fuel the far right’s appeal’ (2012, 48). The NPD’s ‘success in the courts “fighting the system” ’ in the 2003 party ban case has been seen as a case in point, insofar as ban proceedings ‘generated considerable and sustained public attention and was followed by unusually strong electoral showings in Saxony (9.2 per cent of the vote) in 2004 and in Mecklenburg-­West Pomerania (7.3 per cent of the vote) in 2006’ (Downs, 2012, 43). While the argument that mainstream parties’ support for banning parties or otherwise might be linked to a desire to win votes is undoubtedly a plausible argument, it will not be considered in this study. Analysing voting outcomes is complex and would require detailed analysis of survey data in regions (i.e. Northern Ireland, Lower Saxony and Bremen) at points in time (e.g. the 1950s) where such survey data may be unavailable or difficult to obtain. In addition to the goal of winning votes, decisions about whether or not to ban parties may have consequences for achieving mainstream parties’ policy preferences or ‘office seeking goals’ (Müller and Strøm, 1999, 5). Murphy argues, for instance, that despite the constitutional prohibition on the reconstitution of the interwar fascist party in Italy, the Christian Democrats, when dominant in the party system, did not apply anti-­fascist laws to the blatantly neo-­fascist, Movimento Sociale Italiano (MSI, Italian Social Movement) ‘in return’ for the occasional support of the MSI in keeping the Christian Democrats in office (1993, 189–190). Party bans – like the analytically similar cordon sanitaire – may also have mechanical effects changing the ‘mathematical context’ of coalition negotiations (Geys et al., 2006, 979). As Casal Bértoa and I argue, party bans, legalization of banned parties and failed party bans may have significant consequences for the structure of competition in electoral, parliamentary and governmental arenas, where the party in question is a ‘relevant’ party (Casal Bértoa and Bourne, 2017). This includes changes in the class of party system and patterns of interparty cooperation. In Spain, for example, the 2009 ban on the Batasuna successor party, EHAK/PCTV (Communist Party of the Basque Homelands) opened the way for the first and only non-­Basque nationalist-­led government in the Basque autonomous community, which was headed by the Basque branch of the state-­wide Partido Socialista Obrero Español (PSOE, Spanish Socialist Workers Party) with support from the Basque branch of the state-­wide Partido Popular (PP, Popular Party) (Casal Bértoa

Theory and research design   37 and Bourne, 2017, 456). The reappearance of Sortu and EH Bildu in 2012 prevented a repetition of this non-­Basque nationalist government. To the extent these changes could be anticipated, we argued, those parties likely to benefit from party bans – in this example PSOE and PP – would have incentives to support a party ban (ibid., 460–462). Finally, van Spanje’s (2010) work on the ostracism of anti-­immigrant parties provides additional insights for understanding mainstream parties’ choices in relation to party bans. According to van Spanje, ostracism is a strategy for marginalizing parties and occurs when a party systematically rules out cooperation with a specific other party (2010, 356). As such, it is a strategy analytically similar to that of party bans, which also aims to marginalize particular parties albeit by denying it any chance to win votes, influence policy or participate in government. Van Spanje argues that rationales for ostracism are captured in a rational choice framework. Given competition among parties for policy, office and votes, party leaders have incentives to try to ostracize as many of their competitors as possible (ibid., 356). However, it is not always possible for mainstream parties to know which specific parties they may need to cooperate with after an election. As such, a party is less likely to ostracize potential coalition partners, particularly influential parties or those ideologically close to them (ibid., 356). Moreover, supporting ostracism of another party risks electoral losses and intraparty conflict unless it can ‘credibly accuse a particular party of ideologies that are widely perceived as unacceptable’ (ibid., 374). Van Spanje’s analysis shows that mainstream parties are likely to ostracize an anti-­immigrant party ‘if they do not need to cooperate with it’ (ibid., 354). As such small parties are more likely to be ostracized, while one party is more likely to ostracize another as the ideological distance between them grows (ibid., 373). Furthermore, parties are even more likely to ostracize an anti-­ immigrant party ‘if they can convincingly make the case that its ideologies are outside agreed standards of acceptability’ (ibid., 355). As such, neo-­Nazi and authoritarian anti-­immigrant parties are much more likely to be ostracized than a ‘neo-­liberal’ one (ibid., 371). These arguments suggest that in a context of democratic competition, party ban decisions might be influenced by the strategic calculations of mainstream parties with regard to anti-­system parties. More specifically, arguments about mainstream parties’ office and policy seeking goals, and the previously discussed hypothesis focusing on the role of veto players in ban decisions, suggest the following hypothesis: H5. Democracies ban anti-­system parties if partisan veto-­players do not need to cooperate with them to win and maintain office and achieve policy goals. In Chapter 7, I address this hypothesis by drawing on the coalition formulation literature, which has generated various hypotheses about the strategic

38   Theory and research design considerations of parties when it is necessary for them to cooperate with other parties to win office of achieve policy goals. As I discuss in more detail in the chapter, this literature points to three ways of estimating the potential effects of banning a party for remaining mainstream parties’ chances of forming a government. First, it is argued that partisan party-­ban-veto-­players are more likely to support banning an anti-­system party that could only participate in a relatively low number of potential winning coalitions. This coalition potential is indicated by an anti-­system party’s Penrose-­Banzaf score, which measures the relative power of different actors in weighted voting systems when winning coalitions must be formed (Banzaf, 1964). A second way to estimate the effects of banning an anti-­system party on partisan party-­ban-veto-­players’ ability to achieve office and policy goals is to determine whether an anti-­system party occupies the position of a ‘median legislator’ on one or other salient policy dimension. According to coalition theory, the median legislator occupies a central position in a policy space and will be pivotal for coalition formation because the median legislator must be in the majority of every policy vote (Budge and Laver, 1986). Thus, where an anti-­system party comes to occupy the median position on some salient policy dimension, it can be expected that mainstream parties will be more likely to support banning it if the anti-­system party does not occupy the position of median legislator. A third measure for estimating the potential effects of banning an anti-­system party for partisan party-­ban-veto-­players’ achievement of policy and office goals is to examine membership of theoretical ‘minimum winning coalitions’ (Riker, 1962) or ‘minimal connected winning coalitions’ (Axelrod, 1970) following elections. If (1) a partisan party-­ban-veto-­player could not have formed a coalition of these types with the anti-­system party facing ban proceedings at some point in recent past elections, or if (2) the partisan party-­ban-veto-­ player could have formed such coalitions with the anti-­system party, but there were other coalitions of these types that could be formed without it, then the partisan party-­ban-veto-­player would be more likely to support banning the anti-­ system party in question.

Expected relationships between variables This chapter has presented the research design and theoretical framework for the research described in Chapters 3 to 7. The foregoing analysis suggests that anti-­ system parties’ orientations to violence, effectiveness or otherwise of alternatives to bans, processes of securitization and desecuritization, variation in veto-­player preferences and anticipated effects of bans on partisan party-­banveto-­players’ office and policy goals are linked in the following manner. Party bans can be expected to occur when (1) the ambiguity of an anti-­system party about the appropriateness of political violence, and (2) ineffectiveness of alternatives to a party ban, facilitates (3) securitization of that party as an existential threat and identification of the party ban as a solution to that threat, a framing which (4) all veto players accept, including (5) partisan party-­ban-veto-­players, who are unlikely to be negatively affected in terms of policy and office goals.

Theory and research design   39 It is expected that initiatives to ban parties fail if at least one of these conditions is absent. In contrast, legalization of banned parties can be expected to occur when (1) an anti-­system party’s clear renunciation of political violence and (2) existence of effective alternatives to marginalize an anti-­system party, facilitates (3) the desecuritization of anti-­system parties as an existential threat and therefore rejection of the party ban as an appropriate measure, which (4) all veto players accept, including (5) partisan party-­ban-veto-­players, who are now likely to be negatively affected by a party ban for achieving policy and office goals. In the next chapters, I present findings of empirical research which aim to test whether these expectations were observed in three kinds of party ban cases (bans, failed bans and legalizations of banned parties), observed at ten decision points, involving five parties (including where relevant their successors), observed in three countries (Spain, Germany and the United Kingdom). Before turning to this analysis, I describe, in Chapter 2, relevant legal regimes for banning political parties in the country case studies.

Note 1 The effective number of parliamentary parties (ENPP) is a common way of measuring party system fragmentation. Laakso and Taagepera’s (1979) formula measures how many parties find representation in parliament after an election, weighted by the number of seats each party wins (i.e. ENPP = 1/Σsi², where si = the proportion of seats of the ith party). On average for Spain (1977–2016) ENPP = 2.89 and for elections prior to HB and successor bans (2000 2.48) and prior to Sortu and Bildu legalizations (2011 2.6); on average for Basque Autonomous Community (1990–2012) ENPP = 4.46 and for elections prior to Herri Batasuna and successors ban (2001 3.36) and for election prior to Bildu and Sortu legalization (2009 3.29); on average for NI (1945–69) ENPP = 2.2 and elections prior to Sinn Féin and Republican Clubs bans (1953 1.8, 1965, 1.95); for the Westminster Parliament (1945–2015) ENPP = 2.18 and elections prior to Sinn Féin and Republican Club legalizations (1970 2.07, February 1974 2.25); on average for Germany (1949–2013) ENPP = 3.39 and elections prior to ban (1949 4.65) and failed bans (1998 3.31, 2009 4.83, 2013 3.51; on average for Lower Saxony election prior to ban (1951 4.03; on average for Bremen election prior to ban (1951 4.08); on average for NPD ban prior to NPD 2017 ban proceedings in Saxony (2014 3.51, 2009 3.67) and Mecklenburg-­Vorpommern (2006 3.98 and 2011 3.81). Comparative data from the Who Governs in Europe database shows that ENPP for all the cases analysed in the book fall within moderate to low levels of party system fragmentation (Casal Bértoa, 2016).

2 Ban regimes and banned parties in Spain, the United Kingdom and Germany

The decision to ban a political party in a democracy is ultimately a decision about who is entitled to contest political power (Finn, 1990, 54, 56). It involves an explicit and often controversial mobilization of the ‘coercive power of the state’ (ibid., 56). Public authorities will need to weigh the effects of party bans on fundamental rights. Abuses of power need to be avoided. As such, the proscription of political parties is typically the subject of detailed regulation in constitutional and legislative instruments and often subject to judicial ratification. In this chapter, I discuss in detail the legal rules for party bans in Spain, the United Kingdom and Germany and their application in specific ban cases.

Party bans in Spain There are two main legal pathways for banning political parties in Spain, one drawing on party law and the other on the Criminal Code. Since the late 1990s, both of these pathways have been employed extensively to ban parties, organizations and electoral groupings (Tajadura and Vírgala, 2008; Bourne 2018). Article 6 of the 1978 Spanish Constitution provides that the creation of parties and the exercise of their activities are free insofar as they respect the Constitution and the law. Their internal structure and functioning must be democratic. Article 22 recognizes the right of association, while specifying that associations pursuing ends or using means legally defined as criminal are illegal and that secret and paramilitary associations are prohibited. These provisions were elaborated in more detail in Law 54/1978 of 4 December, on Political Parties, which was superseded by Organic Law on Political Parties 6/2002 of 27 July, as amended by Organic Law 3/2015 of 30 March, as well as various iterations of the Criminal Code, notably Organic Law 10/1995 of 23 November and amendments introduced by Organic Law 5/2010 of 22 June. Law 54/1978 on Political Parties was much shorter than its successor, parsimoniously permitting proscription of political parties proven to be ‘illicit associations’ or parties ‘whose organization and activities were contrary to democratic principles’ (Article 5). Organic Law 6/2002 of 27 July, as amended by Organic Law 3/2015 of 30 March (henceforth 2002 Party Law) overhauled the earlier

Ban regimes and banned parties   41 Party Law and provided an additional justification for proscription; namely, conduct threatening to undermine the liberal democratic system. This provision permitted the courts to dissolve parties that violated democratic principles, particularly those that, in a repeated and grave form, aimed to undermine or destroy the regime of liberties, or injure or eliminate the democratic system (Article 9.2). Conduct deemed to violate democratic principles mostly relate to a party’s orientation to violence, as the italicized (my italics) parts of Article 9.2 of the 2002 Party Law reproduced below shows: a

b c

Systematically violating liberties and fundamental rights, promoting, justifying or excusing attacks on people’s lives or physical integrity or exclusion or persecution of people because of their ideology, religion or beliefs, nationality, race, sex or sexual orientation Encouraging, promoting or legitimizing violence as a means to achieve political goals, or to destroy the conditions necessary for the exercise of democracy, pluralism and political liberties Complementing or supporting politically the action of terrorist organizations in the pursuit of aims to subvert the constitutional order, gravely alter public peace, trying to submit public authorities, individuals or social groups or the general population to a climate of terror, or contributing to multiply the effects of terrorist violence and the fear and intimidation it generates.

Article 9.3 spells out in detail the acts which, if repeated and accumulated, would be taken as evidence that acts listed in article 9.2 a–c had occurred. These are almost exclusively actions supporting, encouraging and defending political violence or the actions of terrorist groups, or collaborating with those that did so. The acts listed in Article 9.3 are: a

b

c

d

Giving explicit or tacit support to terrorism; legitimizing terrorist actions pursuing political ends other than through peaceful or democratic means; excusing or minimizing the significance of terrorist actions, or the violation of fundamental rights they entail Actions that foment a culture of adverserialism and civil confrontation, or which aim to intimidate, block, neutralize or socially isolate those who oppose terrorism, and make them live in an atmosphere of coercion, fear, exclusion or privation of basic rights such as freedom of opinion or the right to participate in public affairs (article 9(3)b) Regularly including in party management or electoral lists, people convicted for terrorist crimes who have not publically renounced terrorism; or including a large number of people involved in groups linked to terrorist organizations within party ranks, except when disciplinary measures have been taken to expel them Using symbols, messages or references that can be identified with terrorism or violence

42   Ban regimes and banned parties e f g h i

Ceding to terrorists or their supporters, the rights and privileges granted to political parties Habitual collaboration with groups which systematically act in agreement with terrorist or violent organizations, or with groups which protect and support terrorists Deploying administrative, economic or other resources in public authorities where the party governs to support terrorist and violent organizations Promote, protect or participate in activities that aim to reward, pay tribute or distinguish terrorists, violent people or those who collaborate with them. Protect activity creating disorder, intimidation or social coercion that is linked to terrorists or violent people (article 9(3)i).

The focus on conduct in the 2002 Party Law reflects a conscious rejection of a German style ‘militant democracy’ by the drafters of the 1978 Spanish constitution, those approving the 2002 Party Law and the Courts. Parties writing the 1978 Spanish constitution rejected amendments, such as competence for the Constitutional Court to decide on ban cases, that they thought would establish a militant democracy (Blanco Valdés, 1990: 130). Furthermore, the Spanish constitution contains no explicit provisions authorizing measures against unconstitutional parties and according to its article 168, the constitution is open to ‘total’ revision. The Constitutional Court clarified very early on that, despite this ideological liberty, political parties must pursue their objectives through means permitted by the constitution (de Otto Pardo, 1985, 45; Tajadura and Virgala, 2008: 33). In contrast to the German model of militant democracy, which permitted the proscription of parties for espousing anti-­constitutional ideas, the Spanish model is, then, according to most legal scholars, an ‘open’ or ‘tolerant’ model, in which all ideas and political projects could be pursued, even wholesale reform of the constitution, so long as it was done using democratic procedures set out in the constitution (de Otto Pardo, 1985; Blanco Valdés, 2004, 53 and 57; Hinarejos Parga, 2004, 117–118 and 126). In line with this tradition, the preamble of the 2002 Party Law states ‘any project or objective is understood as compatible with the constitution, so long as it is defended through activities that do not violate democratic principles or fundamental rights of citizens’. Negotiation of the 2002 Party Law opened up the question of whether the new law undermined this interpretation, and was one of the grounds on which the Basque Government challenged the constitutionality of the 2002 Party Law (for other dissenters, see Pérez Royo 2002; Montilla, 2003, 568). Forced to address the question of militant democracy, the Constitutional Court confirmed the view that the Spanish legal order was an open or procedural order and that the 2002 Party Law did not transform it into a militant democracy because only ‘conducts’, such as support for terrorist organizations, were contemplated as grounds for illegalization, rather than a party’s ultimate objectives. With the exception of the electorally inactive Spanish Communist Party (Reconstituted), suspended in 2003 but banned in 2006, the radical Basque nationalist party HB and its successors have been the only parties banned under

Ban regimes and banned parties   43 Party Law in democratic Spain. In the 1980s, the judiciary overruled government attempts to exclude HB from the political arena by denying it registration as a political party. Invoking the 1978 Party Law, the Interior Ministry sought to deny HB registration on the grounds that its statutes contained formal defects including a failure to formally declare respect for the constitution. Soon afterwards, the Public Prosecutor initiated proceedings to ban the party on the grounds it was an illicit association. Both cases were rejected by the Supreme Court (Esparza, 2004, 145; Fernández Hernández, 2008, 176–179). In 2003, the Supreme Court ruled that HB and its successors Euskal Herritarrok (Basque Citizens, EH) and Batasuna (Unity) were subject to the strategy and mandates of ETA and for this reason banned them. This ruling was later endorsed by both the Spanish Constitutional Court and the ECtHR. After their proscription, HB, EH and Batasuna repeatedly sought to return to electoral politics under a different guise. Envisioning this possibility, Article 12.1.c) of the 2002 Party Law prohibits the creation of a new party or use of other registered parties which ‘continue or succeed the activity of a party declared illegal and dissolved’. The 2002 Party Law also amended Ley Orgánica 5/1985, de 19 de junio, del Régimen Electoral General (Organic Law on the General Election Regime, LOREG) in order to prohibit electoral groupings (agrupaciones de electores) from presenting ostensibly ‘independent’ candidates which would continue or serve as a successor to the activity of a banned, dissolved or suspended political party (Article 44.4 LOREG). Ley Orgánica 3/2011, de 28 de enero, further modified the LOREG to extend the prohibition to include parties, federations or coalitions which presented candidates in elections which succeeded or continued the activities of an illegal, dissolved or suspended party (Article 44.4 LOREG) and, among other things, permitted the courts to annul candidacies on these grounds during an election campaign. Indications that a party succeeded another were (a) similarities in organization, structure and functioning; (b) presence of people from banned parties in another party’s organization and electoral lists; (c) overlapping financial resources; (d) any other relevant circumstances such as attitudes to violence and terrorism (Article 12.3 2002 Party Law; Article 44.4 LOREG). The Supreme Court made numerous subsequent rulings to block successor parties, most of which the Constitutional Court endorsed (see Table 2.1). In 2011, the Constitutional Court rejected a Supreme Court ruling that Bildu (Reunite) was an instrument of ETA. Soon afterwards, in 2012 the Constitutional Court overturned a Supreme Court ruling and effectively legalized Sortu. Since then no other party ban cases have been launched. As mentioned, the Criminal Code provides a second pathway for the illegalization of political parties. In its current form, a terrorist organization is defined in the Criminal Code with reference to a goal to subvert the constitutional order; suppress or seriously destabilize functioning of political institutions, economic or social structures of the state; oblige public powers to undertake or not to undertake an act; gravely alter the public peace; seriously destabilize the functioning of an international organization; or provoke a state of terror in

Table 2.1  Failed party bans, illegalization and legalization of parties in Spain Party

Ban year Type of ban procedure

Herri Batasuna 1984 Herri Batasuna 1986 Herri Batasuna, Euskal Herritarrok, Batasuna 2003

Non-registration (overruled) Non-registration (overruled) Parties declared illegal and dissolved

Autodeterminaziorako Bilgunea and others Herritarren Zerenda Aukera Guztiak Partido Comunista de España (Reconstituido)

2003 2004 2005 2006

Disqualification of electoral grouping lists Disqualification of electoral grouping list Disqualification of electoral group candidacies Declared illicit organization and dissolved

Abertzale Sozialisten Batasuna Eusko Abertzale Ekintza Abertzale Sozialistak Eusko Abertzale Ekintza

2007 2007 2007 2007 2008

Party denied registration Disqualification of party lists Disqualification of electoral grouping candidacies Disqualification of electoral grouping candidacies Party declared illegal and dissolved

Euskal Herrialdeetako Alderdi Komunista

2008

Party declared illegal and dissolved

Askatasuna Demokrazia Hiru Milioi Iniziatiba Internazionalista – Herrien Elkartasuna Sortu

2009 2009 2009

Bildu

2011

Disqualification of party lists Disqualification of electoral grouping candidacies Disqualification of electoral coalition lists (overruled by Constitutional Court) Non-registration (overruled by Constitutional Court in 2012) Disqualification of electoral coalition candidacies (overruled by Constitutional Court)

2011

Grounds for ban Formal defects Illicit association Complement to and support for a terrorist organization Batasuna successor Batasuna successor Batasuna successor Integration in terrorist organisation GRAPO Batasuna successor Batasuna successor Batasuna successor Batasuna successor Collaboration with the illegal Batasuna and ETA Collaboration with the illegal Batasuna and ETA Batasuna successor Batasuna successor Batasuna successor Batasuna successor Batasuna successor

Ban regimes and banned parties   45 the population or a part of it (Article 573). It may take the form either of a stable or permanent group of more than two people who jointly and in a coordinated manner divide tasks and functions for criminal purposes or simply as two or more people who jointly aim to commit terrorist crimes (Articles 571, 570 bis and 570 ter). Article 571 defines terrorist organizations as a form of criminal organization, the regulation of which (in Article 570 quáter) provides for the dissolution of terrorist organizations or where appropriate other sanctions, including suspension of activities or closure of its establishments for periods of no longer than 5 years (Articles 33.7 and 129). With reference to membership of a terrorist organization, the Criminal Code makes it an offence to promote, create, organize or direct a terrorist organization or to actively participate in or be a member of one (Article 571). Penalties for terrorist offences take the form of prison sentences and, in some cases, a prohibition on taking up public office, public sector employment or certain professions (Articles 572 and 579). In 1997 the Supreme Court sentenced members of the executive board (Mesa Nacional) of HB to seven years imprisonment for collaboration with an armed group. HB had tried to broadcast a video made by ETA during the 1996 elections, although the sentence was subsequently annulled by the Constitutional Court. In this case, the Supreme Court had nevertheless rejected the charge that the leaders of HB were members of a terrorist organization (Esparza, 2004, 147; Fernández Hernández, 2008, 188; Gil, 2015, 354). Shortly afterward, the Audiencia Nacional, a judicial body specializing, among other things, in terrorism cases, began to articulate and apply a new conception of ETA as a ‘complex structure’ integrating both terrorist commandos and supporting organizations, political parties and networks (Avilés, 2010, 48; Tajadura and Vírgala, 2008, 53). In 2002, the Audencia Nacional suspended the activities of Batasuna for integration within a terrorist complex led by ETA, initially for three years but later extended two further years. In 2006, the Audencia Nacional also declared illicit and dissolved the Partido Comunista de España (Reconstituido) (Spanish Communist Party (Reconstituted)), for integration in the terrorist organization Grupos de Resistencia Antifascista Primero de Octubre (First of October Anti-­ Fascist Resistance Groups). The Audiencia National also suspended the HB successor parties, Eusko Abertzale Ekintza/Acción Nacionalista Vasca (EAE/ANV, Basque National Action) and Euskal Herrialdeetako Alderdi Komunista/Partido Comunista de las Tierras Vascas (EHAK/PCTV) Communist Party of the Basque Homelands) on the grounds that they had been taken over by Batasuna after it was banned under the 2002 Party Law (Tajadura and Vírgala, 2008, 56–61). The suspension of the parties was part of a broader judicial campaign launched by the Audencia Nacional judge Baltasar Garzón in 2002 (sumario 35/2002). From this time, the Courts declared various organizations of the Basque nationalist left (known as the izquierda abertzale) as ‘terrorist organizations’ and thereafter suspended their activities and/or imprisoned their leaders for being part of ETA (Bourne, 2018; Fernández Hernández, 2008; Gil, 2015). In 2012 the Audencia Nacional, and later the Supreme Court, sentenced five leaders of the Basque nationalist left, including former Batasuna leader Arnaldo

46   Ban regimes and banned parties Otegi, to up to six and a half years in prison for membership of a terrorist organization. The Courts held that the condemned were trying to reconstitute the banned Batasuna at ETA’s orders. However, it was not until 2015 that the Audencia Nacional, and later the Supreme Court, formally ruled that HB, EH and Batasuna were terrorist organizations, sentencing 20 people to up to one year and ten months in prison on these grounds (Gil, 2015, 355). As Gil points out, even though the party had been declared illegal through the ‘via civil’ under the 2002 Party Law, in 2003 it had not been determined whether the members of the parties could be condemned for membership of a terrorist organization (Gil, 2015, 355). By this time, however, the Constitutional Court had de facto if not de jure consented to the legalization of Batasuna, by permitting the registration of Sortu as a political party. Many of these cases have been referred to the ECtHR, notably those against HB and Batasuna, the disqualification of electoral lists of Autodeterminazio Bilgunea and others, Herritarren Zerrenda and EAE/ANV. ECtHR rulings have supported rulings of the Spanish courts (Rodríguez-Vergara, 2010; Català i Bas, 2013). As the above discussion has shown, both criminal law and party law paths have been employed as vehicles for illegalization of political parties in Spain. Nevertheless, proscription via party law has been the principal legal vehicle for preventing the participation of parties in elections and will therefore be the focus of analysis in this book.

Party bans in the United Kingdom Sinn Féin and Republican Clubs were banned by authorities in Northern Ireland (NI), in accordance with the powers of the 1920 Government of Ireland Act, but were legalized by the UK government and parliament, which had resumed direct control over NI in 1972 (see Chapter 4 for details). The NI Civil Authorities (Special Powers Act) (henceforth the SPA) of 1922 gave the NI Minister of Home Affairs wide-­ranging powers to ‘take all such steps and issue all such orders as may be necessary for preserving the peace and maintaining order’, including those ‘not specifically provided for in the regulations’. The 1922–43 SPAs permitted proscription of political parties and organizations in pursuit of ‘preserving the peace and maintaining order’. By 1949, the relevant passages of the SPA on unlawful associations was Regulation 24A, which made it an offence for any person to become or remain a member of an unlawful association, or to act with a view to promoting the objects of an unlawful association or seditious conspiracy (Donoghue, 2001, 100–104). Furthermore, the Regulation provided that a person in possession of documents relating to the affairs of an unlawful association, emanating from one of its officers or indicating the person was an officer or member of the association, would be considered guilty of an offence, unless they could prove otherwise. Those convicted for membership of an unlawful association faced imprisonment with or without hard labour and/or fines (SPA, Section 4). In terms of electoral activity, banned parties could not

Ban regimes and banned parties   47 nominate candidates in NI elections but would be able to do so for Westminster elections regulated separately by UK legislation. In addition to various organizations, five parties were banned under the SPA between 1922 and 1972. Upon its founding, the NI state faced acute problems of sectarianism, political violence, rioting and disorder, including a Republican uprising that was ‘a brief but bloody challenge to Unionist supremacy’ (Fitzpatrick, 1998, 117; Patterson, 2006, 3). It was in this context that authorities banned five republican organizations: the Irish Republican Brotherhood, the IRA, the Irish Volunteers, the republican women’s organization Cumann Na m’Ban and the youth organization Fianna Na h’Eireann. In 1931, authorities banned Saor Éire, a socialist party launched by the IRA leadership in place of Sinn Féin (Hanley, 2002, 179–180). In 1936, republican efforts to form a new political party in NI under the name of Cumann Poblacta na h’Eireann (Donohue, 1998, 1112) were quickly thwarted when it too was banned. Sinn Féin was not formally added to the list of unlawful associations until 1956, after the IRA decided to revive the near moribund Sinn Féin. Fianna Uladh was also banned in 1956. This organization was the political wing of the paramilitary group, Saor Uladh, set up after a split within the IRA (English, 2003, 72; Coogan, 2000, 283). Republican Clubs, which were banned in 1967, were initially a successor party to Sinn Féin but eventually became the political wing of the Official IRA. Other non-­party organizations subsequently banned under the SPA were the IRA offshoot just mentioned, Saor Uladh (1955) (English, 2003, 72), the fascist National Guard (1933) and the Ulster Volunteer Force (1966). In Ireland, Éamon de Valera’s Fianna Fail government banned the IRA in 1936 (Patterson, 2006, 20) but, while other republican groups faced restrictions, Sinn Féin was not banned (Hanley, 2002, 94). Sinn Féin has been the subject of party ban proceedings in the United Kingdom since the early twentieth century. In July 1918, the government declared Sinn Féin Organization and Sinn Féin Clubs, along with a number of other republican organizations, ‘dangerous’ associations in accordance with the Criminal Law and Procedure (Ireland) Act 1887 (The Times, 1918). They were so characterized because authorities were ‘satisfied’ that ‘said associations in part of Ireland encourage and aid persons to commit crime and promote and incite to acts of violence and intimidation and interfere with the administration of law and disturb the maintenance of law and order’. Moreover, they were deemed a ‘grave menace designed to terrorize peaceful and law abiding subjects’. In November 1919, following Sinn’s Féin’s December 1918 Westminster election success (winning 73 out of 105 Irish seats), its decision to form the seccessionist Dáil Éireann and the start of the Irish War of Independence (1919–21), Sinn Féin was ‘prohibited and suppressed’ in accordance with the same legislation (The Times, 1919). Following the reintroduction of direct rule, the Westminster parliament passed the NI (Temporary Provisions) Act of 1972, which transferred to the UK Secretary of State for NI the powers of the NI Minister for Home Affairs to proscribe parties under the procedures of the SPA. It was under this legislation that

Table 2.2  Proscription and legalization of parties in Northern Ireland and the United Kingdom Party

Ban year and legislation

Grounds for ban

Legalisation year and legislation

Saor Éire

1931 Special Powers Act 1922 1973 NI (Emergency Provisions) Act 1973

Preserve peace and maintain order –

Cumann Poblacta na h’Eireann 1936 Special Powers Act 1922–1933

Preserve peace and maintain order –

Fianna Uladh

1956 Special Powers Act 1922–1943

Preserve peace and maintain order –

Sinn Féin

1956 Special Powers Act 1922–1943 1973 Northern Ireland (Emergency Provisions) Act 1973

Preserve peace and maintain order 1974 Northern Ireland (Emergency Provisions) Act 1973

Republican Clubs

1967 Special Powers Act 1922–1943

Preserve peace and maintain order 1973 Northern Ireland (Temporary Provisions) Act 1972

Ban regimes and banned parties   49 the Secretary of State for NI legalized Republican Clubs in April 1973. Sinn Féin was initially listed as a banned organization in the 1973 NI (Emergency Provisions Act) (EPA), but legalized by the Secretary of State for Northern Ireland in May 1974. Terrorist organizations could also be proscribed in Great Britain under the Prevention of Terrorism (Temporary Provisions) Acts (PTA), the first of which was enacted in 1974. The EPA and the PTA were periodically renewed and amended until they were repealed by the 2000 Terrorism Act. In both the EPAs and PTAs, an organization’s orientation to violence was the core criteria employed in proscription cases. Section 19(4) of the 1973 and equivalent sections in subsequent EPAs permitted the Secretary of State to proscribe ‘any organization that appears to him to be concerned in terrorism or in promoting or encouraging it’. Section 1(3) of the 1974 and subsequent PTAs permitted the Secretary of State to proscribe ‘any organization that appears to him to be concerned in terrorism occurring in the United Kingdom and connected with Northern Irish affairs, or in promoting or encouraging it’. According to Section 19(1) of the 1973 and equivalent sections in subsequent EPAs, any person who belonged to a proscribed organization; solicited or invited financial or other support for it; knowingly gave or received money or other resources to or from a proscribed organization; invited someone to become a member of, or carry out orders for, a proscribed organization; or (in later versions) arranged or attended meetings to support, or addressed by members of, a proscribed organization, faced prison sentences, and/or a fine. The possession of documents linking someone to a proscribed organization would be taken as evidence that they belonged to that organization. The PTAs mirrored these provisions. Walker has argued that while not expressly excluding judicial review of proscription cases, ‘it is likely to be extremely limited’ (1986, 34–36, 49–50). Schedule 2 of the first, 1973, EPA included many of the organizations already proscribed in NI, with the notable exception of Republican Clubs. In November 1973, the loyalist paramilitary groups, the Ulster Freedom Fighters and the Red Hand Commando were added to the EPA Schedule. In May 1974 Sinn Féin and the Ulster Volunteer Force were removed from the Schedule, and while Sinn Féin never made it back, the Ulster Volunteer Force was written into the Schedule again in 1975. In 1979, a republican paramilitary organization split from the Official IRA, the Irish National Liberation Army (INLA) was added. In 1990, the republican paramilitary organization, the Irish People’s Liberation Organization, an off-­shoot of the INLA, was banned. The loyalist Ulster Defence Association, formed in 1971, remained legal until 1992. Another two organizations were added in 1997, the paramilitary Loyalist Volunteer Force and Continuity Army Council, the latter a split from the Provisional IRA. After legalization of Republican Clubs and Sinn Féin no other parties were banned, only paramilitary organizations directly engaged in violence. In the UK, PTAs banned the IRA alone until 1979, when the INLA was added to its Schedule of banned organizations. Under the Terrorism Act 2000, the Home Secretary continues to have the power to proscribe organizations believed to be concerned in terrorism.

50   Ban regimes and banned parties

Party bans in Germany The proscription of parties in Germany is regulated by Article 21 of the 1949 German Basic Law, as interpreted by the Federal Constitutional Court. Article 21.1 provides that political parties ‘can be freely established’ and gives them a central role in the democratic life of the state through their right to ‘participate in the formation of the political will of the people’. Subsequently, the Federal Constitutional Court elaborated the concept of Parteienstaat (party state) which recognizes parties as ‘agents’ of political will formation and thus ‘integral units of the constitutional state’ (Kommers, 1997, 200; von Schmertzing, 1957, 12). In accordance with the constitutional concept of ‘militant democracy’, Article 21 of the Basic Law also limits party freedoms. In the German constitutional tradition, militant democracy is conceived as the requirement that the state defend the ‘free democratic basic order’ and permits the limitation of certain freedoms to that end (Kommers, 1997, 217). As such, parties’ internal organization must conform to democratic principles (Article 21.1) and they must respect the ‘free democratic basic order’ (Article 21.2). This last provision permits proscription of parties, ‘that by reason of their aims or the behaviour of their adherents, seek to undermine or abolish the free democratic basic order or to endanger the existence of the Federal Republic of Germany’. Such parties would be deemed ‘unconstitutional’ (Article 21.2). The Federal Constitutional Court would determine the constitutionality of a party (Article 21.2). While parties, as constitutional entities, have a duty to affirm the basic values of the constitution, the Federal Constitutional Court has defined a countervailing ‘party privilege’, whereby a party and its officials are to be free from discrimination and interference until declared unconstitutional (Franz, 1982, 63). As Kommers notes, provisions permitting proscription of party bans were: designed to repair a central failing of the Weimar Republic, namely, its tolerance of extremist parties bent on destroying democracy. Recalling the conditions that led to the Hitler state, the founders resolved that the Federal Republic could never be neutral in the face of its mortal enemies. (1997, 218) Similarly, as the Federal Constitutional Court itself declared in its definition of ‘militant democracy’: ‘Article 21(2) … expresses the conviction of the [founding fathers], based on their concrete historical experience, that the state could no longer afford to maintain an attitude of neutrality towards political parties’ (cited in Kommers, 1997, 223; von Schmertzing, 1957, 13). Applications to ban two parties in the 1950s provided opportunities to clarify these provisions. In its 1952 ruling on the constitutionality of Sozialistische Reichspartei Deutschlands (SRP, Socialist Reich Party of Germany), the Federal Constitutional Court defined the term, ‘free democratic basic order’, to be: At the very least, respect for the rights of man as set forth in the Basic Law, above all respect for the rights of one individual to life and free development,

Ban regimes and banned parties   51 the sovereignty of the people, separation of powers, the accountability of government, administration according to the law, the independence of the judiciary, the multiparty principle, with equal opportunity for all political parties, including the right to constitutionally acceptable development, and opposition. (Translated in Franz, 1982, 57; see also Niesen, 2002, 255) In this case, the Court also emphasized the importance of a party’s democratic internal organization, arguing, If a party’s internal organization does not correspond to democratic principles, [one] may generally conclude that the party seeks to impose upon the state the structural principles that it has implemented within its own organization. (Cited in Kommers, 1997, 220) In its 1956 rulings on the Kommunistische Partei Deutschlands (KPD, Communist Party of Germany), the Federal Constitutional Court further elaborated the standards by which it would establish a party’s anti-­democratic character. A party could not be banned merely for articulating views that contradicted the foundational values of the free democratic basic order (McWhinney, 1957, 303; Franz, 1982, 62; von Schmertzing, 1957, 13). However, it was not necessary to prove that the party posed an imminent or actual danger to the democratic system (von Schmertzing, 1957, 14). Nor would it be necessary to present evidence of a party’s ‘concrete undertaking’ to abolish the constitutional order (Kommers, 1997, 223). Rather, the Constitutional Court required evidence showing an ‘intention’, or ‘fixed purpose to combat the free basic order constantly and resolutely’ manifested in ‘political action according to a fixed plan’ (Kommers, 1997, 223; von Schmertzing, 1957, 13). There would need to be evidence of a party’s active, combative and aggressive attitude to the existing political order (McWhinney, 1957, 303). Or in other words: ‘What is necessary is this: The political struggle of a party must be fixed by a purpose and it must be directed; furthermore, it must show a constant proclivity towards the struggle against the established constitutional system’ (von Schmertzing, 1957, 14). Action is therefore essentially ‘preventive’, ‘to forestall the rise of political parties with anti-­democratic objectives’ (ibid., 14). Party bans are also regulated by the 1967 Law on Political Parties (Gesetz über die politischen Parteien), as amended at 22 December 2015. In the Party Law, parties are defined as associations of citizens which exert influence permanently or for longer periods of time on the formation of the political will at federal or Land level and participate in the representation of the people in the German Bundestag or regional parliaments. (Section 2.1) Associations lose their status as a party if they fail to participate in elections at either federal or Land levels with nominations of their own at least once every

Table 2.3  Proscription of parties and failed party bans in Germany Party

Year of ruling

Outcome

Grounds for ruling

Sozialistische Reichspartei Deutschlands 1952

Banned

Former Nazis in leadership, undemocratic internal organization, and anti-democratic and racist political programme.

Kommunistische Partei Deutschlands

1956

Banned

Goal of establishing a dictatorship of the proletariat incompatible with the free democratic basic order

Freiheitliche Deutsche ArbeiterPartei

1994

Case dismissed

Not a political party. Later banned as an ‘association’

Nationale Liste

1994

Case dismissed

Not a political party. Later banned as an ‘association’

Nationaldemokratische Partei Deutschlands

2003

Case dismissed

Some evidence against the NPD was provided by government-paid informers, which prevented a fair trial

Nationaldemokratische Partei Deutschlands

2017

Case dismissed

The NPD pursued the elimination of the free democratic basic order but it was not likely to achieve those aims

Ban regimes and banned parties   53 six years and if they fail to submit their accounts in a six-­year period (Section 2.2). A majority of a party’s members and executive board must be Germans and its registered seat must be in Germany (Section 2.3). A banned party will lose eligibility for state funding from the date of its dissolution (Section 18.7). The Party Law also prohibits parties ‘which continue to pursue the unconstitutional aims of the party banned under Article 21.2. of the Basic Law’, whether these be new organizations or existing parties (Section 33.1–2). The Law on the Federal Constitutional Court (Gesetz über das Bundesverfassungsgericht), as amended at 31 August 2015 further provides that a declaration of unconstitutionality may be limited to a legally or organizationally independent section of a party (Section 46.2). If a party or section of it is declared unconstitutional, it will be dissolved and the establishment of a substitute organization prohibited (Section 46.3). The Federal Constitutional Court may permit confiscation of the banned party’s or section’s property, which will be given to the federation or Land to be used for public benefit (Section 46.3). All the parliamentary representatives of a banned party lose their mandates (Hanschmann, 2001, 104; Kommers, 1997, 223). Sections 84 and 85 of the German Criminal Code prohibit attempts to maintain a banned party or establish a surrogate, with punishment of up to five years imprisonment or a fine for active members or supporters. As Table 2.3 shows, the Federal Constitutional Court has declared two parties unconstitutional, the SRP in 1952 and the KPD in 1956. The SRP was declared unconstitutional and dissolved due to its Nazi leadership, undemocratic internal organization and anti-­democratic and racist political programme. In 1956 the Federal Constitutional Court banned the KPD on the grounds that the party’s goal of establishing a dictatorship of the proletariat was incompatible with the free democratic basic order (Gordon, 1987, 375; McWhinney, 1957, 303). In September 1993, the federal government and Bundesrat called on the Federal Constitutional Court to ban the far-­right Freiheitliche Deutsche ArbeiterPartei (FAP, Free German Workers Party), but the Court rejected the case on the grounds that the FAP was not a political party (Wise, 1998, 314; Flemming, 2003). The day after the Federal Constitutional Court announced its decision, the federal Interior minister banned the organization under the authority of article 9 of the Basic Law (Wise, 1998, 315–316). The Hamburg senate filed a similar case in the Federal Constitutional Court against the neo-­Nazi Nationale List (National List), a case which was rejected on the same grounds as the FAP (Kommers, 1997, 236). In 2001, the federal government and both houses of the German parliament submitted applications for proscription of the NPD, but the Federal Constitutional Court dismissed the case on procedural grounds in March 2003 (Rensmann, 2003, 1121). In 2012, the Bundesrat initiated a new case against the NPD, which the Federal Constitutional Court dismissed in 2017 on the grounds that the NPD was unlikely to achieve its aims, even if it pursued goals which sought to eliminate the free democratic basic order.

3 Herri Batasuna and its successors From ‘terrorist threat’ to carrier of a ‘Copernican shift’ from violence

In this chapter, I present findings relating to the first three hypotheses in the cases of bans on Herri Batasuna (HB), Euskal Herritarrok (EH) and Batasuna in 2003 (henceforth HB and successors), and the legalization of Bildu in 2011 and Sortu in 2012. For purposes of comparison, the 2003 HB and successors ban case study presented in this chapter reproduces the findings of research conducted in the first stage of the research design (see Bourne 2015 and Chapter 2 for more details). I begin with a description of the origins of HB and the terrorist group ETA, the programmatic goals of the izquierda abertzale (Basque nationalist left) movement to which they belong, the evolution of successor parties and their salience in the Basque Country. I then turn to examine the hypotheses: H1. Democracies ban anti-­system parties if these parties have been ‘securitized’ as an existential threat. H2. Democracies ban anti-­system parties if veto players prefer proscription. H3. Democracies ban anti-­system parties if they do not unambiguously reject violence. Data for the case studies consist of parliamentary debates,1 court rulings and newspaper articles principally from the daily El Mundo. Analysis shows how the hypotheses were generated from the HB and successors ban case study and that the conditions under which Bildu and Sortu were legalized varied in a manner consistent with theory. That is, in the legalization cases, it was expected that each of the conditions under which the parties – or their predecessors – were banned would no longer hold.

Herri Batasuna and successors HB was a coalition of parties formed in 1978, in the midst of the Spanish transition to democracy and following the death of the dictator General Francisco

Herri Batasuna and its successors   55 Franco. It has been widely acknowledged that the party was the ‘political wing’ of the terrorist group ETA. ETA emerged in 1959 during the dictatorship, formed by a small group of Basque youths dissatisfied with the civil war generation of Basque nationalists and the dominant Euzko Alderdi Jeltzale/Partido Nacionalista Vasco (EAJ/PNV, Basque Nationalist Party). The group sought to establish an independent socialist state, encompassing Basque territories in Spain and France (Jáuregui, 2006). In the early 1960s, ETA pursued revolutionary war and military defeat of state security services in Basque territories (Ibarra, 1987, 58–61). By 1965, it adopted a ‘spiral of violence’ (or action-­repression) strategy, where violence was used to provoke increased state repression and eventually popular insurrection against the Spanish state (ibid.). Following the death of General Franco, one faction, ETA político militar (ETA political military) renounced violence. However, the largest faction – ETA militar (ETA military) – continued to conceive of the Basque Country as an occupied territory and pursued armed struggle to achieve national liberation. Soon known simply as ETA it rejected new democratic institutions established in the 1978 Constitution (for not recognizing a Basque right to self-­ determination) and the 1979 Autonomy Statute devolving political power to new Basque institutions (for being an obstacle to independence). ETA mounted a massive campaign of terror against the democratization process. Since its first confirmed killing in 1968, ETA has killed more than 800 people and has injured, kidnapped or extorted thousands of others (see Figures 3.1 and 3.2). In a context of democratic consolidation and popular support for new institutions of Basque self-­government, ETA adopted a war of attrition, where it employed violence in an effort to force state authorities to negotiate and accept its demands (Ibarra, 1987, 99–100; Muro, 2008, 166; Domínguez, 2006, 287). These were spelt out in the Alternativa KAS (KAS Alternative) and included: total amnesty; expulsion of state security services from Basque territory; improved living and working conditions for Basques; recognition of Basque national sovereignty, right to self-­determination and independent statehood; and adoption of the Basque language (Euskera) as official and primary language (Ibarra, 1987, 99–100). The strategy assumed that neither the Spanish government or ETA could defeat each other militarily, that Spanish public opinion would support Basque independence if enough pressure was applied, and that an urban mode of warfare and campaign of selective killing would demoralize Spanish authorities and push them to discuss complete withdrawal (Muro, 2008, 166). In 1995, ETA updated the Alternativa KAS with the Alternativa Democrática (Democratic Alternative), which dropped reference to socialism, but continued to insist on government recognition of Basque self-­determination and territorial unity of the seven Basque provinces in France and Spain as a prerequisite for an end to violence. Agreement between ETA and the state on these issues, ETA proposed, could then lead to a ‘democratic process’ in which Basque society could ‘decide its own future in liberty’ (Mees, 2003, 76; Muro, 2008, 154). In 1978, a coalition of parties linked to ETA formed the party, HB. In 1998, EH emerged as an electoral coalition, dominated by HB. It was refounded as

56   Herri Batasuna and its successors Batasuna in 2001. While technically, all three of these parties were banned in 2003, ETA, HB and its successors were part of a movement known as the Movimiento de Liberación Nacional Vasco (MLNV, Basque National Liberation Movement) or, in the terminology more commonly used today, the izquierda abertzale (Basque nationalist left) (Llera, 1992, 183–186). This social movement is present in many spheres of social life – including sports, Basque language schools and bars (Mata, 1993, 104). Various scholars have compared the movement’s brand of radical nationalism with a ‘political religion’ observed in an elaborate set of symbols, liturgy and rituals, including public commemoration of the gudari (Basque soldier) (Casquete, 2009; see also Sáez, 2001). Despite the known relationship between ETA and HB, the latter was legal and permitted to participate in electoral contests for around two decades after the death of Franco (see Table 6.2). When legal, HB won seats in all Spanish general elections it contested, winning as many as five parliamentary seats in 1982. It was one of the largest parties in the Basque Country, winning up to 14 seats in the Basque parliament (as EH in 1998). Following illegalization of HB and its successors 2003, the Basque nationalist left found representation in the Basque parliament through Euskal Herrialdeetako Alderdi Komunista/Partido Comunista de las Tierras Vascas (EHAK/PCTV, Communist Party of the Basque Homelands). This party won nine seats in the 2005 elections but was subsequently banned in 2008 as a successor of Batasuna (see Table 2.1). Despite the tenacity of the government, the ministerio fiscal (henceforth Public Prosecutor) and the Courts to prevent the emergence of successor parties, efforts to keep parties of the Basque nationalist left out of public institutions were initially only partially successful. For example, while 133 of the party lists of Eusko Abertzale Ekintza/Accion Nacionalista Vasca (Basque National Action) were disqualified for 2007 local elections, the party nevertheless obtained a total of 439 local councillors, held 42 mayoral offices, an absolute majority in 24 municipalities, and 5 deputies in Juntas Generales (Provincial parliaments) in Guipúzcoa and Vizcaya (El País, 2008). The party was later banned in 2008 (see Table 2.1), although elected representatives continued to hold their seats until May 2011 elections. It was not until March 2009 that Basque autonomous community elections were held without the presence of a party linked to ETA. By this time, various scholars have argued, exclusion of HB and its successors from participating in elections had left the Basque nationalist left politically isolated and largely inoperative (Murua, 2017, 109; Zabalo and Saratxo, 2015, 372–373; Whitfield, 2014, 6). After a series of short-­lived ceasefires, failed peace processes and negotiations in the 1980s, 1990s and mid-­2000s, ETA announced a halt to armed actions in September 2010 and a definitive end of armed activities in October 2011. There is considerable debate about why ETA unilaterally and unconditionally abandoned armed struggle without a formal negotiated settlement of any kind. Many argue that increasingly effective counter-­terrorism strategies of Spanish and French security services and judicial campaigns banning both political parties and supporting organizations of the Basque nationalist left

Herri Batasuna and its successors   57 fundamentally weakened ETA in both military and operational terms (Alonso, 2016, 20; Domínguez, 2012, 274–292; Murua, 2017, 107–109; Whitfield, 2014, 6–7; Zabalo and Saratxo, 2015, 372). Other factors include changing attitudes of Basque citizens, increasing effectiveness of the police and security operations, a hardening of attitudes against terrorism after 9/11, facilitation by international actors, and the ideological evolution, strategic reorientation and changing balance of power within ETA and the Basque nationalist left itself (e.g. Muro, 2008, 169, 189; Llera, 2011, 40, 44; Domínguez, 2012, 274, 281; Murua, 2017, Whitfield, 2014; Zabalo and Saratxo, 2015). The parties Bildu and Sortu emerged as the new face of the Basque nationalist left in the post-­ETA ceasefire period. Unlike its predecessors, Bildu took the form of an electoral coalition involving a large number of independent candidates and two already legal political parties, namely Eusko Alkartasuna (EA, Basque Solidarity) and Alternatiba (Alternative), both with a long history of explicitly condemning ETA (Iglesias, 2011, 556; Narváez, 2011, 2). After the Constitutional Court overruled the Supreme Court’s prohibition of Bildu (see below), the coalition participated in 2011 municipal, Navarrese parliamentary and provincial elections in Guipúzcoa, Vizcaya and Álava, winning more councillors in the Basque Country than any other party and a high number of those in Navarre (Iglesias, 2011, 557). The formula pioneered in Bildu was re-­used for the 2011 Spanish general election, with the addition of the party Aralar to the coalition. In that election, the new coalition Amaiur (named after a Basque village) won seven seats. After it was legalized in 2012, Sortu joined these parties under the banner of Euskal Herria Bildu (EH Bildu, Basque Country Unite), becoming the second largest electoral force in the Basque Autonomous Community with 21 and 18 seats respectively following the 2012 and 2016 elections, but falling to 2 seats in the 2015 and 2016 Spanish general elections.

Party ban veto players in Spain Since the transition to democracy, there have been two party laws, the 1978 Party Law (Law 54/1978) and the 2002 Party Law (Law 6/2002), as amended by Ley Orgánica 3/2015 (henceforth 2002 Party Law). The banning of HB and its successors and legalization of Bildu and Sortu invoked the 2002 Party Law, and indeed, it is widely acknowledged that this law was enacted to facilitate the ban on Batasuna. The government or Public Prosecutor are entitled to initiate proceedings to ban an already registered party (Article 11.1). In addition, the government is obliged to initiate ban proceedings, after deliberation of the Council of Ministers (equivalent to the cabinet), if one of the two houses of the Spanish parliament, the Congress of Deputies or Senate, call for it (Article 11.1). Criminal courts have jurisdiction to ban parties where there are suspicions that a party has committed offences under the Criminal Code (Article 10, and see Chapter 2). On matters relating to proscription for non-­democratic internal structure or acts violating democratic principles and damaging the ‘regime of liberties’ or the democratic system, a special chamber of the Supreme Court would have

58   Herri Batasuna and its successors jurisdiction (Article 10). The Constitutional Court hears appeals relating to protection of rights (recurso de amparo). The Interior Ministry is responsible for administering a Register of Political Parties. If the Interior Ministry considers a party seeking registration is an illicit organization under the Criminal Code, it could alert the Public Prosecutor (Article 5). If the Public Prosecutor agreed there was a reasonable case against the party, the competent judicial authority would make a ruling on whether registration could take place. The specific Court making a party ban ruling would subsequently be responsible for preventing the creation of new parties or use of existing ones to succeed or continue the activities of the banned party (Article 12). Those involved in the case, or where such a case dealt with inscription in the Register of Political parties, the Ministry of the Interior and Public Prosecutor’s Office, could initiate new proscription proceedings on these grounds (Article 12.3). The Ley Orgánica 5/1985, de 19 de junio, del Régimen Electoral General (LOREG, Organic Law on the General Election Regime, as amended by the 2002 Party Law and Ley Orgánica 3/2011, de 28 de enero), permits the Courts to annul candidacies in specific electoral contests which aim to succeed or continue the activities of banned parties. Article 49.5 of LOREG replicates procedures used for illegalization of parties provided in the 2002 Party Law; that is, annulment of candidates are formally initiated by the government or Public Prosecutor’s Office, while the Congress of Deputies and Senate are authorized to solicit a government-­sponsored application for annulment. A special chamber of the Supreme Court then hears the case, with appeals relating to protection of rights decided by the Constitutional Court. Spain has a parliamentary system of government, although the identification of partisan veto players is complex given the frequency of minority government. Following the 13 general elections between 1977 and 2016, single-­party majority governments were formed on only four occasions. With the exception of 2015, when no government could be formed and new elections were called, single-­ party minority governments have been formed after all other elections. According to veto player theory, minority governments can be classified as single veto players, similar to those governed by single party majority governments (Tsebelis, 2002, 97). This is because those controlling the government have agenda setting advantages and if located centrally in ideological terms parties running minority governments can select among many different partners to ensure their programme is approved by parliament (ibid., 98). This is the case in Spain, especially in relation to party ban proceedings, whether or not it obtains a mandate to do so from one or other house of the Spanish parliament. Spain’s decentralized political system is another relevant parameter for formation of minority governments, given the electoral weight of regionalist parties in the Spanish parliament and, especially in the Basque Country and Catalonia, their ability to control powerful regional governments (Heller, 2002; Field, 2009 and 2014). The minority government of the Partido Socialista Obrero Español (PSOE, Spanish Socialist Workers Party) between 1993 and 1996 and that of the

Herri Batasuna and its successors   59 Partido Popular (PP, Popular Party) between 1996 and 2000 relied on support from the Catalan nationalist Convergència i Unió (CiU, Convergence and Union), the Basque nationalist EAJ/PNV and Canary Island regionalists from Coalición Canaria (CC, Canarian Coalition). On the other hand, the PSOE minority governments under José Luis Rodríguez Zapatero between 2000 and 2004 and 2004 and 2008 relied on much more fluid, shifting and ad hoc alliances (Field, 2009 and 2014). The 1950 ECHR, ratified by Spain in 1979, establishes, among other things, the ‘right to freedom of expression’ (Article 10.1) and ‘the right to freedom of peaceful assembly and association with others’ (Article 11.1). The ECHR permits restriction of these rights by measures that are ‘prescribed by law’ and ‘necessary in a democratic society’, or for purposes of national security or public safety, prevention of disorder or crime, protection of the rights of others, among other things (Articles 10.2 and 11.2). By the time HB and Batasuna referred ban cases to the ECtHR, the Court had already developed a robust doctrine on the prohibition of political parties, particularly in rulings on party bans in Turkey (Koçak and Örücü, 2003; Harvey, 2004; Brems, 2006a). The ECtHR’s jurisprudence on party bans acknowledges the dilemmas inherent in party bans but sets standards explicitly referring to a party’s orientation to violence and their threat to democratic institutions. The doctrine emphasizes the importance of pluralism, including airing of the most controversial opinions, and that political parties could legitimately propose not just changes in legislation, but also to state and constitutional structures. In the exercise of these prerogatives, however, parties were required to use democratic and legal means to achieve these goals, without recourse to violence. To be compatible with the ECHR, parties had to pursue goals in accordance with fundamental democratic principles. The state could reasonably act to prevent realization of political projects incompatible with the norms of the Convention, although measures as serious as banning parties had to be proportionate to the threat posed. Ban of Herri Batasuna, Euskal Herritarrok and Batasuna The decision to launch ban proceedings against HB and its successors in August 2003 took place during a period of majority PP government under the leadership of José Maria Aznar. The decision to ban the party had been proceeded by the 12 December 2000 Acuerdo por las libertades y contra el terrorismo (Agreement for Liberties and Against Terrorism), a pact between the PP and PSOE promising permanent collaboration in combatting terrorism, regardless of which party led government or the opposition (Bew et al., 2009: 227–229). Formal proceedings to ban HB and successors were launched by both the Public Prosecutor’s Office and the government, following an express mandate from the Congress of Deputies (lower house of the Spanish parliament). The parliamentary resolution of 26 August 2002 urging the government to ban the parties was supported by a large majority, including the PP and PSOE (the two largest parties), and small regionalist parties from Canary Islands and Andalusia (CDDS 181, 2002,

60   Herri Batasuna and its successors p.  9091–9109). Votes against (ten in total) included those from Basque nationalist parties, EAJ/PNV and EA, and the leftist Catalan nationalist party Esquerra Republicana de Catalunya (ERC, Republican Left of Catalonia). The leftist Izquierda Unida (IU, United Left) and Catalan nationalist party CiU, among others, abstained. In March 2003, the Supreme Court declared HB and its two successors illegal (Vírgala, 2003, 614). HB and Batasuna launched appeals against the ruling in the Constitutional Court, arguing that the president of the Supreme Court had not been impartial; that aspects of the trial procedure had been unfair; that there was a violation of the presumption of innocence; and that the ruling had violated their freedom of expression, thought and association, among other things. These arguments were dismissed unanimously by the Court. HB and Batasuna then challenged the decisions of the Spanish courts before the ECtHR, which reached judgement in June 2009. In their submission to the ECtHR, Batasuna and HB argued that party bans violated ECHR provisions on freedom of expression and association; that the 2002 Party Law was an ad hoc law which was not accessible or foreseeable; sought to limit political debate in the Basque Country and was applied retroactively; that the ban did not pursue legitimate ends; was not necessary in a democratic society and was contrary to the principle of proportionality (Pérez Sola, 2009; Rodríguez-Vergara, 2010). The ECtHR rejected all these arguments finding in favour of Spain (ECtHR, 2009). Legalization of Sortu and Bildu The government (Ministry of the Interior) or the Public Prosecutor may initiate proceedings in the Supreme Court challenging the registration of a political party suspected of succeeding or continuing the activities of a banned party – which was the case for Sortu. A different procedure, however, applied for an application to anull the right of a party, coalition or electoral grouping to present candidacies in a specific election on grounds that it succeeded or continued the activities of a banned party – which was the case for Bildu. According to Article 49.5 of the electoral law, LOREG, such a case could be initiated by the government or Public Prosecutor’s Office and the Congress of Deputies and Senate could solicit the government to launch such a case. The decisions to initiate ban proceedings against Sortu and Bildu were made during the minority government of José Luis Rodríguez Zapatero, which had won 169 seats in the 2008 Spanish general elections, just seven short of a majority. The government was sustained by shifting alliances rather than a formal agreement but nevertheless stayed in office for close to a full four-­year term (Field, 2009 and 2014). The PSOE government was under pressure to initiate proceedings against Sortu and Bildu from its principal competitor at the national level, the PP (152 seats), which was also supporting a socialist minority government in the Basque Country (El Mundo, 2011b and 2011d). The small regional parties, Unión del Pueblo Navarro (Navarrese People’s Union) (2 seats) and Unión, Progreso y Democracia (UPyD, Union, Progress and Democracy)

Herri Batasuna and its successors   61 (1 seat), continued to call in parliament without success for proscription of Bildu, even after the Constitutional Court effectively legalized the party and, in the case of UPyD, after a new PP had formed a government in 2011(El Mundo, 2011r; CDDS 265 2011, p. 25; CDDS 10 2012, pp. 31–37). On the other hand, the PSOE government faced the threat of non-­collaboration from the Basque nationalist EAJ/PNV, in opposition in the Basque Autonomous Community, but with six seats in the Spanish parliament. Its president, Iñigo Urkullu, warned that if Bildu were not permitted to participate in municipal elections, the EAJ/PNV would take this into account in its relations with the Spanish government, and especially negotiation of the 2012 budget (El Mundo, 2011l). The EAJ/PNV, with only six of the seven votes the Socialist government would have needed for a majority, was not a veto-­player and the government had alternative partners. Furthermore, the EAJ/PNV had traditionally taken a pragmatic stance when making deals with either PP or PSOE minority governments, which was seen as an opportunity to extract concessions in the realm of Basque self-­government and to facilitate deal-­making, particularly with Socialists in the institutions where the EAJ/PNV governed (Heller, 2002; Field, 2014, 306–308). A parliamentary vote was not taken on the question of banning Bildu. In early 2011, the PSOE minority government nevertheless decided to initiate ban proceedings against Sortu, and soon afterwards against Bildu. On 27 March 2011, the Government and Public Prosecutor’s initiated proceedings in the Supreme Court to prevent the inscription of Sortu in the Register of Political Parties. Just three days later the Supreme Court prohibited registration of the new party. For the first time, a party ban ruling was passed by only a majority of judges rather than unanimity (nine in favour, seven against) (Garamendia, 2011, 328). The majority ruled that ETA had managed, instigated and supervised Batasuna’s attempt to create the new party and as such, Sortu was a successor to the illegal Batasuna (Garamendia, 2011, 328–329; Pérez Medina, 2011, 6). In June 2012 – after it had deliberated on the case against Bildu (see below) – the Constitutional Court overturned the decision of the Supreme Court and permitted registration of Sortu (Iglesias Bárez, 2013, 240). Against the majority of six, five judges dissented in three separate opinions. The case against Bildu – formally a coalition under the name Bildu-­Eusko Alkartasuna (EA)/Alternatiba Eraikitzen – was initiated immediately prior to municipal and local elections in the Basque Country and Navarre (and elsewhere in Spain) on 22 May 2011. The Supreme Court and Constitutional Court both reached their respective rulings within an extremely short time-­period. The case was concluded just nine days after the submission of the government’s (27 April) and Public Prosecutor’s (28 April) petition to annul the Bildu candidates previously approved by Electoral Commissions in Álava, Vizcaya, Guipúzcoa and Navarre and Miranda de Ebro (Burgos). In the Supreme Court, a majority ruled that Bildu was designed by ‘Batasuna/ETA’ as a tool to regain representation in Basque political institutions, and that the claim it rejected terrorist violence was not convincing (de Mendizábal, 2011, 4; Caballero, 2011, 3; Narváez, 2011, 1). On these grounds, the Supreme Court majority (with seven dissenting

62   Herri Batasuna and its successors judges) ruled Bildu was a successor to, or continuation of, the illegal Batasuna. Bildu appealed the ruling in the Constitutional Court on grounds that the Supreme Court majority decision violated the rights of participation in public affairs either directly or through freely elected representatives, to equal access to public office and to freedom of ideology, expression and association. Aligning with most of the Supreme Court dissenters, the majority of the Constitutional Court, itself divided, considered that the evidence used by the Supreme Court showed collaboration between EA and Alternatiba and the broader social movement of the Nationalist Left, rather than with the illegal Batasuna (Caballero, 2011, 4). Nor were the majority convinced there was evidence that ETA and Batasuna had authored the creation of Bildu (Caballero, 2011, 4). In sum, there are multiple veto players in Spanish party ban proceedings. The procedures are sequenced so that the agreement of different political actors is crucial at different stages of the procedure. For cases invoking the 2002 Party Law, it is necessary to obtain either (1) an agreement of the majorities in the Congress of Deputies or the Senate, or (2) an agreement within the government and governing party, or (3) a decision by the Public Prosecutor in the initiation or agenda-­setting phase of the procedure. In the decision phase, the Supreme Court can declare a party is illegal. Supreme Court rulings could be overturned on appeal in the Constitutional Court or upon further appeal to the ECtHR declared incompatible with the ECHR. In the case against HB and its successors in 2003, veto players in both phases of the procedure agreed the parties should be banned. Bildu and Sortu, however, did not become legal parties because all veto players agreed this should be so. Only the Constitutional Court rejected continuing to ban parties of the Basque nationalist left. Due to the sequencing of part-­ban veto powers the Constitutional Court’s decision was final.

Proscription of Herri Batasuna, Euskal Herritarrok and Batasuna ETA’s terror campaign and the parties’ ambiguous position on violence The immediate trigger for application of the 2002 Party Law was Batasuna’s refusal to condemn an August 2002 ETA car bombing of a Civil Guard barracks in Santa Pola (Alicante), in which around 40 were injured and two died, including a six-­year-old child (El Mundo, 2002g and 2002l). (Batasuna was at the time of the 2003 ban the current incarnation of HB and public debate tended not to distinguish between HB, EH and Batasuna, as those involved in formal ban proceedings did. As such, I will use this terminology when reproducing public debates.) As mentioned above, the deaths in Santa Pola were two of more than 800 deaths caused by ETA. Figures 3.1 and 3.2 show that most of those killed and injured by ETA occurred during the early years of the transition to democracy when the 1978 Spanish Constitution and institutions of self-­government for the Basque Country and Navarre were set in place.

Herri Batasuna and its successors   63 100

Victims

80 60 40 20 0 –20 1965

1970

1975

1980

1985

1990 Year

1995

2000

2005

2010

2015

Figure 3.1  Mortal victims of ETA, 1968–2010.

120 100 80 60 40 20 0

1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010

Attacks

Source: Ministry of the Interior, Spanish government (www.interior.gob.es).

Year

Figure 3.2  ETA attacks (other than assassinations), 1970–2010. Source: National Consortium for the Study of Terrorism and Responses to Terrorism (START) (2017). Global Terrorism Database [Data file], www.start.umd.edu/gtd. Note Types of attacks included here are armed assaults, bombing/explosion, facility/infrastructure attacks, hostage taking (kidnapping) and unclassified.

Furthermore, various scholars argue that intensification of political violence, the broadening of those targeted by ETA, and political polarization following the realignment of moderate Basque nationalists towards cooperation with the Basque nationalist left in the late 1990s, were among the reasons why tolerating the presence of HB or its successors in the public sphere was questioned (Esparza, 2004, 26; Sáez, 2002; Muro, 2008, 160; Domínguez, 2012, 302). The decision to ban the parties occurred at a time when ETA and its supporters unleashed a wave of violence on the streets of the Basque Country, often referred to as ‘kale borokka’ (street violence, or for the government, ‘urban terrorism’). Government statistics from 2003 record an average of 493 acts of ‘urban terrorism’ each year in the Basque Country between 1999 and 2002, including attacks on members of political parties, public and private property and public transport (Ministerio del Interior, 2003). At the same time, ETA had widened the targets of terrorist attacks beyond police and security service personnel to

64   Herri Batasuna and its successors journalists, civil servants, judges, academics and politicians from the main political parties (Muro, 2008, 155–157). It aimed to ‘socialize the pain’ of nationalist struggle, create social tensions and force Basque citizens to openly take sides (ibid.; Domínguez, 2006, 410–411, 431). The ban on Herri Batasuna and its successors followed in the aftermath of the failed Lizarra-­Garazi Agreement, a political agreement signed by the moderate Basque nationalist parties EAJ/PNV and EA, HB and a series of other organizations mostly close to the Basque nationalist left in September 1988 (Mees, 2003, 139–164; de la Granja, 2003, 315–335). ETA subsequently announced a ceasefire, which lasted 14 months and, in 1999, the PP government of José Maria Aznar had agreed to meet with ETA (Fernández Hernández, 2008, 114; Bew et al., 2009, 225–227). For some, the Lizarra-­Garazi process, modelled on that in Northern Ireland, appeared to offer ‘peace for self-­determination’, a formula by which ETA might be persuaded to abandon arms in exchange for collaboration with other Basque nationalists to achieve independence (Mees, 2003, 139–164). For others, the process was a ‘tregua trampa’ (truce-­trap), as the PP Interior Minister Jaime Mayor Oreja famously described it, or an opportunity engineered by ETA to regroup after debilitating successes of state security forces. ETA broke its ceasefire in November 1999, leading to an intensification of violence in the years immediately prior to the negotiation of the 2002 Party Law and proscription proceedings. International developments, such as the hardening of attitudes to political violence after the 2001 terrorist attacks in the United States provided fertile ground for a hardening of domestic counter-­terrorism policy in Spain, as elsewhere (Esparza, 2004; Muro, 2008; Sáez, 2002; Murua, 2017). In the midst of the campaign to ban it, Batasuna’s spokesman, Arnaldo Otegi, argued that the party ‘did not wear hoods or undertake armed activity’ and that its goal was to initiate ‘a democratic process leading to political change, peace and self-­determination’ (El Mundo, 2003c). Nevertheless, various academic studies – and a key line of argument employed against it in ban proceedings – showed that HB and its successors’ position on the appropriateness of violence as a political strategy was highly ambiguous, given its close relationship with ETA. It was known prior to ban proceedings that the origins of HB lay in ETA’s decision to establish, in 1975, the Koordinadora Abertzale Sozialista (KAS, Patriotic Socialist Coordinator) (Llera, 1992; Mata, 1993, 95–131). KAS came to include trade union, youth and women’s organizations and the coalition, HB, once it was formed in 1978. KAS established the line of action for participating groups, including HB, and the large network of ETA supporters encompassing religious, cultural, ecologist, student and other groups (Llera, 1992; Mata, 1993, 95–131). Scholars described ETA as the movement’s catalyst, theorist, source of cohesion and enforcer (Mata, 1993; Llera, 1992). Its power over the movement, according to Mata, was derived from the legitimacy, or ‘symbolic capital’, generated by its life and death struggle with state security forces, its ability to intimidate dissenters, and tight control over the composition of key bodies, including HB (1993, 112).

Herri Batasuna and its successors   65 KAS – and the broader movement of the Basque nationalist left – embodied a division of labour, where ETA was military vanguard and HB represented the movement in the electoral arena (Mata, 1993, 111 and 124). HB’s role was not so much to pursue power through public office, but to pursue the ‘accumulation of forces’, or to take on social grievances and become a focal point for Basque radicalism of all kinds (Mata, 1993, 124–127). The ‘symbiotic relationship’ between HB and ETA was reflected in the presence of KAS in the directorate of HB and various convicted ETA members among the leadership, elected representatives and candidates of HB and its successors (Mata, 1993, 123–127; Muro, 2008, 127–131). In electoral campaigns, HB assumed the role of mouthpiece for ETA, as its decision to use the entitlement given to all legal parties for free broadcasts during elections to broadcast a video produced by ETA illustrates (Fernández Hernández, 2008, 192). According to Mata, ETA’s supporters tended to see violence as an effective instrument for pushing the government to negotiate independence, for shaping broader political, economic and social conditions in the Basque Country, for communicating the intensity of dissatisfaction with the status quo, and for keeping the Basque nationalist project alive through constant agitation and consciousness-­raising (1993, 279–283). This orientation to violence was grounded on an interpretation of the Spanish transition as a continuation of Francoism, its outcome at best a ‘pseudo democracy’ that was repressive and ‘anti-­Basque’ (Muro, 2008, 133–139; Mata, 1993, 280–282). These views led many radical Basque nationalists to rationalize ETA’s violence by comparing it with violence employed by state (Mata, 1993, 285–286). Securitization as ‘terrorist organizations’: ‘Batasuna is ETA, ETA is Batasuna’ Some have explained the long years of HB’s legality with reference to a political culture inherited from the transition to democracy where, in light of long experience of dictatorship, Spanish political elites were reluctant to limit freedom of association or speech (Ferreres, 2004, 141–142, 153). As de Otto Pardo (1985) put it, the reactions of state institutions against its enemies was seen as the principal threat to the democratic system, rather than the threat from radical ideologies themselves. Esparza argues that in the 1980s and 1990s, political elites sought to harness extremists and radical ideologies through integration in the political system (2004, 18). Vidal argues that elites assumed more tolerant state responses would undermine critiques of the Spanish state underpinning key nationalist grievances (2009, 249). Furthermore, as Esparza observes, the failure of attempts to ban HB in the 1980s discouraged successive governments from initiating further ban proceedings for many years (2004, 146). By 2002–03, however, all party-­ban-veto-­players – including leading parties in parliament, the party of government, the Public Prosecutor, Supreme and Constitutional Courts – had embraced a conception of HB and its successors as security threats.

66   Herri Batasuna and its successors The Audiencia Nacional, a judicial body specializing in terrorism cases but not involved in party ban cases under the 2002 Party Law, played a key role in this evolution. In the 1990s, the Courts and a majority of legal scholars narrowly construed the ‘terrorist organization’2 as a sufficiently permanent collaborative association, aiming to violently subvert the constitutional order or alter the public peace, whose members used arms or explosives for the purposes of mass intimidation or terror (Cancio Meliá, 2010, 167; Gil, 2015, 340). As such, an organization that did not use arms or explosives was not considered a terrorist organization. As part of a more general trajectory characterized by an increase in the types of terrorist offences and the gravity of penalties for such offences (Benavente and Manso, 2014, 217), a broader interpretation of what constituted a terrorist organization took form following publication of Sumario 18/1998 by the Audencia Nacional, under authority of Judge Baltasar Garzón (Gil, 2015, 342; Esparsa, 2004, 141–144). The central argument of the indictment followed the lines of the academic studies cited above. That is, the Court acknowledged that in the early transition, ETA developed the strategy of desdoblamiento (roughly translated as splitting, unfolding or widening), or a division of labour in which ETA would conduct a necessarily clandestine military campaign, while a broader movement of political and social organizations aimed to win support for independence through legal means but hide their dependence on ETA. Referring to this strategy, the Court developed a conception of ETA and many organizations of the Basque nationalist left, including political parties, as part of a common framework or a ‘single criminal enterprise’ in which the parts contributed to a common criminal objective (Gil, 2015, 344). This more expansive conception of the terrorist organization pioneered in the Audencia Nacional, integrating both terrorist commandos and supporting organizations, political parties and networks, effectively amounted to a broadening of the concept of the ‘terrorist organization’ in Spanish criminal law (Cancio Meliá, 2010 and 2011; Avilés, 2010, 48; Tajadura and Vírgala, 2008, 53). It led to the closure of media outlets, businesses and political associations, imprisonment of many dozens of activists, and suspension and illegalization of various political parties in accordance with the Criminal Code (see Chapter 2). Terrorist organizations are a prototypical security threat. They have traditionally been defined in Spanish criminal law with reference to the goal of subverting the constitutional order and gravely altering the public peace (Benavente and Manso, 2014; Esparsa, 2004, 139). Since 2015, the Criminal Code has also included the goals of seriously destabilizing the functioning of international organizations or political institutions, economic or social structures of the state and provoking a state of terror in the population or part of it. The conceptual elision of political organizations of the Basque nationalist left and the terrorist organization ETA thus involved a securitization of those political organizations through the claim that they were integral to and thus as threatening as the terrorist organization ETA. This broadening of the conception of the ‘terrorist organization’, and ultimately of the concept of the ‘terrorist’ itself to include political organizations, was later endorsed in the rulings of Courts beyond the

Herri Batasuna and its successors   67 Audiencia Nacional and became a central claim justifying illegalization of Batasuna under the 2002 Party Law. In debates regarding the 2002 Party Law and the August 2002 parliamentary resolution urging the government to initiate ban proceedings, leading members of both the governing PP and the opposition Socialists frequently justified their pro-­ban positions with claims such as ‘ETA is Batasuna and Batasuna is ETA’, that Batasuna ‘was the same as ETA’, or that Batasuna was the ‘political wing of ETA’ (El Mundo, 2001a, 2001b, 2002a, 2002j, 2002k, 2002l, 2002o, 2002x and 2003a). Others signalled this view by merely referring to ‘ETA-­Batasuna’ (El Mundo, 2002v) and there were frequent claims that Batasuna was a ‘strategic instrument of the terrorist group ETA’ (El Mundo, 2002d). PP President of the government José María Aznar argued: ‘We have to confront, with all its consequences, not just part of terror, but all the terror. And all the terror in Spain also carries the name of Batasuna’ (El Mundo, 2002ae). PP Interior Minister Angel Acebes argued that ‘it is proven [that Batasuna] forms part of a terrorist organization’ and that ‘Batasuna is an essential element of the strategy of ETA’ (El Mundo, 2002ab; CDDS 164, 2002, p. 6300–6302). PP Mayor of Vitoria in the Basque Country, Alfonso Alonso, argued in the Supreme Court that ‘even babies at the breast know Batasuna is part of ETA’ (El Mundo, 2003b). PSOE deputy Juan Fernando López Aguilar argued that ETA was not only constituted by its armed commandos but a ‘terrorist conglomerate’ upon whom depend various “tentacles” which support the armed group’ (El Mundo, 2002u; CDDS 181, 2002, p. 9096). Batasuna’s role in the terrorist organization was described as giving ETA ‘cover and support’ (El Mundo, 2002ab, 2002z), being ETA’s ‘political face’ (El Mundo, 2002y), ‘a propaganda loudspeaker’ (Bermúdez (PP), CDDS 181, 2002, p.  9093), ‘to finance [ETA]’ (ibid. and El Mundo, 2002f ), ‘to recruit young people into operative commandos’ (El Mundo, 2002ab) and identify targets for ETA attacks (El Mundo, 2003a). Insecurity created by ETA attacks was often simultaneously attributed to Batasuna. Interior Minister, Angel Acebes, for instance, argued that ‘ETA and Batasuna generate fear, terror, attacks when they can’ (El Mundo, 2002ab, 2002y). Similarly, PP Minister of Justice, José María Michavila, argued that Batasuna contributed to threats of ETA, generating a climate of terror and fear that ‘suffocated’ the life of the democrats (El Mundo, 2002y). PP secretary general in the Basque Country, Carmelo Barrio argued that ‘the Basques in general, and non-­nationalists in particular are going to benefit enormously from application of the Law of Parties against the blackmail that ETA-­Batasuna submit us to every day’ (El Mundo, 2002v; see also El Mundo, 2003b; Acebes (PP Justice Minister) CDDS, 164, 2002, p. 8300). The Supreme Court, in its sentence of 23 March 2003 banning Herri Batasuna and its successors, supported the government and Public Prosecutors’ arguments that the parties did not just support or collaborate with ETA but were part of the terrorist organization. It considered proven that, although there was a ‘tactical separation’ between the parties and ETA, the parties were nevertheless under the full control of ETA and its delegate, the coordinating body KAS. The parties

68   Herri Batasuna and its successors followed and executed the orders of these organizations. Implementing ETA’s strategy of desdoblamiento, the parties would serve as ‘political’ and ‘political-­ institutional’ fronts with two tasks: Participating in elections and to attract and use ideologically proximate groups to achieve the goals of the terrorist organization by becoming focal points for mobilization on political issues. The parties were ‘instruments’ of ETA, and their leadership was controlled by people who enjoyed the strict confidence of ETA and its delegate, KAS, to the point that ETA selected members of their directing bodies (Mesas Nacionales). There was a clear affinity between ETA and the parties, few disputes and, when small discrepancies about the legitimacy or advantages of the terrorist strategy emerged, ETA reasserted its control over the party. The Constitutional Court, examining evidence accrediting the ‘operative succession’ of HB, EH and Batasuna and their submission to the strategy of ETA, concurred with the Supreme Court. The ECtHR’s ruling also built on acceptance of the core securitization argument that HB and its successors were, with ETA, part of a single entity. It concurred with the Spanish Courts’ conclusion ‘that there was a link between the applicant parties and ETA’ and following a long history of terrorist attacks, that this link between the parties and ETA ‘constitutes a threat to democracy’ and that the threat was a ‘considerable threat’. The conceptual elision of HB and its successors with the terrorist organization ETA by all party-­ban-veto-­players permitted the direct mobilization of the targeted parties’ ambiguity about the appropriateness of political violence in the specific grounds justifying their illegalization. As mentioned in Chapter 2, a key innovation of the 2002 Party Law was addition of specific provisions permitting the illegalization of parties defending the illicit use of violence or supporting those who employed it. During 2002, when debating the 2002 Party Law and August 2002 parliamentary resolution calling on the government to initiate ban proceedings, political parties and civil society groups favouring the ban predominantly justified their position with reference to Batasuna’s defence of political violence or support for ETA. Presenting the draft law to Parliament, Angel Acebes, then PP Justice Minister, argued it was a ‘perversion of democracy’ to allow political parties, ‘as it occurred in Spain’: … to be deposits of arms, nurseries or refuges for assassins, schools for violence or centres for logistical support for terrorists; … an apparatus for the financing, management or propaganda of violent groups; … [which] use institutions to generalize fear, to sew a mesh of terror and exclusion in every village … use public subsidies to develop networks of mafia-­like financing, that begin with kidnapping, robbery and threats … and occupy parliamentary seats or councils to better obtain information. (CDDS 164, 2002, p. 6300) President of the government José María Aznar claimed the 2002 Party Law would ‘permit the state to act forcefully against the terrorists, their helpers and accomplices’ (El Mundo, 2002c). Various PP ministers described Batasuna as

Herri Batasuna and its successors   69 ‘a violent party that shelters violence’ (El Mundo, 2002t), as ‘threatening and intimidating’ (El Mundo, 2002y), argued that Batasuna articulated ‘an authentic exaltation of ETA and the terrorist movement’ (El Mundo, 2002t), or argued it was ‘very difficult’ to understand how a party, such as Batasuna, which did not condemn ETA assassinations, including those of children, could be legal (El Mundo, 2002h). Similarly, leading members of the opposition Socialists described the attitude of Batasuna to terrorist attacks by ETA as ‘criminally passive’ (El Mundo, 2002m), argued that the ban on Batasuna responded to ‘justification of horrible crimes or threats to representatives of Spanish democratic society’ by Batasuna leaders (El Mundo, 2002y and 2002u), or described it as a party that ‘justifies and protects terrorism’ (El Mundo, 2002i and 2002j). Spokesperson of the Asociación Víctimas del Terrorismo (AVT, Association of Victims of Terrorism), Paulino Baena, described Batasuna, whose illegalization it supported, as a ‘party that not only does not condemn the attacks of ETA but justifies and applauds them’ (El Mundo, 2002ag). In its ban ruling, the Supreme Court accepted evidence and arguments submitted by the government and Public Prosecutor that the banned parties, or their leading members, had undertaken acts under nearly each of the provisions of Articles 9.2 and 9.3 of the 2002 Party Law. They had justified or excused attacks against the lives or physical integrity of others (Article 9.2.a); complemented and supported politically the actions of a terrorist organization in the achievement of its goals, or multiplied the effects of terrorist violence (9.2.c); explicitly or tacitly supported terrorism and justified, excused or minimized the significance of terrorist actions (9.3.a); persecuted, intimidated, coerced, neutralized or socially isolated people opposed to terrorist activity, and made opponents live daily in an atmosphere of coercion, fear, exclusion or basic privation of liberties (9.3.b); used terrorist symbols, including that of banned ETA prisoners’ lobby group Gestoras Pro-­Amnistía (9.3.d); collaborated with organizations linked to terrorism (9.3.f ); and promoted or participated in acts of homage to terrorism (9.3.h). Evidence sustaining these conclusions included the known participation of significant numbers of people condemned for terrorist crimes in top positions within the parties and their electoral lists; refusal of Batasuna’s leaders and representatives to condemn terrorist attacks and threatening acts of its collaborators; and speeches or other acts of communications of party leaders, elected representatives and the municipalities it governed which encouraged, justified and protected ETA. In its Constitutional Court challenge to the Supreme Court ruling, Batasuna addressed what was arguably the most controversial item of proof employed by the Supreme Court against it, namely Batasuna’s refusal to condemn terrorist attacks. Along with many other items of evidence against it, Batasuna defended its refusal to condemn ETA’s terrorist attacks as the exercise of rights to free ideology and expression. The Supreme Court had ruled that a refusal to condemn terrorist acts, in the context of the gravity of the situation in which the refusal took place, were not mere expressions of political pluralism. Rather, they were a ‘clear and precise manifestation of rejection of democratic methods in political

70   Herri Batasuna and its successors life and their substitution for other non-­democratic ones’ and a legitimization, excusing and minimization of terrorist acts. The Constitutional Court rejected Batasuna’s claims, ruling, among other things, that it was reasonable to infer from a refusal to condemn terrorist violence, alongside the plurality of acts and conducts considered by the Supreme Court, as ‘a commitment to terror and against organized coexistence in a democratic state’. The issue of condemning violence, among other things, was raised again in HB and Batasuna’s ECtHR’s appeal. As mentioned above, ECtHR jurisprudence establishes that a party’s orientation to violence – and particularly their failure to employ entirely democratic and legal means to pursue their goals – is a key criteria for permissible restrictions on freedom of expression and association, such as illegalization of parties (Harvey, 2004; Brems, 2006a). Prior to its ruling on party bans in Spain, the ECtHR had clearly stated that ‘a political party whose leaders incite to violence … cannot lay claim to the Convention’s protection against penalties imposed on those grounds’ (see in particular ECtHR, 2009). In its judgement in HB and Batasuna vs Spain, the ECtHR accepted that illegalization responded to a ‘pressing social need’, and was therefore necessary in a democratic society. More specifically, it ruled that the conduct of the banned parties ‘bears a strong resemblance to explicit support for violence and the commendation of people seemingly linked to terrorism’; the ‘members and leaders of the applicant parties had not ruled out the use of force with a view to achieving their aims’ and the parties’ conduct could be considered ‘capable of provoking social conflict between supporters of the applicant parties and other political organizations’. These conducts and the parties’ failure to condemn terrorist violence ‘made it possible to conclude that there had been an accommodation with terror’. By ‘providing assistance and political support to the actions of terrorist organizations’, the parties employed unlawful and undemocratic methods to achieve their goals. For these reasons, the ECtHR also argued that interference with freedom of expression and association pursued the ‘legitimate aims’ spelt out in Articles 10.2 and 11.2 ECHR of ‘public safety’ and ‘the prevention of disorder’. Securitization as a threat to the democratic community A second mode of securitization accepted by all party-­ban-veto-­players was that HB and its successors constituted a threat to the democratic community. Many of those supporting the ban in public debates on the 2002 Party Law and the August 2002 parliamentary resolution calling for the government to initiate ban proceedings described Batasuna as undemocratic and conceived of the ban as a measure of democratic defence. ETA and Batasuna were repeatedly described as ‘fascist’, ‘dictators’ or ‘totalitarian’ (El Mundo, 2002z and 2002b), or as posing a threat to the exercise of fundamental rights and liberties of others (El Mundo, 2002q, 2002z and 2002ad; López (PSOE), CDDS 168, 2002, p.  8458). As Socialist deputy Diego López Garrido argued, when a party that collaborates with a terrorist organization enters into the political institutions of the state, it does ‘damage to democracy’, bringing ‘enormous tension’ and ‘a logic of

Herri Batasuna and its successors   71 violence into the heart of the institutions’ (CDDS 168, 2002, p. 8458). The 2002 Party Law, according to Garrido, thus not only sought to avoid terrorist actions or ‘finish with terrorist groups’, but to ‘improve public life and democracy in this country’. PP deputy Javier Arenas, insisted that the party law ‘criminalises those who [i.e. ETA and Batasuna] really criminalize ideas, … because they are killing people for their ideas, because they don’t share the claims of these people they think they have a right to kill or kidnap them’ (CDDS 168, 2002, p. 8460). Referring to targeting of politicians and their families, Carlos Totorica, from PSE EE(PSOE), which was divided on the issue of the party ban, stated, ‘I don’t care if they [ETA-­Batasuna] are 15 or 85 per cent of the Basque population, they don’t have the right to get together to threaten me every day, that is not liberty’ (El Mundo, 2002a). PP Justice Minister, Angel Acebes, argued that Spanish democracy was a ‘precarious democracy’ and it would be a ‘perversion of democracy’ to allow political parties such as Batasuna to be legal (CDDS 164, 2002, p. 8300), while the Socialist parliamentarian Jesús Caldera argued ‘the Batasuna we know today is incompatible with democracy’ (CDDS 164, 2002, p.  8322). Others argued the ban was ‘urgent for the dignity of democracy’ (El Mundo, 2002e), that banning Batasuna ‘reinforced democracy’ (El Mundo, 2002a) and that Batasuna only targeted those who ‘wanted to use democracy to end democracy and liberties’ (El Mundo, 2002c). Foro Ermua, a civil society movement promoting recognition of the victims of ETA, described the 2002 Party Law as ‘an instrument for the defence of democracy’ and for combatting ETA’s ‘dictatorship of terror’ as well as a commitment for ‘liberty against fascism’ (El Mundo, 2002s). This securitization frame underpinned a further set of party ban justifications focusing on the violation of core democratic principles, a justification accepted by all party-­ban-veto-­players. As mentioned in Chapter 2, the 2002 Party Law explicitly links justification for party bans to the threats they pose, not just to the lives of individuals and groups targeted by a terrorist group, but to those people as citizens and the democratic community as a whole. The Supreme Court accepted the arguments of the government and Public Prosecutor, ruling that HB and its successors fulfilled requirements permitting party bans in the 2002 Party Law by failing to respect political pluralism and democratic behaviour (Article 9.1) and by systematically pursuing the deterioration or destruction of the constitutional regime of liberties and preventing the exercise of democratic liberties (9.2). More specifically, the Court ruled that Herri Batasuna and its successors promoted, justified and excused attacks against the lives of individuals and prevented people who did not agree with their beliefs from exercising their rights and liberties (9.2.a). The parties legitimized and promoted the use of violence as a method to achieve political goals and sought to eliminate, constrain or interfere in constitutionally established conditions for the exercise of democracy and pluralism and political liberties (9.2.b). By their actions, the parties complemented and politically supported terrorism, trying to subordinate social groups that did not agree with their ideas, subjecting civil society to a climate of terror and contributing to the multiplication of the effects of ETA’s violence (9.2.c).

72   Herri Batasuna and its successors One of the most prominent arguments about the anti-­democratic nature of the parties referred to premediated campaigns against people publically confronting the positions of ETA by activists of the Basque nationalist left. The courts ruled the campaigns sought to discredit, socially isolate and proffered physical threats against opponents and their families and in so doing systematically undermined the basic liberties of freedom of opinion and participation in public affairs. More generally, the Supreme Court emphasized the damage ETA did to the ‘normal unfolding of democratic institutions and life of citizens’; the threat to political pluralism posed by the targeting of citizens who did not share ETA’s ideas and objectives and who opposed them; as well as the negative impact of threats against politicians from state-­wide parties, such as the PSOE and PP, for the conduct of democratic politics and their rights to participate in public life. The Constitutional Court, when endorsing the Supreme Court’s ruling, further argued that the plurality of acts attributed to Herri Batasuna and its successors justified denying the banned parties the status of, and privileges granted to, political parties in the Spanish constitution. Freedom of association, according to the Constitutional Court, could not be extended to parties ‘associated with terrorism and violence’ and which offered ‘political and ideological support to the actions of terrorist organizations’, nor could freedom of expression be given to those who supported actions of terrorist groups or intimidating conduct. For its part, the ECtHR has established a party’s ‘exceptional’ threat to a democratic system as a core criteria justifying party bans compatible with the ECHR. Prior to its ruling on party bans in Spain, ECtHR jurisprudence had already established that political parties must not only employ legal and democratic means to achieve their goals, but that the changes parties proposed to the law, structures or constitution of the state must themselves be compatible with fundamental democratic principles (Harvey, 1994; Brems, 2006a). As such, in addition to parties whose leaders incite violence, those that put forward a policy which fails to respect democracy or which is aimed at the destruction of democracy and the flouting of the rights and freedoms recognized in a democracy cannot lay claim to the Convention’s protection against penalties imposed on those grounds. (EctHR, 2003) Bans for threats to democracy must be existential threats – that is, applied only when a party ‘seriously endangered the very survival of the democratic system’. This was a core criteria for judging whether the grave incursion on the rights of parties entailed by proscription was proportionate to the impugned conduct of the party. The ECtHR had also established that preventive bans, where the danger of a party to civil peace and a country’s democratic regime was ‘sufficiently established and imminent’, were permissible (ECtHR, 2003). In its HB and Batasuna ruling, the ECtHR accepted the view of Spanish authorities that ‘the applicant parties had constituted a threat to human rights, democracy and pluralism’ and that ‘dissolution had been a means of preventing

Herri Batasuna and its successors   73 the applicant parties from acting against the democratic system and the fundamental freedoms of citizens by supporting violence and the activities of the terrorist organization ETA’. For these reasons, the ECtHR considered the restriction on freedom of association of Article 11 ECHR justified by the permissible ‘legitimate aims’ of ‘protection of the rights and freedoms of others’, in addition to the above-­mentioned goals of ‘public safety’ and the ‘prevention of disorder’. Furthermore, the Court ruled that the ban corresponded to a ‘pressing social need’, and was thus ‘necessary in a democratic society’ and ‘proportional’ to the impugned conduct. The parties pursued actions that were ‘part of a strategy … to achieve a political aim essentially in breach of the democratic principles enshrined in the Spanish Constitution’ and provided ‘assistance and political support to the actions of terrorist organizations with the aim of overthrowing the constitutional order’. For the Court, the acts and speeches imputable to HB and Batasuna showed the party pursued a model of society incompatible with the concept of a democratic society. While accepted by veto players, securitization of Herri Batasuna and its successors as an integral part of a terrorist organization and threat to democracy meriting the exceptional measure of the party ban were subject to challenge. Xabier Arzalluz, President of the EAJ/PNV, the moderate Basque nationalist party then governing in the Basque Country, argued that Batasuna ‘had never set a bomb’ and that the ‘affirmation that Batasuna is the same as ETA is false; we don’t know if some are from ETA, but there are many that reject it’ (El Mundo, 2002aa and 2002af ). Others from the EAJ/PNV argued ‘only ETA is illegal’ and that voting for Batasuna ‘does not imply that you are a murderer’ (El Mundo, 2002p, 2002w, 2002ac). As mentioned above, Batasuna’s spokesperson, Arnaldo Otegi, argued that the party ‘did not wear hoods or undertake armed activities’ (El Mundo, 2003c). Additionally, some challenged the conception of Batasuna as a serious threat to democracy and the appropriateness of developing a new legal pathway, in addition to the Criminal Code, for banning parties through the 2002 Party Law. EAJ/PNV deputy Josu Erkoreka argued, for instance, it was not plausible to argue that parties found guilty of conduct spelt out in the 2002 Party Law actually put Spanish democracy – now a mature and fully consolidated democracy – in any serious danger (CDDS 164, 2002, Erkoreka (PNV), p.  8306, see also Llamazares (IU) p.  8310). Some also criticized the 2002 Party Law itself as undermining democracy. Gaspar Llamazares, spokesman for the leftist IU, rejected the 2002 Party Law on the grounds that it ‘reduced public liberties and procedural guarantees integral to the State of Law’, and produced ‘the contamination of our constitutional model of political parties and a frontal attack on Spanish democracy’ (CDDS 164, 2002, pp.  8309 and 8311). EAJ/PNV argued that ‘mutilating the party system by dissolving a political party is equivalent to deforming the popular will and therefore a perversion of the system of democracy’ (CDDS 164, 2002, p. 8306). Batasuna leader, Joseba Alvarez, argued that ‘without Batasuna there will be no normal elections, no normal institutions and no political legitimacy’ and that the ban was against the will of the majority of

74   Herri Batasuna and its successors Basque society (El Mundo, 2002n). For its part, ETA claimed that illegalization of Batasuna would only serve to aggravate conflict in the Basque Country and distance it from peace (El Mundo, 2002r). While in 2002–03 the arguments against securitization of Herri Batasuna and its successors were those of a minority, by 2011–12, when the legal status of the new party Sortu and coalition Bildu was on the political agenda, these arguments ultimately became those of the veto-­players, as I now turn to spell out.

Legalization of Sortu and Bildu Ceasefire and the Basque nationalist left’s break with ETA The legalization of Herri Batasuna’s successors – first the coalition Bildu and, subsequently, the party Sortu – followed ETA’s declared intention to abandon armed struggle and efforts by political organizations of the Basque nationalist left to demonstrate a break with ETA and an unambiguous commitment to non-­ violence. In September 2010, ETA declared a ‘cessation of armed offensives’. In January 2011, it declared a ‘permanent, general and verifiable’ ceasefire. Soon afterwards, in February 2011, Sortu was formed, followed, in April 2011, by the formation of Bildu. By October 2011, ETA had declared the ‘definitive cessation of armed activity’. These moves had been preceeded by calls from high-­profile international figures – e.g. South African Lawyer Brian Currin and UN Secretary General Kofi Annan – for ETA to abandon violence and Spanish and French governments to enter into negotiations with it over prisoners and disarmament. These interventions were not welcomed by successive Spanish governments. In April 2017, ETA handed over a cache of arms to French security services, overseen by international mediators, and announced its disarmament. Many have argued that a withdrawal of support from ETA’s own base was crucial in its decision to abandon violence (Murua, 2017; Zabalo and Saratxo, 2015; Domínguez, 2012). According to Murua, withdrawal of support was due to a gradual loss of confidence in ETA’s military strategy after the successive failure of negotiations with the government, major police operations, disappointment with the failure of the Lizarra-­Garazi process and ultimately frustration provoked by the failed 2006–07 process (2017, 98). Changes in the form of political violence employed, such as the indiscriminate nature of car bombs, the ‘qualitative leap’ of targeting political representatives and the cruelty of specific acts such as the kidnapping of prison officer José Antonio Ortega Lara and assassination of PP councilor Miguel Angel Blanco were also ‘turning points’ in the Basque nationalist left’s re-­evaluation of armed struggle (ibid., 99). In the context of ETA’s increasing operational weakness, the political wing was gradually able to impose a strategy of non-­violence on the military wing (ibid., 109; see also Domínguez, 2012, 274–292). Between October 2009 and February 2010, leading figures from the Basque nationalist left, including former Batasuna leader Arnaldo Otegi, initiated a strategic debate in the movement defying the taboo of questioning the armed

Herri Batasuna and its successors   75 struggle and the utility of ETA’s use of arms (Murua, 2017, 101; Zabalo and Saratxo, 2015, 375; Domínguez, 2012, 274–292). The move was resisted by ETA, who produced its own contribution in the debate favouring a continuation of the political-­military strategy (in the document Mugarri (Milestone)), but Batasuna was able to lead debate on the basis of its own proposal (Argitzen (Clarifying)). The outcome of the internal debate was support by the majority of Basque nationalist left activists for a new strategy based on non-­violent pursuit of political goals (Murua, 2017, 102; Zabalo and Saratzo, 2015, 375). In September 2010, the Basque nationalist left, Basque political parties, including those that would eventually form Bildu, trade unions and civil society groups signed the Gernika Agreement, which publically called on ETA to end violence and for the Spanish government to begin negotiations with ETA. Sortu’s party Statute locates the origins of the party in these internal debates. The Statute affirms that the party’s commitment to exclusively political and democratic means is firm and unequivocal, unaffected by variation in tactics or contextual factors, and [that it] will therefore defend its ideology on the basis of respect for democratic principles and through legal means of intervention in public life. The Statute specifically states that Sortu will oppose any external ‘subordination, constraints or tutelage’ that aims to convert it into an ‘auxiliary organization of those that promote violence’; that its strategy is incompatible with ‘violent acts or conduct that serve to complement or support them politically’; and that the party ‘rejected, openly and unequivocally, the organization ETA’. The Statute commits the party to the principles of the 2010 Gernika Agreement, restating a commitment to the Mitchell principles of non-­violence (developed during the Northern Ireland Peace Process). Documents establishing Bildu similarly referred to a commitment to ‘exclusively peaceful and democratic means and the defence of all human rights’ and to pursuit of political conflict with reference to a commitment to the ‘Mitchell principles’ of non-­violence (EA and Nationalist Left, 2010; EA, Nationalist Left y Alternatiba, 2011). The two legal parties forming Bildu alongside independents from the Basque nationalist left – EA and Alternatiba – had unquestioned credentials as staunch opponents to ETA and had long democratic traditions. In light of recent experiences, many were sceptical about whether ETA’s 2011 ceasefire was genuinely permanent. In 2006–07, an earlier ‘peace process’ had been initiated under the Socialist government of José Luis Rodríguez Zapatero but had been derailed after ETA exploded a car bomb at Barajas airport in Madrid in December 2006. The process had consisted of negotiations between ETA and government representatives on ‘technical’ issues of prisoners and disarmament and political dialogue between Batasuna, EAJ/PNV and PSOE in a round-­table of parties (Bew et al., 2009; 231–238). This was the most recent of various efforts to end violence through some form of dialogue, all of which had failed.

76   Herri Batasuna and its successors As Murua remarks, it is not easy to determine whether [the 2009–10 internal debates were] a debate among the social base of Batasuna, the party, or among the whole Nationalist Left, the movement. The ambiguous and undefined boundaries between the two concepts swallow both interpretations, and the fact that the majority of Batasuna’s social based also make up the majority of the Nationalist Left allowed the Batasuna leadership to present the debate’s conclusions as the will of the majority of the whole movement. (2017, 102) Against a background of failed peace initiatives, mistrust and suspicion, this ambiguity about the boundaries between the party, movement and armed group was initially a major stumbling block in the legalization of Sortu and Bildu. Until the Constitutional Court reached its decision, the ambiguity confused the issue of whether there really was a breach between the political and military wing of the movement and thus whether it was still plausible to claim, as was the case in the 2003 ban case, that ETA and Batasuna were one and the same. Nevertheless, these internal developments in the Basque nationalist left ultimately opened the way for legalization of Batasuna. In legal terms, the path to legalization of Batasuna required that new incarnations of the party – in this case Sortu and Bildu – satisfied what can be called the condemnation-­of-violence ‘counter-­indicator’. The 2002 Party Law and LOREG established a number of indicators, refined by the jurisprudence of the Supreme Court and Constitutional Court, by which the courts would judge whether a new or existing party, federation, coalition or group of electors took the form of a successor to, or continued the activities of, a banned party. ‘Subjective indicators’ were similarities between banned and succeeding parties in terms of the people who formed, directed, administered or joined them. Objective indicators were similarities between the structures, organization and functioning of banned and succeeding parties and in the sources of their material and financial means. The 2002 Party Law and LOREG also permitted the Courts to take into account other relevant circumstances, particularly the new party’s disposition to support violence or terrorism. From this law, the Courts subsequently developed the concept of the condemnation-­of-violence ‘counter-­indicator’. Even if it could be proven that there were ‘subjective’ or ‘objective’ links between a banned and a new party, a new party would not be banned if it explicitly condemned violence. This was the path to legalization for the parties of the Basque nationalist left. It meant a new party, linked in terms of its personnel and supporters to HB, EH and Batasuna and/or the long list of other banned parties and electoral groupings (see Table 2.1) could return to the public sphere if they unambiguously rejected violence, including that of ETA (Català i Bas, 2013, 567; Iglesias Bárez, 2013). The centrality of the condemnation-­of-violence counter-­indicator in the legalization process focused attention on the changing nature of the security threat posed by Bildu and Sortu. Despite continued appeal to securitization frames in

Herri Batasuna and its successors   77 political debates and legal argumentation, legalization was ultimately achieved as a result of a desecuritization process in which the Constitutional Court imposed a conception of Sortu and Bildu as separate from, and thereby less threatening than, the banned Batasuna and ETA. Desecuritization of Sortu and Bildu and mediation of difference through the public sphere Despite substantial changes in the political context described above, the two main parties in parliament (PP and PSOE), the party of government (PSOE), the Public Prosecutor and majorities of the Supreme Court re-­articulated the securitized conception of first Sortu and then Bildu as an instrument of ETA. For both the Socialist government and the PP, while ETA remained in existence, Sortu’s rejection of violence and of ETA itself was insufficient to justify legalization (El Mundo, 2011h and 2011f ). Attorney General, Cándido Conde-­Pumpido, argued that Sortu ‘should not be legalized while the backdrop of ETA’s defiant shadow remains in political life’ and that given the ‘ambiguity of ETA’s September 2011 ceasefire announcement’ the police, judges and prosecutors should remain ‘vigilant and inflexible’ in the implementation of the law (El Mundo, 2011t). As Basque PP Secretary general, Iñaki Oyarzábal, put it, there could only be ‘generosity’ towards the Basque nationalist left when ETA had disappeared, if it ‘recognized the damage’ it had caused in society, when terrorism and threats had finished, and when ‘los Batasunos’, who still dare not condemn ETA, have taken a few more steps (El Mundo, 2011d; see also El Mundo, 2011v). Both the government and opposition PP remained convinced that like Batasuna, Sortu was a party at the service of ETA and a successor to Batasuna (El Mundo, 2011f, 2011g and 2011i). While acknowledging that ‘things are happening’ in the Basque nationalist left and that the goal of constructing a ‘pro-­ independence political space without violence’ was ‘more or less believable’, the government considered claims that the Basque nationalist left sought to lead ETA away from violence were ‘confused’ and lacked ‘sufficient credibility’ (El Mundo, 2011f and 2011v). Minister of the Presidency, Ramon Jáuregui, further argued that ‘Spanish democratic society has a “right to doubt” the intentions of Sortu, because so many things have occurred over many years’ (El Mundo, 2011h and 2011t), while the PP’s Federico Trillo argued that while ETA existed, ‘we don’t believe absolutely anything from the Nationalist Left’ (El Mundo, 2011g). PP parliamentary spokesperson, Soraya Sáenz de Santamaría, continued to insist that ‘Sortu is Batasuna and Batasuna is ETA’, emphasized that the terrorists should ‘never return to the institutions’ and warned that ‘the terrorists use all types of means to achieve their objective’ of getting into the institutions (El Mundo, 2011g, see also 2011d, 2011k). Sortu was elsewhere described as ‘a successor of Batasuna’ and ‘the umpteenth face of ETA’ (El Mundo, 2011d). PP President of Bizkaia (in the Basque Country), Antón Damborenea, argued that ‘Sortu is a brand, Bildu is another brand, they can have 50 brands, because, in

78   Herri Batasuna and its successors the end, what we are dealing with is Batasuna, which is ETA, as proven by all the European and Spanish courts’ (El Mundo, 2011j). Even after its legalization, Unión del Pueblo Navarro’s (UPN, Union of the Navarrese People) deputy in the Spanish parliament, Carlos Salvador, described Bildu as ‘inheritors of Batasuna’, ‘ETA’s defenders’, ‘friends of ETA’, a ‘philo-­ETA phenomenon’ using ‘mafia methods’ which promoted its objectives with ‘coercion and threats’ (El Mundo, 2011j). Some civil society organizations representing victims of terrorism, such as AVT and Foro Ermua, continued to describe Bildu and Sortu as the ‘political wing of the terrorist group ETA’ or similar, and supported proscription to prevent the ‘return of ETA to the institutions’ and they were subordinate to the directives of ETA (El Mundo, 2011n and 2011q). Similarly, ban proponents continued to describe the parties as threats to democracy and frame illegalization processes as a ‘defence of democracy’, even after legalization. PP president in the Basque Country, Antonio Basagoiti, claimed Sortu defended a ‘totalitarian project’ (El Mundo, 2011v), and in 2013, continued to claim that Sortu ‘are not democrats or decent, because they do not condemn, repent, change or accept the pluralism of Basque society’ (El Mundo, 2013, see also 2011s). Similarly, UPyD leader, Carlos Martínez Gorriarán, claimed that Bildu were ‘a front for ETA’ and that their electoral successes ‘did not convert them into democrats’; rather they were the ‘Trojan horse of fascism’ (El Mundo, 2012). Gorka Maneiro, of UPyD, argued that the legalization of Bildu was a ‘grave error’, ‘whose consequences will be very negative for the democratic system’ (El Mundo, 2011m and 2011p). UPN deputy, Carlos Salvador, called for action against the ‘friends of ETA’, on the grounds that wherever they govern ‘political pluralism, liberty, and a peaceful life disappear’ (El Mundo, 2011r; CDDS 265, 2011, p. 26). A communique from the civil society organization, Foro Ermua, called on the government to ‘take into account the damage done to democracy’ by allowing Bildu to contest elections and to proceed with new ban proceedings (El Mundo, 2011q). The submissions of the government and Public Prosecutor to the Supreme Court, and the ruling of the Supreme Court majority, continued to use the terminology of the securitization frame, referring to the ‘ETA/Batasuna framework’ or the ‘Batasuna/ETA complex’ to describe the leaders of the Basque nationalist left behind creation of Bildu and Sortu. More specifically, the majority of the Supreme Court ruled that Bildu was a coalition designed to circumvent the illegalization of ‘the political wing of the terrorist group ETA and thereby permit access of Batasuna/ETA to representative institutions’. The ‘Batasuna/ETA complex’ had sought to align with, but control, ideologically similar parties, permitting it to develop an electoral presence under cover of legal political parties (i.e. EA and Alternatiba) unstained by past links with illegal parties. Various agreements and evidence regarding their recruitment indicated that Bildu’s ‘independents’ were in fact ‘front men’ for the illegal Batasuna. The control of the ‘Batasuna/ETA complex’ over Bildu was, according to the Supreme Court majority, evident from the fact, among other things, that ‘independents’ within the coalition occupied the most numerous and most relevant positions in the

Herri Batasuna and its successors   79 majority of electoral lists. As far as Sortu was concerned, the Supreme Court ruled that Sortu was created by Batasuna and managed, encouraged and tutored by ETA. In both the Sortu and Bildu cases, majorities of the Supreme Court considered statements explicitly condemning or rejecting political violence, including that of ETA as ‘instrumental’ tactics endorsed by ETA itself to either create the conditions in which a new party could be legalized or to facilitate alliances with parties unequivocally opposed to ETA. In the Bildu case, the Supreme Court majority readily acknowledged that EA and Alternatiba had always unambigiously rejected violence. Nevertheless, in both Bildu and Sortu rulings, the Supreme Court majorities considered proven that ETA itself had delivered instructions on how the rejection of its own terrorist actions were to be presented. It ruled the declarations rejecting violence were formulated ambiguously and in a manner which drew ‘unacceptable parallels’ between ‘the legitimate action of the security services’ and the ‘crimes of terrorists’. The majorities of the Supreme Court in both Sortu and Bildu rulings additionally argued that it was necessary to prevent re-­emergence of new parties or to annul the coalition Bildu’s candidacies in order to guarantee the ‘liberty of citizens’ and protect democratic institutions. In the Sortu case, the majority of the Supreme Court ruled that, even with a ceasefire, ‘the very existence of the criminal group ETA and its political wing’ has ‘disturbed in an extraordinarily grave manner the social peace and normal peaceful coexistence of its citizens’, not only in the Basque Country, but also in the rest of Spain, with the consequence of ‘profoundly and negatively altering the public and constitutional order of the Spanish state’. The majority also argued that tolerance of the use of a legal party to assure the continuity of the ‘complex ETA-­Batasuna’, and allowing it to take advantage of the benefits that legality provided, posed a ‘grave, imminent and objective risk, directly impinging on the basic pillars of our democracy’. Permitting registration of Sortu, constituted an ‘unacceptable threat to life and the fundamental rights of others’ – which the Spanish state was obliged to protect – and this justified limiting the right of political participation for Sortu’s promoters. Furthermore, as the Court majority stated in the Bildu case, illegalization would prevent the independents (‘who were really ETA front men’) from, first, getting access to the electoral census which could be used by ETA for criminal ends and, second, because some candidates, if elected, could hold positions of power in territorial administrations and establish a line of government favourable to the ‘terrorist ends of ETA’. This securitized framing of Bildu and Sortu as successors to Batasuna, their continuity as an instrument of Batasuna/ETA and the threat they purportedly posed to the democratic process was challenged and ultimately superseded by a conception of the new parties as distinct from, and therefore less threatening than, the banned Batasuna and ETA. This process of redefinition, ultimately imposed by the Constitutional Court, involved a core attribute of what Hansen identifies as processes of desecuritization, namely a reconstitution of the ‘friend-­ enemy distinction’ (2012, 534).

80   Herri Batasuna and its successors The Constitutional Court’s Bildu ruling reiterated the importance of separating the ideology of the Basque nationalist left, which could not be a cause of illegalization, from support for violent means or perpetration of terrorist acts to achieve ideological goals, which could be cause for proscription. The Constitutional Court ruled that, with EA and Alternatiba, the Basque nationalist left had created Bildu after its own process of internal debate in which it had rejected violence. In its Sortu ruling, the majority of the Constitutional Court accepted at face value Sortu’s claims to have emerged from the Basque nationalist left, which had distanced itself from ETA, banned parties such as Batasuna and conduct justifying proscription. Focusing on the promoters of the new party – rather than party supporters with links to ETA or banned parties, the majority argued there were no indications of collaboration with either ETA or banned parties. Furthermore, in both cases, the majorities of the Constitutional Court argued there was not proof of sufficient weight, given the threat to fundamental rights of political participation, to support the claim that Bildu or Sortu had been instruments of ‘Batasuna/ETA’. In both cases, the Constitutional Court considered proven – but irrelevant – that ETA and Batasuna had sought to subordinate Bildu and Sortu to achieve their goals. The lack of convincing evidence of ‘subjective’ or ‘objective’ indicators linking Batasuna or ETA, on the one hand, to Bildu, on the other, led the Constitutional Court to deduce that Batasuna/ETA had not actually been able to use Bildu for their own ends. It made a similar argument when rejecting the Supreme Court ruling against Sortu. Against the Supreme Court, the majority in the Constitutional Court argued that Sortu could not be banned because its party statutes and similar manifestations by it promoters favouring non-­violent political means and rejecting ETA’s strategy of violence satisfied the condemnation-­of-violence ‘counter-­ indicator’, outweighing other evidence of succession or continuity of a banned party. It accepted that Bildu had eschewed violence, but did not consider it necessary to consider the issue in depth given the absence of evidence confirming other indicators of succession or continuity. In addition to the long democratic trajectory of EA and Alternatiba, the declaration of Bildu candidates affirming a commitment to use exclusively peaceful and democratic means and to oppose acts of violence and those violating human rights, was deemed sufficient proof of Bildu’s condemnation of violence and terrorism. Indeed, in the view of the Constitutional Court majority the Basque nationalist left had, as a consequences of its internal process of reflection, rejected ETA’s thesis of armed struggle and the tutelage of the terrorist group; rather it had affirmed mass politics, institutional and ideological competition as the only acceptable means to pursue the goal of Basque independence. As the six dissenting Supreme Court judges had put it, the evidence suggested the Basque nationalist left had liberated itself from ETA in a ‘Copernican shift from its traditional attitude of subordination to ETA … towards an absolute and unambiguous rejection of violence for political interventions, exclusively favouring political means’. Finally, in contrast to the Supreme Court, majorities in the Constitutional Court re-­interpreted the source of threat to democracy as the overzealous application of

Herri Batasuna and its successors   81 party ban rules. The Court ruled that, given the seriousness of the limitation on the right of association entailed, banning Sortu and Bildu were disproportionate. This was especially so following introduction of new provisions in the LOREG permitting prohibition of a party, once registered, or their candidates, once elected, for succession or continuity of the activities of a banned party. It also warned that ‘the effort to ensure at all costs the security of the constitutional state, through preventive controls, puts … the very constitutional state at risk.’ A second element of the desecuritization processes identified by Hansen that can be observed in this case was an increasing acceptance that the object framed as ‘threatening’– here parties of the Basque nationalist left – would best be dealt with through the mediation of difference in the public sphere rather than through exceptional measures – here, the party ban (2012, 531). Those against the proscription of Herri Batasuna and its successors have long argued that the best way of dealing with the political claims of the Basque nationalist left was through public debate and political competition. During 2002 Party Law debates, for example, IU deputy Gaspar Llamazares argued that while ETA and its supporters should be dealt with through ‘the work of judges, of the public prosecutor and the police’, it was necessary to allow politicians to work to find solutions to political problems (CDDS 164, 2002, p. 8310). In 2012, IU continued to argue that it had always believed ‘illegalization contributed nothing to bringing the Basque conflict to a strictly political terrain’ (CDDS 12, 2012, p. 31). In the 2002 Party Law debate, EAJ/PNV deputy Josu Erkoreka had similarly argued that the best ‘defence of democracy’ was to apply ‘the Criminal Code to criminal acts … and combat ideologically, all those activities, conduct and acts of political parties that are not criminal’ (CDDS 164, 2002, p. 8306). In 2011, after publication of Sortu’s Statutes, EAJ/PNV president Iñigo Urkullu, called other political forces to be conscious of this ‘important step’ and to ‘do what is possible to bring about the germination of a new time of normalization in political coexistence’ (El Mundo, 2011a; see also Esteban (EAJ/PNV), CDDS 12, 2012, p. 30). ETA’s ceasefire declaration and emergence of parties of the Basque nationalist left formally renouncing violence, including that of ETA, facilitated a broadening of the constituency favourable to this position. Shortly after the emergence of Sortu, leader of PSE EE(PSOE) and president of the Basque Government, Patxi López, argued legalization of Sortu would be ‘good news’, confirming that the party ‘met the democratic and legal requirements to do politics’ and sought to ‘integrate itself into democracy’, even if the party needed to do more to dispel doubts and mistrust due to the frustrations of the past (El Mundo, 2011b, 2011c and 2011e). In addition to expressing a similar need for caution PSE EE(PSOE) spokesperson, José Antonio Pastor, commented that ‘the value of this party is, if it complies strictly with the law … that people who were previously in a political world collaborating with violence, would stop doing it and join … democracy’ (El Mundo, 2011j). After the Constitutional Court ruling legalizing Bildu, the constituency accepting the position that parties of the Basque nationalist left ought to be dealt with through competition and debate in the public sphere came to include almost

82   Herri Batasuna and its successors the entire political spectrum, including the PSOE government and, after December 2011, the PP government of Mariano Rajoy. Party spokespersons frequently spoke of a new period opened by ETA’s ceasefire declaration and many rejected new ban proposals not only on the grounds that they were unlikely to succeed in the Courts but also as a step facilitating peace, reconciliation or negotiation of political differences. The PSOE government continued to wield the threat of illegalization, if the parties of the Basque nationalist left failed to respect ‘democratic rules’ and promised ‘thorough’ and ‘rigorous’ vigilance over the party (El Mundo, 2011r and 2011u). Nevertheless, it rejected calls from the UPN to ban Bildu for acts subsequent to its legalization. Defending this line, PSOE Minister of the Presidency, Ramón Jáuregui argued: ‘Maybe we are facing a unique opportunity to definitively overcome violence, without negotiation, without a political price, and with the triumph of democracy’ (El Mundo, 2011r). He called for encouragement of the Basque nationalist left and recent changes ‘which, probably, represent the fact that this circle has said that they don’t want ETA anymore, although they say it in a manner that we don’t like, but the fact that they are saying it appears important to me’ (ibid.). He further argued that there may be no other possible way to end ‘fifty years of terrorist tragedy’ and called for unity to tackle together this ‘delicate peace process’, converting these steps into something ‘definitive and irreversible’ (CDDS 20, 2011, p. 28). The government of the PP, traditionally the party arguing for proscription in the strongest terms, was less conciliatory in its response to further calls to illegalize Bildu. Answering a call from UPyD to ban Bildu and its sister party Amaiur, PP Minister of the Interior, Jorge Fernández Díaz stated: ‘you don’t need to convince me or the government … about what ETA is, and what Bildu is, or what Amaiur is’ and emphasized that the government’s policy was to continue using instruments available within the law to bring about the end of ETA (CDDS 10, 2012, p.  34; see also Barreda (PP), CDDS 12, 2012, pp.  27–28). Urging the UPyD not to confuse ‘what we find politically repugnant … and what is legally possible’, the PP government argued that in light of the Constitutional Court ruling, there was insufficient proof to launch successful proceedings against Bildu and Amaiur and to do so would be counter-­productive (CDDS 10, 2012, pp. 34–35 and 37). Nevertheless, the PP, along with the vast majority of deputies in the Spanish parliament rejected, by 326 votes to 5 (in a 350-seat parliament), a February 2012 UPyD motion calling for the illegalization of Bildu and Amaiur (CDDS 12, 2012. p.  40). A proposed amendment replacing the UPyD motion, supported by these parties, framed ETA’s ceasefire as a ‘victory for democracy’ and, among other things, urged the government to ‘[encourage efforts towards social coexistence] (BOCG, 2012, 45). In parliamentary debates on the motion, various party spokespersons described party bans as ‘anachronistic’ given the ‘new times’ inaugurated by ETA’s ceasefire and absence of ­political violence (CDDS 12, 2012, Jauregui (PSOE), pp.  28–29; Jané (CiU), p. 29; Esteban (PNV), p. 30). Paul Ríos, spokesperson of the peace movement Lokarri, considered the ‘decision of the Basque nationalist left to advance through solely peaceful and political methods’ was a ‘new political situation’

Herri Batasuna and its successors   83 and that legalization of Bildu and its electoral successes were an ‘opportunity to achieve peace’ (El Mundo, 2011o). And finally, it is pertinent to note that in the Spanish case, desecuritization took the form of what Hansen describes as ‘change through stabilization’, or a ‘slow move out of explicit security discourse which in turn facilitates … a more genuinely political form of engagement’ (2012, 539). It involved, not a resolution of conflict, but a receding of the issue into the background (ibid.). In this case, legalization followed the unilateral decisions of the Basque nationalist left and ETA to abandon political violence in order to meet the requirements of party and electoral laws and was not part of a deal negotiated in a formally constituted peace process. After the Constitutional Court endorsed the legality of Bildu and Sortu, some continued to call for their illegalization, including the Association of Victims of Terrorism (El Mundo, 2015), but the cycle of party bans in Spain appears to be over.

Conclusion The objective of this chapter was to examine hypotheses focusing on the role of securitization and desecuritization discourses, positions of veto players and parties’ orientation to violence in party ban and legalization decisions. In the 2003 HB and successors ban case, from which hypotheses were initially derived Bourne (2015) showed that justifications of, political support for and facilitation of ETA’s strategy of terrorist violence by the banned parties (H3) underpinned their securitization as integral to the terrorist organization and as a threat to the democratic system (H1). Veto players at both the initiation and decision stages of the party ban procedure held this view and it led to the illegalization of the parties (H2). Furthermore, empirical analysis confirmed theoretical expectations regarding party ban legalization cases. ETA’s 2011 cessation of armed struggle and Sortu and Bildu’s declarations explicitly condemning and rejecting political violence, including that of ETA (H3), paved the way for an authoritative Constitutional Court decision legalizing the parties, despite the contrary view of those initiating cases against Sortu and Bildu (H2). The Constitutional Court no longer conceived Sortu and Bildu as integral to ETA and a threat to the democratic community but as vehicles channeling the negotiation of political conflict in the public sphere (H1).

Notes 1 Parliamentary debates referred to are recorded in Diario de Sesiones del Congreso de los Diputados (CDDS) and can be found online at: www.congreso.es/portal/page/ portal/Congreso/Congreso/Publicaciones/DiaSes. 2 The terminology ‘terrorist organization’ is a shorter version of that used in the current version of the Criminal Code, ‘terrorist organizations or group’ and that prior to the 2010 reform of the Criminal Code, ‘armed bands, or organization or groups’.

4 Sinn Féin and Republican Clubs From terrorist pariah to republicanism’s pathway out of violence

In this chapter, I present findings regarding the first three hypotheses in the cases of Sinn Féin (1956) and Republican Clubs (1967), banned by the government of Northern Ireland (NI), and their legalization, respectively in 1974 and 1973, following suspension of those institutions and introduction of direct rule by Westminster in 1972. I begin with a description of the origins of Sinn Féin and Republican Clubs, their evolving relationship with the terrorist organization, the Irish Republican Army (IRA), the fundamental tenets of the republican movement to which they belong, and the parties’ electoral salience. I then turn to examine the hypotheses:

H1. Democracies ban anti-­system parties if these parties have been ‘securitized’ as an existential threat. H2. Democracies ban anti-­system parties if veto players prefer proscription. H3. Democracies ban anti-­system parties if they do not unambiguously reject violence.

I employ various documentary sources: For banning and legalization of Sinn Féin and Republican Clubs, I examined archived government files on the parties from the Public Records Office in Northern Ireland (PRONI), Belfast, and National Archives, Kew Gardens, London, parliamentary debates in NI and Westminster1 and newspaper articles predominantly from the Northern Irish newspapers, Belfast Telegraph and Irish News, and the Irish newspaper, The Irish Times. Analysis confirms that all three hypotheses for the party ban and legalization cases varied in a manner consistent with the theory. That is, in the legalization cases, it was expected that each of the conditions under which the parties – or their predecessors – were banned would no longer hold.

Sinn Féin and Republican Clubs   85

Sinn Féin and Republican Clubs Sinn Féin – in Gaelic ‘We Ourselves’ – emerged in 1905 as a broad-­based organization led by Arthur Griffith. However, the name Sinn Féin has not been held continuously by a single party; rather it has been used by successive parties differing from each other ‘in matters such as leadership, membership, aims and means of obtaining them’ (Pyne, 1969, 30; Coakley, 2003, 15–16). Pyne argues that in the first 21 years of its existence alone there were four successive parties bearing the title Sinn Féin: The ‘Monarchist’, Griffithite, Sinn Féin (1905–17), committed to the idea of a dual monarchy (separate parliaments for Britain and Ireland under a single monarchy), passive resistance and non-­violent agitation (Pyne, 1969; see also Feeney, 2003, 42); the ‘Republican’ Sinn Féin (1917–22), led by Eamon De Valera in the wake of the 1916 Easter Rising, an electoral powerhouse and ‘expression of [Irish] national self-­determination’ which paved the way to independence and the creation of the Free State (Pyne, 1969; Feeney, 2003, 91); the ‘Extremist’, or anti-­Treaty, Sinn Féin (1922–26), created after the Irish Civil War by those, including De Valera, rejecting the 1921 Anglo-­Irish peace treaty which established the Irish Free State, and who would continue to pursue a republic (Pyne, 1969; Feeney, 2003, 156–160); and Fundamentalist Sinn Féin (1926 to 1970), the electorally marginal remnants of the republican movement following De Valera’s decision to participate in Free State institutions and the founding of the new party Fianna Fáil (Pyne, 1969; Coakley, 2003, 16). Coakley argues that the fifth and sixth Sinn Féins – respectively the left-­leaning Officials and the militant, eventually dominant, Provisionals – emerged simultaneously following the split of the republican movement in 1970 over abstentionism and appropriate responses to the explosion of sectarian violence in NI in 1969 (2003, 16). The seventh, or Republican Sinn Féin, was a group splitting from the Provisionals who were dissatisfied by the electoral turn of Sinn Féin under the leadership of Gerry Adams (ibid., 16). The fortunes of the parties claiming the mantle of Sinn Féin have ebbed and flowed, as has their centrality in the republican movement and relationships with the IRA. At its highpoint in 1918, Sinn Féin became, as Feeney argues, ‘the vehicle and political expression of national self-­determination, the political movement the Irish people chose for articulating their desire for independence’ (2003, 91). After the Irish civil war, however, a period of decline and estrangement from the IRA ensued; indeed, the IRA set up alternative political fronts such as Comhairle na Poblachta (1929), Saor Éire (1931) and Cumann Poblachta na hÉireann (1936) rather than rely on Sinn Féin (Feeney, 2003, 174). Sinn Féin probably escaped proscription following the establishment of the NI State in 1921 because, despite its electoral victory in the 1918 Westminster election and inauguration of the Dáil Éireann by elected Sinn Féin representatives, ‘most of the subsequent period … had been marked by isolation and irrelevance’ for Sinn Féin (Tonge, 2006, 102). By the 1940s, Sinn Féin was ‘all but extinct’ (ibid., 103). In 1948, however, the IRA decided to take over the near moribund organization that Sinn Féin had become, not only to appropriate the political

86   Sinn Féin and Republican Clubs capital associated with the party’s historic predecessors, but to delegate to it a series of political roles in support of the military campaign. In the North, this role was to ‘act as a convenient rallying point where northern nationalists could register their numbers and disaffection when a Westminster election was called, after which the party would promptly vanish until the next general election’ (Feeney, 2003, 211). In the 1950s, Sinn Féin experienced a short-­lived revival when it won two seats and 23.6 per cent of the vote in the 1955 Westminster election (see Table 6.3). Its candidates included a number of IRA members imprisoned after a raid on the Omagh military barracks (Belfast Telegraph, 1956o; The Observer, 1956; Irish Times, 1959). Its two elected candidates were subsequently disqualified and by-­elections held. In the early 1960s, a new generation of left-­leaning leaders, notably IRA Chief of Staff Cathal Goulding and Sinn Féin President Tomás MacGiolla, sought to realign republican ideology from its sole focus on the constitutional question of the partition of Ireland towards engagement in economic and social issues from a radical socialist perspective (Feeney 2003, 214). The IRA would still have a role, but it would be that of an armed guarantor of republican social and political gains (Patterson, 1997, 106; Feeney, 2003, 225). In their reorientation towards social and economic issues, the IRA leadership sought to end the tradition of abstentionism, participate in elections and engage in civil society activism through infiltration in tenants associations, youth groups and ultimately the NI Civil Rights movement (Patterson, 1997, 119; Feeney, 2003, 219 and 226). Part of this strategy was the creation of debating societies, such as the Wolf Tone Societies, and Republican Clubs, the latter left-­wing student debating societies centred at universities both North and South (Feeney, 2003, 230–232). In NI, Republican Clubs were also established in nationalist areas and run by Sinn Féin: Just as the IRA had found it convenient to use Sinn Féin as its political mouthpiece because it was a legal political party [in the Republic of Ireland] so the republican movement used Republican Clubs in the North as its mouthpiece because Stormont had … proscribed Sinn Féin. (Feeney, 2003, 231) The new direction posed by the IRA leaderships generated much resistance from a movement imbued with anti-­communism and rank-­and-file enthusiasm for armed struggle (Patterson, 1997, 100–101, 107). In late 1969, the IRA Army Council agreed to abandon its traditional policy of abstentionism, effectively recognizing parliaments at Westminster, Dublin and Stormont (English, 2004, 105–106; Bowyer Bell, 1970, 368). The decision to abandon the long-­cherished principle of abstentionism led dissenters to withdraw and establish the rival Provisional IRA. In January 1970, supporters of the Provisionals walked out of the Sinn Féin ard fheis (party conference) called to ratify the anti-­abstentionist policy and established Provisional Sinn Féin (English, 2004, 105–106; Bowyer Bell, 1970, 367–368).

Sinn Féin and Republican Clubs   87 Tensions within the republican movement came to a head during growing civil unrest in NI in August 1969. Following a build-­up of tensions with the emergence of the NI civil rights movement, a ‘quasi civil war’ erupted in NI, involving large-­scale rioting and sectarian violence, exacerbated by heavy-­ handed and sometimes sectarian policing of the Protestant-­dominated Royal Ulster Constabulary (Bew et al., 1979, 172; English, 2004, 102–104). In August 1969, British troops were sent to NI to assist local security forces. While the Goulding republican leadership had ‘deliberately run down’ the IRA’s military capabilities, the Provisionals set themselves up as an anti-­ imperialist Catholic ‘defence’ organization, acquired new weapons and saw events in the North as a ‘once in a lifetime opportunity to overthrow Stormont by making the Six Counties ungovernable … [resulting in] British withdrawal’ (Feeney, 2003, 252, 255; English, 2004, 120–125). Behind defensive barricades set up in Catholic areas of Belfast and Derry, the Provisionals gained in strength to become the principal republican movement in the North. The Provisional IRA unleashed a wave of terrorist attacks on the British mainland and in NI and according to the Sutton Index of conflict-­related deaths in NI, was responsible for 1,705 deaths between 1969 and 2001, some 50 per cent of all conflict-­related deaths recorded in the period (Sutton, 2017). In May 1972, the ‘rump’ IRA, soon known as the Officials, declared and came to focus most of its attention on politics in the Republic of Ireland. According to the Sutton Index, the Officials were responsible for 53 deaths. In NI, the Officials initially used the party Republican Clubs for their electoral platform. Reflecting their socialist orientation, they later changed their name to The Workers Party Republican Clubs, and in 1982 to The Workers Party. In the 1970s, (Provisional) Sinn Féin, initially a support organization for the IRA, grew into a protest movement mobilizing on issues such as internment (imprisonment without trial), use of torture by the security forces and the political status of republican prisoners (Feeney 2003, 269, 272). Nevertheless, from the 1980s, under the leadership of Gerry Adams and Martin McGuiness, Sinn Féin gradually became a vehicle leading the republican movement out of violence. The party returned to electoral prominence in the early 1980s, following mobilization of Catholic sympathy for the fate of IRA prisoners on hunger strike and adoption of the ‘ballot box and armalite’ strategy (Maillot, 2005, 25). By 1985, after contesting local, NI assembly, Westminster and European levels, Sinn Féin’s share of the vote settled at around 12 per cent and Sinn Féin could claim to represent about a third of nationalist voters in NI (see Table 6.3). Republican Clubs contested elections in Westminster and various NI representative bodies from the early 1970s until today, but has not won seats in Westminster or NI assemblies (see Table 6.4). After an IRA ceasefire in 1994, interrupted but renewed in 1997, Sinn Féin participated in all-­party talks leading to the 1998 Belfast Agreement (also known as the Good Friday Agreement). It took around a decade before the re-­ establishment of envisaged devolved institutions for NI set out in the Agreement worked regularly and Sinn Féin came to play an increasingly important role in

88   Sinn Féin and Republican Clubs the institutional politics of NI. Between 1996 and 2004, its share of the vote rose to just under 20 per cent on average. By 2001, it had overtaken its main competitor the Social Democratic and Labour Party (SDLP) as the main party of NI Catholics in Westminster elections and by 2007 it had done the same for the NI Assembly. The proportional allocation system employed in NI to distribute executive posts meant that Sinn Féin was entitled to hold various ministerial positions during the first sporadic periods of the NI executive’s operation. From 2007 to 2017, former IRA leader Martin McGuinness held the post of deputy first minister. In the Republic of Ireland, Sinn Féin’s first breakthrough since the 1950s was in 1997, when it took its first seat in the Dáil Éireann, after which its vote share has risen at each election. By 2016 it was the third largest party in the Dáil Éireann, with 23 out of 158 seats.

Ban on Sinn Féin and Republican Clubs in Northern Ireland Veto players The NI state, a devolved unit of the United Kingdom with considerable powers of self-­government, was established in 1921 in accordance with the 1920 Government of Ireland Act. The Act established a two-­chamber parliament in NI, a lower chamber, the NI House of Commons, and a Senate, with the latter predominantly composed of senators elected by the lower house. The King was represented by the Lord Lieutenant until 1922 when replaced by a Governor. Among other things, the NI parliament was empowered to make laws for the ‘peace, order and good government’ of NI, which included responsibility for law and order, policing and security (Bogdanor, 1999, 70; Longford and McHardy, 1981, 79). To this end, the parliament enacted the Civil Authorities (Special Powers) Act (Northern Ireland) 1922 (henceforth SPA), which included powers to proscribe unlawful associations. Between 1922 and 1972, the Special Powers Acts (SPA) gave the ‘civil authority’, namely the Minister for Home Affairs for NI: power in respect of persons, matters and things within the jurisdiction of the Government of Northern Ireland, to take all such steps and issue all such orders as may be necessary for preserving the peace and maintaining order. (Section 1) This included the power to make Regulations establishing further provision for the ‘preservation of peace and maintenance of order’ and ‘for varying or revoking any provisions of the regulations’. As such, the initiation of decisions on the proscription and legalization of organizations, which by 1949 took the legal form of an amendment to Regulation 24A on unlawful associations, was the domain of the NI Minister of Home Affairs. In December 1956, in accordance with Order no. 199 under the SPA, the NI Minister for Home Affairs banned Sinn Féin as a threat to the maintenance of ‘peace and order’ and a ‘threat to law and

Sinn Féin and Republican Clubs   89 order and good government’ (Belfast Telegraph, 1956o; The Observer, 1956). In March 1967, the Minister banned Republican Clubs by Order no. 42 under the SPA as a threat against ‘the preservation of peace and maintenance of order’ (Belfast Gazette, 10 March 1967, p. 92). According to the SPA, regulations made under its authority had to be laid before the NI House of Commons and Senate of Northern Ireland and either house could call on the Crown’s representative in NI to annul regulations. However, the parliamentary check on the powers of the Minister for Home Affairs was more theoretical than real in a parliament dominated by a large unionist majority. Between 1921 and 1972, NI had a parliamentary system of government in which a single party, the Ulster Unionist Party (UUP), consistently held an absolute majority in parliament and formed the government (see Appendix 2). Prior to 1956, the government had resisted pressure from hardline Unionists to ban Sinn Féin. In the 1955 Westminster general election Sinn Féin had run members of the IRA as candidates, prompting Unionist backbenchers to call for measures preventing Sinn Féin from standing in NI elections (NI HC Official Report 43, 1955, cc.  1843–1869). At this time, the Stormont government rejected a ban, according to the Belfast Telegraph (1956o), ‘on the grounds that as a political party it is entitled to latitude in speeches’. By December 1956, in the opening phase of the IRA’s Border Campaign, unionists in parliament and government were united in their support for banning Sinn Féin (NI HC Official Report 40, 1956, cc. 3183–3211). From the 1960s, tensions between traditionalists and liberals within the UUP became more palpable and the fragile coalition of diverse interests within the UUP and among unionists more generally began to unravel. In 1963, liberal Unionist Terence O’Neill became NI prime minister, pursuing a modernization agenda and the stabilization of Unionist dominance through a series of minor political reforms (McAllister and Nelson, 1979, 281; Dixon, 2008, 76). Unlike previous prime ministers, O’Neill ‘no longer deferred to ultra loyalist feeling’ (McAllister and Nelson, 1979, 281, 284; Darby, 1976). In 1966, O’Neill’s government banned the paramilitary Ulster Volunteer Force (UVF ), which had openly declared war on the IRA and promised to execute known IRA members ‘mercilessly and without hesitation’ (Darby, 1976, 91). This was the first and only Unionist organization to be banned by the Stormont regime. O’Neill nevertheless appointed the hardliner, William Craig, as Minister for Home Affairs in 1966, who soon decided to ban Republican Clubs. When the vote was put in the NI House of Commons on an opposition motion to set aside the order banning Republican Clubs, Unionists rallied together and the motion was defeated 25 votes to 10 (NI HC Official Report 66, 1967). Various provisions of the Government of Ireland Act formally limited the autonomy of NI authorities (Bogdanor, 1999, 72; Cunningham, 2001, 1). The Act prohibited laws and executive acts interfering with religious equality and included a formal statement that the Westminster parliament remained supreme. There was provision for the Lord Lieutenant, and later the Governor, to withhold approval of legislation (on the advice of the British government). The House of

90   Sinn Féin and Republican Clubs Lords was designated the final court of appeal and the Judicial Committee of the Privy Council as referent body on constitutional issues. Until March 1972, when the powers of the NI institutions were suspended, the legal right of British authorities to interfere in NI politics was rarely used (Bogdanor, 1999, 74–76; Longford and McHardy, 1981, 81). Furthermore, a convention emerged whereby Westminister would not legislate for NI or discuss devolved matters for which British ministers could not take responsibility. As such, while formally in a position subordinate to British authorities, devolved institutions in NI enjoyed a high degree of autonomy in practice. On proscription of Sinn Féin and Republican Clubs, the British government abided by this convention of non-­interference in the affairs of NI (Cunningham, 2001, 2–5), in the case of the Republican Clubs ban, despite a high profile mobilization of NI civil rights activists, Stormont opposition parties and backbench Labour and Liberal MPs (see below). A case dealing with a ban on the Republican Clubs was referred to the House of Lords, but did not ultimately address the question of whether or not it was appropriate to ban the party. In sum, procedures in the United Kingdom devolved primary responsibility for banning parties to authorities in NI and in accordance with the SPA, the NI Minister for Home Affairs and the NI House of Commons and Senate. Formal limits on the powers of NI authorities were constrained by a British constitutional convention favouring non-­interference in matters of NI competence. The dominance of the UUP in the NI political system – particularly its uncontested control over government and parliamentary majority – effectively meant that, until direct rule from Westminster was reintroduced in 1972, the party ban decisions effectively took place in a single veto-­player system. Sinn Féin, Republican Clubs and the republican ‘physical force’ tradition Historically, the legitimacy of the use force to oppose British rule in Ireland has been a fundamental tenet of republicanism, while those employing violence to this end were generally regarded as ‘patriots and freedom fighter, not criminals’ (Feeney, 2003, 277). Armed struggle has been an ‘integral part of the culture of the [republican] movement’, a strategy ‘inherently linked to the movements’ identity’ and part of the ‘cement’ linking the various ranks of the movement (Maillot 2005, 5). When Sinn Féin and Republican Clubs were banned, the role of parties in the republican movement was that of adjunct – or ‘poor cousin’ (ibid., 5) – to the military organization, or as Feeney put it, a ‘rubber stamp for the IRA’ (2003, 211; see also Patterson, 1997, 99, 119). The parties’ close ties to the IRA and its support for acts of political violence facilitated securitization of the parties as an integral part of a terrorist organization and threat to the integrity of the Northern Ireland state. The 1956 ban on Sinn Féin took place within days of the onset of the IRA’s Border Campaign, which ‘broke the relative calm in which Northern Ireland had existed for almost thirty years’, and amounted to the ‘largest concerted attack on

Sinn Féin and Republican Clubs   91 the fabric of the state’ since its inception in 1921 (Flynn, 2009, 49). A series of armed raids in NI and England and the training of IRA volunteers preceded the raids (English, 2004, 72–73; Coogan, 2000, 302). Also known as ‘Operation Harvest’, the campaign sought to initiate a guerrilla-­style war of attrition against British authorities in NI (Flynn, 2009, 49). The IRA sought to establish ‘liberated zones’ within the nationalist border regions of NI, build up popular support and undermine the ‘very fabric’ of NI, through attacks crippling the ability of the state to function effectively, in the hope that this would ‘sap the will’ of the British to retain troops and ultimately political authority over NI (ibid., 49). Local IRA units, supported by volunteers from the Republic of Ireland, attacked air and radar installations, military barracks, police stations, courthouses, communications and customs posts in NI (English, 2004, 73; Coogan, 2000, 298–329; Patterson, 2006, 12). The Border Campaign ended in 1962, at an expense of 17 lives and millions of pounds in property damage and security costs (Coogan, 2000, 303) without obtaining its objectives of forcing Britain to reopen the issue of partition or stirring up support from Irish nationalists north and south (English, 2004, 75; Coogan, 2000, 329; Patterson, 2006, 12). Sinn Féin’s support for the Border Campaign and use of political violence to achieve republican goals was acknowledged by the party itself, in the press, commented upon in parliament and recorded in NI government files. On 12 December 1956, Sinn Féin headquarters in Dublin issued a statement ‘on the uprising in the North’, a statement which the party demanded be read twice after every mass on Sunday 16 December 1956, ‘so all the parishioners hear it’ (Sinn Féin, 1956). The statement read: Irish men have again risen in armed struggle against British Aggression in Ireland. The Sinn Féin Organization states to the Irish people that they are proud of the risen nation and appeal to the people of Ireland to assist in every way they can the soldiers of the IRA. … Constitutional methods alone against armed occupation, civil injustice and victimization, could not possibly be made effective. Only when Ireland is completely free and independent will Ireland be at peace. (Sinn Féin, 1956) In addition, Sinn Féin reportedly organized a ‘national collection’ outside churches in the Republic for ‘gunmen’, described those ‘gunmen’ as ‘freedom fighters’, invited a soldier involved in the raids in NI to a Dublin meeting, and cheered news of further attacks on British installations in NI (Belfast Telegraph, 1956f; Sinn Féin, 1956). In the early 1960s, a new IRA leadership sought, albeit with much rank-­andfile resistance, to transform the republican movement into a revolutionary socialist force. A new focus on social and economic issues and political activity would effectively involve the demilitarization of the republican movement, where arms would be used only against the police and property and to defend protestors

92   Sinn Féin and Republican Clubs (Feeney, 2003, 225; Patterson, 1997, 101 and 105). In other words, the IRA was to be the vanguard of this revolutionary movement, rather than leader of an armed insurgency aiming to force the British out of the North (Patterson, 1997, 106; Feeney, 2003, 225). The IRA chief of staff, Cathal Goulding, announced that the IRA had no immediate military plans and the IRA’s military role was scaled-­down (Patterson, 2002, 109). When banned in 1967, the Northern Directorate of Republican Clubs described themselves as ‘progressive’ and ‘democratic organisations’ (Irish News, 1967b and 1967c). Nevertheless, planned acts of violence continued, even if they were regarded as ‘compensatory’ (Patterson, 1997, 107) or ‘one-­off gestures’ (Feeney, 2003, 241). In addition to military actions in the Republic of Ireland, the IRA bombed two Territorial Army centres in Belfast and Lisburn in May 1967 (around two months after the ban on Republican Clubs), followed by an attack on Territorial Army Headquarters in January 1968 (Patterson, 1997, 109). There were also signs of increasing social unrest. During the 1964 Westminister elections, an Irish flag displayed outside the office of the Republican Party candidate for West Belfast provoked a serious riot when the flag was forcibly removed by the RUC (Feeney, 2003, 235; Darby, 1976, 15). A series of significant republican anniversaries heightened tensions, including (banned) commemorations of the fiftieth anniversary of the Easter Rising against British rule in the Republic of Ireland in 1916, and became an issue of concern for the police and the NI government (Feeney, 2003, 238; Dixon, 2008, 69). It was also in this context that the loyalist paramilitary organization, the Ulster Volunteer Force (UVF ), emerged (Darby, 1976, 91). Direct support for armed insurrection in the case of Sinn Féin, and ambiguity about the appropriateness of political violence in the case of Republican Clubs facilitated two modes of securitization underpinning party ban justifications. Securitization of Sinn Féin as the political wing of a terrorist organization There are some striking similarities in modes of securitization employed in Spanish and NI party ban cases. As in the Spanish case, pro-­ban supporters in NI justified their positions with reference to the close collaboration and ‘identity of purpose’ between the political parties – i.e. Sinn Féin and Republican Clubs – and a terrorist group – i.e. the IRA. In both cases, proscription of both parties was framed as a vital measure to deal with the military security threat posed by ETA and the IRA. The most consequential representation of the parties as a security threat was the ‘open secret’ that Sinn Féin and Republican Clubs were the ‘political wing’ or ‘front organisation’ of the IRA (NI Ministry Home Affairs, 1958, 1959 and 1967; RUC, 1958; Belfast Telegraph, 1956o; Belfast Newsletter, 1956; Manchester Guardian, 1957).2 Sinn Féin’s open support for the IRA’s armed struggle was a central part of the rationale justifying pro-­ban arguments before the IRA campaign (NI HC Official Report 66, 1967, Craig (NI Minister of Home Affairs), cc. 149–150); RUC Inspector General, 1958; NI HC Official Report 43,

Sinn Féin and Republican Clubs   93 1955, Minford (UUP), cc.  1849–1850, Porter (Independent Unionists) cc. 1959–1962). When the Border Campaign began, Sinn Féin voters in NI were feared as a reservoir of supporters for the IRA’, so that when ‘Men … come from Cork, Kerry and other parts of Southern Ireland … they know that when they cross the Border there are people in this Province who are willing to house them and give information which helps them in their efforts to try to overthrow our position.’ (NI HC Official Report 40, 1956, cc. 3198; see also Irish Times, 1954 and NI HC Official Report 41, 1957, Morgen (UUP), cc. 1333–1334) Many suspected that Sinn Féin and Republican Clubs recruited for the IRA (NI Ministry of Home Affairs, 1967; The Observer, 1956; Belfast Telegraph, 1956o; Irish Times, 1957). Restating earlier rationales for banning Sinn Féin when justifying the Republican Clubs ban, Craig argued that the IRA and Sinn Féin had an ‘identity of purpose’ and emphasized Sinn Féin’s support for the IRA’s 1956 Border Campaign (NI HC Official Report 66, 16, cc.  149–150; Irish News, 1967a). Craig argued Republican Clubs were established and run by the Dublin-­based Sinn Féin for the purpose of circumventing the NI Government’s ban on Sinn Féin (NI HC Official Report 66, 1967, cc. 149–150; Irish Times, 1967a; Irish News, 1967a). Republican Clubs operated in the same way as Sinn Féin and returned delegates to Sinn Féin’s annual general meeting, he argued, as was their political goal for the ‘complete overthrow of British rule in Northern Ireland’ (ibid., Irish Times, 1967d; Belfast Telegraph, 1967f ). Craig repeatedly stated that the order banning Republican Clubs sought to ensure the law relating to the IRA and its political wing were observed (in Irish Times, 1967d; Belfast Telegraph, 1967f ). Both the Royal Ulster Constabulary and the NI government regarded the Republican Clubs a direct successor to Sinn Féin. In 1967, the Royal Ulster Constabulary reported to the NI Ministry of Home Affairs that: It was the intention of the I.R.A to use these Clubs as a front organisation in the same way that they used Sinn Féin Cummans (Clubs) for their militant build up prior to the opening of the campaign of violence in 1956. By directing its members to join Republican Clubs, the I.R.A had hoped to provide a solid political wing in Northern Ireland without which, they believe, militant action will not succeed. (RUC, 1967b) Drawing on secret RUC intelligence reports which strongly recommended proscription, NI Minister for Home Affairs William Craig stated categorically that ‘there is not the slightest doubt that the Republican Club Movement is the Sinn Féin organization in another guise and has connections with and supports the I.R.A’ (NI Ministry of Home Affairs, 1967; NI HC Official Report 66, 1967 cc. 149–150; Irish News, 1967a; Irish Times, 1967a).

94   Sinn Féin and Republican Clubs Moreover, Craig – and others – framed both Sinn Féin and the Republican Clubs as an integral part of the military threat posed by the IRA. Craig argued the sole purpose of the bans was to contain and suppress IRA efforts to overthrow the NI state by force (NI HC Official Report 66, 1967, Craig, c. 182; Irish Times, 1967a; Irish News, 1967a). The relationship between both Sinn Féin and Republican Clubs to the IRA was evident, according to the NI government, in the fact that around half the members of Republican Clubs were IRA members (NI HC Official Report 66, 1967, cc. 151–152; Irish Times, 1967a; Belfast Telegraph, 1956b). The parties were described as essentially a ‘front organisation of the I.R.A’ whose objective was ‘recruiting or gaining support for the I.R.A’ and to put out propaganda ((Belfast Telegraph, 1956b and 1967d; Irish Times, 1967b, 1967c and 1967d). Craig argued Sinn Féin ought not be considered ‘a political organisation in the ordinary sense’ because it ‘was more concerned with the bullet than the ballot box’ (NI HC Official Report 66, 1967, cc.  149–150; Irish News, 1967a). Securitization of Sinn Féin and Republican Clubs as a threat to the Northern Ireland state An important contrast between the Spanish and NI ban cases, however, relates to the second mode of securitization employed in the Sinn Féin and Republican Clubs ban cases. Here the UUP government predominantly framed the referent object – or object of threat – as the existence of the NI state itself, rather than democratic institutions and rights of citizens, which was the case for Herri Batasuna and its successors. As Todd (1987) has argued, unionist political culture tended to regarded Catholics, both the minority in NI and the majority-­Catholic Republic of Ireland, with at best suspicion, and at worst hostility, not least in light of the Catholic minorities’ near-­unanimous support for a united Ireland. Each of the two main strands of unionist political culture saw republicanism as inimical, albeit in different ways (ibid.; see also Donohue, 2001, xxv). On the one hand, Ulster loyalists, whom Todd described as ‘probably the most numerous and certainly the most vociferous ideological grouping of Northern Protestants’, were grounded in an evangelical fundamentalism (Todd, 1987, 7). They tended to see republicans as agents of a Roman Catholic Church conceived as ‘an international source of political treason and tyranny as well as a barrier to the true faith’ (ibid.). ‘Any Catholic or Nationalist self-­assertion’, Todd points out, was ‘seen as a stain on the Protestant ethos of the society and an opening for bloody Republican rebellion’ (ibid.). On the other hand, ‘Ulster British’ ideology, associated with liberal unionists and people who tended to distance themselves from the Unionist party until the premiership of liberal unionist Terence O’Neill in 1963, was defined by a ‘patriotic love for British ideals and institutions’ and expressed ‘nothing of the loyalist sense of threat and security’ (ibid., 14 and 11). Nevertheless, nationalists were often seen as ‘potential traitors’, people whose ‘refusal to recognize the

Sinn Féin and Republican Clubs   95 legitimacy of the British state in NI or to participate in its rituals is felt to be insulting at least and traitorous at worst’ (ibid., 19). Republicans were often considered ‘absolutely evil’ and a threat to cherished moral principles of its self-­ understanding as progressive, liberal, democratic and internationalist (ibid., 13 and 18). From its origins, the UUP integrated diverse interests and social classes, which were held together by agreement on the constitutional status quo and the need to maintain Protestant dominance over the nationalist minority (Darby, 1976, 85; McAllister and Nelson, 1979, 282). The party was seen by many unionists as the ‘protector of Protestant security and freedom’ (McAllister and Nelson, 1979, 282) such that for at least the first 40 years of the NI State, ‘virtually any means of control – job discrimination, gerrymandering etc – could be justified’ on these grounds (Darby, 1976, 85; see also Dixon, 2008, 59). Similarly, a ‘simple equation of Catholics as rebels justified their exclusion from any share of power, politics or otherwise’ (Darby, 1976, 85). Oftentimes, the Unionist party sought to neutralize challenges posed by the ‘rigid opposition to Catholic power’ of Ulster loyalists through (at least verbal) moves towards more extreme positions on Catholics’ (McAllister and Nelson, 1979, 283). These themes of republican threat in unionist political culture were evident in many unionist representations of Sinn Féin and Republican Clubs in government statements, the unionist press and parliamentary debates on party bans. NI Prime Minister Lord Brookeborough argued: ‘The real nature of the I.R.A movement is that it is conducting unprovoked, armed aggression against us from a neighbouring state’ (NI HC Official Report 41, 1957, c. 30, see also Topping (UUP), cc. 3189–3190, Minford (UUP), cc. 3185–3186, McCleery (UUP), cc. 3194–3195; Belfast Telegraph, 1956p), The suspicion that the Irish government was ‘secretly in sympathy with what [the I.R.A and the Sinn Féin organization] are doing and wish them every success’ was also raised both publically and privately in meetings between the NI Minister and the Home Office in Whitehall (NI HC Official Report 40, 1956, McCleery (UUP), c. 3193, Porter (Independent Unionists), c. 3195, Faulkner (UUP), c. 3207; Belfast Telegraph, 1956m and 1956p; NI Ministry Home Affairs, 1956b). During its opening stages, the IRA’s Border Campaign was widely regarded as both a threat to life and property of civilians in NI, which could trigger sectarian violence and civil war (NI HC Official Report 40, 1956, cc. 3183–3211; NI HC Official Report 41, 1957, cc. 7–34; HC Deb, 19 December 1956, vol 562, c. 1265; Belfast Telegraph, 1956a, 1956b, 1956c, 1956d, 1956e, 1956h and 1956k; NI Ministry Home Affairs, 1956a). Fear of sectarian violence was such that the NI government and both the Catholic Church and the Orange Order – an organization established to defend the civil and religious liberties of Protestants and integral to both the UUP and Protestant Ulster’s social system (Darby, 1976, 82–85) – called for citizens to exercise restraint (NI HC Official Report 40, 1956, cc. 3183–3211; Belfast Telegraph, 1956j, 1956l and 1956n). More generally, unionists described republicans as ‘evil’, ‘unchristian’, ‘with the brand of Cain on their forehead’, as well as ‘terrorists’, ‘gangsters’, ‘idiots’, ‘criminal’

96   Sinn Féin and Republican Clubs and ‘reckless’ perpetrators of ‘disgusting’, ‘abhorrent’, ‘dire’ and ‘brutal’ ‘outrages’ (Belfast Telegraph, 1956c and 1956d; NI HC Official Report, 18 December, 40, 1956, cc. 3183–3211 and NI HC Official Report 41, 1957, cc. 7–34). Unionists not only invoked themes of republican threats when political violence and social tensions were intense. In 1955 parliamentary debates unionists proposing a Sinn Féin ban refer to the party as a ‘menace’ threatening the ‘future peace of Northern Ireland’; ‘strangers’ and ‘foreigners’ who ‘come into a country and destroy the property and homes of its people’; and as a threat to the constitutional position of NI and its democratic values and liberties (NI HC Official Report 43, 1955, Minford (UUP), cc.  1843–1853, Porter (Independent Unionists), cc. 1859–1864, Thomas Lyons, cc. 1854). When banning Republican Clubs, NI Minister for Home Affairs William Craig made it clear ‘there is no immediate emergency as far as the I.R.A is concerned’ (Belfast Telegraph, 1967b and 1967d; Irish Times, 1967c). Rather, the ban was ‘preventive’, or ‘imposed to make sure that one does not arise – that the IRA is not able to build up its strength, and be in a position to launch an attack’ (Belfast Telegraph, 1967b; Irish Times, 1967a, 1967e, 1967f ). Craig argued there was ‘no alternative’ to using exceptional powers to ban parties as long as political groups wanted to use the ‘gun and violence’ (Irish Times, 1967a). In 1967, Craig summarized threats addressed in the ban of Republican Clubs as follows: ‘We are dealing with a threat to the peace of this community from an organization that has, from the very inception of our state, destroyed life and property in Northern Ireland’ (Irish Times, 1967f and 1967a; NI HC Official Report 66, 1967, cc. 143–202). Similarly, a January 1967 Royal Ulster Constabulary report noted: The membership of some clubs may not be very large at the moment, but it should be remembered that now they have been formed they provide a hard core to which Republicans will look for guidance in normal times and to which they will rally in the event of a worsening of community relations in Northern Ireland. (RUC, 1967c) Unionists additionally saw the Republican Clubs ban as necessary to ‘safeguard the constitutional position’ (Belfast Telegraph, 1967c), and to deal with an organization aiming to ‘overthrow the Northern Ireland government’ (Irish Times, 1967a, 1967d; Irish News, 1967a), or the ‘overthrow of British rule in Northern Ireland’ (NI HC Official Report 66, 1967, cc.  149–150; Irish Times, 1967d; Belfast Telegraph, 1967f; Ulster Constitutional Defence Committee, 1967; Imperial Grand Black Chapter of the British Commonwealth, 1968; Loyal Orange Lodge Institution, 1968). While few challenged the ban on Sinn Féin in 1956, there was considerable mobilization challenging securitization of Republican Clubs involving NI opposition parties, Westminster Labour and Liberal backbenchers and NI civil society groups. In NI, some opposition MPs argued that Sinn Féin had given up on violence (NI HC Official Report 66, 1967, O’ Reilly (Nationalist),

Sinn Féin and Republican Clubs   97 cc.  195–196; Belfast Telegraph, 1967d; Irish Times, 1967a, 1967c). Others argued Republican Clubs were a legitimate political organization, or argued they ‘no longer constituted a threat to anyone’ (NI HC Official Report 66, 1967, Fitt (Republican Labour), cc. 155–156, Diamond (Nationalist), cc. 157–158), Currie (Nationalist), cc. 145–146; Irish News, 1967a, 1967d; Belfast Telegraph, 1967d; Irish Times, 1967c, 1967e). Many opponents argued that the ban was made ‘without solid reasons’ and that there was ‘no evidence’ that Republican Clubs constituted a security threat, claims linked to suspicions raised by Craig’s refusal to release a RUC report purportedly detailing that threat (Irish Times, 1967a, 1967f; Irish News, 1967a, 1967d; Belfast Telegraph, 1967e, 1967g and 1967h; HC Deb 743, 1967, 1918 and HC Deb 750, 1967, Rose (Labour), cc. 1555–1556). In addition, opponents of the Republican Clubs ban characterized proscription as an unacceptable incursion into the democratic rights of free speech and association, regardless of whether they agreed with republican views, and saw it as a sign of broader deficits of democratic governance in NI (NI HC Official Report 43, 1955, Diamond (Republican Labour), cc. 1857 and Morgan (Irish Labour), cc.  1847–1848; Irish Times, 1967a; Belfast Telegraph, 1967e; Irish News, 1967b, 1967c and 1967d). A recurrent theme in critiques of the ban was that it sought to heal rifts within the UUP and to appease loyalist extremists, rather than deal with a real threat to peace and order (Irish Times, 1967a and 1967e; Irish News, 1967a and 1967b; Belfast Telegraph, 1967a). It is notable that the Republican Clubs ban, in particular, provoked some disquiet in Unionist circles. For example, while supporting the government’s decision on banning Republican Clubs, veteran Unionist MP Edmond Warnock argued the IRA no longer constituted a serious threat and that the SPA may no longer be necessary or desirable, should be put in ‘cold storage’ except in real emergencies, and that it would be better to focus on improving inter-­community relations (Belfast Telegraph, 1967f; Irish Times, 1967d; Irish News, 1967e). Similarly, The Belfast Telegraph questioned whether it was appropriate to ‘invoke all the weight’ of the SPA, which ‘should surely be reserved for occasions of emergency when the normal processes of law are inadequate to the situation’ (quoted in Irish Times, 1967b). After the ban on Republican Clubs, a protest meeting, organized by the ‘Northern Directorate of Republican Clubs’, but also attended by Gerry Fitt and Harry Diamond (then Republican Labour MPs) and members of NICRA (Northern Ireland Civil Rights Association), the Belfast Trades Union Council and observed by the London-­based National Council for Civil Liberties, passed resolutions demanding withdrawal of the ban (Irish News, 1967c; Belfast Telegraph, 1967a and 1967e; see also RUC, 1967a). Following the arrest of republicans attending this meeting, NICRA sent a protest telegram to the NI Minister for Home Affairs, UN Secretary General U Thant and the British government (Irish News, 1967c). At Westminister, some 80 Liberal and Labour MPs supported an anti-­ban motion tabled by Gerry Fitt, then Republican Labour MP for West Belfast (Belfast Telegraph, 1967g and 1967h; HC Deb 743, 1967, cc.  1918). Additionally, Fitt and Labour Party MPs tabled parliamentary questions

98   Sinn Féin and Republican Clubs suggesting the British government use its constitutional powers to reign in the NI Minister of Home Affairs (Belfast Telegraph, 1967h; HC Deb 774, 1967, cc. 20–21). When Sinn Féin was banned in 1956 – in the midst of the IRA’s 1956 Border Campaign – British prime minister, Anthony Eden, fully supported NI government’s response, reaffirming that NI was an integral part of the United Kingdom and declaring a willingness to defend NI as it would any part of the United Kingdom (Flynn, 2009, 80–81, 87; HC Deb 562, 1956, cc. 1265; Belfast Telegraph, 1956h and 1956i). British armed forces were mobilized to assist local police and security forces combatting the IRA and the British government initiated diplomatic exchanges with the Irish government over its concerns that the IRA was operating from bases in the Republic (Belfast Telegraph, 1956g and 1956h; Irish News, 1956). In the case of Republican Clubs, Harold Wilson’s first Labour government was no less equivocal, despite mobilization against the ban. Harold Wilson was widely perceived to be sympathetic to Irish nationalism and had responded positively to appeals from NI civil rights activists regarding discrimination in housing and jobs (Patterson, 2006, 195; Cunningham, 2001, 4; Feeney, 2003, 233). Wilson acknowledged ‘widespread concern’ about events in NI and ‘acute concern about … the functioning of democracy over there’ (HC Deb 745, 1967, cc. 1823; see also HC Deb 774, 1967, cc. 20–21; HC Deb 751, 1967, cc. 1686). Nevertheless he confirmed continued adherence to the convention of non-­ interference in matters of NI competence, a preference for bilateral discussions between British and NI governments on such matters and signalled ‘admiration for the courageous stand [NI prime minister] Captain O’Neill has taken on certain issues and at certain times’. Soon afterwards, Secretary of State for the Home Department, Roy Jenkins, restated the terms of the convention: we cannot simply put aside the constitution of Northern Ireland or ignore the historical facts which underlie the present position and present policy. … Under the Northern Ireland constitution, certain powers and responsibilities are vested in the Parliament and Government of Northern Ireland. Successive Governments here have refused to take steps which would inevitably cut away not only the authority of the Northern Ireland Government but also the constitution of the province. (HC Deb 751, 1967, cc. 1685–1686) And finally, a case concerning the Republican Clubs was heard in the House of Lords. In the first prosecution under the 1967 amendment to the Regulation on unlawful associations, Mr McEldowney was charged with membership of Slaughtneil Republican Club (Sills, 1970; Finn, 1990, 57). The case was initially dismissed in a local Magistrates court because there was no evidence that McEldowney or the Slaughtneil Republican Club was ‘at any time a threat to peace law and order’ and ‘insofar as the Police were aware there was nothing seditious in its pursuits or those of its members’ (McEldowney v Forde, HL 18 Jun 1969). The case was then heard

Sinn Féin and Republican Clubs   99 in the NI Court of Appeals, which reversed the decision arguing that only the NI Minister of Home Affairs was entitled to make that judgement (Finn, 1991, 57). At this point, the case was referred to the House of Lords. While previous amendments to the Regulation named specific organizations (such as the IRA or Sinn Féin) as unlawful, the 1967 amendment made unlawful ‘organizations at the date of this Regulation or at any time thereafter describing themselves as ‘Republican Clubs’ or any like organization howsoever described’. This meant the very fact that a person was a member of an association with the name Republican Clubs was sufficient for prosecution, even when authorities acknowledged, as in the case of the Slaughtneil Republican Club, there was no threat to peace or seditious activity. Dissenting judges Lords Pearce and Diplock argued the 1967 amendment was therefore beyond the permissible scope of the SPA, which only permitted proscription to preserve peace and maintain public order (Lords Pearce and Diplock, McEldowney v Forde, HL 18 June 1969). The majority argued, however, that there was no proof the minister acted in bad faith when drafting the Regulation and, therefore, the Court could not challenge the minister’s discretion. Ultimately, the case resolved around a point of law – the discretion of ministers exercising delegated powers. The Lords did not question the issue of whether Republican Clubs should have been banned, nor as Peter Sills remarked, ‘the real issue of the complaint … oppressive police action … arbitrary or capricious use of discretionary power’ (1970, 330).

Legalization of Republican Clubs and Sinn Féin from Westminster Veto players When direct rule from Westminster was reintroduced in March 1972, the NI (Temporary Provisions) Act of 1972 gave the powers of the NI Minister of Home Affairs to the UK Secretary of State for NI, who in 1973 legalized Republican Clubs under an urgency procedure. According to this Act, the Secretary of State could make regulations under the SPA once a draft of the regulations had been approved by a resolution of each House of Parliament or in case of ‘urgency’, regulations could be made without prior parliamentary approval but would cease to have effect after 40 days if not approved by resolution of each House (Schedule, Section 4(3) and (4)). The NI (Emergency Provisions) Acts (EPAs) also made the executive the primary actor in proscription decisions, albeit with parliamentary oversight. More specifically, the EPAs authorized the Secretary of State to add by Order to the list in Schedule 2 (proscribed organizations) ‘any organization that appears to him to be concerned in terrorism or in promoting or encouraging it’ or remove an organization from the Schedule (EPA, Section 19(4) and (5)). Similarly, the Prevention of Terrorism Acts (PTAs) empowered the Secretary of State to add by Order an organization that appeared to be ‘concerned in, or in promoting and encouraging terrorism occurring in the UK and connected with the affairs of Northern Ireland’ to its Schedule

100   Sinn Féin and Republican Clubs 1 as well as remove organizations from it (Section 1(3) and (4). Like the SPA, both the EPAs and the PTAs required that, in normal circumstances, orders were to be approved by a resolution of each House of Parliament prior to enforcement, but in urgent situations, the Secretary of State could issue a temporary Order for proscription of 40 days’ duration. The United Kingdom has a parliamentary system of government in which a single party has controlled a majority in parliament and governed for most of the post-­war period. However, during the early 1970s, when the UK assumed direct rule in NI – and thus responsibility for major decisions on party bans – British government majorities were either absent or rather shaky. In this context, NI MPs, and particularly the UUP, obtained greater influence on government decision-­making. After the February 1974 general election no single party won a majority of seats. Relations between the Conservatives and UUP had traditionally been so close that members of the latter took the Conservative Party whip, ‘in return for which Conservative Central Office provided administrative, electoral and occasionally financial help, as well as a pledge not to stand their own candidates in Northern Irish constituencies’ (Smith, 2006, 74). The suspension of Stormont by Heath’s Conservative government, seen as an act of betrayal by Ulster unionists, led to a fundamental breakdown in the relationship, however (ibid., 71). Following the indecisive February 1974 elections, and unpromising discussions with the Liberals, Heath had offered the Conservative whip to the seven UUP members of the United Ulster Unionist Council (an electoral pact between the UUP, the Democratic Unionist Party and the Vanguard Unionist Progressive Party). As Bogdanor explains, however, the United Ulster Unionist Council response made collaboration in government politically unviable because it required the whip be offered to all 11 of their MPs, including Reverend Ian Paisley, a staunch opponent of proposals for an NI power-­sharing executive, at the time the flagship of the Conservative government’s NI policy (2008, 18). It subsequently led to Heath’s resignation as prime minister (Smith, 2006, 71; Maer and Kelly, 2017, 9; Bogdanor, 2008). Labour’s Harold Wilson then formed a minority government, facilitated by the absence of a formal investiture vote and by the willingness of the Conservatives to abstain on key votes in order to avoid an immediate second election (Bogdanor, 2008, 19; Brazier, 2008, 32; Maer and Kelly, 2017, 9). A second election in October 1974 gave the Labour party a small majority of seats permitting Wilson to form a majority government. By April 1976, however, the majority of the Labour government of James Callaghan, who had succeeded Wilson, had dwindled below a formal majority due to a series of by-­election defeats and defections. By March 1977, the government needed to secure support from the Liberals and the Ulster unionists to stay in power (Maer and Kelly, 2017, 9). In their pact with the Labour party, the Liberals agreed to support the government for a time in any confidence vote, but did not join the government (Bogdanor, 2008, 18). In exchange for supporting the government, the UUP, who ‘played a careful tactical parliamentary game, attempting to extract as much as they could from the government’ (Walker, 2004, 223), won a concession to an

Sinn Féin and Republican Clubs   101 increase in the number of NI MPs returned to Westminister (Walker and Mulvenna, 2015, 251; Norton, 2008, 109). Indeed, Walker and Mulvenna argue that the UUP’s ‘tactical abstentions and occasional support for the government had done much to prolong its life’ (2015, 253). In this context, Dixon argues, ‘both the Labour government and Conservative opposition moved their policy towards a more unionist stance in order to win the support of the Ulster unionist MPs at Westminster at a time when the House of Commons was finely balanced’ (2001, 346; see also Cunningham, 2001, 20). The Labour government was eventually defeated in a vote of confidence in March 1979 by one vote, including that of normally supportive NI MP Gerry Fitt from the Social Democratic and Labour Party (SDLP) (Walker and Mulvenna, 2015, 253; Newsinger, 1991, 56). Fitt voted against the government in protest at the apparent step towards the integration of NI politically into the British system that the increase in NI MPs appeared to signify and a government report on ill-­treatment of terrorist suspects (Cunningham, 2001, 20; Newsinger, 1991, 56). The Conservative Party under Margaret Thatcher won a new majority at the May 1979 general election. Despite this relatively rare moment of fluidity in the formation of governmental and legislative coalitions, bargaining on the fundamentals of British NI policy in general, and party ban decisions more specifically, was constrained by a bipartisan approach among the leading British parties. The Labour party have traditionally been more ‘pro-­nationalist’ and Conservatives more ‘pro-­Unionist’, and there have sometimes been differences in concrete proposals (Dixon, 1995, 2001; Cunningham, 2001). Nevertheless, as Bew et al. argue, the essence of the bipartisan approach, in place since 1922 but only fully disclosed in times of crisis, consisted of a preference to ‘minimize British involvement, politically and militarily’, with action ‘taken only where the alternative appeared ultimately to entail a greater degree of involvement’ (1979, 185). By the 1970s, Dixon argues that the substance of bipartisanship was both a general agreement between the Labour and Conservative parties on the basic principle that consent of people in NI was needed to change the constitutional status of NI, as well as agreement in other areas of constitutional, security and economic policy (Dixon, 1995, 148). This included a commitment to retaining troops in NI while the security situation demanded it, power-­sharing arrangements permitting the participation of nationalists and unionists in a NI government and ‘a rejection, by both major parties when in government, of both integration and support for the promotion of Irish unity’ (Dixon, 2001, 346–348). In relation to the legalization of Republican Clubs and Sinn Féin, the bipartisan approach held in the face of unionist pressure to retain the ban. In May 1973, the Conservative government legalized Republican Clubs using urgency procedures in order that the party could participate in forthcoming elections. The Order was subsequently endorsed in both Houses of the Westminster Parliament (HC Deb 856, 1973, cc. 1147–1177; HL Deb 342, 1973, cc. 740–742). In the House of Commons vote, Unionist politicians spoke out against legalization and four of the five MPs voting against it (excluding two UUP Tellers) were Unionists. Labour MPs and all but one Conservative MP present dominated the 87 yes votes with the Labour Party

102   Sinn Féin and Republican Clubs only challenging the Order insofar as it did not also legalize Sinn Féin. When coming to power, the Labour Secretary of State for NI, Merlyn Rees was able to announce his decision to legalize Sinn Féin in April 1974 as part of a statement of the new government’s NI policy. This policy was approved in the House of Commons and the House of Lords (HC Deb 871, 1974, cc. 1463–1584; HL Deb 351, 1974, cc. 1092–1095). In the House of Commons, only Unionists of varying hues – UUP, Democratic Unionist Party and Vanguard Progressive Unionist Party – and one Conservative MP voted against Rees’ NI policy, while 190 MPs from remaining parties supported it. According to a Background Note prepared by British civil servants, by February 1974 both Conservative and Labour parties had agreed on legalizing Sinn Féin, having consulted with leaders of the NI parties including the UUP’s Brian Faulkner and SDLP’s John Hume (British Government, 1977). The ‘Troubles’, the Provisionals and the Officials The legalization of Republican Clubs and Sinn Féin took place in the opening stages of a bloody 30-year conflict in NI that would cause over 3,600 deaths and many injuries (see Figures 4.1–4.3). Conflict was sparked by the emergence of the predominantly Catholic civil rights movement in the late 1960s demanding reforms to end decades of discrimination in housing, jobs, policing and the political system itself. In 1968, civil rights activists took politics to the streets organizing protest marches around NI, which increasingly ended in violent clashes with the police. Fearing a nationalist agenda behind the movement, loyalists organized counterdemonstrations and sometimes armed confrontations with civil rights demonstrators (Bowyer Bell, 1970, 355–356; Dixon, 2008, 78–83; Darby, 1976, 21). In August 1969, a unionist Apprentice Boys parade in Derry ended in riots and erection of barricades in Catholic areas of Bogside. Disorder soon spread to other parts of NI, developing into intense sectarian violence, forcing many families, most of whom were Catholics, from their homes (Darby, 1976, 22; Feeney, 2003, 255). On 14 August 1969, the British Army was sent to 600

Deaths

500 400 300 200 100 0

1969 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 Year

Figure 4.1 Number of deaths in Northern Ireland, Great Britain and Ireland due to conflict in NI. Source: Sutton Index of Death, Conflict in Northern Ireland (CAIN) database, http://cain.ulst.ac.uk/ sutton/index.html.

180 160 140 120 100 80 60 40 20 0

1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1994 1995 1996 1997 1998 1999 2000 2002 2003 2005 2006 2008 2011

Number

Sinn Féin and Republican Clubs   103

Year IRA attacks excluding assassinations

IRA assassinations

Figure 4.2  IRA attacks and assassinations, 1970–2011. Source: National Consortium for the Study of Terrorism and Responses to Terrorism (START) (2017). Global Terrorism Database [Data file], www.start.umd.edu/gtd. Note Types of attacks included here are armed assaults, bombing/explosion, facility/infrastructure attacks, hijacking, hostage taking (barrier incident), hostage taking (kidnapping) and unclassified.

Attacks

support an exhausted local police force unable to maintain order. According to the Sutton index of deaths, the British Army and the Royal Ulster Constabulary were responsible for 352 deaths related to the NI conflict during 1968 and 2001 (Sutton, 2017). By early 1971 the IRA, now split between the left-­leaning, more politically oriented Officials and the more traditionalist and militant Provisionals, were involved in offensive operations against the British Army (Dixon, 2008, 111). 9 8 7 6 5 4 3 2 1 0

1971

1972

1973

1974 1975 1977 1979 Year Official IRA attacks (excluding assassinations) Official IRA assassinations

Figure 4.3  Official IRA attacks, 1971–79. Source: National Consortium for the Study of Terrorism and Responses to Terrorism (START) (2017). Global Terrorism Database [Data file], www.start.umd.edu/gtd. Note Types of attacks included here are armed assaults, bombing/explosion and unarmed assault.

104   Sinn Féin and Republican Clubs The Provisional IRA, taking a lead in the defence of Catholic communities from the police and loyalist attacks, and aided by Provisional Sinn Féin, quickly won support from many Catholics, especially after the introduction of curfews in 1970 and internment in 1971 largely targeted Catholics (English, 2004, 102–104, 106–115; Feeney, 2003, 254–256). While in 1972 the Official IRA declared a ceasefire, the Provisionals intensified violent attacks in NI and mainland Britain, and were ultimately responsible for some 1,705 deaths during the conflict (Sutton, 2017). Following from the establishment of the UVF in 1966, which was responsible for 428 deaths during the conflict, other loyalist paramilitary organizations soon emerged to defend loyalist areas as violence escalated (Dixon, 2009, 16; Darby, 1976, 24; Sutton, 2017). These included the Ulster Defence Association, formed in 1971 and responsible for 113 deaths, and the associated Ulster Freedom Fighters, responsible for 147 deaths during the conflict (Sutton, 2017). Figures 4.1–4.3 provide a chilling chronology of the scale of violence, at its worst in the years prior to and during the period in which British government decided to legalize Republican Clubs and Sinn Féin. After an escalation of protests following ‘Bloody Sunday’ in January 1972, when British troops killed 14 demonstrators at an anti-­internment demonstration, the British government reintroduced direct rule in NI, effectively suspending the legislative and executive powers of NI institutions. When first Republican Clubs and then Sinn Féin were legalized, British policy-­makers were in no doubt that the scale of political violence in NI was an emergency situation that required an urgent response. In the speech announcing legalization of Sinn Féin in 1974 Secretary of State for NI, Merlyn Rees, highlighted the ‘special problem’ of ‘law and violence’ in NI, the existence of paramilitary forces, bombing, sectarian murders, the ‘717 killed in the last two years alone, the difficulties of people who had lost families and friends to the conflict and damage to property of around £110 million since 1969’ (HC Deb 871, 1974, cc.  1470). In contrast to the Unionist Stormont government, historical memory and contemporaneous experiences of serious political violence and social unrest did not, by 1973–74, facilitate securitization of Sinn Féin and Republican Clubs as threats justifying proscription. Rather, the scale of the emergency in NI and the hope that legalization might help bring violence to an end, underpinned British government decisions to end proscription of Republican Clubs and Sinn Féin. In May 1973, following an Official IRA ceasefire, the British government legalized Republican Clubs on the grounds that they ‘form an organization which wishes to put forward political views and, on the whole, does not pursue them by violent ends’ (Howell (Minister of State for NI), HC Deb 856, 1973, cc. 1175). Moreover, the government argued ‘there are members of Republican Clubs … who in fact condemn violent means’ (HC Deb 871, 1974, cc.  1470) and ‘[m]any of its members do not advocate violence’ (HC Deb 856, 1973, cc. 1174). While the government initially refused to legalize Sinn Féin because it, ‘as an organization, advocates violence as a means of achieving its ends’ (HC Deb 856, 1973, cc. 1174), by April 1974, it took the position that

Sinn Féin and Republican Clubs   105 there are signs that on both extreme wings [i.e. Officials and Provisionals] there are people who, although at one time committed to violence, would now like to find a way back to political activity. It is right to encourage this as much as possible. (HC Deb 871, 1974, cc. 1476) Some opposition and backbench MPs rejected legalization, refusing to accept that the parties were genuinely willing to abandon violence. Government papers acknowledged Sinn Féin openly supported violence and Republican Clubs, despite its ceasefire, had not entirely abandoned it (British Government, 1974a, 1974c, 1979a and 1974b); NI Office (1983). However, by 1974 both Labour and Conservative parties, among others, had come to the position that legalization might contribute to the end of violence. The Conservative government legalized Republican Clubs in time for forthcoming elections ‘to try to encourage further opportunities for debate and discussion as methods of furthering political aims, as opposed to violence and the use of force’ (Lord Windlesham (Minister of State for NI), HL Deb 342, 1973, cc. 741). The government legalized Sinn Féin, according to a government paper, following information that Sinn Féin considered putting candidates forward in February 1974 elections (British Government, 1977), and, according to public statements, in the ‘hope that, by enabling … Sinn Féin to operate openly, people who share the political opinions of these organizations will express them peacefully and within the law’ (Lord Donaldson, HL Deb 351, 1974, cc. 1092–1093). Those supporting legalization claimed it could contribute to combatting violence in a number of ways. Poor election results might diminish the ‘right’ of those supporting violence ‘to speak, to have attention paid to them and to try to threaten with the gun’ (HC Deb 853, 1973, McNamara (Labour), cc. 1347 and Douglas (Labour), cc. 1627). Good election results ‘would show what the position was’ and form the basis for developing appropriate policy responses (Orme (Labour), HC Deb 14 856, 1973, cc.  1150; British Government, 1977; Mason, 1977). Within the republican movement, elections might have strengthened those arguing for political over military means to achieve goals (Orme (Labour) HC Deb 14 856, 1973, cc. 1150). Given a meaningful opportunity to express grievances by voting for republican parties, legalization might undermine support for violence as a political strategy among republicans in NI more generally (Devlin (Unity Party) HC Deb 853, 1973, cc. 1381–1382). An additional advantage of legalization would be the availability of a ‘legitimate mouthpiece … making possible a dialogue other than that of a gun’ (British Government, 1975a). Desecuritization and mediation of difference through the public sphere The integration of Republican Clubs within the Official IRA and what was by 1970 Provisional Sinn Féin, within the Provisional IRA, both of which continued to undertake acts of violence, was widely acknowledged in parliament, the press and government documents. In the 1970s, government briefs stated ‘Sinn Féin is

106   Sinn Féin and Republican Clubs in no real sense an organization separate and distinct from the IRA’, noting a ‘substantial extent of common membership’ (British Government, 1973a, see also 1974b and 1979a and 1979b). They further noted that ‘the whole objective of Sinn Féin is to promote and encourage the objectives of the IRA’, the statements of Sinn Féin leaders ‘clearly associated them with the present campaign of violence’ and ‘various leading members of Sinn Féin are also members of the IRA and associated with terrorist acts’ (ibid.). Some six years after it was legalized, a government brief stated that ‘the OIRA [Official IRA] remains in being and retains a considerable military capacity’, even though it ‘appear to be under firm control of the political leadership’ (British Government, 1979a). Nearly ten years after legalization, Secretary of State for NI James Prior acknowledged in parliament ‘Provisional Sinn Féin acts unconstitutionally and supports violence’ (HC Deb 549, 1983, cc. 549; British Government, 1983b). A 1983 NI Office Note stated it is beyond doubt that some of the most prominent members of Sinn Féin are also, or have been, leading figures in the command structure of the Provisional IRA, and they have not disassociated their political activities from the IRA’s campaign of terror further than the law requires. (NI Office, 1983; British Government, 1978a, 1978b) Nevertheless, British policy-­makers sought to draw fine-­grained distinctions between the political and military wings of the Official and Republican movements. In so doing, policy-­makers undertook a process which Hansen identifies as one of the core attributes of desecuritization (2012, 534), namely ‘reconstituting’ the ‘friend-­enemy distinction’ established by the Stormont government. As the previous section spelt out, a core claim underpinning the Unionist-­dominated Stormont regime’s justification for banning Sinn Féin and Republican Clubs was that they were security threats by virtue of their support for and integration within the IRA. That is, Sinn Féin and its successor, Republican Clubs, along with the IRA were framed jointly as ‘enemies’ of the NI state. Following the resumption of direct rule, by contrast, a core claim underpinning British policy-­ makers’ justification for legalizing the parties was that paramilitaries and parties were different. While the Official and Provisional IRA continued to commit acts of violence, the Republican Clubs and Sinn Féin, as parties, were, at least in theory, capable of contributing to a political solution to conflict in NI. From 1974 differentiation between military and political wings of the republican movements – or between the terrorist organization and political party, or those committing acts of terrorism and those undertaking non-­violent but ancillary acts – became a core element government proscription policy. Aware that the Official IRA ‘remains in being and retains a considerable paramilitary capacity’, a 1979 government paper nevertheless emphasized that leaders of what was by then Republican Clubs–The Workers Party had declared there was ‘no connection between the Party and the Official IRA’ and that the party had ‘sought to distance itself from its militarist past’, regularly participated in elections and ‘totally rejects the militant nationalism of the Provisional IRA’ (British

Sinn Féin and Republican Clubs   107 Government, 1979a and 1979b). Secretary of State for NI, Merlyn Rees, sought to distinguish between, on the one hand, the ‘men of violence’, and on the other, those who, ‘although at one time committed to violence’ could make use of a political party ‘to find a way back to political activity’ (HC Deb 871, 1974, cc. 1476; see also HC Deb 855, 1973, Tope (Liberals), cc. 322–323). Rees later argued ‘the distinction between the IRA and Sinn Féin is a meaningful one, certainly in West Belfast’ (HC Deb 899, 1975, cc.  243). The Conservatives also emphasized the advantages of allowing the ‘political arm of the I.R.A. the opportunity to put forward their political views’ (Lord Windlesham (Conservative) HL Deb 351, 1974, cc. 1094). A 1974 government paper described Sinn Féin (along with the Ulster Volunteer Force) as ‘the most obvious candidates for legalization’ because while the IRA ‘is still obviously committed to violence … Sinn Féin is capable of presenting itself (perhaps more in the Official wing than in the Provisionals) as a nonviolent political organization’ (British Government, 1974a). In 1978, a government brief pointed out that at ‘middle and lower levels, there is much less duplication of membership between PSF and PIRA’ and that there was ‘a significant segment of PSF membership which is dedicated to republican ideals without necessarily participating in PIRA’s violent methods’ (British Government, 1978a). Acknowledging the fine line between perpetration of political violence and ancilliary acts, it was argued individuals committing acts of terrorism could be prosecuted for criminal acts, regardless of whether they were members of legal parties (Howell (Minister of State for NI) HC Deb 856, 1973, cc. 1174; see Lord Donaldson HL Deb 351, 1974, cc.  1092; British Government, 1983g). In this regard, a 1973 policy brief observed ‘most of the really dangerous people in Sinn Féin are also associated with the IRA, and can therefore be prosecuted either for membership of that organization [i.e. the IRA], or more likely, for specific terrorist acts’ (British government, 1973a, 1975a and 1975b; Rees (Secretary of State for NI), HC Deb 871, 1974, cc. 1476). In contrast, a 1983 NI Office Note remarked, the Ulster Volunteer Force, legalized at the same time as Sinn Féin but banned again in 1975, ‘took no pains to keep its terrorist activities separate from its political wing’ (NI Office, 1983). Unionist MPs at Westminster continued see the political and military wings of republican movements as inseparable and to frame the parties as security threats on the grounds of their support for terrorism (e.g. McMaster (UUP) HC Deb 853, 1973, cc. 1388–1389, Kilfedder (UUP) HC Deb 856, 1973, cc. 1175). The distinction between terrorist organization and political party was rather shaky and periodically questioned following new evidence that, particularly Sinn Féin, supported the IRA. For example, in the aftermath of the IRA’s Harrods bomb (17 December 1983), in which six people died and many dozens were injured, the government was under pressure to respond to reports that British public opinion favoured banning Sinn Féin (British Government, 1983f ). Press reports noted ‘growing public resentment at statements by leading Republican politicians linking the bullet and the ballot box’ (Irish Times, 1983) and the ‘apparent condoning of IRA attacks by [Sinn Féin’s] Mr Gerry Adams’ (Belfast

108   Sinn Féin and Republican Clubs Telegraph, 1983). A 1983 Note from the NI Office acknowledged the government was aware that ‘some of … the most prominent members’ of Sinn Féin ‘have been leading figures in the command structure of the Provisional IRA, and they have not disassociated their political activities from the IRA’s campaign of terror’ (NI Office, 1983). Yet, within days of the attack, Secretary of State for NI James Prior commented on Radio Ulster: ‘The IRA are the people who are going to cause us trouble not Sinn Féin’ (British Government, 1983d) and after considering the issue, the government decided not to ban Sinn Féin (British Government, 1983g). Some secret government papers suggest varying frames of republican parties’ relationships with the paramilitary organizations was a strategic choice designed to legitimize government decisions on proscription. In 1973, prior to legalization of Sinn Féin, the civil service considered how the relationship between Sinn Féin and the IRA should be presented if Sinn Féin remained banned, commenting: Much more could be done than has hitherto been attempted to show that Sinn Féin is not in any real sense a separate or autonomous organization, but rather the hat worn by the IRA when it chooses to appear more respectable. (British Government, 1973b) Two years later, following legalization, one civil servant commented that it was ‘useful to maintain the distinction’ between Provisional Sinn Féin and Provisional IRA: With a legal Provisional Sinn Féin (PSF ), ‘Civilian members of PSF can act as a moderating influence on the military wing’ and ‘It provides the PIRA with a legitimate mouthpiece and makes possible a dialogue other than that of the gun’ (British Government, 1975a). Similarly, a 1983 NIO Note stated the preference for prosecuting terrorist crimes over proscription ‘divorces straightforward criminality of terrorist acts from the self-­ascribed political motives of the perpetrators, [which] enables Ministers to contend that all political beliefs or aspirations, however extreme or provocative, are legitimate if peacefully pursued’ (NI Office, 1983). Whatever its motivation, this discursive reworking of the purported relationship between the political and military wings of the Officials and Provisionals amounted to a reconstitution of the friend-­enemy distinction at the core of the Stormont government’s rationale for banning the parties. It did not amount to a reclassification of the parties as a ‘friend’. Many speakers insisted on their profound abhorrence for the parties’ ambiguity or support for violence. Nevertheless, the discursive distancing of the political and military wings of the Official and Provisional republican movements by reference to the former’s potential functional contribution to conflict resolution was a consequential reworking of the friend-­enemy distinction that underpinned justification for legalization. A second element of desecuritization processes identified by Hansen and observed in the legalization of Republican Clubs and Sinn Féin involved a

Sinn Féin and Republican Clubs   109 reconceptualization of the ‘problem’ of Republican Clubs and Sinn Féin’s support for political violence as one more properly dealt with through the mediation of difference through the public sphere (2012, 531). Successive governments pursued the form of desecuritization which Hansen describes as ‘rearticulation’, which involves ‘desecuritizations that remove an issue from the securitized by actually offering a political solution to the threats, dangers and grievances in question’ (ibid., 543). An inclination to deal with Republican Clubs and Sinn Féin through ‘mediation of difference through the public sphere’ (ibid., 534; Buzan et al., 1998, 4) is evident in frequent references by speakers favouring legalization to the virtues of free speech and democratic representation of a wide range of political views. The Labour government, for example, commented that ‘powers of proscription are emergency powers … [which] should be used only when absolutely necessary, and should never be allowed to stifle the free and lawful expression of opinion’ (Lord Donaldson (Parliamentary Undersecretary of State, NIO), HL Deb 351, 1974, cc. 1092). In 1977, Secretary of State for NI Roy Mason stated in relation to republican parties: ‘those who hold political views should have the right to express them and should not be denied the opportunity to pursue their political aims through the normal democratic process’ (Mason, 1977). A 1975 civil service brief warned: One is on a very slippery slope when one begins to proscribe organizations simply (or largely) because the views they uphold are abhorrent to the moderate majority, especially when the group in question is physically and conceptually distinct from the gunmen. (British Government, 1975a) More fundamentally, the core rationale for legalization of Republican Clubs and Sinn Féin in 1973–74, was that it would permit the parties to participate in electoral processes designed as a political solution to the NI conflict. In 1973 the government proposed a new NI Assembly, to be elected using a more proportional electoral system, a NI powersharing executive, and a Council of Ireland, channeling cooperation between the north and south of the island of Ireland (Bew et al., 2009, 45; Cunningham, 2001, 12–16). Prime Minister Edward Heath argued that the aim of the proposals was: to create conditions in which those who wish to prevail in Northern Ireland can do so only by persuading their fellow citizens, by asking for and securing their votes. … Our aim has been to recreate the normal processes of politics, of peaceful argument and fair resolution of that argument at the polling booths. … The whole purpose of this White Paper is to restore the polling booth to its place at the centre of the political life of Northern Ireland – and to banish and defeat the gun, the rocket and the bomb, the terrorists and the hooligans. (HC Deb 853, 1973, cc. 1331–1332)

110   Sinn Féin and Republican Clubs Arguing for legalization, the Conservative government observed that Republican Clubs were putting forward candidates for forthcoming elections and that the government was ‘anxious to see all genuine political activity encouraged’ and ‘welcomes the expression of political views, however unreasonable and unacceptable they may be to one side or the other’ (Howell (Minister of State for NI) HC Deb 856, 1973, cc. 1149; see also HL Deb 342, 1973, cc. 740–741). Republican Clubs subsequently participated in both May 1973 local elections, where it won 2.6 per cent of the vote and six council seats, and the June 1973 Assembly elections, winning a modest 1.8 per cent and no seats. Following elections in June 1973, pro-­agreement parties won 52 of the 78 seats in the Assembly and, following negotiations among the pro-­assembly parties and Irish and British governments at Sunningdale in the UK, a power sharing government was formed. Government documents record that after consultation with key figures in the new NI Executive, who agreed legalization would be helpful, Labour and Conservative Parties concluded it was ‘right for Provisional Sinn Féin to be given the opportunity to contest elections’ (British Government, 1977; Mason, 1977). A government paper suggested Sinn Féin might have stood for a forthcoming June 1974 Assembly by-­election, and that the Secretary of State rejected an approach by Brian Faulkner to postpone legalization until after that by-­election (British Government, 1974b and 1974d). The new NI Assembly and Executive confronted growing signs of unionist dissatisfaction, focusing particularly on fears that institutionalization of relations with the Republic of Ireland was a slippery slope to unification (Bew et al., 2009, 46). In February 1974, 11 out of the 12 NI MPs elected to Westminster had stood on an anti-­Sunningdale platform. In May, following a 14-day general strike organized by the loyalist Ulster Workers’ Council, and withdrawal of support by Ulster Unionist Party members, the new NI institutions were effectively moribund. Soon afterwards, the government proposed a Constitutional Convention in which NI political parties could themselves decide on a form of government (Cunningham, 2001, 17–18). In May 1975 elections for the Convention were held, returning the United Ulster Unionist Council, which opposed power-­sharing institutions and any ‘Irish dimension’, as the largest party with 47 of the 78 seats. While Republican Clubs participated in these elections, winning a modest 2.2 per cent of the vote but no seats, Provisional Sinn Féin did not participate. Soon after publishing a report deemed unacceptable to nationalist representatives and the British government, the Convention was dissolved. Nevertheless, the government continued to hold on to the view that: ‘As long as Provisional Sinn Féin remains deproscribed the possibility that they will contest an election remains’ (British Government, 1977). This hope was at least partially fulfilled, when in October 1982, Sinn Féin did participate in elections to a new NI Assembly, held to initiate a process of ‘rolling devolution’ (Bew et al., 2009, 93–95). Sinn Féin won 5 seats and 10.1

Sinn Féin and Republican Clubs   111 per cent of the vote, standing on an abstentionist platform and refusing to take up seats won. However, rather than providing an alternative to violence, as government officials observed, ‘success with the armalite and success with the ballot box run together and may even reinforce each other’ (British Government, 1983a). The SDLP’s refusal to participate scuppered the initiative (Bew et al., 2009, 95). Despite its concern that SF might overtake the SDLP as the primary party of nationalists in NI, the policy against proscription continued. Government papers show consideration was given to the position of the SDLP in its ban decision, but suggest the government was influenced by the SDLP view that banning Sinn Féin would undermine the SDLP’s position and ‘make elections in the North very difficult’ (British Government, 1983e and 1983f ). Defending the position on the Panorama television programme in 1982, Secretary of State for NI James Prior stated: ‘One of the good things in this election is actually to bring out the real problems we have. Now we know what we’ve got to face and try to deal with it’ (British Government, 1982). Sinn Féin’s electoral engagement was a ‘signs [that] the party itself, is … moving away from violence’ (ibid.). Civil servants warned of the danger that proscription might further alienate those who voted for Sinn Féin from the political and constitutional process (British Government, 1983c) and that ‘The general sledge-­hammer [of proscription] should be avoided so that an outlet for legitimate political activity is not shut off ’ (NI Office, 1983). In addition to arguments that legalization might contribute to an end of violence, that the parties and paramilitaries in the republican movement could play different roles, and a preference for mediation of difference in the public sphere, it is notable that those favouring legalization also cast doubts on the effectiveness of proscription. Proscription was sometimes seen as a ‘judicial shortcut’ (British Government, 1973a; McEldowney v Forde, HL 18 Jun 1969), something of a deterrent (British Government, 1973a; Lord Garnsworthy (Labour), HL Deb 347, 1973, cc.  855), and assuaging public outrage (British Government, 1975b; Mills (Under Secretary of State for NI), HC Deb 865, 1973, cc.  1598; see also Walker, 1986, 44–47; Donohue, 2001, xix). More often, however, it was seen by UK governing parties as ineffective and counterproductive. The Secretary of State for Northern Ireland, Merlyn Rees, opened debate on legalization of Sinn Féin in April 1974 stating ‘we have consistently expressed doubts about the effectiveness of the whole concept of proscription’ (see also Lord Garnsworthy (Labour), HL Deb 855, 1973, cc. 855). Various commentators further remarked that proscription would not prevent the party from contesting elections under a different name (British Government, 1983f, 1983g; NI Office, 1983; Belfast Telegraph, 1983; Stanley Orme, Labour, HC Deb 856, 1973, cc. 1150–1167), that prosecution for membership of a proscribed organization was legally difficult and extremely rare (NIO, 1983), and that ultimately, it would not contribute to ending terrorism (British Government, 1983c, 1983f; The Guardian, 1983; Stratton Mills (Alliance), HC Deb 856, 1973, cc. 1165; Walker, 1986, 46). Proscription could be counterproductive insofar as it would drive organizations underground,

112   Sinn Féin and Republican Clubs making it harder to determine who members of potentially dangerous organizations were (Rees (Labour), HC Deb 865, 1973, cc. 1584–1585; Lord Garnsworthy (Labour), HL Deb 347, 1973, cc. 855), and provoke widespread civil disobedience by Sinn Féin members who might flout the law (British Government, 1975a). It might add fuel to the fire of grievances generated by decades of discrimination (McNamara (Labour) HC Deb 856, 1973, cc. 1167), evoke sympathy, and provide a propaganda weapon, for republicans (HC Deb 865, 1973, cc.  1598; Mills (Under Secretary of State for NI), McNamara (Labour), cc.  1167; Orme (Labour), cc.  1150), and provide further justification for violence on the grounds of exclusion from the political process (Gerry Fitt (then SDLP), HC Deb 853, 1973, cc.  1572, see also McNamara (Labour), cc. 1584–1585).

Conclusion Like the preceding chapter, this chapter aimed to examine hypotheses focusing on the role of securitization and desecuritization discourses, positions of veto players and parties’ orientation to violence in party ban and legalization decisions. As in the case of bans on Herri Batasuna and its successors, the empirical analysis of bans on Sinn Féin and Republican Clubs confirmed hypotheses on securitization, veto-­player and violence variables. Sinn Féin’s support for the IRA’s 1956 Border campaign and the Republican Clubs’ integration within a movement which in the mid-­1960s had not completed the process of renouncing armed insurgency, facilitated the securitization of these parties as integral to a terrorist organization and a threat to the existence of NI as a territory within the United Kingdom. The UUP, a single veto-­player dominating executive and parliamentary institutions in NI and empowered by the SPA and the neglect of British authorities to act alone, banned the parties. However, in contrast to the legalization of Bildu and Sortu in Spain, legalization of Republican Clubs and Sinn Féin following the suspension of NI institutions, did not confirm all relevant hypotheses. Republican Clubs and Sinn Féin were legalized during a security emergency in which hundreds of people were being killed in a sectarian civil war, rather than the context of de-­ escalation that occurred in the Spanish cases. While the Official IRA declared a ceasefire before its political wing was legalized, Sinn Féin openly supported the Provisional IRA, which had intensified its military activities. Paradoxically, the context of quasi-­civil war nevertheless facilitated the desecuritization of Republican Clubs and Sinn Féin in a process controlled by successive Conservative and Labour governments, which adopted a bi-­partisan approach to NI policy. Once they were in charge in NI, Conservative and Labour governments emphasized differences between terrorist organizations and the political parties, seeking to mobilize the latter in the search for solutions to conflict in NI.

Sinn Féin and Republican Clubs   113

Notes 1 Parliamentary debates of the 1921–1972 Stormont parliament referred to are recorded in the NI House of Commons Official Report (henceforth NI HC Official Report) and can be found online at: http://stormontpapers.ahds.ac.uk/index.html. UK parliamentary debates referred to are recorded in House of Commons Hansard (henceforth HC Deb) and House of Lords Hansard (HL Deb) and can be found online at https://hansard. parliament.uk. 2 NI Ministry Home Affairs, 1958, 1959 and 1967; RUC, 1958; Belfast Telegraph, 1956o; Belfast Newsletter, 1956; Manchester Guardian, 1957.

5 The Socialist Reich Party and National Democratic Party of Germany From ‘militant democracy’ to declining anti-­system party threats This chapter examines the banning of the Socialist Reich Party of Germany (SRP, Sozialistische Reichspartei Deutschlands) (1952) and two failed attempts to ban the National Democratic Party of Germany (NPD, Nationaldemokratische Partei Deutschlands) (2003 and 2017). I begin with a description of the origins of the parties, their key programmatic objectives and electoral salience. I then turn to examine the hypotheses: H1. Democracies ban anti-­system parties if these parties have been securitized as an existential threat. H2. Democracies ban anti-­system parties if veto players prefer proscription. H3. Democracies ban anti-­system parties if they do not unambiously reject violence. I employ various documentary sources: In the SRP case, I examined archived government files on the SRP kept at Das Bundesarchiv, in Koblenz, minutes of cabinet meetings, records of German government meetings with the Allies and newspaper articles from Süddeutsche Zeitung and Frankfurter Allgemeine Zeitung. For the NPD case, I examined parliamentary debates,1 court rules and newspaper articles principally from Frankfurter Allgemeine Zeitung. Analysis confirms the three hypotheses for the SRP case, and hypotheses on securitization and violence, but not veto player support for the NPD failed ban cases, which is consistent with theory.

The Socialist Reich Party and the National Democratic Party The Allied occupying powers operated party licensing systems, which permitted the re-­emergence of some extreme-­right parties (Nagle, 1970, 16). These included a party that by 1950 came to be known as the Deutsche Reichs-­Partei

The SRP and NDP   115 (DRP, German Reich Party), which became one of the most radical of the right-­ wing parties emerging after the war. It increasingly espoused more national socialist positions and, for much of the 1950s, adopted a ‘nationalist-­neutralist stance’ strongly opposed to Germany’s pro-­Western course and NATO ties (Nagle, 1970, 17, 27). Another party was the Deutsche Partei (DP, German Party), drawing together regionalist, archconservative groups loyal to the Guelph monarchy, German nationalists and Nationalist Socialists. It participated in Konrad Adenaur’s first CDU-­led federal government coalition until it was effectively absorbed into the CDU (Nagle, 1970, 16–29). The Gesamtdeutsche Block/ Bund der Heimatvertriebenen und Entrechteten (All-­German Block/Federation of Expellees and Disenfranchised, GB/BHE) mostly appealed to refugees and emigrants from communist-­held areas of what was previously German territory, people injured during the war, Nazi party functionaries, and third Reich officials who lost their positions and status through denazification (Nagle, 1970, 21). After obtaining several ministerial posts in Adenaeur’s second government, members of GB/BHE were mostly absorbed into the CDU/CSU and FDP (Nagle, 1970, 24; Backes and Mudde, 2000, 458). The SRP, formed in 1949 and banned in 1951, emerged following a split from the DRP and the departure of that party’s most radical leaders (Nagle, 1970, 19; Fisher, 1974, 129; Braunthal, 2009, 23). As discussed in more detail below, it was a blatantly neo-­Nazi party, claiming the Third Reich remained the legal authority in Germany and claiming West German political parties were ‘lackeys of the occupation powers’ (Fisher, 1974, 130; Zimmermann and Saalfeld, 1993, 52; Nagle, 1970, 20). As Fisher points out, its founders were mainly former Nazi Party functionaries, SS and army officers: For example, SRP chairman, Fritz Dorls was Nazi party member from 1929 and lecturer in Nazi indoctrination; [his] party colleagues included Wolf Count von Westrap, a former member of the SS/Elite Guard, Dr Gerhard Krieger, a member of the SA since 1924 and a high-­ranking Nazi official, and Otto Ernst Remer, an army officer who was promoted by Hitler for his part in crushing the anti-­Hitler putsch in July, 1944. (1974, 129; see also Nagle, 1970, 19) Nagle argues that, ‘through its far more explicit use of Nazi trappings and its strident denunciation of the Bonn republic, the SRP was able to attract many former DRP, DP and [other far-­right party] members’ (1970, 20). The party enjoyed electoral success in 1951 in the states of Lower Saxony, winning 11 per cent of the vote and 15 seats, and Bremen, winning 7.7 per cent and 8 seats in Bremen state elections (see Table 6.5). Two of the five Bundestag representatives of the DRP elected in 1949 joined the SRP shortly after the 1949 Bundestag elections (namely Fritz Dorls and Dr. Franz Richter (alias Fritz Rössler)). At the local level in Lower Saxony, the SRP won an absolute majority in 35 local governments and was the largest party in 375 local governments (Stöss, 1991, 110). After the party was banned, there were no attempts to revive it (Nagle, 1970, 20; Fisher, 1974, 130).

116   The SRP and NDP The emergence of the NPD in 1964, mainly from the remnants of smaller far-­ right parties (Nagle, 1970, 33), marked a new phase in the development of the post-­war German far right (Zimmermann and Saalfeld, 1993; Braunthal, 2009, 24). A significant portion of NPD leaders – 8 out of 18 national committee members – were former Nazi party members (Nagle, 1970, 33–35; Fisher, 1974, 143). As various authors argue, proscription of the SRP and KPD, alongside mainstream parties’ commitment to the West German political system, forced the NPD to proclaim, outwardly at least, a commitment to liberal democracy (Fisher, 1974, 145: Nagle, 1970, 103, 122; Braunthal, 2009, 25). In its party statutes, the NPD formally pledged loyalty to the Basic Law and the fundamental features of the West German political, social and economic system (Fisher, 1974, 146). The party’s official newspaper, Deutsche Nachrichten (German Reports), according to Nagle, took ‘great care not to say anything which could bring the NPD before the courts on charges of advocating the overthrow of the republic’ (1970, 35). Fisher argues the NPD was less extreme than the interwar National Socialist German Workers’ Party (NSDAP, Nationalsozialistische Deutsche Arbeiterpartei), even if there are many other ideological similarities. In 1966, in a context of deflated economic growth, public anxiety and criticism of the government, the NPD had its first electoral success (Nagle, 1970, 4; Braunthal, 2009, 24). In that year, the party cleared the 5 per cent hurdle for the first time, winning 7.9 per cent of the vote and 8 seats in Hessen, which was soon followed by a win of 7.4 per cent and 15 seats in Bavaria. In 1967–68, the party’s success continued, when it obtained seats in Schleswig Holstein (5.8 per cent vote, 4 seats), Rhineland-­Palatinate (6.9 per cent vote and 4 seats), Lower Saxony (7 per cent vote and 10 seats), Bremen (8.8 per cent vote and 8 seats) and Baden-­Würtemberg (9.8 per cent vote and 12 seats). In state elections between 1966 and late 1968, the party won 1.9 million votes and 61 seats in Land parliaments (Fisher, 1974, 141; Nagle, 1970, 68; Kailitz, 2000, 8). During this early phase, the NPD came to establish a presence in 75 per cent of all German cities and all states and its membership rose to around 25,000 by the end of 1966 (Mudde, 2000, 27). NPD success was short-­lived. In the 1969 Bundestag elections, the NPD won 4.3 per cent of the vote, below the 5 per cent electoral threshold permitting it to take up seats (see Table 6.6). The party’s fortunes subsequently declined and it never achieved the level of success of the 1960s again. By 1971, it lost all its seats in state parliaments (Backes and Mudde, 2000, 458) and in the 1972 Bundestag election the NPD won 0.6 per cent of votes, a decline of 1.2 million votes (Fisher, 1974, 142). Until the mid-­2000s, the party’s electoral fortunes remained well below 1 per cent at state and Federal levels and the party itself became, according to Mudde, ‘a sectarian group of militant outcasts’ (2000, 30; see also Braunthal, 2009, 25). In the 1980s, the Die Republikaner (REP, The Republicans) and Deutsche Volksunion (DVU, German People’s Union) emerged as competitors. Formed in 1983, REP sought to establish itself as a modern right-­wing populist party, inspired by the French National Front (Mudde, 2000, 31; Backes and Mudde,

The SRP and NDP   117 2000, 459). Its breakthrough was in 1989 in West Berlin (7.8 per cent vote, 11 seats) and in European Parliament elections (7.1 per cent vote, 6 seats). In the 1990s, the REP performed well in Baden-­Württemberg (1992, 10.9 per cent vote, 15 seats; 1996, 9.1 per cent vote, 14 seats). Since 2001, however, the REP has not won seats in any Land parliament since 2001. It has never won seats in the Bundestag. The DVU was founded in 1987 by the multi-­millionnaire media magnate Gerhard Frey (Mudde, 2000, 62; Backes and Mudde, 2000, 461). Until the emergence of Alternative für Deutschland (AfD, Alternative for Germany) (see below), the DVU achieved the best ever electoral result of an extreme-­right party in the post war period, when it won 12.9 per cent of the vote and 16 seats in the 1998 Saxony-­Anhalt state elections. The DVU was also successful in Bremen (1987, 3.4 per cent vote, 1 seat; 1991, 6.2 per cent vote, 6 seats; 1999, 3.3 per cent vote, 1 seat; 2003, 2.3 per cent vote, 1 seat; 2007, 2.7 per cent vote, 1 seat), Brandenburg (1999, 5.3 per cent vote, 5 seats; 2004, 6.1 per cent, 6 seats) and Schleswig-­Holstein (1992, 6.3 per cent, 6 seats). Since 2005 the DVU has been in an enduring alliance with the NPD which culminated in their merger in 2011 (Decker and Miliopoullos, 2009, 97). According to Decker and Miliopoulos, the greater degree of outrage and attention paid to the NPD, compared to the REP and DVU, was because the NPD ‘only thinly veils its ideological proximity to National socialism’ (2009, 97). Historical revisionism played a larger role in the NPD’s policy portfolio, compared to what was then the DVU and REP, which predominantly focused on xenophobia and anti-­immigrant themes (ibid.). Moreover, under the chairmanship of Udo Voigt (1996–2011), the NPD began to radicalize in terms of ideology and strategy. As I discuss in more detail below, under Voigt, key NPD programmatic commitments included advocating national-­socialism with an anti-­capitalist and national-­revolutionary element; a biological conception of race and associated political demands such as the deportation of Germans with immigrant backgrounds; and an openly anti-­establishment critique (Decker and Miliopoulos, 2009, 101; Backes, 2006, 137–139). The NPD forged ties with a burgeoning extreme-­right subculture including violent neo-­ Nazi groups and more loosely structured Freie Kameradschaften (Free Comradeships) (Backes, 2006, 137–139; Decker and Miliopoulos, 2009, 97; Schellenberg, 2015, 37–38). While the NPD had previously been a party composed ‘more-­or-less exclusively of aging, ex-­Nazis’, its new strategy transformed the NPD from a small party into a ‘movement that makes forays into popular culture … motivated in part by a desire to attract young adherents’ (Schellenberg, 2015, 38). Following reunification of Germany, the NPD focused its attention on the East, where it obtained some important successes. More specifically, it won seats in Mecklenburg-­Vorpommern (2006, 6 seats, 7.3 per cent vote; 2011, 5 seats, 6 per cent vote) and Saxony (2004, 12 seats, 9.2 per cent vote; 2009, 8 seats, 5.6 per cent) (see Table 6.6). It came close to winning seats in 2014 in Saxony (4.9 per cent vote), Saxony-­Anhalt in 2011 (4.6 per cent), Thuringia in 2009 (4.3 per cent) and also in Saarland in 2004 (4 per cent vote). It has been successful in some municipalities. In 2009, for

118   The SRP and NDP example, it won 72 seats in Saxon and 60 in Mecklenburg-­Vorpommern municipal elections, and has won seats in municipalities in Hesse, Thuringia and Saxony-­Anhalt (Schellenberg, 2015, 44). In 2014, the NPD won its first seat in the European parliament with 1.03 per cent of the vote. At present, the NPD’s main competitor is the AfD, founded in 2013 primarily as a Eurosceptical party, albeit with nationalist and anti-­immigrant influences that eventually came to dominate the party (Grabow, 2016, 174; Grimm, 2015; Schmitt-­Beck, 2017, 125). In 2017, the AfD was the first party of the far right to win seats in the Bundestag, becoming the third largest party with 12.6 per cent of the vote and 94 seats. In the 2013 Bundestag election, when the parties’ identity was predemoninantly that of a Eurosceptical party, it had won 4.7 per cent of the vote (Schmitt-­Beck, 2017). The AfD won 7 per cent of the vote and 7 members of the European Parliament in 2014. It has won seats at all Länder elections held since then and an average of 12.55 per cent of the vote in those elections.

Party-­ban-veto-­players As in Spain, there are multiple institutional and partisan party-­ban-veto-­players in Germany and procedures are sequential, so that different political actors are crucial at different stages. According to Article 21.2 of the Basic Law, the FCC (Federal Constitutional Court) alone determines the constitutionality of a party. The Gesetz über das Bundesverfassungsgericht (FCC Law) (as amended at 31 August 2015) provides that the Bundestag, the directly elected federal parliament, the Bundesrat, representing the 16 Länder governments, and the federal government may each separately initiate legal proceedings to ban a party (Article 43). A Land government may initiate party ban proceedings for a party operating within its territory (Article 43). The Act also requires that where a party ban decision is likely to be disadvantageous to the defendant, a two-­thirds majority is required in the FCC (or six out of eight judges) (Article 15.4). Germany has a parliamentary form of government in which majority coalitions are the ‘characteristic mode of governing’ (Klingemann and Volkens, 1992; Saalfeld, 2003, 2010; Roberts, 2006). In the first phase of the German party system (1949 to 1961), the number of parties winning seats in the Bundesrat fell from ten parties to three and the Christlich Demokratische Union/ Christlich-­Soziale Union (CDU/CSU Christian Democratic Union/Christian Social Union) dominated coalition governments formed with small parties. Between 1961 and 1998, coalitions involved two partners, with the Freie Demokratische Partei (FDP, Free Democratic Party) as a junior partner to either the CDU/CSU or Sozialdemokratische Partei Deutschlands (SPD, Social Democratic Party of Germany) with the exception of a grand coalition between CDU/ CSU and SPD (1966 and 1969). Party system fragmentation following emergence of Bündnis 90/Die Grünen (Alliance 90/The Greens, henceforth The Greens) in the 1980s and, following reunification, emergence of Partei des Demokratischen Sozialismus (Party of Democratic Socialism, PDS) and later

The SRP and NDP   119 Die Linke (The Left), led to more varied forms of coalition, including the two-­ party majority coalitions of the SPD and Greens (1998–2005) and the CDU/CSU and FDP (2009–2013) and ‘grand coalitions’ of the CDU/CSU and SPD (2005–2009 and 2013–2017) (Saalfeld, 2010). The Bundesrat, or the Federal Council, represents Länder interests in Federal decision-­making. Each Land has at least three seats, and thus votes, with additional seats allocated in proportion to the population size up to a maximum of six seats. Individual Land governments nominate Land ministers to take up their share of Bundesrat seats and those ministers must vote en bloc. Land-­level coalition governments must therefore agree joint positions for Bundesrat votes or abstain. Party composition of Land governments has become increasingly important in Bundesrat decision-­making, especially when opposition parties in the Bundestag command a majority in the Bundesrat (Roberts, 2009, 152). In 1951, when the Federal Government initiated its case against the SRP, the 1949 Occupation Statute, imposed by the states occupying Germany after World War II (United States, Britain and France) limited German sovereignty. The Occupation Statute proclaimed that ‘the German people shall enjoy self-­government to the maximum possible degree consistent with [the] occupation’ and that the ‘Federal State and the participating Laender (sic) shall have, subject only to the limitations of [the Occupation Statute], full legislative, executive and judicial powers in accordance with the Basic Law and with their respective constitutions’ (Article 1). The Statute maintained the right for the Western Allies to resume in whole or part ‘full authority if they consider that … is essential to security or to preserve democratic government in Germany’, among other things (Article 3). It also established ‘reserved powers’ on a wide range of issues, including ‘respect for the Basic Law and Land constitutions’ (Article 2.f ). Other reserved powers included control of the Ruhr, and thus key drivers of the West German economy, as well as defence, security and foreign policy (Paterson, 1992, 137–152; Glees, 1996, 62). The 1949 Charter of the Allied High Commission transferred authority of the respective commanders in chief of the Western occupying forces to an Allied High Commission, authorised to exercise control over the Federal and Länder governments as provided by the Occupation Statute. The Bonn and Paris Agreements (1952–54) negotiated in the early Cold War context and in light of US determination to re-­arm Germany within multilateral Western structures, revoked the Occupation Statute, abolished the Allied High Commission and proclaimed Germany the ‘full authority of a sovereign state over its internal and external affairs’ (Wiggers, 2004, 107). Nevertheless, Western powers retained important rights over West German security, rights regarding Berlin, to station troops on German territory and final authority over reunification and territorial revisions (Wiggers, 2004, 107; Glees, 1996, 26). In March 1991, the United States, Britain, France and the Soviet Union renounced these external controls in Treaty on Final Settlement with Respect to Germany negotiated with the Federal Republic of Germany and the German Democratic Republic. Germany ratified the ECHR in 1952, entering into force only after the SRP ban. Subsequently, the ECtHR made two rulings on German party ban cases. In

120   The SRP and NDP 1957, leaders of the KPD claimed that the proscription of their party violated ECHR articles 9, 10 and 11 on freedom of thought, conscience and religion, expression, assembly and association. The European Commission of Human Rights, a body assisting the ECtHR from 1953 to 1998, declared the case inadmissible. It ruled that even if it could be established that the KPD pursued its goals through constitutional means, the party nevertheless sought to establish a ‘dictatorship of the proletariat’, which was incompatible with the Convention because it entailed the destruction of rights and liberties (European Commission of Human Rights, 1957). In 2016, the ECtHR dismissed an application against Germany in which NPD claimed it was subject to a de facto ban (ECtHR, 2016). The FCC had rejected an NPD petition for a declaration the NPD was not unconstitutional under Article 21 of the Basic Law. The NPD argued the constant repetition of claims that it was unconstitutional stigmatized the party, which led to infringements of legal rights, such as dismissal of its members for public service employment or insufficient police protection for its events. Available remedies for legal infringements were insufficient, it argued. The FCC ruled that the NPD had not been left without legal remedy for claimed rights violations because the party could apply for redress in administrative, civil and criminal courts. Agreeing with the FCC, the ECtHR dismissed the case. SRP The Western Allies (United States, Britain and France) were concerned about the rise of the clearly neo-­Nazi SRP, pressured Adenauer to initate ban proceedings against the party and threatened to intervene in order to deal with the threat posed by the SRP. A right-­wing federal coalition government dominated by the CDU/CSU and including the smaller FDP and DP submitted the petition to ban the SRP (and KPD) in 1951. Despite initial disagreements (see below), the Cabinet decided to initiate SRP ban proceedings on 4 May 1951 (Bundes­ regierung, 1951). In October 1952, the FCC declared the SRP unconstitutional. NPD I In 2001, a SPD-­Greens federal government coalition, the Bundestag and the Bundesrat submitted applications to ban the NPD. In the Bundestag, an SPD-­ Greens majority – with support from the PDS and some abstentions among the Greens – approved the NPD ban application (DB Plenarprotokoll 14/141, p. 13807). The FDP voted against, while the CDU/CSU supported ban proceedings but argued that separate Bundestag application was not necessary given Federal government and Bundesrat applications (CDU/CSU, 2000). The CDU/ CSU argued ban applications were an executive task, because Bundestag deputies had less intelligence information than Federal or State governments (FAZ, 2000l, 2000h and 2000n). In an initiative led by Bavaria and Lower Saxony, 11 (of 16) Länder voted in the Bundesrat in favour of the ban (Jesse, 2001, 688). Due to the FDP’s stance against the ban, all Land governments including the

The SRP and NDP   121 FDP voted against the ban (i.e. Hesse CDU-­FDP coalition) or abstained (i.e. Baden-­Württemberg CDU-­FDP coalition and Rhineland-­Palatinate SPD-­FDP coalition), while CDU-­led Saarland voted against and the Berlin SPD-CDU coalition abstained (FAZ, 2000a, 2000i, 2000m, 2000g). In March 2003, the FCC dismissed the case on procedural grounds. Applying Article 15.4 of the FCC Act, in which a decision to the disadvantage of the respondent requires a two-­thirds majority of the members of the Court, a minority of three out of seven judges was able to prevent continuation of the case. NPD II In December 2012, the Bundesrat agreed the second attempt to ban the NPD, without support from the Bundestag of federal government. All Land governments supported the Bundesrat application, except the CDU-­FDP coalition in Hesse, which abstained (Bundesrat, 2012; FAZ, 2000n). In contrast to NPD I, some coalitions with FDP partners (in Bavaria, CSU-­FDP; Lower Saxony and Saxony, CDU-­ FDP) supported the NPD ban application in the Bundesrat, including Saxony where NPD had won seats (FAZ, 2012i). Angela Merkel’s CDU/CSU-­FDP federal government did not apply for an NPD ban. There was strong FDP opposition (FAZ, 2013c, 2013d, 2013f ), although Merkel and the Interior Minister Hans-­Peter Friedrich were also publically sceptical about a ban (Spiegel Online, 2012a; FAZ, 2012b, 2012e and 2012i). The Federal Interior ministry would nevertheless help prepare the case against the NPD (FAZ, 2013a, 2013b). The governing parties in the Bundestag defeated SPD and Left Party ban proposals (DB Plenarprotokoll 17/237, p.  29721–29725). In January 2017, the FCC unanimously dismissed the case, arguing the NPD was not likely to acheive its aims.

The banning of the SRP The Nazi shadow and the SRP’s ambiguous orientation to violence In contrast to HB and its successors, Sinn Féin and Republican Clubs, the primary goal of the SRP was to contest and win elections, although it established a quasi-­military arm – the Reichsfront. The blatantly neo-­Nazi SRP emerged – and was banned – at a time when memories of violence perpetrated by the Nazi regime, the catastrophe of World War II, and onset of the Cold War provided a security setting in which the potential for renewed political violence or even all-­ out war seemed a distinct possibility. In this context, the Reichsfront, the SRP’s neo-­Nazi ideology and its political style created ambiguity about its commitment to non-­violent political methods. This ambiguity facilitated securitization of the SRP as a threat to democracy and, together with the SRP’s links with the Soviet Union in the emerging Cold War, the existence of the West German state. The Reichsfront was conceived as an elite, uniformed, paramilitary organization and routinely flanked the party leadership in public appearances (Lee, 1997, 49). The Federal Interior Ministry, the US High Commissioner and the SPD

122   The SRP and NDP described the Reichsfront as analogous to the Nazi’s paramilitary wing, the Sturmabteiling (Storm Troopers) (BMI, 1951; SPD, 1951; USOHCG, 1951, 65). The Interior Ministry saw the principal task of the Reichsfront as organizing Nazi-­ style military marches, conducting propaganda activities according to national-­ socialist principles, and acting as a security force (BMI, 1951). The US High Commissioner saw it as ‘strong arm squads for political rallies’ which had ‘storm-­trooper aspects’ and noted that SRP ‘[m]eetings are replete with military music … insignia and emotional ultra-­nationalistic appeals strongly reminiscent of the Nazi regime’ (USHICOG, 1951, 65, 67). The Reichsfront was banned in May 1951 for its ‘attack on the constitutional order’ and in the same decision the government announced its intention to ban the SRP (Bundesregierung, 1951). The SRP’s open mimicry of Nazi symbols, oratory and performance at electoral rallies that often attracted more than 1,000 supporters frequently provoked counterdemonstrations and riots involving members of the Association of Victims of the Nazi Regime, the German Communist Party, the SPD and trade union activists (Frei, 2002, 252–253; Lee, 1997, 50). The Interior ministry saw this electoral violence as ‘attempts at intimidation of the public’ analogous to the ‘terror used during the Third Reich’ (BMI, 1951). It concluded: ‘By employing the Führerprinzip [Führer principle] and militaristic patterns, by forming militant associations, as well as by means of propaganda according to National Socialist example, a violent confrontation with political opponents is being prepared’. The federal government’s decision to pursue an SRP ban identified ‘terror’ created among voters as the main conduct undermining the free democratic basic order (Bundesregierung, 1951). Ban justifications frequently appealed to the memory of the Nazi’s path to power through elections. In a May 1951 radio interview, Adenauer stated that unlike Weimar, his government’s strategy against extremism would not grant freedom to the opponents of freedom and allow them to bring back chaos (SZ, 1951c and 1951d). In a Cabinet debate on the SRP, Adenauer argued that democratic principles should not be used, as in 1933, to facilitate the ‘death of democracy’ (Kabinettsprotokolle, 1951c). Other cabinet ministers expressed similar concerns, as did the Allies. For example, the CDU’s Hans Lukaschek considered the SRP’s emergence a ‘state of emergency’ domestically and regarding effects abroad (Kabinettsprotokolle, 1951b; see also SPD, 1951). British foreign secretary Herbert Morrison told Adenauer the SRP gave rise to ‘serious concern’ and that people might think ‘the Nazis are coming back to work’ (Akten zur Auswärtigen Politik, 1951a, pp. 266–282). US intelligence officials warned that the SRP might eventually come to power legally by gaining control of the Bundestag, as the pre-­war Nazis had done (Lee, 1997, 51). US Secretary of State Dean Acheson raised the question of what might happen if the SRP launched a coup d’etat (Frei, 2002, 260; Akten zur Auswärtigen Politik, 1951b). Adenauer sought to reassure the Allies that a ‘repetition of events leading to a strengthening of Nazism in the twenties’ would not happen (Frei, 2002, 260). For its part, the FCC ruled that the SRP’s programme, self-­conception and overall style, had an ‘essential affinity’ with the Nazis. Unlike other parties in Germany, ‘the SRP tried to enlist former National Socialist … collecting the

The SRP and NDP   123 particularly hard-­core individuals … not in order to gain positive forces for democracy, but to preserve and propagate National Socialist ideas’ (translation in Kommers and Miller, 2012, 288). In the ruling, the SRP was conceived as an existential threat not just to the democratic system but to the German state itself. The Court described in detail how the Nazi regime not only created a reign of fear and terror, was responsible for the violation of human dignity, defiance of international treaties, the fundamental rights of people, persecution of the opposition and genocide, but also how the regime had brought ‘disaster’ for the state. The SRP was led, the Court ruled, by the same people that allowed Hitler to ‘lead Germany into the abyss’. Securitization of SRP as a threat to the free democratic basic order To ban a party, article 21.2 of the Basic Law requires the FCC be convinced the party is threat to one of two referent objects – the ‘free democratic basic order’ or the ‘existence of the Federal Republic of Germany’ – which necessarily entails characterization of the party as ‘dangerous’ or ‘threatening’. The SRP was securitized as both kinds of threat, although the FCC made its decision on the basis of the threat it posed to the free democratic basic order. As Frei argues, the Lower Saxony elections were a turning point, strengthening the resolve of the federal government and the Allies to ban the SRP (2002, 259–260). Election results were seen in CDU circles as a ‘defeat’ for democracy (Deutschland-­Union-Dienst), with the danger that radical groups might use formal and material resources of the Basic Law to achieve their goals being seen as ‘very great’ (CSU Minister President of Bavaria, Hans Ehard), and preventive action being needed (CSU Franz Strauß) (SZ, 1951a). Two days after the election, Adenauer stated in cabinet that all means should be used to prevent the proliferation of organizations such as the SRP and that establishment of the FCC should be accelerated (Kabinettsprotokolle, 1951c). The Interior Minister’s claim that the SRP sought to undermine the free democratic basic order rested on the following arguments: the SRP claim that imprisoned former Nazi Admiral Karl Dönitz was rightful leader of Germany (FAZ, 1951a); the SRP refusal to recognize the constitutional legitimacy of the Federal Republic and vilification of its constitutional organs (FAZ, 1951a; BMI, 1951); calls by SRP chairman Fritz Dorls for followers to ‘[u]ndermine the authority of the government in Bonn whenever you can’ (BMI, 1951); and SRP efforts to discredit democratic politics, deputies, political parties and the federal government (FAZ, 1951a; see also Nagle, 1970, 20), including claims by SRP leaders that federal government politicians were ‘minions of foreign powers’ and ‘criminals and traitors’ (BMI, 1951). Evidence the SRP justified and sought to revive National Socialism included SRP leaders’ statements such as, ‘Only national socialism was suited to rebuild Germany’ or ‘It would be better for Germany if we had a man like Hitler once again’ (BMI, 1951). More fundamentally, the Interior Ministry argued realization of SRP goals would see the return of the ‘inhuman’ Nazi regime, which had ‘brought

124   The SRP and NDP unspeakable disaster and misery over all of Europe’ (BMI, 1951). The SRP was variously described as a collection of Nazis (SZ, 1951b), as ‘an enemy of the state’ (FAZ, 1951b, 1952b), and ‘as enemies of the democratic order’ (FAZ, 1951a). Threats posed by the SRP (and KPD) were framed as ‘totalitarian’ threats, with neo-­fascism and communism conceived as two equally dangerous elements of totalitarianism (Bundesregierung, 1951). The opposition SPD, supporting the ban, describing the party as ‘enemies of democracy’, arguing that allowing opponents of democracy the chance to exploit formal weaknesses of democracy was itself a threat to democracy (SPD, 1951). The Interior Minister initiated ban discussions in Cabinet, but there were disagreements predominantly among the CDU/CSU in favour of a ban and the more reluctant DP and FDP (Kabinettsprotokolle, 1951b; Frankfurter Rundschau, 1951). DP leader Hans-­Joachim von Merkatz reportedly disagreed with a statement of the Interior Minister describing the SRP as a ‘mass movement of former national socialists’, arguing that compared to Italy, neo-­fascism was unimportant in Germany (SZ, 1951d; see also DP Minister Heinrich Hellwege). In Cabinet meetings, von Merkatz (Kabinettsprotokolle, 1951a), and DP Minister, Hans-­ Christoph Seebohm, stressed political difficulties with measures against the extreme right and especially a SRP ban (Kabinettsprotokolle, 1951b). Merkatz, for instance, argued the SRP was not as bad as the communists and ought not be made martyrs (Kabinettsprotokolle, 1951a). He argued opposition from skilled speakers would probably be more effective than police measures (Kabinettsprotokolle, 1951a). DP leader Hans Mühlenfeld argued that declaring the SRP as unconstitutional was counterproductive, not least because the claims were not publically defended with facts (Mühlenfeld, 1951). There was press speculation that some of the parties governing in Bonn considered forming a government with the SRP, which the DP denied (see also Frei, 2002, 258; SZ 1951d). Nevertheless, these differences appear to have been resolved, following the SRP’s second electoral breakthrough in Bremen in October 1951, by which time the Cabinet came to regard ban proceedings as ‘urgent’ (Kabinettsprotokolle, 1951d). FDP Minister Hermann-­Eberhard Wildermuth insisted the KPD be treated in the same manner as the SRP, as did the DP’s Mühlenfeld, a proposal the Interior Minister accepted (Frei, 2002, 251–276; Niesen, 2002, 255; Kabinettsprotokolle, 1951e). The SRP rejected claims against it (Frei, 2002, 253; Dorls, 1950). Dorls admitted the SRP was deeply critical of the way the Federal Republic was founded, but that this reflected public opinion. He argued there was no proof the SRP sought to subvert civil and constitutional rights and, unlike the communists, the SRP was not an immediate danger. Dorls saw measures against the SRP as ‘political terrorism’, and that the CDU sought to ‘persecute’ the SRP because it was an electoral rival. SRP deputies in Lower Saxony said they supported the idea of the Reich but did not want to resurrect the NSDAP and declared support for a democratic state and a European community of free and equal peoples (SZ, 1951b). In meetings with the Allies, Adenauer described the Lower Saxony election as a ‘serious, but not threatening sign’ (Akten zur Auswärtigen Politik, 1951a,

The SRP and NDP   125 pp. 266–282; see also Frei, 2002, 260), but acknowledged support for the SRP might grow and become ‘very dangerous’ for the German and European ‘future’ (Akten zur Auswärtigen Politik, 1951a). US High Commissioner John McCloy responded that the Allies were ‘very worried’ by the SRP (Frei, 2002, 260) and that leading figures in Congress worried about ‘progress made by Germany on the road to genuine democracy’ (Akten zur Auswärtigen Politik, 1951b). A publication from McCloy’s office stated: While the SRP upsurge in Lower Saxony has not resulted in giving that party a share in the governmental responsibility … and while it appears that the SRP’s achievements do not herald any general advance of the ultra­ nationalist cause in the foreseeable future, it does pose a challenge which cannot safely be ignored by those in places of power and responsibility. (USHICOG, 1951, 67; see similar comments from the British, Akten zur Auswärtigen Politik, 1951a) Referring to the SRP, the British asked Adenauer to estimate when the FCC would be able to ‘accept the matter’ (Akten zur Auswärtigen Politik, 1951a). While the present danger was considered small, McCloy nevertheless insisted the SRP be dealt with (Frei, 2002, 260). According to Frei’s reading of minutes of a meeting between US High Commissioner John McCloy and Adenauer in May 1951: the Allies preferred it [the SRP] being dealt with by the West German government – its ‘enthusiasm’ would lead to ‘mastery’ of the affair. But McCloy did not refrain from informing the Chancellor point-­blank ‘that we will intervene’ if it ‘should ever prove necessary’. And he drew the British and the French directly into the threat of intervention … The ‘old Nazi adventure’ was not to be repeated. (2002, 260) An Information Bulletin from McCloy’s office publically restated this position, although noting a ‘responsible, determined attitude … by the democratic parties’ towards the SRP that the federal government intended to initiate ban proceedings (USOHCG, 1951, 67). Frei argues that ‘much speaks for the assumption that the Allies were the main reason for the West German government trying to stamp out right-­wing radicalism by banning, that otherwise they would have allowed to play itself out via economic growth’ (2002, 263). In 1952, the FCC banned the SRP as a threat to the free democratic basic order. The SRP’s disparagement of, and mode of engagement with, democratic parties made it clear the SRP aimed to eliminate other parties from political life and thereby the multiparty principle. The SRP disregarded behaviour of followers that violated human dignity, the principle of equality and the ‘appalling savagery’ of its anti-­Semitism. Rather than constitutionally guaranteed free expression or genuine political opposition, the Court argued that the SRP’s

126   The SRP and NDP ‘insults’, ‘slander’ and ‘accusation’ levelled against the government, ‘poisoned political life’, sought to shake trust in the Federal Republic and put into question the free democratic basic order as a whole. Hitler employed such a method to eliminate democracy and freedom, the Court observed. Moreover, the SRP was an authoritarian party based on the ‘Führer principle’ of absolute obedience to party leaders, which was reflected in the attempt to establish the Reichsfront (Kommers and Miller, 2012, 288–289). In light of this, the court ruled that ‘if a party’s internal organization does not correspond to democratic principles, one may generally conclude that the party seeks to impose upon the state the structural principles that it has implemented within its organization’ (translation in Kommers and Miller, 2012, 288). If the SRP won power it would, like the Nazi party, eliminate the ‘free democratic basic order’. For the Court, such a party was not deserving of the usual privileges of political parties and proscription would ‘avert the danger of an abuse of this power’ (Kommers and Miller, 2012, 287). Securitization of the SRP as a Cold War threat to West German and European Security After World War II, Germany was divided into four occupation zones administered separately by the Soviet Union and each of the Western Allies, with the goal of eventually establishing a unified state (Görtemaker, 2004; Glees, 1996, 26–58). By 1949, superpower rivalry led to the division of Germany into two separate states, the Western-­oriented Federal Republic of Germany (FRG, ‘West Germany’) and the Soviet-­dominated German Democratic Republic (DDR, ‘East Germany’). The two Germanies became ‘integral parts of the two military-­ political alliance systems that confronted each other across a palisade of tanks, missiles and barbed wire’ and ultimately the threat of nuclear war (Hyde-­Price, 1992, 153; Glees, 1996, 27). Cold War enmity, and threat of war, increased the significance of the SRP’s somewhat unlikely links with Soviet Union. As I spell out below, it underpinned an additional securitization frame justifying the SRP ban, even if this justification was not mobilized in the FCC. Germany’s pivotal position on the European Cold War front line brought the delicate issue of German re-­armament to the top of the political agenda. The US favoured remilitarization of West Germany as a matter of urgency, which eventually took place through North Atlantic Treaty Organization (Glees, 1996, 103–108; Williams, 2000, 357–359). The Western Allies additionally sought to integrate West Germany politically through economic cooperation in ventures such as the European Coal and Steel Community, an early organization that would later develop into a Europe-­wide common market and European Union. Adenaur bought into this strategy, electing to tie Germany into the Western sphere, hoping to link it to a return to full German sovereignty, even at the expense of reunification (Williams, 2000, 355–379; Glees, 1996, 98–99). As Glees observes, even if both the West and the Soviets ‘continued to hope for a single German nation’ even after the Cold War broke out, ‘the West (aided by

The SRP and NDP   127 Adenauer) always preferred a divided Germany, with a strong Federal Republic as a part-­nation, to a united but neutral Germany, which might come under the sway of the Soviet Union’ (1996, 31). Within West Germany, there was considerable and broad-­ranging opposition to remilitarization, encompassing ‘sincere, war-­weary pacifists to Social Democrats, Communists and bitter-­end Nazis’ (Lee, 1997, 73; Glees, 1996, 99, 103–108). The SRP positioned itself clearly in opposition to the Western occupying powers. SRP propaganda described Allied forces as ‘devils from the West’, whose goal was to ‘destroy humanity’ (BMI, 1951). The SRP adopted a national­neutralist stance on the remilitarization issue, entailing the preference for a united Germany possessing its own army, a rejection of both East and West and the hope that Germany could exploit tensions between them to extract concessions from both sides (Lee, 1997, 73; Frie, 2002, 280; Stöss, 1991, 25). SRP leaders met with the KPD, like-­minded but Soviet-­sanctioned groups of former Nazis in East Germany, and Russian authorities themselves (Lee, 1997, 74). The US High Commissioner considered ‘the relationship of Communism to the SRP merits special attention’ in light of their mutual preference for creation of Germany as a ‘Third Force’ between East and West (USHICOG, 1951, 65). There was widespread suspicion that the SRP received financial support from the Soviet Union (Lee, 1997, 74; USHICOG, 1951, 66; Akten zur Auswärtigen Politik 1951a, pp. 266–282; FAZ, 1952a). SRP Bundestag deputy Fritz Dorls rejected claims that the SRP sought to collaborate with the KPD and other leftist organizations, arguing that the SRP was deeply devoted to the fight against Bolshevism (Dorls, 1950). Nevertheless, the SRP’s actions underpinned a securitization frame in which the SRP was seen as a threat to the security of the Federal Republic and Western Europe more generally. The Interior Ministry cited as evidence in its case for banning the SRP Dorls’ calls to support KPD propaganda that the Bonn government’s foreign policy would lead to World War III in order to exploit ‘public anxiety’ about the Cold War (BMI, 1951). Adenauer argued in a meeting with the British foreign minister: ‘This whole development is … dangerous for Europe because relations between [SRP leader] Remer and Karlshorst, the Russians, are absolutely certain’ (Akten zur Auswärtigen Politik, 1951a, pp. 266–282). The Russians, he argued, sought to prevent use of German war potential against the Soviet Union by preventing ‘the final integration of Germany into Western Europe’ and to ‘draw Germany into the Russian sphere of interest in some form’ (Akten zur Auswärtigen Politik, 1951a, pp.  266–282, and 1951b, pp.  379–386). The neutralization strategy supported by the SRP – and others – would quickly allow Russian influence in West Germany, while the building of a ‘fifth column’– groups working to undermine West Germany from within – supported by the Russians was already underway (Akten zur Auswärtigen Politik, 1951a). In public, Adenauer, Interior Minister Lehr and the SPD argued that the SRP was damaging for Germany’s reputation abroad, obliging the government to use all constitutional means against it (SPD, 1951; FAZ, 1951a; SZ, 1951b and 1951c).

128   The SRP and NDP Despite Adenauer’s assurances that even if in the past ‘Germany had hung itself between East and West’, the Federal Government now sought integration into the Western world (Akten zur Auswärtigen Politik, 1951c), the Allies had their doubts. US High Commissioner John McCloy had informed Adenauer in July 1951 that one of the two main concerns of leading figures in the American Congress was: ‘Where is Germany in the conflict between East and West? Does it instinctively lean to the Western world or does it require special arrangements?’ (Akten zur Auswärtigen Politik, 1951b).

Failed attempts to ban the National Democratic Party The political context of the NPD ban attempts were radically different from that of the SRP ban. Germany had established itself as a consolidated and stable democracy. Following the fall of the Berlin wall, Germany had reunified and full sovereignty was restored by the 1990 Treaty on Final Settlement with Respect to Germany. Appeals by pro-­ban figures to the damaging impact of an active extreme right on Germany’s foreign policy identity lacked the urgency of such claims made in regard to the SRP in the early Cold War (e.g. FAZ, 2001e; CDU/CSU, 2000; Stigler (SPD) in DB Plenarprotokoll 14/141 p.  13805; Stoiber (CSU) Bundesrat Plenarprotokoll 756, p. 451). Nevertheless, a rise in extreme-­right violence in the post-­unification period, the radicalization of the NPD, its links with neo-­Nazi activists and associations, and virulent hostility to foreigners created ambiguity about the NPD’s rejection of political violence. This facilitated securitization of the NPD as a threat to the free democratic basic order. While acceptance of this securitization frame found the support of veto players in the initiation phase of the decision-­ process, the FCC did not accept this frame and the party escaped proscription. Neo-­Nazis, xenophobic violence and the NPD’s ambiguous orientation to violence Extreme-­right violence has long been an issue in post-­war Germany (Braunthal, 2009, 77–114; McGowan, 2006, 257; Kailitz, 2000), but an armed insurgency comparable to the IRA and ETA did not emerge. Indeed, there is a much more ambiguous and indirect relationship between acts of political violence, and in some cases terrorism, perpetrated by right-­wing extremists in Germany and political parties (e.g. Jesse, 2001, 689–691; Henkel and Lembcke, 2001, 5–6). Nevertheless, peaks in right-­wing violence or explosions of public outrage about such acts have often triggered party ban debates, including those on the NPD examined here (see, for example, Wise, 1998, 302; Jesse, 2001, 684; Flemming, 2003; Braunthal, 2009, 66; Henkel and Lembcke, 2001, 1). More importantly, the NPD’s ambiguous orientation to violence substantiated the claim that the NPD exhibited an ‘active, combative and aggressive’ attitude to the existing political order, a key criteria established in the KPD case for a party ban. Data on extreme-­right violence in Germany varies depending on the methodologies employed by different agencies (Atkinson, 1993, 154–155), and most

The SRP and NDP   129 scholars accept data is likely to underestimate its extent because victims are reluctant to report them and political motivations can be difficult to prove (Malthaner and Waldmann, 2003, 114; McGowan, 2006, 264, 267). Nevertheless, in the late 1970s and early 1980s there is evidence of a dramatic increase in both non-­violent extreme-­right offences, such as racist and anti-­semitic insults and graffiti, and violent offences committed by right-­wing extremist groups (Zimmermann and Saalfeld, 1993, 62). During this period, racist and right-­wing extremist groups committed various acts of terror, including a bombing incident during the Munich Oktoberfest in 1980 in which 13 people were killed and over 200 injured (Malthaner and Waldmann, 2003, 113; McGowan, 2006). By 1982, the number of offences (violent and non-­violent) stagnated, according to various authors, due to the proscription of various neo-­fascist groups and prosecution of various neo-­Nazi leaders and activists (Zimmermann and Saalfeld, 1993, 62; Malthaner and Waldmann, 2003, 113). Figures 5.1, 5.2 and 5.3 show the evolution of offences perpetrated by right-­wing extremists following reunification of Germany in October 1990. Figure 5.1 shows a steady increase of crimes attributed to the extreme right in the post-­unification period. Figure 5.2 shows a sharp increase of extreme-­right violence in the early 1990s, and another peak beginning in 2013. Figure 5.3 shows the number of murders attributed to the extreme right peaked in the early 1990s and then declined to relatively low levels, while attempted murders attributed to the extreme right remained relatively high (over ten per year) into the 2000s and peaked again in 2016. 25000 Crimes

20000 15000 10000 5000

19 9 19 0 9 19 1 9 19 2 9 19 3 9 19 4 9 19 5 9 19 6 9 19 7 9 19 8 9 20 9 0 20 0 0 20 1 0 20 2 0 20 3 0 20 4 0 20 5 0 20 6 0 20 7 0 20 8 0 20 9 1 20 0 1 20 1 1 20 2 1 20 3 14 20 1 20 5 16

0 Year Total number of politically motivated right-wing crimes Total number of extreme-right crimes

Figure 5.1 Politically motivated right-wing crimes and extreme-right crimes in Germany, 1990–2016. Source: Verfassungsschutzbericht (Annual report of the Federal Office for the Protection of the Constitution) 1992, 1996, 1998, 2012, 2013 and 2016. Note Data was often changed retrospectively in successive years and the most recent figures are included here. The category ‘politically motivated crimes’ was created in 2001 by Federal and Land interior ministers. The main criterion for registering such a crime is its political motivation; that is, whether it is aimed at a person based on political view, nationality, ethnicity, race, skin colour, religion, world view, origin, sexual orientation, disability, appearance or social status. The Verfassungsschutz notes that due to change in methodology, statistics before and after 2001 cannot be compared.

3000 2500 2000 1500 1000 500 0 19 9 19 0 9 19 1 9 19 2 9 19 3 9 19 4 9 19 5 9 19 6 9 19 7 9 19 8 9 20 9 0 20 0 0 20 1 0 20 2 0 20 3 0 20 4 0 20 5 0 20 6 0 20 7 0 20 8 0 20 9 0 20 0 1 20 1 1 20 2 1 20 3 1 20 4 1 20 5 16

Crimes

130   The SRP and NDP

Year

Figure 5.2  Violent crimes attributed to extreme right in Germany, 1990–2016. Source: Verfassungsschutzbericht (Annual report of the Federal Office for the Protection of the Constitution) 1992, 1996, 1998, 2012, 2013 and 2016.

25 20 15 10 5 0

19 9 19 0 9 19 1 9 19 2 9 19 3 9 19 4 9 19 5 9 19 6 9 19 7 9 19 8 9 20 9 0 20 0 0 20 1 0 20 2 0 20 3 0 20 4 0 20 5 0 20 6 0 20 7 0 20 8 0 20 9 1 20 0 1 20 1 1 20 2 1 20 3 1 20 4 1 20 5 16

Number

Note Data was often changed retrospectively in successive years and the most recent figures are included here.

Attempted murder

Year Total number of extreme-right murders

Figure 5.3 Murders and attempted murders attributed to extreme right in Germany, 1990–2016. Source: Verfassungsschutzbericht (Annual report of the Federal Office for the Protection of the Constitution) 1992, 1996, 1998, 2012, 2013 and 2016. Note Data was often changed retrospectively in successive years and the most recent figures are included here.

While in the late 1970s and early 1980s, most individuals and groups committing racist and right-­wing extremist acts were linked to extremist right-­wing and Nazi organizations, by the 1990s violence took a ‘different shape’ (Malthaner and Waldmann, 2003, 113). Violent crimes, motivated by right-­wing and racist ideologies … were committed in most cases rather spontaneously by gangs of local youths, only one-­fifth of perpetrators having contacts with right-­wing or Nazi organizations. Many of the perpetrators identified themselves with the skinhead scene and culture. (Ibid., 113; Braunthal, 2009, 98)

The SRP and NDP   131 Violence mostly took the form of ‘street assaults, riots, violent hooliganism and firebombing attacks, directed mainly against foreigners and accommodation for asylum seekers’ (ibid., 113). Acts of right extremist violence in former East Germany were more or less double that recorded in the West (Ignazi, 2003, 77; Malthaner and Waldmann, 2003, 114; McGowan, 2006, 266). Most violent crimes were committed against foreigners but other victims included the homeless, punks, leftists, gays, the disabled and Gypsies (Braunthal, 2009, 98; McGowan, 2006, 263). Other provocative acts included the desecration of Jewish cemeteries and synagogues, and certain marches and demonstrations (Braunthal, 2009, 103). In the early 2000s, during the NPD I case, incidents referred to in ban debates included hate crimes leading to the death of a Mozambican and three homeless men; a bomb at Dusseldorf station injuring ten immigrants, most of whom were Jews (McGowan, 2006, 366; FAZ, 2003b); firebomb attacks on synagogues in Erfurt and Dusseldorf (although it later became apparent that Islamists were responsible (Flemming, 2003; McGowan, 2006; FAZ, 2003b); vandalism in Jewish cemeteries (McGowan, 2006, 267; FAZ, 2003b) and the desecration of a number of Holocaust memorials, including Buchenwald (ibid.). A trigger for the NPD II case was the discovery in 2011 of the right-­wing terrorist network, Nationalsozialistischer Untergrund (NSU, National Socialist Underground), which killed nine individuals of mainly Turkish origin and a police officer over the period of around a decade, and was responsible for armed robberies and bomb attacks in Turkish neighbourhoods (McGowan, 2014, 203). The cell murdered eight small-­business owners of Turkish origin, a Greek man and a police officer, motivated by racism and xenophobia (Deutsche Welle, 2013). Charges against the group include setting up and supporting a terrorist organization. The NSU campaign was ‘the most lethal campaign of political violence’ since the FRG was founded in 1949 (McGowan, 2014, 204). NPD leader Holger Apfel insisted, following discovery of the NSU murders, that non-­violence was a non-­ negotiable maxim of his party (FAZ, 2012f ). The arrest of Ralf Wohlleben, deputy chairman of NPD in Thuringia, and another former NPD member, for supplying firearms to NSU, as well as expressions of sympathy for the NSU from some in the NPD (FAZ, 2012d, 2012f, 2012j), ignited suspicions that the NPD might be connected to the NSU (FAZ, 2012h, 2012k; Hartman (SPD) DB Plenarprotokoll, 17/237, p. 29718). Such incidents have shocked many Germans, some of whom mobilized in mass gatherings and marches against the extreme right (Malthaner and Waldmann, 2003, 121; Braunthal, 2009, 104, 185–190). Demonstrations in 2000, involving hundreds of thousands of protestors, and often referred to as the ‘uprising of the decent’ (Aufstand der Anständigen) in a reference to Chancellor Gerhard Schröder’s call under this banner for solidarity with Jews and against the far right following the October 2000 attack on a synagogue in Dusseldorf, immediately preceded initiation of the NPD I case (Rensman, 2003, 1120; Decker and Miliopoulos, 2009, 94). Since reunification, acts of extreme-­right violence, and public reactions to them, have been seen as important triggers for

132   The SRP and NDP party ban cases in Germany (Wise, 1998; Atkinson, 1993, 164; Jesse, 2001, 684; Flemming, 2003; Rensmann, 2003, 1120; Braunthal, 2009, 66; Malthaner and Waldmann, 2003; Henkel and Lembcke, 2001, 1). Indeed, some writing on the NPD I case have described the effects of political pressure and ‘state-­led’ mobilization of civil society against extreme-­right violence as a ‘prohibition trap’ (Jesse, 2001, 688; Flemming, 2003) from which a ‘hot-­headed’ political class could not easily step back to ‘soberly’ consider the consequences of proscription once the process began (see also Jesse, 2001, 687; FAZ, 2001b). While such episodes of extreme-­right violence may have facilitated entry of NPD party bans onto the political agenda, a more elaborate set of arguments, many of which pointed to NPD orienations to political violence, were developed in order to substantiate the claim that the NPD exhibited an ‘active, combative and aggressive’ attitude to the existing political order. As mentioned in Chapter 3, in its 1956 rulings on the KPD, the FCC established that a party could not be banned merely for articulating positions contrary to the free democratic basic order (McWhinney, 1957, 303; Franz, 1982, 62; von Schmertzing, 1957, 13). But it was not necessary to prove that a party posed an imminent or actual danger to the democratic system to ban it, nor would it be necessary to present evidence of a party’s ‘concrete undertaking’ to abolish the constitutional order (von Schmertzing, 1957, 14; Kommers, 1997, 223). Rather, the Constitutional Court required evidence showing an ‘intention’, or ‘fixed purpose to combat the free basic order constantly and resolutely’ manifested in ‘political action according to a fixed plan’ (Kommers, 1997, 223; von Schmertzing, 1957, 13). In other words, there would need to be evidence of a party’s ‘active, combative and aggressive’ attitude to the existing political order (McWhinney, 1957, 303). According to ban supporters, the NPD’s active, combative and aggressive attitude was predominantly evident in its transformation from an electoral party into the head of a broad social protest movement incorporating neo-­Nazis and skinheads; its mode of opposition dominated by mass demonstrations and a ‘battle for the streets’; a strategy contemplating the use of force to replace the democratic system; incitement to violence, particularly against foreigners; and perpetration of acts of violence by some NPD members. In NPD I, for example, the federal government, in its submission to the FCC and elsewhere, sought to show that the NPD had evolved from a party into the head of a broad social movement in alliance with violent neo-­Nazis and skinheads (FAZ, 2001c, 2001d; see also Schilly (SPD), Bundesrat Plenarprotokoll 756, p. 462). Many compared the ‘old’ NPD established in the 1960s with the ‘new’, more violent NPD (FAZ, 2001a; Bürsch (SPD), DB Plenarprotokoll 14/141, p.  13790; Gabriel (Lower Saxony), Bundesrat Plenarprotokoll 756, p.  452). In the Bundesrat, then Minister President for Bavaria and CSU chairman, Edmund Stoiber emphasized the evolution of the NPD from an ‘old fashioned party of extremists’ into a party which now incorporated ‘terrorist ringleaders’, and was a ‘pool and shelter for right-­wing extremist, extremely violent skinheads’ (Bundesrat Plenarprotokoll 756, pp. 448–449; see also FAZ, 2001c). Many observed that leading members of banned neo-­Nazi organizations

The SRP and NDP   133 had joined the NPD and its youth branch, the Junge Nationaldemokraten (Young National Democrats) in the early 1990s (FAZ, 2001e and 2001f; Stoiber (Bavaria), Bundesrat Plenarprotokoll 756, pp. 448–449). Various ban supporters argued NPD rallies sought to convey, in accordance with Nazi models, a martial, aggressive and frightening impression through the massive accumulation of uniformed, shaven-­headed, black-­clothed activists, carrying drums and flags (FAZ, 2000b, 2001c and 2001d). In NPD II, the Bundesrat’s submission to the FCC pointed out that collaboration between the NPD and neo-­Nazis continued. In NPD II, the Greens described the NPD as the political or parliamentary arm of a violent Nazi movement (Greens, 2013). The cultivation of links and collaboration with other actors in the German far right was an explicit element of the NPD’s ‘three’ (later ‘four’) pillar strategy adopted under the leadership of Udo Voigt (Flemming, 2003; Braunthal, 2009, 62–66). The ‘pillar strategy’, initially focused on a ‘battle for the streets’ (involving public demonstrations, often commemorating key events in the Nazi calendar), a ‘battle for the minds’ (involving winning public sympathy through propaganda and grassroots populist campaigns) and a ‘battle for the parliaments’ (involving participation in electoral contests) (Braunthal, 2009, 62–66). After 2004 a fourth, ‘battle for the organized will’, came to focus on uniting the broader extreme right under NPD leadership. It was through this strategy, various ban proponents in both cases claimed, that important evidence of the NPD’s active, combative and aggressive efforts to eliminate the free democratic basic order could be found (FAZ, 2001a, 2001c; Die Linke, 2013; Bundesrat Plenarprotokoll 756, Schilly (Federal Interior Minister), p.  462, Stoiber (Bavaria), pp.  448–449; Bundesrat Plenarprotokoll 904, Lieberknecht (Thuringia), p. 549, Tillich (Saxony), pp. 552–553). As SPD’s Ute Vogt put it, to be banned a party must not ‘only’ persecute foreigners or frighten people, but have the political strategy following the logic: ‘First we need the violence, then we conquer the streets, then we conquer the parliaments, and then we have an open path to turn politics around completely, to put an end to the system’ (FAZ, 2001a). In the NPD I case, the applicants claimed that the NPD sought to replace the liberal democratic system, by a ‘coup d’etat if necessary’, and that its supporters did not shy away from the use of force. In NPD II, the Bundesrat argued NPD leaders had spoken of ‘national revolution’ and ‘overcoming the system’, at least partially through the use of force. The NPD supported creation of ‘civil defence’ guards for the ‘defence’ of the majority against certain minorities conceived as security threats, and the state’s purported failure to provide protection for the majority was central to the NPD’s call to a revolutionary overthrow of the Basic Law, the Bundesrat claimed. In both NPD I and II ban cases, pro-­ban supporters referred to the creation of so-­called ‘national liberated zones’ – purportedly an NPD strategy – as further evidence of the NPD’s active, combative and aggressive behaviour. The zones are right-­wing enclaves, located in small towns and villages such as Jamel, Lübtheen and Anklam in East Germany, formed by the purchase of neighbouring properties, and the concentration of people from the right-­wing extremist

134   The SRP and NDP milieu (Braunthal, 2009, 89). According to NPD ban proponents, national liberated zones undermined the legitimacy of the state’s monopoly of violence, created areas of lawlessness, created a ‘climate of fear’ which limited political freedom and hindered the democratic process, and were an ‘intermediate step’ for the achievement of its longer-­term political goals (FAZ, 2001d; Greens, 2013; Özdemir (Greens), Bundesrat Plenarprotokoll 756, p.  13795; see also Oppermann (SPD), Plenarprotokoll 17/237, p. 29705). Many supporting NPD bans emphasized the fear and insecurity created for victims. In the NPD I case, for instance, the Greens’ Cem Özdemir argued that the NPD ban was ‘not about a struggle of political opinion’, but about protecting people: We owe it to the at least 93 victims of right-­wing violence, to crush the organizational infrastructure of right-­wing violence … colored people, members of religious minorities and homosexuals could no longer feel safe on the streets in the evening. (DB Plenarprotokoll 14/141, p. 13792; see also Clement (North Rhine-­Westphalia), Bundesrat Plenarprotokoll 756, p. 456) In NPD II, Stanislaw Tillich, CDU minister president of Saxony argued that the NPD provided financial and logistical resources to violent right-­wing extremists, which threatened and intimated people with differing philosophies, cultures and skin colours (Bundesrat Plenarprotokoll 904, p.  552). SPD’s Thomas Oppermann argued that ‘while democracy in Germany may be strong enough to withstand the anti-­constitutional NPD; the victims of the NPD are not’ (DB Plenarprotokoll, 17/237, p. 29705). As federal interior minister Otto Schily and others argued, by courting violent youth subcultures, and seeking to exploit them to their own ends, the ‘NPD was responsible for an intellectual climate creating the basis for violent attacks by right-­wing extremists on foreigners and other minorities in Germany’ (Bundesrat Plenarprotokoll 756, Schily, p. 462, Stoiber (Bavaria), p. 449; DB Plenarprotokoll 14/141 8, Bosbach (CDU), p. 13792, Gysi (PDS), p.  13798; FAZ, 2001e). In the NPD II case, the Bundesrat similarly argued that NPD ideology not only provided a theoretical framework but a ‘direct action maxim’, which led directly to intimidation and restriction of democratic processes. Christine Lieberknecht, CDU minister president of Thuringia, stated that NPD ideology was the ‘spiritual soil for the murderers of the NSU terrorist cell’, a sentiment others shared (Bundesrat Plenarprotokoll 904, Lieberknecht, p. 549, and Albig (Schleswig Holstein), p. 554; DB Plenarprotokoll, 17/237, Oppermann (SPD), p. 29706; Merkel in FAZ, 2012a). The NPD mobilized extensively against asylum policy, particularly since 2013. Its campaigns were characterized, according to the Bundesrat’s application in NPD II, by a defamatory ideology of hatred and harsh rhetoric denying certain people human dignity, including: the identification of asylum seekers with high rates of criminality and health threats; a rhetorical commitment to excluding Muslims by force, configured as an approval of killing refugees; and

The SRP and NDP   135 the organization of, and participation in, threatening demonstrations, vigils and torchlit marches which sometimes ended in violent clashes with asylum-­ seekers, political opponents and the police. In addition to anti-­semitism, it launched an aggressive campaign against Muslims living in Europe and the building of mosques, and launched a campaign against state support for Sinti and Roma minorities living in Germany. Acts targeting minorities and political opponents were conceived, according to the Bundesrat application, to deter political opponents from the exercise of activities in public, and the fear of violence, threats and social sanctions did in fact have the effect of impairing the democratic process in this manner. The FDP was the harshest critique of NPD ban proposals. In NPD I, the FDP noted with ‘alarm’ the increase in right-­wing extremist anti-­Semitic and xenophobic incidents in Germany (FDP, 2000; Wagner (Hesse), Bundesrat Plenarprotokoll 756, p. 461). However, it argued the real problem was ‘threats to actual people by right-­wing extremists’ which ought to be combatted by the constitutional state, the police and the criminal law rather than a measure designed to deal with an ‘exceptional danger’ to democracy posed by an extremist party (ibid.). Similarly, in NPD II, the CDU/CSU and FDP majority in parliament agreed NPD ideology helped prepare the ground for violent right-­wing extremists (CDU/CSU and FDP, 2013). The Federal Interior Minister Friedrich (CSU) acknowledged there was overlap between the NPD and violent neo-­Nazis (FAZ, 2012a). The CDU/CSU and FDP majority in parliament also acknowledged that discovery of the ‘terrible murders’ of the NSU showed the extreme right was capable of cruel acts of violence and that its violent acts threatened the freedom, security and integrity of individuals (CDU/CSU and FDP, 2013). However, Chancellor Merkel, among others, doubted there was sufficient evidence to link the NPD to the terrorist cell, the NSU (Spiegel Online, 2012a; Wolff (FDP), DB Plenarprotokoll 17/237 25, p.  29716). In the Bundesrat FDP deputy Hartfrid Wolff argued that the NPD ban debate was ‘symbolic’ (ibid.), and more generally arguments from the FPD in NPD I and the government and CDU/CSU and FDP parliamentary majority in NPD II emphasized: the difficulty of proving an ‘aggressive-­combative attitude aimed at overthrowing democracy’ (FAZ, 2012f, 2013b), while others still argued a party ban created a risk of radicalization and a ‘drift into terrorism’ (in DB Plenarprotokoll 14/141, Beck (Greens), p.  13859–13860, Beer and Bettin (Greens) p.  13860–13861). Many in FDP circles and some in the CDU argued that party bans were not effective: party bans could not fight ‘sentiments’, would not address the broader far-­right movement or reduce violence in it, could be circumvented by new parties, or would help other right-­wing parties, such as the DVU and Republicans, to present themselves as more legitimate (FAZ, 2000c, 2000d, 2000e, 2000f, 2000o, 2012l, 2013c, 2012g; Henkel and Lembcke, 2001, 3). In both cases, the NPD claimed to support the rule of law, the state’s legitimate monopoly of violence and non-­violent means to achieve political goals. Its desire to ‘revolutionize’ the political system would proceed through amending laws in a democratic manner rather than revolutionary violence, it claimed. It

136   The SRP and NDP denied involvement in violent or threatening acts, creating ‘zones of dominance’ or an ‘atmosphere of fear’. The NPD argued that its tactics were a completely normal part of democracy. Any self-­defence training of NPD members was needed for defence against left-­wing violence. In both NPD I and II cases, the party called for dismissal of proceedings because, among other things, security service operatives and informants had infiltrated the NPD leadership, ‘contaminating’ evidence against it. In NPD I, it emerged that in their evidence against the NPD, the applicants had used statements by NPD members who were simultaneously paid informants of the domestic intelligence services (Rensmann, 2003, 1121; Flemming, 2003). The applicants acknowledged that some NPD leaders at federal and Länder levels were informants, but that the number of informants was less than 15 per cent of the NPD party leadership (as measured on three specific dates) (ibid.). A minority on the FCC dismissed the application arguing that state security agencies continued to obtain information from undercover agents in leading NPD decisions bodies at federal and state levels, which violated the party’s right to a free, self-­determined process of developing its own defence before the FCC. Furthermore, the fact that the applicants did not disclose names of the undercover agents concerned meant the Court could not judge which parts of the material presented to it originated from undercover agents and which did not. A minority of judges was able to block continuation of the case in accordance with Article 15(4) of the FCC Act because, as mentioned above, such a continuation would be disadvantageous to the NPD. A majority of four judges favoured a continuation of proceedings, arguing there was no indication of any substantial government influence in the NPD’s process of will formation or defence strategy and that there was a public interest in continuing proceedings (Rensmann, 2003, 113). Having learnt its lesson, in NPD II, the Bundesrat was able to confirm no undercover agents sat on the NPD executive board after 6 December 2012; that essential evidence in its case against the NPD was not based on evidence of undercover agents; and that knowledge of the NPD’s legal defence strategy had not been obtained inappropriately. The FCC declared the case admissible. Neverthless, in NPD II the FCC rejected the Bundesrat’s arguments on the issue of violence. It argued that the activities of the party did not exceed the limits of permissible political struggle in a manner sufficient to meet Basic Law Article 21(2) demands that the party was ‘seeking’ to undermine or abolish the free democratic basic order. It was not possible to attribute to the NPD various violent, threatening and xenophobic attacks by people associated with the extreme-­right scene. Evidence did not show the NPD sought to achieve its anti-­ constitutional objectives through violence or the commission of criminal offences. ‘National liberation zones’ did not really exist and the NPD was not able to enforce dominance over territory to the extent that it excluded others from participating in the political decision-­making process. As such, it could not be inferred that the NPD had a plan to apply force against the free democratic basic order or to disregard the state’s monopoly of force. Furthermore, the FCC rejected the claim that NPD actions led to an atmosphere of anxiety capable of

The SRP and NDP   137 impairing the right to free and equitable participation in political will formation. The Court recognized that the behaviour of NPD members or supporters could be intimidating, deliberately provocative or transgress the limits of legality. NPD actions against state asylum and refugee policy, or its xenophobic images or speech, did not exceed the limits of permissibility in democratic discourses and political contestation. Such acts were not frequent enough to create an atmosphere of fear capable of affecting equal participation in democratic processes, regardless of the subjective perceptions of those targeted. The NPD’s ‘battle for the minds’ was not particularly successful at generating sympathy or support for its anti-­constitutional ideas, nor was its ‘battle for the streets’ mobilization around asylum and minority rights issues, or through efforts to associate itself to the PEGIDA movement. Its ‘struggle for the organized will’ had not succeeded in uniting ‘all national-­minded forces’ under its leadership; rather, cooperation with the party and right-­wing groups took place on a case-­by-case basis and the NPD did not have a leadership role in the broader extreme-­right movement. Securitization of the NPD as a threat to the free democratic basic order There was widespread agreement among those supporting both NPD I and II bans that NPD policy was incompatible with the Basic Law. In their submissions the applicants claimed the NPD pursued goals contrary to the value of human dignity and fundamental rights, sought to replace parliamentary democracy and the multiparty system with a non-­democratic, populist, elite-­driven system founded on exclusivist principles of ‘national community’ (Volksgemeinshaft). The anti-­Semitic, racist and xenophobic character of the NPD was widely acknowledged in debates on both NPD I and II bans (FAZ, 2000b, 2001c, 2001d; DB Plenarprotokoll 14/141, Bürsch (SPD), p.  13791, Buntenbach (Greens), p.  13804; CDU/CSU, 2000; Bundesrat Plenarprotokoll 756, Stoiber, (Bavaria), p. 449; Clement (North Rhine-­Westphalia), pp. 457–458; Bundesrat, 2012; Bundesrat Plenarprotokoll 904, Lieberknecht (Thuringia), p. 549, Wowereit (Berlin), p.  551, Tillich (Saxony), p.  552; SPD, 2013). Ban proponents argued that the NPD exhibited what the FCC described, in the SRP case, as an ‘essentially affinity’ with the NSDAP. In its NPD II submission to the FCC, the Bundesrat argued, as did others supporting an NPD ban, that the NPD showed a clear ideological affinity with the NSDAP, in terms of ideology, structure, vocabulary and symbolic references (FAZ, 2012m; Bundesrat, 2012; DB Plenarprotokoll, 904, Lieberknecht (Thuringia), p.  549, Caffier (Mecklenburg-­Western Pomerania), p.  556; Oppermann (SPD), DB Plenarprotokoll, 17/237, p.  29705; Die Linke, 2013). The NPD’s affinity with national socialism was also evident, for various ban supports, in the NPD’s political style – its ‘terminology’, ‘diction’ and public appearance – and sometimes in the literal appropriation of the wording of NSDAP policy programmes from the 1920s (e.g. Bundesrat Plenarprotokoll 756, Stoiber (Bavaria), p. 449, Clement (North Rhine-­Westphalia), pp. 457–458; DB Plenarprotokoll 14/141, Stigler (SPD), p. 13805–13806).

138   The SRP and NDP In pursuing these goals, ban proponents argued that the NPD sought to ‘undermine’ and ‘eliminate’ the free democratic basic order, constituted a significant threat to the democratic community and that, as such, it ought to be banned. In the NPD I ban case, the government argued that ‘the NPD poses a serious threat to the constitutional order in Germany’, and that pursuing the ban was ‘necessary’ to end the NPD’s anti-­constitutional agitation and its commitment to a totalitarian state and society (FAZ, 2001c and 2001d). The Greens’ Cem Özdemir emphasized the diminution of the civil rights of victims of the extreme right, who were likely to withdraw from democratic participation (Özdemir (Greens), DB Plenarprotokoll 14/141, p.  13792). In the Bundesrat, CSU’s Minister President of Bavaria, Edmund Stoiber, argued that an NPD ban would set necessary limits on the far right because ‘we are responsible for the preservation of our constitutional order and civil peace in our country, and we are responsible for the future of Germany’ (Bundesrat Plenarprotokoll, 756, p. 450). Wolfgang Clement SPD minister president of North Rhine-­Westphalia argued that the previously abstract threats against the constitutional order both in terms of democratic foundations, human rights and commitments to human dignity had now become real human rights violations and ‘this must lead us to act now’ (ibid., p. 457). Similarly, in its NPD II submission to the FCC, the Bundesrat argued that proscription was ‘appropriate and necessary for the protection of the free democratic basic order, since “milder” means are not permitted by constitutional grounds because of the [constitutional principle of] party privilege’. The Bundesrat argued an NPD party ban met the ECtHR’s requirement that bans were necessary in a democratic society due to the NPD’s revolutionary goal of overthrowing the current [democratic] system, its ethnic definition of political community and national socialist ideology, among other things. In the Bundesrat, SPD Governing Mayor of Berlin, Klaus Wowereit, argued ‘the NPD is not just any harmless club of right-­wing fruitcakes. It puts an axe to the cornerstone of our peaceful coexistence’ (Bundesrat Plenarprotokoll, 904, p. 551). The ‘dangers’ posed by the NPD were ‘concrete’, adding ‘a democracy must be on guard, a democracy must defend itself ’ (Bundesrat Plenarprotokoll 904, Wowereit (Berlin), p.  551; FAZ, 2012h and 2012m). Die Linke’s pro-­ban proposal argued a ban was ‘necessary’ in order to avert the ‘manifest danger’ posed to democracy by the NPD (Die Linke, 2013). In both ban cases, political leaders and the FCC continued to draw the lesson of the Weimar Republic that the state needed to actively oppose the ‘enemies of democracy’. In both NPD I and II, the FCC restated the view that Article 21(2) was created to prevent a repetition of the ‘catastrophe’ of national socialism and changes since then did not make it irrelevant for warding off dangers to the liberal democratic order. The Court also ruled that acts justifying a party ban need not necessarily be unlawful acts, given the lesson learnt from the Nazi tactics of ‘legal revolution’. A renunciation of a requirement for proving a concrete danger as a condition for party bans drew on lessons learned from the Weimar period, namely that ‘the more ground radical endeavours gain,

The SRP and NDP   139 the harder it is to combat them’ and that by the time a concrete danger existed it could be too late to employ a measure such as a party ban. These themes of the national socialist ‘catastrophe’, the Weimar lesson of what happens if one fails to act, the extremist ‘ruse’ of pretending to respect legality, and the obligation for democrats to act defensively and early to prevent a repetition of the Nazi past, were frequently aired justifications for banning the NPD in both ban attempts (Bundesrat Plenarprotokoll 904, Wowereit (Berlin), p.  551, Haseloff (Saxony-­Anhalt), p.  554, Caffier (Mecklenburg-­Western Pomerania), p.  557, Stoiber (Bavaria), pp.  449–450; DB Plenarprotokoll 14/141 Bürsch (SPD), p. 13791, Stiegler (SPD), p. 13805, Özdemir (Greens), p. 13794, Gregor Gysi, p. 13798–13799; DB Plenarprotokoll 17/237, Oppermann (SPD), p. 29707). Others argued for banning the NPD with reference to a special responsibility of the German state following the Holocaust. SPD Interior Minister, Otto Schily, for example, argued that it is ‘our responsibility’, in relation to those who had the courage to return Jewish religiosity and culture to Germany, to ensure that the organized anti-­Semitism and racism embodied by the NPD are not tolerated (Bundesrat Plenarprotokoll 904, p. 463, see also Clement (SPD NRW), p. 456, Stoiber (CSU Bavaria), p.  450; Stiger (SPD) in DB Plenarprotokoll 14/141, p.  13805). Minister President of Bavaria, Edmund Stoiber, argued during the Bundesrat debate on the NPD ban that the ‘memory of the events of November 9 1938 [Reichspogromnacht/Kristallnacht] reminds us of the consequences of racial hatred, anti-­Semitism and volkisch thought for human rights and human dignity’ and it ‘reminds us to take resolute action against any attempt to revive this demon again’ (Bundesrat Plenarprotokoll 904, pp.  448–449). Similarly, in the Bundesrat’s NPD II submission to the FCC, it argued that combatting the NPD’s ‘relativization of national socialist injustice’ by denying their murder of European Jews through the ban procedure had ‘a special constitutional significance’. The Central Council of Jews in Germany supported ban proceedings in NPD I, as did the representatives of 110 Foreigners Council’s meeting in November 2000, while in NPD II, the Central Council of Jews was deeply critical of the federal government, and especially the FDP, for opposing the ban (FAZ, 2000c, 2000k, 2013e). In both NPD I and II, among those that did not support initiation of party ban applications, there was also widespread recognition that the party was xenophobic, anti-­Semitic, undemocratic and anti-­constitutional (CDU/CSU and FDP, 2013; Westerwelle (FD), DB Plenarprotokoll 14/141, p. 13797; Spiegel Online, 2012a; FAZ, 2012b, 2012f ). Nevertheless, the position of the FDP was, in the words of its chairman and Vice-­Chancellor of Germany, Philipp Rösler, during NPD II, that ‘stupidity cannot be prohibited’ and must be fought politically (FAZ, 2012l and 2013a; FDP, 2000; Westerwelle (FDP), DB Plenarprotokoll 14/141, p.  13797). Federal Government (FDP) justice minister Sabine Leutheusser-­Schnarrenberger expressed the party’s fear of failure in the Courts, arguing in NPD II debates that the outcome of the NPD I case had been ‘warning enough for us’ (FAZ, 2013c, see also 2013d; Spiegel Online, 2012b). Leading members of Angela Merkel’s CDU/CSU-­FDP 2009–13 government were

140   The SRP and NDP publically sceptical about banning the NPD, including Merkel herself and Federal Interior Minister (CSU) Hans-­Peter Friedrich (FAZ, 2012b, 2012e, 2012i, 2013b, 2013c, 2013d, 2013f ). The party ban was frequently described in government circles as ‘risky’ or ‘dangerous’, although the Chancellor reportedly hoped that Bundesrat’s application would be a success (FAZ, 2013h). More specifically, arguments from the FPD in NPD I and the government and CDU/CSU and FDP parliamentary majority in NPD II emphasized the difficulty of proving an intention to eliminate the free democratic basic order (FAZ, 2012f and 2013b); the possibility that the ECtHR might overrule a FCC ban decision (FAZ, 2012b and 2012g); a fear, expressed by the Chancellor among others, that a failed ban process would be harmful for democracy or the state (FAZ, 2012b and 2012c), that even holding ban proceedings could strengthen a party by, for instance, giving it a public platform (FAZ, 2012b, 2013a and 2013b; FDP, 2000); and that political effects of a NPD ban on right-­wing sentiments and activities were ‘difficult to calculate’ (FAZ, 2012g). Doubts were also expressed about the claimed ‘threatening nature’ of the NPD, given its decline in electoral terms (FAZ, 2012e, 2012m, 2013b, 2013d; see also Jesse, 2001, 688; Henkel and Lembcke, 2001, 5, 9). In NPD I, for instance, FDP Bundestag deputy, Guido Westerwelle, argued that while a party ban would be justified in the event of a real threat to democracy, its electoral results showed ‘this threat did not exist’ and that the NPD is the least electorally successful of all right-­wing parties (DB Plenarprotokoll 14/141, p. 13796; FDP, 2000). Similarly, in NPD II, Hessian Justice Minister, Jörg-Uwe Hahn (FDP), described the NPD as a ‘nearly dead party’, a ‘collection of the lost’ and ‘a local phenomenon in Mecklenburg Vorpommern who are evil, but in no way transferable to the entire Republic’ (FAZ, 2013g). CDU Bundestag president, Norbert Lammert, argued that the idea of the NPD as an ‘an acute threat to democracy’ could not be plausibly defended given that its political influence had ‘rarely been smaller’ (FAZ, 2012g; CDU/CSU and FDP, 2013). Ban opponents also emphasized differences between the past, and the present consolidated democratic state in Germany (Jesse, 2001, 685; see also Westerwelle (FDP), DB Plenarprotokoll 14/141, p. 13796; CDU/CSU and FDP, 2013). The FDP’s Ruth Wagner argued that citizen initiatives against xenophobia and anti-­Semitism were a sign that, unlike the Weimar Republic, ‘democrats could vouch for democracy’ (Bundesrat Plenarprotokoll 756, p. 460). In addition to denying a desire to eliminate or impair the free democratic basic order, the NPD claimed in its own defence that its policy did not attack the dignity of foreigners and minorities as human beings, but was based on a call for a return to the citizenship principle of ius sanguinis. The NPD argued its concept of the people was not racially constructed, accepted the multiplicity and equality of people and peoples and that its conception of political community did not threaten individual rights. The NPD rejected accusations of anti-­semitism and affirmed the right of Islam to exist where it had been historically established. The party claimed there was no ‘essential affinity’ with the NSDAP because it did not adopt the organizational ‘leadership principle’, it rejected eugenics and

The SRP and NDP   141 social Darwinism and its programme differed from that of historic national socialism. Rejecting claims it was anti-­democratic, the NPD insisted it supported popular sovereignty and ‘democratization’ of the German political system. The NPD characterized the party ban itself as a threat to democracy, pursued by a ‘self-­appointed prohibition elite’, recruited from the established parties, which sought to dictate which political programmes, ideas and ideology could be legitimately represented. The FCC accepted the argument that in its aims and in the behaviour of its supporters, the NPD sought to eliminate the free democratic basic order. The party opposed political equality. It called for the abolition of parliamentary democracy and its replacement with an authoritarian state through a revolutionary process. It ruled the NPD’s ethnically defined concept of political community (Volksgemeinschaft) was incompatible with the constitutional guarantee of human dignity and violated the guarantee of elementary legal equality. Its policy was racist and directed towards the exclusion and contempt of foreigners, migrants, Muslims, Jews and other social groups. Party goals had an ideological affinity to national socialism, it glorified the NSDAP, employed a similar vocabulary and symbolism to that movement, and relativized its crimes including the Holocaust. However, the Court concluded there was not evidence of sufficient weight indicating the NPD could achieve its objectives and therefore ruled against banning the party. In what amounted to a significant change in the Court’s interpretation of the conditions under which a party could be banned, it ruled that while the NPD was committed to, and had a plan to, eliminate the current constitutional order – a plan it was executing through action that could be considered qualified preparation for the elimination of the liberal democratic order – it nevertheless lacked the capacity and social support needed to implement its unconstitutional goals. The NPD’s limited successes in elections over five decades showed there was no prospect of it winning an election and other parties’ refusal to form coalitions with it or even cooperate with it selectively meant there was no prospect of it governing in the foreseeable future. Its membership was low and the party suffered persistent crises, such as interparty squabbles and financial problems. In many parts of Germany, the organizational capacity of the party severely limited its campaigning capacity and its influence in society more generally was limited. In short, the NPD II ban attempt failed because the Court did not accept a securitization frame.

Conclusion The objective of this chapter was to examine hypotheses focusing on the role of securitization, positions of veto players and parties’ orientation to violence in party ban and failed party ban decisions. Empirical analysis confirmed theoretical expectations. In the SRP ban case, following pressure from the Occupation authorities, Konrad Adenauer’s CDU-­led coalition government agreed to initiate ban proceedings and soon afterwards the FCC banned the party (H2). The SRP’s neo-­Nazi ideology, political style and creation of the quasi-­military Reichsfront

142   The SRP and NDP created ambiguity about the party’s commitment to non-­violent political action (H3) and facilitated securitization of the SRP as an existential threat to the democratic order capable of reviving the ‘catastrophe’ of Nazism (H1). Additionally, contacts with Soviet agents in the emerging Cold War facilitated securitization of the SRP as a threat to the Federal Republic and the Western Alliance (H1). The NPD ban attempts failed because one of the conditions under which party bans are expected to occur did not hold. In both the NPD I and II cases, party ban attempts failed because not all veto-­players – and ultimately the FCC – did not think the party should be banned (H2 disconfirmed). While the FCC dismissed NPD I on procedural grounds, in NPD II the Court accepted the party was anti-­democratic, anti-­semitic and racist, but rejected arguments that the NPD was an existential threat because it was unlikely to achieve its objectives. This was despite ambiguity on the part of the NPD about the appropriateness of political violence for pursuing political goals (H3) and efforts by partisan veto players to securitize the NPD as a fundamental threat to the democratic system and rights and physical integrity of others (H1).

Note 1 Parliamentary debates in the Bundestag referred to in this chapter are recorded in Deutscher Bundestag Plenarprotokoll (henceforth, DB Plenarprotkoll) and can be found online at: www.bundestag.de/dokumente/protokolle/plenarprotokolle. Parliamentary debates in the Bundesrat referred to in this chapter are recorded in Bundesrat Plenarprotokoll which can be found online at: www.bundesrat.de/DE/service/archiv/ pl-­protokoll-archiv/pl-­protokoll-archiv-­node.html

6 Electoral systems as an alternative to party bans

Manipulation of electoral rules may be a particularly potent tool for marginalizing anti-­system parties. Effects of electoral system change in France illustrate this point well. France’s 1986 experiment with a form of proportional representation electoral system permitted the National Front to win 35 seats in the National Assembly with 9.9 per cent of the vote. Two years later, a vote share of 9.8 per cent yielded just one seat under the majoritarian two ballot system (Downs, 2012, 65–66: see also Hainsworth, 2008, 121; Norris, 2005, 107). More generally, Norris has shown that radical right parties are more than twice as successful in gaining seats under proportional representation as under majoritarian electoral systems and that legal thresholds also exerted a critical effect on legislative representation (2005, 114 and 121; see also Carter, 2005). In light of such insights, various authors have argued that it may be necessary for democracies to ban parties where electoral systems are less effective at marginalizing anti-­system parties; or, in other words, party bans may not be necessary where electoral systems are more effective at marginalizing anti-­system parties. It has been argued, for instance, that party bans are necessary in countries like Israel – which uses a proportional representation electoral formula, a single state-­wide electoral constituency and low legal threshold (1.5 per cent) for parliamentary representation – but less so in the United States, with its plurality, single-­member district system, or Germany, which employs a 5 per cent legal threshold for parliamentary representation (Navot, 2008, 747; Gordon, 1987, 395; Pedahzur, 2004, 118; Michael and Minkenberg, 2007, 1119). Others have observed that limited recourse to proscription in the United Kingdom (other than NI) may be due to the electoral system, including the difficulties small parties face entering parliament and incentives against factionalism within mainstream parties (Eatwell, 2000, 187; Fox and Nolte, 2000, 21–22; Downs 2012, 66). More generally, in their study of why some democracies use stronger repressive measures against racist associations (including parties) than others, Bleich and Lambert argue that electoral barriers, particularly the existence of plurality electoral formulas or at least a 5 per cent legal threshold reduce the degree of state ‘repression’ of racist organizations (2013, 136). Downs argued that while ‘bans fail in their intended purpose; greater effect is achieved through imposition of electoral thresholds’ (2012, 23; see also Backes, 2006, 281).

144   Electoral systems as an alternative In this chapter, I examine electoral rules in Spain, the United Kingdom and Germany and their effectiveness at marginalizing anti-­system parties in order to address the hypothesis that: H4. Democracies ban anti-­system parties if alternative forms of marginalization are not effective. I focus principally on two kinds of electoral system effects in this chapter. The first are obstacles affecting entry of anti-­system parties into the parliamentary arena. The second concerns the relative marginalization (and thus influence) of anti-­system parties if they do win seats. In Chapter 7, I address the hypothesis stated above (H4) from a different perspective, namely the strategies mainstream parties deploy against anti-­system parties that actually win seats, and specifically whether they systematically exclude the party from governing and legislative coalitions as an alternative to banning it. In this chapter, I focus principally on some of the key main dimensions of electoral systems – electoral formulas and electoral thresholds, both ‘natural’ and ‘legal’ – as well as advantage ratios comparing vote and seat shares for individual parties. That is not to say that other electoral rules, such as rules on party financing, compulsory voting, access to broadcast media and ballot access are not important for anti-­system party electoral success (Norris, 2005; Carter, 2005; Bale, 2007, 114). Rather, I focus on the barriers for entry into the parliamentary arena and underrepresentation of small parties because these provide the most straightforward way of conceiving of electoral systems as alternatives to proscription. In short, if anti-­system parties struggle to obtain any, or only a few seats, in parliament, despite their electoral achievements, they are less likely to threaten democratic institutions, practices and values. I begin with a brief overview of the theoretical literature on the political consequences of electoral systems. Here I focus on electoral system effects on small parties, which is appropriate given that anti-­system parties are generally small parties (Norris, 2005; Carter, 2005). I then turn to examine electoral systems in the three country case studies and how effectively they have marginalized the anti-­system parties of interest in this book.

Electoral systems and the marginalization of small parties Electoral systems are the set of rules for translating citizen’s votes into seats in the process of electing politicians into office (Farrell, 2011, 4). Since publication of Duverger’s (1954) Political parties and Douglas Rae’s (1971) Political consequences of electoral laws, an extensive field of inquiry investigating electoral system effects on the political system has emerged. Among other things, this literature has focused on electoral system effects on proportionality, party systems, government stability, conflict management and the success of extremist parties.

Electoral systems as an alternative   145 Duverger (1954) distinguishes between ‘mechanical’ and ‘psychological’ effects of electoral systems. Mechanical effects pertain to the vote–seat relationship, or more specifically the way electoral rules systematically underrepresent some, usually smaller, parties in the share of legislative seats compared to their share of votes (Blais and Carty, 1991, 79; Taagepera and Shugart, 1989, 64). They are mechanical in the sense that no human manipulation or strategy is involved in the translation of votes to seats (once, of course, prior to an election, electoral rules have been set). Psychological effects refer to the incentives inherent in different electoral systems that may prompt voters and party elites to change their behaviour. Elites may choose not to run in constituencies they cannot win, while voters, ‘realizing that votes for minor parties are not effectively translated into seats’, may choose to vote for the least unacceptable other party likely to win a seat (Blais and Carty, 1991, 80). The proportionality of an electoral system, or the difference between parties’ seat and vote shares, measures mechanical effects of electoral systems and provides information about the ease with which small parties can enter the parliamentary arena – and thus their chances of influencing legislation and/or participating in government. While there has been some debate about the best way of measuring proportionality, Gallagher’s Least Squares Index of disproportionality tends to be the most commonly used. It is calculated using the formula:

where si is each party’s percentage of seats and where vi is each party’s percentage of votes (Gallagher and Mitchell, 2005, 603). High values on the index indicate that larger parties benefit from a strong mechanical effect at the expense of smaller ones, while conversely, low values suggest parties’ seat and vote share are similar. Table 6.1 shows average disproportionality at state and regional levels in the UK, Spain and Germany. Averaging disproportionality scores over such long time-­periods hides the effects of electoral system changes discussed in more detail below. Nevertheless, Table 6.1 points to broad tendencies in terms of electoral system effects in territories where electoral system reforms have tended to be relatively infrequent and generally confined to alterations that do not move the system into a new class (i.e. between majoritarian, proportional and/or mixed systems). From this data, it is apparent that electoral systems in the United Kingdom have had a significant bias towards larger parties. The Spanish electoral system favours large parties less than those in the UK but to a greater extent than electoral systems in the Basque Country and Navarre or in Germany at the federal and at the regional levels (with a few early exceptions). Efforts to account for varying effects of electoral systems on proportionality have generally focused on the effect of electoral formulas, district magnitude and electoral thresholds.

Table 6.1  Average electoral system disproportionality at state and regional levels in the UK, Spain and Germany State/region

Period

Electoral system

Average disproportionality (Least Squares Index)

UK (NI seats) Northern Ireland

1948–2015 1921–25 1929–72 1973 1975 1982 1996 1998– 1977–2015 1980–2012 1979–2015 1949–2013 1952–2011 1950–54 1958–2011 1950–2013 1990–2014 1951–2015 1950 1954–2013

Single member plurality Single transferable vote Single member plurality Single transferable vote Single transferable vote Single transferable vote PR (closed list), with additional seats for ten most voted parties Single transferable vote PR (closed list) PR (closed list) PR (closed list) Mixed member proportional Mixed member proportional PR (closed list) Mixed member proportional Mixed member proportional Mixed member proportional PR Losers and Surplus1 Mixed member proportional

18.10 7.71 17.53 3.48 4.50 5.96 3.8 3.8 7.34 3.23 3.10 2.94 2.99 5.62 3.31 5.25 4.13 4.03 11.69 2.4

Spain Basque Country Navarre Germany Baden Württemberg Berlin Bavaria Brandenburg Bremen Hesse

Hamburg Lower Saxony Mecklenburg-Vorpommern North Rhine-Westphalia Rhineland-Palatinate Saarland Saxony Saxony-Anhalt Schleswig-Holstein Thuringia

1949–53 1956–2015 1951–2013 1990–2011 1950–2012 1951–87 1991–2011 1952–2012 1990–2014 1990–2011 1950 1954–2012 1990–2014

Losers and Surplus PR Mixed member proportional Mixed member proportional Mixed member proportional PR Mixed member proportional PR Mixed member proportional Mixed member proportional Losers and Surplus Mixed member proportional Mixed member proportional

6.37 4.05 3.22 5.31 2.78 3.97 3.60 4.74 5.41 4.87 3.65 5.48

Notes Data on seat and vote share was drawn from the Elections and Parties in Europe database (www.parties-and-elections.eu), although where the ‘others’ category were more than 5 per cent, additional data was sought from electoral authorities or relevant interior ministry databases. Following Gallagher and Mitchell (2005, 603–605), party scores included in ‘other’ categories have been excluded from calculations, if they were below 5 per cent of total votes. 1 ‘Losers and Surplus’ method is similar to mixed member proportional systems, with the crucial difference that ‘reserve list seats were distributed [by PR, d’Hondt] without consideration of the number of seats already won in constituencies (Massicotte 2003, 3–4).

148   Electoral systems as an alternative Electoral formulas are ‘the counting rules which apply to a given electoral system’ (Farrell, 2011, 239). There are three main types of electoral formulas, each with a variety of subtypes, namely majoritarian electoral formulas (including plurality, double-­ballot and alternative vote); proportional representation (including open and closed list systems, largest remainder and highest averages formulas and single transferable vote) and mixed systems (combining elements of both majoritarian and proportional formulas). It is widely acknowledged that different electoral formulas have differing effects on proportionality and small party access to parliamentary arena. ‘[M]ajoritarian electoral systems are inherently unfavourable for small parties’ (Lijphart, 1994, 21). Unless their support is geographically concentrated, it is hard for small parties to gain representation in majoritarian systems because they need to win majorities or pluralities of the vote in an electoral district. PR (proportional representation) systems tend to do better with regard to proportionality and small party access to the parliamentary arena, although different PR-­subtypes may have varying effects on small party access to the electoral arena (Lijphart, 1994, 21–25; Farrell, 2011, 157). For instance, the d’Hondt highest averages or divisor formula used in party list systems is widely regarded as the least proportional electoral formula in the PR electoral family, and as systematically favouring large parties. On the other hand, the Hare quota or largest remainders system used in party list systems are widely regarded as highly proportional and impartial between small and large parties. Electoral formulas are also likely to have varying psychological effects, with voters less likely to ‘waste their vote’ by voting for smaller parties in plurality systems compared to proportional systems (Blais and Carty, 1991, 83, 89). District magnitude refers to the size of the constituency, measured in terms of the number of seats to be filled. There is wide agreement that ‘district magnitude has the greatest effect on the overall proportionality’ (Farrell, 2011, 6). The effect of district magnitude works differently in majoritarian and PR systems: Where majoritarian formulas are used, increases in district magnitude lead to lower levels of proportionality; where proportional formulas are used, increases in district magnitude bring higher levels of proportionality (Lijphart, 1994, 11, 20; Gallagher and Mitchell, 2005, 614). One of the most obvious barriers preventing small parties entering the parliamentary arena is the creation of electoral thresholds. Two types of thresholds are distinguished in the literature, ‘legal’ and ‘effective’ thresholds. Legal thresholds stipulate a minimum level of support (e.g. number or percentage of votes) that a party needs to win (e.g. at national, regional or electoral district levels) in order to obtain seats. Pippa Norris has argued that low legal thresholds, such as those used in Israel and the Netherlands, positively affect the share of seats won by radical right parties (2005, 122). Effective thresholds, on the other hand, are implicit or ‘natural’ thresholds that, in the absence of specific legal thresholds, nevertheless determine the minimum vote a party needs to win in order to be represented. Effective thresholds can be calculated at the level of the electoral district level (e.g. Lijphart, 1994), or of the assembly as a whole (usually referred to as the national level) (Taagepera, 1998 and 2002). The effective threshold is largely determined

Electoral systems as an alternative   149 by features of the electoral system, especially district magnitude and electoral formula, but also by factors such the number of parties that compete (Lijphart, 1994, 11; Gallagher and Mitchell , 2005, 607–608). At district level, the effective threshold is defined as approximately the mid-­way range between the threshold of representation, or ‘the lowest share of the vote with which a party might win a seat, at which its probability of winning a seat becomes almost imperceptibly greater than zero’, and the threshold of exclusion, or the highest share of the vote with which a party might fail to win a seat, at which its probability of winning a seat falls almost imperceptibly below one (Lijphart, 1994, 25; Gallagher and Mitchell, 2005, 607). A widely used formula for calculating effective thresholds is 75%/(M + 1) (Gallagher and Mitchell, 2005, 607). In this formula, M is the average district magnitude, calculated by dividing the total number of seats in the legislature by the number of constituencies. While the formula for calculating effective thresholds is useful for PR systems, it overestimates the threshold for majority and plurality systems (Lijphart, 1994, 28). A more realistic threshold estimate of between 35 per cent of the vote (Lijphart, 1994, 28; Norris, 2005, 110–111) or 37.5 per cent (Taagepera, 1998; Carter, 2005, 154) are usually employed. While the district level effective electoral threshold is expressed as a percentage, it is nevertheless important to remember that it represents a range of possibilities between the representation and exclusion thresholds. The legal and the effective threshold may not be the same and in such cases, whichever is the higher is considered the effective threshold. Effective electoral thresholds calculated at the assembly level, whether for state-­level or regional assemblies, are of particular interest because they help determine how large a party’s vote must be to achieve minimal representation in an assembly. Following Taagepera, the national electoral threshold is the vote level at which parties have a 50–50 chance to win their first seat (2002, 382). Its calculation takes into account the average district magnitude (M), total assembly size (S) and number of electoral districts (E) and can best be estimated by the following formula T = 75%/((M + 1)*√E) or alternatively T = 75%/(((S/E) + 1)*√E). As Gallagher and Mitchell point out, while the assembly level electoral threshold may be useful for determining the threshold of representation, it is not particularly helpful for understanding the impact of electoral thresholds on disproportionality and the number of parties making it into parliament (2005, 610). Nevertheless, it can in theory be expected that higher effective thresholds will restrict opportunities for small extremist parties, compared to those systems with lower effective thresholds, even if empirical studies on the far right suggest this is not always the case (Norris, 2005, 120–121; Carter, 2005, 154–155).

150   Electoral systems as an alternative In order to examine the effectiveness of electoral systems as alternatives to banning anti-­system parties, I examine the mechanical effects of electoral systems, particularly rules establishing electoral formulas, district magnitude and ‘legal’ electoral thresholds on the marginalization of HB and its successors, Sinn Féin and Republican Clubs, the SRP and the NPD. The effectiveness of electoral systems as an alternative to party bans can be measured in a number of ways. The toughest test, implied by arguments that proscription of anti-­system parties may be more necessary in some countries because their electoral systems successfully marginalize anti-­system parties, would be whether electoral rules successfully prevent these anti-­system parties from winning any seats at all. In this regard, it is more straightforward to identify electoral system effects on individual anti-­system parties than others. It is simplest in relation to legal thresholds, often used in the cases under consideration here. Where a party wins enough votes to pass an assembly-­level ‘natural’ threshold, but not a ‘legal’ threshold, the effect of the latter is readily apparent. In other cases, judgements of electoral system effects on individual anti-­system parties must be extrapolated from more general knowledge about the effects of varying electoral formulas and other parameters such as assembly size. Care is also needed to distinguish between lack of voter support and electoral system effects as such because parties may fail to win seats for the simple reason that they are not popular enough. In order to judge whether a small anti-­system party fails to win seats due to the electoral system or simply a lack of popularity, electoral system effects can be reasonably inferred by comparing the likelihood the party in question would win seats in alternative, more permissive electoral systems employed in the state. On its own, however, this test is not sufficiently nuanced for evaluating how successfully electoral systems may marginalize anti-­system parties. A cursory glance at electoral results of the parties examined in this book shows that many of the anti-­system parties of interest have often won seats, especially if elections at multiple territorial arenas are taken into account. Even when anti-­system parties are successful enough to win seats, they may still be marginalized by electoral systems, which may thereby serve as effective alternatives to party bans. In the first place, they may do so if electoral rules often, if not always, prevent anti-­system parties from winning seats. It is difficult to establish, a priori, an objective threshold at which it could be said electoral systems prevent anti-­system parties winning seats often enough to serve as an effective alternative to party bans. However, a reasonable indicator would be that electoral systems prevent anti-­system parties winning seats more often than not. Second, once they win seats, electoral systems may ensure that anti-­system parties are underrepresented in terms of the seats they win relative to their votes. This form of underrepresentation can be captured by examining an anti-­system party’s ‘advantage ratio’ (A), determined by dividing a party’s percentage of seats by its percentage of votes (A = (% seats)/(% votes) (Taagepera and Shugart, 1989, 68). As Taagepera and Shugart explain, if no seats are obtained, A = 0; if the party obtains seats but less than its proportional share, A is between 0 and 1; if the party obtains more than its proportional share of seats, A is above 1.

Electoral systems as an alternative   151 In sum, electoral systems can be seen as effective alternatives to party bans if they (1) prevent anti-­system parties from winning any seats at all, if they might have won them under more favourable electoral rules; (2) prevent anti-­system parties from winning seats more often than not; and (3) systematically underrepresent anti-­system parties in terms of their seat-­to-vote share. In such circumstances, it is expected that decision-­makers and citizens will reason that electoral systems marginalize anti-­system parties to an extent sufficient to make the option of banning them unnecessary. The following discussion focuses on electoral systems at regional and national/federal levels in each of the three case study countries, given that these territorial levels of governances are in each case the site of important political power. Although some of the parties examined in this book have obtained seats in the European Parliament, which has over time become significantly more powerful in the context of European Union governance, it will not be the focus of attention in this chapter given the very partial influence a small anti-­system party may obtain by winning seats, in this parliamentary arena.

Electoral systems in Spain In Spain, both the lower house, the Congress of Deputies, and the supposedly territorially-based upper chamber, the Senate, consist of directly elected members, although in the case of the Senate, some members are also appointed by regional units of government, the Autonomous Communities. Examination of electoral system effects, however, typically focus on the Congress of Deputies, given the weakness of the Senate, both in terms of its formal powers and tendency to reproduce the political complexion of the lower house. The fundamental features of the electoral system where set during the course of the transition to democracy and negotiation of the 1978 constitution and later incorporated into Ley Orgánica 5/1985, de 19 de junio, del Régimen Electoral General (Organic Law on the General Election Regime, LOREG) (Gunther et al., 2004, 254). Electoral rules reflected concerns of the elite to both avoid excessive fragmentation of the party system, bolster legitimacy of the post-­ Franco regime and represent the interests of significant political and social groups, including those of Francoist reformers themselves (Gunther, 1989, 837; Hopkins, 2005, 375–357). Electoral laws substantially restrict proportionality in Spain, such that its effects are often comparable to that of single-­member plurality systems despite the formal adoption of proportional representation. For the Congress of Deputies, a closed party list system is employed, where voters in each of the 50 mainland provinces and in each of the enclaves Ceuta and Melilla select a single party list, but cannot select among candidates on that list. Within each province a 3 per cent legal threshold applies. However, this threshold has largely been irrelevant outside the two largest districts, Madrid and Barcelona, because it is almost always lower than the effective threshold at the district level (Penadés and Santiuste, 2013, 95; Gunther and Montero, 2009, 106). Seats are distributed at

152   Electoral systems as an alternative the provincial level according to the d’Hondt formula. As mentioned above, this formula systematically favours larger parties and, combined with the large number of relatively small constituencies, its effect on disproportionality can be strong (Hopkins, 2005, 378). Most scholars identify malapportionment as the defining features of the Spanish electoral system. Malapportionment involves ‘awarding some areas of a country more seats in relation to population than others’ (Gallagher and Mitchell, 2005, 14). There are large demographic imbalances among the 52 electoral districts, due in part to the selection of the administrative district, the province, as the basis for the electoral constituencies. Each province is allocated a minimum of two seats (except Ceuta and Melilla which have one seat each), while the remaining 248 seats are allocated with reference to population density. The result is that district magnitude has varied between 1 and 35 seats per province, creating significant differences in terms of proportionality among districts. In the 2015 general election, 76,986 voters in the province of Soria were entitled to select two deputies, while 4,913,893 voters were entitled to select 36 deputies in Madrid. In the event, it took 26,242 votes to elect a deputy in Soria, but 101,170 votes to elect a deputy in Madrid in those elections. These rules systematically favour the larger state-­wide parties at the expense of small parties, except where support for regionalist or minority nationalist parties is geographically concentrated (Gunther, 1989; Lago and Lago-­Peñas, 2000, 232; Penadés and Santiuste, 2013). Indeed, the relatively large number of provinces with small district magnitude means small state-­wide parties can win some representation in the largest provinces, Madrid and Barcelona, but stand little chance in most other districts (Hopkins, 2005, 379). It has also systematically favoured the centre-­ right parties, the UCD and the PP over the PSOE (Penadés and Santiuste, 2013, 91; Gunther, 1989). Large state-­wide parties are not the only beneficiaries of built-­in distortions of the electoral system. It has also favoured strong regionalist and minority nationalist parties with territorially concentrated support. If measured using Taagepera and Shugart’s (1989) ‘advantage ratio’, the Basque nationalists of the PNV, for instance, have been advantaged at every general election since 1977 (Gunther and Montero, 2009, 110). There are many similarities in the electoral systems adopted at state and Autonomous Community (regional) levels in Spain, such as the adoption of proportional representation systems with closed party lists and use of the d’Hondt formula to allocate seats among parties. For elections to the Basque parliament three electoral constituencies corresponding to the Historic Territories of Álava, Vizcaya and Guipúzcoa were established and allocated an equal number of seats: initially 20 and later 25 per Historic Territory. There are significant differences in population between these territories. The smallest, Álava, with an electorate in 2012 of 250,994 has an equal number of seats in the Basque parliament as the larger Guipúzcoa, with an electorate of 575,935, and the much larger Vizcaya, with an electorate of 948, 422 (Pérez Castaños, 2013, 44). Initially, parties had to obtain more than 3 per cent of valid votes in an Historic Territory to be

Electoral systems as an alternative   153 entitled to take up seats in the Basque parliament, although this electoral threshold was increased to 5 per cent in 1983 and then in 2000 revised back down to 3 per cent. The electoral system employed in Navarre is a closed list PR system using the d’Hondt rule to distribute seats. For the first elections in the post-­Franco period in Navarre, 70 deputies from six constituencies were elected, although a single electoral constituency for the autonomous community as a whole was soon established and the number of deputies was reduced to 50 (Oliver Araujo, 2011, 277). A legal threshold of 5 per cent of valid votes was temporarily introduced between 1983 and 1986, after which it was reduced to 3 per cent. While legal thresholds have had limited effect at the state level and in other autonomous communities where district magnitude is small, in Navarre the reduction of the threshold to 3 per cent has had more significant effects (Oliver Ararujo, 2011, 365). Effectiveness of electoral system in marginalizing Herri Batasuna and its successors Electoral systems employed for the Spanish, Basque and Navarrese parliaments have not generally been an effective alternative to proscription. The majoritarian biases of the PR closed list system and use of the d’Hondt highest averages formula employed for the Congress of Deputies are reflected in a relatively high score on Gallagher’s disproportionality index (average LSq = 7.34). The electoral system at this territorial level has an average district magnitude of 6.73 and a relatively low effective national electoral threshold of 1.35 per cent (see Table 6.1 and Appendix 1). Despite advantages for large parties, the electoral system has not prevented HB and its successors from always obtaining seats when they chose to contest elections, due to the geographically concentrated nature of its support base and the irrelevance of the 3 per cent district-­level electoral threshold. HB and successors therefore obtained between 7 and 2 seats between 1979 and 2015 (see Table 6.2). Penadés and Santiuste have observed that the electoral system employed for the Congress of Deputies sometimes worked to the advantage of smaller radical nationalist parties like HB (2013, 104; see also Lago and Lago-­Peñas, 2000, 228). As the advantage ratios presented in Table 6.2 shows, HB has more often than not been underrepresented in terms of the percentage of seats obtained given its share of votes, but on three occasions it has obtained a higher share of seats than its vote share. The electoral system employed for Basque parliamentary elections – i.e. closed list PR system using the d’Hondt highest averages formula but with a much higher district magnitude (ranging from 20 to 25) – produces a much more proportionate distribution of seats (average LSq = 3.23) (see Table 6.1 and Appendix 1). In this institutional context, HB and its successors have always been able to win a significant portion of seats when contesting elections. They have won between 14 and 7 seats, or between 9.33 and 28 per cent of seats between 1980 and 2015 (see Table 6.2). The district level legal threshold of

Table 6.2  Electoral system effects: HB and successors, Congress of Deputies, Basque and Navarrese parliaments, 1977–2015 Election year

Name

Congress of Deputies

1979 1982 1986 1989 1993 1996 2011 2015 2016

Herri Batasuna Herri Batasuna Herri Batasuna Herri Batasuna Herri Batasuna Herri Batasuna Amaiur EH Bildu EH Bildu

Basque Parliament

1980 1984 1986 1990 1994 1998 2001 2005 2012 2016

Navarrese Parliament

1979 1983 1987 1991 1995 1999 2011 2015

% votes

No. seats % seats Theoretical national electoral threshold

Advantage ratio

1.00 1.00 1.20 1.10 0.90 0.70 1.40 0.90 0.77

3 2 5 4 2 2 7 2 2

0.86 0.57 1.43 1.14 0.57 0.57 2 0.57 0.57

1.35 1.35 1.35 1.35 1.35 1.35 1.35 1.35 1.35

0.86 0.57 1.19 1.04 0.63 0.81 1.43 0.63 0.74

Herri Batasuna Herri Batasuna Herri Batasuna Herri Batasuna Herri Batasuna Euskal Herritarrok Euskal Herritarrok Communist Party of Basque Territories EH Bildu EH Bildu

16.60 14.60 17.50 18.30 16.00 17.90 10.10 12.40 25.00 21.13

11 11 13 13 11 14 7 9 21 18

18.3 14.66 17.33 17.33 14.66 18.66 9.33 12 28 24

2.06 1.67 1.67 1.67 1.67 1.67 1.67 1.67 1.67 1.67

1.10 1.00 0.99 0.95 0.92 1.04 0.92 0.97 1.12 1.14

Herri Batasuna Herri Batasuna Herri Batasuna Herri Batasuna Herri Batasuna Euskal Herritarrok Bildu EH Bildu

11.10 10.60 13.70 11.40 9.40 16.00 13.60 14.60

9 6 7 6 5 8 7 8

12.86 12 14 12 10 16 14 16

1.06 1.47 1.47 1.47 1.47 1.47 1.47 1.47

1.16 1.13 1.02 1.05 1.06 1.00 1.03 1.10

Electoral systems as an alternative   155 (ranging between 3 to 5 per cent of votes) has not prevented these relatively large parties or coalitions from winning seats (see Table 6.2). In this relatively proportional system, HB and its successors have usually won a share of seats similar to their share of votes, and on four occasions since the first Basque parliament elections, their advantage ratio shows they were overrepresented in terms of seats won (see Table 6.2). Electoral system effects on the fortunes of HB and its successors are similar in Navarre. A combination of a PR closed list system using the d’Hondt formula and use of a single electoral district since 1983 (average district magnitude of 50) contributes to a relatively high level of proportionality in terms of vote–seat ratio (average LSq = 3.1) (see Tables 6.1 and Appendix 1). HB and its successors have always been able to win a significant portion of seats in Navarre. Between 1979 and 2015, they have won between 9 and 5 seats, and between 12.8 and 10 per cent of total seats (see Table 6.2). The district level legal threshold (ranging between 3 and 5 per cent of votes) and low effective electoral thresholds (at assembly level 1.06 per cent in 1979 and 1.47 per cent in 1983–2015) have not prevented these relatively large parties or coalitions from winning seats. HB and its successors have usually been overrepresented in terms of their seat-­vote ratio (see Table 6.2). In sum, electoral systems in Spain have not been effective tools for marginalizing HB and its successors. Obstacles affecting the entry of anti-­system parties into the electoral arena have not prevented HB winning seats in Spanish, Basque or Navarrese parliaments when they chose, or were permitted to, participate in electoral contests. Furthermore, HB and its successors have been overrepresented in those parliaments more often than not (16 out of 27 elections).

Electoral systems in the United Kingdom Since 1948, the UK has used the single-­member plurality electoral system to elect members to the House of Commons (see Appendix 1). The plurality electoral formula requires that a candidate win the most votes (but not a majority) in a constituency to win a seat in parliament. The district magnitude for UK elections is 1: The territory of the state is divided into constituencies that have ranged between 625 and 659 seats since the end of World War II, each of which elect a single member of parliament. One of the most obvious mechanical effects of the electoral system is the advantage accruing to large parties in terms of the allocation of seats (Mitchell, 2005, 167). Between 1945 and 1970, the Conservative and Labour parties combined won, on average, 91 per cent of votes and 98 per cent of seats, while between 1974 to 2001, following Liberal successes, the two parties’ combined vote share fell to around 75 per cent on average and 92 per cent of seats (ibid.). More recently, between 2005 and 2015, the combined Conservative and Labour vote share has fallen to around 67 per cent, while their seat share has remained on average 84 per cent. In order to minimize the problem of malapportionment, a permanent Boundary Commission was established in 1944 with responsibility

156   Electoral systems as an alternative to periodically review constituency boundaries (Carstairs, 1980, 197). The UK-­ wide parties tend not to participate in NI elections and when they do, they make little impact (Robinson, 2010, 175). Smaller parties with a dispersed vote share have been the main losers from mechanical effects of the electoral system, particularly the Liberals/Liberal Democrats: Between 1974 and 2015, their 19 per cent of the vote won only 4.6 per cent of seats on average. The United Kingdom Independence Party and the Green Party each won a single seat for the first time in the 2015 elections with 12.6 and 3.8 per cent of the vote respectively. Small parties with a geographically concentrated support base higher than the electoral threshold – such as minority nationalist parties in Scotland and Wales – tend to win a more proportionate share of seats, but are still often disadvantaged in terms of the seat to vote ratio. In NI, there have been important differences in the advantage ratios of unionist and nationalist parties (O’Leary, 1998, 107): Between 1974 and 2005, the UUP was always overrepresented, the DUP has almost always been overrepresented, the SDLP was underrepresented in 8 out of 11 electoral contests, while Sinn Féin was underrepresented in the elections it participated in until 2001(see Table 6.3). The 1920 Government of Ireland Act established that 13 MPs would represent NI in Westminster, reduced in 1948 to 12 when University seats were abolished. Until the early 1980s, this seat share was lower than that which its proportion of the population would have warranted on the grounds that important powers had been devolved to NI institutions (Bogdanor, 1999, 70). Despite the imposition of direct rule in 1972, the number of seats for NI did not increase until 1983, when it rose to 17 seats. In 1997, the Boundary Commission allocated an additional MP to NI. The Government of Ireland Act established a bicameral parliament for NI. The lower house, the NI House of Commons, had 52 members, 48 of whom were to be elected by Single Transferable Vote (STV) from 9 constituencies, with an additional 4 seats selected by Queens University Belfast graduates (counted as a single constituency). STV is a proportional electoral formula employed in multi-­member constituencies and permits voters to rank order candidates in terms of preferences (Gallagher and Mitchell, 2005, 293–296). Members of the upper house, the NI Senate, were elected by the lower house, with the exception of the Lord Mayor of Belfast and Mayor of Londonderry who also sat in the Senate. In 1929, the Unionist-­dominated parliament introduced a single-­member plurality system for 48 seats, retaining STV for only the university seats until 1969 when they were converted into four separate single-­member plurality seats. It has been argued that the 1929 electoral reform had major implications for the future development of the NI state insofar as it contributed to the consolidation of the UUP’s dominance over the political system, especially in relation to other parties targeting the Protestant vote, and the predominance of constitutional issues over class struggle as principal dimension of political competition (Pringle, 1980, 118; Coakley, 2009; Bew et al., 2009, 21).

Table 6.3 Electoral system effects: Sinn Féin in the UK House of Commons (1950–2015), Northern Ireland Parliament (1922–72) and Northern Ireland Assemblies (1982, 1996, 1998–2017) Election year

Name

% votes (in NI)

UK House of Commons

1950 1955 1959 1964 1966 1983 1987 1992 1997 2001 2005 2010 2015 2017

SF SF SF Republicans Republicans SF SF SF SF SF SF SF SF SF

0.3 23.6 11 15.9 10.5 13.4 11.4 10 16.1 21.7 24.3 25.5 24.5 29.4

NI Parliament

1921 1925 1933

SF Republicans Republicans

NI Assembly

1982

NI Forum NI Assembly

Note 1 Calculated for list seats.

No. seats

% seats (in NI)

Theoretical assembly electoral threshold (for NI seats)

Advantage ratio

0 2 0 0 0 1 1 0 2 4 5 5 4 4

0 16.67 0 0 0 5.88 5.88 0 11.1 22.2 27.7 27.7 22.2 22.2

10.83 10.83 10.83 10.83 10.83 9.1 9.1 9.1 8.84 8.84 8.84 8.84 8.84 8.84

0 0.70 0 0 0 0.43 0.52 0 0.69 1.02 1.14 1.09 0.91 0.76

20.5 5.3 7.7

6 2 1

11.5 3.8 1.92

3.83 3.83 5.2

0.56 0.72 0.25

SF

10.1

5

6.41

2.89

0.63

1996

SF

15.47

17

15.45

2.95

0.99

1998 2003 2007 2011 2016 2017

SF SF SF SF SF SF

17.63 23.52 26.16 27.45 24 27.9

18 24 28 29 28 27

16.66 22.2 25.93 26.85 25.9 30

2.53 2.53 2.53 2.53 2.53 2.95

0.94 0.94 0.99 0.98 1.08 1.08

1

158   Electoral systems as an alternative In the arena of local government, the Unionist government in NI also abolished PR and redrew electoral boundaries in the early 1920s, such that it ‘gerrymandered the nationalists out of winning a large portion even of those authorities which they could reasonably have hoped to control’ (Whyte, 1983, 5–7; Pringle, 1980, 199). Nevertheless, according to Pringle, there is less evidence that electoral boundaries for the NI parliament were manipulated to marginalize nationalists (1980, 195). The single-­member plurality electoral system punished small parties but did not substantially alter the balance of forces between nationalists and unionists (ibid.). Since the return of direct rule in NI in 1972, successive plans to reinstate an Assembly for NI departed from the conviction that the single-­member plurality electoral system employed in NI contributed to political tensions and that a more proportional electoral formula must be employed (Coakley, 2009, 253; Cunningham, 2001, 14). STV was the formula of choice for electoral contests determining the composition of most of the short-­lived bodies and fora established prior to the 1998 Good Friday Agreement as well as those for the current NI Assembly (see Appendix 1). STV has the attraction of permitting the election of a number of parties representing different traditions within a single constituency (Robinson, 2010, 162). For all STV electoral contests in NI using STV candidates from more than one party have almost always been elected, and parties representing both nationalist and unionist traditions have very frequently been elected in each constituency (Bogdanor, 1999, 78; Robinson, 2010, 71). Effectiveness of electoral system in marginalizing Sinn Féin and Republican Clubs Electoral systems employed for Westminster and NI parliament/assembly elections have not been an effective alternative to the proscription of Sinn Féin and Republican Clubs, even if in theory they have tended to be disadvantageous for Sinn Féin. Seen over the long term, the Westminster electoral system has not generally prevented Sinn Féin from entering parliament when it decided to contest elections. The single-­member plurality electoral system employed in Westminster elections, with its high disproportionality score for vote–seat ratios (average LSq = 18.10 for NI) (see Table 6.1) and an average effective threshold at district level of 37.5 per cent (see Appendix I) tends to favour large parties. Nevertheless, parties such as Sinn Féin, with a geographically concentrated support base, tend to win a more proportionate share of the seats. As Table 6.3 shows, Sinn Féin won at least one seat in 9 of the 14 elections it contested in the post-­war period, obtaining between two and five seats. It was, however, underrepresented more often than not in terms of seats won given its vote share in NI (i.e. in 6 out of 9 elections where it won seats). Except for 1933, Sinn Féin did not participate in electoral contests for the NI parliament under the single-­member plurality system in place between 1929 and 1972. In 1933, this highly disproportionate system (average LSq = 17.53), with a district magnitude of 1 (except for Queens University) and an effective threshold

Electoral systems as an alternative   159 of 5.2 per cent at assembly level, did not keep Sinn Féin from winning seats (see Table 6.1, Table 6.3 and Appendix 1). Contesting the election under the name of Republicans, Sinn Féin candidates won one seat, but as the advantage ratio of 0.25 shows, the party was underrepresented given its vote share (see Table 6.3). When a single transferable vote or a variation of PR closed list systems were used, Sinn Féin always won seats when it contested elections. These were more proportional formulae (LSq = 4.44 on average for ad hoc elections between 1973 and 1996) (see Table 6.1). Average district magnitude was relatively high (ranging between 5 and 6.5) and Sinn Féin easily crossed the assembly-­level effective electoral threshold (which ranged between 2.89 and 3.83 per cent for STV and PR elections prior to 1998 assembly elections). By the mid-­1990s, its share of seats obtained was nearly the same as its share of votes and in recent elections it has been overrepresented (see Table 6.3). In contrast to Sinn Féin, Republican Clubs/The Workers Party has never won seats in Westminster or NI Assembly elections. Its share of the vote has always been small, at best 3.1 per cent of the vote in NI in October 1974 Westminster elections and 2.7 per cent in 1982 NI Assembly elections (see Table 6.4). Over the longer term, however, Republican Clubs/The Workers Party’s limited electoral fortunes are more likely to be the result of low votes rather than electoral system effects. Republican Clubs/The Workers Party has failed to obtain seats in both elections using the highly disproportionate single-­member plurality system used for Westminster elections and the relatively proportional single transferable vote or PR closed list systems used in the NI elections it has participated in. Looking specifically at the 1950s and 1960s, when Sinn Féin, and what was then its successor, Republican Clubs, were banned, it could be argued that at least in theory, the electoral system employed in Westminster elections were an effective instrument for marginalizing the parties. In the five elections Sinn Féin/ Republicans contested between 1950 and 1966, it won seats on only one occasion, despite an average vote share during that period of 12.3 per cent (see Table 6.3). In terms of its seat–vote ratio, Sinn Féin was underrepresented in parliament when it did win a seat in 1955. In NI, Sinn Féin and Republic Clubs did not contest elections in the period from the mid-­1930s until the resumption of direct rule by Westminster and so electoral effects cannot be calculated. Nevertheless, marginalization of Sinn Féin and Republican Clubs through the electoral system was more theoretical than real at the time they were banned due to the parties’ lack of interest in electoral politics as a vehicle for achieving political objectives. Sinn Féin did not – and still does not – take up its Westminster seats because it did not recognize the legitimacy of the House of Commons in the regulation of NI affairs. It could not, therefore, be plausibly argued that the electoral system was an effective alternative to banning the parties, when the parties themselves made little use of parliament as an avenue for political influence. Electoral systems continued to remain ineffective means of marginalizing Sinn Féin and Republican Clubs at the time the parties were legalized. In the 1970s, Republican Clubs, which participated in Westminster and NI elections, was a very minor party and never won seats, while Sinn Féin did not contest

Table 6.4 Electoral system effects: Republican Clubs/The Workers Party in the UK House of Commons (1974–2015) and Northern Ireland Assemblies (1982, 1996, 1998–2017) Election year

Name

% votes (in NI)

Seats

Theoretical assembly threshold (for NI seats)

UK House of Commons (NI Seats)

February 1974 October 1974 1979 1983 1987 1992 1997 2001 2005 2015

Republican Clubs Republican Clubs Republican Clubs Workers Party Workers Party Workers Party Workers Party Workers Party Workers Party Workers Party

2.1 3.1 1.7 1.9 2.6 0.5 0.3 0.3 0.2 0.4

0 0 0 0 0 0 0 0 0 0

10.83 10.83 10.83 9.1 9.1 9.1 8.84 8.84 8.84 8.84

NI Assembly

1973

Republican Clubs

1.8

0

2.89

Constitutional Convention

1975

Republican Clubs

2.2

0

2.89

NI Assembly

1982

Workers Party

2.7

0

2.89

NI Forum

1996

Workers Party

0.47

0

2.951

NI Assembly

1998 2003 2007 2011 2016 2017

Workers Party Workers Party Workers Party Workers Party Workers Party Workers Party

0.24 0.2 0.14 0.18 0.2 0.2

0 0 0 0 0 0

2.53 2.53 2.53 2.53 2.53 2.95

Note 1 Calculated for list seats.

Electoral systems as an alternative   161 elections at all. In NI, as discussed above, the British government consistently selected proportional electoral systems to avoid the marginalization of republican parties in elections to representative bodies in NI, institutions which the parties were not popular enough to participate in (Republican Clubs/The Workers’ Party) or not willing to join until the late 1990s (Sinn Féin).

Electoral systems in Germany Electoral system for Bundestag elections The design of the Federal Republic’s electoral system reflected the preferences and experiences of the post-­war occupying powers and political compromises among German parties, especially the CDU/CSU, which preferred a single-­ member plurality system, and the SPD and FDP, which were more inclined to proportional representation (Carstairs, 1980, 168; Pulzer, 1983, 95–96; Merkl, 1963, 85–89). The electoral system also reflects the paramount concern, as Pulzer put it, of ‘shutting the Weimer stable door’ of high party-­system fragmentation and unstable multiparty government (1983, 85; Roberts, 2009, 57). The compromise formula consisted of a Mixed Member Proportional (MMP) electoral system, used for Bundestag elections since 1949. While there have been various electoral reforms, these nevertheless have left the basic character of the electoral system intact (Roberts, 2009, 56; Capoccia, 2002b, 173). From the outset, citizens voted for two kinds of Bundestag member, constituency seats and regional party list seats. In the first election in 1949, voters selected 60 per cent (240 seats) of members directly in single-­member districts in accordance with the plurality rule and the remaining 40 per cent (160 seats) from closed party lists. In that election voters had one vote for a constituency candidate which was then automatically taken as a vote for party list candidates. The d’Hondt formula was used to calculate the parties’ overall seat share, although the constituency seats already won by the party were counted as part of their party’s seat share. An electoral threshold applied so that only those parties winning 5 per cent of the vote at the Land level took up party list seats. A further reform of the electoral law in 1953 increased the size of the assembly to a minimum of 484 seats, half (242 seats) for single-­member constituencies and half (242) for party list seats. Voters now also had two votes, one for constituency seats and one for party list seats, giving them the opportunity to ‘split’ their votes between different parties if they so chose. The new law also adjusted the 5 per cent electoral threshold, so that parties could only take up list seats if they won 5 per cent of the vote in the whole country. An exception was made for parties that won at least one constituency seat. From 1956, parties obtaining at least three constituency seats could obtain a share of list seats, even if they missed the 5 per cent threshold. While in 1949, vote counting and seat allocation took place with reference to the regional level alone (Capoccia, 2002b, 173), following reforms in 1953 and 1956 a party’s total seat share was determined, first, by the number of its candidates winning a plurality of votes in the

162   Electoral systems as an alternative constituency seats. Second, a party’s list votes – cast for closed lists drawn up at Länder level – are summed to determine the party’s national share of votes and thus its national share of Bundestag seats. Finally, a party’s total number of constituency seats is deducted from its share of party list seats to ensure the number of seats obtained corresponds proportionally to its vote share. Prior to the 1987 general election, the Hare-­Niemeyer electoral formula replaced the d’Hondt formula for allocating list seats, which was itself replaced by the Sainte-­Laguë/ Schepers formula for the 2009 election. Until the 2013 electoral reform, a party obtained ‘surplus’ or ‘overhang’ (überhangmandate) seats if the number of constituency seats it won at Land level exceeded the number of seats it was entitled to receive from its regional quota of list seats. Following a Federal Constitutional Court ruling that this system was unfair, the 2013 electoral reform permitted those parties that lost out in terms of seat-­vote ratios due to the distribution of overhang seats to be compensated. Under the new rules, the disadvantaged parties would also receive additional seats so that overall, the seat-­vote ratio of the whole system would be more proportional. This procedure was already used extensively at the Länder level prior to its introduction at the Federal Level (Massicotte, 2003, 8). Appendix 1 summarizes the main features of the German electoral system. Länder electoral systems where the Socialist Reich Party contested seats: 1950–52 A two-­tiered mixed electoral system, also known as the ‘Losers and Surplus method’ (Massicotte, 2003), was initially employed in Schleswig-­Holstein when the SRP won 1.6 per cent of the votes but no seats. This system combines single­member plurality with PR in single-­member constituencies, with each voter casting a single vote (ibid., 4). While most representatives were elected by the single-­member plurality method, the others were allocated by PR, using the d’Hondt formula, with reference to ‘reserve lists’. These lists were made up of votes for parties that did not win a constituency seat or votes over and above that needed to achieve a plurality for a party that did win a seat (ibid., 4). As Massicotte points out, the system produced outcomes that provided a substantial bonus to the leading party at the expense of the smaller ones (ibid., 4). For the 1950 elections, 46 representatives were elected in single-­member plurality constituencies, with the remaining 23 distributed by PR. From this election, a legal threshold of 5 per cent (or one constituency seat) was set as a prerequisite for inclusion in reserve-­list seat allocation (ibid., 15). North Rhine-­Westphalia has employed an MMP system since 1947. In the 1950 election where the SRP won 0.2 per cent of the vote and no seats, voters chose a single constituency candidate with PR seats also allocated on the basis of each party’s individual candidates (ibid., 3). Constituency seats have constituted the largest portion of seats; between 1947 and 1975 there were 150 constituency seats out of a total of 200 (wahlrecht.de). An electoral threshold of 5 per cent has been employed since the first post-­war elections (Massicotte, 2003, 14).

Electoral systems as an alternative   163 The Hare-­Niemeyer method was employed for distributing the PR seats (wahlrecht.de). Lower Saxony, where the SRP performed best winning 11 per cent of the vote and 16 seats, also employed a MMP system (Massicotte, 2003, 3). Of the total 149 seats, 95 were constituency seats and the d’Hondt electoral formula was used. An electoral threshold was not introduced until 1959. In Bremen, the other Land where the SRP performed well with 7.7 per cent of the vote and 8 seats in 1951, a PR system was employed. The 100 parliamentary seats were equally divided between two constituencies, the cities of Bremen and Bremerhaven, the d’Hondt electoral system was used and a 5 per cent threshold was applied at constituency level (ibid., 6; Roberts, 2006, 105). From the establishment of Baden-­Württemburg in 1952, where the SRP won 2.4 per cent of the vote but no seats, a distinct form of MMP has been used (Massicotte, 2003; Roberts, 2006, 103). Voters cast a single vote in one of 70 single-­member constituencies and the candidate with the most votes is elected. Proportionality is retained by, first, adding the votes for each party’s constituency candidates in the state as a whole and calculating the distribution of additional seats in proportion to that vote share. Parties that do not get more than 5 per cent of the vote at Land level will not get any of the 50 PR seats. The seats to which the other parties are entitled are then divided among four administrative districts in proportion to the number of votes the parties receive within each district. After taking into account the 70 seats already won by direct mandate, the remaining seats allocated to the parties are then awarded to the candidates within each party with the highest number of votes (Weber and Häuser, 2008, 32; Massicotte, 2003, 13). PR seats were distributed according to the d’Hondt formula. Electoral systems where the National Democratic Party contested elections 1990–present As mentioned above, Baden Württemberg adopted a distinct form of MMP system in 1952 and it has been used ever since (Massicotte, 2003; Roberts, 2006, 103). Since 2011 the Sainte-­Laguë formula has been used to distribute PR seats. By the 1950 Land election, Bavaria had an MMP system (Massicotte, 2003, 5; see also James, 1998; Roberts, 2006, 104). Voters have had two votes from the outset, one for constituency seats and one for list seats. The ratio of constituency to list seats has been roughly equal, with 104 single-­member constituencies and 100 open-­party list seats between 1950 and 1998 and 92 constituency and 88 list seats since 2003. In contrast to other Länder using MMP, list members have been allocated among seven constituencies of varying size, based on administrative districts. The computation of total seats each party receives is made on the basis of aggregate first and second votes (ibid., 7). The d’Hondt formula was used up to and including the 1990 election, when the Hare-­Niemeyer formula was adopted. From 1946 to 1973, parties had to cross a 10 per cent threshold in one of the seven constituencies in order to be able to take up any seats, although by the 1974 election, a Land-­wide 5 per cent threshold was introduced (ibid., 14; James, 1988, 35–36).

164   Electoral systems as an alternative In 1958, Berlin changed from a standard PR electoral system to a MMP one (Massicotte, 2003, 5). In the 1970s a two-­vote system was introduced (ibid., 11). Since adoption of MMP, the size of the legislature had been altered a number of times (ibid., 8). Parties may offer voters a single Land-­level list or multiple lists in the various administrative districts (Bezirke) (ibid., 13). The d’Hondt formula was used until 1990, after which it was replaced by the Hare-­Niemeyer formula (ibid., 12). Since 1958, a 5 per cent legal threshold has applied at the Land level for allocation of list seats, although parties winning at least one constituency seat are exempt from this rule (ibid., 14). As mentioned above, Bremen has adopted a PR system in 1947, with seats divided between two constituencies. It continues to be used. From 2003, the number of seats was reduced to 83 and the electoral formula changed from d’Hondt to Saint-­Laguë. A 5 per cent threshold is applied at constituency level (Roberts, 2006, 105; Massicotte, 2003, 6). Voters currently have five votes, which can be used to select either candidates or an entire party list. There has nevertheless been a long tradition of closed party list voting in Bremen. In 1956, Hamburg adopted a PR system employing a single electoral constituency for election of 120 representatives, with seats allocated by the d’Hondt formula. Since 1953 a legal threshold prevents parties not winning 5 per cent of the votes in the Land from taking up seats. In 1991, the Hare-­Niemeyer formula was introduced and the number of seats increased to 121. By 2008, the 121 parliamentary seats were distributed via a two-­tier PR system in which 71 representatives were elected in 17 multi-­member districts, with the remaining 50 elected with reference to Land-­level lists. At present, each voter has ten votes, five for candidates and five for party lists, and the Sainte-­Laguë electoral formula is used to distribute both constituency and list seats. Since 1954 the electoral system in Hesse has been a MMP system. From 1970 to present, the parliament’s 110 seats were divided equally into 55 constituency and 55 list seats. In 1991 Hesse introduced a two-­vote system and in 1983 the Hare-­Niemeyer formula was selected for allocation of list seats. There is a 5 per cent Land level electoral threshold. Lower Saxony has, as mentioned above, employed a MMP system since 1947. By 1988, separate votes for constituency seats and Land lists had been introduced (Massicotte, 2003, 11). From 1986 elections, the d’Hondt formula has been used. Parties failing to win 5 per cent of the vote in the Land cannot take up seats. The distribution of constituency and list seats has varied over time; from 1982 to 2003 there were 100 constituency seats and a minimum of 155 seats overall, while from 2008 there were 87 constituency seats and a minimum of 135 seats overall. Similarly, North Rhine-­Westphalia has employed an MMP system since 1947 (ibid., 14). Until 2005, voters chose a single constituency candidate, with PR seats also allocated on the basis of each party’s individual candidates (ibid., 3). Subsequently, each voter got two votes, the first for a constituency candidate and the second for a Land-­level party list. Constituency seats have always constituted the largest portion of seats; for instance, between 1980 and 2000, there were 151 constituency seats out of a total 201

Electoral systems as an alternative   165 seats; and since 2005 elections there have been 128 constituency seats out of a total of 181 seats. An electoral threshold of 5 per cent has been employed since the first post-­war elections (ibid., 14). Until 2005 the Hare/Niemeyer method was employed (ibid., 12), after which Sainte-­Laguë was adopted. In order to obtain Land seats, a party must obtain at least 5 per cent of total valid second votes (ibid., 8). Rhineland-­Palatinate employed a list PR system until 1989, when it adopted a MMP system (ibid., 5). Since 1989, seats have been divided almost equally between 51 single-­member constituency seats under the plurality rule and 50 Land-­list PR seats. Voters have two votes, one for a candidate and one for list seats. Seats are allocated with the Hare-­Niemeyer formula, although from elections in 2011 the Sainte-­Laguë method was employed. A 5 per cent Land level threshold is applied. While seats are distributed at the Land-­level, parties may nevertheless choose to re-­allocate seats among four districts or to have a single land list (ibid., 13). Saarland has used a PR system since the end of World War II. In 1955, a 5 per cent Land level threshold was introduced. By 1960 a two-­tier system has been used whereby 40 members were elected on the basis of party lists in three constituencies, while the remaining ten were elected from Land-­level party lists. In 1980, the number of constituency seats was increased to 51. Also since 1960, voters have had a single vote through which they selected both a party’s constituency and Land level seats. Seats are distributed with reference to the d’Hondt formula, with ten land list seats then used to correct any distortions between the Land and constituency vote shares (ibid., 6). By 1954, Schleswig-­Holstein had adopted a MMP system. From the 1950 election, a legal threshold of 5 per cent (or one constituency seat) was a prerequisite for inclusion in list seat allocation, although since 1957 parties representing the Danish minority have been exempt from this requirement (ibid., 15). The ratio of constituency to Land list seats has been fairly stable, but numbers have changed slightly. Voters had one vote until the 2000 election, after which they could cast two votes, one each for constituency and Land list seats. The d’Hondt electoral formula was used until the 2012 election when the Sainte-­Laguë formula was introduced. Following converging trend towards adoption of MMP systems in Germany, the five Eastern Länder adopted this electoral system for Land elections in 1990. All Eastern Länder divided parliamentary seats equally (or almost equally) between constituency seats, to be elected under single-­member plurality rules, and closed list PR seats distributed in a way intended to promote proportionality. Brandenburg and Thuringia divided their 88 seats, and Saxony its 120 seats equally between constituency and list seats, and Mecklenburg-­Vorpomern has 36 constituency and 35 list seats. Saxony-­ Anhalt initially allocated 49 constituency and 50 list seats, but these were reduced to 45 constituency and 46 list seats for elections in 2006 and 2011, and to 43 constituency seats and 44 list seats for the 2016 elections. All five Eastern Länder initially adopted the Hare-­Niemeyer electoral formula for

166   Electoral systems as an alternative allocation of List seats, although Saxony changed to the d’Hondt formula in 1993 (ibid., 12). Only those parties that obtain 5 per cent of candidate votes are entitle to a share of Land list seats, although in Brandenburg and Saxony they are entitled to a share of such seats if they win, respectively, one or two constituency seats. In Thuringia, parties representing the Sorbian minority are exempt from the 5 per cent threshold. Appendix I provides a summary of the main features of these electoral systems. Effectiveness of electoral systems in marginalizing Socialist Reich Party Various authors have argued that Germany’s electoral system and especially its 5 per cent national electoral threshold for party list seats provide significant protections against anti-­system parties and thereby reduces the need to ban them (e.g. Navot, 2008, 747; Gordon, 1987, 395; Pedahzur, 2004, 118). In the following, I argue that electoral systems were not an effective alternative to proscription in the case of the SRP, which at its height performed better in electoral terms than the NPD ever did. In the case of the NPD, however, it is clear that the electoral system, especially the widespread use of legal thres­ holds, has been very successful at marginalizing the NPD in terms of entry into the parliamentary arena. The SRP did not contest elections at the federal level. No federal elections were held during its short life (October 1949–October 1952). However, the SRP did participate in five elections at the Land level. In these Länder, a straight PR electoral system (Bremen), or variations of MMP systems combining plurality and PR formulas, were employed (see Appendix I). These produced low to moderately proportional outcomes, ranging from LSq = 3.89 (Bremen) to LSq = 5.17 (Lower Saxony). All except Lower Saxony employed 5 per cent legal thresholds at either Land or constituency levels, and in the case of Schleswig-­Holstein a party winning one constituency seat could also be included in the count for PR (reserve list) seats. In North Rhine-­ Westphalia, the SRP’s poor electoral showing of only 0.2 per cent of the vote meant it could not cross the low assembly-­level effective threshold (0.37 per cent). In Schleswig-­Holstein, where the SRP won 1.6 per cent of the vote, comparatively high assembly-­level effective thresholds (5.53 per cent constituency, 3.3 per cent reserve list) produced by the Losers and Surplus method, meant that the SRP could not obtain either constituency or list seats, let alone cross the 5 per cent Land-­level legal threshold. That this result was not just due to a poor electoral showing can be seen by estimating the SRP’s chances of obtaining seats if different electoral rules were applied: For example, with the 1.6 per cent vote share the SRP obtained in Schleswig Holstein, it could have crossed the assembly-­level effective thresholds in each of the other Länder where it contested elections in the 1950s, and in Lower Saxony, where no legal threshold applied, it would probably have obtained seats. In Baden Württemberg, however, where the SRP won 2.4 per cent of the vote, the SRP

Electoral systems as an alternative   167 Table 6.5  Electoral system effects: Socialist Reich Party (1950–52) Election % votes No. seats % seats National Advantage Year electoral ratio threshold Schleswig-Holstein North Rhine-Westphalia Bremen Lower Saxony Baden-Württemberg

1950 1950 1951 1951 1952

1.6 0.2 7.7 11 2.4

– –  8 16 –

– –  8 10.13 –

3.13 0.37 1.04 0.5 0.62

– – 1.04 0.92 –

won enough votes to be, theoretically at least, entitled to seats according to the assembly-­level effective threshold (0.62 per cent), although the 5 per cent Land-­level legal threshold prevented it from getting these seats. In the two Länder where the SRP performed well and won seats – i.e. Bremen (7.7 per cent votes, 8 seats) and Lower Saxony (11 per cent votes, 16 seats) – the SRP easily crossed assembly-­level effective thresholds (1.04 per cent Bremen, 0.5 per cent Lower Saxony) and 5 per cent constituency thresholds in Bremen. While in Lower Saxony the SRP’s 16 seats saw it underrepresented in relation to its vote share (advantage ratio: 0.92), this was not the case in Bremen where it was slightly overrepresented (advantage ratio: 1.04) (see Table 6.5). From the foregoing, it can be argued that while the SRP remained a relatively small party, with less than 5 per cent of the vote, electoral systems were an effective alternative to proscription insofar as they kept the SRP from obtaining seats (i.e. Schleswig-­Holstein, 1950 and Baden Württemberg, 1952). Where the SRP won more than 5 per cent of the vote, the electoral system was a much less effective alternative to proscription. However, it cannot be argued that electoral systems prevented the SRP for winning seats more often than not, given that in the four elections where it crossed the assembly-­level thresholds, it won seats half of the time. Electoral system rules helped minimize the impact of SRP electoral successes in Lower Saxony insofar as the SRP was underrepresented in terms of its vote to seat ratio. In Bremen, however, electoral system effects meant the SRP was slightly overrepresented in terms of its vote to seats ratio. Consequently, it cannot it be argued that electoral systems systematically disadvantaged the SRP in terms of seat share where it did win seats, given that in one Land parliament its was overrepresented, while in the other it was not. Effectiveness of electoral systems in marginalizing the National Democratic Party of Germany By 1990, most electoral systems at the Federal and Land level were a variation of the MMP system, with the exception of Bremen, Saarland and Hamburg, which used straight PR systems (see Appendix 1). The disproportionality of vote

Table 6.6  Electoral system effects: National Democratic Party of Germany, Bundestag (1965–2017) and Länder (1990–2016) Election

Election year

Bundestag Bundestag Bundestag Bundestag Bundestag Bundestag Bundestag Bundestag Bundestag Bundestag Bundestag Bundestag Bundestag Bundestag Baden-Württemberg Baden-Württemberg Baden-Württemberg Baden-Württemberg Bavaria Bavaria Bavaria Bavaria Berlin Berlin Berlin Berlin Berlin Brandenburg

1965 1969 1972 1976 1980 1983 1987 1990 1998 2002 2005 2009 2013 2017 2001 2006 2011 2016 1994 1998 2008 2013 1999 2001 2006 2011 2016 1990

% votes 2 4.3 0.6 0.3 0.2 0.2 0.6 0.3 0.3 0.4 1.6 1.5 1.3 0.38 0.2 0.7 1.0 0.4 0.1 0.2 1.2 0.6 0.8 0.9 2.6 2.1 0.6 0.1

No. seats

% seats

National electoral threshold

Advantage ratio

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

0.15 0.15 0.15 0.15 0.15 0.15 0.15 0.11 0.11 0.13 0.13 0.13 0.13 0.13 0.62 0.62 0.62 0.62 0.37 0.37 0.41 0.41 0.57 0.57 0.57 0.57 0.57 0.84

– – – – – – – – – – – – – – – – – – – – – – – – – – – –

Brandenburg Brandenburg Brandenburg Bremen Bremen Bremen Bremen Hamburg Hamburg Hamburg Hamburg Hesse Hesse Hesse Hesse Hesse Lower Saxony Lower Saxony Lower Saxony Lower Saxony Mecklenburg-Vorpommern Mecklenburg-Vorpommern Mecklenburg-Vorpommern Mecklenburg-Vorpommern Mecklenburg-Vorpommern Mecklenburg-Vorpommern Mecklenburg-Vorpommern North Rhine-Westphalia North Rhine-Westphalia North Rhine-Westphalia North Rhine-Westphalia

1999 2009 2014 1995 1999 2011 2015 1997 2004 2011 2015 1995 1999 2008 2009 2013 1990 1994 2008 2013 1990 1994 1998 2002 2006 2011 2016 1990 2000 2005 2010

0.7 2.6 2.2 0.1 0.3 1.6 0.2 0.1 0.3 0.9 0.3 0.3 0.2 0.9 0.9 1.1 0.2 0.2 1.5 0.8 0.2 0.1 1.1 0.8 7.3 6.0 3.0