The African Mission in Burundi (AMIB): A Study of the African Union’s Peacekeeping Success and ‘Triangular Area of Tension in African Peacekeeping’

Introduction

This article explores the African Union’s (AU) role as a security and peace consolidation actor in Africa, especially the way in which it relates to its peacekeeping operations. Against the backdrop of the AU’s Peace and Security Architecture (APSA), the article examines the African Mission in Burundi (AMIB) in order to explain its achievements and challenges, which Bogland, Egnell and Lagerström (2008) referred to as a ‘triangular area of tension in AU peace operations’. In addition, I briefly analyse the lessons learnt from the mission for improving future peace operations in Africa. AMIB was the pioneering AU armed peace operation and its deployment was authorised in 2003 before the inauguration of the Peace and Security Council (PSC) of the AU. AMIB mirrored the pan-African organisation’s ambition to intervene in African conflicts where the United Nations (UN) either was not too interested or delayed in responding to a volatile conflict situation in which there was no comprehensive peace agreement. Based on the assertion in the epigraphs above, three analytical questions are pertinent and addressed in this article. First, did AMIB balance the triangular area of tension in African peacekeeping, namely, the AU’s ambitions, the organisation’s peacekeeping capacity and its member states’ political will and agendas? Second, were the optimisms embedded in the APSA in terms of its ability to guarantee African security realised with the AU’s experiences in Burundi? Third, what lessons were learned from AMIB in order to address this triangular area of tension in African peacekeeping?

This article addresses these questions but this cannot be done in the absence of an understanding of the AU security agenda and the context in which AMIB was deployed and operated. First, I discuss both the APSA and the conflict history of Burundi. The conflict history is discussed through a periodisation of Burundi’s cycle of conflict. This is followed by the analysis of Africa’s mediation efforts that led to the signing of the Arusha Peace and Reconciliation Agreements for Burundi (the Arusha Agreement) in August 2000 and the subsequent ceasefire agreements made between the Transitional Government of Burundi (TGoB) and the armed groups in October and December 2002, which provided the basis for AMIB’s deployment. ¹ Then I examine AMIB right from its conceptualisation to deployment. The mission’s positive and negative experiences are also explored against its objectives, and the aforementioned triangular area of tension in African peace operations. I assess AMIB in order to question the extent to which the APSA’s rhetoric has been put into practice. I conclude with a reflection on lessons learnt from AMIB by the AU as a prelude to how to address this area of tension for better future peacekeeping performances in Africa.

The African Union and the African Peace and Security Architecture

The AU formally replaced the Organisation of African Unity (OAU) in July 2002. Established in May 1963, the OAU was to find solutions to problems confronting Africa at the time. Principally, the OAU was to end colonial rule in Africa, promote African unity, settle conflicts between its member states through peaceful means and work towards the improvement of the well-being of the African people (Imobighe, 1989). The OAU Charter emphasised the principles of non-interference in the internal affairs of its members and the sovereign equality of African countries among others (Article III, The OAU Charter, 1963). While the organisation was able to perform creditably in the area of decolonisation where it was crowned with dismantling the apartheid system in South Africa in the early 1990s, much cannot be said in term of its performance in other areas of conflict management. Throughout of its existence (1963–2002), the OAU was hamstrung by its ‘non-interference’ principle. This principle made it difficult for the organisation’s involvement in matters regarded as the internal conflict affairs of member states. Its conflict management effort in interstate conflict did not fare better, while its peacekeeping operation in Chad in the early 1980s was a miserable failure. Additionally, the OAU lacked the resources (both human and material) and technical capacity to undertake complex peacekeeping. The OAU member states’ political will to really support the organisation was lacking as African leaders were fond of pursuing national and/or personal agendas that often contradicted the OAU Charter. These factors and the new challenges confronting Africa at the end of the Cold War—globalisation, AIDS/HIV pandemic and democracy and good governance—made the OAU irrelevant in the evolving post-1989 international order and therefore its replacement was important (Sesay, 2008).

As part of their efforts to replace the OAU with a vibrant and proactive organisation, African leaders signed the AU legal document, the AU Constitutive Act (the Act) in Lomé, Togo, in July 2000. The Act replaced the OAU’s non-interference principle with that of non-indifference to conflict. Although the Act retains some of the OAU’s principles, especially those dealing with sovereignty and territorial integrity of its member states (Article 3b), and the principle of non-interference by any member state in the internal affairs of another member state (Article 4g), these principles do not prevent AU intervention in internal armed conflict situations. For, Article 4(h) of the Act empowers the AU ‘to intervene in a member state pursuant to a decision of the Assembly in respect of grave circumstances, namely: war crimes, genocide and crimes against humanity’ (OAU, 2000). Thus, Article 4(h) represents one of the cornerstones of the credibility of the AU as a security actor. Equally important is that the Act emphasises the ‘[t]he right of member states to request intervention from the Union to restore peace and security’ (Article 4j). With the provisions of the various sections of Article 4, Africa moved away from unqualified respect for state sovereignty to an approach where the duty to protect populations and ‘the right to intervention’ shapes AU’s security management agenda.

At the inaugural AU Summit in July 2002, African leaders adopted the ‘Protocol Relating to the Establishment of the Peace and Security Council of the AU’ (the PSC Protocol) (Africa-Union, 2002). The PSC Protocol institutionalised the AU conflict management mechanisms to enforce the Act’s new norms—the APSA. At the heart of the APSA is the PSC that is established as a standing decision-making body for conflict management. Replacing the Central Organ of the OAU Mechanism, ² the PSC functions as a collective security and early warning arrangement to facilitate timely and efficient response to conflict situations in Africa (Article 2(1), AU, 2002). The council is composed of 15 countries, of which 5 countries—1 country each per geographical region (Central, East, North, Southern and West Africa)—are elected to serve for a 3-year term, while the remaining 10 countries are to serve for a 2-year term. ³ This arrangement ensures continuity of the council. The Panel of the Wise (PoW), the Continental Early Warning System (CEWS), the Peace Funds (the Fund), an African Standby Force (ASF) including a Military Staff Committee (MSC) and the AU Commission (the Commission) support the PSC. Therefore, the APSA provides an all-encompassing set of instruments to address African security needs by African actors (Makinda & Okumu, 2008).

The PoW was inaugurated in December 2007. The Panel supports the PSC and AU Commission Chairperson in the area of conflict prevention through its advisory roles (Article 11(1), AU, 2002). It consists of five members—one from each of Africa’s five sub-regions. These are well-respected African personalities who have already contributed significantly to African peace, security and development. The CEWS anticipates and prevents conflicts. It is designed to gather and analyse information that will help the AU prevent conflicts in a timely manner (Article 12(1), AU, 2002). The CEWS builds on the Regional Economic Communities’ (RECs)/Regional Mechanisms’ (RM) early warning mechanisms. The CEWS furnishes the AU Commission Chairperson with information and data through which he or she may advise the PSC on potential conflicts and threats to African security and recommend the best course of action (Article 12(5), AU, 2002). In terms of its structure, the CEWS consists of the Observation and Monitoring Centre (OMC) known as ‘The Situation Room’ based at the AU Commission in Addis Ababa, and the Observation and Monitoring Units (OMUs) of the RMs. The Situation Room is to be linked directly to the RMs’ OMUs through appropriate means of communication. The OMUs are to continuously collect and process data at their respective levels and transmit them to the Situation Room. Based on the data collected through the multi-levelled African early warning system, the AU can then take prompt actions in response to a threat of violent conflict that could disturb African security (Engel & Gomes, 2009). The Peace Funds was established to provide funding for AU security activities. The Fund is financed directly through requisitions from the AU’s regular budget, including arrears of contributions and voluntary contributions from states and private sources (Article 21(2), AU, 2002). The ASF is the implementing arm of the PSC. It is made up of regional brigades ready for rapid deployment at appropriate notice. The ASF empowers the AU to conduct prompt and robust peace missions in response to complex emergencies requiring quick military deployments (within 30–90 days, depending on the type of operation) as decided by the PSC. The MSC advises and supports PSC and the ASF on all matters relating to military and security (Neethling, 2005).

The AU works with Africa’s sub-regional organisations known as the RECs in the operationalisation of the APSA. Therefore, the RECs are the pillars of the APSA. The PSC Protocol puts an accent on the AU’s cooperation and effective coordination with the RECs’ security mechanisms—known as RMs. Overall, the APSA seeks to guarantee African security with its three main imperatives—‘Try Africa First’, Responsibility to Protect (R2P) and ‘Africa Must Unite’. Thus, the APSA functions as a proactive security regime to address Africa’s democratic and governance deficits and also strives to promote socio-economic development and guarantee human security in Africa (Badmus, 2012).

A Synopsis of Burundi’s Cycle of Violent Conflict and Africa’s Peace Mediations

Burundi, with a population of about 8 million, comprising approximately 85 per cent Hutu, 14 per cent Tutsi and 1 per cent Twa, gained its independence in 1962 from Belgium (CIA). The tense relations among these groups, especially between Hutu and Tutsi, have plunged Burundi into a cycle of conflicts. The Burundian conflict is rooted in the unequal distribution of state power and socio-economic benefits along ethnic fault-lines. The country’s major institutions such as the military and judiciary are privatised and controlled by a small section of the country’s population. The alienated segment of the population becomes frustrated because its interests are perceived as being marginalised. Despite the Hutu’s demographic advantage, the minority Tutsi have dominated Burundi’s military and political–economic lives in both the pre- and post-independence periods (Lemarchand, 1995) and ethnicity has become an instrument in the hands of the political elite for political and socio-economic competition. The Tutsi’s control of power has on many occasions caused Hutu uprisings.

In April 1972, Hutu insurgents from Tanzania crossed into south of Burundi and carried out systematic attacks (genocide) on Tutsi killing between 2,000 and 3,000 Tutsi. A large-scale reprisal from the army led to the death of 100,000–200,000 Hutu (Manirakiza, 2005, pp. 45–45; Nyankanzi, 1998). In September 1987, Major Pierre Buyoya, a Tutsi, seized power in a bloody coup. This incident was followed by another Hutu uprising against Tutsi peasants, organised by the Hutu political organisation—Party for the Liberation of Hutu People (PALIPEHUTU)—in August 1988, in the north of the country (The African Centre for the Constructive Resolution of Disputes [ACCORD], 2007). The uprising caused massive reprisals conducted by the military against the Hutu in which many were killed and over 100,000 people became homeless. The international community condemned the tragic events and strong external pressures forced Burundi to carry out democratic political reforms. By the end of 1988, Buyoya embarked on democratic reforms, which culminated in the introduction of a multiparty system in 1992. The new democratic experiment was put to the test in June 1993 and a new Hutu-dominated non-armed political party—the Burundi Democratic Front (FRODEBU)—defeated Buyoya’s party—Union for National Progress (UPRONA)—while its leader, Melchior Ndadaye, became Burundi’s first democratically elected Hutu president. Ndadaye formed a government of national unity (Southall, 2006). Thereafter, Burundi witnessed the outbreak of a civil war in October 1993 following the assassination of Ndadaye by the extremist elements within the Tutsi-dominated Burundian army.

Consequent to the death of Ndadaye, the Hutu massacred many Tutsi. The tense situation resulted in retaliations by the combination of the army, gendarmerie and militias in the indiscriminate killings of Hutu peasants and officials (Bellamy & Williams, 2005). The wave of inter-ethnic killings after Ndadaye’s death were, as Jackson (2006) noted, ‘built on ground already poisoned by decades of colonial divide-and-rule, ethnicised violence around independence in 1962, and a failed insurrection in 1972’. The violence claimed between 250,000 and 300,000 lives (Daley, 2006). Amid the tense political atmosphere, a Hutu politician, Cyprian Ntaryamira, succeeded Ndadaye (Southall, 2006).

Burundi was further sucked into a vortex of conflict in April 1994 when Ntaryamira and his Rwandan counterpart Juvenal Habyarimana were both killed when their airplane was shot down over Kigali on their way from a regional peace meeting in Tanzania. Another political crisis followed Ntaryamira’s death and this prompted the UN to begin negotiations with the parties. In September 1994, the UN brokered ‘the Convention of Government’ that called for the establishment of a coalition government with a president from FRODEBU and a prime minister from UPRONA. Pursuant to this agreement, Sylvestre Ntibantunganya—a Hutu politician—was appointed president in October 1994 (Boshoff, 2010). Afterwards, the transitional arrangement suffered setbacks. As Ntibantunganya tried to accommodate the demands of the military, he felt the wrath of the radical elements within FRODEBU who believed that they had been alienated by a creeping coup. The radical elements eventually split to form the National Council for the Defence of Democracy (CNDD) and its armed group Forces for the Defence of Democracy (FDD). This development, together with UPRONA’s withdrawal from the parliament and government, resulted in a new round of violence in which many people were killed, while many became refugees in neighbouring countries.

The military–political situations in Burundi prompted the leaders of the Great Lakes region to hold the first regional conference on Burundi in Kigali, Rwanda. In November 1995, these leaders announced the formation of a ‘Regional Peace Initiative on Burundi’ (RPI), and former president of Tanzania Julius Nyerere was asked to lead the peace efforts. At Nyerere’s request the regional leaders, with the OAU and UN’s support, met with the representatives of FRODEBU, UPRONA and other smaller parties in Arusha, Tanzania, to discuss the security situation in Burundi. At the meeting, Ntibantunganya was reluctant to accept the regional initiative of sending peacekeeping forces to Burundi and the idea was rejected by the Burundian army (Boshoff, 2010). These peace efforts, unfortunately, failed to stop the Tutsi insurgency. On 25 July 1996, the early successes of the regional efforts were upset by the Burundian army’s overthrowing of Ntibantunganya and the reinstalling of Buyoya as president (Adebajo, 2011).

The economic sanctions imposed on Burundi by regional leaders forced Buyoya to withdraw from the Arusha peace process and he opted for internal settlement of the political crisis. The sanctions were interpreted by the Tutsi as evidence of Nyerere’s support to the Hutu cause. Buyoya undertook efforts to generate a national dialogue while he also forged close partnerships for peace among the government, judiciary and civil society organisations. Buyoya unbanned political parties and restored the 1993 parliament. He negotiated directly with the CNDD and successfully engaged with FRODEBU and by May 1998, the FRODEBU members that remained in Burundi re-entered the government. Then Buyoya agreed to revive the Arusha process.

In June 1998, Nyerere convened the Arusha II meeting that was attended by the government, national assembly and 17 political parties including UPRONA and FRODEBU. However, the Arusha talks sparked off splits in the CNDD and PALIPEHUTU when factions that had the allegiances of the majority of their armed wings, the FDD and National Liberation Force (FNL), respectively, demanded recognition for these armed groups as independent organisations. ⁴ The Regional Initiative decided to continue the negotiations with the political organisations in the absence of PALIPEHUTU—FNL and CNDD—FDD to secure a political agreement, hoping that a ceasefire agreement with the military factions would be secured later. The former president of South Africa, Nelson Mandela, became the new facilitator in November 1999 following the death of Nyerere.

Against all odds, ⁵ Mandela successfully negotiated the Arusha Agreement on 28 August 2000. The agreement is not a comprehensive one in that only 13 out of the 19 delegations signed it. ⁶ The remaining six parties that refused to sign were all Tutsi parties and due to regional pressures the remaining parties eventually signed the agreement at a regional summit in September 2000. The Arusha Agreement established a framework for a transitional settlement of the conflict leading to national elections.

The Arusha Agreement did not lead to a reduction in conflict and the idea of deploying an African force to provide VIP protection to returning politicians taking part in the peace process did not materialise. The idea suffered a setback because many African states were unwilling to contribute troops because no ceasefire agreement was in place. To implement the agreement, Mandela persuaded Pretoria to deploy troops—South African Protection Support Detachment (SAPSD)—in Burundi. The SAPSD’s deployment started in late October 2001. Once completely deployed, its force strength stood at just over 700 men. Despite its initial opposition from the Burundian army, SAPSD was able to overcome this challenge and then focused on its assigned functions. The SAPSD operated on a very limited mandate as a protection force, and under the rules of its mandate, troops would be withdrawn in case they become the object of attack.

The SAPSD stabilised Bujumbura, the Burundian capital, but it was not enough to ameliorate the Burundian security situation overall. SAPSD was Bujumbura based, it neither protected the civilian population nor performed a broader peacekeeping function, while the humanitarian situation continued to worsen. Between October and December 2002, two ceasefire agreements were signed but the belligerents did not respect them. With the deteriorating security and humanitarian situations, the deputy president of South Africa, Jacob Zuma, who had been helping Mandela in facilitating the peace process, requested the UN and donor community’s contributions to the initiative of deploying an African force in Burundi. South Africa’s proposal was later accepted and it set the stage for the AU peace operation in Burundi.

The African Mission in Burundi

AMIB’s Achievements and the ‘Triangular Area of Tension in African Peacekeeping’

Judging by its first objective of supervising the implementation of ceasefire agreements, AMIB could be credited for achieving this important role. The presence of the African force helped to deter further political violence and stabilised the country with the exception of Bujumbura Rural where Rwansa’s PALIPEHUTU—FNL remained very active (Murithi, 2009). Aboagye (2004) estimated that about 95 per cent of the entire country was relatively stable at the end of the operation. The relative stability in Burundi brought about by the presence AMIB was conducive to moving the peace process forward. AMIB halted the escalation of violence and was able to manage the violent aspects of the conflict. The reduction of political violence was replaced by acts of criminality in late 2003 but AMIB was able to handle this and also helped in overseeing the implementation of the ceasefire agreements and facilitated the CNDD—FDD’s participation in the peace process. In this respect, with its limited resources, the AU committed a significant share of its resources to VIP protection; this was for the leaders of the armed groups returning to Burundi to participate in the peace process. If not for AMIB’s intervention, Burundi would have witnessed a more dangerous violent conflict (beyond what it was before AMIB operation) with far more devastating consequences, as stated in the first epigraph of this article.

AMIB did not achieve much in terms of its DDR objective. The mission was envisioned to implement the World Bank-funded DDR programme, which involved cantonment and disarming of about 20,000 ex-combatants. Soon after its deployment, AMIB, at the invitation of the World Bank, joined the Multi-country Demobilisation and Reintegration Programme (MDRP) and became part of a joint planning group for the implementation of the DDR programme. But due to a number of challenges, including the delay by the TGoB in meeting the World Bank’s requirements, the implementation of the DDR programme did not commence until December 2004, 6 months after the expiration of AMIB’s mandate. Despite these challenges and its limited resources, AMIB went ahead to implement its DDR mandate. The first cantonment area was set up in Muyange in June/July 2003. The second cantonment site was not established until towards the end of AMIB operation in May 2004 (Boshoff & Vrey, 2006). The challenge of the triangular area of tension in African peacekeeping manifested because the AU lacked resources to sustain its force and as a result, its mission was unable to canton a large number of ex-combatants. Afterwards, the cantonment area ran out of food, medical supplies and lacked tangible infrastructure. For sustainability of the Muyange cantonment site, the HoM used his influence to secure assistance from international donors. At the Muyange cantonment area, AMIB was able to assemble and disarm 189 members from CNDD–FDD of Jean-Bosco Ndayikengurukiye and PALIPEHUTU–FNL of Alain Mugabarabona. By November 2003, the number of disarmed ex-combatants increased to 228 (Boshoff, 2010). In addition, AMIB was successful in finding suitable DC areas and Pre-Disarmament Assembly Areas (PDAAs). This effort culminated in AMIB’s identification of 11 PDAAs. Former fighters cantoned at Muyange were subsequently transferred to the PDAAs in December 2003 and January 2004.

AMIB’s third objective stated that the mission was to strive towards ensuring that conditions were created for the deployment of a UN peacekeeping mission. How was AMIB’s performance in this area? The mission’s performance in this respect could be described a success. The political–security situations in Burundi were volatile when AMIB was deployed. The absence of a comprehensive peace agreement meant that AMIB had to operate in a fluid security environment. The African mission mitigated the escalating violent conflict and stabilised about 95 per cent of the entire country by the end of its operation; the UN Evaluation Team recognised this achievement when they concluded in February 2004 that the conditions in Burundi were now appropriate for the deployment of a UN peacekeeping mission. On 21 May 2004, the UN Security Council (UNSC) passed Resolution 1545 which authorised the deployment of a UN peace operation in Burundi (ONUB by its French acronym) with an annual budget of $333.2 million. This was in line with the agreement with the AU that AMIB would give way to a multidimensional UN peace mission in Burundi. The fact that at the expiration of AMIB’s mandate the UNSC was able to authorise the deployment of ONUB to replace the African mission attests to AMIB’s success in creating conditions suitable for the presence of a UN peace mission (Bellamy & Williams, 2005). On 1 June 2004, the UN officially took over the peace mission with a peacebuilding mandate. The rebadged AMIB peacekeepers formed their advance party while other contingents were from Nepal, Pakistan and Kenya (Williams, 2006). One important point of note here is that AMIB helped to stabilise the political and security situations in 2004, and laid the foundations for a more multidimensional peacebuilding process in mid-2004. The ONUB operation ended in December 2006 after it successfully completed its mandate, and the UN Integrated Office in Burundi (BINUB by its French acronym) replaced ONUB with the mandate to coordinate international assistance.

In its fourth objective of contributing to political and economic stability in Burundi, AMIB’s achievement in this regard is partial. Undoubtedly, AMIB helped to manage the conflict but the mission was constrained in many ways, among which were the attacks on civilian population and the cantonment areas. The December 2003 Human Rights Watch Report blamed both the Burundian army and the armed factions for deliberate attacks on the civilian population in violation of IHL. These challenges negatively affected AMIB’s ability to fully realise its objective. Besides, the displaced people based in the eight Burundian provinces and refugees in the three (refugee) camps in Tanzania were supposed to be integrated into society and given their means of engaging in livelihood activities as defined in terms of their accessibility to allocated land. Unfortunately, those disarmed ex-combatants in the DCs were not provided with economic opportunities because the AU lacked resources and this made their reintegration into society problematic (Murithi, 2008). AMIB was unable to accomplish this objective. A critical look at this mandated objective raised the question: What is the domain of military peacekeepers? Traditionally, the primary role of military peacekeepers is to interpose between belligerents and serve as external guarantors of a ceasefire agreement in order to avert further bloodshed and create an enabling environment in which peacemaking and post-conflict peacebuilding can take place through the efforts of civilian and humanitarian missions. AMIB, as an integrated peace mission, performed the traditional peacekeeping role but could not, due to its resource limitations as part of the challenges of the triangular area of tension in African peacekeeping, contribute meaningfully to peacebuilding efforts and to some extent this also applies to peacemaking, although the peace process was ongoing behind the scenes.

The AU’s achievements in Burundi should not be over-romanticised since AMIB’s experiences raised a number of concerns. AMIB operation brought to the fore what Bogland, Egnell and Lagerström (2008) call a triangular area of tension in AU peace operations, that is, the resource and capacity constraints often associated with African peacekeeping, which did not allow the mission to fully implement its mandate. The UN Secretary General (UNSG), in a report on Burundi, acknowledged: ‘The financial and logistic constraints under which AMIB is operating prevent the force from fully implementing its mandate.’ ⁹ AMIB’s lack of required financial resources originated from three sources. First, within Africa, the AU member states were not enthusiastic in providing the requisite funds to the mission. A clear indication of this was the AU Memorandum of Understanding (MoU) where it ruled that the TCCs were to be self-sustained for up to 2 months of the operation. Based on this ‘self-sustainment concept’, both Ethiopia and Mozambique were able to deploy their troops with external support. The implication of the self-sustainment concept of AMIB operation for the TCCs was uncovered by de Coning when he succinctly asserted that ‘Even then their (that is, Ethiopian and Mozambican contingents) deployment was delayed, and once deployed their operational status was affected by ongoing financial constraints and uncertainty’ (de Coning, 2004). Second, the AU itself was financially and logistically incapacitated. Its resources limitations were also due to the fact that since its inauguration, the AU had been operating with a budget deficit, which made the institution rely on external donors for its peace operations (Okumu, 2009, p. 105). Therefore, the institution was unable to provide sufficient funds for its peace mission and relied on financially incapacitated TCCs to fund the deployment and sustenance of their troops. Third, the UN’s indifferent attitude towards this problem confirmed its unwillingness to sufficiently fund the operation. This can be partly explained by the fact that the AU was newly established and its future was not completely certain. Given the questionable character of the OAU, coupled with the fact that AMIB was the AU’s pioneering mission, donors were not enthusiastic in adequately supporting AMIB in the way the AU had expected. Powell (2005) argued that the AU was asked to reduce AMIB’s budget when it was first presented to the UNSC because according to the UN, the budget was too large, and that the human and material resources proposed for AMIB by the AU, in the opinion of the UN, were too ambitious for African peacekeeping. However, in addition to being insufficient, the funds were disbursed very slowly, which was also a source of the problems the AU faced with AMIB operation.

The operational budget of AMIB was estimated at around $110 million for 1 year (Boshoff & Dara, 2003, p. 43). This amount is exorbitant, in the African context, considering the dwindling nature of the economic situation across many African countries. It is important to emphasise the fact that only five African countries—Nigeria, South Africa, Gaddaffi’s Libya, Egypt and Algeria—pay 75 per cent of the AU budget. The AU Commission’s budget for 2004 was approximately $32 million (de Coning, 2004). At the end of the operation, AMIB’s total budget stood at $134 million. With its financial resource limitations, the AU relied on external donors to fund AMIB’s budget. Aboagye (2004, p. 15) writes that

…the pledges from the partners, amounting to some $50 million, fell far short of the budget. Even worse, actual donations into the trust fund amounted to just $10 million, even though this excluded in-kind assistance from the US ($6.1 million) and UK ($6 million), to support the deployment of the Ethiopian and Mozambican contingents respectively. ¹⁰

These financial limitations, coupled with the piecemeal approach adopted by the EU and other donors in disbursing the money pledged, hindered AMIB’s operational performance.

AMIB was also challenged by the AU’s lack of institutional capacity and logistics. As I mentioned earlier, AMIB was authorised when the AU was newly established, and the APSA’s institutions were just evolving. These conditions incapacitated the AU in organising deployment. The AU’s lack of capacity led South Africa to provide leadership and plan the mission. Besides, the late arrival of the Ethiopian and Mozambican troops also revealed the AU’s lack of logistical resources that hindered the efficacy of the peace operation at the initial stages. The troops would have been quickly deployed if the AU itself had been well resourced and equipped for its peace operations. The AU’s lack of logistics, especially modern equipment, negatively affected the mission’s ability to deliver on its revised RoE for civilian protection, as it was unable to move out of secured areas.

Since AMIB’s deployment was based on the understanding that the mission will be replaced by a UN peace mission when the Burundi security situation becomes normalised enough for such a deployment, the AU established a strategic-level AU/UN engagement to mobilise resources, as well as to receive in-theatre administrative and logistical assistance from the UN system. These included technical capacity support that AMIB received from the UN Mission in the Democratic Republic of Congo (MONUC). The UN support to AMIB included benefits from MONUC’s experience in the areas of public information, headquarters administration and DDR. Also, as part of the operational-level collaboration, AMIB consulted with international agencies such as the EU, UNICEF and World Bank. The consultation was crucial for AMIB to source funds and other assistance for the DDR programme and also in its implementation of the ceasefire agreement tasks.

AMIB operation also revealed how the determination and commitment of a lead state can fill the gap created by the capacity deficiencies of a peace operation-authorising institution and make such a peace mission a reality. It should be recalled that the appointment of Nelson Mandela as the new facilitator for the Burundian peace process in November 1999 saw the inroad of South Africa’s diplomacy and resources in Burundi. Mandela did not only successfully negotiate the Arusha Agreement but also used his good offices to secure Pretoria’s consent in deploying SAPSD to prevent the peace process from falling apart. AMIB would have been an impossible mission without the leadership, and human, military and financial resources from South Africa (Williams, 2006). The AU was hamstrung by its lack of resources and also in its force generation inability. A study conducted by a Durban-based NGO African Centre for the Constructive Resolution of Disputes (ACCORD) found that South Africa played a significant role in the transitional operation and was ‘the largest force present on the ground, contributing approximately 1,500 troops, which proved a determining factor for the deployment of the mission’ (ACCORD, 2007, p. 31). Probably this is why a Swedish defence analyst argued that although AMIB was an AU mission on paper, in reality it was a mission that wholly relied on the leadership of one single TCC, South Africa (Svensson, 2008, p. 17).

Given all the numerous challenges mentioned above that were skilfully negotiated, AMIB could be described as a relatively successful peace operation when compared with the previous African peacekeeping operations, especially under the OAU regime. Overall, one would describe AMIB as a partially successful peace operation, for the mission was able to achieve the traditional peacekeeping goals; however, it could not achieve much in its DDR and economic stability mandate. AMIB raised the hope that Africans can manage African conflicts despite the fact that the APSA’s institutions were in the process of being established when the mission was deployed.

Lessons Learnt and Future Outlook

This article explored AMIB to question the role of the AU as a security actor in Africa, and the optimism embedded in the APSA, in terms of its ability to guarantee African security through a proactive rather than reactionary approach to the continent’s security challenges. AMIB was deployed in a very dangerous security environment and at a time when the APSA institutions were just evolving. The mission recorded some successes, notably in stabilising security in the country and thereby creating favourable conditions for the promotion of other important tasks, such as the DDR programme, security sector reform and other institutional reforms, such as human rights and economic stability. All these positive outcomes can be related to a succeeding UN mission for the advancement of peace in Burundi. Despite these achievements, the challenges facing the AU in relation to mounting and sustaining its peace operations have brought to the fore what Bogland, Egnell and Lagerström (2008) call a triangular area of tension in AU peace operations. As far as AMIB operation is concerned, I conclude that the AU has not been able to bridge the gap between its ambition and capacity for running totally successful peace operations.

AMIB revealed the commitment of the AU, as an institution, to implement the peace and security norms embedded in its Constitutive Act and the PSC Protocol. The AU and RPI recognised that the existence of peace in Burundi would contribute to and improve the security situation in the Great Lakes region. Consequently, the pan-African institution deployed AMIB despite the fact that there was no comprehensive peace agreement in place. Without question, AMIB’s deployment demonstrated the AU’s ambitions and will to deliver on its promises of securing the continent and its people. Furthermore, the AU’s commitment to implement its security framework is part of the institution’s efforts to fill gaps in the continent’s peace and security agenda, especially in mobilising a peacekeeping mission in African conflicts when the UN is reluctant to intervene.

The AU’s ambition was overshadowed by its capacity limitations and a lack of political will on the part of AU member states. AMIB’s challenges included a lack of resources, logistics and funding, all of which contributed to the AU’s capacity weakness to manage peace operations efficiently. Despite the fact that the maintenance of international peace and security is the responsibility of the international community, the fact that AMIB was envisaged to give way to a UN peace mission is a clear indication of the AU’s acceptance of its organisational weakness as well as its resources and capacity limitations. Besides, the AU member states’ unenthusiastic stance in relation to providing sufficient financial support to the mission has raised serious questions about African ownership of and control over the operationalisation of the APSA and AU peace operations. The gap between the AU’s ambitions on the one hand and its capacity along with member states’ agendas on the other is a reminder of the regional-sceptics’ arguments in terms of the reliability and effectiveness of the AU’s integrated peace and security agenda and the running of regional peace operations. The triangular area of tension remains a challenge for the AU, in the operationalisation of its peacebuilding agenda and to ultimately free Africa from virulent armed conflicts.

In spite of these challenges, I conclude this article with a reflection on some lessons that were learned from AMIB’s experiences by the AU for planning future AMIB-like peace operations in Africa. The most pertinent of these are discussed here. The AU’s experiences in Burundi underlined the importance of prompt AU interventions in African armed conflicts for peace and stability in Africa. In Burundi, the AU deployed AMIB first as a stabilisation force in preparation for a multidimensional UN peace mission to be mounted later when the conditions allowed for such a deployment. This is a major achievement since the deployment of a UN mission takes a long time. The prompt deployment of AMIB filled the gap between the outbreak of the conflict and the presence of ONUB that took over AMIB in May 2004.

AMIB suffered a lack of financial resources. Financial constraints compelled the AU to rely heavily on international funding partners for assistance for its peace operation in Burundi. Thus, AMIB was a donor-driven and donor-dependent peace operation and this situation demonstrated that the AU did not perform well in the area of financing its peace operation in Burundi. Excessive reliance on external donors for the sustainability of the AU’s peace missions is problematic as such a practice risks loss of control over such support and, by extension, its peace operations in the African context. Besides, the piecemeal fashion in which the donors were disbursing pledged funds to the AU for AMIB operation calls for the AU to look for alternative ways (besides the member states’ contributions) to fund and sustain its peace missions, pending the arrival of donor contributions, which may take up to 6 months. The dangers of excessive reliance on external donors for peace operation are illustrated by the Ethiopian and Mozambican experiences in AMIB. Once these contingents were deployed in Burundi with the assistance of the US and the UK, respectively, their sustainability became the responsibility of South Africa. Thus, the leadership role played by South Africa helped AMIB to be a relative success story. The South African role in Burundi underscored the importance of the AU’s need of the resources of a lead African nation (or nations) to be involved in the mission for the (mission) sustainability, at least for the first few months of African peace missions. ¹¹ AMIB could not have been a relatively successful peace mission without Pretoria’s commitment and resources. Nigeria’s leadership role in the Economic Community of West African States’ (ECOWAS) peace missions in West Africa is also a good example of this effort. In the 1990s, the Nigerian-led Economic Community of West African States Monitoring Group’s (ECOMOG) troops were deployed in Liberia and Sierra Leone in the absence of proper logistics and financial arrangements in place. Lack of resources denied many ECOWAS member states to deploy to these missions but Abuja shouldered responsibility as the lead nation of the ECOWAS peace operations in both countries. The costs of the operations were largely borne by Nigeria. AMIB and ECOWAS examples underlined the needs for predictable funding for African peace operations and also the needs for the AU and other African sub-regional organisations to reflect on how best to finance their future peace operations and also how the larger international community can fund African peace missions expediently and adequately to ensure that the African institutions are not set up to fail during the onset of their peacekeeping roles.

AMIB has had some successes because it operated on a clear mandate and RoE. Powell offered a critique of AMIB’s mandate when she said ‘…AMIB was tasked with a mandate it could not possibly fulfil and its resources were not aligned with its requirements.’ Powell’s argument suggests that a clear mandate alone is not enough by itself for a successful peace operation. At the strategic level, mission planners need to operate a peace mission on a realistic and robust mandate to cater to the exigencies on the ground in line with resources, both human and material, that are required to be at the AU’s disposal. In Burundi, AMIB directed efforts towards the implementation of the key aspects of the peace agreement at the initial stage of the peace mission. Additionally, AMIB operation underlined the important role of the offices of the Special Representatives of Chairperson of the AU Commission. The credibility and diplomatic acumen of the political head of AMIB, Ambassador Bah, helped the African mission tremendously in achieving cooperation from the warring parties. The experience, credibility and respect that the Burundian political players have for Ambassador Bah helped the mission to achieve its mandate and also helped AMIB and the UN to work together in a harmonious manner in this theatre of peace operations. Besides, the influence and credibility of the facilitators of the Burundian peace process, Julius Nyerere and later Nelson Mandela, contributed to the success of the mission. AMIB precedent showed that peace operations and peace processes tended to be successful when the peace negotiators and HoMs are influential and credible personalities with extensive political and diplomatic skill and experience.

Finally, one important method that must be obviously accepted in African peacemaking processes is the recognition of and ownership by local population. A peace process must be people-centred and also owned by the people for the benefit of the people. In Burundi, the peace process was owned by the people of the country, for the representatives of local populations and the Burundian civil society groups/agencies were pivotal in the peace process while post-conflict peacebuilding efforts were people-centred as well. The post-conflict peacebuilding efforts targeted assistance towards those affected by the conflict. Ownership of the Burundian peace process by the government and the people of Burundi and the way the DDR process and national rebuilding were conceptualised in the context of the peace operation helped AMIB to be a success story overall.