Abstract
Focusing on political parties, this article highlights divergent conceptualizations of key elements of transitional justice that are part of the current contestation of the dealing-with-the-past process in Burundi. Speaking to the emerging critical literature on transitional justice, this article attempts to look beyond claims that there is a lack of political will to comply with a certain global transitional justice paradigm. In this article, transitional justice is conceived of as a political process of negotiated values and power relations that attempts to constitute the future based on lessons from the past. This paper argues that political parties in Burundi use transitional justice not only as a strategy to protect partisan interests or target political opponents, but also as an instrument to promote their political struggles in the course of moulding a new, post-conflict society and state.
Since the country's independence in 1962, Burundi has experienced several cycles of violence. An opportunity for peace finally arose with the signing of the Arusha Peace and Reconciliation Agreement in 2000. In order to highlight measures for reconciliation and against impunity, the Arusha agreement contains provisions on transitional justice through the Truth and Reconciliation Commission (TRC) and a judicial mechanism in the form of an International Judicial Commission of Inquiry (IJCI) to be followed by an international criminal tribunal. However, neither the TRC nor the judicial mechanism to try those responsible for grave human rights violations has been implemented.
Transitional justice has become a prominent element of the liberal peace-building approach (Sriram 2009). It aims to promote social and political integration and reconciliation, to enhance the rule of law, to fight impunity and to increase trust in government institutions. This normative model is mainly based on humanitarian law, international criminal law and human rights law (Bell 2009). However, some argue that Burundi lacks the political will that is necessary to implement such a normative model of transitional justice (Human Rights Watch 2009). According to advocates of this model, such an “implementation” deadlock might be due to the fact that transitional justice touches on fundamental interests, especially those of individuals who have been implicated in past violence. However, as argued in detail elsewhere (Rubli 2011), transitional justice in Burundi might also be contested because the political actors’ understandings of the basic concepts of transitional justice, such as justice, reconciliation and truth, do not conform to international transitional justice norms or the liberal peace-building model.
This article understands transitional justice as a political process of negotiated values and power relations that attempts to constitute the future based on lessons from the past. Transitional justice is not a value-neutral process (Bell 2009) but instead reflects certain normative beliefs and values about what a post-conflict society and state should look like. The production of particular truth narratives and lessons from the past might be used as an instrument for political struggles to construct, forge and mould society and the political apparatus, and/or as a strategy to protect partisan interests. Hence, as the title of this article suggests, transitional justice is used in the efforts to (re)make the social world and the state.
The article is based on a previously published paper that argues that transitional justice in Burundi may be contested not only because it touches upon the fundamental interests of some of the decision-makers but in addition due to different understandings of key elements and concepts of a dealing-with-the-past process (Rubli 2011). Relying on additional empirical data, the present article develops the argument that transitional justice can be used as an instrument to mould the post-conflict society and the state in a way that includes context-specific conceptualizations of justice or reconciliation. 1
I would like to thank Briony Jones, Sara Hellmüller, Rina Alluri, Markus Höhne and the two anonymous reviewers for their useful and detailed comments and feedback. The opinions expressed are solely those of the author. Research for this article was supported by the doctoral programme Global Change, Innovation and Sustainable Development associated with the NCCR North-South programme of the Swiss National Science Foundation (SNSF).
This article is structured in three parts: The first part outlines the evolution of the transitional justice process in Burundi and explains why some consider the political will to follow the global, normative transitional justice model to be absent. Speaking to the emerging critical literature on transitional justice, this article attempts to look beyond such claims that political will is lacking, and conceives of transitional justice as a political process of negotiated power relations and values. Focusing on the conceptualization of justice, the second part highlights the political parties’ different understandings of justice that are associated with the dealing-with-the-past process in Burundi. Divergent conceptualizations and normative values inherent in a domestic dealing-with-the-past process are related to broader questions of international interventions and their transfer and imposition of certain concepts, ideas and norms to other societies. In the last part of the article, I argue that transitional justice and the produced “truths” and interpretations of the past are used not only as a strategy to protect self-interests, but also as an instrument for the political struggles in the course of moulding a new, post-conflict society and state. Looking at the strategies, discourses and concepts that various actors evoke can help us understand how transitional justice is linked to and embedded in broader state-formation processes and meta-debates about the organization of both society and the political apparatus.
The data for this analysis, including qualitative interviews, various public speeches, news articles and radio broadcasts, were collected during extensive fieldwork in Burundi between 2009 and 2011. During this period, I conducted in-depth interviews 2 with presidents, vice-presidents, secretary-generals and speakers of political parties, including the CNDD, CNDD-FDD, FNL, FRODEBU, FRODEBU-Nyakuri, MSD, UPD and UPRONA. 3 It is evident that political parties are not unitary actors and that the opinions expressed during the interviews may not reflect the official stance of the party, as most of the parties do not have an official position regarding transitional justice. However, all the interviewees occupy high-ranking positions in their respective parties and are authorized to speak in their names. The information collected during the interviews has been analysed against the backdrop of other public statements by these parties.
Since the Burundian elites are used to expressing themselves in French, all interviews were conducted in this language. All quotes in this article have been translated into English by the author.
Conseil National pour la Défense de la Démocratie (CNDD), Conseil National pour la Défense de la Démocratie – Forces pour la Défense de la Démocratie (CNDD-FDD), Forces Nationales de Libération (FNL), (Sahwanya-) Front pour la Démocratie au Burundi ([Sahwanya-]FRODEBU), Front pour la Démocratie au Burundi-Nyakuri (FRODEBU-Nyakuri), Mouvement pour la Solidarité et la Démocratie (MSD), Union pour la Paix et le Développement (-Zigamibanga) (UPD [-Zigamibanga]) and Union pour le Progrès National (UPRONA).
Burundi's Transitional Justice Process
Burundi has experienced several cycles of violence since the country's independence in 1962. In 1965 an unsuccessful coup d'état by a group of Hutu gendarmes triggered retribution by the Tutsi-dominated national army. This pattern repeated itself several times in the following decades. In 1972 a Hutu-led insurrection, caused by the more-or-less systematic exclusion of Hutu from government institutions, triggered a violent response by the national army and led to the killing and disappearances of many Hutu intellectuals (Uvin 2009). In 1988, in an outburst of violence, around 20,000 Hutu were killed by the national army. After democratization efforts at the beginning of the 1990s, a civil war broke out in 1993 with the assassination of the first democratically elected president, Melchior Ndadaye (Daley 2007).
In August 2000, Burundian political parties signed the Arusha Peace and Reconciliation Agreement, but it did not end the violence, as the Hutu-dominated rebel movements at the time, the CNDD-FDD and the FNL-Palipehutu, were not included in the peace negotiations (Sculier 2008). The agreement included provisions on transitional justice – namely, a TRC, which would shed light on the truth about grave violence, promote reconciliation and forgiveness, and clarify the entire history of Burundi. An IJCI would also be set up to investigate and establish the facts relating to genocide, war crimes and crimes against humanity. Based on its findings regarding the occurrence of such acts, an international criminal tribunal was to be set up to implement trial processes and punitive measures for those held responsible (Arusha Agreement 2000: art. 6 and 8, chap. 2, prot. 1). While the TRC and IJCI were meant to be set up during the transitional period (2001–2005), neither of them has been established so far. The transitional government did not consider transitional justice a priority; instead, its preoccupation was with ending the violent hostilities, integrating the rebels into the state structures and preparing the elections and the new constitution (BUJ-II-a-8). This also holds true for the armed groups (CNDD-FDD and FNL); they did not focus on “talking about truth”, but rather on obtaining “a position of strength” through their integration into the government and state structures (BUJ-II-a-8).
In 2004 the Parliament of Burundi passed a law on the establishment of the TRC, but it was never implemented (Vandeginste 2011). In the same year, the UN sent an international assessment mission to evaluate the advisability and feasibility of the IJCI (Vandeginste 2009). The resulting Kalomoh Report (2005) called for a reconsideration of the Arusha formula (TRC, IJCI and the international criminal tribunal) and proposed a twin mechanism, consisting of a TRC and a judicial process. Following the endorsement of the Kalomoh Report, the UN and the new CNDD-FDD-dominated government negotiated the implementation of the report's recommendations in 2006 and 2007. The main issues of discord were the question of amnesty for war crimes, crimes against humanity and genocide; the independence of the special tribunal's prosecutor; and the interrelationship between the TRC and the tribunal (Ndikumasabo and Vandeginste 2007). In 2007, the negotiating parties eventually agreed to hold popular consultations on transitional justice. These consultations on the modalities and composition of the TRC and the tribunal were finally conducted in 2009 (Comité de Pilotage Tripartite 2010).
In June 2011 the Burundian president nominated a Technical Committee, whose responsibility it was to draft the law that would establish the TRC and its functions. As of November 2012, the draft law was expected to be studied soon by the Council of Ministers and the National Assembly (IWACU 2012). However, the judicial mechanism, the special criminal tribunal, has as of today not yet been conceptualized, and when (and whether) this mechanism will be established remains unknown.
Thus, 12 years after the signing of the Arusha agreement, none of the transitional justice provisions have been implemented. This has led practitioners and advocates of a global transitional justice policy (Nagy 2008) to argue that there is no political will by Burundian leaders to establish the TRC and the special criminal tribunal (Human Rights Watch 2009; BUJ-II-b-1). Underlying such claims is the assumption that some political actors are likely to contest the principles of transitional justice due to the fear that they could be held responsible for past crimes. A transitional justice process, particularly criminal prosecution, is seen by many political actors as a direct threat. If investigated and/or put on trial, political actors’ reputations would be at risk. They might also experience a loss of their power or political position if they are found responsible for human rights violations, not to mention long prison sentences if a special tribunal finds them guilty of having committed certain crimes. Such arguments of a lack of political will to “deal with the past” according to international transitional justice norms all stem from a logic of rational choice. Consequently, advocates of a global transitional justice policy argue that actors who do not benefit from transitional justice or who may face consequences because of it are likely to act as “spoilers”, who will try to circumvent or “manipulate” it to meet their needs (cf. Subotić 2009; Vandeginste 2010). The intuitive assumption is that the more power actors hold, the more capable they are of shaping transitional justice mechanisms in a way that serves their interests.
Several authors have highlighted the fact that power relations influence any transitional justice process and that the resulting institutional design is an outcome of prevailing power constellations (cf. Sieff and Vinjamuri Wright 1999; Rubli 2010). Since the Arusha peace talks took place, power constellations have undergone considerable changes in Burundi. The former rebel groups CNDD-FDD and FNL, which were not included in the negotiations in Arusha, transformed themselves into political parties in 2005 and 2009, respectively. The power balance between different political parties may be largely determined by their representation in the government and the parliament. The first post-transition elections in 2005 were won by the CNDD-FDD; however, the UPRONA and the FRODEBU held a combined 30 per cent of the seats in the National Assembly (African Elections Database 2011). In 2010 the CNDD-FDD won the elections with a majority, and UPRONA was represented by only 17 seats in the National Assembly. Other political parties, among them the FRODEBU and the FNL, boycotted the presidential and parliamentary elections. Forming a coalition, they claimed that the communal elections were rigged (ADC-Ikibiri 2010). They are no longer represented in the government and parliament (African Elections Database 2011).
Due to its electoral victory, the CNDD-FDD currently holds a powerful position and could in principle impose its stance on transitional justice to the detriment of other political parties. Nevertheless, the approach to transitional justice taken by the CNDD-FDD-dominated government appears to be in line with the global transitional justice paradigm. 4 Certainly factors of path dependency of the transitional justice policy as it was enshrined in the Arusha agreement (strongly influenced by the FRODEBU and the UPRONA, the two strongest parties during the peace talks) and pressure from international donors may play a role in shaping Burundi's official policy. As Subotić (2009) has shown for the case of the former Yugoslavia, the alleged adherence to the global transitional justice model might be guided by ulterior political motives, including obtaining financial aid or international legitimacy. Although claims of attempts to manipulate and instrumentalize are widespread, relying on such an analysis is shortsighted. Elster (1998) argues that social actors not only are driven by interests including securing advantages for specific individuals (including themselves) and groups, but in addition, that their actions might be motivated by reason, passion and/or emotions. By “reason”, which is highly subjective, Elster has in mind “any impartial consideration of the common good or of universal rights” (1998: 34). Hence, it would be wrong to assume that political parties simply manipulate transitional justice institutions to suit the party members’ interests; instead it is likely that they are motivated by normative considerations as well. This becomes clear when we consider that transitional justice is a political process, as the following paragraph outlines.
It is difficult to judge the extent to which the CNDD-FDD's position influenced and is reflected in the government's policy on transitional justice. However, such an analysis would go beyond the scope of this article, which primarily focuses on political parties.
Transitional Justice as a Political Process
Some authors of the emerging critical literature point to the ways in which transitional justice is highly contested (McEvoy and McGregor 2008). This is due, in part, to the fact that it involves negotiations, compromises and responses to political, legal and moral dilemmas (Sriram 2009). On the one hand, norms, institutionalized rules and laws regulate our behaviour and shape our political relations, our language and even the way we think; thus, they have the capacity to regulate violent behaviour and expose arbitrary state practices (McEvoy 2007). In the transitional justice language, they fulfil the functions of the “never again” or “non-recurrence” premise (cf. Joinet 1997). This is based on the assumption that transitional justice measures are supposed to reform the system that allowed gross human rights violations to occur, and to design a legal and political system that prevents violent conflict. On the other hand, formalized norms and laws represent a way of conceptualizing and articulating how we would like the social world to be (McEvoy 2007). Thus, transitional justice is not a mere “(value-)neutral process” (Bell 2009: 6) to deal with past human rights abuses, but instead reflects certain social and normative values. Transitional justice should be understood as an inherently political process because it is mainly in the field of politics that we decide how a society should be organized and how norms and perceptions will be translated into legally binding institutions. Hence, transitional justice, as it is understood in this article, is a political process of negotiated values and power relations which endeavours to constitute the future based on lessons from the past. Consequently, various actors attempt to shape a transitional justice process, along with its mechanisms and outcomes, so as to ensure that they are favourable to them and that they reflect their ideological preferences. Before elaborating on why transitional justice is used as an instrument for political struggles, the next paragraph outlines different conceptualizations of justice by political parties.
Conceptualizations: Retributive, Social or Reconciliatory Justice?
As a political process or social engineering project (Rubli 2012), transitional justice reflects different perceptions and conceptions about justice and reconciliation, or, more generally, about what a post-conflict society should look like. This makes it an inherently normative concept. Based on international criminal law, human rights law and humanitarian law, transitional justice appears to be influenced by a Western understanding of justice and conceptualized in a rather retributive and adversarial way (Lambourne 2009) which often pits victims against perpetrators and where the accused is usually sentenced to prison. Some authors examine ways in which different actors conceptualize justice. For example, regarding “opinion leaders” in Uganda, Allen finds that their critiques of transitional justice reveal different understandings of justice; reconciliation is preferred to retribution, and amnesty and truth-telling are more acceptable than punishment of the guilty (Allen 2007). In the case of Burundi, based on two large-scale studies of the rural population, Ingelaere and Kohlhagen (Ingelaere and Kohlhagen 2012; Ingelaere 2009; Kohlhagen 2009) find that terms like “norm” and “law” are usually not associated with justice, peace or equity, but instead with ideas of constraint, arbitrariness or political power. Using a different epistemological framework to reflect on what is labelled “transitional justice”, rural Burundians evoke concepts that diverge from those of the global transitional justice model to refer to similar objectives. The authors conclude that
it is not simply a matter of forgetting the past; it is a matter of dealing with the past differently (Ingelaere and Kohlhagen 2012: 53–54).
This also might hold true for Burundian political parties, who might contest the transitional justice process because their conceptualization of justice does not fit with that of the normative global transitional justice model.
While Burundi's two biggest former rebel groups, the CNDD-FDD and the FNL, might reject the idea of a tribunal because some of their members might be among the accused (Watt 2008), neither party conceptualizes justice as retributive and punishing. A memorandum on transitional justice issued by the CNDD-FDD in May 2007 states that
the choice has to be made between national reconciliation through the Truth and Reconciliation Commission and repression by means of the special criminal tribunal (CNDD-FDD 2007).
The party suggests some kind of middle path that
consists of favouring reconciliation and submitting only the disputes which could not be resolved through reconciliation to the special criminal tribunal (CNDD-FDD 2007).
If the perpetrator of the crime has acknowledged the facts and asked for forgiveness, and the victim has granted it, then the CNDD-FDD would consider judicial accountability through the tribunal to be unnecessary (CNDD-FDD 2007). This view has further been explained by a representative of the party: Reconciliation does not start at a precise moment, but instead is a steadily evolving linear process to which something new is added everyday (BUJ-II-a-6). For the CNDD-FDD, the reconciliation process between Hutu and Tutsi has already progressed considerably and the cleavage between them has been closed, or at least significantly reduced. According to one party representative (BUJ-II-a-6), reconciliation in Burundi has already reached a certain level, and a special criminal tribunal would “destroy what has already been achieved in terms of reconciliation” because individuals would simply be accused, and these accusations would once more divide the people (BUJ-II-a-6). Justice promoted by the tribunal would risk re-framing the conflict once more in ethnic terms by opposing (Hutu) perpetrators to (Tutsi) victims (BUJ-II-a-6). Instead, a tribunal's justice must contribute to reconciliation.
The tribunal's perceived conception of a punitive and retributive form of justice does not fit with the CNDD-FDD's conception of reconciliatory justice which the party believes Burundians need. This different conception of justice might have implications for the eventual approaches to transitional justice, including judicial mechanisms. The party may consider the tribunal to no longer be necessary because the reconciliation process would at that point be too advanced and disputes would have already been resolved. Or, the tribunal's mandate might focus more on reconciliatory justice. We might interpret the CNDD-FDD's reconciliatory approach to dealing with the past as an attempt to avoid the prosecution of its own members in the special criminal tribunal for past human rights violations. But the party's focus on reconciliation might also be a normative preference for a reconciliatory justice approach to dealing with the past. Despite the fact that many members of the CNDD-FDD are accused of being involved in corruption and extrajudicial killings, the CNDD-FDD-dominated government firmly expressed its support for the fight against impunity and for prosecuting criminals in these spheres. 5 Considering the party's alleged preference for prosecuting corruption and extrajudicial killings, its preference for reconciliation in regard to dealing with the past cannot solely be explained by the fact that it is trying to protect its own members from prosecution.
While the government's preference for prosecution in cases of corruption and extra-judicial killings is firmly expressed in public, this rhetoric does not seem to be rigorously applied. For example, a recent report by a governmental commission (2012) finds that none of the cases mentioned by civil society organizations qualify as extrajudicial killings.
The FNL also strongly opposes punitive justice and the special criminal tribunal which punishes perpetrators; instead, it proposes that those who ordered the crimes should show regret and remorse and ask the population for forgiveness (BUJ-II-x-1). In addition to the fact that its members are accused of having committed crimes (Human Rights Watch 2010), there are three possible reasons underlying this rejection of a tribunal. The first is rather pragmatic: The party believes that if everybody who has committed a wrongdoing in the past is accused, then there would be only very few innocent people left (BUJ-II-x-1). Second, the FNL does not trust the Burundian justice system, as it considers it to be biased and partisan. As a party member puts it,
like the army was mono-ethnic, also the justice [the judicial sector] was mono-ethnic, thus talking about the independence of the magistracy would be very difficult (BUJ-II-a-4).
Finally, for the FNL, some of the past crimes that it believes should be dealt with are difficult to define, as they concern the exclusion of one ethnic group from education, economic wealth and access to the state (BUJ-II-a-4). The FNL, which claimed to have fought for social justice (for the Hutu), is convinced that the planned tribunal would not address such past social injustices. Thus, the perceived punitive justice of the tribunal does not correspond to the FNL's perception of the social justice needed for Burundians. Instead, the FNL proposed that popular consultations on transitional justice should lead to a social contract (Dar es Salaam Agreement 2006), hence they should be used as a foundation of a “new” society.
In contrast, the UPRONA advocates for a more retributive approach towards transitional justice. A representative of the party explains that justice should come first, because nothing else would dissuade those who have killed people from repeating their criminal act (BUJ-II-a-1). Hence, punishment constitutes a guarantee of non-recurrence and a measure for preventing fears of genocide among some of the Tutsi population in the Great Lakes region. “We can speak about negotiations, reconciliation and forgiveness” only after justice has been applied (BUJ-II-a-1). According to a representative of the UPRONA, forgiveness cannot be enforced (BUJ-II-a-1) and it may not prevent recurrence, because someone who asks for forgiveness may not be sincere (BUJ-III-x-1). In this sense, forgiveness is equated with a lack of punishment and with impunity for past crimes. While the UPRONA considers the TRC and knowing the truth necessary for reconciliation and to break the cycle of violence, discovering the truth should not simply be used to advance forgiveness. In contrast to the view of the CNDD-FDD, the process of requesting and being granted forgiveness does not ensure reconciliation, because in the UPRONA's view a minimum of judicial accountability is needed to reconcile and create the “new” Burundian society.
To conclude, these examples show that political parties do not appear to share the same understanding of justice, nor do those diverse understandings necessarily reflect a retributive and punishing approach of dealing with the past as the global transitional justice model promotes. However, these divergent conceptualizations of key components of transitional justice do not necessarily mean that Burundian political parties reject a dealing-with-the-past process, per se. Like in the case of rural Burundians studied by Ingelaere and Kohlhagen (2012), political parties seem to formulate and pursue concepts and objectives for dealing with the past differently than the global transitional justice model advocates. Hence, it is important to look beyond claims of a lack of political will and an alleged rejection of transitional justice by political elites. Instead, discussions around transitional justice should be seen in the context of meta-debates about the organization of society and the state.
Transitional Justice as an Instrument for Political Struggles
If we accept that transitional justice is a political process attempting to constitute the future based on lessons from the past and that it is embedded in broader meta-debates, it becomes clear that transitional justice and its outcomes might be used as an instrument to advance political struggles in order to mould the social world and the future state apparatus. Various actors negotiate, shape and compete over the nature and direction of a transition, as
whoever can win the transition, can win the peace, and whoever can win the peace, can win the war (Bell 2009: 25).
Leebaw (2008) argues that historical lessons are framed in relation to the needs of the present to avoid volatile conflict and to (re-)create a nation. Such historical lessons, produced “truths”, and interpretations of the past are then translated into institutions and institutionalized norms that organize and guide society and the state. Thus, the past is framed, produced or suppressed in such a way that it can serve as the basis to construct and mould a society and political apparatus that reflect the vision of a particular actor.
Transitional justice has the capacity to adjudicate the rights and wrongs of the conflict and to assess and judge individual guilt and social and institutional responsibilities (Bell 2009). Consequently, through transitional justice various political actors attempt to depict an official past and a dominant truth narrative that can be a basis upon which to construct the “new” post-conflict society and state according to their visions. Hence, some actors may use transitional justice as an instrument for political struggles and to legitimize their visions in the process of organizing, constructing and moulding society and the political apparatus. This vision of a “good” society and state is not only shaped by beliefs and values, it also embraces a situation where a person need not fear the negative effects of transitional justice. Individuals or a group might have different perceptions of what constitutes a “crime” and of how such a crime should be punished. If they fear prosecution for such crimes, they would probably not support a transitional justice policy that punishes these crimes. They might try to circumvent or at least shape the transitional justice process so as to ensure their own well-being. Moreover, they might direct transitional justice mechanisms against others for strategic interest-based calculations or for normative reasons (cf. Elster 1998).
However, on a conceptual level, it is difficult to know whether political parties evoke certain discourses and positions only in an attempt to legitimize their interests and aspirations, or rather because those discourses reflect the party's understanding and conceptualization of justice, truth or reconciliation. But actors themselves may not be conscious of whether they use certain discourses in order to legitimize their interests or whether they are an expression of certain beliefs and understandings.
It is important to note that a party's stance on transitional justice can change over time. The Burundian political landscape is characterized by many splits and fragmentations within political parties. For example, the president of the FRODEBU at the time of the Arusha negotiations, Jean Minani, formed his own political party, the FRODEBU-Nyakuri, in June 2008. His stance on transitional justice seems to be closer to that of the CNDD-FDD than of his former party, the FRODEBU (BUJ-II-a-3, BUJ-III-a-3). For another example, during the Arusha negotiations in October 1998 the UPRONA split into two factions, of which one was led by the hardliner Charles Mukasi (Watt 2008), who continues today to promote a strong punitive approach toward transitional justice. The FRODEBU also seems to support the tribunal more now than it did prior to the elections in 2010. However, it would go beyond the scope of this article to discuss in detail the evolutions in the positions of political parties since the start of the peace process.
In shaping the dealing-with-the-past process and forging the future society and the state, actors mobilize resources at their disposal and refer to existing and (re-)invented discourses. “Resources” refers to the material basis of political action, which includes, among other things, bureaucratic capacities, organizational skills, specific knowledge about the topic of transitional justice, financial resources and control over the use of physical violence (Hagmann and Péclard 2010). In addition to resource mobilization, actors are active subjects in producing, or referring to, existing or (re-)invented transitional justice norms and discourses (ibid.). Such a discourse constitutes the integration of “local”, “customary” and/or “traditional” forms of justice into a formal transitional justice process (e.g. Shaw et al. 2010; Huyse and Salter 2008), which might be evoked to justify divergent conceptualizations of justice, truth and reconciliation as propagated by the global transitional justice model. As an instrument for political struggles, on the one hand, the instruments of transitional justice may be used against other political parties or to protect one's own. Such arguments have led advocates of a legalistic conception of transitional justice (cf. McEvoy 2007) to claim that transitional justice is simply manipulated and instrumentalized. On the other hand, transitional justice and its constructed version of the past constitute an opportunity to legitimize political claims and aspirations. The following paragraphs demonstrate how political parties use transitional justice as a strategy to protect themselves or to target political opponents; how truth is constructed; and how the design of transitional justice institutions are negotiated in a way that is useful in efforts to (re)make the social world.
Some political parties in Burundi dread the possibility that the TRC may produce a truth that protects political adversaries. For example, a representative of the FRODEBU, a political party which has been part of an extra-parliamentary coalition since the elections in 2010, said that if the UPRONA and the CNDD-FDD are implicated in crimes in some way, would they not design “a TRC that protects themselves”, given that they both hold governmental power today (BUJ-II-a-3)? According to a representative of the CNDD-FDD, the truth should be used to rehabilitate certain individuals that have been unjustly accused (referring to its own members), but also to identify the criminals (BUJ-III-x-1). A representative of the FRODEBU (BUJ-III-a-1) evokes the perceived intention of the CNDD-FDD to focus the investigation of the TRC primarily on the events of 1972 and 1993, prior to its own creation (1998), thus leading to only the UPRONA and the FRODEBU being investigated (BUJ-III-a-1). He adds that the FRODEBU would focus on the period starting in 1962, which should be extended until today, because after the signing of the ceasefire (with the CNDD-FDD in 2003), and even after the elections in 2010, violence continued (BUJ-III-a-1). For the FNL such transitional justice would be an “unjust justice”. A representative reflects:
Would it be only justice for a certain political context, and would [real] justice only start to function afterwards (BUJ-II-a-4)?
In the meantime, it is likely that some individuals have managed to circumvent condemnation due to their high-ranking positions, influential power and other self-protecting systems (BUJ-II-a-4). Although not explicitly, this interviewee (BUJ-II-a-4) is referencing members of the CNDD-FDD who have been accused of having committed human rights violations, but have never stood trial. Double standards may emerge due to the fact that crimes committed in the past will be prosecuted but risk going unprosecuted when committed in the present, as transitional justice only applies to a certain time period (BUJ-II-a-4). Hence, for the FNL, transitional justice does not contribute to restoring the country's judicial system and rule of law, as the transitional justice literature suggests (e.g. Van Zyl 2005).
Further, questions related to timing and sequencing are important and can have an impact on the power balances in a post-conflict society. During the Arusha negotiations, some of the Tutsi-dominated parties firmly requested that the tribunal be put in place before the elections in 2005 (Rutamucero 2006), as they expected that Hutu politicians (especially those who joined rebel groups) would face criminal prosecution, which would end their political careers. Once sentenced or jailed, they would no longer be political competitors in elections for pro-Tutsi parties (Vandeginste 2010). Thus, these political parties have been accused of using the concept of transitional justice, especially the special tribunal, to strengthen their power and gain more political influence through elections by “eliminating” political adversaries and competitors. In this way, the Mukasi wing of the UPRONA reiterated in 2009 its request to establish the special tribunal before the 2010 elections (UPRONA 2009).
Transitional justice might be used not only to target political opponents, but also as an instrument to construct certain “truths” and “facts” in order to legitimize political claims and to mould society and the state according to a certain ideal or vision. For example, the FNL's understanding of truth is not a simple one in the sense of “knowing what happened”; indeed, according to a party member (BUJ-II-a-4), there are two different phenomena: the reality and the truth. Knowing the truth is a process which is always unfinished, whilst reality is constituted by facts. He exemplifies this by saying that a corpse found in a river would be a fact. In contrast, truth would be the process of knowing who killed this person, under what circumstances, with what motive or intention, and whether the murderer acted on the command of somebody else (BUJ-II-a-4). By distinguishing between the reality (the violence and crimes) and the truth (the motives), the party might try to morally and politically justify certain past crimes and the party's violent rebellion.
The CNDD-FDD also uses transitional justice discourse to mould a certain image of itself. By constantly highlighting and focusing on reconciliation, the party tries to further strengthen its image as a national populist party that represents both ethnic groups and that brought peace and reconciliation to Burundi (cf. BUJ-II-a-6; Reyntjens 2005). 6 Therefore, the ultimate aim of uncovering the truth should further enhance the reconciliation process:
This is in contrast to their former claims that they had fought for the Hutu cause (The Economist 2005).
This truth is used in a wise way in the sense that it would lead Burundians to reconcile (BUJ-III-x-1).
If the “discovered” truth would cause conflicts to re-emerge, then the TRC would not be useful for Burundi (BUJ-III-x-1). Thus, the TRC should produce a truth that would reconcile, bridge the gaps between former adversaries (meaning, ethnic groups) and construct a harmonious society.
The design of transitional justice institutions, including deciding what period of time they will cover, largely determines the “truth” that will be “found” and the interpretation of the historical “facts” that will serve as a basis for organizing the “new” society and state. During the Arusha negotiations, the period of investigation constituted one issue of discord between the political parties. The predominantly Hutu parties, known as G7, requested that the investigated period start in 1965 (Fondation Hirondelle 2000), when the appointed Hutu prime minister was killed, and a failed coup d'état staged by Hutu military and gendarmerie officers led to retaliatory acts by the army (Vandeginste 2010). In contrast, the predominantly Tutsi parties, known as G10, demanded that the period under investigation start in 1993, when the assassination of the then-president, Melchior Ndadaye, triggered revenge killings of Tutsi (Fondation Hirondelle 2000). Eventually, the political parties present in Arusha decided that the TRC and the IJCI should cover the period from the date of independence on 1 July 1962 up to the date the Arusha agreement was signed, in 2000 (Art. 6 and Art. 8, Chap. II, Prot. I). However, the national consultations and the peace agreement with the FNL have extended the proposed period of investigation until 4 December 2008, when Burundi's last rebel group agreed to transform itself into a political party (cf. Comité technique 2011: art. 6). In a press conference on 20 January 2012 the UPRONA proposed that the TRC's period of investigation should start in 1958 – when Burundi's first political parties were created – and end the day when the law establishing the TRC is passed (Urakeza 2012). If this amendment is ratified, current human rights violations (including those committed by the government) would also be investigated. Moreover, by starting in 1958 with the creation of the first political parties, the process could emphasize the view of independence that favours the role of the UPRONA. While the UPRONA imposed itself as a single party in 1966 (OAG 2009), it had played a significant and positively perceived role in leading the country to its independence. This lengthening of the investigation period may have implications for the historical contextualization of facts and the production of an official “truth” regarding past events. Hence, the suggested period of investigation can be used by political actors to emphasize their positive role in Burundi's history and to provide the basis for the legitimization of a certain vision of society and the state.
Some smaller Tutsi-dominated parties requested during the Arusha negotiations that the killings of Tutsi in 1993 be classified as a “genocide”. If the truth produced by transitional justice supports this view, the official narrative will give more legitimacy to the ethnic quota required by the constitution. The constitution of 2005 stipulates that the National Assembly should be composed of 60 per cent Hutu and 40 per cent Tutsi members, and that the ratio in the defence and security forces be 50 per cent Hutu and 50 per cent Tutsi (Vandeginste 2010). Though some argue that this ethnic quota would protect Tutsi against ethnic extermination, it amounts to an over-representation of Tutsi in political institutions compared to their share of the population (14 per cent Tutsi; Sculier 2008).
Conclusion
This article has analysed transitional justice as a political process of negotiated values and power relations. First, it has demonstrated that the contestation of Burundi's planned transitional justice institutions may not be solely due to political parties’ self-interests but can be considered as an expression of different conceptualizations of justice, truth and reconciliation. While this piece has focused on outlining the political parties’ divergent values and understandings, it would be valuable to look further into their construction. Generally, the political parties in Burundi appear rather detached from the population, and their ideological programmes do not seem to be emergent from or responding to their electoral constituencies. Indeed, further research on the aggregation of preferences and the formulation of programmes combined with large-scale studies on the expectations of ordinary Burundians (e.g. Ingelaere 2009; Uvin 2009) is needed.
Some authors of the emerging critical literature on transitional justice have analysed the normative assumptions guiding the global transitional justice model (e.g. Nagy 2008; Tiemessen 2011; Sriram 2009) and point to ways in which justice is conceptualized on a local level (e.g. Allen 2007; Ingelaere 2009). These discussions are related to the wider question about transferring or “imposing” certain concepts, ideas and norms on African societies. Local contestation of (or even resistance toward) them results in claims of manipulation, instrumentalization or lack of political will, which lead to unintended (negative) consequences. Most international guiding documents on transitional justice emphasize that such a process needs to be context-specific, locally grounded and socially acceptable. However, it seems still to be the case that local or traditional mechanisms are considered appropriate and worthy of international funding only if they possess a certain kind of formality, resemble Western courts or can be adapted to meet “universal” standards (Allen 2007; Baines 2010; Rubli 2012). However, the “local” is defined by disputes over values, practices, memories and efforts to (re)make the social world (Leebaw 2008; Colvin 2008), and they may be an “invention of tradition” by local elites (Allen 2007). In the course of giving social meaning and relevance to norms and concepts, they are always filtered, adapted and manipulated to fit a particular local cultural context.
In this article, transitional justice is conceived of as a political process that produces an official narrative of the past and historical lessons upon which the future is constituted. Along those lines, I have argued that political parties in Burundi use transitional justice as an instrument for political struggles. Since transitional justice has the capacity to adjudicate rights and wrongs, thereby reflecting certain normative beliefs, it can serve as an instrument to (re)make the social world. However, the article does not aim to neatly distinguish whether transitional justice is used as a strategy to protect self-interests or as an instrument to mould the post-conflict society and the state according to a political party's vision. Methodologically, such a distinction would be very difficult to make. For example, in the case of one political party accusing another of being responsible for criminal acts, demands for prosecution might be an expression of the belief that those perceived to be guilty need to be punished in the “new” society/state, which does not tolerate impunity, or such demands might be an instrument to sideline political opponents. Making such a distinction and focusing on the instrumentalization and manipulation of transitional justice can only be achieved if one evaluates a specific transitional justice process against a particular ideal and norm. However, if we accept that transitional justice reflects values and beliefs and is part of, and embedded in, broader processes of forging and remaking society and the state, then it is clear that transitional justice does not necessarily follow a linear trajectory or converge on a certain ideal (Hagmann and Péclard 2010). Hence, it is more important to look at the strategies, actions and discourses that actors evoke, refer to and use to comprehend particular ongoing debates and contestations of transitional justice. Indeed, further research is needed on the framing and conceptualization of key elements of transitional justice and on how they reflect and relate to meta-debates about the organization of society and the state. This includes questions of how to understand the construction of justice, truth and reconciliation conceptualizations, and how to reconcile their differences without falling back into a culturally relativistic approach. Understanding divergent conceptualizations can be a step toward framing and adapting formal transitional justice processes to particular contexts and ensuring that they are socially meaningful. Finally, looking at the strategies, actions and discourses of multiple actors and how they use transitional justice as an instrument for political struggles may help us better understand how key concepts and processes of transitional justice relate to political legitimacy, history and state formation in order that more comprehensive peace-building processes can be designed.
BUJ-II-a-1 (June 2010), representative of UPRONA
BUJ-II-a-3 (September 2010), representative of FRODEBU
BUJ-II-a-4 (October 2010), representative of FNL
BUJ-II-a-6 (October 2010), representative of CNDD-FDD
BUJ-II-a-8 (September 2010), government official
BUJ-II-b-1 (June 2010), representative of civil society
BUJ-III-a-2 (October 2011), representative of UPRONA
BUJ-III-a-3 (October 2011), representative of FRODEBU-Nyakuri
BUJ-III-a-5 (October 2011), representative of FRODEBU
BUJ-III-a-4 (October 2011), representative of UPRONA
BUJ-III-a-1 (November 2011), representative of FRODEBU
Broadcast interviews, Bujumbura, Burundi
BUJ-II-x-1 (April 2010), Radio Isanganiro.
BUJ-III-x-1 (January 2011), La Benevolencija.