Abstract
The unrest in Burundi following President Nkurunziza's controversial re-election has put the country high on the international agenda, but research on the resurgence of turmoil is still in its infancy. A workshop held on 3 and 4 March 2016 in Freiburg, Germany, whose focus was Burundi after the 2015 elections, aimed to go beyond short-term accounts of the current unrests and extend past theorising in an attempt to address the current conflict. Special attention was paid to the interaction between external attempts to address the crisis and domestic contestation. The issues examined included the construction of Burundi as a case of successful transition to peace, the conflict's neglected legacies, and the (perceived) inadequacy of past approaches to address the current crisis, such as security sector reform, power sharing, and term limits. This also allowed us to rethink approaches developed to understand peacebuilding in Burundi since 2000.
Since the presidential elections in 2015, Burundi has witnessed its worst political and humanitarian crisis since the country's transition to peace after the civil war that started in 1993. The unrest preceding and following the elections has cost more than 400 lives and resulted in more than 240,000 refugees. Several initiatives for dialogue and de-escalation have failed to produce substantial results. So far, neither the various dimensions of past conflict dynamics nor the limits of regional and international peacebuilding endeavours have been well elucidated. We still lack a clear understanding of the failure of conflict transformation and reconciliation after the civil war. How power sharing has precisely affected political contestation beyond the issue of ethnic cleavages also requires further analysis. In a nutshell, the current unrest is forcing scholars to rethink the interplay between multifaceted forms of external intervention and domestic conflict dynamics.
Several workshops and symposia as well as concrete engagement on the part of Burundian researchers underline scholars' willingness to both critically engage with the state of the art and take into consideration more practical implications of their work. This report thus pursues several goals. It summarises some results of the Freiburg workshop, 1 situating this endeavour in the context of a growing “Burundian Studies” community and the implications of that community for case-based research more generally, and it thereby introduces the contributions to follow in this particular section of the journal.
Workshop “Domestic Dynamics of Contention and External Attempts to Address the Crises: Burundi after the 2015 Elections,” Arnold Bergstraesser Institute, Freiburg, Germany, 3–4 March 2016.
Burundian Studies: Past Achievements and Present Questions
Burundi's (post-)conflict development has motivated a body of social science research that has stimulated broader conceptual and theoretical debates. First, existing studies have examined the ethnic dimension of the civil war (see i.a. Ndikumana 1998; Uvin 1999) and the construction of the ethnic dimension (e.g. Daley 2006). More recent work has subsequently looked into how ethnic power-sharing provisions contained in the Arusha Peace and Reconciliation Agreement and in the 2005 Constitution have contributed to reducing post-war interethnic cleavages (e.g. Samii 2013). Policy briefs on the current crisis suggest that the ethnic quotas in the parliament, the government, the police, and the army have indeed been successful in preventing another outbreak of ethnic violence thus far (cf. Sterck 2015). However, power sharing has also frozen conflicting identities, not least because profound conflict transformation has been hampered by an absence of transitional justice and reconciliation (Ingelaere and Kohlhagen 2012; Vandeginste 2010). Hence, we still lack a clear understanding of legacies of the civil war and whether the process of conflict transformation was sustained or reversed during the recent crisis.
Second, and closely related to the first issue, the outcome of the Arusha peace process – namely, a power-sharing arrangement along the lines of Lijphart's consociational model, has been critically reviewed by several scholars in terms of its achievements and shortcomings (Lemarchand 2007; Reyntjens 2005). While it was arguably key in achieving the short-term objective of ending the civil war and at least initially reduced the destabilising impact of elections, power sharing has failed to pave the way for more ambitious objectives such as democracy, the rule of law, and political pluralism (Grauvogel and Simons 2015; Vandeginste 2009, 2011). Yet, the question of how power sharing has precisely affected political contestation in Burundi beyond the issue of ethnic cleavages requires further analysis.
Third, scholars have closely followed the process of peacemaking and peacebuilding in Burundi over the past 20 years (i.a. Campbell et al. 2014; Khadiagala 2003; Ndikumana 2005; Street et al. 2008). In that context, the interplay between international, regional, and domestic conflict management has attracted special attention. Both regional and domestic forces laid claim to actively shaping the Burundian crisis context, thereby rendering friction almost inevitable (Wodrig and Grauvogel 2016, see also Daley 2007; Grauvogel 2015; Wodrig 2014). Curtis (2013: 72) speaks of the “peacebuilding paradox” to describe how international, regional, and local actors have produced governance arrangements in Burundi that were at odds with their inclusionary rhetoric (see also Hirblinger and Simons 2015). These approaches to conceiving of the relationship over the years between international, regional, and domestic efforts to find a lasting solution to the violent struggle in Burundi raise the question of whether we are observing similar contradictions between international, regional, and domestic endeavours during the current crisis.
The Burundian Studies community seeks to address these open questions with respect to Burundi's post-conflict trajectory while at the same time attempting to make sense of the current unrest. An increasing number of contributions at the key conference of African Studies in Europe, the European Conference on African Studies (ECAS), illustrate this. While ECAS 5, which took place in Lisbon in 2013, saw a total of four papers focusing on Burundi, 2 at ECAS 6 in Paris in 2015, 3 there were ten individual contributions as well as an entire panel consisting of five papers that dealt with Burundi. In addition, a Contemporary Burundian Studies Symposium took place in Ghent on 15 and 16 October 2015. An open letter on the endangered freedom of thought outside and within academia originated from the Ghent symposium. 4 It expressed concern that universities in Burundi can no longer provide spaces for debate and analyses. Those Burundian researchers who have not fled the country are paralysed by fear of intimidation and repression, which often hinders their freedom of expression. In view of this, it is necessary to provide room – within Burundi and, if necessary, outside Burundi – for reflection on the current developments in the country.
See http://cei.iscte-iul.pt/ecas2013/index.shtml (30 June 2016).
See www.ecas2015.fr/ecas-program (30 June 2016).
See www.lesoir.be/1181042/article/debats/cartes-blanches/2016-04-14/quereste-t-il-liberte-pensee-au-burundi (30 June 2016).
Country-Specific Research in Times of Crises
This is occurring at a time when demand for country-specific policy advice is being fuelled by the current crisis. The need for such research does not fall on deaf ears. The researchers themselves reflect the practical implications of their work and positions, as illustrated by the open letter. But the volatile situation in Burundi renders much-wanted predictions and scenario building a difficult endeavour. Without denying that these tensions exist, the workshop in Freiburg sought to address the challenge by enabling a structured dialogue between practitioners and researchers. It took place back to back with an actors' conference organised by the foundation for development cooperation in Baden-Württemberg (Stiftung Entwicklungs-Zusammenarbeit Baden-Württemberg, SEZ), 5 which brought together members of the Burundian diaspora and practitioners, especially in the area of development cooperation. Participants from the actors' conference were present during the academic paper presentations, and vice versa.
Actors' Conference “Was können wir jetzt tun?” (“What can we do now?”), Freiburg, Germany, 3 March.
A policy panel concluded the actors' conference. Paul Seger, the former chair of the UN Burundi Configuration (and Swiss permanent representative at the UN), Georg Schmidt, the Africa Director at the German Federal Foreign Office, and Barbara Kemper and Sylvia Servaes, both (former) development cooperation practitioners, discussed what different external actors could do in the aftermath of the 2015 elections. Four issues played a crucial role during the debate and served as key input for the subsequent scholarly workshop:
First, many participants condemned the prioritisation of stability over conflict transformation after the end of the civil war. According to this view, the resulting culture of perpetuated impunity has undermined the process of conflict transformation. Second, practitioners underscored the importance of economic development as a prerequisite for sustainable peace (critically, see Stern and Ojendal 2010). In that context, the largely ignored fate of youth and (returned) refugees was also problematised. Third, the nature of cooperation in times of crises was reviewed. The current turmoil in Burundi has led to a partial suspension and readjustment of bilateral and multilateral aid. A number of Western donor countries have suspended election funding and security cooperation as well as announced cuts to development aid. All participants stressed the need for engagement below the level of the state during the current crisis, but also warned against unrealistic expectations as regards the chances of influencing the domestic dynamics of contestation from “the outside” through aid conditionality. Some contributors also noted the politicisation of – the sometimes seemingly technical and even apolitical – current development cooperation. Fourth, all panellists emphasised the need for a regionally and locally driven solution to the conflict, which should build on the political will of key Burundian actors and mediation by regional actors. Yet, whether such a solution should include President Nkurunziza, who secured his re-election in violation of the constitutional term limits, remained highly controversial.
Domestic Dynamics of Contention and External Attempts to Address the Crisis
Based on these discussions, the presenters addressed the workshop's leading theme: the interplay between external attempts to address the unfolding crisis in Burundi and the domestic dynamics of contention. As outlined above, research on Burundi has explored the diverse unintended consequences of the regionally and internationally promoted post-war order. This points to two key issues: First, one needs to understand the ways in which international as well as regional peacebuilding and post-conflict reconstruction contributed to the development of a fragile and frequently contradictory post-war order. Insights on these attempts' “inherent limitations in creating any meaningful transformation of the political space” (Daley 2007: 333) can also help us understand the interplay between various forms of external intervention and domestic conflict dynamics during the present unrest.
These underlying issues were addressed in two panels. The first panel tackled the (local) civil war legacies and the resurgence of conflict, violence, and mistrust. Despite negative repercussions of the protracted civil war and, even more importantly, the subsequent lack of transitional justice, problematic trends were obscured by what Susanna Campbell described as the “need for success.” According to her, in the wake of the “unexpected success of Arusha,” the international community, and especially Western donors, ignored negative patterns that became visible from 2006 onwards. This construction of Burundi as a success case – to which, in the view of Stef Vandeginste, academia also contributed – has subsequently reduced the international community's leverage vis-à-vis the Burundian government.
Yet, this willingness to ignore emerging problems was also backed by symbolically relevant advances. In her analysis of the Burundian army, Nina Wilén showed how the integration of Hutu rebels into the armed forces, which used to be one of the key symbols of Tutsi minority rule before the Arusha Agreement, was indeed a success story: Current developments (the abortive coup d'état on 13 May 2015 notwithstanding), such as the transformation of the ex-Forces Armées Burundaises (FAB) and rebel movements into an integrated army 6 and that army's progressive professionalisation, have constituted a remarkable accomplishment. Together with the peaceful conduct of elections in 2005 and 2010, these developments help explain why the regime's increasingly authoritarian nature has been “overlooked.”
Currently, Burundi contributes a significant number of troops to two missions, the United Nations Multidimensional Integrated Stabilisation Mission in the Central African Republic (MINUSCA) and the African Union Mission to Somalia (AMISOM). Moreover, smaller police contingents form part of the UNOCI (United Nations Operation in Côte d'Ivoire), UNISFA (United Nations Interim Security Force for Abyei), UNAMID (United Nations–African Union Hybrid Mission in Darfur), and MINUSTAH (United Nations Stabilisation Mission in Haiti).
The tendency to turn a blind eye to problematic post-war issues also became visible with respect to two further issues – namely, the return of refugees and the development of (former) rebel movements. Before 2015, the international community and the Burundian government hailed the repatriation of large numbers of refugees, even though many of them complained about broken promises, as Andrea Purdeková argued. Accordingly, she did not interpret the current refugee crisis as an ad hoc response to growing insecurity in Burundi, but stressed how the narratives of those now allegedly “fleeing peace” are profoundly political and an indication of the flawed process of post-conflict resettlement. This process, whereby the stabilisation through power sharing at the macro-level has obscured more nuanced micro-dynamics, also applies to everyday life in former rebel strongholds. In Bujumbura Rural, the pro-Hutu rebel movement FNL (Forces Nationales de Libération) remained “relevant in the post-ethnic era.” In his contribution, included in this special section, Tomas Van Acker explores how an international discourse that had vocally condemned the FNL for its human rights violations during the civil war has tended to conceal the movement's persistent legitimacy and charisma in this part of Burundi (Van Acker 2016). In a nutshell, various actors' determination to make Burundi a peacebuilding success story has obstructed the international community's view of problematic post-conflict developments.
The second panel focused on macro-level conflict resolution and post-war stabilisation efforts. The presenters examined how approaches such as the security sector reform, power sharing, and term limits did not fail, but have fallen short of adequately addressing political cleavages during the current crisis, which differ from the civil wars' major fault lines. In that vein, Filip Reyntjens (in this issue) characterised the 2015 political crisis as a “failure of democracy, not as a failure of constitutional engineering” (Reyntjens 2016). Consociationalism has contributed to overcoming the country's ethnic divide, and people have generally resisted certain politicians' attempts to infuse ethnicity into the current struggle (see also Sterck 2015). In Reyntjen's view, this is an astonishing evolution, considering that ethnicity once constituted the country's most lethal cleavage.
However, other key components of the Arusha Agreement, most importantly the term limits, have failed to serve their potential function as safeguards for democracy during the present crisis (Vandeginste 2015, 2016 [in this issue]). Similar to other problematic tendencies described above, and as emphasised by Stef Vandeginste, the question of a third mandate did not suddenly arise prior to the 2015 elections. As early as 2012, prominent members of the ruling party, the CNDD-FDD (Conseil National Pour la Défense de la Démocratie – Forces pour la Défense de la Démocratie), stressed that Nkurunziza's re-election would be considered as a violation of the Constitution. This was followed by unsuccessful attempts to replace or amend the Constitution in 2013 and 2014 – in other words, long before President Nkurunziza's announcement that he would run for a third term in office in 2015 attracted broad international attention. Moreover, the dispute about Nkurunziza's third mandate is deeply engrained in Burundi's (post-)conflict trajectory: as the term limits were laid down in the Arusha Agreement, opponents of the president's re-election regarded this as an attack on the peace treaty that was instrumental for establishing Burundi's post-conflict order.
Along a similar track, Gérard Birantamije examined how the security sector reform, which has constituted a cornerstone of international peacebuilding since 2000, has failed to prevent a renewed politicisation of the armed forces. Ostensible advances such as the appointment of a civilian minister of national defence in May 2015 were a smokescreen for underlying difficulties. Most importantly, the reintegration component of the DDR (disarmament, demobilisation, and reintegration) programme constituted “just a letter” in his view, and it has never been fully implemented. Accordingly, over the course of the current crisis it has been easy for the government to remobilise the – on paper demobilised – former rebels to carry out violent attacks on the opposition. The importance of Burundi as a troop contributor to the African Mission in Somalia (AMISOM) has further reduced the African Union's willingness to criticise such shortcomings, showing once again how external actors have become entrapped in their focus on “success.”
Throughout the second panel, the role of regional actors and (sub) regional organisations in addressing the current crisis was debated. Policymakers present during the discussion stressed the need for regional mediation, but several presentations pointed to vocal disagreement amongst regional actors, which has impeded effective regional responses thus far – most importantly with respect to the term limits debate. A similar degree of regional disunity already existed during the 1990s, as I described in a presentation on lessons learned from regional sanctions against Burundi during the civil war. In summary, the second panel revealed how macro-level peacebuilding approaches that were instrumental in dealing with some of the civil war's root causes, particularly the ethnic divide, have proven inadequate to address shifted political cleavages and micro-level insecurities during the current crisis.
Conclusion: General Lessons and Particular Context
These insights into the interplay between international and regional peacebuilding efforts as well as their (unintended) repercussions for the current crisis raise the question of in what respects Burundi is an example of the challenges that come along with liberal peacebuilding. The international community's construction of a success case narrative appears to constitute an important, still under-researched phenomenon beyond the case of Burundi. The workshop's particular setting, bringing together scholars and policymakers, allowed for reflection on the complicity of diplomacy, NGOs, and academia during this process, but further research along these lines is necessary (see also Bliesemann de Guevara 2014 on the related issue of knowledge production in conflicts).
In addition, the apparent failure of power sharing to sustain the process of democratisation and the willingness of various actors to turn a blind eye to the regime's increasingly authoritarian nature because consociationalism appeared to work sheds further light on this increasingly popular instrument (for a summary, see Mehler 2009). Power sharing has been used as an instrument both to end violent conflict and to achieve stable democracy (Binningsbø 2013), but the trajectory of Burundi suggests that it has been more effective with respect to the former goal. Furthermore, the tendency to focus on high-level politics while ignoring local and societal concerns when assessing power sharing (Simons et al. 2013) has proven highly problematic in places not limited to Burundi.
Nevertheless, the Burundian post-conflict trajectory and relapse into unrest is also characterised by notable particularities. To provide just one example discussed during the workshop, the fact that the regime was able to present a legal as opposed to a political argument for its hold on power during the term limit debate in Burundi is unusual in the sub-Saharan African context. Other countries' presidents who have sought or are currently seeking a third term in violation of their states' constitutions have tended to justify their move by citing political necessities or by claiming, truthfully or not, that the population wishes to re-elect them (Simons and Tull 2015). By contrast, Nkurunziza refers to a law-based position, whereas his opponents are forced to make a political case for democratic renewal and the need to preserve the spirit of Arusha (Vandeginste 2015, 2016). Even though this constitutes a Burundian specificity, which is important to understand in itself, it is also illustrative of a more general tendency of regional and international actors to quickly denounce visible violations of a constitution, but position themselves more cautiously vis-à-vis other types of democratic misconduct that are more difficult to measure (see also von Soest and Wahman 2015).
The debates in Freiburg underscored the need for historically contextualised single-case studies, both to contribute to the better understanding of unique developments and to use as a springboard for further theorising about external intervention into fragile post-war settings and its (unintended) repercussions. However, the discussions also showed that even case-based research more often than not contains an element of (implicit) comparison (Zanker and Newbery 2013). The following contributions by Reyntjens, Vandeginste, and Van Acker are revised versions of their presentations in Freiburg.