Jump to ContentJump to Main Navigation
Democracy and the State in the New Southern Europe$

Richard Gunther, P. Nikiforos Diamandouros, and Dimitri A. Sotiropoulos

Print publication date: 2006

Print ISBN-13: 9780199202812

Published to Oxford Scholarship Online: May 2009

DOI: 10.1093/acprof:oso/9780199202812.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2019. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: null; date: 17 September 2019

Multilevel Governance and the Transformation of Regional Mobilization and Identity in Southern Europe, with Particular Attention to Catalonia and the Basque Country

Multilevel Governance and the Transformation of Regional Mobilization and Identity in Southern Europe, with Particular Attention to Catalonia and the Basque Country

(p.235) 6 Multilevel Governance and the Transformation of Regional Mobilization and Identity in Southern Europe, with Particular Attention to Catalonia and the Basque Country
Democracy and the State in the New Southern Europe

Iván Llamazares

Gary Marks

Oxford University Press

Abstract and Keywords

South European countries share a tradition of centralized government, which was reinforced by authoritarian regimes. However, democratization was accompanied by regional mobilization, particularly in Spain, where the double pressure from the EU and empowered regional governments weakened the central state. Examples are Catalonia and the Basque Country, which have strong ethno-territorial movements that took advantage of Spain's integration into the EU. European integration provided regional actors with new avenues for mobilization. The Basque Country is subjected to a destabilizing contest between nationalists and non-nationalists, while Catalan nationalists have been more pragmatic, without loosing their assertiveness and their focus on obtaining ever greater autonomy. Survey data show that the share of Basques and Catalans who consider themselves exclusively Spanish has declined dramatically, while after 1979 the share of those who claim having balanced multiple identities has risen. While separatism has become weaker over time, European integration has strengthened territorial identities.

Keywords:   multi-level governance, territorial politics, regional mobilization, nationalism, multiple identities, Basque Country, Catalonia

The territorial organization of decision‐making in Europe has been transformed in the last twenty years under the influence of European integration and subnational decentralization. With the creation, enlargement, and deepening of the EU, decision‐making competencies in a range of policy areas, including regional policy, agricultural policy, the internal market, and monetary policy have been reallocated from the national to the European level. At the same time, central state executives in several European countries, including Spain and, to a lesser extent, Greece, Italy, and Portugal, have shifted a variety of competencies to regional, local, or urban governments.

This chapter examines the consequences of these developments for regional governments and regional political parties. The first set of questions we ask has to do with how the changing territorial organization of decision‐making has influenced regional political activity. How have European integration and regional decentralization shaped the structure of political opportunity for regional parties and regional governments? To what extent have they been drawn directly to the European level, and how can one explain the resulting pattern of representation? The second set of questions we ask has to do with the goals of regional actors. To what extent has their participation in an emerging European polity influenced the ambitions of regionalists for political autonomy and the conceptions of institutional alternatives available to them?

(p.236) The countries of Southern Europe—Greece, Italy, Portugal, and Spain—are an interesting focus for such questions because they share membership of the EU and many sociocultural features, yet have contrasting systems of territorial relations. Variation in the territorial relations of these four countries is just about as great as that within the EU as a whole. At one extreme, Spain is one of the most ethno‐territorially diverse countries in Western Europe, while Portugal and Greece are among the most homogenous. Catalonia and the Basque Country have a combination of strongly entrenched regional governments and electorally successful nationalist parties; Northern Italy has a weaker regional party and much weaker subnational competencies; while Portugal and Greece are unitary states with largely national parties.

An examination of territorial politics in Southern Europe is therefore a study in contrasts. This is the first theme of this chapter. Our answers to the questions posed above emphasize the very different starting points of subnational actors both across and within the countries of Southern Europe, and we find few signs of convergence in the territorial structuring of authority among these countries. Our second and more contentious theme is that the role of national states is being transformed in some subtle—and not so subtle—ways. In regions where the reallocation of competencies and the mobilization of regional parties has progressed furthest, particularly in Catalonia and the Basque Country, we find that subnational governments are intimately involved at the European level, that the national state has lost appeal as an institutional form for realizing political autonomy, and that there has been a significant shift toward multiple territorial identities. In short, the central state is in decline in Southern Europe, both as a normative model for regional autonomy and as an actor representing domestic territorial interests in the EU.

From National Decision‐Making to Multilevel Governance

The reallocation of decision‐making competencies away from territorial states that has taken place over the past three decades in Southern Europe is an historically anomalous development that reverses a centuries‐long process. The overall direction of power redistribution and the locus of power creation in the process of state‐building from the thirteenth to the twentieth centuries was toward the center. While the absolute degree of centralization varied across Western European states, all became more centralized as monarchs and state builders struggled to monopolize legitimate authority, to create secular (p.237) hierarchical systems of justice, and to deepen and widen taxation. This process was greatly intensified with the rise of nationalism across Europe and the identification of centralized extraction, provision of welfare, and control of various sectors of the economy and society with the interest of the nation as a whole.1

State‐building followed divergent paths in Southern European countries.2 Portugal and Spain developed state‐like institutions at a very early historical stage, whereas Greece and Italy did not become independent and unified countries until the nineteenth century. In the Iberian Peninsula after the secession of Portugal from the Spanish crown in the seventeenth century, central states evolved gradually within stable territorial boundaries. In Spain and Portugal, political unification preceded the development of modern nationalism. However, these two countries differed in that the Spanish crown encompassed a much more heterogeneous collection of cultures and political institutions than the Portuguese crown. In Spain, encroachment by the center had to surmount serious resistance on the part of peripheral movements. The revolt of the Catalans in 1640, the War of Succession at the beginning of the eighteenth century, and the continuous political conflicts and civil wars of the nineteenth century were decisive chapters in a drawn‐out struggle leading to a modern centralized state.

The Greek and Italian experiences were very different. In both countries nationalism played a critical role in the expulsion of foreign powers and the emergence of centralized political institutions. However, there are also important contrasts between the Greek and Italian paths. In Greece, absorption of territories previously dominated by foreign states continued until well into the twentieth century. In Italy, a national state was constructed from discrete political units by the end of the nineteenth century.

Although their historical paths were distinctive, all Southern European states were developing highly centralized political systems by the beginning of this century. This trend was reinforced by the centralizing tendencies of authoritarian regimes suffered by each of these countries in the twentieth century.

The contention that guides this chapter is that this development has been reversed over the past two decades. Of course, as with the previous process of central state‐building, we find large and persistent variations across countries. But there are strong grounds for believing that the overall direction of development in Western Europe is toward the devolution of competencies both up the European level and down to various levels of subnational government. We describe the resulting pattern of policymaking as one of multilevel governance in which (p.238) governments at different territorial levels share, rather than monopolize, public decision‐making over wide areas of policy. The past several decades have seen the development of highly variable and nonhierarchical patterns of interaction among diverse actors in dispersed territorial arenas.3

As we noted above, policymaking is highly heterogeneous across policy areas. In some policy areas, such as monetary policy and agricultural subsidies, major decisions are made exclusively at the European level; in others, such as fiscal policy and most areas of social policy, they remain largely at the national level. In many areas of service provision, subnational governments play the major role; while in others, such as cohesion (structural) policy and the environment, governments at several levels participate in decision‐making. This policy patchwork is overlaid by wide territorial variation, for there are sharp differences in the extent to which regions are mobilized politically, either by subnational governments or by regional parties. In Greece, Portugal, and Spain territorial relations have been shaped by the democratic transition that took place in each country during the 1970s. In Portugal, and particularly in Spain, decentralization has gone hand in hand with democratization. Decentralization has been promoted by the negative experience of centralism under former authoritarian regimes, by regional movements (most notably in Spain, and, to a lesser degree, in Portugal's Atlantic islands during the 1970s,4 and by the need to strengthen democratic consolidation by integrating local and regional elites within the new political system.

As we detail below in our discussion of ethno‐territorial movements in Catalonia and the Basque Country, regionalization has been most ambitious in Spain. However, steps in this direction have been taken in all Southern European countries. In Portugal, the Constitution of 1976 called for the creation of the autonomous regions of the Azores and Madeira, each of them having its own executive and an elected assembly with significant competencies.5 The Constitution of 1976 also mandated administrative regions and regional assemblies in the rest of Portugal, though this was not put into practice until 1991.6 Regional Coordinating Committees were set up in Portugal in 1977 to design and implement regional development plans, a function that has been strengthened as a result of the EU's cohesion policies (Opello 1992). However, in 1998, a referendum to create elected regional councils was defeated.

Decentralization in Greece has been the most modest. Despite administrative reforms in 1982 and 1986 the regional councils [nomarchia] and administrative regions [perifereies] remain of marginal (p.239) importance.7 The most important decision makers at the regional level are prefects who are appointed by central government. Council assemblies at the level of the prefecture (nomarchia) were created in 1982, but until the elections of October 1994, they were not directly elected, but selected by the governing party from among local officials, and their decisions can be annulled by prefects. However, against the background of the extreme centralism of the past, particularly under the military dictatorship of 1967 to 1974, the reforms of the 1980s can be considered a noteworthy innovation. The most important development in Greece was the passage of a new law in 1997, which reorganized the more than 6,000 municipalities and communes into 1,033 larger municipalities and enlarged the competences and jurisdiction of municipal authorities.

Italy has not experienced a recent democratic transition nor has it had a legacy of strong ethno‐territorial movements. However, it shares with other Southern European countries the experience of authoritarian rule and a corresponding reaction against centralized governance once it ended. Pressures for decentralization have been a continuing feature of the Italian polity. The regionalization process started with the 1948 Constitution, which distinguished between ‘special’ and ‘ordinary’ regions (see Leonardi 1992). Special regions (Trentino‐Alto Adige, Val d'Aosta, Sardinia, Sicily, and Friuli‐Venezia Giulia) were created to deal with ethnic, territorial, and linguistic conflicts, and they were given their own institutions in the 1950s (see Putnam 1993). The establishment of regional governments in ‘ordinary’ regions did not take place until 1970, largely because key Christian Democratic interests opposed the policy and because fundamental institutional change was impeded by the polarization of Italian politics. In 1978, the administrative resources of regional governments and the share of public spending devoted to them increased significantly (Leonardi 1992: 231). Despite such steps, Italian regions remain weak. They lack significant decisional competencies, have limited administrative capability, and receive nearly all their resources from central government.8 This gap between expectation and reality has helped to trigger regional mobilization, most notably in the Northern Leagues. Regional demands have also been voiced by the Conference of Regional Presidents which was established in 1983. In the early 1990s, the Conference of Regional Presidents called for regional financial autonomy, clear‐cut areas of competency, regional representation at the national level and the introduction of subsidiarity as a constitutional principle.9 The same dissatisfaction has led Italian regions to cultivate transnational links with other regions and demand direct representation in the EU (Leonardi 1992; Grote 1996).

(p.240) Regional Mobilization in the EU

A key feature of multilevel governance is that subnational governments are no longer constrained to dyadic interactions with central state actors, but connect instead with a variety of actors in different arenas. Until the post‐World War II era the logic of conflict for regions that attempted to resist the process of state‐building was bipolar, pitting recalcitrant regions against nation states intent on consolidating their grip on their respective territories. Even when ethno‐territorial movements within the same state shared the goal of defending themselves against state encroachment, they were rarely able to coordinate their activity.10 The defining feature of such movements was their particularism, the defense of exclusive cultures and institutions, and as a result they found it difficult to establish durable cross‐regional alliances. Regions that straddled national borders (e.g. the Basque Country) developed along different lines in each country and regionalists were drawn into political conflict with their respective national state. National states not only structured political conflict within their territories but served as a barrier to alliances between subnational actors located in different countries and between subnational actors and foreign states.

In the EU, relations among territorially defined political actors are characterized by multipolarity in which actors participate in subnational, national, and supranational political arenas simultaneously.11 These arenas connect subnational actors with subnational actors in other states, with supranational actors, with their respective national governments, and even, in some cases, with foreign governments. The result is an intricate web of relationships. State executives no longer mediate relations between subnational actors and extranational actors. Supranational, national, and subnational arenas are more accurately described as interpenetrative than as simply interconnected (Jones and Keating 1995). The boundaries between these arenas are diffuse rather than nested. Instead of being nested exclusively within their respective national arenas, subnational actors are now engaged in a variety of transnational relationships in diverse supranational, national, and subnational arenas.

The shift from bipolarity to multipolarity has transformed coalitional possibilities. Whereas under bipolarity, regional actors were confronted by centralizing states on a whole range of issues, in the emerging European polity they are embedded in a pluralistic setting in which they may ally with different actors at different levels on different issues.12

(p.241) The issues that divide mobilized ethno‐territorial actors from state actors are no longer mutually reinforcing or predictable. Regional actors face a fluid situation characterized by cross‐cutting rather than mutually reinforcing cleavages. For this reason, we hypothesize in the following section that multipolarity has a profound influence on the political orientation of regional actors, favoring pragmatic bargaining and demands for regional autonomy rather than separate statehood.13 But multipolarity by no means tames regional demands. In fact, it does the reverse. Insofar as regional actors have new sources of political influence and the means to build ethno‐territorial constituencies, they may continuously upgrade their demands for an increased share of competencies and resources. Multilevel governance has undermined violent nationalism, but it has also intensified regional demands, and prompted a series of experiments with regional autonomy short of state creation.

The situation we are describing is unprecedented in modern Europe, but it is worth emphasizing that we are not forecasting the demise of the territorial state. European integration and decentralization have pulled some important competencies away from national states, but they have by no means negated the immense resources still concentrated in the hands of central state executives. State executives remain the most influential political actors in the EU by virtue of their representation in the Council of Ministers and the European Council. The difference now is that national arenas are just one among a variety of channels that are available for subnational actors.

The political reality of state executive power within the multilevel polity described above creates an ambiguous, double‐edged, situation for regional actors, particularly those entrenched within their respective states. On the one hand, they are concerned that state executives may preempt their competencies by making laws at the European level over which they have little control. State executives have been able to project their power into the supranational arena by virtue of their dominance of treaty making in the EU.14 Entrenched domestic interests, including particularly subnational governments, are fearful of being outflanked by state executives as decision‐making is shifted to the European level. The result is an ongoing struggle among contending institutions at the supranational, national, and subnational levels concerning decision‐making, voting rules, and the role of subnational governments at the European level.

On the other hand, regional actors see enhanced opportunities for influence in many areas of policymaking. While national governments dominate the process of institutional creation, they do not dominate many areas of policymaking (see Marks, Hooghe, and Blank 1996: (p.242) 343–78). The capacity to create new institutions does not imply the capacity to control how those institutions actually work. Recent years have seen the creation of new policies, especially in the area of cohesion policy, that are not exclusively determined by national governments and which bear directly on subnational actors and institutions (Marks 1992). European integration has, in a nutshell, created powerful supranational institutions and has provided regional actors with new avenues for influencing decision‐making.

Regional actors have mobilized in the European arena through diverse channels. Local and regional governments from several countries have set up autonomous offices directly in Brussels; subnational governments have established numerous formal and informal networks; in regions designated for EU cohesion funding, subnational officials help design and implement economic development plans alongside national and Commission officials; and subnational governments are represented in highly visible, though primarily symbolic, assemblies—most notably the Committee of the Regions established in the Maastricht Treaty.

Subnational governments in Southern Europe, and particularly from Spain, have taken advantage of these channels. They are represented in the Assembly of European Regions, the Conference of Peripheral Maritime Regions, the ‘Four Motors for Europe’, Eurocities, the Association of Regions of Traditional Industry, C‐6 (a network of six French and Spanish Mediterranean cities), the European Association of Border Regions, the Union of Capital Regions, associations covering the Alps and the Pyrenees, the Association of Frontier Regions, and the Coalfields Communities Campaign. Catalonia has been particularly active in creating subnational networks. Catalans have chaired the Assembly of European Regions and the Council of Municipalities and Regions, and Pasqual Maragall, President of the Catalan Government, has served as President of the Committee of the Regions (Muñoz 1993).

But, and this is a distinctive feature of multilevel governance, regions do not engage in these activities equally. There is no congruence, or even convergence, in the political role of regions in the EU. Instead, there are enormous disparities in the level of organization, financial resources, political autonomy, and party‐political mobilization of regions across Southern Europe. The creation of new avenues for regional mobilization brings into focus persistent differences in the organizational and political capacities of regional actors. At the one extreme, Spanish Comunidades Autónomas (along with German Länder and Belgian regions) are well funded, strongly institutionalized, entrenched within their respective states, and active in the European arena. At the other extreme, regional governments are much (p.243) weaker in Greece or Portugal. If one wishes to find subnational political actors in the latter countries, one has to look to town and city mayors who, even if they are aware of opportunities in the European political arena, lack the resources to mobilize there. There is, therefore, little sign of the kind of territorial convergence implied in the notion of a ‘Europe of the Regions’. In perhaps no other area of political institutionalization in the European polity is there greater diversity than in the territorial politics of the member states. Multilevel governance seems to be leading not to uniformity but to continued diversity as contrasting regional actors are brought together within an overarching polity.

The Transformation of Ethno‐Territorial Movements

What is the effect of multilevel governance on territorial conflict? In this section we examine this question by looking at two regions in particular: the Basque Country and Catalonia. These are regions with strongly entrenched ethno‐territorial movements that go back well before European integration was underway and where, as a consequence, the effects of European integration and multilevel governance may be analyzed in historical perspective.15 The Basque Country and Catalonia have the strongest and most durable ethno‐territorial movements in Southern Europe, and are two of the four most influential movements (along with the Flemish and Scottish) in Western Europe. In both the Basque Country and Catalonia, ethno‐territorial movements challenge their respective central states directly in demanding new competencies. For these reasons they are interesting cases in which to examine the dynamics of multilevel governance.

The consolidation of democracy in Spain went hand in hand with far‐reaching decentralization. The reasons for this go beyond the pressures mounted by Basque and Catalan nationalists. Those pressures were effective due to the vulnerability of the democratizing political elite during the transition process and the widespread rejection of the Francoist model of traditional Spanish nationalism involving the suppression of peripheral cultures.16 Regionalism in Spain has been a dynamic process. Concessions to Basques and Catalans precipitated demands for decentralization on the part of other, less distinct, regions that feared they would lose resources and influence if they did not mobilize on their own account. Central policymakers believed that they could tame Basque and Catalan pressures by creating federal institutions for the country as a whole.

In one sense, those pressures have been tamed, but in another sense they have been intensified. Violent nationalism has been decisively (p.244) weakened where it once had widespread appeal, in the Basque Country. Participation in a system of multilevel governance and the multipolar character of political coalitions in this new setting have made political–territorial conflicts less explosive and more mundane. The logic is essentially the same as that underlying the moderation of revolutionism under liberal democracy: channels for authentic participation and a share of control reduce the intensity of political opposition. However, at the same time, these new channels and regional powers have ratcheted up demands for self‐government and provided ethno‐territorial parties more political leverage to attain their goals. Moderation of means—operating within the political system—does not imply timidity of goals. In both the Basque Country and Catalonia, the demands articulated by nationalist movements go beyond federalism. The constitutional and financial reforms that have been undertaken intensify, rather than dampen, the political struggle about how to structure a multilevel polity. This underlying similarity exists despite sharp contrasts in the character and context of ethno‐territorial nationalism. Whereas the claims of Catalan nationalists concern the territory of the autonomous community of Catalonia (a former principality within the crown of Aragón), the territories claimed by Basque nationalists are divided into two states (France and Spain) and into two autonomous communities within Spain (the Basque Country and Navarra). Basque nationalists have been far more extreme in their tactics and goals than Catalans. Independent statehood has long been a staple demand of the Basques; Catalans have been readier to explore independence short of statehood. Finally, and perhaps most importantly, there is a strong terrorist organization in the Basque territories (ETA) that has no parallel in Catalonia.

Basque Nationalism: Bargaining and Confrontation Under the Threat of Terrorism

The Basque nationalist movement has long rejected the Spanish state and demanded full independence.17 After Francoism, radical nationalists were willing to use violence in pursuit of their goals. The more moderate Partido Nacionalista Vasco (Basque Nationalist Party [PNV]), which is the largest party in the Basque Country, at first refused to legitimate the Spanish democratic transition. The PNV campaigned, with some success, for abstention in the 1978 constitutional referendum.

Although the PNV was initially reluctant to endorse the Spanish democratic path, once democratic regional institutions were set up, (p.245) the party was gradually integrated in a multilevel polity. With the exception of the militant Euskadi Ta Askatasuna (ETA) and its political wing, Herri Batasuna (HB), Basque nationalists opted for enhanced self‐government within Spain rather than the creation of an entirely separate state.

The coalition between the PNV and the Socialist Party (PSE–PSOE) in the Basque regional government was perhaps the most telling example of the changing nature of alliance and opposition in the new institutional framework.18 This coalition governed from 1986 to 1998, except for a brief period in 1991 when a nationalist coalition, composed of the PNV and two nationalist parties, Eusko Alkartasuna (EA) and Euskadiko Ezkerra (EE), prevailed. Despite the potential for conflict between the PNV and PSE on the national question, the coalition proved remarkably resilient—in sharp contrast to the nationalist coalition which lasted less than ten months.

Multilevel governance is conducive to complex alliances and cross‐cutting cleavages rather than a single line of combat. The coalition between the PNV and PSE–PSOE operated at the regional and local level, but not in the national political arena. After the 1993 General Election, the PSOE turned to the Catalan CiU rather than to the PNV for support. But this did not prevent the PNV from moderating its attacks on the Socialist government. Whereas the Conservative Partido Popular (PP) and the left‐wing Izquierda Unida repeatedly called for the resignation of the Socialist prime minister, Felipe González, the PNV supported González and his Socialist government. Although the Socialists resisted Basque demands for more political autonomy, the PNV feared (wrongly as it turned out) that a conservative government would centralize authority. After the victory of the PP in the 1996 general elections, the PNV supported José Maria Aznar's government on budgetary and other issues in return for concessions on certain nationalist demands. However, this did not prevent the PNV from continuing regional, provincial, and local coalitions with the PSE in the Basque Country.

Between 1979 and 1998 the PNV gradually dropped its demand for full independence and national sovereignty. José Antonio Ardanza, President of the Basque government from 1985 until 1998 and a leader of the PNV, argued that multilevel governance made traditional Basque demands for independence and statehood less pressing:

Nations—and I am referring both to those nations that identify with a state and to those without a state—will no longer consider the state as their only model of self‐fulfillment. The maintenance or achievement of sovereignty and statehood will no longer serve as a priority and obsessive objective….


We think that in the new Europe we want to build, it is possible and convenient to concentrate our efforts on national construction free from statist obsessions. As far as we are concerned, we are ready to untie the knot that has indissolubly linked nation and state.19

Regionalism has weakened violent nationalism. Both ETA, the terrorist organization, and HB, the radical national party, lost support from the mid‐1980s as regional self‐government was introduced. Reformist nationalist parties that participated in government were drawn into closer collaboration against violence.20 The Pactos de Ajuria Enea of 1988 united all Basque political parties except HB in support of continued regionalism and against violence. The Basque government and its police force were controlled by a Basque nationalist/socialist coalition and they, rather than the Spanish (or French) police, were in the front line against nationalist violence. It was one thing for Basque terrorists to attack Spanish police, but it was quite another thing for them to confront a Basque police force that was widely regarded as legitimate in the region. The police campaign against ETA has been increasingly successful. The successive imprisonment of ETA leaders created disarray within the organization. At the same time, antiviolence movements have gained considerable support in the Basque Country. The murder of several members of the PP, especially of the city counselor of the Basque city of Ermua, Miguel Angel Blanco, in 1997, after he was kidnapped and held hostage, led to a wave of civic revulsion that surprised ETA and HB and induced them to alter their strategy.

Paradoxically, the weakening of ETA and HB paved the way for the creation of a broad front calling for national independence. In September 1998, the PNV, EA, and HB, alongside other nationalist groups, concluded the Lizarra agreement which declared that the Basque conflict could only be resolved on the basis of self‐determination for the citizens of Euskal Herria. Those living in the Basque territories should have ‘the last word’.21 Neither the national frontier between Spain and France, nor the preferences of those living outside the Basque territories should weigh in finding a solution to the Basque question. The Lizarra declaration also called for the immediate creation of new institutions to meet ‘the aspirations of sovereignty’ of Basque citizens.22 By signing the Lizarra agreement the PNV moved toward a radical assertion of sovereignty and self‐determination, though it did not demand a separate state.

A few days later ETA called for a truce, the result, according to ETA, of a secret compact between ETA and the PNV signed in August 1998. According to ETA, the PNV and EA agreed ‘to sever their ties with (p.247) Spanish parties and create a unique and sovereign institution that would include all Basque territories’ (i.e. in France and Spain).23

These agreements among Basque nationalists led to the establishment, in February 1999, of an assembly of nationalist representatives in local institutions (udalbiltza).24 They also led to the formation of a coalition among the PNV, EA, and EH in the Basque parliament, even though EH did not enter the Basque government and continued to contest its legitimacy on the grounds that it reinforces the division between the Basque territories.

The transformation of the PNV was paralleled by a realignment of the Basque party system. After the Lizarra agreement, HB reorganized itself as Euskal Herritarrok (EH) in an attempt to broaden its appeal to disillusioned militants and nationalists who had publicly condemned violence. The reorganized party increased its share of the vote in subsequent regional (1998) and provincial, local, and European elections (1999).25 At the same time, the PNV and EA formed an electoral coalition to stem the loss of votes to EH on one hand and the Socialists and PP on the other. The PSE and especially the PP sought the votes of moderate nationalists who had formerly supported the PNV, and indeed both parties gained support in regional and local elections at the expense of the PNV and EA. The net result was a sharp reversal for the PNV. It lost control over several local and municipal governments, including the Basque capital (Vitoria‐Gasteiz) and the provincial government of Alava, which shifted to the PP.26

Although the PNV went so far as to declare the Basque statute null and void, demand radical restructuring of the Basque institutional framework, and sign a parliamentary agreement with EH, ETA suspended the truce in November 1999.27 However, this did not change the position adopted by the PNV leadership. In December 1999, a few days after ETA suspended the truce, the PNV reaffirmed its will to move toward a common political–institutional framework for all the Basques.

Nevertheless, due to the persistence of violence, the internal divisions of the nationalist forces, and the weakness of nationalist support in certain Basque territories, the nationalist front failed to unleash a process leading to self‐determination, and ended up collapsing among increasing reproaches between democratic and violent nationalists. The folding of the nationalist front preceded the polarized regional elections of March 2001, in which the coalition formed by PNV and EA benefited from the sharp decline of radical separatists and managed to improve its position in the Basque parliament.28 After these elections, the PNV and EA constituted a new coalition government under (p.248) the leadership of Juan José Ibarretxe.29 However, these were also the regional elections in which electoral support for all nationalist parties dipped to its lowest level since the statute of autonomy of 1979.30

In 2004, the Basque parliament passed a proposal to drastically reform the Basque statute of autonomy, the so‐called Ibarretxe Plan. According to this project, the Basque Country would enter a status of ‘free association’ with Spain. This proposal was rejected by the Spanish parliament with the votes of the Socialist and Popular parties. Following this rejection by the Spanish parliament, Juan José Ibarretxe called new elections aimed at strengthening the institutional and bargaining position of both the PNV and the Basque government. However, in the 2005 Basque elections the PNV experienced a serious setback. Although the PNV–EA electoral coalition remained the most voted for political force, it lost more than one hundred and fifty thousand votes (falling from 42.7 to 38.6 percent of the vote) and four seats in Parliament. By contrast, the Socialist Party, which was now adopting a less confrontational profile and proposing a negotiation on a renewed statute of autonomy compatible with the Spanish Constitution, experienced a strong rise in both votes (from 17.9 to 22.6 percent) and seats (from thirteen to eighteen). The radical nationalist left, now represented by the Communist Party of the Basque Territories (EHAK‐PCTV) attained also better results than in the 2001 elections (12.5 percent of the votes in 2005 vs. 10.1 of Batasuna in 2001) and seats (nine in 2005 vs. seven in 2001). The electoral advance of the radical separatist left took place after a long period without ETA killings and was paralleled by declarations made by the radical left underscoring that the times of confrontation should be replaced by politics and negotiation among all political forces in the Basque Country. The PNV setback makes it very difficult for this party to reach stable majorities in parliament and to advance unilaterally toward the implementation of the Ibarretxe Plan.

Recent developments have shown both the failure of nationalist sectors to give rise to a new political framework and the resilience of nationalist orientations among the population of the Basque autonomous community. Neither nationalists nor non‐nationalists can unilaterally impose a stable institutional framework, and all political groups, and first and foremost supporters of radical nationalism, will have to eventually accommodate in a very plural political setting. This development may be ultimately facilitated by the increasing weakness of ETA, and by the awareness, on the part of the separatist left, that a strategy based on institutional and political participation must displace the politics of violence and confrontation.

(p.249) Catalan Nationalism: Pragmatism and Assertiveness

In contrast to Basque nationalism, Catalan nationalism has historically been oriented toward greater political autonomy within the Spanish state rather than the creation of an entirely separate state.31 This facilitated the integration of Catalan demands into the Spanish Constitution of 1978 and the Statute of Autonomy of the following year. Convergència i Unió (CiU), a moderate nationalist center‐right coalition, became the major political party in the Catalan regional parliament in 1980, and its leader, Jordi Pujol, has presided over successive Catalan governments since that time.32

The CiU has combined pragmatism and assertiveness. From the beginning of the Spanish transition Catalan nationalists pressed their demands in ways that were compatible with the Spanish Constitution. In fact, the part of the Constitution that specifies the position of regions resulted from a compromise with Catalan nationalists and, in particular, their speaker in the Spanish Parliament, Miquel Roca. Nevertheless, there has been continual conflict between the government and Catalan nationalists. After the failed military coup of 1981, PSOE felt obliged to reassure Spanish nationalists by setting limits on regionalism. The same years saw an escalation of Catalan demands for greater political autonomy.

Like the PNV, the CiU has been willing to enter governing coalitions with national and other parties. But only recently has this been necessary for the CiU at the regional level. From 1984 to 1996, the CiU had absolute majorities in the Catalan Parliament, and from 1996 to 1999 the party could rely on the support of several smaller parties without having to enter a formal coalition. However, in the regional elections of 1999, a left‐wing coalition led by the former mayor of Barcelona, the Socialist Pascual Maragall, received more votes than the CiU. The CiU remained the largest party in the Catalan Parliament due to the overrepresentation of rural areas, but it was compelled to gain the support of the PP and/or the center‐left nationalist Esquerra Republicana de Catalunya (ERC). In November 1999, Pujol was reelected to the Presidency of the Catalan Government with the support of PP deputies, while ERC deputies abstained.

Relations between Spanish and Catalan parties have reflected the party identity and parliamentary strength of the Spanish government. From 1982 to 1993, the PSOE had absolute majorities in the Spanish parliament and the result was a standoff between two entrenched governments (Morata 1995). By contrast, from 1993 until 1996 the PSOE depended on the parliamentary support of the CiU, which in (p.250) exchange demanded a higher level of financial transfers from the central state on the grounds that there is a large gap between the tax revenue extracted from Catalonia and central state expenditure there.33 Under the new system, regional governments received 15 percent of the direct taxes that the Spanish government collects in their respective territories (see El País, February 6, 1994). The alliance between CiU and the PSOE became more apparent—and more precarious—in the wake of the corruption scandals implicating several well‐known incumbents, including the former governor of the Spanish central bank and the chief of the Civil Guard. Despite growing discontent among the electorate and deep political crisis in the government, the CiU supported Felipe González's refusal to resign. This placed the CiU in a difficult political position insofar as it ran the risk that some of its voters might switch to the PP. The CiU supported the González government until the end of 1995, when the CiU failed to vote for the budget thus prompting new general elections for 1996, one year before the end of the legislature.

After Aznar's limited victory in the 1996 elections, the CiU maintained a similar relationship with the PP. After the elections, Catalan nationalists helped to elect Aznar as president of the Spanish government, and they supported the government budget from 1996 to 2000. This close cooperation followed years of tensions between the CiU and the PP.34 After the 1996 agreement between these parties, the percentage of direct taxes that reverted directly to regional governments rose to 30 percent. In addition, after 1996 regions could themselves introduce exemptions or targeted reductions of certain taxes.35 In the general elections of the year 2000, the PP attained the majority in the Spanish parliament. After these elections, no stable parliamentary alliance between CiU and the PP was formed. However, further steps toward the fiscal autonomy of regional governments were adopted. In 2001, a new fiscal agreement between the Spanish government and fifteen autonomous communities (including the Catalan) was reached. By virtue of this agreement, autonomous communities directly receive 33 percent of the national income taxes and 35 percent of the value‐added taxes that are collected in their respective territories. In addition, regional governments will now have full legislative control over their shares of income taxes.

The CiU's electoral success has generated opportunities for participation in governing coalitions, and this has given rise to internal strains within the party. Before the 1993 general election, the CiU was riven by the Socialist party's offer of a coalition. After an intense debate on the issue at the October 1992 congress of Convergència Democràtica de Catalunya (a constituent of CiU), the General (p.251) Secretary of CDC, Miquel Roca, resigned on the grounds that he favored closer cooperation with the Socialists.36 The issue arose again after the election when the Socialist party repeated its offer. The CiU still refused in principle, but agreed to cooperate on a case‐by‐case basis. This strategy was controversial, both within the moderate wing of Convergència and particularly within its Christian democratic wing, Unió Democràtica de Catalunya. However, the prestige of the President of the Catalan government, Jordi Pujol helped to neutralize dissent and establish a stable pattern of parliamentary association without participation in the Spanish government.

The ongoing conflict between the CiU and the Spanish government over competencies and the CiU's adeptness at exploiting existing political channels created the perception that a CiU‐led regional government is the most effective way to get more for Catalans. This helps to explain the frequency of split ticket voting in Catalonia, with the Socialist party receiving the majority share of votes in national elections and the CiU most often receiving the majority at the regional level (see Montero and Font 1991: 7–34; Riba 2000).

Rather than satisfying demands for political autonomy, Catalan self‐rule, unmediated by overarching political parties, seems to have deepened ethnic identity, reinforced the center‐periphery cleavage, and ratcheted up political ambitions.

In 1978, the CiU bargained for a Spanish ‘state of autonomies’, an arrangement that fell far short of federalism. Twenty years later, the goal of a Spanish federation has been taken up by many CiU leaders. In 1994, a document published by the Catalan government proposes that all organs of the Spanish state in the region should become subordinate to the Catalan government which would assume nearly all the competencies of the Spanish executive within Catalonia. The Catalan President would antecede the Spanish Prime Minister as the main representative of the Spanish State in Catalonia. In the proposed framework, the Catalan President would have a direct relationship to the Spanish King, who would continue to play a symbolic, moderating role.37 This proposal which points in the direction of a loose Spanish confederation, is consistent with statements by Jordi Pujol, President of Catalonia, stressing that the situation prior to 1714, in which the monarchy was the chief link among the different parts of Spain, is more suitable for Catalonia today than the current system.38 In July 1998, in the Declaration of Barcelona, the CiU joined the PNV and the Bloque Nacionalista Galego in a common demand for the recognition of the national status of Catalonia, Euskadi and Galicia, and for a deepening of the devolution process. This declaration was followed by other steps, such as the Agreement of Santiago de Compostela, which demanded a new (p.252) interpretation of the Spanish Constitution in order to fit the multinational character of Spain.39

The Catalan elections of 2003 and the Spanish general elections of 2004 involved crucial changes in the relative importance of Catalan nationalist parties. They also created a new scenario for the advance of demands of autonomy. After the 2003 Catalan elections, and even though the CiU attained more seats than any other party, a coalition government including the PSC, the nationalist ERC, and the left‐wing Iniciativa per Catalunya‐Els Verds was formed. This new government was presided over by the former major of Barcelona, the Socialist Pascual Maragall, and included also a strong institutional presence of the ideologically independentist ERC. The 2004 Spanish general elections further increased the political importance of the ERC. Not only did this party experience a sharp rise in these elections, but also became a key and necessary parliamentary ally for the Socialist party. These changes in the Catalan and Spanish political landscape are determining important institutional changes. These changes include the development of a new statute of autonomy for Catalonia and the initiation of a process of constitutional reform in Spain.

Toward Multiple Identities in the Basque Country and Catalonia

Might multilevel governance affect territorial identities as well as the intensity of territorial conflict? In this section we examine the extent to which individual identities have been transformed in the context of multilevel governance. Our guiding hypothesis is straightforward: individual identities to a territorial community are strengthened by experience of legitimate authority exercised by actors representing that community. Our presumption is that political identities are somewhat flexible over time, and that they are shaped by the political activities in which citizens are engaged—as participants in the body politic, as voters, as members of social movements, interest groups, and associations that are mobilized in the territorial arena, and as subjects of authoritative decisions.40 To the extent that competencies are diffused across different territorial levels within a system of multi‐level governance, we hypothesize the emergence of multiple identities, that is coexisting identities to several territorial communities.

Historical variation in identity is almost certainly linked to political conflict. Exclusive national identities emerged with the development of strong states that were intent on gaining the total allegiance of those living in their territories. In the experience of Western Europe, war was (p.253) the crucible in which local, regional, and national identities were melded into strong national identities, pitting one nation against another. Multiple identities, by contrast, arguably moderate conflict among communities to the extent that they share an overarching loyalty. The logic of multiple identity appears to parallel that of cross‐cutting cleavages which diminish the intensity of conflict because they join members of otherwise contending groups (Lipset 1981/1959).

There is some evidence for the existence of such multiple identities among Europeans. Surveys conducted by Eurobarometer in November 1991 and May/June 1995 find on average that individuals report as strong an attachment to their region as to their country. Only in Denmark, Finland, Ireland, and the United Kingdom is attachment to country stronger than attachment to region, while in Greece, Italy, and Portugal the two were evenly matched. The surveys include the usual battery of social background questions, and a study commissioned by the EU reports that strength of regional attachment is positively associated with youth, education, and positive attitudes to the EU (Ravet 1992: 26).

However, data on change in individual orientations over a number of years is difficult to come by. The questions on territorial attachment that were posed in Eurobarometer surveys in 1991 and 1995 were not asked before, nor have they been repeated since. In searching for comparable data over time, one faces a rather sharp trade‐off between breadth of territorial coverage and comparability over time. Fortunately, survey items included in a number of DATA S.A. surveys in the 1970s and 1980s were repeated in surveys carried out by the Centro de Estudios Sociológicos in 1996. This gives us the possibility of examining changes over a period of critical political change in two regions that are of particular interest to us (see the footnote on Table 6.1 for details). The first survey was conducted in 1979, the year in which the statutes of autonomy for Catalonia and the Basque Country were passed. The second survey was conducted roughly fifteen years after the establishment of regional governments in the Basque Country and Catalonia. Finally, the third survey was conducted during Aznar's second term as prime minister, in a period marked by a PP absolute majority in the Spanish parliament and serious polarization on the Basque political scene. Tables 6.1 and 6.2 show the overall frequencies of territorial identities in the Basque Country and Catalonia.

These tables reveal that from 1979 to 1996 the percentage of people having balanced multiple identities, that is, who claim to be both Spanish and Basque or both Spanish and Catalan, increased by 10.2 percent in the Basque Country and 3.6 percent in Catalonia.41 From 1996 to 2002, this percentage increased by 0.7 percent in Catalonia, but (p.254)

Table 6.1 Identities* in the Basque Country (Comunidad Autónoma Vasca)





Only Basque




More Basque than Spanish




Both Basque and Spanish




More Spanish than Basque




Only Spanish




Don't know/no answer




N =




*Respondents in the 1979 survey were asked to choose among the following: ‘Español,’ ‘más español que vasco,’ ‘tanto vasco como español,’ ‘más vasco que español,’ and ‘vasco.’ Response options were worded and ordered somewhat differently in the 1996 CIS survey: ‘Me siento únicamente español,’ ‘me siento más español que vasco,’ ‘me siento tan español como vasco,’ ‘me siento más vasco que español,’ or ‘me siento únicamente vasco’.

Sources: For 1979, Linz 1986: 43 (based on data from DATA survey for Richard Gunther, Giacomo Sani and Goldie Shabad, as described in Gunther, et al. 1986: 425–45); for 1996, CIS survey (2,228); and for 2002, CIS survey (2,455).

Table 6.2 Identities* in Catalonia





Only Catalan




More Catalan than Spanish




Both Catalan and Spanish




More Spanish than Catalan




Only Spanish




Don't know/no answer




N =




*Response options in the two surveys were identical to those described in Table 6.1, except that ‘catalán’ replaced ‘vasco.’

Sources: For 1979, Linz 1986: 43 (based upon data from DATA survey for Richard Gunther, Giacomo Sani, and Goldie Shabad, as described in Gunther et al. 1986: 425–45); for 1996, CIS survey (2,228); for 2002, CIS survey (2,455).

increased by 2.5 percent in the Basque Country. Table 6.3, which combines responses of all those who have multiple identities, whether balanced or one‐sided, shows dramatic changes from 1979 to 1996. In this period, in the Basque Country multiple identities rose by 25.9 percent, and in Catalonia by 23.8 percent. By contrast, from 1996 to 2002 multiple identities dropped 14.1 percent in the Basque Country and 4.5 percent in Catalonia. However, Basque figures for 2002 must be (p.255)

Table 6.3 Percentages of respondents in the Basque Country and Catalonia with multiple identities*





Basque Country








*These percentages are the sums of those selecting the second, third, and fourth response options in Tables 6.1 and 6.2.

taken with special caution due to the dramatic increase in the percentage of DK/NA answers (from 4 to 15.1 percent). It did not take into account DK/NA answers, and in the Basque Country multiple identities would have dropped from 72.8 percent in 1996 to 65.7 percent in 2002.

In both regions, the proportion of the population who regard themselves as exclusively Spanish has declined sharply and is now small: 5.4 percent in the Basque Country and 12.9 percent in Catalonia. Likewise, in Catalonia, the proportion who are exclusively Catalan is just 11 percent. In the Basque Country it is higher, but has fallen from 38 percent in 1979 to 20.7 percent in the 1996. A large majority of Catalan and, in more recent years, Basque citizens, view their peripheral and Spanish identities as coexisting rather than mutually exclusive.

Without overinterpreting these data, we can say that they are consistent with the general argument we have developed concerning multilevel governance and our hypotheses concerning multiple identities. We do not merely find that exclusive Spanish identity has declined in the Basque Country and Catalonia, but we find as well that multiple identities have increased in both regions after 1979. However, whereas from 1979 to 1996 there were dramatic increases in nested identities, from 1996 to 2002 multiple identities have declined, particularly in the Basque Country. The 1998 shift of the PNV to a policy aiming at Basque sovereignty is one of the most crucial factors affecting these developments. In addition to this, the fact that Aznar's government adopted a response of uncompromising resistance against these upgraded nationalist demands, presenting itself as the main bulwark against the risks of secession and emphasizing the links between those demands and ETA, may have contributed to the decrease of nested identities. Even if we assume that the decisions made by political leaders were conditioned by their electoral opportunities (the competitive advantages of adopting staunch positions regarding the Basque crisis, both in the Basque Country and outside of it) and institutional capabilities (the absolute majority attained by the PP in the Spanish (p.256) Parliament in 2000), our data show that the strategies adopted by political elites can have direct effects on the evolution of territorial and cultural identities.42


The emergence of multilevel governance in Southern Europe has created a new context for peripheral nationalism as well as new ways for regional governments to press and express their interests. The emerging European polity offers them expanded opportunities for influence and new potential coalition partners in diverse political arenas. In short, peripheral nationalists are being drawn into a variety of multipolar relationships that reward bargaining more than confrontation. On the other hand, the creation of a multilevel polity has by no means tamed demands for regional empowerment. The sheer political weight of several regional movements has increased over the past three decades in Italy and especially in Spain, and regionalization is on the agenda even in highly centralized countries, such as Greece and Portugal.

Over the past decade, the channels available to regional actors have multiplied beyond recognition. The shift of competencies to the European arena has exerted a powerful magnetic force on subnational actors, particularly those in Spain which are entrenched within their national political arena. Alongside the Committee of the Regions, established under the Maastricht Treaty, and encompassing subnational actors across Southern Europe, are less publicized, but more potent, channels of regional representation: regional offices, transnational regional coalitions, and associations of every kind. Regional actors may now mobilize in a trans‐European polity, with diverse associational and coalitional possibilities, alongside their respective national polities.

The cases of the Basque Country and Catalonia suggest that nationalist movements can escape from the view that the national state is the dominant, indeed the only, political form in which national aspirations can be realized. The emergence of multilevel governance encompassing states and regions within an overarching polity poses a competing vision in which regions and supranational actors are empowered at the expense of central states. Scenarios of multilevel governance do not challenge the defining feature of states as institutions monopolizing legitimate coercion in a given territory. But this minimal notion of state sovereignty no longer captures the realities of power in Europe, and it is less and less useful for distinguishing among alternative modes of governance. As both the former Basque regional president, José (p.257) Antonio Ardanza, and the former leader of the Northern League, Gianfranco Miglio, have pointed out, regionalization and European integration are eroding sovereignty and paving the way for a new kind of polity in which the material, legal, and symbolic character of the state will be transformed (Ardanza 1993; Miglio 1993).

European integration poses an alternative to national states as the basis for ordering political life in contemporary Europe. To the extent that regions find a way of gaining meaningful autonomy in the European multilevel polity short of statehood, so the national state is less likely to be perceived as the culmination of ethno‐territorial ambitions. Even from the perspective of nationalist parties which have (at least rhetorically) separatist leanings, the new European polity creates a framework in which the national state loses its exclusive appeal, both symbolically and as a strategic goal. Along these lines, Jordi Pujol, the former President of Catalonia, has argued that Catalonia's problems will not be solved through secession, but rather by restructuring the Spanish political system in a way that reflects its multinational character (El País, February 9, 1994). The weakening of state control has led some peripheral nationalists to argue that traditional demands for independent statehood are becoming irrelevant. In José Antonio Ardanza's words, ‘we are ready to untie the knot that linked nation and state’ (Ardanza 1993: 38).

This is not to suggest that peripheral nationalist movements are withering away. Separatism has become weaker, but territorial movements in general are at least as assertive as they were in the past. Territorial identities are, if anything, becoming stronger, not weaker, in the process of European integration. Instead of subsuming national identities under a common European identity, the continued existence of strongly entrenched state executives in the Council of Ministers may actually enhance the sense of divergent national interests. This, we have argued, sharpens subnational identities, as distinct ethno‐territorial groups countermobilize to press their particular demands at the European level through a variety of formal and informal channels. Regional empowerment is potentially self‐reinforcing. Competencies not only provide regional governments with the capacity to make policy, but they give them the opportunity to deepen their own bases of support. Political patronage, cultural, and linguistic policies are as important for regional governments as they were for state builders in shaping identities. As Levi and Hechter indicated, regional empowerment is immensely helpful for the electoral growth of ethno‐territorial groups, which can now allocate scarce resources to foster ethnic mobilization (Levi and Hechter 1985).

Regional empowerment has led the Catalan, Basque, and even Galician governments to ratchet up their demands. The proposals of (p.258) the former Galician regional President (a Conservative politician who was a minister in the Franco regime) and of the former Catalan President for a thoroughgoing federal transformation of the Spanish state indicate that subnational demands have grown wider since the Constitutional reform of 1978.

While we diagnose a common direction of development across Southern Europe, there are few grounds for supposing that these countries will converge. In the first place, their starting points are vastly different. The powers of subnational governments and the degree of regional mobilization vary from country to country about as much as any political variable one can think of. Second, the creation of a common set of political opportunities in the EU has had very different effects depending on initial dispositions. Entrenched subnational governments are best positioned to exploit the changing structure of political opportunity. New channels for the mobilization of regional governments at the European level have actually widened the gap between strongly articulated Spanish regional governments and weak subnational units in Greece and Portugal which are virtually silent. Heterogeneity of territorial relations across Southern Europe is therefore not a passing phase, but a durable feature of the emerging Euro‐polity.



(1.) For a more detailed discussion of state‐building and European integration, see Marks 1997.

(2.) For a detailed analysis of the similarities in the economic, social, and political histories of Southern European countries, see Giner 1986.

(3.) On the characteristics, types, and implications of multilevel governance, see Hooghe and Marks 2003.

(4.) On the conflicts leading to the special political situation of the Azores and Madeira, Gallagher 1979. See also Guilherme Reis Leite 1992.

(5.) These transfers enabled the autonomous regions to pass legislation in matters of their special interest, to regulate national laws, and to initiate legislation in the national parliament. See Pereira 1995.

(6.) However, not all the members of the regional assemblies would be elected by direct suffrage. Some would be elected by the municipal assemblies of each region. See Opello 1992: 162–86.

(7.) See Papageorgiou and Verney 1992: 139–61; Verney 1994; Featherstone and Yannopoulos 1995; Ioakimidis 1996.

(8.) For instance, whereas the central administration employs more than 2,000,000 civil servants, only about 80,000 civil servants work for the regional administration (see Desideri 1995). Although the regions control between 20 and 25 percent of public spending, more than 90 percent of (p.259) that amount is directly transferred from the central government. In addition, most of the regional income has to be directed toward areas in which regional governments lack autonomous competencies (see Hine 1993: 268–9). On the characteristics of the taxation system of the special regions, see Desideri 1995.

(9.) These demands were expressed in the documents ‘Per un nuovo stato regionale’ (1991) and in the ‘Carta delle Regioni d'Italia.’ See Leonardi 1992.

(10.) Even though some peripheral movements attempted to establish links with national political projects, as some wings of Catalan nationalism did in the first third of this century, the centralizing trends of national states promoted bipolar conflicts between peripheral movements and their respective states. We use the term ethno‐territorial movement to distinguish ethnic movements that are territorially based. Hence, it enables us to exclude a movement that has a distinctly ethnic character, but whose constituency lacks a connection with a defined territory, a category which would include many religious and almost all immigrant groups.

(11.) We should note that the concept of multipolarity here is different from the concept as it is used by neorealists, in that we encompass not only nation states, but a variety of institutions having heterogeneous constituencies and competencies across different levels of government often in non‐exclusive territories.

(12.) On the fluid and complex character of the relationships among territorial actors within the new European polity, see Hooghe 1995.

(13.) On pragmatic bargaining among subnational, national, and supranational institutions in the Belgian case, see Hooghe 1995.

(14.) See Moravcsik 1993: 473–524; and Marks, Salk, Ray, and Nielsen 1996: 164–92.

(15.) In this respect the Basque Country and Catalonia are unlike Northern Italy, where party‐political mobilization, in the form of the Northern League, is a recent development.

(16.) On the causes and characteristics of the process of regionalization in Spain see Pérez‐Díaz 1993: 194–204. An analysis of the relationships between the party systems of Catalonia and the Basque Country, and the Spanish democratic transition can be found in Gunther, Sani, and Shabad 1986.

(17.) On the characteristics of Basque nationalism and the diffusion of separatist orientations in the movement, see Linz 1986; Díez Medrano 1995; and Juaristi 1997. A study that emphasizes the rhetorical character of Basque separatism and traces its historical origins is Juaristi 1994: 127–30.

(18.) After the absorption of a part of a former left‐wing nationalist organization, the current name of the PSOE‐PSE is PSE (Partido Socialista de Euskadi).

(19.) ‘Las naciones—y me refiero tanto a las que se identifican con un Estado propio como a las que no disponen de él—dejarán de ver en el Estado su único modelo de realización. El mantenimiento o la consecución de su soberanía y estatalidad dejará de ser un objetivo prioritario y obsesivo.… Pensamos que, en la nueva Europa que queremos construir, es posible y (p.260) conveniente concentrar nuestros esfuerzos en una construcción nacional descargada de obsesiones estatalistas. Por nuestra parte, estamos dispuestos a desatar el nudo que unía indisolublemente la Nación al Estado’ (Ardanza 1993: 37–8). For the independentist orientations of the President of the PNV, Xabier Arzalluz, see El País, April 3 and April 7, 1994. Interestingly enough, some other declarations by Xabier Arzalluz point in a direction similar to that of Ardanza. See Unzueta 1994: 166.

(20.) In the Spanish general elections, the percentage of vote for HB in the Basque autonomous community dropped from 17.78 percent (1986) to 16.98 percent (1989), 14.83 percent (1993), and 12.47 percent (1996). In the autonomous elections, HB attained 17.47 percent in 1986, 18.33 percent in 1990, and 16.29 percent in 1994. Data excerpted from the official website of the Basque government. See http://www. euskadi.net.

(21.) With the exception of the Basque branch of Izquierda Unida, Izquierda Unida‐Ezker Batua (IU‐EB), all the political forces that joined the Lizarra agreement had nationalist orientations. IU‐EB experienced a dramatic setback in the regional and local elections that followed the Lizarra agreement. In the Basque parliament, IU‐EB lost four out of its six seats in the 1998 elections.

(22.) Excerpt from the Declaración de Lizarra, September 12, 1998, see http://www.iu‐ebberdreak.com/docu/lizarra.html.

(23.) See http://www.gara.net/99/eta/gehi2.2k.html.

(24.) Euskal Herriko Udal Ordezkarien Biltzarra. This structure allowed nationalists to surpass the borders that separate all the Basque territories. In addition, this assembly aimed at strengthening the position of nationalists by giving the same weight to small villages (where nationalists are strong, particularly in Guipúzcoa and Vizcaya) and large capitals (where nationalists are weaker than non‐nationalists).

(25.) In the 1998 autonomous elections EH received 17.91 percent of the vote in the Basque autonomous community, 1.62 percent more than in the 1994 autonomous elections.

(26.) After the elections of 1999, only one of the four capitals (Bilbao) of the Spanish territories claimed by Basque nationalists remained in nationalist hands. At the time of writing, the mayors of Vitoria‐Gasteiz and Pamplona are PP and the mayor of San Sebastian is a socialist. In the general election of March 2000, the PP received a greater share of the vote than the PNV in every Basque provincial capital city and in most major towns. In the provincial capital of Vitoria, the PP received 40.9 percent of the vote, whereas the PNV received just 16.8 percent.

(27.) El País, November 28, 1999. The main reason ETA alleged to suspend the truce related to the slowness and lack of determination of the PNV and EA in the fulfillment of their compromises. According to ETA, a new agreement was reached in August 1999 among ETA, the PNV, and EA. By virtue of this agreement, Basque nationalist forces would work to call for a national Basque parliament elected by citizens of all Basque territories. That parliament should elect the Basque president and lead the process toward a (p.261) sovereign Basque state. However, the PNV argued that neither the PNV nor EA actually did sign that agreement because they considered it unfeasible. See El País, November 28, 1999.

(28.) EH dropped from fourteen to seven parliamentarians, and from 17.9 to 10.1 percent of the vote. The coalition formed by PNV and EA attained thirty‐three seats, one more than the seats won by PP and PSE.

(29.) In October 2001 IU‐EB joined the PNV‐EA government, assuring it thirty‐five of the seventy‐five seats of the Basque parliament.

(30.) Nationalist parties received 57.8 percent of the vote in the 2001 regional elections.

(31.) On the emergence of Catalan nationalism, see Solé Tura 1970.

(32.) The competing Catalan nationalist party, Esquerra Republicana de Catalunya (ERC), has played a secondary role, receiving less than 10 percent of the vote in regional elections. The ERC has campaigned for separatism, though it has not supported violence. In this critical respect, the Catalan political scene is very different from that of the Basque Country.

(33.) According to CiU, 28 percent of the money the Spanish state extracts from Catalonia as taxes never return to this autonomous community. See http://www.convergencia.org/pactefiscal/index.html.

(34.) This tension was aggravated by the PP attacks on the linguistic policies of the Catalan government and on the support that the CiU gave to the González government.

(35.) The agreements of the PP and nationalist parties involved many other decentralizing reforms. These reforms included the transference of competencies in employment policies and infrastructures, as well as regional access to the Spanish policy‐making process in European issues.

(36.) See Cambio 16, no. 1090, October 1992. See also Antich 1994.

(37.) A version of this document can be found in El País, February 11, 1994. Although Jordi Pujol refused to dismiss the document, he acknowledged that its goals were probably too ambitious.

(38.) Josep Antoni Durán Lleida, leader of Unió Democràtica de Catalunya (the party that forms part, with Convergència Democràtica de Catalunya, of CiU) holds a similar project: a confederation of Catalonia with the Basque Country, Galicia, and the rest of Spain. That view explicitly excludes independence (see Cambio 16, no. 1098, December 1992: 31).

(39.) In the 1999 Catalan elections the CiU decided not to run on the basis of the Declaration of Barcelona, and, so far, it has not engaged in a single confrontation with the Spanish government in order to bring about fundamental restructuring of the Spanish state. Although the declaration of Barcelona was never suspended by the parties that signed it, it lost public relevance after 1999 due to both the different tracks followed by Basque and Catalan nationalists, and the dependence of the CiU government on PP support in the Catalan parliament during the final years of the Pujol governments. In December 2004, a new declaration was issued by the BNG, CiU, and PNV calling for the establishment of a multinational state. The political impact of this declaration was clearly limited by the (p.262) fact that the Catalan ERC, now a key player in the Catalan and Spanish political arenas, did not sign it.

(40.) One may also expect that the exercise of authority that is regarded as illegitimate may intensify identities in an alternative territorial community.

(41.) Other surveys conducted in 1990 show very similar figures for balanced nested identities. See Ferrando, López‐Aranguren, and Beltrán 1994: 16.

(42.) It must be kept in mind that changes in the strategy of the Basque Nationalist Party were not triggered by growing public demands for self‐determination nor, as the CIS 1996 data clearly show, by the development of exclusively Basque identities.