a Solution to Ethno-National Conflicts?
This is a very interesting essay on federalism. By presenting the Belgian and Brussels case, Maarten Theo Jans suggests, very plausibly, that we could solve many up-to-now intractable problems by implementing personal federalism, and this, we should add, in place of territorial federalism or, even worst, central state territorialism.
This is a proposal that should be not only better explored by everybody but should also represent matter of extensive experimentation in order to assess its value and to verify if it can really put to rest endemic conflicts and clashes.
If that were the case, it would be tantamount to a criminal failure to act to neglect this possibility and to carry on with state centralism and state territorialism that has already generated so many miserable situations and so much destruction and death all over the world.
Territorial definitions and the political control over territory are central to both the concepts of statehood and nationalism. The triangular definition of a state encompasses a population, a territory and a government which rules over that territory and its inhabitants. The concept of a state intimately links the idea of political power to full control over a given territory. Territory has become such a crucial element in the formation and organisation of states that it is hard to imagine the exercise of state power without reference to its exclusive rights over a particular territory. In a similar vein, the concept of ethno-nationalism has a strong territorial basis. Ethnicity and nationalism entail a developed sense of commonality between people who feel that they share a common cultural platform which can consist of characteristics based on language, custom, religion and even race. Features like common language, historical experiences and customs are markers which establish the myth of common ancestry of the ethno-national group members. All ethno-national groups are characterised by a sense of common origin which is reflected in a number of common characteristics.
This myth of common descent is, thus, substantiated by commonly shared cultural features and inspires feelings of in-group solidarity which differentiate the ethno-national group from other such groups with allegedly different histories, languages, culture and descent. Among those markers that create the contours of an ethno-national identity, territory, in the sense of the place of origin, fatherland or home, is often the cornerstone of the identity. Territorial notions are particularly potent identity-builders, because they provide the idea of the cradle or fatherland of an ethno-national group. The reference to a territorially defined space makes the myth of common origin tangible. It sets the origin of the group in a particular place and provides the group members with “roots”. When members of an ethno-national group feel that territory which they consider to be theirs is jeopardised or are faced with competing claims from other groups, the group reactions can be virulent. The emotional nationalist reactions over territory are not merely based on land issues per se, but touch upon the cornerstone of ethno-national identities. By losing land or full control over it, the ethno-national group also loses part of its self-definition and of the imaginings which bind individuals and forge them into an ethno-national group. Territorial disputes between ethno-national groups, therefore, automatically turn into disputes where group survival is felt to be at stake, a struggle in which no group is willing to lose or to accept a compromise.
The central role of territory in ethno-nationalism is also linked to the political goals of ethno-national movements. Ethno-national groups seek to rule themselves and not to be ruled by others. The idea of having their own particular history, a common present, also generates a desire to create and control their own future. The most obvious means of establishing such ethno-national self-rule is the creation of state structures with all the attributes of an independent state. The desire to have a state of their own also prompts ethno-national groups to attach an overwhelming importance to the notion of territory. Without a delimited and recognised territory over which complete authority can be exercised, state structures bear little meaning. As Gellner indicates, nationalism is an ideology which seeks to create congruence between state and group borders. Though implicit, territory is central to Gellner’s qualification of nationalism, since both nationalism and statehood are intimately linked to notions of territory.
In view of the prominence of territory in ethno-national and state concepts, it is no surprise that ethno-national conflicts often focus on territory and that the institutions developed to regulate such conflicts emanate from the same territorial reasoning. Nevertheless, other institutional techniques, whose pacifying effect is not primarily based on separating ethno-national groups in different territories, do exist. Among these extra-territorial techniques one finds institutions based on power-sharing, on electoral techniques, on group rights and on personal federalism. Despite the often unsatisfactory experiences with purely territorial solutions to ethnic conflicts (population transfers, creation of ethnically homogeneous areas, secessions, etc.), solutions based on personal federalism are seldom contemplated. The parties themselves often insist on territorial settlement of the conflict and find it hard to envisage extra-territorial solutions, certainly when territorial notions are used to mobilise and rally peoples around the issue at stake. Moreover, there is very little practical experience with personal federalism in the organisation of state power. Since there are very few and only partial real world examples of personal systems of governance, policy-makers are less inclined to engage in such institutional experiments. In spite of the multitude of reasons why extra-territorial solutions to ethno-national conflicts are used so rarely, attention needs to be drawn to the advantages of such systems and to the few examples where extra-territorial thinking has been put into practice. Territorial and personal forms of autonomy will be discussed in the remainder of this paper. Subsequently, the case of Brussels will be presented in the light of its extra-territorial institutional features. Finally, an attempt will be made to show what contribution personal federal techniques could make to the institutional settlement of the Georgian-Abkhazian conflict.
Federalism is characterised by a territorial distribution of power. The state is divided into territorial entities (regions, provinces, Länder, communities, cantons, etc.). State power is not centralised in one focal point, but is dispersed over a number of sub-states or regions. These regions typically possess almost all the attributes of a state; they have governments, legislative chambers, civil services and, most importantly, policy-making powers granted to them by the Constitution, or a basic law which cannot be altered without a degree of regional consent . The fractional distribution of political power over a number of non-central entities makes federalism particularly attractive to ethno-national groups seeking self-rule.
In unitary states there is one forum where power is concentrated, namely the national government. Ethno-national groups that constitute a minority have very limited possibilities to participate in substantial policy-making in the central government since the numerical majority is formed by another group (or other groups). Federalism offers ethno-national minorities alternative means of controlling certain policy levers without interference from national majorities.
In fact, closer scrutiny of federal principles shows that ethno-national minorities can benefit in terms of both representation and participation. In a federation, minorities are represented in the federal first chamber, as they would have been in the parliament of a unitary state. Moreover, the ethno-national group is represented in the regional parliament and government in which the group constitutes a majority. In a federal system the regions often also receive equal representation in the federal second chamber or senate. Finally, as contributors to policy, the regional governments participate as equal partners in the extensive intergovernmental policy networks which are typical of federal systems. In sum, federal systems provide enhanced possibilities of participation and representation for ethno-national groups concentrated in a region. Where a federation sets out four venues for minority participation, a unitary state merely offers minorities a limited presence in the parliament and the prospect of perpetual exclusion from governmental power. The potential benefits for minorities in federations exist in principle but their realisation also largely depends on the translation of principles into specific institutional structures. The same can be said of unitary states: in principle they provide fewer possibilities for minorities, but this does not exclude the possibility that specific institutions to accommodate ethno-national groups can be introduced in unitary structures.
By granting a minority group control over territory and a number of constitutional powers, federalism can also satisfy the intrinsic desire of ethno-national movements to determine their own destiny on the land which they see as their own. Moreover, the constitutional nature of federal division of power guarantees that the institutional provisions from which the ethno-national group benefits cannot easily be withdrawn (constitutional amendment formula and the role of the constitutional court).
Territorial federalism does reveal a number of institutional qualities that can contribute to the settlement of ethno-national conflicts. However, the usefulness of federalism is limited to those cases where ethno-national minorities are territorially concentrated. In the cases where the ethnic population spread is more pronounced and where ethno-national groups are actually dispersed and mixed over the state territory, traditional federalism based on territorial principles is much less useful and might even prove to be counter-productive. It is also important to acknowledge the intrinsic limitations of federalism as well as its advantages.
Federalism entails the tracing of internal borders. The territorial entities need to be identified as beneficiaries of political power. The drawing of lines and the construction of regions brings about issues with a high conflict potential. First, it needs to be established which territory belongs to which group. When populations are mixed, competing claims can be expected. There are no obvious criteria to determine internal borders; delimitation will therefore suffer from a degree of arbitrariness and dispute. In the case of mixed populations, each ethno-national group will view its own demands as highly legitimate; such juxtaposed claims are difficult to resolve because they force the parties into a zero-sum situation. Control over a territory by one group will be seen by other groups as a direct loss. It is important to see that the territorial logic of federal organisation actually brings territorial disputes strongly to the fore. In other words, drawing internal borders can exacerbate the very conflict that it seeks to resolve. Other sensitive questions which federal structures generate relate to the number, size and shape of the regions. Should each geographic area constitute a political region, or should regional self-rule merely be granted to those groups that demand it? Is each ethno-national group in the state entitled to a regional government? These are all questions which become important when federalism is used to regulate ethno-national tensions. The territorial nature of federalism stimulates a reformulation of the conflict in territorial terms, with all the difficulties that this entails. Federal theory and practice offer no answers to the questions raised by the delimitation of internal borders. The territorial logic of federalism tends to perpetuate territory as a contentious issue.
The creation of regions to accommodate minority problems is usually not an end-point in the development of ethno-national disputes. Even if questions concerning the number, size and shape of the regions are settled, new minority problems tend to surface when ethno-national groups are dispersed and regions are ethnically heterogeneous. Take the example of the Flemish population in Belgium: Flemings constitute a national majority but amount to only 15% of the Brussels population. As long as the Brussels area was an undifferentiated area without direct political significance, the Flemish population was a sociological minority in Brussels but did not constitute a political minority. As the Brussels area became a constituent region in the Belgian federation, the Flemish population in Brussels became a political minority in the new region. This shows that the territorialisation of minority issues brings about new minority problems when territories are ethnically heterogeneous. In this respect, federalism can actually make the political conditions of minorities worse rather than better. A national minority of 20 % can find its numbers reduced to, say, 5% in a regional sub-state. In this case federalism will cause the political representation and participation of the ethno-national group to deteriorate.
Territorial federalism is a useful tool for institutional conflict regulation, but it has its limitations and is largely inadequate when populations are mixed and when they entertain competing claims over land. The basic premise of federal structures is the distribution of power over territorial entities. This has a pacifying influence when territories are undisputed but it is ineffective when the territorial delimitations themselves are subject to inter-group rivalries. The inadequacy of the territorial solution calls for a search for alternative solutions which dissociate power and self-rule from territory.
A possible institutional solution to ethno-national conflicts involving competing claims over land could be found in the form of personal federalism (also called extra-territorial federalism). Personal federalism implies that the recipients of state power would be population groups rather than territories. In these state forms, communities could retain substantial autonomy in regions with mixed populations through formation of separate political institutions. The underlying principle of personal federalism is that governmental power is not distributed over territories but over population groups. The limits of governmental jurisdiction are determined by group membership, not by territorial borders. Authority does not stretch over territories, but covers a group of people. In a certain sense policy-making in personal federalism targets the population directly; thus ethno-national affiliation and not the place of residence determines the scope of governmental jurisdiction. Territorial organisations of power operate more indirectly, territories are used as a proxy for populations. Via governmental jurisdiction over a territory, the population living in that territory is automatically subjected to state regulations. Personal arrangements refer directly to the population with no territorial intermediary.
The great advantage of extra-territorial solutions is that a degree of self-rule can be granted to a group without assigning exclusive control over land or territory. In this way several ethno-national groups can enjoy autonomy in the same territory. Territorial claims are not required to establish group autonomy. Territorial control and claims, therefore, lose their relevance for the ethno-national groups. It is clear that the dissociation between territory and political control can be highly useful in ethnically mixed territories. Matters such as education, cultural and language policies, religion, social welfare, aspects of civil law, health policy, taxation, media, sports, etc. can be separately administered by each of the ethno-national groups over the same territory. Legislative and executive councils with jurisdiction over the above-mentioned prerogatives can be created for each of the groups. In this way ethno-national groups obtain a substantial share of autonomy and tangible political power without entering into perilous territorial issues.
Personal arrangements do require some kind of identification of the members of the ethno-national groups. However, it is not necessary to categorise all residents in a given territory into an ethno-national group. Only those individuals who choose to admit to membership of an ethno-national group should be recipients of personal autonomy. The scope of autonomy works differently in the territorial system; the jurisdiction of the regional governments extends to all inhabitants of the territory irrespective of their ethno-national affiliation, even to those residents who do not belong to the regional ethnic majority. Personal systems use voluntary ethno-national membership to determine the limits of governmental jurisdiction.
It is not even necessary for individuals formally to declare their ethno-national allegiance in order to organise a personal system. Instead of categorising the population according to ethno-national affiliation, the governmental institutions and services themselves can be ethno-nationalised. Personal governments will create a number of specific ethno-national institutions and services which target one group but which are in fact open to all individuals who wish to benefit from them. Individuals are free to use services organised by another community as long as they are willing to comply with the cultural specifics of these institutions. In this way personal governments can be limited to regulating ethno-national institutions and those individuals who chose to use these institutions. Such an institutional construction, which primarily regulates institutions and not persons, can be useful for those cases where it is difficult to place individuals in ethno-national groups in a straightforward way. In most multi-ethnic states mixed marriages and people of mixed descent might be unable or unwilling to declare themselves as members of one single ethno-national group. By ethno-nationalising institutions rather than people such difficult choices can be avoided.
In the cases where the delineation of the population groups is less problematic, sub-nationalities based on language or personal choice can be created. Those individuals who declare their ethno-national affiliation come within the jurisdiction of a corresponding personal government.
Forms of personal autonomy are appropriate for regulating matters with no direct link to territory. However, not all policy matters can be dealt with without regard to territory. Infrastructure, transport, ecological matters and natural resources are but a few of the many issues which by their nature are not directly linked to the ethnic or cultural identity of individuals, but are intrinsically territorial. These issues typically affect all inhabitants of a region irrespective of their ethno-national membership. For example, it is not feasible to have two personal governments regulating road-building activities separately on the same territory. Therefore, personal autonomy will always have to be combined with pure territorial forms of government. Personal governments can only control a limited amount of prerogatives due to the intrinsic territorial entrenchment of many policy issues.
The fact that personal autonomy needs to be combined with territorial government highlights one of the main weaknesses of these systems. Personal arrangements tend to be highly complex institutional constructions. Institutional complexity is prone to obscurity and jurisdictional disputes. Mixtures of territorial and personal arrangements can imply a substantial amount of ethno-national self-rule but will also require a high degree of intergovernmental co-operation to resolve institutional grey areas.
Personal autonomy can be a beneficial element in a general institutional settlement of an ethno-national conflict where population groups are geographically dispersed. However, personal federalism does not provide a full institutional solution in itself. Extra-territorial techniques remain dependent on some kind of territorial organisation of state power and will need to be used in conjunction with other pacifying mechanisms such as power-sharing, veto rights, constitutional guarantees, etc. Even personal federalism can not entirely do away with the constraints imposed by land and territory.
The Case of Brussels
In the previous paragraphs the territorial and personal organisation of autonomy was discussed in essentially theoretical terms. The case of Brussels provides an interesting example of how extra-territorial arrangements can contribute to the regulation of apparently insoluble ethno-national differences. For more than thirty years the Brussels issue was the main stumbling-block in the federal reform process. Brussels was a political obstacle which it took the Belgian political elite more than thirty years to overcome. Paradoxically, today the institutional fabric of Brussels is probably the single most important factor which keeps the Belgian federation from splitting up. The institutional stumbling-block became the binding fabric of the Belgian state. The ethno-national conflict over Brussels actually extended to three interrelated issues.
First, both Flemings and Francophones claim Brussels as their territory. Flemings rest the legitimacy of their claims on the fact that the Brussels area is an enclave in Flemish territory. Moreover, Brussels was historically a Flemish-speaking town, which was gradually Frenchified. The Frenchification of the 19th and the early 20th century reduced the Flemish-speaking population of Brussels to a linguistic minority of 15%. Frenchification also extended to the municipalities surrounding Brussels. The increased numbers of French speakers in and around Brussels were seen by Flemish nationalists as a loss of territory or of Flemish soil. Francophones use the demographic numbers to substantiate their claim over Brussels territory: 80% of the Belgian population in Brussels is French-speaking. Historically there has always been a substantial Francophone presence in Brussels. Over time French speakers have also come to see the Belgian capital as theirs.
Secondly, the Flemish nationalist movement historically strove for linguistic and cultural equality for the Flemish majority in the country. The demographic dominance of the Flemish population was not translated into a corresponding share of political power. The fact that until 1960 the Belgian political elites were predominantly French-speaking was at the root of the Flemish demand for self-rule. Flemish nationalists felt that they would only be able to counter the linguistic, cultural and political dominance of the French speakers through the creation of an autonomous Flemish sub-state where the Flemish language would be the only language in the public and social sphere. Francophones also advocated federalisation of the country, but for a different reason. Francophones sought control over economic policy-making in order to restructure the old 19th century industrial complex of the (Francophone) south of the country without Flemish interference. The creation of an autonomous southern Walloon region would enable Francophones to tackle their regional economic difficulties independently of Flemish economic policy choices. Despite the emerging national consensus on the desirability of a federal reform, there was serious disagreement between Flemish and Francophones regarding the institutional organisation of the federal design. On the one hand, Francophones advocated a federation with three components: a Flemish region, a Walloon region (Francophone) and a Brussels region (predominantly Francophone). Francophones argued convincingly that these regions corresponded to a sociological reality and that Brussels, since it is largely Francophone, could not be part of a Flemish region but should constitute a separate region in its own right. On the other hand, the Flemish claimed a federation with two regions based on the dual linguistic composition (although there is a German-speaking minority on the Belgian-German border) of the country. Flemish nationalists viewed Belgium as a country with two communities; they felt that the future Belgian federation should fully reflect that linguistic bi-polarity through the creation of only two sub-states. Flemish nationalists argued that since Brussels was historically Flemish, and as an enclave on Flemish soil harbouring a Flemish minority, it should be an integral part of the Flemish region. Moreover, the Flemish vehemently opposed a federation with three entities because that would set up two predominantly Francophone regions against the Flemish region. The Flemish feared that a tripolar federation would put the Flemish in a political minority again, a fate they wanted to avoid at all costs. To summarise, although Flemish and Francophones endorsed federalism, they held entirely different views on the status of Brussels in the federation. The Flemish felt that Brussels belonged to Flanders, Francophones wanted to keep Brussels aloof from Flemish political control and demanded the formation of a separate Brussels region.
Thirdly, apart from the institutional status of Brussels in the federation the socio-political situation of the Flemish minority in Brussels was another thorny issue which complicated the entire ethno-national dispute. Flemish nationalists were dedicated to protecting and consolidating the limited Flemish presence in Brussels. Given the history of Frenchification in Brussels, Flemish nationalists were determined not to abandon the remaining Flemish minority in the capital. In order to consolidate the de-facto Flemish presence in Brussels, bilingual services, power-sharing and guaranteed minority representation were high on the Flemish political agenda. The Francophones were extremely reluctant to have their impressive demographic majority in Brussels (roughly 5:1 Francophones:Flemish) curtailed by a wide variety of measures aimed at protecting the Flemish minority. Moreover, the Francophones argued that the rights granted to the Flemish minority in Brussels should be mirrored by parallel rights for the Francophone minorities in the Flemish municipalities around Brussels.
The joint effect of the above-mentioned three issues turned Brussels into a political problem that seemed insoluble. In spite of the difficulties, institutional arrangements were devised that gave both sides some satisfaction. The institutional engineering led to a complex structure that incorporates elements of territorial as well as extra-territorial techniques. The eventual institutional solution for Brussels addressed (1) the competing territorial claims, (2) the question whether the federation should be based on two or three components, and (3) the position of the Flemish minority in Brussels.
The divisive dispute over the number of federated entities (two or three) was converted to an institutional structure with twice three components. The federation consists of three territorial entities (the regions) and three partly territorial/personal units (the communities). The regions - Flanders, Wallonia and Brussels - were designed in the following way: these territorial governments have jurisdiction over territorial matters such as environment, economic policy, infrastructure, urban planning, etc.
In addition to the regions, three community governments were created. The communities are partly personal in conception. The Flemish community consists of the inhabitants of the Flemish region (territorial) and the Flemish inhabitants of Brussels (personal), the German community covers the German language area (territorial) and the French community covers the inhabitants of Wallonia without the German area (territorial) and the Francophone inhabitants of Brussels (personal). The community governments have jurisdiction over policy matters relating to ethno-national characteristics of persons, namely, language, culture, education, welfare, etc., for as far as Wallonia and Flanders are concerned the communities operate in a territorial way: all inhabitants of those regions are considered to be Francophone and Flemish respectively. In the Brussels region the community governments operate in a personal way: the Flemish community government does not regulate all inhabitants of Brussels, but its authority extends only to the Flemish inhabitants of Brussels.
It is only in Brussels that the personal characteristics of the Belgian federation come to the fore. The Brussels area actually has three  governments: a regional government (territorial), a French community government (personal) and a Flemish community government (personal). In Brussels the community governments separately regulate the affairs of their language communities which occupy the same territory. If territory is not used to designate the boundaries of government jurisdiction, another indicator has to be used. In this respect the Flemish and Francophone community governments in Brussels display another peculiar characteristic. Because of the difficulties in clearly defining ethno-national affiliation (mixed marriages, mixed origins, etc.), it was decided not to create sub-nationalities or to define community membership. Instead, the line between communities was drawn on the basis of the institutions that fulfil community tasks. Individuals do not register officially as Flemish or Francophone, but educational, cultural and welfare institutions are registered as Flemish or Francophone institutions. The community governments rule over unilingual institutions and services and those individuals who chose to use these institutions. This implies that an inhabitant of Brussels does not have to decide whether he/she is Flemish or Francophone but whether he/she is going to opt for unilingual Flemish or Francophone education, or whether she/he is going to vote for a unilingual Flemish or Francophone party. The fact that the institutions (not the population) are divided according to ethno-national affiliation allows individuals to maintain shifting identities and to use the facilities of both ethno-national groups. The “twice three” solution was acceptable to Francophones because it made Brussels a separate region outside Flanders, it was also acceptable to the Flemish because the existence of two large community governments (and a small Germanophone government) reflected the linguistic duality of the country and the Flemish community government guaranteed a continued Flemish presence in Brussels.
The issue of Flemish minority protection was tackled by the creation of a personal community government in Brussels, as mentioned above, but it still left the Flemish in a weak position as far as regional matters were concerned. Since both the Flemish minority and Francophone majority would have to govern jointly over regional affairs, additional protective measures were put in place. The Brussels regional government (territorial) consists of five members, two of whom are Flemish, irrespective of the electoral success of the Flemish political parties. Moreover, the Brussels regional government decides by consensus, which implies a de facto veto right for the Flemish ministers. The settlement provides for an “alarm bell procedure” which suspends legislative activity when a three quarter majority of a language group feels that its interests are threatened by a particular governmental or parliamentary action. These provisions largely guaranteed continued political representation of the Flemish minority and amounted to a very significant Francophone concession to Flemish demands.
The combination of personal self-rule and the power-sharing provisions encapsulated in a global federal settlement (the creation of communities and regions) reduced the competing claims over the Brussels territory. The above-mentioned institutions and a clear bilingual policy in regional affairs created the conditions under which both Flemish and Francophones could accept sharing a territory which they initially regarded as their sole birthright. Brussels remained predominantly Francophone as a separate region, but the Flemish minority gained substantial self-rule in Brussels over issues that matter to them (language, culture, education, etc.) and entrenched political representation in regional matters, where the Francophone majority was neutralised due to power-sharing techniques.
The case of Brussels shows that even the most intractable conflicts can be brought to a peaceful outcome. Moreover, institutional engineering can offer a number of context-specific solutions, which may appear highly complex but which do work. In particular, the use of personal institutions has proved to be a way out of juxtaposed claims over territory. The case of Brussels also reveals the limitations of personal scenarios: they are highly complex, they need to be applied in conjunction with other pacifying measures and, most importantly, they entail continued co-operation between conflictual population groups.
Personal Federalism. The Way Out of the Georgian-Abkhaz Conundrum?
The discussion of personal arrangements proved useful in the case of Brussels. It provided a way out of a stalemate which was caused by competing territorial claims over an ethnically mixed area. In this respect the general principles applied in Brussels may also be beneficial in other cases where ethnically mixed territories limit the usefulness of traditional territorial solutions. In the following paragraphs an attempt will be made to analyse how a personal arrangement could contribute to an institutional settlement of the Georgian-Abkhaz conflict.
The inclusion of personal federalism in an institutional settlement might look like this: (1) Abkhazia has substantial self-rule within a Georgian-Abkhaz federation; (2) within Abkhazia areas predominantly populated by Georgians or Abkhazians are ruled by a corresponding Georgian or Abkhazian territorial government; (3) within Abkhazia, personal governments are created for the mixed areas where neither group has the majority; (4) Georgians and Abkhazians share power in the regional Abkhaz government; (5) all federal actions in relation to the Abkhazian region require the agreement of the Georgian and Abkhazian community governments in Abkhazia.
The proposal would entail delineation of predominantly Abkhazian and predominantly Georgian areas as well as identification of the areas in Abkhazia where neither group is predominant. The areas where Abkhazians or Georgians are the largest group can be governed by a corresponding Abkhaz or Georgian local government which has jurisdiction over all territorial matters. The mixed areas can be ruled by extra-territorial or Abkhaz and Georgian community governments. These personal governments would not be in charge of a region or area, but would rule over those inhabitants who profess membership of the Abkhaz or Georgian group. The governments based on personal federalism (Abkhaz and Georgian community governments) could be put in charge of all ethnically sensitive areas (language, education, etc.). Of course, in the nature of things personal government cannot apply to all fields of political regulation. A number of clear territorial matters (natural resources, pollution, transport, communication, regional public infrastructure, criminal law, etc.) cannot be governed by personal entities. The territorial policy matters which relate to the mixed areas should, therefore, be regulated by an overarching Abkhazian government of ethnically mixed composition.
To summarise, the institutional structure would comprise:
(1) An Abkhazian community government with jurisdiction over ethnically sensitive issues covering a) the Abkhazians in the predominantly Abkhaz areas; b) the Abkhazians in the mixed or predominantly Georgian areas.
(2) A Georgian community government with jurisdiction over ethnically sensitive issues covering a) the Georgians in the predominantly Georgian areas; b) the Georgians in the mixed or predominantly Abkhazian areas.
(3) A territorial government for the entire Abkhazian region which would have jurisdiction over a) general issues which affect the region as whole; b) territorial policy issues (infrastructure, energy, environmental affairs, etc.) in the mixed areas. This Abkhazian territorial government should consist of representatives of the Georgian and Abkhazian communities on an equal basis or with mutual rights of veto.
(4) A provision in the Georgian constitution that all federal political action relating to the autonomous Abkhaz region requires the prior consent of the Abkhazian regional government (with an equal Abkhaz and Georgian composition).
According to this proposal, Abkhazians would enjoy substantial self-rule in the predominantly Abkhazian areas through their community government. Because of the extra-territorial community government they would not have to sacrifice any territory, nor would they have to abandon Abkhazians living in predominantly Georgian areas. The power-sharing arrangement in the regional government would give them decisive political importance irrespective of their demographic numbers. The Abkhazians could no longer be relegated to the status of a political minority. Moreover, they would have a constitutional guarantee that decisions emanating from Tbilisi could not be applied to Abkhazia without their prior consent.
Georgians could also gain significantly from this proposal. Georgia’s territorial integrity would be reinstated, with the qualification that the Georgian central government could intervene in Abkhazia only with the prior consent of both the Georgian and the Abkhazian community governments. Refugees could return to Abkhazia and would not be forced to settle in predominantly Georgian areas, but could actually return to their initial places of residence. The returning refugees would not have to fear Abkhazian political domination because they would be ruled by a Georgian community government and a joint Abkhaz-Georgian regional government.
The proposal makes no mention of the Armenian and Russian population groups in Abkhazia. Within the logic of the above proposal there seem to be two options for the Russian and Armenian populations: without preference for either option, the weaknesses of the alternatives should also be taken into account. First, following the provisions made for the Georgian and Abkhazian groups, separate community governments could also be created for the Russian and Armenian populations. This would result in equal treatment for each group, but would probably also increase the complexity of the system to the extent that it might become unmanageable. The presence of additional community governments for Russians and Armenians would have the further advantage of working against the bipolarity of Georgian-Abkhazian relations. A second option could involve incorporating the Russian and Armenian populations into one of the other community governments. Since the current Abkhazian government claims to provide ample guarantees of representation and participation for Armenian and Russian inhabitants of Abkhazia, representatives of these populations could be included into the government of the Abkhazian community and in the Abkhazian delegation to the government of Abkhazia. Ethnic Abkhazians would have to ensure appropriate inclusion of Armenian and Russian minority interests in their midst. With regard to the Armenian and Russian minorities, the use of personal federalism would merely formalise the situation as it is today, and the question remains whether this is sufficient for the Russian and Armenian populations. The inclusion of Russians and Armenians in an Abkhazian community government would (numerically) strengthen the Abkhazians in their dealings with the Georgian community government.
The downside of this proposal is the fact that the institutional structure can only function with a substantial amount of inter-ethnic co-operation. Conflicts will not cease to arise between the different ethnic groups, and since no group will be institutionally capable of overruling the other community, co-operative attitudes and compromise-orientated behaviour will be essential. Moreover, the proposal presupposes renewed multi-ethnic coexistence between populations that were previously at war with each other. Georgians and Abkhazians would live in the same streets and towns again. It will be very difficult to blot out the history of rivalry and violence which both groups experienced both before and during the war. Nevertheless, other ethnically divided countries with similar painful histories of inter-ethnic violence and atrocities (South Africa, Northern Ireland, Malaysia, etc.) have gone a long away towards resuming pacified multi-ethnic coexistence. It would be a sign of unwarranted pessimism to claim that Abkhazia could not do the same.
 The required regional consent is determined by the constitutional amendment formula in force in a given state. Often federal amendment formulas operate with qualified majorities which make constitutional reform dependent on sub-state agreement.
 The Brussels area is actually governed by four governments if the federal government is taken into account.
E. Gellner, Nations and Nationalism, Oxford, Basil Blackwell, 1983.
R. Lapidoth, Autonomy. Flexible Solutions to Ethnic Conflicts, Washington D.C., United States Institute of Peace Press, 1997.
M. Brans, H. Van Hassel, K. Rimanque, De Vlaamse Gemeenschap in Brussel. Detoepassing van art. 65 en 66 van de Brusselwet, Leuven, ACCO, 1992.