Why Violence in the South Kivu Highlands Is Not ‘Ethnic’ (And Other Misconceptions About the Crisis)

Students from a school near Minembwe, June 2007. (Photo Julien Harneis)

Judith Verweijen is a lecturer at the Department of Politics & International Relations at the University of Sheffield (UK). Her research looks at the interplay of armed mobilization, violence and conflicts around natural resources. Her main focus is on eastern DRC, where she has conducted extensive field research since 2010.

On August 10, 2020, the United Nations Joint Human Rights Office in the Democratic Republic of Congo (UNJHRO) published a report on the unfolding crisis on the highlands (Hauts Plateaux) at the intersection of Fizi, Uvira, and Mwenga territories in South Kivu. Curiously, the report focuses on a limited area of the highlands only: it omits the Bijombo area, where fighting first escalated mid-2018, after three years of conflict. 

Despite this omission, the report gives an indication of the enormous toll that the crisis has taken: it documents the destruction of at least 95 villages, 128 deaths from summary and extrajudicial executions,  47 victims of sexual-related violence, and the looting and killing of thousands of cattle. This violence has led to a dire humanitarian situation, with over 110,000 people displaced. 

The UNJHRO report does not provide much analysis of the drivers of this violence. It acknowledges that the conflict and its origins stem from multiple factors at the national and sub-regional level, but is limited to the inter-community aspect. This is unfortunate, as it advertently gives off the impression that this is the most important factor. 

Similar to other conflicts in eastern DRC, the crisis on the Plateaux is characterized by profound complexity. It involves a range of different drivers of conflict and violence that play out at different levels–from the local to the sub-regional. Narratives emphasizing simple explanations provide only one piece of this complex puzzle. Here are three such narratives and why on their own they are incomplete, if not inaccurate. 

1. The violence on the Hauts Plateaux stems from “ethnic” or “inter-communal” conflict

Ethnic identity has featured prominently in explanations for the recent violence. From this perspective, it stems from animosities between the Banyamulenge on the one hand, and groups that label themselves “autochthonous”–including the Babembe, Banyindu, Bafuliiru, and Bavira on the other. 

This narrative needs to be nuanced. There are certainly numerous conflicts on the Plateaux that pit Banyamulenge against other groups. These conflicts relate to contestations around local authority and control over land and resources, including the taxation and regulation of markets, mines, and cattle movements.

However, these conflicts do not always turn into armed violence. Violence is committed first and foremost by armed groups and “local defense” militias. True, these armed actors claim to defend particular ethnic communities and are often supported by members of these communities seeking protection. Yet most ordinary citizens are not involved in planning, organizing, directing, inciting, or committing violence. We therefore cannot ascribe such violence to “ethnic groups” writ large. More importantly, we must pinpoint and analyze when, why, and how conflicts turn violent. As an extensive body of research shows, violence labelled as “ethnic” is often driven by a range of other motives and objectives, including inter-personal conflict, economic and political competition and disputes about land and other property. 

Another problem with the “ethnic conflict” narrative is that it assumes there are two homogeneous blocks: the Banyamulenge and groups calling themselves “autochthonous.” Yet these groups themselves have numerous internal divisions, which are reflected in the plethora of armed groups linked to either side. 

There are at least three Banyamulenge armed groups: the Twirwaneho–a coalition of local militias, which are also developing a political branch; the Gumino, led by Shaka Nyamusharaba; and an armed group commanded by FARDC deserter Michel Rukunda, aka “Makanika,” which has numerous Banyamulenge youth from the regional diaspora (Kenya, Rwanda, Burundi) in its ranks. The armed groups linked to the Babembe, Bafuliiru, and Banyindu are even more numerous. They include the Mai-Mai of Ebuela Mtetezi–which encompasses Bembe commanders who previously had their own groups, such as Aoci and Ngyalabato; the Mai Mai Mulumba; the Mai Mai “Mupekenya” under the command of Kati Malisawa, and a range of mostly Fuliiru and Nyindu groups operating under the label “Biloze Bishambuke.” The latter include the groups of Ilunga, Kashomba, Mushombe, and in the Minembwe area, those led by Luhala Kasororo and Assani Malkiya. 

These armed groups operate in broad coalitions, but there are regularly tensions and occasionally even clashes between groups that are supposedly on the same side. For instance, on August 2, the Biloze Bishambuke under the command of Ilunga clashed with the troops of  Kati Malisawa near Maheta village, allegedly due to a dispute over stolen cattle. This indicates that certain armed group leaders, and the political actors that help mobilize and support them, also have different agendas than protecting their communities. They often aspire to enhancing their own political and economic clout and some have national political aspirations. This further undermines the argument that the violence is primarily driven by “ethnic conflict.” 

2. The violence on the Hauts Plateaux is related to the creation of the commune rurale of Minembwe 

Another popular explanation for the violence, which is closely linked to the ethnic conflict narrative, is that it stems from the creation of the “rural commune” of Minembwe– a non-customary decentralized local governance entity. The commune became operational at the start of 2019, following decrees issued in 2013 and 2018, and the nomination of its leaders in February 2019. 

The commune is undoubtedly a source of conflict. It is located in Fizi territory, on lands that members of the Babembe community consider to be theirs. They therefore see the creation of the commune as encroachment on or the occupation of their ancestral grounds. Some have also contested the designation of the mayor, who is  Munyamulenge. But most importantly, the creation of the commune is seen as the first step for the resurrection of the territoire (sub-provincial administrative entity) of Minembwe.

During the Second Congo War, the rebel administration of the Rwanda-backed Rassemblement congolais pour la démocratie (Congolese rally for Democracy or RCD) created  Minembwe territory, which encompassed large parts of the Hauts Plateaux and the adjacent Moyens Plateaux. The territory fulfilled a longstanding wish among the Banyamulenge, whom the colonial authorities had denied a chiefdom or groupement–local governance entities generally formed along ethnic lines. Consequently, they became subject to rule by customary chiefs from other communities. The territory, where they dominated the administration, resolved this. Moreover, anticipating future elections, the territory, which is an electoral district, would have allowed the Banyamulenge to increase their political representation in parliament. Being a minority in each of the three territories that comprise the Hauts Plateaux, they had struggled to get their candidates elected. Finally, the territory brought local administration closer to the people in this isolated region, allowing them to obtain birth certificates and other official documents there. 

The creation of the territory–which was formally abolished in 2007– was heavily contested by other groups, who perceived it to break up their ancestral lands. In addition, it appeared to confirm a conspiracy theory that the Banyamulenge were the vanguard of a foreign invasion attempting to expropriate and displace “autochthonous” groups, and usurp their local authority. Members of these groups therefore have bad memories of Minembwe territory. In addition, it has left a legacy of leadership conflicts. Numerous former appointees have continued to behave as de facto local authorities, even if they no longer hold an official position. 

For these reasons, Minembwe territory has a highly symbolic function, as a marker of division and violence. The commune rurale evokes similar strong feelings, being deeply inscribed in struggles around local authority and identity. It has also become a national political affair. While prominent Banyamulenge leaders––including Azarias Ruberwa, currently the minister of decentralization–– endorse the commune, many Bembe, Fuliiru and Nyindu politicians, such as Pardonne Kaliba, have denounced it. The commune has also stirred heated debate among Congolese in the diaspora. 

Yet the violence on the Hauts Plateaux, as well as the emergence of most of the armed groups involved in the current fighting, predates the creation of the commune. Violence on the Plateaux has been a regular occurrence since 1996. The current cycle started in 2016 and escalated mid-2018. This escalation first occurred in Bijombo groupement. This groupement is not included in the commune rurale, whose surface is many times smaller than the (abolished) territory of Minembwe. Bijombo also has distinct conflict dynamics. The latter revolve to a large extent around the position of the chef de groupement–for which there are multiple contenders linked to different ethnic groups. Another site of significant violence is the Itombwe area, which is equally not included in the commune. 

In sum, even though it is an important source of conflict and figures prominently in the belligerents’ discourses, the commune is only one of many factors in the current fighting. It does not explain why and when armed groups emerged on the Plateaux and who they target with their violence. 

3. The violence on the Hauts Plateaux is the result of foreign interference

As documented by, inter alia, the UN Group of Experts on the DRC and Radio France Internationale, the coalitions of belligerents fighting on the Plateaux include foreign armed groups, notably the Burundian groups Résistance pour un état de droit au Burundi (RED-Tabara) and Forces nationales de libération (FNL), as well as the Rwanda National Congress (RNC). They have occasionally also included soldiers linked to the governments of neighbouring countries, which have moreover hosted recruitment and supply networks. At the same time, eastern DRC has a history of wars kickstarted by foreign interference. The conclusion that the trouble on the Plateaux results from foreign meddling is therefore easily drawn. 

Yet this explanation glosses over the many conflicts around local authority mentioned above. It also overlooks the role of provincial, national and diaspora political actors in supporting armed mobilization and polarization. Moreover, the language of “foreign interference” is somewhat misleading. It suggests that all power resides on the side of foreign forces, who manipulate Congolese intermediaries as they see fit.

This reading obscures that Congolese armed group leaders and political actors have significant room for maneuver regarding what foreign forces they ally with. Occasional shifts in such alliances testify to this leeway. These shifts also demonstrate that such alliances are mutually beneficial. Through their foreign allies, Congolese groups gain in military capabilities, for instance, by acquiring heavy weaponry. This, in turn, allows these groups to better advance their position within conflicts around local authority and access to resources. As such, the involvement of foreign actors cannot be seen in isolation from local dynamics of conflict and violence; they are interwoven and mutually reinforcing. That said, foreign interference has clearly contributed to a significant escalation of the violence, even if it has not caused it. 

What, then, is at the root of this terrible violence? There are a number of interlocking mechanisms at work. First, the narrative of “ethnic violence” has become a self-fulfilling prophecy: all types of conflicts and incidents of violence are seen primarily through an ethnic lens, even if there are also other factors at work. This activates a second mechanism, which is the attribution of collective responsibility for individual acts of violence. As a result, civilians are attacked in retaliation for violence committed by armed groups. This blurring of the boundaries between armed groups and civilians is an important driver of cycles of revenge violence. Widespread impunity has further aggravated this: as individual perpetrators are not held to account, the blame is shifted towards groups as a whole. 

Another key mechanism is militarization, or the tendency by local leaders and politico-military elites to resort to force in order to gain the day in disputes and power struggles. This does not only involve politicians, business people and military leaders in the DRC, but also governmental actors and other elites at the level of the Great Lakes Region. 

The emergence and persistence of armed groups, however, is not only the result of militarization: it also stems from local security dilemmas related to mutual distrust between communities. The presence of armed groups seen to defend particular ethnic communities prompts members from other communities to equally sustain armed groups. The same logic propels these armed groups to maintain a military balance of power, which motivates attacks to weaken the enemy. Local security dilemmas crucially hinge upon a generalized lack of trust in the state security forces, which are accused of partiality by all sides. It is also rooted in a history of violence dating back to the Congo Wars, which has instilled bitter feelings and deep distrust between different groups. 

These various mechanisms play out at different levels and become mutually reinforcing. For instance, the involvement of foreign armed actors is in part the result of the strategies of politicians and military leaders operating at the national level. Once present, these foreign forces exacerbate local security dilemmas and conflicts around local authority and resources. In this manner, dynamics of conflict and violence at different levels become entwined. Monocausal explanations, such as the lazy trope of “ethnic violence” do no justice to this complexity. In fact, they may exacerbate the situation. They further essentialize identities and legitimize attributing responsibility for armed group violence to civilian communities. When describing violence in eastern DRC we must therefore try harder to find an adequate analytical language. 

Who’s Stealing the FARDC’s money?

On the afternoon of February 20, the green Congolese army jeep was speeding along National Road 2, between Goma and Rutshuru-center. In the jeep were seven FARDC soldiers and some precious cargo: the wages of the 3416th regiment. Over 100,000 USD in cash.

But as it approached the village of Rwaza, the convoy was brought to an abrupt halt. “The attackers had blocked the road and positioned themselves to stop the arrival of reinforcements,” explained a FARDC officer in his office at the 34th Military Region’s headquarters in Goma. “When the jeep arrived, any retreat was effectively blocked off. None of the passengers survived. It was the work of professionals.”

Is it conceivable that such an ambush could take place without collusion from within the army? An investigation has since been launched by the military prosecutor’s office. But the Sukola-2 operational sector spokesperson, Major Guillaume Ndjike, has already accused the Democratic Forces for the Liberation of Rwanda (FDLR-Foca). In his view, they are the main enemy forces operating in the area.

However, data analysis by the Kivu Security Tracker (KST) suggests that Rwaza has not been part of the area of operations of this group for at least six months. The areas surrounding the village have, however, been impacted by incidents involving the FARDC and the Nyatura FDP (Forces de défense du peuple) militia, who are allied with the FDLR.

The red star indicates where the ambush took place on February 20, and the colored dots show the incidents recorded by the KST over the previous six months.

Whatever the case may be, the attackers vanished into the Virunga National Park with their loot. All that remained was the soldiers’ resentment, deprived as they were of their meagre salary. This feeling of resentment was violently expressed in the hours and days that followed: FARDC soldiers retaliated indiscriminately, killing four civilians – all from the Hutu community – between February 20 and 23 according to information collected by the KST.

At the 34th Military Region, no one refutes that civilians were killed around the same time as the ambush. But, according to one officer, these were collateral victims of crossfire.

However, the incident is nevertheless a source of embarrassment. Especially as this is the second time in less than a month that soldiers have protested and called for what is rightfully theirs in Rutshuru territory (North Kivu). The last time, it was in Nyanzale on January 27, after soldiers of the 3407th regiment accused their commanding officer of misappropriating part of their food rations. Soldiers had “gone on strike,” shooting into the air and refusing to intervene during a kidnapping. This attitude so annoyed local inhabitants that a “ghost town” (ville morte) day was declared on January 28. Four people, suspected of being kidnappers, were lynched by the mob. However, the army also denies that there was any cause for concern there. “The commanding officer did not misappropriate anything,” explained Major Ndjike. “He simply took some welcome and practical steps: the frozen fish destined for soldiers was going to rot before it could be delivered to this remote area. He therefore decided to sell the fish in order to buy some meat in its place. Only one soldier complained and fired a shot at one of his friends. He was arrested.”

Such eye-opening events are in fact only the visible part of a much larger phenomenon: the chronic inability of the FARDC to pay a decent and regular wage to its soldiers. Often what is at stake are methods for misappropriation and fraud, where some of the cash destined to pay troops is misappropriated by commanders, or where commanders fraudulently claim troop numbers well in excess of reality, to embezzle the wages of fictional soldiers.

The use of such methods – and particularly inflating troop numbers – has been widespread since the FARDC was created in 2003. At the end of the Second Congo War, this army was created by bringing together the troops of several warring parties after the signing of the Sun City Agreement: the Congolese Armed Forces (FAC), the Congolese Rally for Democracy/Liberation Movement (RCD/ML) and National (RCD/N), the Movement for the Liberation of the Congo (MLC) and the Mai-Mai groups (in French). Each party found it advantageous to artificially increase the number of its combatants, to exaggerate its troop numbers – and its capacity to misappropriate. An assessment carried out by South Africa in 2004 estimated that between 30% to 50% of the new FARDC troops were fictitious.

Despite efforts to improve the system, including by deploying a biometric census and separating the payment chain from the command chain with the support of the European Union (EUSEC) from 2005, problems have remained. Successive waves of integrating rebels, sometimes in haste, such as the National Congress for the Defense of the People (CNDP) in 2009, have added to the ongoing confusion and inflation of troop numbers. In 2011, the regimentation process (the restructuring of the Congolese army from brigades into regiments) which aimed to remove fictitious troops, again revealed the scale of the problem. For example, in the province of South-Kivu, the official recorded troop count of 35,000 was re-evaluated at less than 16,000.

In 2012, the government of Prime Minister Matata Ponyo attempted to put an end to the misappropriation of wages by using the banking system to pay soldiers’ wages directly into a bank account rather than through their commanding officers.

Although this system, still used today, did prevent some embezzlement and improved the timeliness of payments, other problems remain. First, not all regiments have access to this system. Those which were created recently, or those based in remote areas far from any bank branches, can still opt to be paid in cash. This is the case for the 3416th Regiment, which explains that their wages, physically transported, were able to be stolen in an ambush.

Second, commanders may be tempted to hide their losses to keep control of the wages of dead (“ghost”) soldiers or deserters. “Using the banking system to pay soldiers’ wages directly into their bank accounts, which in theory prevents this practice, is sometimes subverted because commanders can fabricate powers of attorney to withdraw the money,” explains a source who for many years worked for one the banks used by the FARDC. “Widows regularly complain that their deceased husbands’ wages have been withdrawn without them knowing who made the withdrawal.”

Lastly, only wages are paid directly into bank accounts. The various bonuses are still paid in cash. And they quickly add up: operational funds, funds for provisioning during operations, command responsibility bonus, special intelligence funds, funds for conducting operations, funds for healthcare and funerals, funds for provisioning… In her thesis, published in 2015, the researcher Judith Verweijen listed seven different types of bonuses, which provide as many opportunities to embezzle. In total, these funds represent a greater amount of money than the wages of soldiers themselves.

This issue is therefore vital for Félix Tshisekedi, who has set his sights on improving the living conditions of soldiers to increase his popularity within the army, with whom he had little contact before becoming president.

Also, the former head of military intelligence, Delphin Kahimbi, on January 9, during an operation against the Allied Democratic Forces (ADF), had warned that “dissatisfaction” was “insidiously growing” in the ranks of the FARDC, “due to non-payment of bonuses and other funds (FP, FHF, etc.) for nearly three months, exacerbating to some extent the precariousness of their living conditions” in a leaked letter (in French).

Is this evidence of a cause and effect relationship? An increase of ten USD per month came into effect in January 2020. According to a source at the Ministry of Finance of North Kivu, soldiers with the lowest incomes are now paid 156,000 Congolese Francs per month, or 92 USD.

However, for such measures to have an impact – and allow Félix Tshisekedi to gain an upper hand in his power struggle with Joseph Kabila (in French) – this money still needs to get to its recipients.

In this regard, steps have been taken since the last presidential elections. The 2020 Finance Law lists approximately 205,000 “base salaries of regular staff” for “military, police and security” compared with 221,000 the previous year, which suggests troop number lists have been cleaned up.

Also, military hearings have been conducted in recent years on the issue of misappropriation of funds. This was the case of General Fall Sikabwe, Commander of the 3rd FARDC Defense Zone (in French), who was called to Kinshasa to be questioned on suspicions of misappropriation of funds – a relatively rare event at this level of the military hierarchy.

President Félix Tshisekedi also appears to be pressured to take this approach by the United States, a “strategic partner for peace and prosperity” of the DR Congo. On February 27, the US Ambassador, Mike Hammer, tweeted: “As we have consistantly (sic) said, those who are corrupt, commit violations of human rights, or disrupt the democratic process should be held accountable,” in reaction to the news that General Delphin Kahimbi had been suspended as military intelligence chief (in French).

However, will these hearings and warnings suffice to remediate the way the FARDC handles its money? Judith Verweijen has misgivings. “The issue of embezzlement of wages is only a small part of a vast system of generating revenue within the army,” she explains. “Commanders generate considerably more income by means of trafficking or illicit taxation, at roadblocks or mining sites. And they have to give some of this income to the senior military staff who appointed them. This system is therefore deeply entrenched at all levels and I am not convinced that increasing soldiers’ wages is enough for it to stop”.

Photo: 401kcalculator.org