TREATISES AND DOCUMENTS JOURNAL OF ETHNIC STUDIES
RAZPRAVE IN GRADIVO REVIJA ZA NARODNOSTNA VPRAŠANJA
81 / 2018, p. 21-41
Karl Kössler
Governing Divided Societies through Territorial Autonomy? From (too) Great Expectations to a Contextualist View
Territorial autonomy remains important for the governance of divided societies. The question is rather which understanding of it dominates the political and scholarly debate. This contribution discusses different notions of territorial autonomy that have shaped this debate. It argues that a too narrow focus on concepts of ethnic-territorial autonomy such as multinational federalism fails to recognize challenges that studies on autonomy as a prescription for governing divided societies need to face. These are the issues of secessionism, political polarization and internal minorities within the autonomous territory. The paper concludes with an assessment of the track record of autonomy and highlights the importance of taking into account for such an assessment a number of legal and non-legal context factors. Put differently, territorial autonomy itself is only one of many factors contributing to (un)successful governance of divided societies.
Keywords: territorial autonomy, ethnic autonomy, divided societies, multinational federalism, constitutional design.
Upravljanje razdeljenih družb s teritorialno avtonomijo? Od (pre)velikih pričakovanj do kontekstualističnega pogleda
Teritorialna avtonomija ostaja pomemben pristop za upravljanje razdeljenih družb. Zato je ključna razprava, katero razumevanje teritorialne avtonomije prevladuje v političnih in znanstvenih razpravah. Prispevek obravnava različne koncepte in vrste teritorialne avtonomije. Ugotavlja, da preozka usmeritev na koncepte etnično-teritorialne avtonomije, kakršen je večnacionalni federalizem, ne uspe prepoznati izzivov, ki so jih zaznale in s katerimi se soočajo študije o avtonomiji, ko obravnavajo upravljanje razdeljenih družb. Med izzivi so secesionizem, politična polarizacija in (notranje) manjšine v avtonomnih teritorijih. Razpravo sklene s kritično oceno delovanja in učinkov avtonomije ter poudarja pomen upoštevanja pravnih in nepravnih dejavnikov pri takšnih ocenah. Drugače povedano, teritorialna avtonomija je le en od mnogih dejavnikov, ki lahko prispevajo k (ne)uspešnem upravljanju razdeljenih družb.
Ključne besede: teritorialna avtonomija, etnična avtonomija, razdeljene družbe, večetnični federalizem, ustavni dizajn (ustavna ureditev).
Correspondence address: Karl Kossler, Institute for Comparative Federalism, Eurac Research, Viale Druso 1, 39100 Bozen/Bolzano, Italy, e-mail: karl.koessler@eurac.edu.
ISSN 0354-0286 Print/ISSN 1854-5181 Online © Inštitut za narodnostna vprašanja (Ljubljana), http://www.inv.si
81 / 2018 TREATISES AND DOCUMENTS JOURNAL OF ETHNIC STUDIES
K. KÖSSLER Governing Divided Societies through Territorial Autonomy? From (too) Great Expectations
1. Introduction
"Constitution-making is in large part about making bets about the future" (Simeon 2009, 2). What complicates the process of federal constitution-making in ethno-culturally divided societies1 is that in these cases there is usually a higher likelihood than in others that the bets will be acutely controversial from the differing perspectives of different groups. The typically heightened political polarization in such societies is likely to entail a scenario in which extremist views are advanced, with one group portraying federalism as absolute good and another as absolute evil. The fact that a federal constitution is frequently seen as Pandora's box with the ultimate consequence of state disintegration by some and as panacea by others, is also linked to questions of timing. All too often, divided societies begin serious negotiations about viable federal arrangements only at a late stage - or even too late.
Some of the most intractable conflicts of our times, such as in Syria and Ukraine, bear testimony to this. Indeed, in many cases federalism is introduced not to prevent a (violent) conflict but to end it: political polarization, not least in regard to federalism, will thus have been intensified by the experience of armed confrontation. It should not surprise us, then, that "[m]ost of the ethnic wars of the last half century have been fought over issues of group autonomy and independence" (Gurr 2000, 195). Evidence from (violent) conflicts in divided societies demonstrates that federalism, or other forms of autonomy in a broad sense,2 have either been a bone of contention from the outset or became one in the course of the confrontation.
Against this backdrop, this article recognizes the continued relevance of federalism to divided societies, but argues that the debate about it, among scholars and practitioners alike, needs to be reframed. First, the article sets out to critically examine the rise since the 1990s of multinational federalism, ethnic autonomy and similar concepts and to assess the main arguments typically advanced against it (section 2). In so doing, the paper identifies and discusses pitfalls in the seemingly endless academic and political debate about the viability of multinational federalism. Some of the pitfalls pertain to the character of the debate itself, and others to the methodological approach, which has been prevalent so far (section 3). The conclusion points to the limitations of multinational federalism, ethnic autonomy and the like as prescriptions for constitutional design in divided societies, which requires appropriate consideration of at least seven crucial contextual factors to be successful.
RAZPRAVE IN GRADIVO REVIJA ZA NARODNOSTNA VPRAŠANJA 81 / 2018
K. KOSSLER Upravjanje razdejenih družb s teritorialno avtonomjo? Od (pre)velikih pričakovanj do ...
2. The Rise of Multinational Federalism and its critics
2.1 The Weight of History and the conceptualization of	23
Multinational Federalism
Without doubt, the issue of secession hangs like a sword of Damocles over federalism in divided societies (Kossler 2015, 263-265). Indeed, a considerable number of scholars claim that a federal constitution would just be a final step on a disintegrative path towards secession. By analogy with Albert Hirschman's seminal treatise Exit, Voice, and Loyalty (1970), they have seen federalism as pushing minority groups towards the "exit" option and as providing insufficient incentives for the "voice" option, that is, for making sincere attempts to restore well-functioning relationships within the state rather than heading for the exit. Scholars have advanced this argument since the 1990s in particular, when it seemed to be confirmed by the disintegration of the three communist ethno-federations, namely the Soviet Union, Yugoslavia and Czechoslovakia (Roeder 1991; Bunce 1999).
Bearing in mind the impact of path dependency, it is important to note that communism's characteristically ethnocentric view of autonomous territories as homelands for the specific groups that constitute a majority there clearly precedes the establishment of the Soviet Union. This view was already conceptualized and advanced by Lenin and other Bolsheviks in the early twentieth century, not least of all in their controversies with, and in contrast to, Austro-Marxist ideas of non-territorial autonomy (Bowring 2015, 146-52).3 It is equally important to acknowledge, however, that in practice the Soviet Union implemented the concept of ethnic federalism only to a limited extent, which also provides a significant lesson for federal systems in the twenty-first century. The reality of mixed settlement areas simply did not allow for the realization of the concept, because in subnational entities it inevitably produced what is today known as "internal minorities" (Choudhry 2008b, 158) or "intra-unit minorities" (Watts 2007, 232).
The fact that the envisaged ethnic-territorial overlap proved to be an illusion is demonstrated by simple figures. For example, in the Russian Soviet Federative Socialist Republic (RSFSR), by far the largest ofthe 15 Soviet republics, no more than 41 out of 127 officially recognized nationalities constituted a majority in what was supposed to be "their" territory and thus enjoyed autonomy in Lenin's sense (Codagnone & Filippov 2000, 266). Yet in spite of the apparent lack of implementation of ethnic federalism, the idea, as illusionary as it was, has lived on. It is obvious, for instance, that the notion "nationality regional autonomy" as set out in Article 4 of the Constitution of the People's Republic of China formally espouses this idea. The provision first states that " [a] ll nationalities in the People's Republic of China are equal," and then goes on to stipulate that
81 / 2018 TREATISES AND DOCUMENTS JOURNAL OF ETHNIC STUDIES
K. KÖSSLER Governing Divided Societies through Territorial Autonomy? From (too) Great Expectations
"[r]egional autonomy is practiced in areas where people of minority nationalities live in compact communities."
Precisely at the time when the collapse of the communist ethno-federations prompted the scholars mentioned above to see federalism in divided societies as a recipe for state disintegration, others started claiming the opposite. Undoubtedly, the period since the 1990s has witnessed what has been conceptualised with some variation as multinational, plurinational, ethnic and post-conflict federalism or as ethnic autonomy.4 As in the earlier case of Lenin's ethnocentric view, proponents of these concepts regard federalism primarily as a tool for minority protection. To this end, according to multinational federalists, the territorial structure should reflect ethno-cultural diversity so that internal self-determination of minority groups can prevent their external self-determination, that is, their secession. Thus, multinational federalists recommend that subnational boundaries be drawn or redrawn in such a way that nationwide minorities, at least large ones with a compact area of settlement, are transformed into regional majorities within "nationality-based units" (Kymlicka 1998, 125).5
Importantly but unsurprisingly, this minority-focused idea was advanced not only by federalism scholars but experts in minority rights (Henrard 2005, 134). As their argument goes, the two essential components of minorities' effective participation in political life are autonomy regarding their own affairs and participation in decision-making concerning affairs of the polity as a whole. This closely resembles Elazar's "self-rule and shared rule" formula, with the crucial difference, however, that everything revolves around self-rule and shared rule for a (minority) group rather than that of a territory (Elazar 1994).
2.2 Three Main critical Arguments: State Disintegration, Political Polarization and Internal Minorities
Even if multinational federalism still enjoys considerable popularity, over time it has faced several critical arguments. One group of scholars has focused on the widespread view, not least among policy-makers, that the application of this concept would be a stepping stone to the disintegration of the state concerned. The fact that states often view secession with suspicion hardly comes as a surprise, as they are guided, of course, by an inherent interest of self-preservation. Constitution-makers therefore tend to shy away, in the spirit of Madisonian political thought, from adopting provisions that could defeat the "basic enterprise" (Sunstein 1991, 633) of any constitution, which is self-preservation. While very often they see secession clauses as such provisions, they are often also suspicious of multinational federalism in this regard.
A further concern is that, apart from the loss of state unity itself, disintegration hardly ever occurs without large-scale violence. Indeed, only few federal systems, such as the West Indies Federation (1962), the Federation of Rhodesia and Nyasaland (1963), Malaysia (1965) and Czechoslovakia (1993),
RAZPRAVE IN GRADIVO REVIJA ZA NARODNOSTNA VPRAŠANJA 81 / 2018
K. KOSSLER Upravjanje razdejenih družb s teritorialno avtonomjo? Od (pre)velikih pričakovanj do ...
were dissolved in a largely peaceful manner (Watts 2008, 185). States sometimes nevertheless go beyond legitimate concerns about state unity or preservation of peace and exhibit, mostly for historical reasons, excessive anxiety about federal arrangements. Countries in Central and Eastern Europe, for example, are characterized, owing to the legacy of Lenin's abovementioned views on the matter (see section 2.1) and the collapse of communist ethno-federations, by a particularly strong political climate of "autonomy-phobia" (Palermo 2012, 82).
Those arguing that multinational federalism would facilitate secession regard it as a concept providing nationality-based subnational units with important incentives and political resources they could act and draw upon in the event of an attempt at secession. An autonomous subnational parliament, government and public administration that wields power in significant policy fields could be transformed easily into a strong national institution of an independent state. Moreover, nationality-based subnational units could use their jurisdiction over crucial issues such as culture, language use, education and media to push a proj ect of minority nation-building beyond the limits of the current state.6 Another critique is that regional and minority parties would thrive under the conditions of a multinational federation and sooner or later demand secession, either out of conviction or out of political and economic self-interest (Snyder 2000, 327).
Most proponents of multinational federalism do not deny the risk of secession. Their claim instead is that granting sufficient autonomy for nationality-based units would dissuade them from pursuing independence in the first place. In this sense, its proponents see the concept of multinational federalism as paradoxical: "while it provides national minorities with a workable alternative to secession, it also helps to make secession a more realistic alternative" (Kymlicka 1998, 142). Put differently, multinational federalism is perceived as increasing the capacity of minority groups to secede yet as intended to decrease their will to do so. Even if a failure of a federal system is not so easy to define, unless state collapse makes it self-evident,7 proponents of multinational federalism sometimes admit that countries adopting this concept may be more prone overall to failure than other federations. It is emphasized, though, that multinational federations do not experience difficulties because they are federal, but that they are federal because they experience difficulties that make a unitary design unfeasible (Watts 2007, 230-231).
A second group of critics of multinational federalism claims that the entrenchment of ethnicity as the basis of the territorial structure would perpetuate these differences and result in polarized political discourse. This claim is linked to some extent with the abovementioned secession argument, in that polarization between the two extremes of centralization and separatism is seen as pervading ordinary politics: in such a scenario, the scope of routine politics open to pragmatic compromise is massively reduced. The factors motivating these confrontational dynamics are identified in the official recognition that multinational federalism grants to "competing nation-state projects that pit ho-
81 / 2018 TREATISES AND DOCUMENTS JOURNAL OF ETHNIC STUDIES
K. KÖSSLER Governing Divided Societies through Territorial Autonomy? From (too) Great Expectations
meland governments against the common-state government" (Roeder 2009, 209). In line with what has been called "groupism" (Brubaker 2004, 2), that is, the tendency to perceive bounded groups as basic constituents of society and politics, ordinary political questions would thus tend to be seen in a polarized way through a "minority lens".
The concern that polarization would make a culture of compromise impossible had also been voiced by Daniel Elazar, one of the founding fathers of modern federal studies, with regard to what he called "ethnic federations." He even claimed that "ethnic nationalism is probably the strongest force against federalism," given that
ethnic federations are among the most difficult of all to sustain and are least likely to survive because constituent units based on ethnic nationalisms normally do not want to merge into the kind of tight-knit units necessary for federation. It may be that confederations of ethnic states have a better chance of success (Elazar 1994, 167).
Apart from these considerations of viability and failure, Elazar identified a more profoundly deep-rooted contradiction in the rationale for ethnic federations, namely that "ethnic nationalism tends to subordinate all free government to its uncompromising position. Federalism is a democratic middle way requiring negotiation and compromise. All aspects of society fostering uncompromising positions make federalism more difficult, if not impossible" (Elazar 1994, 168).
In addition to the secession and polarization arguments, a third line of argumentation has advocated a nuanced view of multinational federalism cognisant of the inherent gap and contradiction between the concept's strong ethnic-territorial link, presuming the homogeneity of nationality-based entities, and a reality in which the latter boast in many cases considerable ethno-cultural diversity. Acknowledging that the concept was and still is successful in some divided societies, certain scholars have highlighted weaknesses of multinational federalism from the perspective both of theoretical considerations and comparative empirical evidence (Palermo 2015; Kossler 2015). The point is that the more the law has reinforced the ethnic-territorial link,8 which is the essence of the concept, the more that adequate legal recognition of the internal diversity within a subnational entity has been compromised.
In terms of this logic, the dominant group(s) are seen as owning the autonomous territory as a homeland and, by implication, enjoying territorially based power instead of sharing it with other groups. Paradoxically, while multinational federalism legally recognises diversity on the national scale, it therefore typically fails to acknowledge diversity on the subnational scale. On the one hand, the concept entails dedication to fighting the unitary nation-state model; on the other, it often replicates precisely this model at the subnational level in what may be called "nation-regions" and merely shifts problematic majority-minority relations to a lower level of government.
RAZPRAVE IN GRADIVO REVIJA ZA NARODNOSTNA VPRAŠANJA 81 / 2018
K. KOSSLER Upravjanje razdejenih družb s teritorialno avtonomjo? Od (pre)velikih pričakovanj do ...
This inherent problem with any constitutional design relying on the ethnic-territorial link was already identified a century ago by Karl Renner: "Ifyou live in my territory, you are subject to my domination, my law, and my language! It is the expression ofdomination, not ofequal rights" (2005 [1918], 27-28). Renner's solution to the problem of minority groups in someone else's homeland was non-territorial autonomy. The latter is, however, quite a weak instrument from the perspective of contemporary empirical evidence. All too often, it is attractive to policy-makers (of the national government) precisely for its symbolic rather than practical utility and "the chance that non-territorial arrangements will fall short oftrue autonomy" (Coakley 2016, 182). Besides non-territorial autonomy, there are other constitutional design options to prevent or at least mitigate the marginalization of internal minorities within subnational entities (Kossler 2018). One way to achieve this is to set external substantive limits on majori-tarianism, such as through an extensive bill of rights in the national constitution or by providing national government with powers to intervene on behalf of internal minorities. Another way is to place internal procedural limits on majori-tarian decision-making at the subnational level through regional power-sharing.9 Interestingly, the view that multinational federalism may be a double-edged sword precisely because of the problem with internal minorities has also been espoused by scholars who were initially supporters of the rationale for multinational federalism. Yash Ghai, for example, had defined forms of autonomy "as [a] device to allow minorities claiming a distinct identity to exercise control over affairs of special concern to them while allowing the larger entity to exercise those powers that cover common interest" (2005, 38). In a more recent publication, however, he appears to have reservations regarding the rationale of autonomy for a nationwide minority group and recognizes its potential negative implications for internal minorities:
Autonomy is a response to marginalisation, or oppression, but can itself all too easily become an instrument for the marginalisation of others. /.../ Starting as a response to discrimination, it sets up its own orthodoxy. Justified in the name of diversity, it tends to entrench boundaries between cultures. Instead ofdefining identity as a composite of different values and multiple affiliations, identity is perceived as made up of a singular and exclusive affiliation (Woodman & Ghai 2013, 485).
3. Multinational Federalism as Viable constitutional Design? Pitfalls of an Endless Debate
Some critics of multinational federalism have claimed, based on the disintegration of the communist ethno-federations, that multinational federalism has "a terrible track record" (Snyder 2000, 327). Indeed, at first glance it seems temp-
81 / 2018 TREATISES AND DOCUMENTS JOURNAL OF ETHNIC STUDIES
K. KÖSSLER Governing Divided Societies through Territorial Autonomy? From (too) Great Expectations
tingly straightforward to come to this conclusion. Of all the communist states in Central and Eastern Europe, it was only federations that broke apart and all three of them did, i.e. the Soviet Union, Yugoslavia and Czechoslovakia.
Yet it remains a matter of contention whether factors other than the federal structure were decisive in the (violent) break-up of these states. Proponents of multinational federalism claim that the three communist states are "false negatives" (McGarry & O'Leary 2009, 9) and that critics have constructed a case against the concept by relying on what were sham federations. This would be so because the constituent units of these federations were forced together without regard to self-rule or shared rule and were characterized by disregard of the rule of law, by the lack of neutral judicial umpires regarding the distribution of powers, and by the political principle of democratic centralism, which was superimposed on and offset any decentralization in constitutional terms. Moreover, those pointing to the communist ethno-federations as negative precedents are accused of using implausible counterfactuals by claiming that multinational federalism was unnecessary in view of better (unitary state) alternatives. A further claim is that they get historical causation wrong, as the break-up of these states followed, at least in the Yugoslav case, a trend of centralization. As for the Czechoslovakian velvet divorce, this was in many respects an idiosyncratic case. If it had not been for strategic political gains of the leaders of the Czech and Slovakian parts of the federation, the latter might not have been dissolved. Nevertheless, this is precisely what happened, without a referendum and hence in disregard of constitutional provisions and, quite likely, against the will of the people in both parts of the country (Stein, 1997). Thus, unless for political rather than federal-structural reasons, Czechoslovakia might well still exist today.
This early debate on the merits of multinational federalism in the 1990s, which was fueled by and focused on the dissolution of the communist ethno-federations, later gave way to a more nuanced discussion. It has involved, for example, a more careful consideration of single secession-inducing and secession-preventing factors, among which certainly is the federal design but also others such as the political will and mobilization capacity of separatists, as well as economic and sociological determinants (Erk & Anderson 2012). Yet even though the debate on multinational federalism has matured and become more differentiated, it is still highly questionable whether it could lead to any reliable results. It seems, in other words, that this confrontation - given the manner in which it has unfolded so far - might be not only endless but also pointless. This is so due to a number of problems that have beset the controversy. They fall into one of two categories: problems related to characteristics ofthe debate itself, and others connected to a questionable methodological approach.
RAZPRAVE IN GRADIVO REVIJA ZA NARODNOSTNA VPRAŠANJA 81 / 2018
K. KOSSLER Upravjanje razdejenih družb s teritorialno avtonomjo? Od (pre)velikih pričakovanj do ...
3.1 The Debate: Equation with Federalism and Overemphasis on Territorial (Re)Organization
Even if there has been, as just described, an evolution towards more nuanced analysis, much of the debate about the virtues and vices of multinational federalism continues to be highly polarized. Thus, discussion is often obstructed by an oversimplified confrontation between believers in the concept and their critics. In this light, it has appeared either as a cure or curse and, as this polarized debate has occupied so much space, multinational federalism often has come to be conflated and confused with federalism (in divided societies). But while the latter is a broader research field, the former is, as mentioned above (see section 2.1), merely a specific concept that aims to transform nationwide minorities into regional majorities within nationality-based units.
This transformation entails a second problem related to the debate on multinational federalism, as by definition the concept focuses, in terms of constitutional design, on the issue of territorial demarcation. At this point, it is important to acknowledge that questions of how to draw and redraw subnational boundaries are far from irrelevant for divided societies. The overemphasis on the issue seems problematic, though, in that a blinkered view may entail a failure to pay attention to other, equally important matters of constitutional design. Cases in point are the abovementioned limits on the majoritarianization and margi-nalization of internal minorities within subnational entities (see section 2.2). It is true that the issue of territorial (re)organization may have a profound impact on the viability of a federal system, even more so in view of certain further consequences in divided societies. Some subnational entities may be neutral with regard to religion, whereas others adopt one or even more official religion.10 But this zero option does not exist in the case oflanguage. Any entity must choose one or more official languages in which to conduct legislative affairs, deliver public services, provide education and administer justice. Solving territorial questions is particularly intractable if it is attempted in the wake of violent conflict and involves decisions about the ownership ofvaluable economic resources. The failure to define the boundaries of the Kurdistan region clearly, in the context of the disputed territories mentioned in article 140 of the Constitution of Iraq, is testimony to these problems (Galbraith 2008).
In India, by contrast, subnational boundaries have been repeatedly redefined since 1956 according to the rationale of multinational federalism. This process of creating new states is indeed often regarded as having succeeded in accommodating territorial claims based primarily on linguistic identities (Caste-llino & Domínguez Redondo, 2006). Such an identity-based territorial structure was not only diametrically opposed to the claim of Jawaharlal Nehru, the first Indian Prime Minister, that only heterogeneous states would protect national unity and prevent disintegration (King 1997, 138). It also goes against
81 / 2018 TREATISES AND DOCUMENTS JOURNAL OF ETHNIC STUDIES
K. KÖSSLER Governing Divided Societies through Territorial Autonomy? From (too) Great Expectations
the advice of scholars who seek to avoid a transformation of ethnic divisions into political cleavages. Claiming that such a transformation would hamper cooperation between groups and empower extremists over moderates, the proposed alternative is to "make moderation pay" (Horowitz 1990).
As for territorial demarcation, it is thus recommended that subnational boundaries be drawn in such a way that they produce heterogeneous entities, with Nigeria's Second Republic usually cited as a leading example.11 In practice, however, this much-debated normative question of whether subnational boundaries should follow the prescription of multinational federalism (as in India) or not (as in Nigeria's Second Republic) is to a considerable degree subject to political feasibility. All too often, these boundaries are dictated by power relations in political (and often military) terms rather than following theoretical blueprints. This is simply a further reason for not overemphasizing territorial (re)organization and so losing sight of other important issues of constitutional design.
3.2 The Methodology: Biased case Selection and Overbroad categorization
Besides those within the debate itself, a second set of problems has also made it impossible to find a general, clear and reliable answer to the question of whether multinational federalism is a cure or curse. These issues pertain to research methodology. Fervent advocates of the concept as well as their equally ardent opponents have tended to be biased in their case selection by focusing on countries that confirm their assumptions (Choudhry & Hume 2011, 368). As mentioned above, opposition to multinational federalism largely originated from early analyses of the collapse of the communist ethno-federations and, albeit to a much lesser degree, of the equally failed post-colonial federations. The latter were often established in the process of decolonization by a departing imperialist power (e.g. the British in Nigeria) or even by two such powers in collaboration (e.g. the British and French in Cameroon). Thus, a series of newly imposed federal systems came into being in Africa, Asia and the Caribbean, most of which proved to be short-lived.12 Advocates of multinational federalism, by contrast, have tended to take recourse in more encouraging counterexamples, such as Canada, Belgium, Spain and sometimes India.
Another methodological problem is related to categorization, or more precisely to the dichotomy between those systems that follow the rationale of multinational federalism in order to create nationality-based units and those that do not and are variously termed mononational or territorial federations. However, this seemingly obvious dichotomy, one relying again on the criterion of territorial demarcation, obscures the fact that multinational federations do not constitute a coherent and clear-cut category. Indeed, the categorization does not account
RAZPRAVE IN GRADIVO REVIJA ZA NARODNOSTNA VPRAŠANJA 81 / 2018
K. KOSSLER Upravjanje razdejenih družb s teritorialno avtonomjo? Od (pre)velikih pričakovanj do ...
for the enormous variety of systems fulfilling this criterion in numerous other respects. Multinational federations simply seem to vary far too much in their constitutional designs for them all to be lumped together in a single uniform category.
The crucial problem is that, in view of these considerable dissimilarities, they can hardly be expected to have similar dynamics in terms of the risk of disintegration and the other the alleged problems mentioned above (see section 2.2). Why should we expect, for example, that the constitutional designs of India, Canada and Belgium, all widely recognized as multinational federations, have the same effects, even though they differ immensely from each other, for example, concerning the degree of (de)centralization?
In the case of India, the national parliament may temporarily legislate even on subject matters that are constitutionally assigned to the states if two-thirds of the second chamber, namely the Rajya Sabha, deem this "is necessary or expedient in the national interest" (Article 249 ofthe Indian Constitution); even more notorious are the powers of the President of India to impose emergency rule, powers through which India can be, in the words of B. R. Ambedkar, "both unitary as well as federal according to the requirements of time and circumstances" (Sabha, Ambedkar, cited in Shiva Rao 1968, 810). These powers may be invoked in the event of a national emergency, that is, "a threat to the security of India or any part of its territory caused by war, external aggression or armed rebellion" (Article 352), a state emergency due to "the failure of the constitutional machinery in a state" (Article 356), or a financial emergency (Article 360). Even though in two seminal judgments the Indian Supreme Court established some limits to these powers,13 they remain a political option.
This stands in contrast to Canada, where, according to the Supreme Court (1981), the potentially centralizing power ofthe national government to disallow and reserve provincial legislation (section 90 of the 1867 Constitution Act), "although in law still open, [has] to all intents and purposes, fallen into disuse."14 While Canada is in fact today one of the most decentralized (multinational) federations in the world, Belgium still surpasses it in this regard. First, in contrast to nearly all other federal systems, it has no supremacy clause. Given that the "decrees" and "ordinances" adopted by the subnational parliaments are granted the same legal force as the formal "laws" ofthe national parliament (Articles 127130 and 134 of the Belgian Constitution), conflicts over material jurisdiction are decided, in the absence of legal hierarchy, through consultation in the Council of State or by the Constitutional Court (Articles 141-142), (Delperee 1993, 133-143, 138).
Secondly, Belgium's subnational entities have the power to regulate international cooperation, including the conclusion of treaties, for all matters that fall within their internal competences (Articles 127-128 and 167). This alignment of internal and external powers has entailed, in combination with the transfer of
31
81 / 2018 TREATISES AND DOCUMENTS JOURNAL OF ETHNIC STUDIES
K. KÖSSLER Governing Divided Societies through Territorial Autonomy? From (too) Great Expectations
32
4. Conclusion: The Potential and Limits of Constitutional Design
This article has sought to uncover some weaknesses of multinational federalism as a concept and to critically analyse its viability for constitutional design. Yet this is not to claim the concept is doomed to fail in each and every case. Individual situations are so tremendously different, with so many intervening variables -historical, political, economic and social variables - that there is no direct and straight line between the design of a constitution and its impact on the ground. Indeed, this is why "[t]he world of constitutional predictions is littered with failed predictions and unanticipated consequences" (Simeon 2009, 2).
Some proponents of multinational federalism have explicitly recognized that its impact is highly dependent on context. They have therefore taken a more nuanced approach by regarding the concept's success as contingent on its facilitation or even enablement by certain favourable conditions.15 According to such scholars, the paradox of multinational federalism is that it "may provide cultural minorities with greater resources to engage in collective action, leading to a rise in protest events, at the same time it may erode the demand for sovereignty" (Hechter 2001, 146). As for the solution to this paradox and that of the question of what accounts for the fact that there are successful and obviously unsuccessful cases, the scholars in question point to the necessity to give appropriate consideration to contextual factors as critical exogenous determinants. Generally, more realistic approaches to and more modest expectations of the power of constitutional design seem important.
First, the degree to which the largest group is demographically preponderant is regarded as a contextual factor that makes a difference. As the argument goes, such a situation seems to make it easier for the largest group to grant concessions to minority groups without feeling threatened and to make it strong enough, in political and economic terms, to oppose secessionism successfully (O'Leary 2001). Apart from the demographical situation itself, however, a related issue concerns how territorial organization deals with this situation.
For at least three decades observers have warned against the inherent instability of dyadic (multinational) federations with merely two subnational
extensive powers through successive state reforms, an extraordinary degree of subnational autonomy, leading some observers to describe Belgium as having a "borderline constitution" (Mancini 2008, 576) that blends federal features with confederal elements. In short, there is a tremendous disparity between India, on the one hand, and, on the other, Canada - and even more so Belgium - in the degree of (de)centralization, a disparity making it a rather unlikely supposition that these three multinational federations are animated by very similar federal dynamics.
RAZPRAVE IN GRADIVO REVIJA ZA NARODNOSTNA VPRAŠANJA 81 / 2018
K. KOSSLER Upravjanje razdejenih družb s teritorialno avtonomjo? Od (pre)velikih pričakovanj do ...
entities. A bipolar constellation of this kind tends to preclude the possibility of subnational entities developing a variety of alliances among themselves regarding different political issues and instead to produce a political process that accretes around, and reinforces, a single cleavage (Duchacek 1988). As a result, even bargaining on issues that are not germane to identity politics can be perceived by the demographically and politically non-dominant entity (e.g. West Pakistan before 1971 or Slovakia before 1992) as an all-or-nothing, zero-sum game, one which they are bound to lose.
While, therefore, a large group should not be concentrated in one subnational entity within a dyadic federation, also splitting up that second entity does not seem to contribute to stability. It has been claimed that any constellation in which a large group dominates, demographically and in other respects, by means of a "core ethnic region" (Hale 2004, 166)16 would be inherently instable. The argument is that such a constellation creates a "dual power" situation, with politicians from this core region being able to challenge and profoundly influence the national government; in turn, the failure of the central government to represent the interests of the smaller subnational entities adequately would fuel secessionism as a defensive reaction.
In terms of empirical research, this argument mainly builds on the collapse of several federal countries with core ethnic regions, either dyadic ones (e.g. Czechoslovakia) or those with more than two subnational entities (e.g. the Soviet Union and the First Republic in Nigeria). Conversely, the survival of countries without such a hegemonic region (e.g. Canada, Switzerland, Spain and Nigeria's Second and Third Republics) are typically adduced as examples in support this argument.
Secondly, there is another socio-demographic factor that impinges on the stability of multinational federations, namely the relationship between different cleavages such as ethnic, linguistic or religious differences. First, cleavages are often highly fluid because they are socially constructed at a given point in time. In South Africa, for instance, the primary cleavage was during apartheid, the one between black and white, whereas it is only in the post-apartheid era that divisions within these two categories became more apparent and politically salient (Murray & Simeon 2007, 709-710). Secondly, particular cleavages, such as language and religion, may be either reinforcing, as in Belgium as well as in Bosnia and Herzegovina, and hence deepen the alienation between different groups, or cross-cutting, as is the case for the most part in Switzerland or India (Dardanelli & Stojanovic 2011; Arora 2010, 211), and thus attenuate each other. In this context, it has also been argued that a federal system should, against the rationale of multinational federalism, deliberately avoid taking ethno-cultural diversity as the basis for its territorial structure because this would mean a reinforcement of existing cleavages with a territorial one (see section 3.1).
Thirdly, the historical dynamics at work at the founding of the federal system play a role. Systems qualifying as what has been called putting- or
81 / 2018 TREATISES AND DOCUMENTS JOURNAL OF ETHNIC STUDIES
K. KÖSSLER Governing Divided Societies through Territorial Autonomy? From (too) Great Expectations
forced-together federations clearly appear to be more fragile than coming-or holding-together federations.17 Coming-together federations, established by previously independent entities, and most holding-together federations, formed to accommodate demands of (certain) subnational entities, are typically characterized - from the start or belatedly - by an element of voluntary union. Putting- or forced-together federations lack this element at the beginning, and in most cases forever, with the result that a sense of local ownership of the constitution cannot take root.
Fourthly, economic prosperity has considerable impact. Even if it does not constitute a necessary condition for the success of multinational federalism, it certainly reduces the salience of conflicts over the distribution of critical and/ or scarce resources. Conversely, economic failure may provide an incentive, especially for more prosperous subnational entities, to secede from a state which is seen as holding them back (McGarry & O'Leary 2009, 19). Such a crisis of legitimacy of the state concerned - economically and, as a result, politically - occurred during the collapse of the planned-economy system in the communist ethno-federations, with the relatively prosperous Baltic republics, as well as Slovenia and Croatia, taking the lead in the break-up of their respective countries. Similar processes also contributed significantly to the rise of Catalan secessionism in the context of Spain's economic and financial crisis.
Fifthly, geopolitical dynamics and the involvement of international actors, in particular of kin-states (Palermo & Sabanadze 2011) or what we might call "kin-regions," have a substantial and sometimes decisive impact on the prospects of multinational federations. Some groups have ethnic kin who dominate another state or a region of another state, as is commonplace, for example, in Central and Eastern Europe. This may affect multinational federations critically for better or for worse. A case in point is Bosnia and Herzegovina, where inter-group and intergovernmental relations have often been influenced for the more than two decades of the country's existence by moderating or radicalizing political dynamics in neighboring Serbia and Croatia.
In addition to any kin-states or "kin-regions," other states and international organizations may also have an interest in intervention. Overall, the record of such international engagement is mixed at best. It may have a positive impact where interventionism is balanced by the encouragement of local ownership. Such a role of benign intervention by facilitating a peace agreement was played, for instance, by Ireland and the United States regarding the 1998 Belfast Agreement (McGarry & O'Leary 2008). This agreement then provided the basis for the subsequent devolution ofpowers to the Northern Ireland Assembly and Executive.
By contrast, the balance between external intervention and local ownership has been lacking from the outset in Bosnia and Herzegovina. For example, it was often only the three international judges sitting on the country's Constitutional
RAZPRAVE IN GRADIVO REVIJA ZA NARODNOSTNA VPRAŠANJA 81 / 2018
K. KOSSLER Upravjanje razdejenih družb s teritorialno avtonomjo? Od (pre)velikih pričakovanj do ...
Court who tipped the scales in crucial cases regarding the federal system.18 Moreover, in view of internal stalemate, the High Representative, appointed by the Peace Implementation Council (PIC),19 has used his power repeatedly to issue decisions that even amended the constitutions of the country's two autonomous entities (OHR 2002a; OHR 2002b). Thus, the role of this High Representative has changed "from that of a supervisor of the peace implementation process to its main actor" (Woelk 2012, 119). Although it served to overcome continued obstructionism and disagreement between elected representatives, such inter-ventionism is unlikely to have positive effects in the long run. It simply exempts local actors from the need to negotiate and compromise, something that cannot be permanently substituted for by international actors.
Sixthly, the political context of a stable system that cherishes democracy and the rule of law appears to enhance multinational federalism's chances of success. This is primarily because such a system is more likely to foster the political culture of pluralism, and enable the open negotiations, that are essential to federalism (see below).
Moreover, it entails, through popular elections, enhanced democratic legitimacy and accountability of the political actors involved, that is, at both the national and subnational levels of government (Nordquist 1998, 69).
In a number of cases, however, federal systems have been established not in a stable democracy but during a transition towards democracy. Cases in point are the abovementioned post-colonial federations, to which critics of multinational federalism often refer; further examples are found in the wake of the "third wave" of democratization, especially in the 1990s, e.g. Ethiopia and the Russian Federation (Huntington 1993). Whereas an "authentic multinational federation is democratic" (McGarry & O'Leary 2007, 202), with such systems mostly being successful, "non-authentic" democratizing federations have often witnessed enthusiasm and solemn declarations giving way to very weak self-rule and shared rule in practice.
Among other things, this is due to the fact that democratization favours the resurfacing of minority identities that were suppressed under authoritarian conditions. In the new context of political competition, they may be nurtured and exploited by nationalist parties. Such dynamics were evident in the violent break-up of the Soviet Union and Yugoslavia and, more recently, in the Civil War in Nepal (1996-2006) where the Maoist party had championed its so-called "Ethnic Policy of Nepal" from 1995 onwards (Singh & Kukreja 2014, 405).
Finally, a seventh contextual factor which is critical to the functioning any federal constitutional design is the existence of a federal political culture (Burgess 2012). To be sure, such a culture may arguably be nurtured to some extent by constitutional design, as law certainly has, according to socio-legal studies, some capacity to influence and reconstitute people's values (Post 2003). Yet it is highly unlikely that such a political culture will take root in situations where,
81 / 2018 TREATISES AND DOCUMENTS JOURNAL OF ETHNIC STUDIES
K. KÖSSLER Governing Divided Societies through Territorial Autonomy? From (too) Great Expectations
particularly after a (violent) conflict, the parties to the federation are entirely-averse to the spirit of federalism and the negotiation and compromise it entails. These ultimate ties that bind are therefore, to a considerable degree, something presupposed by constitutional design. While key elements of a federal culture based on federal loyalty, mutual consideration and negotiated compromise may be entrenched in constitutional provisions (e.g. Belgium) or constitutional jurisprudence (e.g. Spain),20 this of course does not secure compliance with these rules in practice. For these rules to be observed rather than be dead letters, there needs to be a pre-legal political culture which is capable of sustaining them.
Some have gone as far as to claim that this culture is the only thing that keeps federalism alive. Federalism would be nothing more than a "constitutional legal fiction which can be given whatever content seems to be appropriate at the moment," provided only that this fiction succeeds in luring politicians into accepting a national government: "Once the central government is actually in operation, however, what maintains or destroys local autonomy is not the more or less superficial feature of federalism but the more profound characteristics of the political culture" (Riker 1969, 142, 146). Though this statement on the superficiality of federalism as a feature of constitutional design may seem exaggerated, there arguably is some truth in it. Federalism may indeed be interpreted as a "constitutional bargain" (Riker 1964, 1) between future national and subnational government leaders. A culture of negotiation in good faith remains crucial, then, for the (successful) establishment and existence of a federal system and, in some cases, as the Canadian Supreme Court (1998) has ruled, even for its dissolution.21
In the past, the vast literature on multinational federalism and ethnic autonomy has focused mostly on the impact of these concepts' constitutional design imperatives on the viability of federal systems. As there is no such impact which is unidirectional and unfiltered, research on federalism and other forms of autonomy in divided societies should place more emphasis on taking due account of the crucial contextual factors outlined in this article.
References
Arora, B., 2010. Republic of India. In L. Moreno & C. Colino (eds.) Diversity and Unity in Federal Countries. McGill-Queens University Press, Montreal, Kingston, 200-226.
Belgian Constitution, 2017. Legal Affairs and Parliamentary Documentation Department of the Belgian House of Representatives, https://www.dekamer.be/kvvcr/pdf_sections/publi-cations/constitution/GrondwetUK.pdf (accessed 5 October 2018).
Bermeo, N., 2002. The Import of Institutions. Journal of Democracy 13, 96-110.
Bowring, B., 2015. From Empire to Multilateral Player: The Deep Roots of Autonomy in Russia. In T. Malloy & F. Palermo (eds.) Minority Accommodation Through Territorial and Non-Territorial Autonomy. Oxford University Press, Oxford, 133-160.
Brubaker, R., 1996. Nationalism Reframed. Cambridge University Press, Cambridge.
RAZPRAVE IN GRADIVO REVIJA ZA NARODNOSTNA VPRAŠANJA 81 / 2018
K. KOSSLER Upravjanje razdejenih družb s teritorialno avtonomjo? Od (pre)velikih pričakovanj do ...
Brubaker, R., 2004. Ethnicity without Groups. Harvard University Press, Cambridge.
Bunce, V., 1999. Subversive Institutions: The Design and the Destruction of Socialism and the State. Cambridge University Press, Cambridge.
Burgess, M., 2012. In Search of the Federal Spirit: New Comparative Empirical and Theoretical Perspectives. Oxford University Press, Oxford.
Castellino, J. & Domínguez Redondo, E., 2006. Minority Rights in Asia. Oxford University Press, Oxford.
Choudhry, S. & Hume, N., 2011. Federalism, Secession and Devolution: From Classical to Post-Conflict Federalism. In T. Ginsburg & R. Dixon (eds.) Research Handbook on Comparative Constitutional Law. Edward Elgar, Cheltenham, 356-386.
Choudhry, S., 2008a. Bridging Comparative Politics and Comparative Constitutional Law: Constitutional Design for Divided Societies. In S. Choudhry (ed.) Constitutional Design for Divided Societies: Integration or Accommodation? Oxford University Press, Oxford, 3-40.
Choudhry, S., 2008b. Does the World Need More Canada? The Politics of the Canadian Model in Constitutional Politics and Political Theory. In S. Choudhry (ed.) Constitutional Design for Divided Societies: Integration or Accommodation? Oxford University Press, Oxford, 141-172.
Coakley, J., 2016. Conclusion: Patterns of Non-Territorial Autonomy. Ethnopolitics 15(1), 166-185.
Codagnone, C. & Filippov, V., 2000. Equity, Exit and National Identity in a Multinational Federation: The "Multicultural Constitutional Patriotism" Project in Russia. Journal of Ethnic and Migration Studies 26(2), 263-288.
Constitution of India, 2007. National Portal of India, https://www.india.gov.in/my-government/ constitution-india/constitution-india-full-text (accessed 5 October 2018).
Constitution of the People's Republic of China, 2004. National People's Congress of the People's Republic of China, http://www.npc.gov.cn/englishnpc/Constitution/node_2825.htm (accessed 5 October 2018).
Constitutional Court of Bosnia and Herzegovina, 2000. Partial Decision U 5/98 III of 1 July 2000, http://www.ustavnisud.ba/dokumenti/_ en/u-5-98-12209.pdf (accessed 5 October 2018).
Constitutional Court of Spain, 1982. STC 18/1982, http://hj.tribunalconstitucional.es/en/ Resolucion/Show/60 (accessed 5 October 2018).
Constitutional Court of Spain, 1986. STC 11/1986, http://hj.tribunalconstitucional.es/en/ Resolucion/Show/574 (accessed 5 October 2018).
Dardanelli, P. & Stojanovic, N., 2011. The Acid Test? Competing Theses on the Nationality-Democracy Nexus and the Case of Switzerland. Nations and Nationalism 17(2), 357-376.
Delpérée, F., 1993. Le fédéralisme de confrontation à la Belge. In J. Kramer (ed.) Föderalismus zwischen Integration und Sezession. Nomos, Baden-Baden, 121-148.
Duchacek, I., 1988. Dyadic Federations and Confederations. Publius: The Journal of Federalism 2(18), 5-31.
Elazar, D., 1994. Federalism and the Way to Peace. Institute of Intergovernmental Affairs, Kingston.
Erk, J. & Anderson, L., 2012. The Paradox of Federalism: Does Self-Rule Accommodate or Exacerbate Ethnic Divisions? Routledge, New York.
Federal Constitution of the Swiss Confederation, 1999. The Portal ofthe Swiss goverment, https:// www.admin.ch/opc/en/classified-compilation/19995395/index.html (accessed 5 October 2018).
81 / 2018 TREATISES AND DOCUMENTS JOURNAL OF ETHNIC STUDIES
K. KÖSSLER Governing Divided Societies through Territorial Autonomy? From (too) Great Expectations
Galbraith, P., 2008. Unintended Consequences: How War in Iraq Strengthened America's Enemies. Simon & Schuster, New York.
Ghai, Y., 2005. Autonomy as a Participatory Right in the Modern Democratic State: Public Participation, Autonomy and Minorities. In Z. A. Skubarty (ed.) Beyond a One-Dimensional State: An Emerging Right to Autonomy? Martinus Nijhoff Leiden, Boston, 3-45.
Gurr, T. R., 2000. Peoples Versus States: Minorities at Risk in the New Century. United States Institute of Peace Press, Washington DC.
Hale, H., 2004. Divided we Stand: Institutional Sources of Ethnofederal State Survival and Collapse. World Politics 56(2), 165-193.
Hechter, M., 2001. Containing Nationalism. Oxford University Press, New York.
Henrard, K., 2005. "Participation," "Representation" and "Autonomy" in the Lund Recommendations and their Reflections in the Supervision of the FCNM and Several Human Rights Conventions. International Journal on Minority and Group Rights 2/3(12), 133-168.
Hirschman, A., 1970. Exit, Voice, and Loyalty: Responses to Decline in Firms, Organizations, and States. Harvard University Press, Cambridge.
Horowitz, D., 1985. Ethnic Groups in Conflict. University of California Press, Berkely.
Horowitz, D., 1990. Making Moderation Pay: The Comparative Politics of Ethnic Conflict Management. In J. V. Montville (ed.) Conflict and Peacemaking in Multiethnic Societies. Lexington Books, Lexington.
Huntington, S., 1993. The Third Wave: Democratization in the Late Twentieth Century. University of Oklahoma Press, Oklahoma.
Indira Nehru Gandhi vs. Raj Narain, Supreme Court of India, 1975 AIR 1590, 1975 SCC (2) 159, https://indiankanoon.org/doc/1240174/ (accessed 5 October 2018).
King, R. D., 1997. Nehru and the Language Politics of India. Oxford University Press, New Delhi.
Kossler, K., 2015. Conclusions: Beyond the Illusion of Ethno-Culturally Homogenous Territory. In T. Malloy & F. Palermo (eds.) Minority Accommodation Through Territorial and Non-Territorial Autonomy. Oxford University Press, Oxford, 245-272.
Kossler, K., 2016. Beyond Majoritarian Autonomy? Legislative and Executive Power-Sharing in European Regions. In M. Nicolini, F. Palermo & E. Milano (eds.) Law, Territory and Conflict Resolution. Brill Nijhoff, Leiden, Boston, 39-66.
Kossler, K., 2018. Hegemonic or Shared Autonomy? Two Approaches and Their Implications for Constitutional Design. In A. G. Gagnon & M. Burgess (eds.) Revisiting Unity and Diversity in Federal Countries: Changing Concepts, Reform Proposals and new Institutional Realities. Brill Nijhoff, Leiden, Boston, 399-423.
Kymlicka, W., 1998. Is Federalism a Viable Alternative to Secession? In P. B. Lehning (ed.) Theories of Secession. Routledge, London.
Lipset, S.M., 1960. Political Man: The Social Bases of Politics. Doubleday, New York.
Mallory, J., 1984. The Structure of Canadian Government. Gage, Toronto.
Mancini, S., 2008. Rethinking the Boundaries of Democratic Secession: Liberalism, Nationalism and The Right of Minorities to Self-Determination. International Journal of Constitutional Law 6, 553-584.
McGarry, J. & O'Leary, B., 2007. Federation and Managing Nations. In M. Burgess & J. Pinder (eds.) Multinational federations. Routledge, London, New York, 180-211.
McGarry, J. & O'Leary, B., 2008. Consociation and its Critics: Northern Ireland after the Belfast Agreement. In S. Choudhry (ed.) Constitutional Design for Divided Societies: Integration or Accommodation? Oxford University Press, Oxford, 369-408.
McGarry, J. & O'Leary, B., 2009. Must Pluri-National Federations Fail? Ethnopolitics 8(1), 5-25.
RAZPRAVE IN GRADIVO REVIJA ZA NARODNOSTNA VPRAŠANJA 81 / 2018
K. KOSSLER Upravjanje razdejenih družb s teritorialno avtonomjo? Od (pre)velikih pričakovanj do ...
Murray, C. & Simeon, R., 2007. Recognition without Empowerment: Minorities in a Democratic South Africa. International Journal of Constitutional Law 5, 699-729.
Nordquist, K. 1998. Autonomy as a Conflict-Solving Mechanism: An Overview. In M. Suksi (ed.) Autonomy: Applications and Implications. Kluwer Law International, The Hague, 3-29.
O'Leary, B., 2001. An Iron Law of Nationalism and Federation? A (Neo-Diceyian) Theory of the Necessity of a Federal Staatsvolk, and of Consociational Rescue. Nations and Nationalism 7(3), 273-296.
OHR - Office of the High Representative, 2002a. Decision on Constitutional Amendments in the Federation of Bosnia and Herzegovina, 19 April 2002, http://www.ohr.int/?p=66761 (accessed 5 October 2018).
OHR - Office of the High Representative, 2002b. Decision on Constitutional Amendments in Republika Srpska, 19 April 2002, http://www.ohr.int/?p=66769 (accessed 5 October 2018).
Palermo, F. & Kossler, K., 2017. Comparative Federalism: Constitutional Arrangements and Case Law. Hart Publishing, Oxford.
Palermo, F. & Sabanadze, N. (eds.), 2011. National Minorities in Inter-State Relations. Brill Nijhoff, Leiden, Boston.
Palermo, F., 2012. Central, Eastern and South-Eastern Europe and Territorial Autonomy: Are They Really Incompatible? In A. G. Gagnon & M. Keating (eds.) Political Autonomy and Divided Societies: Comparative TerritorialPolitics. Palgrave Macmillan, Basingstoke, 81-97.
Palermo, F., 2015. Owned or Shared? Territorial Autonomy in the Minority Discourse. In T. Malloy & F. Palermo (eds.) Minority Accommodation Through Territorial and Non-Territorial Autonomy. Oxford University Press, Oxford.
Post, R., 2003. Law and Cultural Conflict. Chicago-Kent Law Review 78(2), 485-511.
Renner, K., 2005. State and Nation. In E. Nimni (ed.) National-Cultural Autonomy and its Contemporary Critics. Routledge, London, New York, 15-48.
Riker, W. H., 1964. Federalism: Origin, Operation, Significance. Little, Brown & Company, Boston.
Riker, W. H., 1969. Six Books in Search of a Subject or Does Federalism Exist and Does it Matter? Comparative Politics 2(1), 135-146.
Roeder, P. G., 1991. Soviet Federalism and Ethnic Mobilization. World Politics 43(2), 196-232.
Roeder, P. G., 2009. Ethnofederalism and the Mismanagement of Conflicting Nationalisms. Regional & Federal Studies 19(2), 203-219.
S. R. Bommai vs. Union of India, Supreme Court of India, 1994 AIR 1918, 1994 SCC (3) 1, https://indiankanoon.org/doc/60799/ (accessed 5 October 2018).
Shiva Rao, B., 1968. The Framing of India's Constitution, Vol. 5. Indian Institute of Public Administration, New Delhi.
Simeon, R. & Conway, D.-P., 2001. Federalism and the Management of Conflict in Multinational Societies. In A. G. Gagnon & J. Tully (eds.) Multinational Democracies. Cambridge University Press, Cambridge, 338-365.
Simeon, R., 2009. Constitutional Design and Change in Federal Systems: Issues and Questions. Publius: The Journal of Federalism 39(2), 241-261.
Singh, M. P. & Kukreja, V., 2014. Federalism in South Asia. Routledge, New Delhi.
Snyder, J., 2000. From Voting to Violence: Democratization and Nationalist Conflict. Norton, New York.
Stein, E., 1997. Czecho/Slovakia: Ethnic Conflict, Constitutional Fissure, Negotiated Breakup. University of Michigan Press, Ann Arbor.
81 / 2018 TREATISES AND DOCUMENTS JOURNAL OF ETHNIC STUDIES
K. KÖSSLER Governing Divided Societies through Territorial Autonomy? From (too) Great Expectations
Stepan, A., 1999. Federalism and Democracy: Beyond the US Model. Journal of Democracy 10(4), 19-34.
Sunstein, C. R., 1991. Constitutionalism and Secession. University of Chicago Law Review 40	58(2), 633-670.
Supreme Court of Canada, 1981. Re: Resolution to amend the Constitution, [1981] 1 SCR 753, https://scc-csc.lexum.com/scc-csc/scc-csc/en/item/2519/index.do?r=AAAAAQ_ AKcGF0cmlhdGlvbgAAAAAAAAE (accessed 5 October 2018).
Supreme Court of Canada, 1998. Reference re Secession of Quebec, [1998] 2 SCR 217, https://scc-csc.lexum.com/scc-csc/scc-csc/en/item/1643/index.do (accessed 5 October 2018).
Watts, R., 1966. New Federations: Experiments in the Commonwealth. Clarendon Press, Oxford.
Watts, R., 2007. Multinational Federations in Comparative Perspective. In M. Burgess & J. Pinder (eds.) Multinational Federations. Routledge, London, New York.
Watts, R., 2008. Comparing Federal Systems. McGill-Queen's University Press, Montreal, Kingston.
Woelk, J., 2012. Bosnia-Herzegovina: Trying to Build a Federal State on Paradoxes. In M. Burgess & G. A. Tarr (eds.) Constitutional Dynamics in Federal Systems: Sub-National Perspectives. McGill-Queen's University Press, Montreal, Kingston, 109-139.
Woodman, S. & Ghai, Y., 2013. Comparative Perspectives on Institutional Frameworks for Autonomy. In Y. Ghai & S. Woodman (eds.) Practising Self-Government: A Comparative Study of Autonomous Regions. Cambridge University Press, Cambridge, 449-486.
Notes
1	The term divided societies refers to the many cases, in which ethno-cultural diversity is not only a pure fact, but has actual political salience because it forms the basis of political identity and mobilization (Choudhry 2008a, 4—5).
2	On federalism and autonomy, see Palermo and Kössler (2017, 58-61).
3	Non-territorial autonomy in the context ofAustromarxism is mostly associated with Karl Renner and his model of a Nationalitätenbundesstaat, i.e. a federation of nationalities. See Renner (2005, 15-47).
4	For an introduction, see McGarry and O'Leary (2007, 180-211).
5	Kymlicka contrasts them with "regional-based unity."
6	With reference to the Soviet Union, see Brubaker (1996, 9).
7	Apart from such obvious indicators of failure as state disintegration, factors that may justify regarding a federal system as dysfunctional include circumstances such as internal instability, violence, chaos and stalemate. See Simeon and Conway (2001).
8	For three main abstract approaches to the ethnic-territorial link, see Palermo (2015, 14-19).
9	For a comparison ofEuropean examples, see Kössler (2016).
10	The case of Switzerland is interesting because of the variety of cantonal arrangements regarding the official position ofreligion. Article 72(1) ofthe 1999 Swiss Constitution makes the regulation of this matter a prerogative of the cantons, and these indeed use this autonomy to follow very different models. While there is, in line with the French tradition, no officially recognized cantonal church (Landeskirche) in Geneva and Neuchätel, other cantons grant this special status to one or more religious communities. These are mostly the Roman Catholic Church and the Swiss Reformed Church, as well as, in some cases, the Old Catholic Church and Jewish congregations.
RAZPRAVE IN GRADIVO REVIJA ZA NARODNOSTNA VPRAŠANJA 81 / 2018
K. KOSSLER Upravjanje razdejenih družb s teritorialno avtonomjo? Od (pre)velikih pričakovanj do ...
12 For an early analysis ofpost-colonial federations, see Watts (1966).
11 See Horowitz (1985, 613-621). For an earlier exposition of this argument, see Lipset (1960, 91-92).
ea
13	After the government of Indira Gandhi declared a state ofemergency based on Article 352 of the Constitution (2007), the court invalidated the Thirty-Ninth Amendment, which was supposed to prevent her prosecution (Indira Nehru Gandhi vs. Raj Narain). Nearly 20 years later, the judges countered the misuse ofArticle 356, especially in the 1970s and 1980s, as a tool for the national government to dismiss state governments controlled by opposition parties (S. R. Bommai vs. Union oflndia).
14	Re: Resolution to amend the Constitution, [1981] SCR 753, 802. None of the two powers has been invoked since 1943. See Mallory (1984, 371).
15	For an overview ofmany ofthese conditions, see McGarry and O'Leary (2009, 19).
16	Hale (2004, 166) defines a "core ethnic region" as one possessing at least 50 per cent ofthe entire federation's population or 20 per cent more than the second largest region.
17	While Alfred Stepan's term "putting-together" (1999, 19) was supposed to capture the unique character of federal states that were formed through coercion, Nancy Bermeo's similar term "forced-together" (2002, 108) aims to place particular emphasis on the influence of external actors and the element ofsystemic frailty.
18	See, for example, the "constituent peoples" case, Partial Decision U 5/98 III of 1 July 2000. The Constitutional Court is composed oftwo Bosniak, two Croat and two Serb judges, as well as three international judges.
19
The PIC is an international forum of 55 states involved in the peace process in Bosnia and Herzegovina.
See, for instance, Article 143 of the Belgian Constitution: "In the exercise of their respective responsibilities, the federal State, the Communities, the Regions and the Joint Community Commission act with respect for federal loyalty, in order to prevent conflicts of interest." By contrast, in Spain it was the Constitutional Court that recognized (STC 18/1982; STC 11/1986) duties of cooperation and loyalty to the constitution (fidelidad a la constitución). On these and similar principles, see Palermo and Kossler (2017, 249-53).
"The clear repudiation by the people of Quebec of the existing constitutional order would confer legitimacy on demands for secession, and place an obligation on the other provinces and the federal government to acknowledge and respect that expression of democratic will by entering into negotiations and conducting them in accordance with the underlying constitutional principles already discussed." (Reference re Secession of Quebec, [1998] 2 SCR 217, paragraph
41
21