By Denisa Kostovicova*
Contemporary conflicts are seldom contained within nation states. The cross-border nature of violence has prompted scholars to call for a regional approach to transitional justice (TJ). This article furthers our understanding of the regional dimension of postconflict justice by addressing the puzzle of whether discourse in regional-level debates is different from that in national- or local level debates. It applies a communicative deliberation perspective by focusing on the patterns of talk during civil society deliberations on postconflict justice. Empirical evidence is based on the unique regional civilsociety- led process in the Balkans that aims to establish a regional fact-finding commission, known by its acronym RECOM. Critiquing state-centred approaches to the study of TJ, this study conducts mixed-method research and applies text analytic techniques appropriate for the investigation of a large volume of regional-level data. Quantitative text analysis is carried out by applying a dictionary method to over half-a-million words of RECOM’s textual data. Findings suggest that regional-level debates have a higher level of restorative justice discourse than nonregional debates. Qualitative analysis of justice discursive frames explores this finding further, and shows that deliberators’ endorsement of the regional approach to TJ rests on a set of historical, present and future considerations, underpinned by a sense of regional consciousness.
KEYWORDS: regional approach, postconflict justice, civil society, text analysis, Balkans
Contemporary conflicts are seldom contained within nation states. Victims and perpetrators often come from different sides of a border, as in the Balkans, the Great Lakes Region and the Middle East. This phenomenon has implications for postconflict transitional justice (TJ). Chandra Lekha Sriram and Amy Ross have identified an ‘impunity gap’ in the pursuit of postconflict justice, whether by judicial or nonjudicial instruments. State-centred TJ mechanisms are unable to offer satisfactory justice in the context of cross-border violence in the post-Cold War period. Scholars have responded to the implications of the cross-border nature of conflict by calling for a regional approach to TJ. These calls point to the need for long-overdue dialogue between TJ scholars and those in conflict studies, where the study of the regional dimensions of conflict has entered its third decade. At the core of the concept of a regional conflict is recognition of the ‘many and multifaceted linkages’ between conflicts in neighbouring countries. Regional dimensions of conflicts have become prominent with the rise of globalization. Global interconnectedness has erased a clear distinction between the inside and outside of nation states, with significant implications for contemporary conflicts. As Nadine Ansorg sums up, wars ‘expand to a multiplicity of transnational actors of violence and security that correlate in complex relations and often compete for political control and the monopoly of violence in a region.
Regional dimensions of conflict are consequential for peace efforts. The crossborder nature of the commission of war crimes and human rights violations poses a challenge for postconflict justice. While juridical separation between victims and perpetrators has presented obstacles to legal redress, the physical separation of ethnic groups into ethnically homogenized states or substate entities helps entrench ethnocentric interpretations of truth and justice. It follows then that constraints posed by borders in the pursuit of truth and justice can be more appropriately overcome by adopting a regional approach to TJ. But what precisely would a regional approach entail? Apart from focusing on a need to overcome an exclusive focus on the nation state, scholarly calls for a regional approach to TJ have thus far remained vague. A regional approach could be understood both in relation to regional institutions as well as regional civil society. Similarly, the scholarly inquiry could focus on regional norms, identity or discourses. In sum, theorizing about a regional perspective on TJ is at its inception.
This article contributes theoretically, empirically and methodologically to answering the question raised by this special issue: is a regional approach consequential for TJ, and, if so, how? I focus on the discourse of regional civil society actors. Specifically, I address the puzzle of whether debates about TJ held at a regional level are different from debates held at a national level. To assess the merits of a regional approach, I conduct a mixed-method investigation of the content of civil society deliberations and apply text analysis to a large volume of text data on TJ. This study furthers the TJ scholarship by showing that a regional approach to TJ has an impact on how actors deliberate about TJ: regional-level debates have a higher level of restorative justice discourse, while deliberators’ endorsement of a regional approach is underpinned by a sense of regional consciousness.The article draws on empirical evidence from a unique, regional civil society initiative in the Balkans that advocates the establishment of a regional fact-finding commission, the Coalition for the Regional Commission for Establishing the Facts about War Crimes and Other Serious Human Rights Violations in the former Yugoslavia from January 1991 to the end of 2001, known by its acronym RECOM. By doing so, this study critiques and complements predominant state-centred approaches to the study of postconflict justice and makes a broader contribution to the field of TJ in three ways: first, it captures the agency of local societies in contrast to focusing on norm-promoting global actors and institutions; second, with its ‘bottom-up’ perspective and focus on civil society, it adds to analyses focused on the institutions ofthe state; and, lastly, it overcomes the limitations of methodological nationalism, for which a nation state serves as ‘the main grid through which social scientists see the world. Methodological nationalism is inappropriate for studying the regional dimension of TJ. This study recognizes the need to ‘abandon inherited habits of formulating questions and pursuing answers in terms of territorial-national-state-society entities, and offers a model for analyzing regional-level TJ data.
The article is structured as follows. I first review the debate on the role of civil society in relation to postconflict TJ and make the case for the study of processes unfolding within civil society from a communicative deliberative perspective that focuses on the patterns of discourse. I proceed by introducing the case study of a regional TJ-seeking process in the Balkans. I then present the study’s data and methods, followed by presentation and analysis of the findings. In the conclusion, I reflect on theoretical and methodological implications of the study of TJ from a regional perspective, and chart avenues for future research.
CIVIL SOCIETY, DELIBERATION AND TJ
Appreciation of Ju¨rgen Habermas’ theory of communicative action helped TJ scholars recognize dialogic interaction as a means of reconstructing divided societies through nonconfrontational engagement with the violent past. In contrast to strategic action, where actors are interested solely in achieving their interests, at the crux of communicative action is the idea of communicative actors sharing knowledge in order to arrive at mutual understanding. Drawing on Habermas, scholars have centred on two interrelated questions: where and how can communication promote postconflict justice? The where question has been animated by Habermas’ idea of the public sphere, defined as ‘the sphere of private people com[ing] together as a public, which affirmed the importance of civil society in TJ processes. The how question has led scholars to focus on specific requirements of a deliberative communicative exchange, and why deliberation would be conducive to the goals of TJ, such as peace and reconciliation. Many scholars have singled out the discursive conception of civil society, embodied by the ideas of public sphere and deliberation, as most promising for TJ. One of the main challenges for TJ scholarship and practice is stakeholders’ (both perpetrators and victims) perception in postconflict environments that the practice of TJ is illegitimate. Yet, the success of TJ instruments ‘depends upon their ability to establish and maintain legitimacy with ordinary citizens. Therefore, civil society is seen as key to promoting and legitimizing postconflict justice as a mediator between distant international criminal tribunals and domestic publics, as well as between universal human rights norms and local contexts. In domestic settings, civil society challenges nationalist narratives that promote ethnocentric interpretations of conflict and hinder postconflict reconciliation. Civil society can thus help foster an ‘accountability environment by democratizing the debate about the criminal past during transition from conflict to peace.
Critical perspectives have challenged the role of civil society in promoting TJ. The NGO-ization of civil society resulting from external donors’ engagement with ‘a narrower slice of the civil society cake, which in reality has a plurality of forms (groups, associations, networks, etc.) has undermined civil society’s ability to promote accountability. Seen as being in the pockets of international donors and distant from other grassroots organizations, NGOs are derisively labelled as ‘peace-profiteers. Also, according to critics, these NGOs’ adoption of universal human rights language and norms has alienated local constituencies from TJ initiatives.
Consequently, NGOs have been seen as directly responsible for delegitimizing TJ. The two perspectives on civil society that either endorse or critique civil society activism have been studied within a diffusion paradigm, which refers to the spread of TJ as norm and practice. Here the focus has been on the ability of civil society to promote universal norms of justice and human rights in order to transform postconflict states and societies. This outward-oriented paradigm overlooks processes that unfold within civil society. The theory of democratic deliberation helps us address this lacuna in the scholarship and look inside a civil society initiative to understand civil society’s role in legitimizing TJ.
To this end, particularly advantageous is the conceptualization of the ‘civil public sphere’ as ‘“space”, where individuals without official status can communicate and attempt to persuade one another through argumentation and criticism about matters of general concern. Where matters of criminal past are concerned, the fact that civil society as a deliberative space is removed from the state is consequential. Postconflict states (or at least some of their elements) are by rule implicated in the past violence, as peace rarely coincides with complete regime change. Furthermore, postconflict states are often defined by consociational arrangements that institutionalize ethnic logic. Conceiving civil society as space enables people from different ethnic groups, who also often live in segmented communities, to meet. It also allows ‘communicative freedom, which releases deliberators from pressure to comply with ethnically centred views. Civil society as a space paves the way for the transformative logic of communication both at a micro-individual and macro-social level. In the Habermasian perspective, the question of where the communication unfolds is linked with the question of whether and how procedural requirements of deliberation are fulfilled during communication. One answer to this question has come from theorists of democratic deliberation, who have traditionally studied the institutional political sphere in established democracies and have recently turned to the study of divided societies. They were puzzled by the question of whether deliberation, defined as a reasoned exchange of arguments that considers views of others, can take place despite the legacy of conflict, and whether participants in ethnically mixed deliberative settings would inevitably side with their ethnic kin. Testing this assumption on empirical evidence, scholars have found that deliberation across the ethnic divide is challenging, but possible. This result is by no means intuitive. Deliberation requires high procedural standards during a communicative interaction, such as reciprocity, reason, respect, equality and other-regarding logic, when weighing and articulating arguments. These demands are particularly high for divided postconflict societies. Conflicts tend to entrench ethnically centred perspectives. Empirical studies of deliberation across ethnic divides have spurred theorizing about prospects for ‘deliberative reconciliation. Scholars of deliberative democracy and peacebuilding have converged on a view that deliberation can help reconstruct postconflict societies along inclusive civic, as opposed to exclusive ethnic, lines But the notion of deliberation as a form of ‘communicative justice comes with a major caveat. Scholars who have studied divided societies have premised their studies of deliberation explicitly on the avoidance of divisive subjects. We know that divided societies can deliberate about education, for example, but do not know how deliberators engage directly with the legacy of war crimes. I address this theoretical gap and contribute to the empirical study of deliberation engaged directly with the legacy of mass atrocity. I draw on Donald Ellis’ explicit focus on communication, which is implicit in deliberation. A communicative deliberation approach aims to ‘transmit new ideas and information but also to cultivate improved communicative competence and more fully engaged collaboration designed to manage conflicts. Claire Garbett has pointed out the importance of coherence between a TJ process and its outcome. This study looks at legitimation of TJ as a means of managing conflict by studying the alignment between the TJ discourse and its outcome. To do so, this article’s empirical investigation turns to ‘patterns of talk’ in civil society deliberations on TJ. The regional dimension of these patterns is investigated by studying a regional civil-society-led TJ initiative in the Balkans, reviewed in the following section.
BACKGROUND TO THE RECOM INITIATIVE
The aspiration to establish a regional mechanism for TJ to address the criminal legacy of the Balkan wars, a textbook example of regional conflict, defines the RECOM process. This process started in 2006 when three prominent NGOs from the region – Documenta from Croatia, the Research and Documentation Centre from Bosnia and Herzegovina, and the Humanitarian Law Centre from Serbia – initiated a debate involving all ethnic communities in the Balkans over what would be the most appropriate response to the legacy of mass atrocity in the region. These debates gave birth to the idea of a regional fact-finding commission. The subsequent founding of the Coalition for RECOM in 2008 formalized the process of regional justice-seeking. The Coalition has attracted the membership of some 2,000 human rights groups and individuals throughout the region. The perceived failings of the International Criminal Tribunal for the former Yugoslavia (ICTY), the primary instrument of TJ at the time, galvanized support for the idea of a regional, victim-centred approach to TJ. Activists responded to its inability to bring about reconciliation in the region and to acknowledge the victims. While the ICTY has removed the possibility of impunity, ICTY proceedings and verdicts have been politicized by local elites and publics unwilling to give up an ethnic perspective on victimhood. Moreover, as a retributive TJ instrument, it was never geared towards the particular needs of victims for acknowledgement of their suffering. The RECOM initiative conducted consultations with a range of stakeholders, including civil society associations, victims, veterans, lawyers, artists, journalists, academics and youtin its inclusion of the entire region. Initially, the consultations focused broadly on how to address war crimes in the region. Once the idea of a regional restorative approach was articulated, focus shifted to the draft statute of the proposed regional commission. The RECOM initiative conducted 134 one- or two-day meetings at the regional, national and local levels from 2006 to 2011. It also held 10 larger regional meetings (forums). The statute of the regional commission was adopted in 2011. Encouraging respect for all speakers regardless of their ethnicity and inviting ex
The RECOM process generated unique regional-level data that allow us to identify and analyze regional as opposed to nonregional (national and local) patterns of discourse. Despite their availability on RECOM’s website, the consultation transcripts have not been studied rigorously.
This neglect is due in part to the sheer volume of RECOM data. The corpus of some four million words is simply too large to make manual content analysis feasible or practicable. One way to address this problem is to apply computer-assisted quantitative text analysis, which approaches text as data and is suited for tackling large volumes of text. The language of these transcripts has also stood in the way of this technique. Translations of mutually unintelligible languages in the region are available (e.g.,from Bosnian–Croatian–Serbian into Albanian, and vice versa). However, availability of transcripts in mutually intelligible Bosnian, Croatian and Serbian does not help overcome the problem. The violent breakup of the former Yugoslavia resulted in ‘cultural and linguistic separation.’ Close variants of what used to be the Serbo- Croat language spoken in Serbia, Croatia, Montenegro and Bosnia and Herzegovina before the wars have grown even further apart. Therefore, preprocessing of RECOM’s textual data for this study unfolded in three stages.
First, 20 transcripts were selected, encompassing all consultations deliberating about the draft statute of the regional commission. Secondly, the entire corpus, amounting to 581,292 words (Table 1), was translated into Serbian by the author, with some research assistance. Linguistic and lexical uniformity made the RECOM data suitable for computer-assisted quantitative textual analysis methods. Thirdly, the metadata corresponding with the variations in the level of consultations (regional versus nonregional) were added
Focusing on the regional dimension of TJ, this article investigates whether alignment between the communicative process and its outcome is a feature of regional or nonregional consultations. The mixed-method study combines quantitative and qualitative text analysis of RECOM’s data: a dictionary method, which is a quantitative text analysis method and – to the author’s knowledge – the first application of such a method in the TJ field; and a qualitative discursive framing method, itself assisted by a computer-assisted technique. This study applies an explanatory sequential design, where quantitative results are used to select the direction for a qualitative study, while qualitative data ‘help explain quantitative results that need further exploration.
Text analysis includes ‘techniques for the social scientific study of communication. The selection of an appropriate quantitative text analytic model hinges on the research question and the quantities of interest. Various text analysis methods have been widely used in political science. Scholars have analyzed large volumes of text, including parliamentary debates, committee meetings, party manifestos, political speeches and newspaper reports. The digital revolution and the rise of social media have added to the variety of data, now available on an unprecedented scale. The analytic techniques of quantitative text analysis have been used to further our understanding of conflict, but they have yet to be used to further scholarship on TJ. According to Roel Popping, a ‘dictionary is a concrete representation of the investigator’s theory as it is related to verbal data. As a quantitative text analysis method, the dictionary is used to classify texts into categories. As Justin Grimmer and Brandon Stewart specify, this means ‘inferring either the category of each document, the overall distribution of documents across categories, or both. Researchers may apply existing dictionaries,or, alternatively, create ‘a problem-specific dictionary.’
For this study, I constructed a dictionary to measure the restorative dimension of the RECOM consultations, in order to assess whether restorative discourse aligns with the restorative aim of this TJ process. Dictionary-based methods use the relative frequency of keywords to measure the extent to which each category is present in a document or set of documents. The dictionary method approaches text data as a bag of words. This approach, which allows the transformation of text into quantitative data, regards words as semantically independent and disregards word order. The power of word-count-based methods relies on ‘the invaluable role that good statistical work can play in bringing out patterns in the data.’ The substantive content of the dictionary is at the heart of the research design. Its construction represents a challenging step in the research process, in contrast to its automated application, which is straightforward. As an analytical construct, dictionaries ‘must point to but also distinguish between phenomena.’ In other words, they need to capture variation in phenomena of interest. This is accomplished by identifying words that distinguish categories. In this study, the dictionary-based measurement of the restorative dimension of the RECOM consultations centres on a relatively narrow concept category of restorative justice. It is a nonhierarchical, single-category dictionary. The concept category is defined as ‘existing of words and phrases that are aggregated and that represent a theoretical concept relevant for the investigator’s theory.’ The a priori strategy in the construction of a dictionary is appropriate for a theory-based dictionary, where theory is used to compile at least a preliminary set of entries forming the basis for a search instrument.63 Such a dictionary is user-defined, whereas a rigorous theory-led approach serves to address criticism that dictionaries necessarily contain a subjective component. Theoretical insights from the TJ scholarship on restorative justice guided the identification of initial entries. The preliminary list was refined by taking into account local meanings of universal concepts associated with restorative justice (see Appendix Table A1).
Finalization of the list and of the specific form of the search terms included in it requires extensive empirical verification to ensure that ‘words are assigned to concepts with high validity.’ An important step in the construction of a dictionary is disambiguation of words. This is accomplished by checking how words are used in the context and applying a range of techniques to pin down a theoretically relevant meaning of a search term. Input from the expert coder is needed for this task as it cannot be automated, resulting in a ‘hybrid technique.’Search entries in the dictionary for measuring the restorative dimension of the RECOM’s consultations were explored in their context to minimize the ambiguity in the dictionary. The included words are in the form of wildcard search entries. I checked to ensure that this strategy did not have the unintended effect of introducing ambiguity. This format was chosen in order to deal with the highly inflected form of the Serbian language. Lastly, the dictionary was applied to RECOM’s corpus, taking individual words as a unit of measurement and a document (i.e., text of a consultation) as a unit of analysis. The analysis was carried out using quanteda, an R package for the quantitative analysis of textual data.
Discursive Frames Analysis
A qualitative discursive frames analysis addresses the question of how participants in the regional justice-seeking process construct and understand a regional approach to TJ. It provides further explanation of the result of the quantitative text analysis that identifies the restorative dimension of TJ discourse at regional-level debates. Frames analysis has been utilized in political communication and social movements’ research. According to Dennis Chong and James Druckman, the major premise of the framing theory is ‘that an issue can be viewed from multiple perspectives and evaluated on different bases.’ The study of representation and meaning is therefore at the core of frames analysis. Reversing the logic of social movement research, with its focus on the construction of grievance, frames analysis in this article probes the construction of redress for past wrongs. This construction sheds light on broader processes of justice-seeking, since frames analysis also provides insights into the processes of legitimizing different policy positions. Frames analysis is interpretive, since it allows us to trace cognitive patterns and multiple meanings. By doing so, it ‘problematizes and challenges existing authoritative views and framings of reality. The strategy for identifying frames was twofold. The initial formulation of frames was deductive. The prior deductive categories, which were assigned to text, were theoretically defined, drawing on arguments that a regional approach to TJ is needed because of the changed nature of conflict. It was combined with an inductive identification of frames, which involves an iterative process of formulation, subsumption and revision of categories emerging from text. Identification of frames was conducted by using a keyword-in-context (KWIC) function, in order to be able to tackle the large volume of text contained in the RECOM corpus. This computer-assisted technique provides ‘lists of all occurrences of a word and the context (e.g. sentence etc.) around it.’ Computerized searches were conducted based on a selection of relevant words and phrases, such as ‘regional approach’ and ‘regionally,’ in order to identify lexical environments in which they occur. Ultimately, specific frames concerning a regional approach to TJ were linked to arguments and concepts associated with them. Discourse and communication are central to the production of reality and allow a researcher to study ‘aspects of the constructive process and its products.’ Accordingly, the aim of discursive frames analysis is to identify the ‘conceptual scaffolding’ of the regional approach to TJ.
FINDINGS AND ANALYSIS
A Restorative Dimension of Discourse and the Regional Level of Justice-Seeking
Through applying the dictionary method to the RECOM data, this study investigates whether the coherence between the outcome – the creation of a restorative TJ instrument – and restorative discourse during deliberations is a feature of regional or nonregional deliberations. The theoretical interest in the restorative justice dimension of discourse (and not of retributive justice terms, for example) stems from the fact that RECOM’s deliberations on the statute dealt with the remit of a regional fact-finding commission after this restorative justice instrument had already been endorsed. It was a response to the lack of legitimacy of retributive TJ embodied by ICTY trials. The alignment between the process and the outcome provides insight into internal discursive legitimacy within deliberations on TJ. Therefore, I constructed a dictionary to measure the restorative dimension of discourse comparatively on regional-level as opposed to nonregional-level debates. The results of applying the dictionary show that regional-level consultations have a
slightly higher level of restorative justice discourse,86 measured by the proportion of words in the restorative dictionary, after applying tf-idf weighting (Table 2). This finding is important for a deliberative democratic perspective on TJ. It demonstrates that at the regional level, as opposed to the nonregional level, the debate about TJ has more content dealing with restorative justice, and, in that sense, is more closely aligned with the aim of establishing a restorative TJ instrument. The
regional-level debates were also characterized by greater ethnic diversity of participants. This indicates that increasing the multiethnic composition in debates on TJ is conducive to better-focused discussions on overcoming the conflict. In sum, the findings show that the level at which justice-seeking takes place has a consequential impact on how the communicative process unfolds. In line with the rationale behind the mixed-method research design, the next section uses discursive frames to explore the quantitative analysis result that regional-level debates are different from nonregional-level debates.
Why a Regional Approach? Discursive Frames Analysis
A qualitative analysis of discursive frames probes how support for a regional approach to TJ is constructed. The focus here is on the buy-in into the process in the statute-drafting stage, after a regional approach was accepted by the RECOM Coalition. Seven frames can be identified, of which only one aligns closely with the scholarly argument proposing a regional approach to TJ based on the changed nature of conflicts fought in the post-Cold War period. Discursive frames and illustrative quotes by the speakers during the RECOM consultations are presented below.
Character of Conflict Frame
This frame is closely aligned with the scholarly argument that the cross-border nature of crimes requires a regional response. It was summarized succinctly by a participant in the RECOM process who believed in the need for a regional approach to TJ ‘because we were killing each other regionally.’ This view is echoed in the RECOM consultations, as speakers refer to ‘multinational, regional wars’ fought on the territory of the former Yugoslavia. For them, the ‘regional character’ of the conflict implies that ‘perpetrators are in one state, evidence in another, and everything else is elsewhere.’ This regional dimension, therefore, necessitates a regional approach to TJ, because a regional commission will ‘encompass all countries of former Yugoslavia, and have access to all witnesses, and especially victims, who are in other states. Therein lies its power.
The empathy frame, which is the most prevalent frame, reflects a consensus by all participants in the RECOM consultations that victims lack acknowledgement of their pain and suffering, both from members of other ethnic groups and also within their own communities. The persistence of an exclusive claim to victimhood in the region partly results from the physical separation of victims and perpetrators in different states, but also from consociational institutional arrangements that support ethnic separation, such as those in Bosnia and Herzegovina and in Kosovo. As one speaker put it:
The victims are acknowledged at a national level, but this is not sufficient to address their need that the other side should hear about their suffering. Victims’ faith in humanity is seriously shaken by the fact that the other side – the side that is responsible – is unwilling to hear about what had happened to them. Thus, the victims need an acknowledgement at a regional level and their voice should be heard at a regional level.
In addition, the empathy frame also refers to internal dynamics within the Balkan states. In this sense, it has less to do with ethnicity and more to do with domestic politics. Victims feel marginalized and manipulated by their own political elite, who acknowledge victims only when it is politically expedient and without following up with policies advancing the rights of victims. Consequently, the empathy frame refers to restoration of dignity to all victims through the incorporation of multiple perspectives, ensured by a regional process which offers recognition of the suffering even of those who do not belong to one’s own ethnic group. As another speaker explained, ‘the commission has to be regional because the victims do not have a nationality.’
The credibility frame reflects the participants’ view that the legitimacy of justice intervention is paramount. The pursuit of TJ may look like the adoption of a global norm, but it may actually have little or nothing to do with either justice or fairness, and everything to do with political expedience. For the participants in the RECOM process, where impartiality is of utmost importance, national commissions have become discredited. A monoethnic commission that would deal with a multiethnic reality is not likely to be credible, since it would be always objected for its partiality. The argument in favour of the RECOM is that a national commission is unable to address causes or effects that are attributed to other states. It seems to me that it would be riskier to have several national commissions than one regional commission. However, a regional approach to TJ is not necessarily seen as a rejection of all national-level postconflict justice initiatives. Participants are unequivocal in their respect for official initiatives and instruments that advance postconflict justice and consider national- and regional-level initiatives to be ‘complementary, while the regional level has a higher dimension that is more comprehensive and encompassing.’
The necessity frame refers to practical obstacles concerning the attainment of justice at a national level. These obstacles have an ethnic as well as a procedural dimension. The ethnic dimension has to do with difficulties encountered by victims seeking justice in the states where they are and were a minority during the conflict, although that minority may have had support from a neighbouring state, as was the case with Serbs from Croatia. Thus, a mother seeking justice for her killed son said: I cannot get either moral or financial satisfaction in the state where I live. That means that. I can only hope for the establishment of the commission that will work fairly, and I hope that most of those present here or at least their families will live to testify before that commission one day. This frame also has a procedural dimension, where regional cooperation is seen as a necessity because of wide disparities among Balkan states in their ability to provide justice. These include disparities in their legal capacities, laws and definitions (e.g., of victims and perpetrators) and different legal entitlement for victims. A regional approach that would define and introduce uniform standards through the regional commission’s national offices is therefore seen as a necessary and practical solution. Such standardization, which includes a uniform model of collecting and processing information on human rights violations, is considered essential for establishing a set of credible facts and is an important precondition for moving forward.
The cognitive frame demonstrates that the advocacy of a regional approach resonates with participants’ understanding of regionalism as an experience of life in a common state, that is, the former Yugoslavia. One participant articulated this understanding by saying that ‘we are a region that still lives in the heads of the people who remember that they used to live in the same state.’ At the same time, such an understanding of regionalism is perceived to have a ‘huge potential that can help us do something regionally.’ The association of region with life in the common state before the war corresponds to an understanding of regionalism as coexistence in close territorial proximity after the war. As another speaker put it, regardless of religious, national, cultural and other differences, we still have to live together and to cooperate. . .on the territory of the region of the former Socialist Federal Republic of Yugoslavia. A retroactive assigning of regionalism onto the reality of the defunct federal Yugoslav state, which itself was never considered a region either by people who lived in it or by outsiders, is of interest for the question of how regions emerge as cognitive constructs. For the purpose of this article, this frame is interesting because it demonstrates that the pursuit of TJ rests on, and, simultaneously, contributes to creating regional consciousness.
The conflict-prevention frame is closely related to the cognitive frame, which recognizes that people are destined to live together. According to one speaker, the processes of facing the past are closely connected, and should always include a perspective on what is happening to our first neighbour,. . .but also to all our neighbours in neighbouring states. Participants in the RECOM consultations associate a regional approach with conflict prevention because they recognize the risks for peace and security of avoiding facing the criminal past. This realization is a historical lesson, as explained by a speaker: this region is a region where one trauma is laid on top of another trauma, and then when they are all piled up, they are swept under the carpet, so that every war that breaks out is bloodier than the previous one. Therefore, a regional approach to TJ offers the possibility of ending the historical cycle of violence. The future of the region where we all live together depends on our readiness, but also on our humanity to accept responsibility and establish the truth. People who fought in the war know themselves that they’d spent that time in several states created during the breakup of former Yugoslavia. A regional approach is singled out because a ‘regional commission, whilst establishing facts regionally, through an official mechanism, can be a barrier to denial of crimes, which is very important for the victims.’ Participants are aware that the establishment of such facts in the aftermath of conflict is unprecedented in the region. Many believe that establishing these facts would contribute to ‘the prevention of future conflict, that is, final stabilization of the region, its full democratization and regionalization.’ Such a future is based on repairing relationships, since, according to one participant, ‘our regional relations as countries, but also our relations as people, have been broken.’ The future-oriented, transformative understanding of a regional approach is examined next.
This frame reveals that for participants in the RECOM process the pursuit of TJ goes beyond a societal transformation focused on the legacy of mass atrocity. In this respect, the pursuit of TJ, as Ruti Teitel puts it, is constitutive of transition. For those involved in justice-seeking in the Balkans, this implies adoption of liberal values as a criterion for reclaiming a position in the European family of nations in general, and in the European Union (EU) in particular. According to one speaker, the assertion of a new European identity is a rejection of ethnic logic: European elites view us as a semi-periphery of democracy and liberal consciousness. This is the first step in the direction of us to think in a European way, so that Europe can see that we too can resolve something together and reach joint conclusions and solutions concerning war criminals and their renunciation. That means, war criminals are all equal, whether it is an Albanian, Serb, Croat or other, and all victims are equal. This qualitative analysis of the discursive framing of a regional approach indicates that the buy-in to this particular justice strategy goes far beyond the nature of the conflict frame, which responds to the cross-border character of contemporary conflicts. As this analysis shows, the regional approach to TJ is justified by a wide range of arguments. They provide a broad basis, or ‘scaffolding,’ on which the idea of a regional commission rests. A unifying assumption cutting through all these is the idea of a regional consciousness that allows participants to recognize their common history of conflict, common contemporary and regionwide challenges (foremost among them the marginalization of victims of all ethnicities) and their common future destiny. Paradoxically, the fact that Yugoslavia was not perceived as a region prior to the outbreak of conflict in the 1990s is inconsequential. The results of discursive frames analysis add a new perspective to the findings of quantitative text analysis. Consultations held at a regional level brought together members of all the different ethnic communities. Frames analysis indicates that these consultations brought to life the sense of regionalism. The sense of commonality – common regional identity, common suffering and common striving for redress – appears to have focused communicative deliberation on the restorative task at hand.
This article explored a regional approach to TJ, heeding the call of scholars that the regional nature of contemporary conflicts necessitates a regional approach to postconflict justice. It examined the regional dimension of TJ through the lens of discourse and considered how regional debates, as opposed to national or local debates, have affected TJ-seeking. This effect was measured by analyzing communicative deliberation within a regional civil-society-led justice-seeking process. The study found that regional-level debates demonstrate a higher level of the restorative dimension of discourse, which aligns them more closely with the aim of the restorative justice-seeking process. Analysis of the meaning of a regional approach to TJ, as constructed by participants themselves, sheds additional light on these results. It shows that participants’ buy-in to the regional TJ process rests on a broad set of considerations defined by historical, present and future factors, all of them underpinned by a regional consciousness. Regional-level debates can be seen as an embodiment of that consciousness, focusing minds on the restorative aim of the process. The restorative dimension of regional-level justice-seeking was identified by viewing civil society as deliberative space, an approach that brought the patterns of communication into analytical focus. Martina Fischer reminds us that we still do not know how societies establish discourses that promote some degree of consent on facts on war crimes. By focusing on the regional dimension of deliberation, this article revealed a blindspot in the scholarship on deliberation in divided societies, which has stayed firmly fixed in a national paradigm. The findings indicate that concepts of deliberation and reconciliation should be reexamined from a regional perspective while taking into consideration the idea of regional deliberative reconciliation. Despite tackling a vast amount of textual data from a justice-seeking process involving nearly 6,000 people from different ethnic groups, this study is ultimately the study of a deliberative ‘mini public.’ A study of the dynamics of communication in a broader public sphere, or macro deliberation, was beyond the scope of this study. Future empirical research needs to investigate comparatively micro and macro levels of deliberation, especially since the deliberative legitimacy achieved during deliberation may not extend automatically to those who did not take part in the process. How do micro and macro levels of deliberation interact, and what effects do they each have on the legitimation of TJ?
This study has shown that we need to sharpen our tools for analyzing micro-level discourse in order to learn more about deliberation and TJ. Scholars have acknowledged that a focus on methods is key to pushing the frontiers of knowledge in the TJ field. This applies to the analytic possibilities both of new methods, such as quantitative text analysis, which is yet to be embraced by TJ scholars, and of mixed methods. This application of a relatively simple quantitative text analysis method allowed identification of the restorative dimension of a regional-level discourse while overcoming the limits of a state-centred epistemology and methodology in the study of TJ. The next step is to apply more sophisticated text analytic techniques to drill deeper into the RECOM data longitudinally and comparatively, including at the level of speakers’ utterances, to further deepen our understanding of the regional approach.
Lastly, mixed-method research on TJ has led to new knowledge in the field of TJ, through its underlying principles of deeper exploration and validation. Mixing text analysis methods is not just theoretically productive. This study confirms Victoria Baxter’s observation that multimethod research design in the study of TJ is also necessary in a field where many of the questions being asked ‘fall outside the scope of conventional research strategies,’ and, therefore, require innovation in both research design and methods.
*Associate Professor of Global Politics, London School of Economics and Political Science
1 This research was supported by the Leverhulme Trust (Research Fellowship RF-2015-262). The article benefited from comments by participants in the Eighth Annual OxPeace Conference, St John’s College, Oxford (2016); the ‘Reimagining Conflict-Transformation’ workshop, the Centre for Conflict, Rule of Law and Society, Bournemouth University (2016); and ‘Norms and Local Agency in Transitional Justice in the Western Balkans’ panel, British International Studies Association Annual Conference, London (2015). I received valuable advice from Kenneth Benoit and Helen Addison on preprocessing the RECOM data for quantitative text analysis, which I gratefully acknowledge. I am indebted to Tom Paskhalis for assistance with quantitative text analysis and thank Barnaby Perkes and Ivona Ladjevac for research assistance in earlier stages of this project. I also thank Pierre Hazan and anonymous reviewers for their comments, and Aaron Glasserman for editing assistance.
(Published in The International Journal of Transitional Justice by Oxford University Press, 26. 12. 2016. )