Abstract
A common feature of the present-day constitutions of the Western Balkans is the effort to solve conflicts of ethnic character using predominantly legal tools, mostly in a parliamentarian way. However, the practice shows that most legal regimes based on instruments that give preference to the interests of one or more equally strong ethnic groups can be built mostly to the detriment of democratic states. Effective and functional state institutions and ethnic power-sharing in multi-ethnic states seem to be in conflict with each other. Rule-of-law-based models can only function properly if parties have mutual trust and can solve their internal conflicts through compromises among themselves. Political agreements based on mutual trust are more effective in the long term as legal instruments. There are examples of such arrangements in multi-ethnic states of Western Europe (Belgium, Northern Ireland, and Switzerland). Analysis of the solutions of the Western Balkans countries and their comparison with these Western examples shows clearly that hard legal tools (vetoes) do not soften but sharpen conflicts, while informal arrangements based on mutual trust are more productive.
1 Introduction
Ethnic conflicts and their institutional solutions – often imposed by external powers – have characterized the past and recent history of the Western Balkans, ranging from the 1878 Congress of Berlin to Dayton and Kosovo. After WWII, maintaining ethnic peace was an important political feature of socialism in multi-ethnic Yugoslavia. The country, to which five of today's six Western Balkans countries belonged, was a federal state, but even if the federal units corresponded more or less to the ethnic ones, full ethnic-territorial autonomy was never achieved, partly because of the many ethnically mixed territories. In 1981, the share of the different ethnic groups in Yugoslavia was the following: Serbs (36.3%), Croats (19.7%), Bosniaks (9%), Slovenes (7.8%), Albanians (7.7%), Macedonians (6.0%), Montenegrins (2.6%), Hungarians (1.9%), Yugoslavs (5.4%) and Others (3.56%). Although power-sharing arrangements and autonomies existed on paper, ethnic peace was maintained predominantly by state forces – the tools of the authoritarian regime of the communist leader Josip Broz Tito.
The fall of socialism opened a Pandora's box of ethnic conflict, and the region soon became a battlefield. After the armed ethnic conflicts of the 1990s and 2000s, the region experienced constitutional consolidation. As a price of the peace, various ethnic-based power-sharing methods were introduced into the constitutional systems of the region, based mainly on Western initiatives and advice.
There was strong external pressure both when the constitution of Bosnia and Herzegovina was framed in 1995 and when the federation of Serbia and Montenegro adopted its constitution in 2003. Some scholars tend to say that Kosovo's Constitutional Framework (2001) was also de facto adopted under external pressure, and the constitution that Albania adopted in 1998 was also framed with foreign (chiefly US) assistance. The Western Balkan countries differ in their religious and ethnic background, but their present-day constitutional setups have plenty of similarities due to the adoption of Western standards during the creation of their constitutions (e.g., parliamentary form of government, proportional voting system, ceremonial head of state, independent constitutional court). Citizens did not have a direct say in the adoption of these arrangements. Although independence was achieved by referendum in Bosnia and Herzegovina (1992), North Macedonia (1991), and Kosovo (1991), none of these countries had referendums on their constitutions (such referendums were only held in Serbia in 1990 and 2006, and in Albania in 1994 and 1998).
Of the six Western Balkans countries, I will analyse conciliatory parliamentary solutions in three ethnically complex states: Bosnia and Herzegovina (2013: Bosniaks: 50%, Serbs: 31%, Croats: 16%), North Macedonia (2002: Macedonian: 65%, Albanian: 25%) and Kosovo (2011: Albanians: 93%, Serbs: 1.5%) and compare them with Western European multinational countries' solutions.
A common feature of the present-day constitutions of the Western Balkans is the effort to solve conflicts of ethnic character using predominantly legal tools, mostly in a parliamentarian way. The European Union propagates the model of democratic states based on the rule of law, with guarantees for minorities as the only alternative to separatist and nationalist endeavours. Attractive though that model may be in principle, it is extremely difficult to implement in practice.
In spite of the massive assistance and expertise provided by international donors, the rule of law still lags behind, 1 even if it has certainly struck root. Ethnic-based politics is still a strong group-forming factor in the Western Balkans, even if militant nationalism has lost its appeal to a considerable share of voters. In the past one or two decades, violent ethnic conflicts have been replaced by ethnic confrontation in the political sphere. Some domestic political forces in the region urge the reinstatement of forced constitutions – again – by external powers, for they see this as the only way out of the stalemate. 2 In contrast to this development, equally crucial ethnic conflicts in Western countries were solved over a longer time frame (Belgium: 1960s–2000s, Northern Ireland 1980s–2000s), and, what is more important, based on the cooperation and agreement of local citizens.
2 Theoretical framework of ethnic power-sharing in multinational states
In states inhabited by multiple nations (nation-states with minorities as well as multinational states where more ethnic groups enjoy equal rights), where territorial and personal autonomy has not led to resolution (e.g.: geographical borders cannot be defined unambiguously), democratic principles are usually nuanced by ethnocratic representative elements. In a multi-ethnic or multinational state, instead of differentiating between minority and majority roles, only majority roles exist: multiple groups claim majoritarian positions, forming a grand coalition. Those who are not willing to recognize these political power deals often find themselves outside such coalitions, not belonging to any of the state-building groups.
The following institutional power-sharing arrangements are usually available in nation-states with minorities as well as in multinational states:
-
Electoral or other representative arrangements are the most common (these may either be adopted by formal legislation or based on political agreement):
-
quotas for candidates or within government bodies (parliament, government),
-
‘reserved seats’ (eligibility for a given number of seats irrespective of election results),
-
ethnic advocacy by the upper chamber of parliament,
-
the ethnic-based creation of electoral district borders.
-
-
Special majority voting requirements within representative bodies:
-
qualified majorities (two-thirds, four-fifths, etc.),
-
multiple (dual, triple, etc.) majorities (majority both within a minority group and within the whole organ).
-
-
Procedural arrangements:
-
ex-ante instruments: ethnic vetoes imposed by minority groups before the final act (promulgation) is concluded, conciliatory mechanisms between parliamentary chambers,
-
ex-post instruments: remedies by external bodies (constitutional courts).
-
The strength of these instruments extends from soft, mediational, reconciliation proceedings (Belgium) to the complete blocking of decisions (Bosnia-Herzegovina).
After World War II, scholars were keen to establish rules and guarantees for linguistic-ethnic groups living within a state, combining the above tools in different forms. Arend Lijphart described four (constitutional) legal mechanisms in the 1970s that characterize the parliamentary power partition in multinational states:
-
grand coalition (more ethnic groups participate in the government regardless of the results of elections, proportional to population),
-
minority veto (the majority does not have unlimited power over minorities),
-
commensurate representation (taking into account the share of membership in government bodies), and
-
segmental autonomy (different groups can decide on certain issues – e.g., language, or culture). 3
However, apart from this categorisation, the main thesis of Lijphart is as follows: in order to achieve successful multinational societal existence, it is not the institutional solutions that are determinative but a common, solid agreement between the elite groups that stabilizes the system. If this persists, even states that are ethnically heterogeneous may be democratic.
Lijphart's four recommendations, as listed above, were used precisely during the creation of constitutional institutions and rules in the Ahtisaari plan, which established the state arrangement of the independent Kosovo, 4 but one can recognize their influence in the legal solutions of North Macedonia 5 and Bosnia and Herzegovina as well. In practice, however, Serbs in Kosovo are fundamentally negative towards the state of Kosovo (boycotting elections, non-recognition of the state). The ethnic rights offered them are only able to solve ethnic conflicts at a slow pace; however, collaboration is a better stabilizing factor than any other legal mechanism.
Abdel-Fattah compared two multinational states, Bosnia and Herzegovina, and Lebanon, looking for connections between the formalization of ethnic decision-making by veto rights and political decisiveness, and the efficiency of state operation. The result was that Bosnia and Herzegovina suffer more from legislative paralysis than Lebanon since ethnic vetoes are more formalized and easier to impose. 6 The more formal an ethnic veto appears to be, the harder it is to find a compromise. In Lebanon, the different linguistic-ethnic groups are less well equipped with enforceable laws; instead, cooperation procedures are used. It needs further research why Lebanese equilibrium was not durable: at a certain point, Shiite and Sunni groups demanded more rights, which was one of the reasons for the war in 1975.
The Yugoslav Wars in the 1990s also provided an opportunity for Roeder and Rothchild to examine the problems of multinational states. After examining agreements on the division of power after conflicts caused by war in ethnically divided states, they concluded that the same institutions which seemed appropriate for putting an end to a violent conflict are more likely to prevent the consolidation of peace and democracy in the long term. They described the inconsistency between short-term benefits and long-term expenses as the dilemma of power sharing. 7 As we will see, this precisely describes the post-Dayton situation of Bosnia and Herzegovina, called ‘institutionalized ethinicism’ by Bieber. 8 Elite groups preferred the hard division of power based on ethnicity rather than fostering democracy and equality. Institutionalized ethnocracy is therefore counterproductive: it often does not serve for the solution of conflicts but promotes their manoeuvrability or even conservation. No reconciliation is possible if groups represented by their elites are not interested in putting an end to problems. They will often tend only to achieve legal guarantees for their groups, which could be presented as clear victory over others. They will not choose to rely on solutions that depend on the kindliness of others, trust, or mutual compromise.
In practice, the ethnocratic legal structure created for the major groups (Serbs, Bosniaks and Croats) in Bosnia and Herzegovina ultimately paralyzes the operation of the state. This situation has even elicited the protest of the population several times, regardless of group affiliation. Similar phenomena could also be observed in Kosovo: from the end of 2013 until mid-2015 the parliament was more or less inoperable, and not even early elections could make a change. Yet the fundamental difference between the two countries is that while the ethnic veto can be applied at any time in Bosnia and Herzegovina, it can only be used in an itemized form with regard to certain legislative issues in Kosovo. Thus, Kosovo is not paralyzed by veto rights – the crises are more political than systemic in nature –, while in Bosnia and Herzegovina ethnic-based decision-making leads to the serious inoperability of the state.
Lijphart also suggested some institutional solutions for ethnically divided societies during the construction of post-conflict constitutions. 9 A parliamentary form of government, federalism, and decentralisation are core elements of his advice. Interestingly, there are several elements on Lijphart's list that do not reinforce the legal toolbox of classic democracy but even tend to narrow it down: few or no referendums, an electoral system with proportional representation, fixed lists, constructive motions of confidence, consideration of ethnic aspects during the formation of the government, head of state elected by parliament. Could there be a trade-off between democracy and ethnic peace? There is an underlying consideration behind Lijphart's list: the full implementation of democratic processes could lead to the empowerment of a dominant group. Democracy without minority rights is not sensible to minorities, Therefore, some issues should not be decided only by the majority (e.g., language policy). It seems – especially from the respective legislation in the Western Balkans – that multinational states usually disprefer forms of direct democracy. The main rule is representative democracy in the Western Balkans as well. In contrast, many forms of highly acclaimed and widely used direct democracy exist in Switzerland. This also suggests that, as we shall see, Switzerland is more similar to a nation state than to a multinational one.
Personal autonomy could be the real alternative to veto rights and quotas. In the case of an exacerbated ethnic conflict, parties regard regional autonomy or complete secession as the biggest victory, and as the most presentable result for their own group. The Dayton agreement – which created almost independent states within Bosnia and Herzegovina, following the sole logic of geographical autonomy with extremely complex internal borders – reflected this clearly. If the necessary financial resources are available, personal autonomy could provide a realistic alternative, albeit not being as spectacular as regional autonomy or independence. A good example of this can be observed in Belgium. In order to achieve personal autonomy, one must give up the quest for statehood, since one of the conditions of a real state is territorial sovereignty. Personal autonomy can only mean a quasi-state existence for a community with a population with limited power, and without an exact territory. Yet, in countries where community membership is not exclusive and multiple identities are common (e.g., Bosnia and Herzegovina, and Ukraine), territorial autonomy is a real alternative.
In the following empirical section, the conciliatory mechanisms of multi-ethnic state parliaments will be presented, first in Western Europe, then in the Western Balkans. The comparison focuses on the question to what extent legal arrangements assist the co-operation of dissenting parties, and whether they are rather designed to block legislation and be used as weapons against ethnic/political opponents.
3 Ethnic conciliatory mechanisms in parliaments of Western Europe
3.1 Belgium
It is well-known that Belgium is a state that was gradually decentralized, developing in the last two decades from a unitary to a federal state. The country is characterized by multilingualism and a strong local identity (Flemish and Walloon citizens make up the population in almost equal proportions). The country's territory is divided into the (territorial) regions of Wallonia, Flanders, and Brussels (the latter is officially bilingual), while the population is comprised of three linguistic (personal) communities (Walloon, Flemish, and German). 10 All the regions, as well as the communities, can decide independently on certain issues; they have elected bodies (parliament) and administrations; they control their own budgets. Including the federal parliament, there are seven popular representations operating in the country. The ethnic principle is not reflected directly in electoral rules: i.e., it is not relevant which ethnicity the given candidate belongs to. Citizens' membership is based on the territorial principle in the regions; in the communities, it is based on the individual principle (voluntary basis). Indirectly, the ethnic principle prevails at electoral district boundaries.
Because of this widespread autonomy, there is no real need to apply an ethnic veto during decision-making. In areas where protection by ethnic veto might be needed (e.g., in cases that fall under federal authority), the so called ‘alarm-bell’ procedure is available (see below), which was named the ‘soft veto’ by Kelleher, and which does not mean complete blocking nor the postponement of decision-making, but obligatory additional reconciliation through the rethinking of decisions. 11 This soft quality decreases the possibility of the abusive use of the veto and situations when disputing parties are split into winners and losers.
The lower chamber of the federal parliament, the House of Representatives, is created as a result of democratic elections, while the Senate is the forum for safeguarding the interests of the regions and the communities. Electoral districts are created along linguistic boundaries during the election of the House of Representatives: there are five Flemish, five Walloon, and one bilingual district (that of Brussels) out of the eleven electoral districts. The number of mandates differs in the eleven electoral districts depending on the size of population, and varies between four (Luxembourg) and twenty-four (Antwerp). In total, 79 seats are allocated from the Flemish district, 49 from the five Walloon districts, and 22 from the bilingual Brussels district (the total number is 150. So, it follows that the Flemish majority is ensured definitely in the House of Representatives, which rightly reflects the population ratio. However, the legislation only refers to numbers and districts, not ethnic affiliation. In contrast to the rigid rules in Bosnia and Herzegovina, the laws do not exclude the possibility that the Walloon delegates can obtain a majority in the House of Representatives. This is, of course, highly unlikely: in order to achieve this, Walloon delegates must be elected in Flemish districts as well.
Since the last constitutional reform in 2014, the Senate has had 60 members and been composed of 50 delegates representing regions/communities, as well as 10 senators co-opted by them. Twenty-nine out of the fifty are elected by the Flemish parliament, ten by the parliament representing the French-speaking community, eight by the Walloon parliament, two by the French delegates of the Brussels parliament, and one by the German community's parliament. There is no ethno-linguistic ‘census’ here; the linguistic equilibrium is a result of the method associated with the electoral rules. Before 2014, the 71 members of the Senate were elected directly from the Flemish and the Walloon electoral districts, with only 21 from the parliaments of the communities. Since 2014, the Senate has not been a directly elected body but the representative of the parts that form the federation. The two chambers have not been equal since 1995: the lower house has the final word and the Senate uses its veto rights only with certain issues. After the creation of the chambers, not only political but also linguistic groups are formed: the constitution requires that the elected delegates of both houses are members of the French or the Dutch linguistic group.
Concerning the composition of the federal government, similarly to that of Bosnia and Herzegovina, it is not proportional to the population, but has an ethnic-parity composition: it is made up of an identical number of French- and Dutch-speaking members who make decisions unanimously, while both parties have veto rights. The 15-member federal government is made up of 14 ministers in addition to the prime minister, with an equal number (fifty-fifty) of Flemish and Walloon members.
To sum up, the composition of the regional/community parliaments, as well as the democratic composition of the federal parliament, is not influenced directly by the ethnocratic rules of the electoral districts, which are not only based on the federal principle of local-central power division but on linguistic-ethnic boundaries as well. In addition to the election rules, balancing-of-interests procedures exist in Belgium as well, which create the platform for ethnic groups to protect their rights in relation to other groups. Three such procedures exist in the Belgian parliament(s).
-
The ‘alarm-bell’ procedure.
The ‘alarm bell’ could be rung by any of the houses of the federal parliament: with the exception of laws about the budget and those requiring a qualified majority, at least two-thirds of delegates belonging to one of the linguistic groups can make a motion with reasoning before a final vote, according to which the proposal may seriously undermine the relationships between the communities. 12 In the case of such a motion, the legislative procedure is suspended, and the motion must be handed over to the federal government, which makes a statement about the motion within thirty days, in which it can ask for an amendment of the bill from parliament. There is an associated qualitative restriction: one member of one parliamentary linguistic group can initiate it only once during the debate on the law.
The procedure, which was established in 1970 for the protection of French-speaking delegates, has been used only twice in practice: in the middle of the 1980s, and in 2007, when French-speaking delegates applied it in order to prevent the redefinition of the borders of electoral districts which would have affected the French-speaking community negatively. However, the case was not resolved and a government crisis occurred, which was followed by early elections. Negotiations on the issue between the two linguistic communities resulted in the sixth state reform, which was implemented between 2012 and 2014. There is a variant of the ‘alarm bell’ in Brussels as well. However, while the federal procedure protects the rights of those speaking French against the Flemish majority, this version serves the interests of the Flemish minority in Brussels. It was introduced by law in 1989 (it does not have any constitutional basis) when the Brussels region started functioning in practice. This form of alarm bell has never been used in practice.
-
2. Balancing-of-interests procedures between regions/communities.
According to article 143 of the federal constitution, the federal state, the communities, and the regions act according to the principle of ‘federal loyalty’ in order to avoid conflicts of interest. Formally, the Senate may declare a well-justified opinion in the case of a conflict of interests, which may occur among the parliaments. Based on the principle of federal loyalty, if everybody acts as a loyal member of the alliance, conflicts of interest should not occur. This procedure, which could be triggered by any parliament (e.g., the Flemish parliament against the federal one, the Walloon parliament against the Flemish, etc.) has been used in practice more frequently than the alarm-bell process. It does not allow for parliaments to completely block each other's motions, but if more parliaments stand up against the motion of one parliament, then the motion may be stopped.
Obtaining the Senate's opinion, which is the next step in the procedure, is not mandatory; it is only of a consultative nature. Finally, a reconciliation committee may be established to discuss the case. This decides unanimously, and in the case that no agreement is reached, the parliamentary procedure may proceed according to the general rules. The procedure therefore does not mean veto rights, and does not resolve the conflict of interest on the highest level, although it is useful for winning time. The procedure is regulated in detail by the law of 9 August 1980, which can only be modified with a qualified majority. It is worth noting that the procedure is not to be mixed up with the debate on authority, which is regulated by the constitution. In these cases the subject of debate is the authority of an institution in relation to certain decisions. This procedure is decided by the Federal Constitutional Court, which – unlike in Bosnia and Herzegovina – otherwise does not have any role in balancing-of-interests procedures between the regions/communities.
In sum, the electoral system in Belgium does not include any rules regarding the ethnic affiliation of candidates or voters. It operates with the tools of electoral geography; it takes care of the proportional representation of the linguistic-ethnic communities indirectly. There are no special decision-making rules in parliamentary decision-making (e.g., voting in relation to ethnic quotas, ethnic veto), and decisions requiring qualified majority are listed by the federal constitution. Balancing-of-interests procedures exist but are rarely used in practice; their existence only increases the threatening possibility of a dispute and points the parties towards compromise. A significant difference is that in Bosnia and Herzegovina the constituent nations are present in all committees, and the balancing-of-interests procedure takes place within a given entity. In Belgium, however, it is triggered by an external actor – i.e., decisions made by another parliament could also be the subject of a dispute.
3.2 Switzerland
Switzerland is a distinct and unique case for study because it is hard to locate in relation to the dichotomy of multinational state/nation states. 13 A significant part of the literature considers it a nation state – albeit being multilingual it does not have the characteristics of a multinational state. 14 A strong common identity is indeed a feature of Switzerland, regardless of residence or spoken language (similarly to the USA). I consider it to be a multinational state, and this is reflected in the ethnocratic legal solutions that can be found in Switzerland as well, although with less significance.
The alpine federal state has four official languages (German, French, Italian, and Romansh) but the country is not composed of ‘nations’; i.e., independent ethnic-linguistic groups. According to the first paragraph of the constitution, Switzerland is composed of ‘the Swiss people (Schweizervolk) and the cantons’. It is important that the definition of the nation extends to the whole population and to every citizen; the linguistic identity German, French, etc. is only of secondary rank. Twenty-two of the 26 Swiss cantons are monolingual: 17 German-speaking, four French-speaking, and one canton (Ticno) is Italian-speaking. Since cantonal borders typically do not follow the Latin Germanic linguistic border, three out of the remaining four cantons are bilingual (Bern, Fribourg/Freiburg, Valais/Wallis are German- and French-speaking), and finally, one canton (Graubünden) is trilingual (German, Italian, Romansh). The country is distinct from this perspective as well: in Belgium, Bosnia and Herzegovina and other multinational states of federal character the linguistic border is the administrative border at the same time, and vice versa.
Thus, the German-speaking cantons are in the majority, which leads to a German-speaking majority in both of the houses of parliament. The 200 members of the lower chamber (Nationalrat) are elected proportionally to the population, whilst the 46 members of the upper chamber (Ständerat) uniformly provide two delegates (in the case of the so-called half-cantons, only one). No declared, legally binding minority rights apply to the non-German cantons. There is no reconciliatory mechanism in place – neither national veto right nor linguistic-ethnic parity, as there are no linguistic-ethnic groups by legal definition. Democratic logic and proportionality prevail, which is distorted by some ethnocratic elements.
Regarding the composition of the federal government, there are no prescriptions for the representation of linguistic groups, or their granted membership. Nonetheless, it is an unbroken practice (not confirmed by any formal rules) that two out of the seven members of the government must not be German-speaking. For a long time this normally meant having two French-speaking members – since 1848 there have been no Italian-speaking members of the government, nor Romansh. 15 The representation of the minorities in the federal government is ensured on an ad hoc basis based on informal agreement instead of formal quotas.
However, formal quotas – although in small numbers – also exist in Switzerland, in terms of the internal regulation of bilingual cantons. In Bern's parliament, one seat is reserved for a French-speaking delegate for the Bernois region. Bern is also the only canton to apply a formal and direct quota (since 2006); namely, in the Bienne-Seeland ethnically diverse electoral district, for the sake of the French-speaking inhabitants. We only find indirect ethnocratic rule in Wallis canton within the government: the constitution ensures that three of the five cantonal ministers must originate from three different regions of the canton. Since two out of the three are French-speaking and one is German-speaking, this provision grants indirectly that German speakers are also represented in the government. However, this constitutional provision does not require that the respective minister should be de facto German-speaking (in Bern's case, French-speaking); it only defines the region, not the candidate's affiliation. Graubünden, although trilingual, is the only canton where the government's ethnic composition is not a political issue – there are neither formal nor informal linguistic-ethnic representational rules there. 16
To sum up, in Switzerland the nationality is ‘Swiss’ and is not an ethnic category. Groups (communities) of those speaking the same language do not appear in any formal institutional setting. There are no balancing-of-interests mechanisms between the different linguistic-ethnic groups, and even fewer legal tools; a linguistic-ethnic quota can only be found on lower levels of administration. The constitution does not mention linguistic representation – it only tackles the cantons and the Swiss people. In the monolingual cantons, the minority protection of those speaking a minority language is unknown. Instead of these formal tools, however, informal arrangements do exist in order to ensure ethnic participation, first of all in relation to the composition of the federal government.
4 Ethnic conciliatory mechanisms in parliaments of the Western Balkans
The public sphere teems with ethnic-based power sharing solutions in Bosnia and Herzegovina, Kosovo and, occasionally, North Macedonia – which, in some fields, obstruct democratic processes. The ethnic composition of state bodies is predetermined (by quotas), while ethnic considerations are present in decision-making (dual majority) and in compensatory procedures (veto, and the ‘vital national interest’ procedure).
Interestingly, from the standpoint of minorities (i.e., persons not belonging to any of the three dominant groups), the most complex case, Bosnia and Herzegovina, is similar to that of a nation-state – or more precisely, a ‘three-nation-state’ – discriminating against all other ethnic groups. According to the preamble of the state constitution, the Bosnian, Croatian, and Serb nations are the ‘constituent nations’, whereas ‘others’ are regarded as minorities (e.g., Roma, Albanian, etc.). Thus, the grand coalition of the three ethnicities appears to be a nation-state from the other minorities' point of view, and equality before the state only applies to the three constituent nations. Consequently, and by definition, none of the three ethnicities are minorities, even if one of them is in a numerical minority, as all three are equal depositories of the definition of ‘majority nation’. This approach to the equal rights of the three state-building nations leads to inequality and hierarchy between citizens: the three state-building nations are the depositories of state power. The constitution recognizes citizens belonging to a minority as equal, albeit without providing the opportunity for them to fill certain state positions, which are reserved for the representatives of the state-building nations.
In an ideal case, such institutionalized ethnicity 17 solutions would be the result and not the basis of the peaceful coexistence of ethnic groups, and they would be conceived as voluntary behavioural patterns rather than mandatory rules. As an EU report stated about the Federation of Bosnia and Herzegovina, ‘legislative work is adversely affected by intransigent and ethnically oriented positions taken by the country's political leaders'. 18 Due to ethnic borderlines in the political field, clear political coalitions and stable majorities in parliament are often missing, therefore cooperation between parliament and government is insufficient. In some cases, laws proposed by the government have been blocked later in parliament by political actors who back the government coalition. In addition, no proper mechanism exists in the legislative process for coordination between the state parliament and the parliaments of the two entities of the country, the Federation of Bosnia and Herzegovina, and Republika Srpska. 19 The Dayton Accord was silent on such nuances – today a major obstacle to democratic functioning. According to Reinhard, 20 besides economic issues, the region's most important and urgent tasks that require addressing in terms of constitutionality concern the judiciary and parliamentary structures.
The parliamentary structure within Bosnia and Herzegovina is rather complex: there is a bicameral state-level legislative, the Parliament of Bosnia and Herzegovina, 21 and both entities have parliaments respectively: that of the Federation is a bicameral legislative body, while the Republika Srpska parliament has only one chamber. 22 (The Federation consists of ten autonomous cantons with elected legislatures – their analysis is outside the scope of this paper.)
In political practice, ethnic conflicts paralyse parliamentary functions. The number of legislative proposals (bills) in the parliaments of Bosnia and Herzegovina is less than in any other country in the Western Balkans. The reason is the inactivity of the state government due to internal ethnic conflict. Furthermore, law-making is slower and less predictable in the parliaments of Bosnia and Herzegovina and the Federation (with adoption times amounting to one to two years). As a result, legislative work is unduly slow, and outcomes uncertain. 23
There are other solutions in the region for reconciling ethnic conflict. As democracy and equality before the law might become distorted in multi-ethnic societies, constitutional provisions should affirm the political representation of minorities or seek a kind of ethnic equilibrium. As for the first requirement, examples can be found in Kosovo and North Macedonia; for the latter, there are examples in Bosnia and Herzegovina. In Western Europe (Belgium, Switzerland, and Northern Ireland), the following solutions can be observed: arrangements for determining the composition of public authorities (quotas and reserved seats), decision-making arrangements (double majorities) and conciliatory procedures (veto and the so-called ‘vital national interest’ procedure, see below). In the Western Balkans there are examples of employing all those solutions simultaneously, yet with a lack of mutual trust and confidence: in Bosnia and Herzegovina the equality and mutual vetoes of the ‘state-building nations’; in Kosovo and North Macedonia the commensurate representation of the minority groups in almost all state institutions. Even the latter two differ: while in North Macedonia the Albanians received such rights upon their own demand, in Kosovo these rights were provided to Serbs against their will, satisfying the international community. They do not recognize the state, therefore never asked for such rights. The example of the neighbouring multinational Montenegro is different: politics is not only polarized along ethnic lines, but in a cross-cutting manner, including with regard to certain basic questions, such as the relationship with Serbia and the role of the Orthodox Church.
In a liberal democracy, the state must grant the right of self-determination to individuals. The latter should be able to choose the ethnic group to which they wish to belong. Consequently, individuals also have the right to refuse such ‘labelling’ and not to identify themselves with any of the ethnic groups (i.e., to join ‘others’). In systems which force the population to join ethnic groups, some might not be able to find their place. Therefore, it is important that the balance of power between such groups does not jeopardize individuals' integrity or their right to self-identification. It is not only Bosnia and Herzegovina which provide for that: Belgium and Northern Ireland also cluster their parliamentary representatives into linguistic-ethnic groups on a compulsory basis. This is not only disadvantageous for those who reject such division, but slightly modifies the will of voters because they did not necessarily choose representatives based on their ethnicity. Such grouping also preserves ethnic boundaries. In Switzerland, however, these ethnic groups do not exist within the population: there is no formal representation of Swiss-Germans, Swiss-French, or Swiss-Italians in any constitutional provision. The Venice Commission repeatedly criticized the ethnic arrangements in Bosnia and Herzegovina on all levels since they run counter to the principles of democracy. Yet there has been no reaction. A huge constitutional reform package that was developed upon the European Commission's proposal failed in the last vote in 2006.
The European Court of Human Rights condemned Bosnia and Herzegovina in 2009, since those who belong to particular ethnicities are excluded from certain state offices. 24 In Kosovo and North Macedonia there are similar solutions concerning majority-minority relations (although ‘majority-minority’ terminology is avoided: in Kosovo they prefer to use ‘national communities’; in North Macedonia they call them ‘communities that do not form the majority’). Kosovo's constitution – as opposed to that of Bosnia and Herzegovina – does not specify nations and ethnic groups as state-forming factors. Instead of a ‘nation-state’ it defines itself as a ‘state of free citizens’, 25 which grants freedoms and rights equally to all its citizens. In the case of Kosovo and Bosnia and Herzegovina, the state symbols (coat-of-arms, flag) express a multinational existence (e.g., a star motif); these symbols were created with the help and support of the international community. They do not belong to any of the ethnic groups; they are neutral, artificial symbols.
Below, I will analyse the ethnocratic solutions of the constitution, comparing them to Western-European solutions for balancing ethnic interests, with parliamentary mechanisms in focus. I will not cover the field of fundamental minority rights (use of language, cultural autonomy, etc.), which nevertheless exists in law in all of these states.
4.1 Ethnic composition of state institutions
The constitution of Bosnia and Herzegovina grants equal representation, regardless of the prevailing election results, to the three constituent nations (and only to them) in some positions (upper chamber of the parliament, and state presidency). Similar rules exist in Kosovo: the Serb minority is entitled to ten seats in parliament, regardless of elections results, and some seats for other minorities are also reserved. In contrast, North Macedonia sought to display ethnic representation through a double majority in parliamentary voting and by abolishing the electoral threshold, thereby promoting the advancement of ethnic parties. The composition of Bosnia and Herzegovina's constitutional court involves types of ethnic scaling as well. Some detailed examples of the ethnic nature of representation are as follows.
The head of state of Bosnia and Herzegovina is a position filled by a presidency of three members, where each member represents one of the three ethnicities. Formally, it is not the ethnicities but the territorial units (entities) that elect the members of the presidency. But, due to the strong ethnic character of the territorial demarcation, ethnic representation prevails de facto. For example, the Serb member of the presidency is elected by the citizens of Republika Srpska, not only the Serbs that live there. On the other hand, Serbs living in other entities have no say in the election of the Serb member.
The government of Bosnia and Herzegovina is formed by a fixed number of ministers: nine cabinet members in addition to the prime minister, all ethnicities having three ministerial positions. Similar solutions also exist in Western Europe (Belgium). There is a similar regulation in place in Kosovo, without fixing the number of cabinet members: a Serb and another minister that represent a different minority group must have at least one seat. In the case of twelve or more ministers, three nationality ministries must be established. Additionally, there are two Serb and two deputy ministers that represent other nationalities. Küpper criticizes this solution rightly, claiming that this cabinet principle bends towards ethnocracy. 26 The Federation's constitution regulates similarly when setting the number of ministers at 16, among whom there should be eight Bosnian, five Croatian, and three Serb ministers.
In Western Europe, even in multi-ethnic countries, it is rather uncommon to fix the number of ministers and set ethnic quotas. In Switzerland, the government became linguistically variegated on a voluntary basis without legal prescription, based merely on customs (i.e., self-created rules).
But this kind of ‘ethnic engineering’ has the most impact on the institution that otherwise should be the most democratic: parliament. Bosnia and Herzegovina's National Assembly is a bicameral body composed of a 42-member House of Representatives (28 elected from the Federation and 14 from the Republika Srpska) and the 15-member House of Peoples (ten members are elected by the upper house of the Federation, five Bosnian and five Croatian members respectively, the remaining five members are elected by the parliament of the Republika Srpska). Whilst in the House of Representatives the delegates are organized by fractions of political parties, in the House of Peoples Bosnian, Croatian and Serbian ethnic fractions operate. In the House of Peoples, the presence of nine members is necessary for a quorum, and there should be at least three members present from all of the three ethnicities. The position of the parliamentary presidency changes on a rotating basis among the delegates of the three constituent nations. Imperatively, the electoral districts are formed along ethnic borders. The Federation's parliament is composed of a bicameral House of Representatives similar to the state level, and of a House of Peoples, which represents the interests of the constituent nations. In Kosovo, out of the predefined number of five parliamentary vice-presidents, one should be Serb, and one is required to be a representative of the other minorities.
Another institution of Republika Srpska worth mentioning is the Council of Peoples, which was established due to pressure from the High Representative for Bosnia and Herzegovina, who oversees the implementation of the Dayton Peace Accord (the decision was therefore mostly ignored and refused by Serb institutions and stakeholders). The Council was created in order to establish a forum in the ethnically relatively homogeneous Republika Srpska (similarly to the House of Peoples in the Federation), and to advocate the interests of the constituent nations. In theory, the Council of Peoples was considered the second chamber; in practice, it has less power: it does not take part in the formal legislative procedure and cannot even make a legislative proposal. The Council is only part of the legislative process through the ‘vital national interest’ procedure: it can veto the laws passed by the parliament within a specified time.
In Kosovo, the ethnic logic is present in the other branches of state power as well: 15% or at least three Supreme Court judges must belong to a national community (a minority); two members of the 13-member Judicial Council, which supervises the judicial system, are Serb, and two members belong to other national communities. The composition of the courts and the prosecutors must take into account ethnic proportions; one of the deputies of the ombudsman also belongs to the minority.
However, in Western Europe there are more sophisticated rules about the ethnic composition of state institutions. Usually, it is not the quota that is predefined before the elections which is the prevailing one, but delegates are divided into linguistic-ethnic groups after elections. These form voting blocs during the voting sessions, which require a majority in addition to the majority of the whole house. I present two such solutions here in brief.
-
After the elections in Belgium, all the delegates of the lower chamber of the federal parliament, regardless of their party and fraction membership, become members of ether the Flemish or the French linguistic group (delegates from Brussels can choose).
-
In Northern Ireland, delegates are divided into three groups in parliament: those who support union with Great Britain (unionists), ones who support integration with Ireland (nationalists), and ‘others’. Therefore, the division is not based on ethnic background but more on judgements about an ethnically motivated question (independence). The third group is formed of those that refuse to choose between the two, or who do not wish to take sides – e.g., because they do not recognize this division as legitimate. According to the voting rules, if 30 delegates call for it (a ‘petition of concern’), any vote can become an inter-community decision. In this case, besides the qualified majority in the parliament (60%), 40% of the members of both groups who are present and vote are necessary. Therefore, a political group that achieves a dominant position in one of the communities could veto any decision of the parliament. This Northern Irish solution does not take into account those that who not do politics along ethnic boundaries consciously and who refuse to be identified with any of these groups. The group of ‘others’ does not have veto rights; therefore, they give up a position of power voluntarily, which implies that they start their electoral campaign with a handicap. The Northern Irish power division does not resolve but maintains and prolongs ethnic boundaries.
The ethnic ‘manipulation’ of the composition of elected institutions seems to erode equality before the law and democracy, and could distort the results of elections. Affirmed by this, the prescription of the ethnic composition of the state's institutions cannot be justified or successful, and we cannot view even Western-European solutions as reassuring.
4.2 Ethnic voting (‘double majority’) during decision-making
In the parliaments of Bosnia and Herzegovina, Kosovo, and North Macedonia, besides parliamentary decision-making based on simple majority, qualified majority requirements also apply in order to avoid one of the ethnicities being outvoted against their will or in their absence.
Complicated voting regulations are in place defined in the constitution in the parliamentary chambers of Bosnia and Herzegovina: ‘The delegates must make efforts to achieve that besides the simple majority, at least one third of the delegates of all entities must be present.’ 27 If this condition is not fulfilled, even a simple majority can trigger a special procedure: the session is suspended for an hour while the house speaker tries to ensure the necessary majority. In case this is unsuccessful, the house speaker and the two deputies (the representatives of all three ethnic groups) will try to achieve the prescribed majority within three days. If this effort fails, the decision must be made with the simple majority of those present, provided that the votes of none of the entities who vote against a motion reach two-thirds of all votes. If this fails as well, the question will be adjourned to the next session. Thus, based on the ethnic voting rules, delegates could block a decision simple by not attending. This makes parliamentary decision-making unpredictable and uncertain. The requirement that takes into account the votes of the entities equally applies to both Houses, but it is more frequently used in the House of Representatives.
Besides voting rates, the rules of quorum are of an ethnic nature as well. In the House of Peoples, the presence of nine out of fifteen delegates is expected, provided that three delegates are present from all three ethnic groups. The House of Representatives has not had an ethnic condition for the quorum since 2007 (22 out of the 42 delegates are necessary for a quorum). Previously, the presence of one-third of the members of each ethnic group was also required. This was abolished by the High Representative so that none of the ethnic groups may block parliamentary work; it had occurred that the parliament did not achieve a quorum even with half of the representatives present due to the non-attendance of one of the ethnic groups. According to the statutory interpretation of the High Representative, if a delegate is present in the hall but does not vote, this should be considered abstention. In the joint committees established by the chambers of the state parliament there is quorum requirement based on ethnicity: these 12-member committees can create a quorum from seven people or more who are present, as long as one person is present from all three constituent nations in the House of Peoples. Decision-making takes place by simple majority, which must include one-third of the House delegates in addition to the one delegate from each constituent nation.
In Kosovo, besides the absolute majority of all delegates, the absolute majority of the minority delegates is also necessary for voting, especially in relation to legislative subjects that are sensitive in terms of minorities 28 (unlike Bosnia and Herzegovina, the constitution lists these issues exhaustively: minority, cultural issues, and local government legislation). Originally, a relative majority was required for both; however, a constitutional amendment raised this to an absolute majority. Minority representatives have 20 reserved seats, ten of which belong to the Serb community. Because of this, Serbs alone are not capable of raising a veto; in order to achieve this the vote of at least one delegate that represents another minority is necessary. Based on the new majority regulation, the ten Serb representatives cannot have a majority within the minority group. To amend the constitution, besides having two-thirds of all representatives, two-thirds of minority delegates are also necessary.
According to the similar provisions of the North Macedonian constitution, a majority of delegates ‘who do not belong to any of the communities forming the population's majority’ is necessary for adopting certain laws: besides the relative majority in plenary, the relative majority of those belonging to a minority group is also necessary. 29 For legislation related to local government, apart from the two-thirds majority of all delegates, the absolute majority of minority delegates is needed. Since the Albanian community is itself a majority among all minorities, those belonging to the smaller minorities can never raise a veto; in other words, only Albanians have real veto rights in practical terms. The interests of the other groups are only granted if they coincide with the Albanians' interests. In comparison to the rules in Kosovo, a further difference is that in North Macedonia the minority delegates do not have a legal definition – namely, minority representation is not created by quotas, but simply by the election results of the ethnic parties which compete with other political parties. Thus, minority representation here is not a legal but a political category.
As also stated by the Venice Commission, the double majority decision-making prescribed by the constitution is only functional if it is not possible to change ethnic affiliation voluntarily and temporarily – i.e. if the majority or minority status of delegates is pre-determined. 30 Otherwise, the ethnic majority prescription could be circumvented by convincing a fraction or a political group to claim that they are of another ethnicity. The Venice Commission argues for ethnic representative groups that are based on registered membership, which is inherent to the rules prescribing the double majority. Since in North Macedonia minority delegates get into parliament not through legal but political (electoral) ways, representatives who falsely claim their ethnicity cannot be excluded, leading to the potential manipulation of voting results.
As seen in Kosovo and North Macedonia, rules do not provide permanent veto rights for all issues to the ethnic groups, but rather include minorities in certain decisions. Majoritarian decision-making mechanisms prevail, instead of legal tools based on equality between constituent nations.
Special voting rules exist in Western Europe as well. In Belgium, a double majority – namely, the two-thirds support of both houses and majority votes from all linguistic groups (Flemish, French, and German) – is necessary for making certain decisions as defined in the constitution. The laws that create the current federal state organization were also passed by such a form of voting. 31 However, obtaining a wider consensus is not without precedent. Parliamentary decision-making is bound to a qualified majority and is a solution that cannot be objected to on democratic principles. As seen in Lijphart's thesis, there is no place for referendums in any of these states about issues that are ethnically sensitive. The regulation does not use the term ‘minority’ but applies the term ‘national community which is not in majority’ instead.
4.3 The vital national interest (VNI) procedure in Bosnia and Herzegovina
Besides the rules of ethnic composition and the regulation of ethnic decision-making defined by the constitutional institutions, ‘vital national interest’ (VNI) proceedings are pronounced ethnocratic elements in the functionality of the state of Bosnia and Herzegovina. Practically, these are institutionalized veto rights that allow one of the constituent nations to fully block a decision of parliament on the state level, as well as on the level of entities. Through the ‘vital national interest’ procedure one of the three constituent nations can veto legislation or parliamentary decision-making by referring to a vital national interest – however, the concept of national interest is not legally defined.
The fundamental national interest procedure existed earlier in the Yugoslav federal parliament as well. If a proposal about the interests of the republic or an autonomous province or about the equality of nations or nationalities was on the agenda, and one of the representatives of the republic or an autonomous province requested it, a special reconciliation procedure could take place. Besides this, the upper chamber of the Federal Parliament decided through consensus – i.e., reconciliation of the delegates of the republic and province was necessary.
VNI procedures are currently in place on both state and entity levels in Bosnia and Herzegovina. On the state level, the state constitution; in Republika Srpska, the constitution; and in the Federation the constitution and the house rules of the House of Peoples include these provisions. The procedure is regulated in roughly the same way in both entities.
-
In the House of Peoples (second chamber) of the state parliament, a majority of the Bosnian, Croatian or Serb delegates (three out of the total five) can declare any decision as threatening their national interests. In this case, the House Speaker establishes a reconciliation committee in order to resolve the debate with one delegate from each of the three ethnic groups. If this does not lead to a result within five days, the procedure is handed over to the Constitutional Court, which discusses it in urgent procedure. 32 The final decision whether the acceptance of the decision really harms the vital interest of an ethnic group is declared by the Constitutional Court. However, this rarely happens; the procedure in practice is rather used by the parties for postponing decisions.
-
In the Federation, in the House of Peoples, practically all decisions could be declared harmful to the vital interest of a constituent people by two-thirds of any ethnic group. The procedure can be initiated by the president or his/her deputies of the House of the Peoples. The proposal must be voted on within the ethnic groups, and all sessions must have a majority. If an agreement is not reached, or the amendments are not accepted by the House of Representatives, a reconciliation committee is established from the members of both houses which decides through consensus. If the settlement is not accepted, the proposal must be considered rejected, and it is returned to the proponent. The last word about a proposal which has already been accepted by the committee is had by the Constitutional Court, which decides in a special seven-member council with a two-thirds majority (two members from each ethnicity and one person representing ‘the other ethnic groups’). If harm to a national interest is not declared, the decision must be considered accepted.
-
As seen above, the VNI procedure in the Republika Srpska takes place ‘outside’ the parliament by the People's Council, which was established due to external pressure. All decisions accepted by the parliament are brought in front of the People's Council, which has one week to engage in further examination. If the Council initiates the procedure, it moves similarly to the Federation (reconciliation committee with the participation of the parliament and the Council), and ultimately the Constitutional Court decides.
The main problem with the ‘vital national interest’ procedure is that what is considered significant harm to interests is not defined. This rather depends on the opinion of those affected, and the interpretation of the Constitutional Court. In practical terms, this is used as a political veto. Abdel-Fattah's research about the practice of Bosnia and Herzegovina has shown that almost all potential questions could be associated with a vital national interest. 33 Similar procedures exist in the multinational Northern Ireland as well; however, unlike in Northern Ireland, there is no pre-defined list of topics which could be vetoed or postponed referring to a national interest. 34 On the other hand, the disadvantage of a closed list is that it grants veto rights, even if the issue is not about minorities. For example, in Kosovo minorities have veto rights related to the complete educational and local governmental legislation because this involves issues which affect minority interests. 35
An important question regarding the presented procedures, voting methods, and vetoes, is whether there are mediation procedures for the case of complete deadlock, or simple complete blocking or acceptance (‘take it or leave it’) is possible. As we shall see, in Belgium, the relevant legislation specifies mandatory mediation between the disputing parties, whilst in Bosnia and Herzegovina cases inevitably split the disputing parties into losers and winners.
It is worth mentioning that in the Brčko District, which has significant autonomy in Bosnia and Herzegovina, a sophisticated form of ethnic control is in place: the ‘vital national interest’ procedure does not exist; instead, the thirty-one-member parliament of the district takes every decision with at least three-fifths of its members. Aside from this, a form of ethnic voting prevails according to which at least one ‘confirmatory’ vote is necessary from all ethnic groups in certain legislative fields (e.g., education, religion, and budgets): if three to five members of a group are present, two confirmatory votes are needed; if six to eight members are present, three, and so on. Bieber compared the solution in Brčkó with the Mostar Canton, which is plagued by severe internal ethnic conflict. Mostar applies rigid, legal solutions (vetoes), and, as he concluded, power division is more successful through less formal methods than territorial fragmentation or the complex balancing-of-interest-based procedures. 36
As a consequence of the procedures presented above, in Bosnia and Herzegovina the state functions slowly and unpredictably. Legislation can be blocked at any time based on ethnic conflict, and the legislative procedure is held captive by competition between the various ethnic groups. In Kosovo and North Macedonia, only the previously presented quotas of territorial composition and double-voting rules exist, but not the complex ethnic veto system known from Bosnia and Herzegovina. So, differently from Bosnia and Herzegovina, ethnic voting is not based on voluntary decisions but may occur only in relation to pre-defined issues in these two countries, where ‘occupied seats’ take up only a fraction of the whole legislature.
It needs to be highlighted that in Bosnia and Herzegovina all three ‘constituent nations’ have equal rights in both entities; i.e., not only in the one in which they form the majority. One of the decisions of the Constitutional Court of Bosnia and Herzegovina should be mentioned as an example here, 37 according to which the Court declared the provisions of a statute of the city of Sarajevo as unconstitutional. Paragraph 21 (3) of the original statute declares that Bosnians and Croatians, as well as ‘others’ must each have 20% representation in the city council, regardless of the election results. If elections do not lead to such a result, seats must be filled from the list of candidates. Therefore, the provisions list the Serbs as ‘others’, even though they are to be considered a constituent nation throughout the whole of Bosnia and Herzegovina, including the Federation, inhabited mostly by Bosnians and Croatians, where the city of Sarajevo is situated.
5 Conclusions
The investigation of the phenomena of ethnic quotas and ethnic preference in political representation in the Western Balkans compared to Western European arrangements showed that any legal regime that is based on instruments that award preference to the interests of one or more ethnic groups and enable them to completely block legislation can only be built to the detriment of democratic states. If ethnic/local identities are stronger than national identities as a result of an ethnic way of thinking, the political landscape is dominated by mostly ethnic-oriented parties. As a result, effective democratic, compromise-based decision-making procedures are often missing or neglected in the Western Balkans. The Western world, including the EU, propagates the model of a democratic state with guarantees for minorities as the only alternative to separatist and nationalist endeavours. Yet these guarantees and rule-of-law-based models can only function properly if the parties have mutual trust and can solve their internal conflicts through compromises among themselves. Political agreements based on mutual trust are more effective in the long term as legal instruments.
Effective and functional state institutions and ethnic power-sharing seem to conflict in the Western Balkans. Both are relevant criteria in relation to the EU perspective. The main question about the future of these countries is whether their citizens are able to see themselves in a win-win position as members of the European community even if they are required to give up some long-deserved ethnic privileges. Future perspectives were the number-one incentive in historical times when Swiss people united for a better future, and later when Belgian citizens decided to cooperate despite the veto rights they enjoyed. Hopefully, a better future is also to be discovered for the Western Balkans.
Literature
Abdel-Fattah, D. , The Comparative Effect of Minority Vetoes on Shared Governance in Post-Conflict Consociational Societies: Case Studies of Bosnia-Herzegovina and Lebanon. (Manuscript, College of William and Mary 2010).
Bieber, F. , ‘Institutionalizing Ethnicity in the Western Balkans’ (2004) 19 ECMI Working Paper < https://www.ecmi.de/fileadmin/redakteure/publications/pdf/working_paper_19.pdf> accessed 15 December 2020.
Bieber, F. , ‘Local Institutional Engineering: A Tale of Two Cities, Mostar and Brčko’ (2005) 3. International Peacekeeping 420–33.
Bieber, F. , ‘Power Sharing and International Intervention: Overcoming the Post-Conflict Legacy in Bosnia and Herzegovina’ in Weller, M. and Metzger, B. (eds), Settling Self-Determination Disputes (Brill 2008) 193–241.
Dardanelli, P. , and Stojanovic, N. , ‘The acid test? Competing theses on the nationality – democracy nexus and the case of Switzerland’ (2011) 2 Nations and Nationalism 357–76.
Deschouwer, K. , Ethnic structure, inequality and governance of the public sector in Belgium (Unrisd 2004).
Helbling, M. and Stojanovic, N. (eds), ‘Themed Section: Switzerland. A Nation-State or a Multi-National State?’ (2005) 4 Nations and Nationalism 712–820.
Kelleher, S. , ‘Minority Veto Rights in Power Sharing Systems: Lessons from North Macedonia, Northern Ireland and Belgium’ (2005) 13. Adalah’s Newsletter < https://www.adalah.org/uploads/oldfiles/newsletter/eng/may05/ar2.pdf> accessed 15 december 2020.
Korenica, F. and Doli, D. , Constitutional law in Kosovo (Kluwer Law International 2012).
Küpper, H. , ‘Das normative Demokratiepotenzial der kosovarischen Verfassung’ in Clewing, K. and Dzihic, V. (eds), Eigenstaatlichkeit, Demokratie und ‘Europa’ in Kosovo – Analysen und Perspektiven . (Südosteuropäische Arbeiten 2015).
Lijphart, A. , ‘Constitutional Design for Divided Societies’ (2004) 2 Journal of Democracy 96–109.
Lijphart, A. , Democracy in plural societies: A comparative exploration (Yale University Press 1977).
Mendelski, M. , ‘Where does the European Union make a difference? Rule of law development in the Western Balkans and beyond’ in Elbasani, A. (ed) EU Enlargement and Europeanization in the Western Balkans (Routledge 2013) 101–18.
Priebe, R. , ‘Beitrittsperspektive und Verfassungsreformen in den Ländern des Westlichen Balkans’ (2008) 3 Europarecht 310–19.
Roeder, Ph. G. and Rotchild, D. S. , Sustainable Peace Power and Democracy After Civil Wars (Cornell University Press 2005).
Stojanovic, N. , ‘How to solve the dilemma of power sharing? Formal and informal patterns of representation in the Swiss multilingual cantons’ (2008) 3 Representation 239–53.
Szabó, Zs. , ‘The Parliamentary Legislative Process in the Accession Countries of the Western Balkans’ (2015) 2 Jahrbuch für Ostrecht 375–418.
Mendelski (2013) 13.
Bieber (2008) 198.
Lijphart (1977) 35.
See especially articles 57–61 of the Constitution of Kosovo; Korenica and Doli (2012) 46.
For example, Constitution of North Macedonia Amendments X., XII., Constitution of Bosnia and Herzegovina Articles IV. 3., V. 2.
Bieber (2004) 10.
Lijphart (2004) 99.
Constitution of Belgium, Art. 2.
Constitution of Belgium, Art. 54.
Stojanovic (2008) 246.
Bieber (2004) 15.
Bosnia and Herzegovina 2009 EU Progress Report.
Bosnia and Herzegovina 2010 EU Progress Report.
Priebe (2008) 315.
In Bosnia and Herzegovina a fourth parliament also exists: the Brčko District Assembly (Skupština Brčko distrikta) is a unicameral legislative body of some 80,000 citizens located around the town of Brčko, and at the same time a municipal council.
Although the RS parliament is unicameral, some control functions of an upper house are vested in the Council of Peoples, which exercises its power to protect the vital national interests of the constituent peoples, as outlined in the Republika Srpska constitution. The Council does not participate in the law-making process, but controls laws, regulations, and legal acts previously adopted in the National Assembly for the purpose of defining whether an adopted law, regulation or other legal act violates any vital national interest of the constituent peoples.
For a detailed analysis of legislation data see Szabó (2015) 375–417.
Sejdić and Finci v. Bosnia and Herzegovina (27,996/06 and 34,836/06).
Preamble to the Constitution of the Republic of Kosovo.
Küpper (2015) 12.
Constitution of Bosnia and Herzegovina, Art. IV. 3) b).
Constitution of the republic of Kosovo, Art. 81.
Constitution of North Macedonia, Art. 69., 114., and 131.
Venice Commission opinion No. 508/2007.
Constitution of Bosnia and Herzegovina, Art. IV. 3. (e).
Kelleher (2005) 15.
Küpper (2015) 13.
Bieber (2005) 420.
Constitutional Court of Bosnia and Herzegovina, Decision U 4/05 on 22.04.2005.
