Present-day Mai-Mai groups have their immediate roots among the armed groups that appeared in the early 1990s. This period was marked by a fledgling democratisation process that triggered the political manipulation of conflicts around land, local authority and access to resources in the east of the country then known as Zaire. Many of these conflicts grew out of the colonial period, when the colonisers granted certain groups customary chiefdoms but excluded others, and organised the mass migration of Rwandans to Congo to work on plantations and in mines.

These processes left a legacy of disputes that were often expressed in identity-based terms, pitting populations identifying themselves as ‘indigenous’ (or ‘autochthonous’) against speakers of Kinyarwanda (Rwandophones, encompassing Hutu and Tutsi), who were labelled ‘foreigners’. Within this context, armed groups were seen to defend the land and rights of particular (ethnic) communities, and were closely linked to the leaders of those communities.

The mobilisation of these groups drastically increased during the First (1996–97) but particularly in the Second (1998–2003) Congo War, both of which were unleashed by the invasion of insurgencies supported by the Rwandan government. During the Second War, the Rwanda-backed Congolese Rally for Democracy (RCD) managed to occupy large swathes of the east. Based in remote rural areas, Mai-Mai groups waged a guerrilla war against the RCD, whose leadership was dominated by Congolese Rwandophones.

Given the strong dislike for the RCD among large parts of the population, Mai-Mai groups drew massive popular support, portraying themselves not so much as an armed group but as a ‘popular movement’ engaging in self-defence. As Kapopo, former commander of a Mai-Mai group in Itombwe said: ‘The Mai-Mai, it’s not only us. It’s everyone, it’s every patriot who defends the Congo. It’s not a movement linked to a person, it’s the state of mind of every patriot, of those who defend their country and do not want aggression and the pillage of natural resources’ (interview, Bukavu, 27 March 2011).

Despite this popular ethos, Mai-Mai groups became increasingly disconnected from the communities they had emerged from and started to pursue their own agendas. Several groups liaised with bigger politico-military movements, which sometimes also provided military training. Many became caught up in competition over access to natural resources, such as gold, coltan and timber. Being based in the west and having no troops in the rebel-occupied east, the Congolese government also used Mai-Mai groups as proxies to fight the RCD, providing them with arms and ammunition. This diversification of sources of income and support made Mai-Mai groups less dependent on and accountable to local communities, contributing to a rise in abuses against the very civilians they claimed to defend.

Government recognition, even though largely symbolic, fostered an important sense of entitlement among Mai-Mai groups. It promoted a self-image of the Mai-Mai as patriots essential for defending the country’s territorial integrity, and who by extension deserved compensation for their efforts. Yet through the peace process and ensuing transition, the Mai-Mai’s expectations were largely frustrated. A first obstacle was their participation in the Inter-Congolese Dialogue (ICD) held under the auspices of the Organisation of African Unity and facilitated by Sir Ketumile Masire, former President of Botswana. The dialogue was an attempt to both establish peace and forge a new political dispensation for Congo. In addition to the belligerents (the government, rebel movements and other armed groups), it included political parties and civil society groups. Yet many of these groups were embroiled in intense struggles about which groups should participate and who should represent them, which complicated the negotiations.

After Mai-Mai representatives were excluded from an ill-fated attempt to begin the ICD in Addis Ababa in 2001, the Mai-Mai were invited to a second effort in Pretoria in 2002. But who was to represent this heterogeneous category, which on paper looked like a few large coalitions of fighters, but in reality consisted of dozens of groups whose commanders often acted autonomously?

Only six Mai-Mai delegates were asked to attend. Moreover, Kinshasa tried to influence the composition of the delegation to ensure it was favourable towards the government. As a result, many Mai-Mai groups did not feel represented during the peace process, which culminated in the signing of a peace agreement named the Global and All-Inclusive Accord. These groups considered themselves to be non-signatories and so not bound to respect the agreement’s provisions. As Mai-Mai leader Amuri Yakotumba said: ‘We, the Mai-Mai, we were not signatories of those agreements. We were not invited so these agreements do not concern us’ (interview, Sebele, 14 December 2011).

Mai-Mai representation was also a problem in relation to the allocation of positions in the army and political institutions during the transition (2003–06). The accord had been signed by the government, four rebel movements, 28 different political parties representing the ‘unarmed opposition’, the umbrella category of ‘civil society’ and finally, ‘the Mai-Mai’. The Mai-Mai were listed as an ‘entity’ (entité) in the peace accord, rather than a ‘component’ (composante) like the biggest political-military movements, and so were entitled to comparatively few political positions. Similarly, within the army, they were given a low quota in the allocation of command-and-staff positions. In part owing to their fragmentation, the Mai-Mai had limited political weight in Kinshasa, causing them to miss out in the division of political and military functions still further.

Once in government in Kinshasa, those acting as the Mai-Mai’s political representatives largely pursued their own interests, losing touch with their bases in the east. The same pattern developed regarding those appointed to higher-level functions in the army. Not seeing their grievances and concerns articulated, many lower-level officers and rank-and-file lost faith in their leaders. In addition, the division of scarce positions in political and military institutions led to endless quarrels both among and within already fragmented Mai-Mai groups. These quarrels induced discontent and schisms, and ultimately contributed to the return to arms of those feeling disadvantaged.

Many Mai-Mai attributed their relative marginalisation in the political and military arena to their identity as autochthones. They believed that they were being discriminated against and saw this as evidence that the new state institutions were infiltrated and dominated by ‘foreigners’, in particular Congolese Rwandophones. In the words of Mai-Mai leader Fujo Zabuloni: ‘the Mai-Mai today have no functions. We fought during the war, but then we found that there was no work for us in the army. So we went back to the bush … The question is: Why did they bring foreigners into the army?’ (interview, Kisanga, 12 February 2012).

Another reason for Mai-Mai combatants to boycott the army integration process was that some of them never wanted to work as soldiers deployed throughout the Congo’s vast territory. Rather, they wanted to stay in their zones of origin, close to their families and communities. In low-level positions in the army in distant territory, they would never command the same status and respect, nor have the same levels of influence and income. Thus, many Mai-Mai chose not to join the army but rather to be demobilised.

Additionally, numerous Mai-Mai groups – like other belligerents such as parts of the RCD – refused to integrate their armed wings fully into the army out of fear that local communities would be left unprotected. With memories of atrocities fresh in mind, some of which had appeared ethnically targeted, conflicts and suspicion among communities remained high. Few people fully trusted the newly formed national army, often doubting its neutrality. As political attention was focused predominantly on the transitional process in Kinshasa, there were limited initiatives to address the conflicts that had pitted communities against each other in the east.

Afraid to lose political and economic influence, civilian leaders and businesspeople linked to the Mai-Mai often encouraged Mai-Mai groups not to fully disband. While the war was proclaimed over, eastern Congo remained militarised – in particular the two Kivu provinces, where force was still a shortcut to political power and resources. For instance, links to armed groups was an important asset for holding on to mineral-rich zones or important positions in the local administration. To both attract attention in Kinshasa and consolidate influence in their local constituencies, politicians had to show that they had leverage over armed groups. This logic was not limited to Mai-Mai networks but played among all the former warring factions.

The 2006 general elections marked the end of the transition and revealed that it had failed to consolidate peace. A number of Mai-Mai factions, like Yakotumba’s, abandoned their commitments to army integration and reconstituted themselves as armed groups, declaring little faith in the newly elected government. A major Rwandophone rebellion, the Tutsi-led National Congress for the Defence of the People (CNDP), reared its head in North Kivu, which in turn prompted a further proliferation of Mai-Mai groups in the area.

To explain why they took up arms, these groups generally invoked reasons of community self-defence necessitated by ongoing insecurity, dissatisfaction with the army integration process, and grievances related to conflicts around land and local authority. Yet for several military leaders, status, income and influence were also at stake. Leading an armed group gave them visibility, including in Kinshasa and among international donors, and made them people to reckon with.

The pertinence of these more opportunistic motivations became particularly clear when the news broke that a large conference to bring peace to the Kivu provinces would be organised by the Congolese government in Goma in January 2008. The conference culminated in the signing of a ceasefire agreement by around two dozen groups and paved the way for the creation of a government-initiated stabilisation and reconstruction programme (Amani) which was sponsored by donors.

In the hope of gaining a place at the conference and access to positions in the army, the administration or the Amani programme, some entrepreneurial individuals created new armed groups or reconstituted dormant ones. A good example is commander Mahoro, who during the Second War had served under Mai-Mai commander Nyakiliba. After the war, he participated in the army integration process and received the rank of major, but never obtained a function. Therefore, when he heard of the Goma Conference, he decided to launch a new movement called the Mai-Mai Mahoro. In his own words, the conference was an opportunity ‘to have my rank of colonel fully recognised and to no longer be en dispo [not having an active function in the army]’ (interview, Bijombo-Ishenge, 18 November 2011).

Similar feelings of marginalisation would push a large number of ex-Mai-Mai officers in the army to desert and create new armed groups after 2009. That year a peace deal was concluded between the government and the CNDP – as well as between other armed groups and the government. The CNDP integrated into the army and obtained numerous influential positions in the command chain. This reconfirmed the belief among ex-Mai-Mai that Rwandophones were systematically favoured.

CNDP integration also went hand-in-hand with large-scale military operations against armed groups, which fundamentally disrupted existing political and military power balances in many areas. The result was serious instability, which paved the way for the emergence of new Mai-Mai groups citing the need for community self-defence. An army restructuring process in 2011 caused many ex-Mai-Mai to again feel disadvantaged by the army. Elections that year gave further impetus to the proliferation of Mai-Mai groups, a dynamic that persists today. The accords signed on 23 March 2009 include the signatures of less than 20 Mai-Mai groups. But an analysis mapping armed groups published in December 2017 lists over 70 Mai-Mai groups, and their numbers have grown further since then.

Given the multitude and heterogeneity of the parties involved, the peace process that aimed to end the Second Congo War always faced an uncertain outcome. The government re-established administrative control over former rebel-held areas, foreign armies withdrew, and many rebel groups were disbanded or transformed into political parties. But in the east, violence continued.

Although by no means the only reason, the power-sharing arrangement on which the peace process and transition were based importantly fed into this instability. By making the use of force convertible into positions in or influence on the state apparatus, power-sharing incentivised violence and unleashed a relentless competition for power, positions and resources, not only among but also within factions. This competition affected all ex-belligerents, but it hit the Mai-Mai particularly hard. Dozens of groups that had never formed a unified structure now had to divide up positions among themselves in the army and political institutions, positions that appeared fewer and less important than those allocated to their main adversary, the RCD.

Competition for power during the transition unfolded in a climate of ongoing conflict, distrust and violence. Some of this violence was clearly intended to protect economic and political gains made during the Congo Wars. Yet in the east, it was also related to a legacy of conflicts around land and local authority that were often instrumentalised in relation to identity. The focus on dividing positions in the state institutions seems to have deflected attention from addressing these conflicts and the related distrust between communities, on which Mai-Mai groups strongly drew for recruits and resources.

Excluding the Mai-Mai from the peace process altogether would have been likely to have produced worse outcomes. But questions can be raised in respect of the ways in which they were included. Should they have been treated as a single faction on a par with unified rebel movements? Should they have been given command positions and the option to integrate into the army, or rather have been demobilised? Could more holistic approaches to Mai-Mai demobilisation have been adopted that simultaneously addressed conflicts and distrust between communities?

While there are no easy answers to these questions, the case of the Mai-Mai does provide a few pointers as to how to address the inclusion of local and fundamentally fragmented groups into peace processes. First, a comprehensive study must be made of the level and nature of factions’ fragmentation and how this affects (purported) leadership and the implementation of commitments. Second, power-sharing arrangements need to examine whether and how they can be implemented in ways to avoid further fragmentation. Third, more attention must be paid to the reasons underlying fragmentation – be it self-interested leadership, deep-rooted conflicts, or the fact that armed groups issue from communities that have never known centralised leadership. Finally, the story of Mai-Mai fragmentation also urges us to look beyond overarching, dichotomous conflict narratives (in this case: of autochthonous versus Rwandophone groups), to uncover the multi-faceted and diffuse conflicts that may hide behind them.