By Nicole Ball

November 30, 2014

Download this Brief as a PDF:
English | Français | Português

Summary

Security sector reform (SSR) is increasingly put forward as a solution for a broad spectrum of African states facing security challenges. Yet, for a variety of reasons, there are relatively few examples of successful SSR implementation. The case of Burundi’s Security Sector Development (SSD) program, while still a work in progress, is noteworthy for having advanced both security sector effectiveness and democratic account­ability since its inception in 2009. The success of SSR programs depends heavily on the ability to address politically sensitive issues. For the most part, SSR programs have fared poorly in this regard. Burundi’s SSD program, however, proactively addresses the politics of reform at both the policy and operational levels on a daily basis. Insights from Burundi’s SSD program can inform other African SSR initiatives.

Burundi’s 12-year civil war cost approximately 300,000 lives, devastated the nation’s physical and institutional infrastructure, and tore at the social fabric of this ethnically divided population. Efforts to heal these wounds thus started from a polarized political environment and high levels of distrust. Compounding these challenges, Burundi would have to overcome a long legacy of military domination of politics. The Arusha Peace and Reconciliation Agreement, which inaugurated the transition process in 2000, called for significant reform of the security sector, including the integration of rebel factions into the armed forces. A power sharing agreement in 2004 coupled with the decision by the predominantly Hutu rebel group, Front de Libération Nationale, to transform itself into a political party in 2008, ushered in a period of relative stability and peace in Burundi. The armed forces have subsequently made important strides in becoming ethnically integrated and professional.

Nonetheless, serious challenges remain. The political rules of the game in Burundi are still not fully agreed upon. The political elite remains divided. The ruling party has yet to fully embrace democratic norms and continues to use the police for political ends.¹ Moreover, for many Burundians, a large rift persists between the security sector and society at large. In parts of the country, the public harbors a strong resentment of the security sector, especially the police, whom many perceive as agents of repression.² At times, Burundians’ lack of confidence in the security sector has resulted in a willingness to resort to vigilantism or “mob justice.” Perpetrators of such violence have justified the need to take the law into their own hands on the basis of police corruption, incompetence, and favoritism.³ In order to consolidate peace and security for its citizens, Burundi would need to embark on an ambitious security sector reform (SSR) program.

The Burundi-Netherlands Security Sector Development (SSD) program aimed for such transformative change when it was launched in 2009 with its explicit support for the development of more democratic and accountable governance of the security sector. The program had little in the way of successful models to draw on, however. Most previous SSR efforts had focused on training and equipping security forces and given little sustained attention to strengthening governance of the sector. The SSD program, therefore, had to break new ground in integrating democratic governance objectives into the existing SSR paradigm.

What is Security Sector Reform and Why is it Important?

The SSR concept was developed in the 1990s to inject a governance component into traditional security assistance. It was based on two closely linked relationships. First, it recognized that a safe and secure environment engenders successful economic and political development. Second, a safe and secure environment requires sound governance of the security and justice sectors. Countries where governance of the security and justice sectors has been suboptimal have tended to experience higher rates of impunity by security and justice sector actors. As a result, the state and its citizens are less secure.⁴

The centrality of governance to the SSR agenda has since been reconfirmed by actors as diverse as the United Nations, the African Union, the Organization for Economic Co-operation and Development (OECD), and the U.S. Department of the Army.⁵ This has resulted in broad agreement conceptually that the objective of SSR is to create a nationally owned process to promote the effective and efficient provision of security and justice in line with citizens’ needs and where security and justice providers are accountable to the state and its people, operating within a framework of democratic governance, rule of law, and respect for human rights.⁶

Implementing SSR

While there has been growing appreciation for the importance of SSR, there have been few clear successes in implementation. There is still a strong preference within security assistance programs for a traditional “train and equip” approach that is very light on the governance aspects of SSR. There are good reasons to provide operational support. However, if not also linked to improving governance, operational support alone risks bolstering the repressive capability of unaccountable security forces. In fact, many SSR programs have lacked either the ambition or the capacity to become transformative initiatives aimed at improving the democratic governance of the security sector.

In addition to limited awareness and political will, there are several geopolitical reasons why governance has taken a backseat to operational support.

• The “war on terror.” In 2001, the SSR agenda was in its infancy. An approach prioritizing governance had not yet taken root among the major donors of security and development assistance. It was therefore easy to default to a Cold War-era approach to security assistance.
• The financial crisis. Donors are increasingly focusing on “value for money” approaches to programming that require measurable outputs and predictable rates of expenditure. This approach does not mesh well with lengthy, unpredictable, and difficult-to-quantify programs to strengthen security sector governance.
• Failure of important international actors in the security arena to buy into the SSR concept. While OECD donors may express varying degrees of support for security sector governance, nontraditional donors—such as China, Russia, Turkey, and the Gulf States—are almost uniformly focused on building operational capabilities.

Given SSR’s limited track record of implementation, one program stands out for the lessons it holds of how SSR can be applied in a fragile, conflict-affected state: the Burundi–Netherlands Security Sector Development (SSD) program.

SSD—Promoting the Governance Agenda

The SSD program was established in April 2009 with the signing of an 8-year Memorandum of Understanding (MoU) between the governments of Burundi and the Netherlands. An important impetus on the Burundian side was the military’s interest in reconstituting and professionalizing the force and rebuilding its reputation after the war. The political leadership may have also viewed the SSD program as a means of diversifying their support away from Belgium and France and of enhancing the country’s international standing. Thus, the MoU laid out a number of strategic objectives for the program, all of which would strengthen the governance of Burundi’s security sector (Figure 1).

Structure and Management

The SSD program has been implemented in 2-year phases by a program team made up of international advisors and Burundians. Burundians have played an increasingly greater role in the management of the program over time. The program consists of three pillars: defense, public security, and governance. During the first 4 years of the program, the governance component was managed by a very small Program Management Unit (consisting of a program director and an assistant) and a Governance Advisory Group. In late 2014, two project managers were added to the PMU. The advisory group consists of representatives of Parliament and other oversight bodies, the executive branch, the military, the police, the intelligence service, and civil society, thus reflecting the cross-cutting nature of security sector governance.

The program has a three-tiered management structure: 1) the ministerial-level Political Committee to conduct high-level political dialogue between the partners (Figure 2); 2) the senior official-level Steering Committee, charged with approving work programs, monitoring their progress, and discussing program strategies and budgets; and 3) the working-level Program Management Units, one for each of the program’s three pillars—charged with developing and overseeing projects and guiding the future direction of the program. Each pillar also has a National Coordinator who acts as the link between the program and the government on a day-to-day basis. The work of the SSD program has been supported by an international consulting firm charged with providing financial management and administrative and technical support to the execution of the program.

Approach

The SSD program is guided by loosely defined strategic objectives (Figure 1) and a shared vision between the two governments of a transparent, accountable, democratically governed, fiscally sustainable security sector delivering security and justice to all Burundian citizens.⁷ The program had no logframe, results framework, or business case at the outset. Rather, it adopted a highly flexible problem-solving approach, taking conditions on the ground as its starting point and building on them to progressively achieve the shared vision and objectives. At the end of each 2-year phase, the program has evaluated its progress and adjusted accordingly. To this end, the SSD program began with concrete activities during Phase 1 (2009-11) to build the trust and relationships that would later be necessary to tackle the thorny governance issues involved in security sector reform. For example, the program helped refurbish army kitchens and build the capacity of the police to maintain their communications equipment. SSD has continued to support activities that enhance the operational effectiveness of the military and the police. One of the major activities of the defense pillar, for example, was a defense review conducted between 2011 and 2014. Program priorities between 2014 and 2017 include, among other things, an improved vehicle maintenance system for the police, building the police’s counterterrorism capacity, and enhancements to the army’s logistics system.

Having established credibility by generating tangible benefits to Burundian security actors, the SSD program progressively promoted activities to change the attitudes and behaviors of key Burundian actors consistent with democratic control of the security sector. A 2013 evaluation of the SSD governance component characterized the main features of the SSD program as:

• an integrated approach to security encompassing defense, public security, intelligence, and governance with links to the justice sector
• a demonstrated commitment to enable the defense and security forces to respond to the population’s security needs but also a recognition of the value of nonstate security providers and the need to better understand their role
• a focus on strengthening the capacity of key security actors to engage in the process of reform—for example, by educating them on governance-related issues (such as the role of Parliament in overseeing the budgeting process or improving the quality of the security sector’s legal framework) and by providing critical infrastructure (vehicles, computers) and technical assistance (expertise in drafting legislation and developing defense policy and strategic documents) to enable security sector actors to fulfil their legally mandated tasks
• a focus on strengthening the integrity of key security actors with a view to ending impunity—for example, developing a code of ethics and ethics courses for the police and the armed forces, promoting dialogue among security personnel about the application of the code, and inviting civil society actors to take part in evaluating troops’ adherence to ethics norms when dealing with civilians
• a dedication to ensuring the sustainability of results generated by the program
• a mechanism to coordinate all SSR interventions in country to maximize the complementarity of activities and objectives.⁸

What the SSD Program Achieved in its First 4 Years

While the MoU clearly gave a high priority to effecting change in security sector governance, it was by no means assured that this would be the outcome. Skilled and dedicated program staff and conducive political dynamics within the governments of the Netherlands and Burundi enabled the SSD to evolve toward addressing difficult political challenges facing the development of the security sector in Burundi. However, despite important gains in creating the foundation for democratic governance of the security sector, the ultimate outcome in terms of greater accountability and transparency is far from certain. Nonetheless, the SSD program demonstrates better than any other SSR effort to date that an explicit governance-focused approach to SSR is possible.⁹

The SSD program has not been the only security and justice sector program in Burundi. Activities ranging from police and justice reform to training the armed forces have been carried out in conjunction with a range of partners including Belgium, France, Germany, the United Nations, and the United States.¹⁰ The SSD program has, however, most consistently addressed the governance aspects of SSR. As a result, it can be credited with contributing to the changes in the governance environment which were evident by late 2013 (Figure 3). These changes have helped increase the potential for
transparency and accountability in the sector, improved the perception of the security bodies (especially the army) among the population, and strengthened the quality of security provision (for example, by helping to reduce the incidence of extrajudicial killings through policy dialogue with the Ministry of Public Security and the National Police).

While all parts of the SSD program have contributed to these outcomes, the program would have recorded considerably less progress in the area of governance without a mechanism for ensuring that governance-related issues could be addressed on an equal footing with capacity building for the security bodies. SSD’s governance pillar has been able to support capacity building of important oversight actors such as Parliament, civil society, the media, the Office of the Ombudsman, and the Constitutional Court. It has also been able to play an important role in sensitizing the defense and public security actors to governance needs.

Key Takeaways of How the SSD Program Promotes Better Security Sector Governance

There are four main reasons why the SSD program has improved security and justice governance in Burundi: 1) the program has put politics center stage; 2) it has established results progressively; 3) it has prioritized the gradual development of ownership; and 4) it has matched time-frame with ambition and environment.

Putting Politics Center Stage

The success of SSR programs depends heavily on the ability to address politically sensitive issues. This ability is particularly important for addressing the governance aspects of SSR. For the most part, SSR programs have fared poorly in this regard. The SSD program, however, proactively addresses the politics of change at both the policy and operational levels on a daily basis.

The MoU placed considerable importance on political dialogue (discussing how best to implement the program and overcome any political constraints). Although the MoU envisioned a partnership between the two governments, it took them some time to find the right balance in their relationship. Now, political dialogue occurs between the two MoU governments, between the Burundian government and other international partners, and among Burundian stakeholders. It is clear from the SSD program results that there can be no effective SSR unless the political challenges to change processes are squarely addressed.

The success of this dialogue process has required a balancing act on the part of the SSD program, facilitated in large part by the political skill of program staff. Because the program supports both governments in implementing the MoU, the program must make sure the Dutch political agenda does not drive programming decisions (other than setting limits on the use of funding). The program must also make sure the Burundian government adheres to the objectives of the MoU and not simply see the SSD program as another source of funding to meet the material needs of the country’s security sector. The program’s relationship with both governments has experienced some hiccups, but on the whole, the program has succeeded in maintaining the necessary balance between pushing forward program objectives and maintaining a relationship of trust.

The engagement of the Dutch government through its embassy in Bujumbura has been crucial in addressing potential blockages to program activities. The Dutch Embassy has taken responsibility for addressing the political issues surrounding highly contentious activities aimed at implementing the MoU. In doing so it enables the program to focus on helping the parties arrive at solutions rather than becoming part of the political dispute.

There has clearly been an evolution in the Burundian government’s engagement, one that is linked in no small measure to the Burundians’ growing responsibility for managing the program. The National Coordinators of the three SSD program pillars, each of whom represents a government ministry, are playing an increasingly important role, including in discussions between the partner governments on contentious political issues.

Nonetheless, the commitment of senior Burundian political leadership to democratic security sector governance remains uncertain. In order to address this concern, the SSD program began to implement a two-pronged strategy in 2014: 1) decentralize SSR by educating local government officials and the population beyond Bujumbura on SSR objectives such as the importance of transparency, accountability, and respect for human rights as well as the roles of the respective actors in creating a safe and secure environment for all Burundians; and 2) develop mechanisms to enhance the understanding of these same issues among the most senior cadre of Burundian political leaders.

Establishing Results Progressively

The SSD program develops work plans on a 2-year basis and within each 2-year phase it can respond to emerging needs. Accordingly, the program is not tied to objectives that made sense in year one but are no longer feasible or salient in year five. This is particularly important for governance where issues continually surface and the ability of the program to address any one of them evolves as trust and relationships mature. This flexibility proved helpful when the Burundian government decided that it wanted to conduct a defense review midway through Phase 1 and again when the National Intelligence Service and the National Security Council expressed interest in participating during Phase 2.

The SSD program has been able to define and achieve results progressively because its programming approach is flexible (not tied to a logframe, for example), it has a long-term (8-year) time horizon, and relationships of trust have been built between the two governments and the program. Trust building is facilitated by the program’s ability to operate simultaneously on two distinct but interrelated tracks. It provides tangible benefits (training, certain types of equipment, infrastructure) prioritized by national stakeholders at the same time as it assesses the most appropriate way to address the highly political transformational aspects of change (Figure 4). Additionally, the program has leveraged delivery of these tangible benefits to open windows of opportunity for addressing longer-term change issues. For example, the SSD program agreed to construct a building for the Constitutional Court, which has been housed in a particularly decrepit building, with the understanding that the government accord budgetary independence to the Court by 2016 as a first step in severing the administrative ties between the Court and the Ministry of Justice.

Prioritizing the Gradual Development of Ownership

Responsibility and authority for identifying the strategic direction, developing activities, and managing the program have been progressively transferred to Burundian stakeholders. Burundians are increasingly the public face of the program. The ultimate objective is an entirely Burundian-managed program, which Burundians appreciate and compare very favorably with other internationally supported security and justice programs (Figure 5). Ownership is not just a matter of the government leading a change process, however. If the security and justice needs of all citizens are to be addressed, then ways of involving the broadest range of actors in the process need to be found. From the outset, the intention was to engage civil society in the SSD program. However, the inclusion of civil society organizations (CSOs) was particularly contentious, and not only, as might have been expected, with the security forces. Some parliamentarians in the defense and public security committees initially believed that CSOs had no right to participate in security discussions with governmental officials. Since CSOs were not elected by the population to represent them, it was argued, they had no legitimate role to play.

The key to changing attitudes was dialogue. Thus, the SSD program (particularly its governance pillar) sought opportunities to bring the diverse actors together. Skilled facilitators were employed to make the dialogue as constructive as possible. The experience of other African countries was examined, often with the support of individuals from those countries or by study trips abroad. Senegal, a francophone country also undergoing a democratic transition, was particularly relevant in this regard. Reflective of an emphasis on South-South engagement, Senegal hosted a study visit for Burundian parliamentarians. A Senegalese legal advisor also provided training on legal drafting for parliamentarians. When international experts were involved, they were often paired with Burundian counterparts, accelerating the sharing of insights and lessons learned in both directions.

Over time, the concept of human security began to take hold. Once it became evident that security had multiple dimensions beyond a purely military orientation, it became easier to accept the roles that other actors could play in creating an environment of security for all Burundians. And as the different program stakeholders engaged with one another more frequently, they came to understand and appreciate the different contributions each brought to the table (Figure 6).

Within the SSD program there is now acceptance of the importance of discussing security issues outside the restricted circle of uniformed service personnel. Security has become “everyone’s affair.”

Matching Timeframe with Ambition and Environment

SSD’s 8-year timeframe has enabled Burundian stakeholders to understand what SSR implies and begin to adjust attitudes and behaviors so that sustainable change can take root. Yet, attitudes and thus behaviors typically change gradually. Moreover, because governance raises inherently sensitive subjects, it is unrealistic to assume that results will be achieved quickly. The SSD program shows that what may seem like a very minor change to established democracies can, in fact, be a substantial milestone in countries just setting out along the road to democratic governance. Numerous stakeholders have noted that had the SSD program tried to address the governance-related structural issues foreseen in the MoU (e.g., transparent budget management, legal framework of the security sector) from day one, it would have failed. Part of the problem was the unwillingness of key political actors to accept change. But also critical was that key Burundian stakeholders simply did not understand what the necessary reforms consisted of, why they were important, and how to implement a process of change. As one Burundian stakeholder stated, when people do not understand something, their default position is “no.” Once they understand why change is needed and how they can support the process of change, they become more open to it. This process of education, however, is a lengthy one, which is why The SSD program’s 8-year timeframe is so important.

Notes

1. ⇑ For a review of the Burundi peace process see Howard Wolpe, “Making Peace after Genocide: Anatomy of the Burundi Process,” Peaceworks no. 70 (Washington, DC: U.S. Institute of Peace, March 2011).
2. ⇑ Willy Nindorera, Security Sector Reform in Burundi: Issues and Challenges for Improving Civilian Protection, Centre d’Alerte et de Prévention des Conflits (CENAP)/North-South Institute (NSI) Working Paper (Ottawa: NSI, July 2007). Willy Nindorera, “La police nationale, le renforcement démocratique et la consolidation de la paix au Burundi,” Canadian Journal of Development Studies/Revue canadienne d’études du développement 32, No. 1 (March 2011), 79-93.
3. ⇑ Mob Justice in Burundi: Official Complicity and Impunity (New York: Human Rights Watch, March 2010).
4. ⇑ Louis-Alexandre Berg, “Security Governance and the Recurrence of Violence after Civil War” (draft paper prepared for presentation at the APSA Annual Meeting, Washington, DC, August 28, 2014), cited with permission from the author. Berg examines all cases of civil war termination since the Cold War and concludes that “changes to security sector governance that broaden civilian oversight and reduce factional control reduce the chance that civil war recurs….”
5. ⇑ Report of the Secretary-General on “Securing peace and development: the role of the United Nations in supporting security sector reform” (UN doc. A/62/659–S/2008/39[)], January 27, 2008). African Union Policy Framework on Security Sector Reform, African Union Commission (adopted January 2013), 6. Security System Reform and Governance: Policy and Good Practice, Development Assistance Committee (DAC) Guidelines and Reference Series (Paris: OECD, 2005), 12. U.S. Army, Stability Operations, FM 3-07 (Washington, DC: Government Printing Office, 2008).
6. ⇑ “SSR Glossary,” The Geneva Centre for the Democratic Control of Armed Forces’ International Security Sector Advisory Team.
7. ⇑ Mémorandum d’Entente entre le Gouvernement de la République du Burundi et les Ministres des Affaires Etrangères, de la Coopération au Développement et de la Défense des Pays Bas sur la Développement du Secteur de la Sécurité, April 9, 2009.
8. ⇑ Nicole Ball and Fabien Nsengimana, Evaluation du Volet Gouvernance du Programme DSS, October 28, 2013. Nicole Ball, Putting Governance at the Heart of Security Sector Reform: Lessons from the Burundi-Netherlands Security Sector Development Program, Conflict Research Unit (CRU) Report (The Hague: Netherlands Institute of International Relations Clingendael, March 2014).
9. ⇑ Ibid.
10. ⇑ Nicole Ball, Jean-Marie Gasana, and Willy Nindorera, From Quick Wins to Long-term Profits? Developing better approaches to support security and justice engagements in fragile states: Burundi case study (Paris: International Network on Conflict and Fragility (INCAF), March 2012).

Nicole Ball is a Senior Fellow at the Center for International Policy. She has worked on issues relating to security sector reform for over 30 years and conducted assessments and evaluation relating to democratic security sector governance in a dozen African countries.

⇑ back to top ⇑