Last week, the International Crisis Group (ICG) published a new report on the eastern DRC in which it zooms in on a conflict over customary power in the Ruzizi Plain (Uvira territory, South Kivu).  Apparently, ICG no longer tries to weigh in on broader diplomatic, political and military developments in the DRC, like the recently signed Framework Agreement and the current army restructuring process, but has decided to focus uniquely on conflict dynamics at the micro level. This choice reflects ICG’s understanding of the “root causes” of violence in the eastern DRC, which it describes as “local competition between communities for land and economic opportunities” (p.i).  As the report concludes: “far from the national and regional preoccupations of the national government and the United Nations, ethnic spaces are being redrawn by violence” (p.24). In order to prove this point, the report subsequently presents an “anatomy” of the conflict around the customary chiefdom of the Ruzizi Plain, portraying it as primarily resulting from ethnic tensions between the (“immigrant”) Barundi and the (“autochthon”) Bafuliro that have existed since the colonial era.

However, this portrayal greatly simplifies the complex and multi-layered causes of conflict and violence in the Ruzizi Plain and the adjacent Moyens Plateaux of Uvira. Not only has the majority of violence in this area not been directly related to the conflict around the chiefdom or ethnic tensions, it has crucial regional and national dimensions that the report glosses over. It achieves this reductionist interpretation through a highly selective and at times inaccurate presentation of the events. Given the importance of (as ICG puts it) “investing in knowledge before acting” (p.20), the inadequate analysis provided in the report is problematic. Based on extensive field research and the ongoing monitoring of developments in Uvira since 2010 for academic purposes, I will highlight three main interrelated weaknesses of ICG’s analysis.

1. The ICG report confounds “conflicts” and “violence”

It is undeniable that the assassination of the mwami (paramount chief) of the Barundi in April 2012 and the enthronization of his successor have triggered instability and led to a series of violent demonstrations and tit-for-tat massacres in the Ruzizi Plain. However, the far-out majority of violence in the Plain and the adjacent Moyens Plateaux over the past years has not been directly related to this or other ethnic conflicts. Rather, it is the result of the ongoing activities of the Burundian rebel group FNL, various Fuliro-led Mai Mai and self-defence groups, the Congolese army (FARDC), the Rwandan-led FDLR, groups of demobilized from the Congo Wars, and local bandits.

While some of these armed actors have occasionally drawn upon ethnic narratives for the purposes of mobilization, and have been involved in conflicts around local governance, an analysis of their practices and alliances (in Uvira) learns that these are by no means uniquely informed by ethnic motives or customary and land conflicts.  Rather, they seem to a large extent driven by military competition and geared towards violent accumulation. Their peaks in activity have not been directly related to customary, ethnic or land conflicts. In 2009, the Kimia II operations pushed the FDLR, which used to control most of the Moyens Plateaux of Lemera and had strong influence in the Ruzizi Plain, towards the Itombwe forest[1], causing major upheaval. Another upsurge in violence could be detected in 2011, when three Fuliro army and police deserters-the Colonels Nyerere Bunana and Baleke Sumaili and Major Bede Rusagara-re-entered the maquis with the encouragement of businessmen and politicians from Uvira, capitalizing upon the voids left by the FARDC regimentation process.  In 2012 and 2013, the operations of the Congolese and Burundian army (FDN) against the FNL, which has been one of the main authors of violent attacks and banditry in the Ruzizi Plain, trigged important instability, leading to massive displacement and the permanent occupation of Kiliba/ONDS (in the Plain) by the FDN.

Again, this violence did not result from the customary conflict in the Ruzizi Plain or inter-ethnic tensions. Clearly, there is a need to disaggregate the variables “violence” and “conflict”, which  ICG seems to confound. Conflicts around local governance, land and economic opportunities do not automatically lead to violence, nor is all violence in Uvira necessarily related to such conflicts. Furthermore, the relation between violence and the customary conflict in the Plain appears to be the opposite of what ICG suggests: were it not for the context of ongoing violence and the presence of militia, whose members were directly involved in the assassination of the mwami and the subsequent escalation, one can wonder to what extent the conflict around the chiefdom would have led to violent incidents in the first place.

2. The ICG report downplays the crucial regional and national dimensions of the violence in the Ruzizi Plain and the Moyens Plateaux of Uvira

The important role played by foreign armed groups and militaries in the production of violence in the Ruzizi Plain, as highlighted above, reveals that this violence has a distinctly regional dimension. This is all the more so since nearly all Fuliro-led armed groups in Uvira have ties to Burundian armed factions, while profiting from the trans-border trafficking networks that the ICG report rightly identifies as a key source of instability. In particular the group of Bede Rusagara, an ex-CNDP FARDC deserter, has a pronounced regional character. As a key ally of M23 in South Kivu, it has extensively profited from recruitment and financing networks in Rwanda, Burundi and Uganda (S/2012/843:21-23).[2]

But there are also important national dimensions to armed group activity in Uvira. The development of the self-defence militia of the mwami of the Bafuliro (FALL), together with Bede’s group centrally implicated in the unrest around the chiefdom, is telling in this respect. Although it later evolved into mostly a power tool of the mwami, the FALL’s creation in 2009 was predominantly a reaction to the insecurity caused by the Kimia II operations and the failure of the FARDC to protect the population against retaliatory attacks by the FDLR.[3] Furthermore, the FARDC has not only tolerated the FALL, it has also provided it with ammunitions and carried out joint operations (S/2011/738, p.73), thus playing a crucial role in its development.

The ICG report’s failure to acknowledge the key role of the FARDC in the ongoing violence in Uvira is also manifested in its neglect of the army’s contribution to the escalation of the conflict around the chiefdom of the Ruzizi Plain. In particular the FARDC’s perceived lack of neutrality has aggravated the conflict in important ways, as it has created incentives for the protagonists to take recourse to self-defence militia.[4]  Established business interests and patronage alliances have further undermined the credibility of the FARDC’s interventions, including that of the Commander of the Luberizi base, who owns at least 15 hectares of land in the Plain, and the officers of the 111th regiment involved in the (competition over the) various trans-boundary trafficking networks.[5] Yet, the ICG report is all but silent on the role of the FARDC, locating agency almost exclusively in “local actors”.

This reading of the events is all the more remarkable as the report explicitly recognizes the catalysing effects of the 2011 national elections on the conflict, even describing the latter as “collateral damage” of the elections (p.11). While it is an open question in how far the reinstallation of the mwami of the Barundi results from an electoral strategy of Joseph Kabila, as the report suggests (p.13),[6] it is certain that pre-and post-electoral tensions have significantly contributed to the escalation of the conflict, not least through the intensified financing of armed groups and militia by electoral candidates both before and after the elections.[7] Interestingly, the report does not mention that the mwami of the Bafuliro, Ndare Simba, ran himself for the 2011 legislative elections, but failed to get elected. Some of my sources have identified this electoral disappointment as an important cause of his strategy to escalate the conflict with the Barundi, thus hoping to maintain his power in the face of dwindling legitimacy.[8] In general, while the report does mention intra-Fuliro tensions, it fails to recognize the extent to which these have led to “ethnic outbidding” as partly triggered by national political events and processes in both past and present.

3. The ICG report obliterates the socially constructed nature of ethnicity and overstates its importance

In its presentation of “the historical roots” of the current conflict, the report paints an inaccurate picture of a Fuliro people who have clashed since the colonial era with the “immigrant” Barundi (footnote 16, p.4). Not only is it questionable in how far a population that was present on the Kivus’ soil before the delineation of international boundaries can be labelled “immigrants”, this presentation of facts is based on a misreading of the complex roots of the Fuliro people and its chiefdom. Much like the Barundi, the Bafuliro were not a coherent “tribe” before the arrival of the colonizers, but are the product of a long process of the regrouping, territorialization and assimilation of various relatively mobile groups (including those coming from present-day Burundi)[9] and clans, and the imposition of a paramount ruler among various (quasi) autonomous chiefs (Muchukiwa 2006: 14-18, 90; Depelchin 1974: 48, 50). Therefore, the creation of the enlarged Chiefdom of the Bafuliro was a process that provoked almost as much resistance as the creation of the enlarged Chiefdom of the Barundi (Muchukiwa 2006: 83-90).  This makes it difficult to speak of a coherent opposition of “the Bafuliro” against “the Barundi” in the colonial era, as suggested by the ICG report[10] (p.4), specifically as it also occurred that certain Barundi sided with Fuliro chiefs in their fight against other Fuliro chiefs (Depelchin 1974: 93-94).

It was only in the immediate post-independence period that a more explicit and politicized Bafuliro/Barundi antagonism materialized under the influence of newly unleashed power competition, electoral pressures, administrative reorganization, and the Simba rebellion. However, it would be reductionist to suggest that the 1961 unrest in Uvira, and the Simba revolution that unfolded from 1963 onwards, were primarily related to ethnic-based tensions, although these certainly played an important role.[11] As pointed out by Verhaegen (1966: 273-275), these events were also the product of agitation against customary chiefs  seen as complicit  with the colonial authorities and their policies of taxation, forced labour, and land expropriation against the backdrop of increasing economic hardship due to a stark decline in cotton prices.

The antagonisms between Barundi and Bafuliro were reinforced in the 1980s, not least through the creation of the cités (urban administrative entities) of Sange and Kiliba, which entailed the amputation of around three quarters of the territory of the Chiefdom of the Ruzizi Plain and led to an important loss of its revenues (Muchukiwa 2006: 173).  In the 1990s, the tensions flared up under the influence of the announced transition to a multi-party democracy, which triggered a political competition that became primarily channelled along ethnic lines, in part as a result of conscious manipulation by Mobutu (Lanotte 2003: 29). In the Ruzizi Plain, this was manifested in popular manifestations provoked by a contentious census of Rwandophone populations, forcing the Barundi chief to flee in 1991.[12] The two subsequent Congo Wars militarized the conflict as one side became linked to the RCD and the other to the Mai Mai, further politicizing ethnic identities.

While these events show the long lineage of the Barundi/Bafuliro conflict in the Ruzizi Plain, they also demonstrate that it have often been national political developments, administrative reorganization and electoral pressures that have rendered ethnicity politically salient, and not only local competition around land and economic opportunities. Furthermore, a nuanced historical analysis should also uncover the non-ethnic dimensions of these conflicts.  As pointed out by Depelchin (1974:98-100), who emphasizes that the social fabric of the Barundi and the Bafuliro is closely interwoven through inter-marriages, cattle exchange and patronage ties, intra-ethnic conflict has historically been as important as inter-ethnic conflict in the Ruzizi Plain. This observation still holds today, and many of the socio-economic factors ICG correctly identifies as nourishing tensions in Uvira, including pressures on land and the re-opening of the Sucrerie of Kiliba (p.8), by no means feed conflict that falls exclusively along Barundi/Bafuliro fault lines.  This also applies to what is at present a crucial source of instability in Uvira: the negotiations surrounding the integration into the FARDC of a plethora of Fuliro warlords, which foster fierce power struggles between the political-economic networks that these military leaders are tied into. Again, it is intra-ethnic competition, fuelled by incentives stemming from national policies (in this case army integration) that produces unrest, and not “a will to local power by ethnic groups” (p.10).


Rather than being based on a nuanced analysis that recognizes the multi-dimensionality  and complexity of conflict dynamics and the production of violence in the eastern DRC, ICG’s presentation of the events in the Ruzizi Plain and the Moyens Plateaux of Uvira is moulded around a single narrative-that of “local ethnic and land conflicts” as the “root causes of violence”.  Certainly, ICG’s insight that a grasp on micro-dynamics is crucial for understanding conflict and violence in the eastern DRC is valuable, and it is right to point to the conflict-generating nature of local (land) governance, as partly stemming from the institution of customary chiefdom. Yet, this micro-focus becomes problematic if it leads to the downplaying of a range of crucial interlocking supra-local factors that are especially of importance for understanding when and why conflicts translate into violence.

It is difficult to see how without an end to manipulation by national and provincial politicians and businessmen, army reform, and the demobilization of the multitude of competing Fuliro armed groups (which ICG does mention as solutions-but of tertiary importance), and perhaps most importantly, a halt to the activities of the FNL, peace can be restored in the Ruzizi Plain. Furthermore, one wonders in how far the primary solutions proposed by ICG-a reorganization of local governance, the reigning in of the power of customary chiefs, and land management reforms (which are all “national policies” that require state-building for their implementation, even though ICG labels them as “local solutions for local conflicts” (p.19)), will in the current climate of ongoing armed group activity not provoke more, rather than less violence.  This prioritization of policy solutions appears to rest upon an altogether feeble analysis of the instability in Uvira, making it doubtful in how far ICG itself lives up to the imperative of “improving knowledge of the local context” (p.20) that it so vigorously advocates.


ADEPAE and SVH (2011) Congolese Refugees From South Kivu : Challenges of return in the territories of Fizi and Uvira,

CIRESKI (2012) Etude analytique sur la milice FALL, unpublished document

Depelchin, J. (1974) From Pre-Capitalism to Imperialism: A History of Social and Economic Formations in Eastern Zaire (Uvira Zone, c.1800-1965). Stanford University, PhD Dissertation

Lanotte, O. (2003). République Démocratique du Congo.Guerres sans frontières. De Joseph-Désiré Mobutu à Joseph Kabila. Brussels: GRIP and Éditions Complexe.

Muchukiwa, B. (2006). Territoires ethniques et territoires étatiques. Pouvoirs locaux et conflits interethniques au Sud Kivu (R.D.Congo). Paris: L’Harmattan.

UN Security Council (2012), S/2012/348,  Interim Report of the Group of Experts on the Democratic Republic of Congo

UN Security Council (2012). S/2012/843, Final Report of the Group of Experts on the Democratic Republic of Congo

UN Security Council (2011). S/2011/738, Final Report of the Group of Experts on the Democratic Republic of Congo

Verhaegen, B. (1966). Rébellions au Congo. Tome 1. Brussels and Leopoldville: C.R.I.S.P., I.R.E.S. and I.N.E.P.

[1] For that reason, the FDLR no longer controls mines in the Moyens Plateaux as the report erroneously states (p.9), this is also not mentioned in §96 of S/2012/843, as is claimed in footnote 56.

[2] This complicates dominant readings of the M23 as being uniquely linked to “Rwandophone” interests, since the Mufuliro Bede has been a key defender of the “autochthon” Fuliro side in the conflict around the chiefdom. At the same time, it has (loosely) collaborating with the “Rwandophone” Banyamulenge forces of Nkingi Muhima (S/2012/843:22).

[3] The creation of the FALL is therefore not related to the 2011 FARDC regimentation process, as the report states (p.12). Furthermore, its members are not composed in majority of ex-Mai Mai Zabuloni (although they contain some), but of ex-Mai Mai Kayamba, Abdou and Ruarara, which were the main Mai Mai leaders in this part of Uvira during the Second War. Findings based on fieldwork in the Chiefdom of the Bafuliro in March and May 2010.

[4] The FARDC’s lack of perceived neutrality is to a large extent related to internal power struggles channelled along ethnic fault lines, which have been strongly exacerbated by CNDP and PARECO (dis)integration. In this case, it is also related to the power struggles unleashed by the departure of Col. Bernard Byamungu, the previous commander of the military sector of Uvira, who participated in the ex-CNDP orchestrated mutiny that started in this territory in March 2012 (S/2012/348: 18-20).

[5] Conclusions based on fieldwork in Luberizi and Sange, November 2011.

[6] My own sources cannot confirm or refute this reading. However, it seems implausible Kabila would rely for electoral support on a group (the Barundi) that has almost no electoral weight due to its clear minority status (20%).

[7] This includes for instance Mwami Ndare Simba’s use of the FALL for electoral campaigning, as documented by the NGO CIRESKI (2012), and the activities of failed candidate Emmanuel Ndigaya Ngezia (aka La Fantaisie), who has tried to capitalize upon the announced integration of Bafuliro and other Mai Mai leaders by attempting to create a unified political platform for these groups.

[8] Personal communication with members of civil society organizations based in Uvira, November 2012.

[9] For instance, the Bazige/Bahungu (Muchukiwa 2006:17-18).

[10] There have surely been important armed clashes between certain Barundi and Bafuliro chiefs (e.g. Mokogabwe and Lubisha) in the colonial era. These took mostly place in 1920/1921 and not in 1929 as the report suggests (p.4) (Muchukiwa 2006:153).

[11] Remarkably, the ICG report fails to mention that the anti-Barundi rhetoric of the radical provincial MP Musa Marundura was partly a strategy to cling to the power he had obtained in 1961, after Mwami Henri Simba of the Bafuliro had fled the political unrest (Verhaegen 1966: 269-270).Perhaps this omission is related to the fact that the ICG report refers to a Verhaegen volume on the Mulele rebellion in Kwilu (Bandundu) (footnote 19), and not to his classic work on the Simba rebellion in Uvira (Verhaegen 1966).

[12] It was only under the reign of the AFDL in 1996 that the mwami of the Barundi could return (see also ADEPAE & SVH 2011:41). The ICG report glosses over these developments by erroneously stating the Barundi chief was dismissed from power by the AFDL in 1996 (p.5).

Judith Verweijen is Senior Researcher at the Nordic Africa Institute, Uppsala.