Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Introduction

This paper aims at presenting a new research project, which is, however, still at the very beginning. Therefore, empirical results cannot be presented. Rather, in the paper the basic analytical framework is outlined that is to be applied to the cases, thereby hoping to open a discussion on basic concepts, relevant variables, and suggested causal relationships.

Recently, we can observe an increasing number of regionally concentrated minority groups that define their group identity based on ethnic, national, religious, ideological, or linguistic unity. Those groups live, however, on the territory of a larger nation state. Consequently, requests for autonomy or threats of separation generally accompany those group feelings and find their expression in acts of conflict, ranging from a lower intensity—such as declarations or demonstrations—to outright violence in the form of terrorist attacks, if not civil wars. How can such groups be successfully integrated without suppressing their group identity and without jeopardizing the integrity of the nation state? This development poses a major challenge for the Western-type nation-state (Benz 2008; Obinger et al. 2005), which seemed to have triumphed ultimately after the breakdown of communism 20 years ago.

As a possible remedy to appease intrastate conflicts of the kind described, it has often been proposed to introduce or enlarge federal institutions and to decentralize competencies from center to periphery, for it is hypothesized that by granting minority groups more rights of autonomy, they can be successfully integrated (Agranoff 1999; Bakke and Wibbels 2006; Bermeo 2002; Chapman and Roeder 2007; Elkins and Sides 2007). Most recently, similar proposals have been made with regard to conflicts in Nigeria and Iraq (Elkins and Sides 2007; Schrijver 2006) for, traditionally, it is regarded as one particular strength of federal arrangements to safeguard unity in diversity. On the other hand, this hypothesis is violently contradicted theoretically, as well as empirically. It is argued that more subnational autonomy provides more resources for the leaders of protesting groups who will consequently reinforce their protest (e.g. Hale 2008). Also, the knowledge that their protest has successfully induced the central government to devolve competencies may fuel their hopes to gain even more if they continue. Both hypotheses are plausible and have been proven empirically but contradict each other fundamentally (for a balanced discussion of the two hypotheses, see, e.g., Elkins and Sides 2007). In the light of those contradictions, which recommendations can be given to governments trying to deal with multinational conflicts?

This is the basic question guiding the research presented here. Obviously the straightforward recommendation that the introduction of power-sharing institutions and the devolution of competences appease group conflicts is not always true. In order to arrive at a better understanding of the dynamics of the policies of accommodating group conflicts, it is necessary

  1. 1.

    To enlarge the temporal perspective and take into account a longer period of time. Short-term reform processes and reactions to them are embedded in a long historical stream of struggles for competences between center and periphery. In this long-term process, sequences or phases of group request, reforms initiated by the central state, and reactions by the protesting groups can be interpreted as waves of action and reaction following each other in shorter or longer distance.

  2. 2.

    To reconstruct the sequence of events, as well as the actor orientations and strategic decisions in great detail in single cases. A narrative of the things that happened, when they happened, and why they happened from the perspective of the actors involved is the most promising way for theory construction, for only based on this evidence is it possible to derive hypotheses about the causal mechanisms at work that represent at least distantly the real complexity of causal and temporal interrelationships.

It is thus the aim of the research project to come to an understanding of the dynamics and mechanisms at work guiding the interplay between regional and central actors in multinational states. The conditions under which requests are formulated, reform policies are selected and implemented, and the effects that these policies have on state-periphery relations are to be specified in greater detail.

Empirically, we proceed by conducting comparative case studies. Although variants of federal institutions are currently being discussed in real-life politics, especially in conflict-prone societies in Africa, in the Middle East, and in Eastern Europe, we start by limiting our research interest to established democracies in Western Europe. It is of course problematic to transfer results gained in one type of political system to another—the so-called travelling problem (Sartori 1970) is well known to all comparativists. But acknowledging that the processes under investigation are highly complex, by comparing cases within Western Europe, we can control at least some of the basic systemic variables. Cases are selected in Spain and in the UK. Spain is most interesting for investigating those dynamics of minority requests, decentralization reforms, an asymmetric distribution of competences, and renewed ‘rounds’ of conflict. While in the short run conflicts seem to be appeased by reforms, in the long run the autonomous communities formulate new and farther reaching requests, thus pushing the conflict between center and periphery further and jeopardizing the stability of the Spanish state (Agranoff 1996; Colino 2009; Grau i Creus 2000). In the UK, where contrasting cases are selected, conflicts are seen to be far less violent than in Spain, and the overall tendency of recent developments is less disintegrative (Bradbury 2006; Jeffery and Wincott 2006; Trench 2005). Thus, even if structural conditions are similar for regions in Spain and in the UK, there must be a reason for those differences, which deserves further investigation.

In the following, I will first outline our analytical framework (‘Analytical Framework’ section). In particular, I specify the temporal horizon of the analysis (‘Sequences of Events’ section), as well as the theoretical background for distinguishing reform approaches, namely the distinction between integration and accommodation (‘Reform Options’ section), which is basically rooted in the old dispute between liberalism and communitarism in political theory (Kymlicka 2005). The mechanisms at work that are to be reconstructed in our research are rather complex and cannot be adequately represented by bivariate causal relations. Thus, in order to understand the logic of those complex relationships, the sequences of events under investigation need to be embedded in broader historical, institutional, structural, and cultural contexts. Therefore, in ‘Relevant Influences on the Reform Process’ section, those sets of variables are explained that are deemed to be important in influencing the development of the sequences of events. ‘Analytical Framework’ and ‘Relevant Influences on the Reform Process’ sections outline together the theoretical conception and assumptions of our research, which however are still waiting for their empirical test and possible reformulation. The ‘Case Selection’ section explains briefly the rationale of case selection. Instead of providing a (necessarily premature) conclusion, the ‘Research Agenda’ section gives then a short overview of the future research agenda.

Analytical Framework

Analytical Perspectives

In approaching the cases under investigation, I take a certain analytical perspective in two respects:

  1. 1.

    The temporal perspective: The processes to be analyzed have a very long time horizon. Singular events are embedded in a long-lasting power struggle between center and periphery. It is thus necessary to take into account historical roots of minority groups, as well as long-term effects of single reform events. When talking about the ‘process’, this long-term perspective is intended. From this process, shorter periods of time can be distinguished. The term ‘sequence of events’ is used for describing the phases immediately before, during, and after a decentralization reform, which form in their whole a kind of ‘wave’ in the long-term process. The single events can be manifestations of protest, requests by minority groups, reactions of the central government, formal or informal rounds of negotiation and decision, and finally reform acts, agreements, or treaties. The sequence in which those events occur is likely to be of great importance for the results of those ‘waves’ (Falleti 2005).

  2. 2.

    The individualistic perspective: It is not intended to impose one ‘objective’ external perspective on strategic options of action or in evaluating the possible ‘success’ of reform measures. Rather, it serves our aim to reconstruct processes and dynamics that preferences, strategies, options for action, and situational definitions are seen through the lenses of relevant actors. Relevant actors are political decision-makers (elites) at the levels of the central state, as well as the region. Focusing on elite perceptions does not necessarily mean to neglect the influence of popular opinions. Rather, it is assumed that—although a broad mobilization of the population on autonomy issues is possible and possibly important for their success—the relevant actors are elite actors. This is plausible because the subject matter is complex and relatively remote from everyday living experiences. Thus, elites are more likely to take a reasoned opinion and then to act as political entrepreneurs and opinion leaders. In this regard, the elite perceptions are more important for understanding the dynamics of the process.

Sequences of Events

As was said, within the long-term process of conflict, reform, and (possible) appeasement, the focus of the case studies is on sequences of events around one or several specific reform measures. These reforms are usually the result of negotiations between a regional group and the central government. They may be more or less far-reaching, decentralizing many or few competences, or granting extended or limited rights of autonomy. The distinction between competences and rights of autonomy is meant to highlight the difference between political and fiscal competences on the one hand—rights of self-rule and rights of shared rule (Marks et al. 2008), as well as tax levying and spending powers—and rights of cultural autonomy on the other—e.g., official recognition of a native language, independent organization of schools, holidays, or religious services (e.g. Roach 2005). Furthermore, those reforms can aim at integration or rather at accommodation of the protesting groups (see the ‘Reform Options’ section).

From this analytical starting point, it is also taken into consideration how those measures are implemented and which effects they have—whether and how they influence subsequently the self-perception of the regional group, their requests, or potentially further protests. From the perspective of the regional actors, the success of the reform measure will be evaluated according to the degree to which their requests have been satisfied. From the perspective of the central actors, the success of the reform measures will be evaluated according to the degree of national stability and societal peace that they helped to (re-)establish. Indicators for the level of acceptance of reform measures can be, e.g., the results of subsequent regional elections (whether radical/separatist parties gained more votes than before), as well as acts of protest (political manifestoes, founding of new parties, demonstrations, or violent protest to a varying degree).

On the other hand, the focus is also directed ‘backwards’, trying to understand which actions or events lead to the reform measure. Generally, it can be assumed that such reforms did not come out of the blue sky but that they have a history of requests, protests, and negotiations. This history is most enlightening in trying to understand which things happened, how they happened, and which results they produced. It can be useful, e.g., to point out which requests were formulated by the minority groups and on which arguments they were founded. In this regard, the look backward includes an account of regional history, as well as a reconstruction of the mechanisms of minority mobilization; an analysis of the forms, frequency, intensity, and sequence of protest; a narrative of negotiations—between governments, party leaders, or individual politicians at the two levels of government—were they formalized or rather personalized? All those sets of variables are necessary to get a more complete picture and a better understanding of the mechanisms at work in this struggle between center and periphery.

Reform Options

The choice for implementing any reform at all, as well as for selecting a particular type of measure depends on the ‘framing’ of the actors involved at both levels of government. A reaction to minority requests can range from suppression of the conflict in order to reinforce the status quo over more or less far-reaching changes in the territorial order to granting of rights and competences to the regional minority to outright secession of the protesting region. While the two extreme cases (suppression of protest and secession) are no dominant options in Western European States, the ‘middle ground’ of the dimension still encompasses various degrees of acceptance of group differences. Two major ‘paradigms’ are ‘integration’ and ‘accommodation’ (McGarry et al. 2008). In an integrationist approach, group differences are accepted but relegated to the private domain. Integration aims, thus, not at suppressing group differences but on a liberal ideology of guaranteeing individual equality of rights instead of granting collective rights to groups. In the framework of this ideology, it is the state’s duty to protect individual rights, but not to secure the survival of groups.

In contrast to this, an accommodationist approach allows the manifestation and realization of group differences also in the public/political domain. ‘Accommodation, minimally, requires the recognition of more than one ethnic, linguistic, national, or religious community in the state. It aims to secure the coexistence of different communities within the same state, though supporters of accommodation may support secession or partition if accommodation is impossible’ (McGarry et al. 2008: 52).

In relation to the territorial dimension of group requests for reform, it can generally be assumed that a regional group aims at getting possibly far-reaching rights of autonomy, while actors of the central state tend to concentrate power at the central level. Based on this assumption, the distinction between an integrationist and an accommodationist attitude allows for the formulation of several hypotheses:

  1. 1.

    Regional actors will request accommodationist reforms, while central actors will rather offer integrationist measures.

  2. 2.

    Regional actors will request integrationist reforms only if they expect that accommodationist requests will have no chance of success.

  3. 3.

    Central actors will offer accommodationist measures only if they expect that integrationist offers will not be accepted.

Consequently, the requests that regional actors formulate, as well as the offers that central actors make, depend heavily on their relative assessments of the chances of success. Expected chances for success, again, depend on their relative bargaining power. The bargaining power of regional actors can be low, e.g. when they cannot make credible threats of secession, when they lack internal or external support, or when they are economically dependent on transfer payments by the central state. The bargaining power of central actors, on the other hand, can be low when the central government needs regional support in coalition building, when the central state is economically dependent from the region, or when they face an increased pressure for action due to credible secession threats or upcoming elections.

Relevant Influences on the Reform Process

The perceptions of the actors involved, their estimates of their bargaining power, as well as the general climate between the conflicting groups are obviously influenced by several external factors. Those contextual factors need to be taken in particular consideration in understanding the dynamics of the process. Although empirically the distinction is sometimes hard to take, analytically it is useful to distinguish between institutional and cultural contextual factors.

Institutional factors serve to describe the situational environment in which the conflict takes place. The most important among them are:

  • The initial level of rights and competences available to the group as compared to the aspirational level. It can be described by (a) the list of competences and rights of which they dispose and (b) the degree of asymmetry in the distribution of rights and competences between the regions. Furthermore, it makes a difference (c) whether those rights and competences extend to self-rule or include shared rule, i.e., whether the group has, e.g., veto power in subsequent rounds of institutional reforms (Petersohn 2011); asymmetry might also be seen by some actors as a value in and of itself that needs to be maintained, in a way that not the absolute extent of rights of autonomy and political competencies is relevant but the distance to other regions (Hombrado 2011), an attitude that is likely to generate a dynamic of its own;

  • The degree to which power-sharing institutions (Lijphart 1969, 2004; Norris 2002) form an integral part of the institutional structure. Bicameral parliaments, proportional electoral systems, and a federal organization of the state are generally assumed to curb intragroup conflicts;

  • The existence of a vertically integrated party system providing for cross-cutting cleavages (Goodin 1976) as compared to a regionalized party system that might reinforce separationist cleavages (Swenden 2009).

Cultural factors are not always easily distinguished from institutional factors. Among those are cognitive and behavioral patterns shaping the manner in which conflicting groups interact:

  1. 1.

    If a state has, for example, a long tradition of consensus democracy (Lehmbruch 1967) there is obviously an overlap to the power-sharing institutions mentioned above, and conflicts can be more easily accommodated by securing the participation of all relevant groups in decision-making. Such consociational features can be more procedural than institutional. A perfect example is Switzerland, where the consociational culture is deeply engrained in politics and society.

  2. 2.

    Another important aspect of ‘culture’ is the degree of internal and external support that minority groups receive for their requests. First of all, the popular support within their own group is relevant for underpinning requests for more autonomy. In this regard, the distinction is useful whether the movement is only elite driven or whether there is a broad popular sense of belonging and of suppressed needs for autonomy. Second of all, external support is just as important. In this regard, the EU policy played a major role in the past 20 years. With its explicit regional policy, the normative basis for regional requests for more autonomy, as well as intraregional cooperation, was strongly enforced (Balme 2004; Börzel 2002).

Case Selection

In trying to understand complex multivariate mechanisms, the appropriate number and selection of cases are always a tricky problem. Large-N-comparisons cannot be accomplished with the necessary depth, single case, or small-N-studies make it hard to generalize the results. Therefore, I opted for a middle road, choosing a design of a two-by-two comparison, that is, two regions in two states, thereby controlling for a maximum of third variables with a relatively small sample. In comparing Spain and the UK, two of the most interesting cases in Western Europe were selected: Both are traditionally described as unitary states but have a multinational character and have, in the past few years, initiated far-reaching reforms that resulted in altering fundamentally their institutional structure. Their regional units now dispose of a large number of competencies, which are, however, distributed asymmetrically among them. Therefore, Spain and the UK were also denominated as ‘union states’ (McGarry et al. 2008) or as ‘hybrid states’ (Loughlin 2009). This asymmetry is likely to spin the disintegrative dynamic further.

In the second step, two regions in each of the states were selected, making a sample of four regions altogether. One region in each state is commonly known to be a leading actor in the process of requesting more autonomous competences from the state—Catalonia in Spain and Scotland in the UK. Those are contrasted with another region in each state where the strive for autonomy is less expressed, which has a less pronounced autonomous tradition but which profits now from the general political climate of decentralization, enlarging its competences, so to say, by clutching the coattails of the leading regions. There are a number of regions for which this description is suitable. Among those, Wales in the UK and Andalusia in Spain were selected. Both have a relatively weak economy, and both can at least point to several origins of difference from the central state in terms of language and culture.

This way, different couples of cases can be formed that have different traits in common: In terms of historical experience, for example, the Spanish regions have lived a sustained period of massive suppression under the Franco regime, followed by a very liberal democratic constitution, which did not even define the number, borders, or names of the communities (Schrijver 2006). In the UK, the situation was fundamentally different because Wales and Scotland joined the Union voluntarily and Scotland always retained special rights. However, the conservative government from 1979 on, with its restrictive antiunionist politics, hitting the regional industries painfully and was perceived as a kind of siege. In terms of economic strength and concomitant bargaining power, as well as a history of autonomous rights, Scotland and Catalonia have lots in common in contrast to Wales and Andalusia.

One might ask why the Basque Country in Spain and Northern Ireland in the UK were not included in the sample: Both cases are highly insightful for investigating the dynamics of struggles for autonomy. Both regions, however, are populated by cross-border minorities, whose situation is so different from groups that are mainly concentrated within one country, that it seemed wise to reduce complexity and leave those cases out. It is, however, intended to do follow-up research enlarging the sample and including those cases (as well as possibly several others).

Research Agenda

As was pointed out at the beginning of this paper, the research project is only about to get started, thus so far, there is no empirical evidence available. The agenda for the next 2 years’ research will be to investigate those processes in the four selected regions by conducting in-depth comparative case studies. The case studies aim at reconstructing the different phases of the processes as described above and enriching the evidence by taking into account the possibly relevant context factors. The logic of the two-by-two-cases design makes it possible to distinguish idiographic from systematic evidence and to arrive at preliminary hypotheses about the dynamics of minority requests for competences and autonomy, on the one side, and federal politics of accommodation or integration, on the other.