Transitional Justice, Federalism, and the Accommodation ofMinority Nationalism

Will Kymlicka

Revised Draft, May 2009

1. Introduction:

The dramatic growth of the field of transitional justice(hereafter TJ) since the late 1980s is connected to the ideathat TJ can serve as an instrument to consolidate democratictransitions. This instrumental justification is not the onlyrationale for adopting TJ. One can make a plausiblenormative argument that there are moral duties to expose thetruth about gross human rights violations, to prosecutetheir perpetrators, and to compensate their victims, even ifsuch programs have little or no effect on the largerprocesses of democratic transition. But the internationalpopularity of TJ is undoubtedly connected to the idea thatit also serves as an instrument of democratization, whichhas increasingly become the accepted goal of internationalorganizations.1

There are many ways in which TJ has been said to helpconsolidate democracy, but one familiar claim is that insocieties that have been scarred by ethnic animosity orreligious intolerance, TJ can help to reshape identities,and in particular, to weaken those aspects of people’sidentities that were the source of violence and conflict,and to replace them with a strengthened sense of sharedidentity related to common membership in the nationalpolitical community. For example, the ethnic identities of“Hutu” and “Tutsi” had become interpreted in Rwanda asinherently antagonistic and monolithic, and hence were1 See Paige Arthur, “How ‘Transitions’ Re-shaped Human Rights: A Conceptual History of Transitional Justice,” HumanRights Quarterly Vol. 31/2 (2009): 321–67, on the relationship between the principled/normative arguments for TJ, and the instrumental/causal arguments.

1

capable of being mobilized for acts of genocide. Processesof TJ, it is said, can reduce the salience of these divisiveidentities, depoliticizing them, and replacing them with ashared inclusive national identity as “Rwandan” based onequal citizenship in the polity. Similarly, the SouthAfrican TRC was often defended as a way of weakening thesalience of inherited racial identities, and of building anew inclusive pan-racial sense of South African `rainbownationhood’.2

This is often described as the `nation-building’ dimensionof TJ – it can be a vehicle for weakening inherited identitydivisions, and for building new national identities.Countries will only become peaceful democracies if allcitizens come to share a sense of identification with, co-ownership of, and perhaps even pride in, their politicalcommunity. This is difficult to achieve in contexts wheremembers of one group have traditionally viewed the state asalien and oppressive, dominated by another identity group.Nation-building forms of TJ, it is widely believed, can helpovercome this legacy, and enable all people to feel thatthey have a rightful place in the new national imaginary.

Of course, TJ can only serve this nation-building functionif the national identity is indeed an inclusive one – if themembers of different ethnic groups can see a place forthemselves within this new national identity. If not,nation-building can simply exacerbate the problem. This hasbeen recognized by practitioners of TJ. For example, whilethe Guatemalan truth commission (Commission for HistoricalClarification or CEH) firmly insisted that truth-telling andreparations can play a nation-building role in helping torelegitimize the Guatemalan state as the locus of people’sloyalties, it acknowledged that this will only work if thestate repudiates earlier `monocultural’ ideas of Guatemalan2 Brandon Hamber & Hugo van der Merwe, “What is this thing called Reconciliation?”, (Centre for the Study of Violence and Reconciliation, South Africa, 1998), available at: http://www.csvr.org.za/wits/articles/artrcbh.htm

2

nationhood, based almost exclusively on Ladino identity.Such exclusionary ideas of nationhood, it argued, havehistorically denigrated the value of indigenous peoples, andrendered them publicly invisible, and thereby helped tocreate the conditions under which severe human rights abusesbecame more likely.3 Nation-building of this sort has beenthe cause of historic injustice toward indigenous peoples,not the solution to it. To avoid this danger, Guatemala’sreconciliation framework insists that it is promoting adifferent sort of nation-building, one that is‘pluricultural’ rather than monocultural, and henceaccommodating of indigenous difference.4 According to theCEH, TJ can be used to build a new sense of Guatemalannationhood that transcends and weakens the division betweenLadinos and indigenous peoples, but only if this sense ofnationhood is explicitly multicultural.5

The Guatemalan case stands in interesting contrast with theRwandan and South African cases, where nation-building TJwas used to delegitimize inherited ethnic and racialidentities. In the CEH’s model of multicultural nationhood,explicit public recognition must be given to indigenousethnic identities (eg., in the form of language rights,customary law, legal protections of Mayan sacred sites ordress). Indigenous peoples must be included in the nation,not just as undifferentiated individual Guatemalan citizens,3 The Peruvian CRV made a similar argument – ie., that the centuries of denigration and invisibility of indigenous peoples made human rights abuses more likely during the conflict between the military and Shining Path. 4 Commission for Historical Clarification, Memory of Silence(Guatemala, 2000), available at: http://shr.aaas.org/guatemala/ceh/report/english/toch.html5 For the appeal to multicultural nation-building in Guatemala’s CEH, see Jeff Corntassel and Cindy Holder, ‘Who’s Sorry Now? Government Apologies, Truth Commissions and Indigenous Self-Determination in Australia, Canada, Guatemala, and Peru’, Human Rights Review 9/4 (2008): 465-89; and the chapter by Rubio-Marin et al in this volume.

3

but also collectively, as a constituent component of thenation. The Rwandan government, by contrast, did not giveexplicit recognition to its constituent ethnic groups –rather it attempted to suppress public discussion of ethniclabels, which it dismissed as inauthentic colonialinventions used for divide and rule purposes. Its aim hasbeen to “extinguish” ethnicity, not simply to build newnational identities that can encompass diverse ethnicidentities.6 Similarly, the ANC often treats racialidentities as the artificial product of apartheid which canand should fade in the new democracy.

This contrast between the Guatemalan and Rwandan/SouthAfrican approaches points to an important choice facing allattempts to use TJ for nation-building purposes. Bothapproaches can be seen as pursuing the project of“citizenization” – that is, they both seek to replaceearlier uncivil relations of distrust, animosity andintolerance with relations of democratic citizenship.7 Andthey both therefore seek to weaken the divisive nature ofidentity conflicts – that is, they both seek to reduce thelikelihood that identity cleavages will be mobilized in waysthat are violent or destabilizing. But they differ in howto achieve this. For many commentators, the best way toavoid the destabilizing political mobilization of ethnicityis to reduce the political mobilization of ethnicity toutcourt – that is, to depoliticize ethnicity. This is theroute chosen in Rwanda and South Africa, and has beenpursued in other countries as well, often reflected inprohibitions on ethnic political associations or parties, orprohibitions on the voicing of particular types of ethnicdemands, or on the public discussion of `sensitive’political topics.

6 See the chapters by Chapman and Fullard and Rousseau in this volume.7 On “citizenization”, see James Tully, "Introduction", in Alain Gagnon and James Tully (eds) Multinational Democracies(Cambridge University Press, Cambridge, 2001), p. 25.

4

But this is not the only option. Another option, reflectedin the Guatemalan case, is to acknowledge and legitimize thepolitical mobilization of ethnicity, but to channel thismobilization in peaceful and democratic ways. On thisapproach, the best way to avoid destabilizing ethnicmobilization is precisely to stabilize it – ie., toregularize and normalize ethnic politics, to treat it as anaccepted and everyday part of democratic political life,subject to the same rules and conditions as any other formof democratic participation and claims-making. This caninvolve many different forms of public recognition ofethnocultural difference, including both substantivecultural rights (eg., language rights, recognition of `usesand customs’, accommodation of religious practices) andprocedural mechanisms of consultation, representation andparticipation (eg., advisory councils, guaranteedparliamentary representation, various forms of self-government). These policies do not reduce the politicalsalience of ethnicity, but seek to reduce the potential forviolence or instability associated with identity cleavages,by normalizing ethnic politics within a larger democraticconstitutional order that all citizens can identify with andendorse, and that compels all political actors to respectconstraints of human rights and democratic accountability.

It is an interesting and important question – beyond thescope of this chapter – to determine which of these twoapproaches to citizenization is more appropriate under whichconditions. Each has obvious risks. The first strategy ofde-ethnicization carries the risk that the `national’identity which in theory is supposed to transcend ethnicdifferences will in practice be strongly associated withjust one group. For example, while the Rwandan TJ mechanismsexplicitly avoid ethnic categories, the putatively non-ethnic categories (eg., of victim or perpetrator) havebecome strongly ethnically coded.8 More generally, the ideathat there can be a purely ‘civic’ form of nationhood,8 This dynamic in Rwanda is discussed in the chapters by Chapman, Aptel and Fullard/Rousseau in this volume.

5

neutral amongst the languages, identities and practices ofthe various ethnic groups in the state, has been stronglydisputed. Claims to the construction of a civic or post-ethnic form of nationhood often simply mask the ways inwhich the language and culture of the dominant groupcontinue to shape the national identity. Moreover, there issomething paradoxical about pursuing democraticconsolidation by prohibiting the democratic mobilization ofethnic groups. Any country that fears the exercise ofdemocratic freedoms by its ethnic minorities is likely todevelop a stunted form of democracy. As Istvan Bibo put it,referring to the way Central European countries in theinter-war period attempted to limit the politicalmobilization of their minorities:

In a paralyzing state of fear which asserts thatfreedom's progress endangers the interests of thenation, one cannot take full advantage of the benefitsoffered by democracy. Being a democrat means,primarily, not to be afraid: not to be afraid of thosewho have different opinions, speak different languages,or belong to other races… The countries of Central andEastern Europe were afraid because they were not fullydeveloped mature democracies, and they could not becomefully developed mature democracies because they wereafraid.9

However, the option of normalizing ethnic politics through amulticultural conception of citizenization also carriesrisks. In some circumstances, there is simply no way toguarantee that ethnic politics will stay peaceful anddemocratic. No matter what `paper guarantees’ exist, therisk that ethnic politics will degenerate into violence maysimply be too high. (This presumably was part of therationale in Rwanda). But even if one is confident that9 Istvan Bibo, “The Distress of East European Small States”,in Karoly Nagy (ed) Democracy, Revolution, Self-Determination (Social Science Monographs, Boulder, 1991), p.42.

6

ethnic politics will remain peaceful and democratic, one mayworry that multicultural models of TJ will not achieve their`nation-building’ or `community-building’ goals. They maylead to `parallel societies’ – groups living side-by-side inrelative peace, but unable or unwilling to work together,not feeling any sense of shared purpose or solidarity. As deGreiff notes, this picture of mere co-existence doesn’t liveup to our ideals of “reconciliation”.10 And we may alsoworry that institutionalizing ethnic politics can havedetrimental effects within ethnic groups, leading to thereification, essentializing or freezing of ethnicidentities, as self-proclaimed ethnic leaders attempt toimpose their own views of the group’s `authentic’ culture oridentity on group members, particularly where these leadersare not properly constrained by norms of democraticaccountability.

Much has been written on the pros and cons of these modelsof citizenship, as defenders and critics debate the severityof these risks.11 I cannot hope to resolve those debateshere. For the purposes of this chapter, I want to focus onone particular context where this issue arises: namely, incases where a country undergoing a democratic transitioncontains a strongly-mobilized minority nationalist movementseeking some form of self-government on a territorial basis,either through federalization or even independence. Considerthe case of the Catalans during Spain’s democratictransition in the 1970s, or the case of the Kurds in Iraqtoday, or the Acehnese in Indonesia.

These cases raise some acute dilemmas about the relationshipbetween transitional justice, citizenization, and nation-building. In the rest of the chapter, I would like to10 See the discussion of the modus vivendi amongst Macedonians and Albanians in Pablo de Greiff, “Theorizing Transitional Justice” (International Center for TransitionalJustice, 2008).11 For a helpful overview of these debates, see Paige Arthur’s chapter in this volume.

7

outline these dilemmas, and suggest avenues for futureresearch to help address them. I will begin by describingone familiar model of accommodating substate nationalism –namely, multination federalism – and how this model cangenerate a distinctive, but potentially successful, form ofcitizenization. I will then explore some of the risksassociated with adopting this model under conditions oftransition, and conclude with some reflections on theimplications for the practice of transitional justice.

2. Democratic Multination Federalism

While all countries contain a plurality of identities andidentity cleavages, the phenomenon of territorially-concentrated minority nationalism is a very distinctive sortof challenge, and we need to understand its specificity.12

In these cases, the members of a regionally-concentratedgroup exhibit a nationalist consciousness – that is, theyconceive of themselves as forming a `nation’ within a largerstate, and mobilize behind nationalist political movementsor parties to achieve recognition of their nationhood,either in the form of an independent state or throughterritorial autonomy within the larger state. To be sure,the depth of this nationalist consciousness varies frommember to member and over time, and the leaders ofnationalist movements often exaggerate the depth of supportthey have amongst group members. Yet we also know that, whenthese nationalist parties are able to compete in free andfair elections, they often do gain the support of theplurality or majority of the group on a consistent basis12 For the purposes of this paper, I am focusing on substate nationalist minorities (like the Catalans, Quebecois, Kurds and Acehnese) rather than indigenous peoples (like Aboriginals in Canada, Sami in Norway, or Maori in New Zealand), in part because the indigenous case is discussed in other chapters in this volume. However, as I note below, there are some important similarities between the two categories (and indeed there are ongoing scholarly and legaldebates about whether or how to distinguish them).

8

(eg., in Flanders, Quebec, Catalonia, South Tyrol etc). Sowe should neither exaggerate nor underestimate the depth ofnationalist consciousness.

In such cases, we can say that the state is, sociologically-speaking, a `multi-nation’ state, although states themselveshave been reluctant to acknowledge this. The presence ofsuch minority nationalist movements has traditionally beenseen as a threat to the state, putting into question itslegitimate right to rule all of its territory andpopulation. In a world of `nation-states’ where stateslegitimate themselves by reference to norms of the self-determination of peoples, to have a group which claims thatit forms a distinct `people’, with its own rights to self-determination, is often perceived as a serious threat.Indeed, many commentators have argued that democraticstability is impossible in such cases. Democracy is rule by`the people’, but this requires an agreement that thecitizens of a state do indeed form a single `people’ whoexercise their popular sovereignty through a common statewithin agreed borders. Without agreement on the unit ofself-determination, democracy becomes unstable.13

As a result, many states in the past have adopted a widerange of measures to erode any sense of distinct nationhoodamongst their territorially-concentrated minorities, forexample by restricting minority language rights, abolishingtraditional forms of regional self-government, andencouraging members of the dominant group to settle in theminority group's homeland in an effort to outnumber theminority even in its traditional territory. Where theseassimilationist policies failed, as they often have, moreradical solutions have been considered, such as partition –breaking up multination states into two or more states thatare more nationally homogenous - and/or population transfer,13 For a recent version of this argument, see Claus Offe, "Political Liberalism, Group Rights and the Politics of Fearand Trust", Studies in East European Thought 53 (2001): 167-82.

9

to encourage members of minorities to move to a neighbouringkin-state. Indeed, international organizations themselveshave sometimes endorsed these solutions (eg., after the twoworld wars), and some academic commentators supportpartition or population transfer as the more realisticalternative to naïve hopes for stabilizing multi-nationstates.14

While the challenges of minority nationalism are severe,they are by no means rare. On the contrary, as Walker Connornotes, the phenomenon of minority nationalism is a trulyuniversal one. The countries affected by it

are to be found in Africa (for example, Ethiopia), Asia(Sri Lanka), Eastern Europe (Romania), Western Europe(France), North America (Guatemala), South America(Guyana), and Oceania (New Zealand). The list includescountries that are old (United Kingdom) as well as new(Bangladesh), large (Indonesia) as well as small(Fiji), rich (Canada) as well as poor (Pakistan),authoritarian (Sudan) as well as democratic (Belgium),Marxist-Leninist (China) as well as militantly anti-Marxist (Turkey). The list also includes countrieswhich are Buddhist (Burma), Christian (Spain), Moslem(Iran), Hindu (India), and Judaic (Israel).15

Given its pervasiveness, we can reasonably expect that moreand more cases of transitional justice will arise incountries that confront the challenge of minoritynationalism.

14 See, for example, Chaim Kaufmann, “When All Else Fails: Ethnic Population Transfers and Partitions in the Twentieth Century”, International Security 23/2 (1998): 120-56.15 Walker Connor, "National Self-Determination and Tomorrow's Political Map", in Alan Cairns et al (eds) Citizenship, Diversity and Pluralism: Canadian and Comparative Perspectives (McGill-Queen's University Press, Montreal, 1999), 163-4.

10

Given this background, it’s important to determine whetherthere are in fact ways of accommodating minority nationalismthat are consistent with democratic stability. Are thereways of constructing relations of democratic citizenship insuch countries? In my view, we have growing evidence for thesuccess, at least under favourable conditions, of one modelof accommodation, which I will call "multination andmultilingual federalism". This model has two key features:(a) it involves creating a federal or quasi-federal subunitin which the minority group forms a local majority, and canthereby exercise meaningful forms of self-government; (b)the group's language is typically recognized as an officialstate language, at least within their subunit, and perhapsthroughout the country as a whole.16

At the beginning of the twentieth-century, only Switzerlandand Canada had adopted this combination of territorial16 In Spain, Belgium, Canada and Switzerland, territorial autonomy for national minorities was achieved through a federalizing of the state, so as to create a federal subunitthat was dominated by the national minority. The United Kingdom, by contrast, did not federalize the entire country,but created quasi-federal forms of territorial autonomy for Scotland and Wales. A similar quasi-federal autonomy regime exists for the Swedish minority in Finland, the German minority in South Tyrol, and for Puerto Rico in the United States. I use the term ‘multination federalism’ to cover both federal and quasi-federal forms of autonomy. It is important to distinguish such "multination" federations fromother federal systems where internal subunits are not designed to enable minority self-government, such as the continental United States, Germany, Australia, and Brazil. In these countries, none of the subunits was designed to enable a national minority to exercise self-government over its traditional territory. For more on the difference between multination federalism and other forms of federalism, see my Politics in the Vernacular: Nationalism, Multiculturalism, Citizenship (Oxford University Press, Oxford, 2001), chap. 5.

11

autonomy and official language status for substate nationalgroups. Since then, however, virtually all Westerndemocracies that contain sizeable substate nationalistmovements have moved in this direction. The list includesthe adoption of autonomy for the Swedish-speaking AlandIslands in Finland after the First World War, autonomy forSouth Tyrol and Puerto Rico after the Second World War,federal autonomy for Catalonia and the Basque Country inSpain in the 1970s, for Flanders in the 1980s, and mostrecently devolution for Scotland and Wales in the 1990s.

In each case, the adoption of territorial autonomy fornationalist groups was initially controversial, and viewedas potentially destabilizing. But I think we now havesufficient experience to say that this model ofaccommodation can indeed be a vehicle for citizenization, atleast in the Western cases mentioned above. Where it hasbeen adopted in the West, the historic relations ofanimosity and exclusion between states and nationalistminorities have been reduced and replaced with relations ofdemocratic citizenship.

To prove this conclusively would require detailedmeasurements of the quality or resilience of relations ofdemocratic citizenship, but for the purposes of thischapter, let me just suggest four relevant indicators of“citizenization”:

1. Peace and individual security: the multination federationsreferred to above are managing to deal with theircompeting national identities and nationalist projectswith an almost complete absence of violence or terrorismby either the state or the minority.17

17 The Basque Country is the main exception, although the ETAcampaign of violence began in the 1960s and 1970s as aresponse to the highly-centralized Fascist regime, and isunlikely to have emerged had Spain been a democraticmultination federation. Similarly, in Cyprus violence arosein the 1960s in response to attempts by the state to subvert

12

2. Democracy: ethnic politics is now a matter of ‘ballotsnot bullets’, operating under normal democraticprocedures, with no threat of military coups orauthoritarian regimes which take power in the name ofnational security.18

3. Individual freedom: these reforms have been achievedwithin the framework of liberal constitutionalism, withfirm respect for individual civil and political rights,including freedom of speech, conscience, politicaldissent, and increasingly gay rights, gender equalityand so on. Since substate governments are subject to thesame constitutional constraints as the centralgovernment, they have no capacity to restrict theseindividual freedoms in the name of maintaining culturalauthenticity or cultural purity, and in any event theevidence suggests they have no wish to do so.19 4. Inter-group equality: multination federalism has promotedequality between majority and minority groups, where“equality” is understood here as non-domination, suchthat one group is not systematically vulnerable to thedomination of another group. Equality in this sense is

the multination federal arrangements adopted at independence.It was the abrogation, not the adoption, of multinationfederalism that generated violence.18 Saul Newman, Ethnoregional conflict in democracies: mostlyballots, rarely bullets (Greenwood Press, Westport, 1996).19 Indeed, minority subunits are sometimes more liberal thanthe central government. Policies on gender equality or gay rights, for example, are more progressive in Scotland than the rest of Britain; more progressive in Quebec than in other parts of Canada; and more progressive in Catalonia than other parts of Spain. Moreover, support for cosmopolitan values is also typically higher in these substate regions than in other parts of the country, as is support for strengthening the role of the European Court of Human Rights, or other international human rights instruments. For some of the evidence, see my Multicultural Odysseys: Navigating the New International Politics of Diversity (Oxford University Press, 2007).

13

multidimensional, and can be assessed along economic,political and cultural lines. Multination federalism hashelped create (i) greater economic equality betweenmajority and minority; (ii) greater equality ofpolitical participation and influence, so thatminorities are not continually outvoted on all issues;and (iii) greater equality of cultural respect andrecognition, reflected for example in more equitablerecognition of the minority’s language and culture inpublic space, and in reduced levels of prejudice anddiscrimination between groups. In short, the historicpatterns of economic disadvantage, politicalsubordination and cultural marginalization that havecharacterized the status of these national minoritieshave been reduced.

As measured by these criteria, Western cases of multinationfederalism surely qualify as successful examples of creatingrelations of democratic citizenship in states with minoritynationalism.20 This is confirmed, I think, by the fact thatthere has been virtually no significant movement in any ofthe Western democracies to roll back either the territorial20 For more detailed evaluations of these new models of multination federal states, and the sort of democratic citizenship they support, see Alain Gagnon and James Tully (eds) (2001) Multinational Democracies (Cambridge UniversityPress, Cambridge, 2001); Scott Greer, Nationalism and Self-Government : The Politics of Autonomy in Scotland and Catalonia (SUNY Press, 2007); Monserrat Guibernau, Nations Without States: Political Communities in a Global Age (Polity Press, Cambridge, 1999); Siobhan Harty and Michael Murphy, In Defense of Multinational Citizenship (UBC Press, Vancouver, 2005); Wayne Norman, Negotiating Nationalism: Nation-Building, Federalism and Secession in the Multinational States (Oxford University Press, Oxford, 2006); Ferran Requejo Multinational Federalism and Value Pluralism: The Spanish Case (Routledge, London, 2005); Stephen Tierney, Constitutional Law and National Pluralism (Oxford University Press, Oxford, 2004).

14

autonomy or the official language rights granted to substatenationalist minorities. If this experiment in accommodatingsubstate nationalism had been a failure or a threat todemocratic stability, we surely would have seen somebacklash and retreat from it in those countries that haveadopted it. Instead, we see the reverse trend: it has notonly become more securely entrenched in those countrieswhere it has historically been adopted, but is alsospreading to new countries. The success of this model isalso confirmed by the repeated statements of variousinternational organizations - such as the UN, Council ofEurope, and OSCE – that multinational federalism andterritorial autonomy are `best practices’ for the democraticinclusion of national minorities.21

To be sure, like any form of multicultural citizenship,multination federalism poses risks. In particular,multination federalism carries the risk of creating`parallel societies’ that merely co-exist, rather thanleading to a stronger sense of community and solidarity.Indeed, we might say that the whole purpose of multinationfederalism is to create parallel societies – ie., to enablenational minorities to sustain themselves as distinct andinstitutionally complete societies within the larger state.Yet the evidence suggests that the institutionalization ofsuch parallel societies has not undermined solidarity, atleast as reflected in the ability to cooperate in theproduction of public goods and redistribution. Countriesthat have adopted multination federalism have fared no worsethan others in building or sustaining their welfarestates.22

21 For examples of this endorsement by international organizations, see my Multicultural Odysseys, chapter 6.22 See Keith Banting and Will Kymlicka (eds) Multiculturalism and the Welfare State: Recognition and Redistribution in Contemporary Democracies (Oxford University Press, Oxford, 2006), for a review of this evidence.

15

So I would argue that the model of multilingual, multinationfederalism emerging in the West is a promising example ofhow to accommodate minority nationalism in a way thatdeepens relations of liberal-democratic citizenship,reducing inter-group hierarchies while protecting individualfreedom. And yet this is a very distinctive form ofcitizenization, which requires us to rethink our standardideas of democracy and political community, including theideas of citizenship that have informed existing models ofTJ. As I noted earlier, traditional conceptions of democracyhave presupposed that the state is the embodiment of asingle `nation’ or `people’ which is the bearer of rights ofself-determination exercised through popular sovereignty.Democracy requires a consensus that citizens form a singularsovereign people. This presupposition has been implicit inmost TJ processes as well, even in the `multicultural’conception of nationhood endorsed by the Guatemalan CEH.While it emphasized that the Guatemalan people were diverseand multicultural, and that these differences needed to bepublicly acknowledged and respected, it nonethelesspresupposed that rights of sovereignty and self-determination rest with the Guatemalan people as a singularwhole, not with any ethnic groups inside the state. Indeed,it said that one goal of `reconciliation’ was precisely torelegitimize the Guatemalan state as the bearer of popularsovereignty and self-determination. Similar assumptionsunderlie accounts of the `nation-building’ function of TJ inother countries. Indeed, we could say that one aim of TJ isprecisely to resolve and foreclose debates about thelegitimate right of the state to rule over all of thecitizens and territory of the state. The success of TJ canbe measured, in part, by whether this legitimacy is nolonger disputed.

Multination federations, however, require a different model.Minority nationalist movements are characterized preciselyby their claims to peoplehood or nationhood, and theadoption of multination federalism reflects anacknowledgement of the need to accommodate this nationalist

16

aspiration and identity.23 Territorial autonomy bothacknowledges this sense of minority nationhood, and providesthe institutional means to reproduce and consolidate it(eg., by enhancing regional control over education, publicsymbols, public media). The result is to open up, ratherthan to foreclose, the question of the larger state’slegitimacy. Once a substate group is (implicitly orexplicitly) recognized as a distinct nation or people, thequestion arises as to how the state’s assertion of authorityover the group’s members and historic territory is to bejustified. Nationalist leaders within the minority may arguethat the larger state never acquired legitimate authority,since the incorporation of the minority into a larger statewas the result of conquest or coercion rather than voluntaryconsent. Or even if there had been an initial agreement ortreaty at some earlier point of history, the larger statemay have violated the terms of that agreement, stripping theminority of its historically agreed-upon rights. Or even inthe (null?) case of incorporation based on voluntaryagreements that have been scrupulously respected,nationalist leaders can still claim that the right to self-determination is not a one-shot affair in which onegeneration’s decisions bind all future generations inperpetuity, but rather is an ongoing right of each people to“freely determine their political status and freely pursuetheir economic, social and cultural development”.24

23 This acknowledgement as a ‘nation within’ can be either implicit or explicit. Compare the Spanish constitution’s distinction between the Spanish `nation’ and regional `nationalities’ with the Canadian parliamentary recognition of Quebec nationhood. But even if there is no formal recognition, the adoption of multination federalism rests onan acknowledgement of the sense of nationhood amongst members of the substate national minority, and hence of the need to enable the minority to act upon this self-identity. 24 See S. James Anaya, Indigenous Peoples in International Law (Oxford University Press, New York, 2004) on self-determination as an ongoing, not one-shot, right.

17

On the basis of these (and other) arguments, minoritynationalist leaders can contest the legitimacy of the largerstate’s claim to authority. Multination federalism providesthe conceptual vocabulary and political space for this sortof ongoing contestation, both within the minority andbetween the minority and the larger state. Indeed, itcreates the possibility that the substate government willinclude self-declared secessionists, as has happened inQuebec, Scotland and Catalonia, who openly argue for theredrawing of international boundaries. In short, multination federalism opens up and suspends,rather than resolves and forecloses, questions about thebases of legitimate authority. It creates the institutionalframework in which multiple and conflicting claims tolegitimate authority, popular sovereignty and national self-determination can be advanced and debated. Political actorsin a multination state will disagree about which politicalauthority is inherent or delegated (eg., whether the centralstate gains its authority by delegation from the consent ofthe sovereign constituent national groups, or converselywhether regional autonomies gain their authority bydelegation from the consent of the sovereign central state),and about the nature of the agreements that have founded thestate (eg., whether it is a `union’ state, a devolved state,a federation, a confederation etc). For some minoritynationalist actors, the larger state is a usurper of theinherent sovereignty of the substate nation; for others, thelarger state is a `union’ or `confederal’ state that arosefrom the voluntary pooling of the sovereignty of itsconstituent units; for yet others, regional autonomy issimply one form of devolution or decentralization which asovereign state can pursue. These sorts of disagreements arethe background to, and bread and butter of, ethnic politicsin many multination states.

This indeterminacy regarding the nature of sovereignpeoplehood in multination states may puzzle politicaltheorists who think that agreement on peoplehood is a

18

precondition of democracy. However, it appears that mostcitizens in multination Western states take thisindeterminacy and contestation in stride. As we’ve seen,this phenomenon of conflicting nationalist projects is notinherently unstable or violent. It appears that citizens ofmultination federations in the West are capable of managingwhatever conceptual ambiguities and political controversiesarise from the fact of overlapping claims to sovereignty andself-determination. Public opinion polls reveal a seeminglystable pattern of what David Miller calls “nestednationalisms”.25 For some purposes, in some contexts, mostScots view themselves as a distinct people with self-originating claims to sovereignty and self-determinationthat the British state must respect; but for other purposes,in other contexts, they view themselves as members of aBritish nation that also possesses self-originating claimsto sovereignty and self-determination. And we can even seethe emergence of yet a third possible layer, as the EuropeanUnion comes to be seen as the rightful bearer of rights ofpopular sovereignty on certain issues. Each is anappropriate locus of loyalty and of self-determination, andeach can legitimately adopt policies aimed at consolidatinga sense of “we, the people” amongst its members. Theseidentity-building projects can sometimes be contradictory:British politicians may promote a common British identity ina way that downplays the significance of regionalidentities; Scottish politicians may promote a sense ofScottish nationhood that views British authority as usurpedor derivative; and EU politicians may imply that bothBritish and Scottish national identities are anachronisms inan increasingly post-national European demos. And yetcitizens themselves seem able to reconcile these identities,loyalties and claims to legitimate political rule, and geton with the business of building (multi-level) politicalorders that are peaceful, democratic, and free.

25 See David Miller, Citizenship and National Identity (Polity Press, 2000).

19

In short, in the context of Western multination federations,citizenization is not about resolving disputes overlegitimacy, but about learning to live with their ambiguousand contested character, and building peaceful anddemocratic forums for continuing that conversation.

I have been focusing so far on cases of territorially-concentrated minority nationalist groups for whom some formof federal/regional autonomy is possible, but it is worthnoting that we can see a similar dynamic in NorthernIreland, where the 1998 Good Friday peace agreement adopteda non-territorial model for accommodating Irish-Catholicsubstate nationalism. Since Irish Catholics are notregionally concentrated within Northern Ireland,nationalists have not pursued regional autonomy, but rather(a) consociational power-sharing within Northern Ireland;and (b) trans-border linkages to the Republic of Ireland.From an institutional point of view, this non-territorialand trans-border model of accommodating substate nationalismis very different from multination federalism, but theunderlying conception of citizenization has remarkablesimilarities. While the Agreement is firmly committed toconstructing relations of democratic citizenship, itacknowledges that this cannot be achieved by resolvingissues of peoplehood and sovereignty. In recognition of thefact that many Irish Catholics do not see themselves as partof a British people,26 the Good Friday Agreement not onlyallows for a future referendum on unification with Ireland,but also explicitly states that: “The two Governmentsrecognise the birthright of all the people of NorthernIreland to identify themselves and be accepted as Irish orBritish, or both, as they may so choose, and accordinglyconfirm that their right to hold both British and Irishcitizenship is accepted by both Governments and would not beaffected by any future change in the status of Northern26 Only 8.5% of Irish Catholics acknowledge a British identity. See John Cockley, “National identity in Northern Ireland: stability or change?”, Nations and Nationalism 13/4(2007), 573–97.

20

Ireland” (Article 1.6). One could not get a clearerstatement of the principle that citizenization in amultination state requires acknowledging and suspending,rather than resolving and foreclosing, issues of thedefinition of peoplehood and sovereignty.

It is too early to tell whether the innovativeconsociational and trans-border features of the Good FridayAgreement will work, but so far at least, there are groundsfor cautious optimism. In any event, the Northern Irelandexperience seems to reflect the emerging consensus in theWest that new forms of citizenship are both possible andnecessary in multination states.

3. Federalization, Democratization and Transitional Justice

Given the positive experiences with multination federalismin the West, it’s not surprising that this model has oftenbeen recommended for democratizing countries around theworld that confront minority nationalisms. And indeed Ithink it is likely that an increasing number of TJ processeswill arise in countries in which federalization is acomponent of peace agreements or democratic transitions.Federalization or regional autonomy has been part of severalrecent peace agreements (eg., Iraq, Sudan, Indonesia,Bosnia) and many commentators expect it to be part ofpotential agreements in other countries in the future (eg.,Cyprus, Nepal, Burma, Sri Lanka). It is important to ask,therefore, whether or how TJ can contribute to theconsolidation of these distinctive relations of multi-nationcitizenship. We know what kinds of TJ processes have beenadopted to promote `nation-building’, but what kinds of TJprocesses would help promote `multi-nation building’?

Unfortunately, this question has rarely been studied in theliterature.27 So my discussion will inevitably be sketchy27 The literature on TJ in Bosnia touches on this question, since the adoption of TJ has been stymied by unresolved contestations around its ethno-federal structures (eg.,

21

and speculative. The first question is whether processes ofTJ provide an appropriate forum for even raising `thenational question’. One might think that TJ should simplyavoid the national question in multination states, and focusentirely on vetting and prosecuting perpetrators of humanrights abuses and on compensating their victims, withouttaking a stand on, or even attempting to understand,conflicting claims of self-determination and peoplehood. InIraq, for example, a process of TJ might focus onidentifying those guilty of crimes against humanity underSaddam Hussein’s regime, disconnected from any assumption oreven aspiration that this will help re-build a sense ofIraqi nationhood, or relegitimize the Iraqi state as thelocus of loyalty or popular sovereignty (or, for thatmatter, legitimize Kurdish nationalism). On this view, itwould unduly politicize TJ programs, particularly inmultination states, if they took it upon themselves to tryto engage in (multi)-nation-building.

Ivana Franović, Dealing with the Past in the Context of Ethnonationalism: The Case of Bosnia-Herzegovina, Croatia and Serbia (Berghof Occasional Paper Nr. 29, October 2008, Berghof Research Center for Constructive Conflict Management, Berlin); Iavor Rangelov and Marika Theros, Maintaining theProcess in Bosnia and Herzegovina : Coherence and Complementarity of EU Institutions and Civil Society in the Field of Transitional Justice (Working Group on Developmentand Peace, FriEnt, Bonn, 2007). However, this literature is generally hostile to the ethno-federal structure imposed by the Dayton Agreement, and so does not ask whether or how TJ can help to make that structure work. Instead, the literature asks whether or how TJ can be used to overcome Dayton, on the assumption that true democratization (and true TJ) requires depoliticizing ethnicity. According to Rangelov and Theros, for example, the Dayton Agreement “contributed to the continued politicization of ethnicity that undermines the process of transitional justice” (p. 3).On this view, we can have ethnic federalism or we can have TJ, but not both. I will return to this below.

22

However, this approach has obvious drawbacks. For one thing,it leaves unexplored the deeper cultural and politicalforces that made human rights violations possible in thefirst place. From the perspective of many Kurds, the humanrights violations they suffered were not just the result ofthe abusive exercise of state power by a militarydictatorship, but were also the result of long-standing anddeeply-rooted supremacist Arab nationalism. Without thelatter, the former would not have been possible. TheArabization campaigns and Anfal genocide were made possibleby the widespread support amongst the general Iraqipopulation for such ideologies. Just as the Guatemalan andPeruvian TRCs insisted that human rights violations by themilitary were made possible by centuries of denigration ofindigenous peoples, so too we might think that human rightsviolations in Iraq were made possible by long-standingassumptions that Iraq is an Arab state that belongsexclusively to the Arab people. If so, then simply vettingor prosecuting individual officials, without examining the`national question’, may fail to address the underlyingcauses of human rights violations, and hence fail to ensuretheir non-recurrence. If members of the Arab majority arenot asked to critically reflect on their assumptions ofnationhood, and the way it was complicit in human rightsviolations, then the conditions for future violations remainin place.28

Of course, Kurdish nationalist leaders may be exaggeratingthe significance of this factor in order to justify theirclaims to autonomy or independence. Commentators often note(and bemoan) the extent to which minority nationalists areinvested in a `victimization’ narrative that exaggeratestheir history of mistreatment, and ignores more complex28 See the chapters by Aptel and Wilke in this volume for discussion of the role of TJ in encouraging reflection on complicity by the larger society. As Arthur puts it, TJ needs to explore the sources of `ordinary violence’ against particular groups that make possible the outbreak of `extraordinary’ violence.

23

patterns of cooperation and conflict within and betweengroups.29 This, indeed, is precisely why it might beimportant for processes of TJ to tackle the nationalquestion, in order to provide an open and impartial forumfor airing these contesting national narratives andclaims.30

In any event, the goal of avoiding the national question isprobably unrealistic. Even the most basic logistical issues,such as the physical location of TJ processes, or thelanguage in which they operate, or the level of governmentthat authorizes them, will be inflected with ethnonationalconnotations.31 If processes of TJ are located in Baghdadand held in Arabic, they will inevitably be seen as effortsat relegitimizing the authority of the central state. Ifthey are located in Erbil and held in Kurdish, they willinevitably be seen as efforts at legitimizing Kurdishnationalism.32

Indeed, these are often the most controversial issuesregarding TJ in multination states. A recent UNDP survey inKosovo, for example, revealed a remarkable consensus acrossethnic lines on the basic principles of TJ, such as29 For the Yugoslav case, see the discussion in Franović, Dealing with the Past.30 For a similar argument, see Arthur’s chapter in this volume.31 On the significance of these choices about location and language, see Aptel’s chapter.32 The use of TJ to legitimize Kurdish nationalism has been widely-noted: “What happened at Halabja, and the international community’s muted response to it, has become arallying cry for those seeking an independent Kurdish state based in northern Iraq – an argument for a Kurdish state just as the Jewish Holocaust was part of the rationale for creating the state of Israel” (Mark Mackinnon, “The painful lesson of betrayal: Iraq 20 Years after the Halabja Massacre”, Globe and Mail (Toronto), March 17, 2008, p. A11).

24

uncovering the truth about missing persons and about warcrimes; the importance of reconciliation between ethnicgroups; and the need to vet police and judges. There wereprofound disagreements, however, about who should be doingthese things, and where. Whereas 72% of ethnic Serbsbelieved that the Serbian courts in Belgrade should be incharge of war crimes committed in Kosovo (therebyreaffirming Serbia’s claim to rule Kosovo), only 4% ofethnic Albanians supported this option, preferring eitherlocal Kosovo courts or the international ICTY (therebydisputing Serbia’s claim over Kosovo).33 Decisions about thelocation of TJ inevitably reflect contested ethnonationalclaims to sovereignty and territory.

We can see the outlines of similar debates over TJ inAceh.34 The 2005 peace agreement contained a provision for aprovincial-level TRC, which would undoubtedly be used as avehicle for legitimizing Acehnese nationalist struggles forautonomy (particularly since the ex-rebel Acehnesenationalist GAM party won the provincial elections). So thecentral government in Jakarta has attempted to re-establishits control by passing a law saying that the provincial TRCis an “inseparable part” of an anticipated national TRC,which will undoubtedly be used as a vehicle forrelegitimizing the Indonesian state. The result, to date,has been a standoff. Here again, the basic principles of TJare not in dispute, but the implementation of TJ has beencaught up in the broader disagreement over who exercisesappropriate authority or jurisdiction in Aceh.

Confronted with these duelling claims to jurisdiction overTJ, commentators often accuse the various actors of “playingpolitics” with TJ, and argue that the pursuit of human33 UNDP, Public Perceptions on Transitional Justice: Reporton Transitional Justice opinion polling survey conducted inApril May 2007 in Kosovo (Pristina, 2007).34 See Ross Clarke, Gulah Wandita Samsidar, “Considering Victims: The Aceh Peace Process from a Transitional Justice Perspective” (ICTJ, 2008).

25

rights violations in Kosovo or Aceh should not be hostage tosuch nationalist agendas. But there is no neutral orapolitical way to institute TJ. Decisions about where TJ isheld, in what language, under the control of which level ofgovernment, all inevitably involve taking a stand on issuesthat are the subject of ethnonational contestation inmultination states.35

The reality is that all political processes in a multinationstate – including TJ - inevitably become filtered throughthe lens of competing nation-building projects. The best wayto deal with this, therefore, may not be to deny or ignorethis dimension, but precisely to address it head on. Indeed,TJ could provide one of the first lessons for citizens in anewly-democratic multination state in how to learn to livewith the ambiguities of contested nationhood.36 Since35 Defenders of the ICTY might say that holding trials in The Hague (and in English) avoids this problem, by providinga neutral location and language. But as various chapters in this volume suggest, the very remoteness of the ICTY undercuts its ability to promote new models of reconciliation (or citizenization) on the ground. Moreover, the evidence suggests that local communities in the former Yugoslavia view the ICTY as (implicitly or explicitly) passing judgement on the legitimacy of collective ethnonational projects, and not simply judgement on the guilt of individuals. Indeed, Dan Saxon argues that the ICTY“has failed as a pedagogical tool” precisely because it seeks to sidestep, rather than engage with, the ethnonational sentiments that shape people’s perceptions of its work (Dan Saxon, “Exporting Justice: Perceptions of the ICTY among the Serbian, Croatian and Muslim Communities in the Former Yugoslavia.” Journal of Human Rights 4/4 (2005): 559-572).36 Moreover, as several other chapters in this volumeemphasize, attempts to de-ethnicize or de-politicizeindividual participants in TJ processes – to force them intothe universal human rights category of `victim’ to theexclusion of the identities under which they struggled and

26

multination states will only become peaceful and democraticif citizens learn this lesson, why not start the learningcurve with TJ? Why not adopt an explicitly multinational andfederalized conception of TJ?

What would this mean? As I said earlier, the question hasrarely been addressed in the literature on TJ (or in theliterature on multination federalism). Moreover, we havevery few real-world examples to study. While there are manyexamples of Western states that have adopted multinationconceptions of citizenship, none of them was accompanied byprocesses of TJ. If we examine the shift to multinationfederalism in the United Kingdom, Canada, Switzerland,Belgium, and Spain (and related shifts to territorialautonomy for substate national minorities in South Tyrol,Puerto Rico and the Aland Islands), none involved processesof TJ. (As I’ll discuss below, there are examples of TJbeing adopted in relation to indigenous peoples, but not inrelation to substate nationalist minorities). In most cases, this is because the pluralization of thestate to accommodate substate nationalist groups occurredlong after its democratic consolidation, and hence was notpart of a transition from authoritarianism to democracy (orfrom violence to peace). There was no period of gross humanrights violations that preceded this shift, and hence nocall for TJ.

However, there are two exceptions to this generalization:Spain and Northern Ireland. In both cases, the dramaticrestructuring of the state to accommodate substatenationalism was part of a broader transition from fascism todemocracy (Spain) or from violence to peace (NorthernIreland). These are the two Western cases where one mighthave expected processes of TJ to accompany the emergence ofmultination citizenship. It is striking, therefore, thatneither in fact adopted TJ as part of their transitions.

suffered – often simply means that these processes no longerconnect up with their actual self-understandings.

27

In Spain, federalization (to accommodate Basque and Catalannationalisms) was undertaken as part of a broader process ofdemocratic transition from Fascist rule. There were calls atthe time for transitional justice, most strongly in theBasque Country,37 presumably as a tool to remember andlegitimize Basque nationalist struggles. However, the majorpolitical actors involved in the transition rejected TJ aspotentially divisive and destabilizing.

Interestingly, claims for TJ mechanisms are now resurfacingin Spain, thirty years after the transition, and predictablyminority nationalists have again been at the forefront ofthis development. The first TJ law in Spain was adopted inCatalonia (the Law on Democratic Memory, or Llei delMemorial Democràtic), and a Catalan separatist party(Esquerra Republicana de Catalunya) was one of the initialco-sponsors of the new Law on Historical Memory at thecentral level.38 Survey evidence suggests public support forTJ remains higher in minority nationalist regions,39 so that37 Paloma Aguilar, “Justice, Politics and Memory in the Spanish Transition”, in A. Brito, C. Gonzalez Enriquez, and P. Aguilar (eds) The Politics of Memory and Democratization (Oxford University Press, Oxford, 2001), 92-118.38 The ERC “demanded the creation of a Committee of the Truth and asked the King to apologize in the name of the Spanish state for the crimes committed against Catalonia during the Civil War, the implication being that Spain, not the Francoists, were responsible for postwar repression in Catalonia” (Sebastian Balfour and Alejandro Quiroga, The Reinvention of Spain: Nation and Identity since Democracy, Oxford University Press, 2007, p. 143n11). See also Stephanie R. Golob, “Volver: The Return of/to Transitional Justice Politics in Spain”, Journal of Spanish Cultural Studies Vol. 9/2 (2008): 127-41. The ERC subsequently withdrew its support for the law, on the grounds that it didnot go far enough.39 Paloma Aguilar, Laia Balcells and Héctor Cebolla, “Individual, Family and Regional Determinants of Attitudes

28

negotiating minority nationalist issues will be an importantpart of the evolution of TJ in Spain.

In that sense, Spain may become the first case in the Westof TJ operating in the context of multination federalism.Indeed, the recent process of negotiation between thevarious centralist parties and regionalist parties overthese TJ measures is a paradigm of `multination citizenship’in action.40 But the fact that these developments areoccurring thirty years after the transition is revealing.Some commentators suggest that they are better understood ascases of “post-transitional justice” rather thantransitional justice, precisely because they are emerging asa result of the successful long-term embedding of democraticnorms, and the rise of a new post-transition generation ofpolitical leaders, rather than as tools of transition perse.41 It is precisely because Spain has thirty years of thesuccessful practice of multination citizenship that it nowfeels confident in tackling a multination form of TJ. Theidea that the central government would negotiate with a

Towards Transitional Justice: An Empirical Analysis of the Spanish Case” (Paper presented at the Midwest Political Science Association Meeting, April 2nd 2009, Chicago, IL).40 These negotiations are exemplary not just for the way theyinvolve a peaceful and democratic form of negotiation between minority nationalists and centralist parties, but also for the democratic way they handled divisions within minority regions. Catalan nationalists are divided between the left-wing ERC and the conservative CiU, who have very different views about `the politics of memory’ (eg., over the role of the Catholic church). A truly democratic form ofmultination citizenship must provide space for these internal divisions within minority national groups as well as for divisions between minority nations and dominant groups, and this appears to have been the case in the process of negotiating the Law on Historic Memory.41 Paloma Aguilar, “Transitional or Post-transitional Justice? Recent Developments in the Spanish Case”, South European Society and Politics, Vol. 13/4 (2008): 417-33.

29

separatist party over the terms of TJ would have beeninconceivable in 1977, but it is now part and parcel of awell-established practice of federal and coalitionalpolitics. In short, the recent emergence of TJ in Spainseems dependent on the fact that a practice of multinationcitizenship is already well-established; it does not provideany lessons about whether or how TJ can be used to helpestablish that practice in the first place.

We see a similar dynamic in the Northern Ireland case. As Inoted earlier, while this is not a case of multinationfederalism - since the accommodation of substate nationalismis occurring through consociational and trans-bordermechanisms rather than regional autonomy – it shares withSpain the intention to construct new relations ofmultination citizenship as part of a broader democratictransition. And here too there have been calls for TJ,primarily from victims’ groups and civil societyorganizations. Yet, as in Spain, the main political actorsdecided not to include TJ as part of the 1998 Good Fridaypeace agreement, in large part because the recognition of“victims” requires agreement on the legitimacy of the stateand of armed rebellion, which is precisely what is indispute between the national groups. Recognizing that thisissue was irresolvable, the Good Friday agreement decided tofocus on building new security institutions, rather thanholding people accountable for past (mis)-uses ofviolence.42

To be sure, the call for TJ remains strong in civil societyin Northern Ireland, and we may yet see the emergence of a42 For debates around TJ in Northern Ireland, see Fionnuala Ni Aolain and Colm Campbell (2005) “The Paradox of Transition in Conflict Democracies”, Human Rights Quaterly, Vol. 27: 172-213;Colm Campbell, Fionnuala Ni Aolain, and Colin Harvey, “The Fronteirs of Legal Analysis: Reframing the Transition in Northern Ireland”, Modern Law Review, Vol. 66/3 (2003): 317-45.

30

`post-transition’ form of TJ down the road. But as in Spain,this is likely to be the outcome of a process ofcitizenization, not a source of that process. If the variousconsociational and trans-border mechanisms in the GoodFriday agreement designed to promote `parity of esteem’ inNorthern Ireland are successful, then perhaps thesepractices of multination citizenship will make possible theemergence of TJ.

So to date at least, we have no examples from the West ofhow TJ has been used as a vehicle for building of newrelations of citizenship for substate nationalist groups.Given this absence, it is worth considering the role of TJin a closely related context – namely, indigenous peoples.In much of the social science and legal literature, substatenationalist movements such as the Catalans, Quebecois andScots are sharply distinguished from those of indigenouspeoples, such as the Aboriginal peoples of Canada orAustralia, the Sami in Scandinavia or the Maori of NewZealand. Yet the two types of groups raise some of the sameissues. Like minority nationalisms, indigenous peoples oftenclaim to be distinct `peoples’ or `nations’ with inherentrights of sovereignty and self-determination which predatethe state. And, as with minority nationalisms, this claim isinvoked to contest the larger state’s claim to legitimatelyrule indigenous peoples and their territories. Moreover, aswith minority nationalisms, various Western states havestarted to acknowledge (if only in a grudging and partialway) the legitimacy of these claims. We see trends towardsthe recognition of indigenous customary law, for example, orthe recognition of treaty rights, or the affirmation ofself-government rights – all of which (implicitly at least)recognize indigenous peoples as possessing a legal andpolitical standing that predates and constrains that of thelarger state. And these trends have recently been endorsedin the 2007 UN Declaration on the Rights of IndigenousPeoples, which affirms the right of indigenous peoples toself-determination and to autonomy. So here too we see newmodels of citizenization emerging that seek to acknowledge

31

rather than foreclose disputes over sovereignty andpeoplehood.

It is interesting, therefore, to ask whether or how TJ hascontributed to this process of defining new models ofcitizenship for indigenous people. At first glance, itappears that TJ has been more important in this context thanfor substate nationalist groups, since there have beenseveral recent apologies and reconciliation programs inrelation to indigenous peoples. However, if we dig deeper,it seems that existing models of `reconciliation’ withindigenous peoples have rarely addressed the issue ofcontested sovereignty. Reconciliation has focused ondiscrete injustices – such as the `stolen generations’ inAustralia or residential school abuses in Canada ormassacres in Guatemala and Peru – but not on the deeperquestion of contested sovereignty. Corntassel and Holderreview recent reconciliation programs relating to indigenouspeoples in Canada, Australia, Guatemala and Peru, and arguethat in all four cases, these programs failed to seriouslyacknowledge the contested nature of the state’s authorityover indigenous peoples and their territories.43 As I notedearlier, the Guatemalan CEH, while promoting a more`pluricultural’ conception of nationhood, still sought torelegitimize a single sovereign Guatemalan people, ratherthan recognizing a plurality of legitimate indigenous claimsto sovereignty or authority. The same is true in Australia44

and Canada.45

This is not to say that reconciliation programs forindigenous peoples have been a failure. The more discretehuman rights violations (such as the Stolen Generations) forwhich these programs have been adopted are enormouslyimportant in their own right, and they almost certainly havebrought healing to many survivors. And they may have also43 Corntassel and Holder, “Who’s Sorry Now?”.44 Damien Short, Reconciliation and Colonial Power: Indigenous Rights in Australia (Ashgate, 2008).45 See Jung’s chapter in this volume.

32

helped to create a more sympathetic public opinion forindigenous peoples to make a wider range of claims,including those of contested sovereignty. Acknowledging theways in which settler states have abused their authorityover indigenous peoples may make people more receptive toclaims that settler states never legitimately acquired thisauthority in the first place. In that sense, TJ processesaddressing discrete injustices may operate alongside othermore explicitly political and legal tracks that focus onrenegotiating the underlying claims to sovereignty.Indigenous peoples may agree to suspend their objections tothe assumptions of state sovereignty implicit in TJprocesses if they are confident that they are able tocontest those assumptions in other forums, and if theybelieve that TJ processes can serve more immediate goals ofhealing and public education.

However, to date at least, TJ processes have not themselvesprovided a space for indigenous peoples to raise theselarger questions about state authority and sovereignty. Theycontinue to operate within a framework that takes stateauthority over indigenous peoples as a given. Indeed,critics argue that these exercises in `reconciliation’ aremore concerned with re-affirming the legitimacy of settlerstates in a post-colonial age than with redressing the coreinjustices of colonialism. They are seen as attempts todampen or undercut, rather than to acknowledge and accept,indigenous peoples’ contestation of state authority.46 46 See, for example, Andrew Schaap, “Reconciliation”, in Brian Galligan and Winsome Roberts (eds) Oxford Companion toAustralian Politics (Oxford University Press, Melbourne, 2007); Paul Muldoon, ‘Reconciliation and Political Legitimacy: The Old Australia and the South Africa’, Australian Journal of Politics and History 49/2 (2003): 182-96; Muldoon, “The Very Basis of Civility’: On Agonism, Conquest and Reconciliation”, in Will Kymlicka and Bashir Bashir (eds) The Politics of Reconciliation in MulticulturalSocieties (Oxford University Press, Oxford, 2008), pp. 114-35.

33

This may be a premature judgement. I do not think we haveclear evidence to assess whether TJ mechanisms to deal withdiscrete injustices against indigenous peoples are helpingor hindering broader processes of renegotiating citizenship.We do not know whether TJ processes are preparing the groundfor addressing issues of self-determination in other forums,or distracting attention from those more fundamentalissues.47 But to date at least, it seems clear that TJ hasnot itself provided the vehicle by which indigenous peoplesengage in that struggle.

In short, in the case of both indigenous peoples andminority nationalisms in the West, TJ has not been used as avehicle for addressing underlying questions about how tobuild citizenship under conditions of contested authority.Western democracies have started to acknowledge thatcitizenization for substate national minorities andindigenous peoples requires opening up, not foreclosing, thecontested nature of state claims to legitimacy, sovereigntyand self-determination. However, TJ has not been a part ofthis process, and we have no precedents or models of how TJcan be used to advance this process.

All of this suggests that we are entering uncharted waters.We live in an era where both TJ and multination federalismare increasingly popular options for post-conflictsocieties. We are therefore likely to see an increasingnumber of cases where the two go together, and where TJaccompanies a shift to multinational federalism. Potentialcases include Iraq, Indonesia, Sudan, Nepal, Cyprus and Sri47 Defenders of reconciliation in Australia argued that it would help prepare the ground for negotiations of a new `treaty’ relationship with Aboriginals. However, according to Short, there is no evidence that the 10-year reconciliation process has had any effect in enhancing support for a Treaty process, or for other positive political rights (Short, Reconciliation and Colonial Power, p. 153).

34

Lanka.48 It is important, therefore, to ask how TJ cancontribute to processes of federalization, and vice-versa.

Unfortunately, we have little to go on. While there aresuccessful Western examples of multination federalism, theyhave not to date employed TJ mechanisms. I will attempt todraw some preliminary lessons for the practice of TJ inmultination settings below, but it’s important to begin bynoting the many ways in which the sort of multinationfederalism that has emerged in the West is likely to differfrom the sort of multination federalism that is emergingaround the world. The reality is that multination federalismfaces obstacles and challenges in much of the world that arenot present in the West, at least not to the same degree,and these obstacles have implications for the role of TJ.

I have elsewhere discussed a number of these obstacles,49

but let me focus here on three: the securitization of state-minority relations; the absence of human rights guarantees;and the lack of democratic accountability.

(i) securitization: Where states feel insecure in geo-political terms, fearful of neighbouring enemies, they areunlikely to treat fairly their own minorities. Morespecifically, states are unlikely to accord territorialautonomy to minorities that they view as potentialcollaborators with neighbouring enemies. In the past, thishas been an issue in the West. For example, prior to WorldWar II, Italy, Denmark and Belgium feared that their German-speaking minorities were more loyal to Germany than to theirown country, and would support attempts by Germany to invade48 And in a much longer time-frame, we can imagine the same combination eventually occurring as part of post-conflict ordemocratization transitions in China, Burma, Iran, Thailand,Pakistan, Afghanistan, Philippines – all are countries in which peace/democratization is likely to involve both a reckoning for past human rights abuses and claims to regional autonomy by substate nationalist groups.49 See my Multicultural Odysseys, particularly chapters 6-7.

35

and annex areas of ethnic German concentration. Thesecountries worried that Germany might invade in the name ofliberating their co-ethnic Germans, and that the Germanminority would collaborate with such an invasion.

Today, this is a non-issue throughout the established Westerndemocracies with respect to historic national minorities andindigenous peoples, although it remains an issue with respectto certain immigrant groups, particularly Arab/Muslim groupsafter 9/11. It is difficult to think of a single Westerndemocracy where the state fears that a national minoritywould collaborate with a neighbouring enemy and potentialaggressor.50 This is partly because Western states do not haveneighbouring enemies who might invade them. NATO has removedthe possibility of one Western country invading itsneighbours. As a result, the question of whether nationalminorities and indigenous peoples would be loyal in the eventof invasion by a neighbouring state is moot.

Of course, Western democracies do have long-distancepotential enemies – such as Soviet Communism in the past,Islamic jihadism today, and perhaps China in some futurescenario. But in relation to these long-distance threats,national minorities and indigenous peoples are on the sameside as the state. If Quebec gains increased powers or evenindependence, no one in the rest of Canada worries thatQuebec will start collaborating with Al Qaeda or China tooverthrow the Canadian state. An autonomous or independentQuebec would be an ally of Canada, not an enemy.

In most parts of the world, however, minority groups arestill seen as fifth columns collaborating with neighbouringenemies. This is particularly true where the minority is50 If we move outside Western Europe, Cyprus and Israel are consolidated democracies which still exhibit this dynamic ofviewing their historic Turkish and Arab minorities as potential collaborators with external enemies, and not coincidentally have been unable to agree on minority autonomy.

36

related to a neighbouring state by ethnicity or religion, orwhere a minority is found on both sides of an internationalborder, so that the neighbouring state claims the right toprotect "its" minority. Consider the ethnic Serbs in Bosnia,or Kashmiris in India.

Under these conditions, ethnic relations become“securitized". Relations between states and minorities areseen, not as a matter of normal democratic debate andnegotiation, but as a matter of state security, in which thestate has to limit the democratic process to protect itself.Under conditions of securitization, minority politicalmobilization may be banned, and even if minority demands canbe voiced, they will be rejected by the larger society andthe state. After all, how can groups that are disloyal havelegitimate claims against the state? So the securitization ofethnic relations erodes both the democratic space to voiceminority demands, and the likelihood that those demands willbe accepted.

In most Western countries, by contrast, ethnic politics havebeen “de-securitized". Ethnic politics is just that - normal,day-to-day politics. Relations between the state and minoritygroups have been taken out of the "security" box, and put inthe "democratic politics" box. This is one essentialprecondition for multinational citizenship to emerge and takeroot.

(ii) Human Rights Protection: A second precondition concernsthe security, not of the state, but of individuals who wouldbe subject to self-governing minority institutions. Statesare unlikely to accept minority self-government if they fearit will lead to islands of local tyranny within a broaderdemocratic state.

This too has been a worry in the past in the West, where somehistoric national minorities were seen as carriers ofilliberal political cultures. And this fear persists inrelation to some recent immigrant groups. But at least in

37

relation to national minorities, it is now widely assumedthat there is a deep consensus across ethnic lines on basicvalues of liberal-democracy and human rights. As a result, itis assumed that any self-government powers granted tonational minorities will be exercised in accordance withshared standards of democracy and human rights. Everyoneaccepts that minority self-government will operate within theconstraints of liberal-democratic constitutionalism, whichfirmly upholds individual rights. Where national minoritieshave gained autonomy in the West, their self-governinginstitutions are subject to the same constitutionalconstraints as the central government, and so have no legalcapacity to restrict individual freedoms in the name ofcultural authenticity, religious orthodoxy or racial purity.Not only is it legally impossible for national minorities inthe West to establish illiberal regimes, but they have nowish to do so. On the contrary, all of the evidence suggeststhat members of national minorities are at least as stronglycommitted to liberal-democratic values as members of dominantgroups, if not more so.

This removes one of the central fears that dominant groupshave about territorial autonomy. In many parts of the world,there is the fear that once national minorities acquire self-governing power, they will use it to persecute, dispossess,expel or kill anyone who does not belong to the minoritygroup. In Western democracies, this is a non-issue. Wherethere is a strong consensus on liberal-democratic values,people feel confident that however issues of federalism aresettled, their own civil and political rights will berespected. No matter how the claims for autonomy areresolved, people can rest assured that they won’t be strippedof their citizenship, fired from their jobs, subjected toethnic cleansing, jailed without a fair trial, or deniedtheir rights to free speech, association and worship. Putsimply, the consensus on liberal-democratic values ensuresthat debates over accommodating diversity are not a matter oflife and death. As a result, dominant groups will not fightto the death to resist minority claims. This, too, is a

38

precondition for the successful adoption of multinationalcitizenship.

Where these two conditions are absent, it is extremelyunlikely that states will voluntarily accept federalizationto accommodate minority nationalism. Indeed, there may belittle reason to hope or expect that federalization wouldactually achieve the sort of success we see in the West. Ina condition of regional insecurity surrounded byneighbouring enemies and hostile international powers,autonomy arrangements can be a threat to the very securityof the state. And in the absence of democraticconsolidation, autonomy arrangements can be a threat to thevery life and liberty of individual citizens who belong tothe “wrong” group. Under these circumstances, the intendedgoal of multinational federalism – namely, replacing uncivilrelations of enmity and exclusion with more equitablerelations of liberal-democratic citizenship – may besubverted. Institutions and policies designed to promote“citizenization” in multination states may be captured byinternal and external actors who seek to perpetuate andexacerbate relations of enmity and exclusion.

(iii) Democratic Accountability: A third factor thatinfluences the prospects for successful citizenization inmultination federations concerns the democraticaccountability of the elites who claim to speak for substatenational minorities. Throughout the Western democracies,members of national minorities can choose, in free and fairelections, between a range of political parties,representing quite different views on both ideological andnationalist issues. Voters in Catalonia, for example, canchoose between left-wing and right-wing Catalan nationalistparties, but they can also vote for centralist Spanishparties (of both the left and right). All of these partieshave reasonably fair opportunities to organize, raise funds,access the media and so on, and citizens can vote as theysee fit without fear of violence or intimidation. So ifnationalist parties receive a plurality of votes, we have

39

good reason for thinking that this reflects the depth ofnationalist identities in the population, and that theseparties have a democratic mandate to pursue nationalistpolicies. Indeed, it is precisely the fact that nationalistparties have consistently received such a democratic mandatein free and fair elections that finally compelled Westernstates to accept the need to adopt new multination forms ofcitizenship. In the past, states had the hope or expectationthat minority nationalism was simply an elite phenomenon,disconnected from any real popular discontent or aspiration,but this belief has been repudiated by the consistentability of nationalist parties to demonstrate theirdemocratic mandate in free and fair elections.

In much of the world, however, there are lingering doubtsabout the democratic mandate of minority nationalist elites,and about the depth of nationalist identities andconsciousness. Consider again the case of Aceh. In therecent peace agreement, the previously-separatist GAM rebelsagreed to renounce violence in return for significantautonomy within Indonesia. The international community hasgenerally applauded this as a successful case ofaccommodating a distinct sense of “peoplehood” or“nationhood” amongst the Acehnese, through something akin tothe Western model of multination federalism. But manycommentators dispute this assumption that the Acehnesedisplay a “nationalist” consciousness. The Acehnese wereactive participants in the Indonesian national liberationmovement, and historically viewed themselves as foundingmembers of the Indonesian nation. While GAM emerged as aneffective insurgent force in the 1970s, many commentatorsinsist that most Acehnese sought only democratization andeconomic development, not independence or “national”recognition. On this view, GAM only managed to maintain thefaçade of popular support for its nationalist agenda bysuppressing or intimidating any rivals, and theinternational pressure on Indonesia to negotiate with GAMfurther suppressed the space for democratic voice and

40

dissent within Aceh.51 In this context, there are genuinequestions about the extent to which GAM reflects the realaspirations of the group members they claim to speak for.

Some commentators go further and argue that this is true ofmost or all ethnonationalist movements – ie., that they restupon elite manipulation and intimidation, without a genuinedemocratic mandate - and that citizenization thereforerequires contesting ethnonational narratives andloyalties.52 This is clearly an overgeneralization: I see noreason to assume that the average voter for the ScottishNationalist Party is any more manipulated or misinformedthan the average voter for the Conservative or Labour Party.But it’s a legitimate concern about many specific cases,whether GAM in Aceh, or Islamic militants in Kashmir, or theTamil Tigers in Sri Lanka. In the absence of free and fairelections and a free press, we have no reliable mechanismsfor assessing the democratic mandate of minority nationalistelites. While states are often too quick to dismiss thelevel of popular support for minority nationalistmovements,53 minority nationalists are also too quick toclaim such support, leaving it systematically unclear whohas a democratic mandate.

In short, in much of the world, the prospects formultination federalism are undermined by concerns about51 See, for example, Elizabeth Drexler, Aceh, Indonesia: Securing the Insecure State (University of Pennsylvania Press. 2008); John Bowen, “Normative Pluralism in Indonesia:Regions, Religions and Ethnicities”, in Will Kymlicka and Baogang He (eds) Multiculturalism in Asia (Oxford UniversityPress, 2005), 152-169.52 For versions of this argument in the Bosnian case, see Franović, Dealing with the Past; Rangelov and Theros, Maintaining the Process in Bosnia and Herzegovina.53 In the case of Aceh, GAM has won two post-agreementelections, which suggests that there was a significant depthof nationalist consciousness, although there remainquestions about its use of intimidation.

41

geopolitical security, human rights protection, anddemocratic accountability. Moreover, none of these concernsis likely to be resolved in many of the countries undergoingtransitions to multination federalism in the foreseeablefuture. Given these obstacles, one may wonder why any of thesecountries are in fact contemplating federalism. The simpleanswer, in most cases, is that substate minorities havetaken up arms and seized territory, and are in a position toinsist on territorial autonomy as part of a peace agreement.Unlike the Western cases, these are not examples ofcountries agreeing to federalize after a peaceful anddemocratic debate. Rather, they are federalizing at thepoint of a gun.

As a result, we cannot assume that ideas of federalism orterritorial autonomy have any legitimacy amongst the generalpopulation. They are instead likely to be viewed as coercedconcessions to belligerent minorities and/orunrepresentative elites, not reasonable accommodations ofclaims of justice, and as concessions that pose anexistential threat to the state and its individual members.The experience to date suggests that such arrangements arefragile at best.

This is not to say that the alternatives are any better. Onthe contrary, attempts to cling to the fiction of unitarypeoplehood and uncontested sovereignty under theseconditions have an equally dismal track record, and oftensurvive solely through repression. Indeed, in most casesmultinational federalism is being considered preciselybecause all the other options have already been tried andfailed. Under these circumstances, multination federalismmay appear as both inevitable and impossible, or to put itless dramatically, as a matter of lesser evils and longodds.

4. Lessons and Implications

42

This, then, is the difficult terrain in which processes ofTJ are likely to unfold in multination states that aresimultaneously democratizing and federalizing. Torecapitulate the key points of the analysis so far, I haveargued that:

(a) Traditional accounts of democracy (and of thedemocratizing potential of TJ) have assumed thatdemocratization requires a consensus amongst citizensthat they form a single sovereign people, and that TJcan contribute to democratization in part by buildinginclusive national identities that enable all citizensto identify themselves as members of that singlesovereign people.

(b) This model is unlikely to work in societiescontaining sizeable, mobilized and territorially-concentrated national minorities who assert their ownnationhood or peoplehood. In such cases, successfuldemocratization is likely to involve new and innovativeforms of `multination citizenship’ that open up andsuspend, rather than resolve or foreclose, disputesover sovereignty and nationhood.

(c) While we have successful examples from theconsolidated Western democracies of how to constructsuch innovative forms of multination citizenship,particularly through multination federalism, theirsuccess has depended on certain preconditions (such asgeopolitical security, human rights guarantees anddemocratic accountability) that are absent in much ofthe world. In the absence of these conditions, modelsof multination citizenship are unlikely to be adoptedvoluntarily, and if adopted under threat of violence orinternational pressure, are unlikely to have theirintended `citizenizing’ effects.

(d) We have no precedents or models, either in theconsolidated Western democracies or in transitioncountries, for the use of TJ to help buttress suchmultination forms of democratic citizenship.

43

What are the implications of this analysis for the design ofTJ? Can we draw any conclusions about which forms of TJ arelikely to contribute to the democratic consolidation ofmultination federalism, and which are unhelpful orcounterproductive?

If my analysis is broadly correct, the most obviousconclusion is simply a call for caution and modestexpectations. We know very little about the circumstances inwhich multination federalism in transition countries islikely to succeed, and even less about the role that TJ canplay in that process.

However, if only to stimulate further debate, let meconclude with a few tentative suggestions. One option is totemporarily forego TJ in cases of transitions to multinationcitizenship. This is more or less the approach taken inSpain and Northern Ireland, the two Western cases that mostclosely approximate the circumstances in transitionalcountries. On this model, TJ need not be permanentlyrejected, but rather is deferred to a “post-transition”phase, once practices of multination citizenship have beendeveloped through other means. Perhaps effective andsustainable forms of TJ cannot be developed where citizenshave not yet learned how to deal with unresolved issues ofpeoplehood and sovereignty. As I said, we are far fromhaving clear evidence about this. Perhaps Spain and NorthernIreland are exceptional in this regard, rather than examplesof a more general rule. But their relatively-successfulexperience to date suggests that this option is worthconsidering, and that we should be hesitant to assume thatTJ is a necessary component of democratic transitions inmultination countries.

However, from the perspective of victims, and indeed ofjustice more generally, this option is obviouslyunsatisfactory. As I noted at the beginning of the chapter,the fundamental argument for TJ is the intrinsic moral

44

obligation to redress wrongs, an obligation rooted inbroader human rights principles. Justice delayed is justicedenied, and so victims and civil society organizations, andtheir international allies, are likely to push for TJ in thetransitional phase, whether or not it fits into some socialscientific account of the most effective `sequencing’ ofreforms to achieve democratic consolidation. Moreover, evenif we have no examples of how TJ contributes to aspecifically multination form of citizenship, we have goodreasons to believe that TJ can contribute to buildingcitizenship in a more minimal or thinner sense – forexample, by challenging a culture of impunity, reaffirminghuman rights principles, and acknowledging the legitimacy ofdissent and diversity.54 None of these addresses thedistinctive challenges of contested sovereignty inmultination states, but they are important in themselves,and indeed can be seen as providing some of the resourcesneeded to address the more complex challenges of multinationcitizenship.

So we need to consider possible ways to make TJ work in thecontext of building multination citizenship. The keychallenge here, I think, is to prevent TJ from beingcaptured by, and subordinated to, the dynamics ofethnonational contestation. As I’ve tried to indicate,virtually all the decisions about the design andimplementation of TJ are likely to implicate contestedassumptions about sovereignty and territory. Decisions aboutwhere to hold TJ processes, which language to use, and whichlevel of government to administer them, all implicate claimsto sovereignty and territory that are contested inmultination states. If a centrally-organized TJ is givenjurisdiction over the whole country in a multination state,this may well be perceived by national minorities as adenial of their status, and as structurally biased againstthem. Conversely, enabling substate groups to administertheir own TJ – as in Kurdistan, Aceh or Kosovo – may well beperceived by the larger state as endorsing their54 See de Greiff, “Theorizing Transitional Justice”.

45

ethnonational claims to sovereignty. Supporters andpractitioners of TJ may have had no intention or desire tobe seen as taking a stand on such contested claims, but thisis the frame within which their actions will inevitably beperceived. After all, it is often precisely thesecontestations over sovereignty and territory that engenderedthe violence in the first place, and so local actors will beexquisitely sensitive to any hint or implication that TJ isbeing implemented in a way that either advances or subvertstheir claims. When TJ is perceived in this way as endorsingone side’s claims to territory and sovereignty, the entireprocess will quickly be delegitimized in the eyes of otherkey actors. If so, TJ will not only fail to promotemultination citizenship, it will fail to achieve even itsmore basic goal of gaining public acknowledgement ofwrongdoing.55

The question then is how TJ can avoid being instrumentalizedin these ethnonational struggles, and thereby delegitimized.A first step is to ensure that practitioners of TJ areconscious and reflective about the ways in which theirdecisions inevitably, if unintentionally, implicatecontested ethnonational claims. And this in turn requiressetting aside many of the assumptions of traditional modelsof TJ, which take for granted that the goal is some sort of`nation-building’. In many models of TJ, it is taken asobvious and self-evident that the goal is the consolidationof a shared sense of membership in, and co-ownership of, asingle sovereign nation. This assumption infuses therhetoric as well as the practice of much of the TJ55 See Saxon, “Exporting Justice”, for the failure of the ICTY to gain public acknowledgement amongst its intended audience for its judgements of wrongdoing, due to the way the entire process has been read through ethnonational filters. (For example Croatian war criminals condemned by the ICTY have been welcomed home as national heroes). Since the ICTY is seen as being biased against Croatian claims to sovereignty and territory, its judgements are seen as suspect.

46

literature. All of this needs to be carefully reviewed andrevised to make clear that decisions about theimplementation of TJ in multination states are not intendedto pre-empt or prejudge contested issues of territory andsovereignty.

But being self-conscious of these risks is not sufficient.After all, the concrete challenge remains: how precisely canTJ avoid being implicated in these contestations? TJ needsto operate in specific locations and languages, and needs tobe authorized by specific levels of government. As I’veemphasized, there are no neutral spaces, languages orjurisdictions that float free of ethnonational contestationin multination states.

I don’t see any obvious way around this problem, except thebanal suggestion that wherever possible such decisionsshould be based on consensus and dialogue, not unilateraldictate by either the central government or substate groups(or of course international organizations). In principle, itshould be possible to design innovative forms of TJ that areendorsable by all parties despite their differing views ofsovereignty. Perhaps one could have a model of TJ that movesbetween different cities and regions, operates in differentlanguages, and is authorized (and staffed) by multiplelevels of government, in such a way that all sides can seeit as consistent with their own views of state legitimacy.Where all parties agree to the process, contested questionsabout who has the ultimate authority to take this or thatdecision over this or that territory can be finessed.Parties may disagree about which level of government hasoriginal or ultimate sovereignty over particular peoples orterritories, but this disagreement can be finessed if anethos and practice of consensus develops. This ethos ofconsensus (and hence of creative ambiguity regardingsovereignty) is a familiar feature of multination states inthe West that enables them to function even though groupshave very different views about the underlying sources ofsovereignty. We can also see this at work in some of the

47

cases of TJ for indigenous peoples, where efforts are madeto secure indigenous consent to state-administered TJprocesses, and to emphasize that these processes are notintended to pre-empt or foreclose debates in other forumsthat allow for the contestation of state authority.

But of course this appeal to an ethos of consent justrestates or relocates the problem. The experience to datesuggests that this ethos is often not present duringtransitions in multination states. That is what we see inAceh, for example, or in Bosnia. States and substate groupsare too jealous of their powers, and too distrustful ofother’s, to accept the sort of vague or ambiguous formulasthat allow all sides to finesse their disagreements. As aresult, calls for TJ remain blocked, stymied by the need forconsent.56

Indeed, appealing to an ethos of consensus and of creativeambiguity over sovereignty as the basis for an effective TJis putting the cart before the horse. This ethos is, in manyways, simply another name for the practice of multinationcitizenship, and the task is precisely to build such apractice where it does not yet exist. If a climate of trustand consensus across ethnonational lines had already existedin countries like Sudan, Iraq, Indonesia, Bosnia or SriLanka, they would not have undergone years or decades ofethnonational civil war. Insofar as effective TJ requiressuch an ethos or practice of multination citizenship, thenit surely can only come at the end of the transition56 Gaining the consent of indigenous peoples to such TJ processes, despite their contesting of state authority, may be easier because in most of these cases, judgements of wrongdoing and victimhood are clear-cut. No matter who runs the TJ process regarding the Stolen Generation or residential school abuses, it’s uncontested that the perpetrators arefrom the settler society, and that the victims are indigenous peoples. In the case of ethnonationalist violence in Bosnia, say, or Aceh, it’s much more contested who are the victims and who are the perpetrators. The ultimate identification of perpetrators and victims is likely to depend on who precisely is in control of the TJ process. And in this context, groups may be much less inclined to yield on or suspend contested issues of sovereignty.

48

process, not the start (as the Spanish and Northern Irelandcases suggest).57

Given these obstacles to creating consensual and effectiveTJ at the formal state level, we could perhaps considerpursuing TJ in other forums. One option is to take TJ out ofthe hands of the state and shift it to the internationallevel, under UN jurisdiction, as with the ICTY. Theinternational level has the advantage of seeming to offer aneutral location, a neutral language, and a neutral level ofgovernance, standing at an equal distance from allethnonationalist claims to sovereignty or territory. Butthis appearance of neutrality is often deceiving, since thevery act of creating international tribunals is seen asmaking an exception to the normal rules of state57 Of course, even if a state proceeds unilaterally without the consent of a substate minority, it can nonetheless take steps to make TJ open and inclusive of minority members. Forexample, it can take proactive steps to accommodate the language and religion of minorities (holding bilingual hearings, publishing in minority languages, recognizing minority religious holidays and rituals etc). This sort of `multicultural outreach’ is indeed quite common. And for some minority groups it may be sufficient to generate a sense of trust and legitimacy in the process. But this is unlikely to work in the case of nationalist minorities. It is precisely the distinctive feature of such groups that their concern is not only or primarily with cultural accommodation, but rather with asserting claims to sovereignty and territory that compete with state claims. Inthis context, it is the sheer fact of the assertion of unilateral state authority over the minority and its territory that is problematic, regardless of how much that assertion is garbed in a multicultural form. In all ethnically diverse societies TJ faces the challenge of cultural accommodation and multicultural outreach, but the distinctive challenge in multination states concerns the contested sources of the sovereign power needed to adopt TJ in the first place.

49

sovereignty, and hence as implicitly or explicitlydelegitimizing assertions of state authority.58 In anyevent, even if distance creates the appearance ofneutrality, it also undermines the pedagogical aims of TJ.If TJ is to have transformative effects, wrongdoing must bepublicly acknowledged by the communities involved, and notjust by a remote international body.

Another more radical option is to pursue TJ at the level ofcivil society, amongst alliances of victims’ organizationsand human rights groups that cut across ethnonational lines.We see examples of this in the former Yugoslavia, and somecommentators have argued for greater international supportand funding of these “bottom-up” processes of TJ activism,as a way of bypassing the inability of state structures toget beyond their anxieties about conflicting claims tosovereignty and territory.59 But these ideas remainexperimental, and even their proponents acknowledge that atsome point TJ has to penetrate formal state structures if itis to achieve its goals of redressing wrongs and buildingcitizenship.

In short, we have a number of possibilities for thinkingabout the relationship between TJ and multinationalcitizenship: we might defer TJ to a post-transitional phase;we might try to develop consensual models of TJ that finesseissues of contested sovereignty; or we might relocate TJ tointernational levels or to civil society. However, all ofthese are essentially untested, and not backed by eitherclear models or firm evidence. Much research and58 This is one reason why Kosovar Albanians are moresupportive of the ICTY than either Serbians or Croatians.The former see it as supporting their contestation ofSerbian state legitimacy; the latter see it as contestingthe legitimacy of Serbian and Croatian statehood. See Saxon,“Exporting Justice”.59 See Rangelov and Theros, Maintaining the Process; and Franovic, Dealing with the Past.

50

experimentation is needed to help clarify the differentoptions, and there various risks and opportunities. All wecan safely say at this point is that all the options requireus to radically and creatively rethink current assumptionsabout how TJ relates to state legitimacy and nationhood.60

60 An earlier version of this paper was presented at workshops at the ICTJ and the Oxford Centre for the Study ofInequality and Democracy, and I’m grateful to the participants for the many helpful comments and questions. Special thanks to Paige Arthur and Pablo de Greiff for theirdetailed suggestions and advice.

51