THE USAID/OTI COMMUNITY-FOCUSED 
REINTEGRATION PROGRAMS IN THE 
DEMOCRATIC REPUBLIC OF CONGO 
AND BURUNDI
FINAL EVALUATION
Prepared for the Office of Transition Initiatives
Bureau for Democracy, Conflict, and Humanitarian Assistance
United States Agency for International Development
Submitted by;
SOCIAL IMPACT, INC.
HDA-I-20-03-000124-00
February 24, 2006
2
TABLE OF CONTENTS
Acronyms 3
Executive Summary 4
Part A: The OTI DRC Program (SE*CA) 12
 
I. Background 12
II. Evaluation Objectives and Methodology 13
III. Findings 14 
IV. Conclusions 28
V. Recommendations 30
Part B: The OTI Burundi Program (CPRI) 32
I. Background 32
II. Evaluation Objectives and Methodology 32
III. Findings 33
IV. Conclusions 48
V. Recommendations 49
Part C: The OTI Burundi and DRC Programs Compared 51
Annex A: List of Persons Interviewed 58
Annex B: Selected Bibliography 62
Annex C: Evaluation Scope of Work 64
3
ACRONYMS
AAA Agro Action Allemande
ASI African Strategic Initiative
BLTP Burundi Leadership Training Program
BNL Business, Numeracy, and Literacy
CBLP Community-based Leadership Program
CEI Commission Electorale Indépendante (Congo)
CENI Commission Electorale Nationale Indépendante (Burundi)
CESVI Cooperazione e Sviluppo – Italy (Cooperation and Development)
CFR Community-focused Reintegration
CI Community Initiative
CIAT International Committee for Accompaniment of the Transition
CF Community Facilitator
CGIC Comité de Gestion de l’Intégration Communautaire 
CONADER Commission Nationale de Désarmement, Démobilisation et Réintegration
CMM Office of Conflict Management and Mitigation (USAID)
CNDRR Commission Nationale pour la Démobilisation, Réinsertion, et Réintegration
CPRI Community-based Peace and Reconciliation Initiative
DDR Disarmament, Demobilization and Reintegration
FARDC Forces Armées de la République Démocratique du Congo
IDP Internally Displaced Person
IEC Independent Election Commission
IFES International Foundation for Election Systems
IRI International Republican Institute
MDRP Multi-Country Demobilization and Reintegration Program
M&E Monitoring and evaluation
MLC Mouvement de Libération du Congo
MONUC Mission des Nations Unies en République Démocratique du Congo
NDI National Democratic Institute
OMAC Organisation des Média de l’Afrique Centrale
ONUB Mission de l’Organisation des Nations Unies pour le Burundi
PADCO Planning and Development Collaborative International
PMP Performance Monitoring Plan
PO Program Officer
QIP Quick Impact Projects 
RCD Rassemblement Congolais pour la Démocratie
REDSO Regional Economic Development Services Office (USAID)
RSF Radio Sans Frontière (Bonesha)
RTNB Radio Télévision Nationale Burundaise
SE*CA Synergie d’Education Communautaire et d’Appui à la Transition
SSI Semi-structured interview
VST Vocational Skills Training
YES Youth Education and Skills Program
YRTEP Youth Reintegration Training and Education for Peace Program
WWICS Woodrow Wilson International Center for Scholars
4
EXECUTIVE SUMMARY
The OTI country programs evaluated here are two of three Community-focused Reintegration (CFR) 
programs launched by OTI in early 2004 in Burundi, Democratic Republic of Congo, and Liberia.
All three are based upon an earlier approach used in Sierra Leone. As OTI prepared to hand over 
program activities in Burundi and DRC to other USAID entities or donors, it sought to explore the 
following questions through an independent evaluation:
(1) Did OTI contribute to the advancement of transition processes in Burundi and DRC? If so, how 
and in what ways could it have improved this?
(2) Were the programs strategic in responding to shifts in the transition processes?
(3) Did the programs meet their stated goals and objectives?
Beyond the evaluation of two country experiences, this evaluation also seeks to shed light on the
appropriateness of CFR in OTI programming. 
The two-person evaluation team of Philip Boyle and Edward Rackley carried out preliminary
interviews concerning the DRC and Burundi programs directly or by telephone in Washington, D.C. 
from October 31 through November 4, 2005. In the field, the team was in DRC from November 6 
through November 18 and in Burundi from November 19 to December 1. In both countries, team
visits began with briefings and interviews in the capital city, then moved to project sites in the
interior. Visits were made to activities or accomplishments of all program components. 
OTI DRC
The political transition in DRC has been underway since April 2003. National elections, the
culmination of this process, are scheduled for April 2006. Phase 2 of OTI’s DRC program, Synergie 
d’Education Communautaire et d’Appui à la Transition (SE*CA), began in March 2004 and will
conclude in July 2006, unless USAID/DRC provides funds to extend the program to July 2006 or 
beyond. However, DRC’s transition is far from over, and the basic democratic requirements of
minimum state capacity, order, and disincentives to violence are still lacking. These absences are 
compounded by citizens’ distrust of transition actors and manifest distrust among the political actors 
themselves. None of this uncertainty should eclipse an array of more positive indicators for which
OTI can claim some responsibility.
Conclusions
1. OTI’s use of the Community-focused Reintegration (CFR) approach allowed a focus on entire
communities, not the ‘peace spoilers’ alone, as do traditional DDR programs. In highly militarized
areas with ex-combatants, OTI contributed to reintegration by first gaining the trust of communities
through its YES trainings. In areas where there were no ex-combatants, the capacity-building and
community grants components of SE*CA achieved their objectives of reintegration and
reconciliation of community members through broad-based, participatory activities. 
5
2. Working at the national and local levels, OTI opened up long-isolated communities, reconnecting
and implicating them in national political processes, such as voter registration and the constitutional 
referendum. Many communities reported that reconnecting with the outside world through the
various modes offered by OTI was the first real peace dividend they had experienced since the
transition began in 2003.
3. As a result of OTI’s intervention, beneficiary communities now grasp that a successful political
transition requires civic engagement and public accountability, actions for which they alone are
responsible. It did not escape them, however, that OTI – not the national government – created
SE*CA and delivered its messages, projects, and hope to the community.
4. OTI stabilized and invigorated war-torn communities through YES group trainings, labor￾intensive, community-led rehabilitation projects, and Radio Listening Clubs. The multi-component 
CFR approach proved its relevance in the Congolese context, where small grants alone are not
enough to stabilize communities and increase transition awareness. 
5. The program contributed to the reintegration of ex-combatants in highly militarized areas, such as 
Ituri and Maniema. This it did by first gaining the trust of communities and by spending significant 
time with them. In areas where ex-combatants or military deserters were present but not officially 
recognized, OTI prepared communities to receive and accept ex-combatants once demobilized. In
areas where there were no ex-combatants, the YES training facilitated a process of reintegration 
between communities and local authorities, whose relations had been marked by distrust, mutual 
exploitation, and fear.
6. OTI also facilitated the reconciliation of stigmatized and disenfranchised groups with other 
community members. Such reconciliation included: (1) rape survivors and their husbands, families
and the wider community; and (2) former combatants, military deserters, and child soldiers with
their families and the wider community. YES training modules laid a solid foundation for conflict
resolution without recourse to violent revenge, a common mode of settlement in rural and urban
communities alike. 
7. The six-month YES capacity-building course clearly galvanized community members, who
remained poised for further interaction with SE*CA representatives. Once the training was over,
however, SE*CA moved on to new areas leaving a serious void. This was true even in communities 
that had received some small grants. In Cycle 2 communities, however, budget cuts severely 
reduced these projects, leading to severe disappointment and anger. 
8. Regarding preparations for national elections, including the recent constitutional referendum, OTI 
was instrumental in mobilizing voter registration. Where YES trainings occurred, turnout for voter
registration was 80%; where there was no training or sensitization, people resisted and police were
brought in to force them to register.
9. OTI was able to be flexible and increasingly appropriate in local media strategies. Since OTI’s 
internet centers revealed themselves to be ineffective tools to support the transition, it developed
community-based Radio Listening Clubs. This is to say that internet centers were not the most
6
effective way to reach the masses, since they were urban-based and required user literacy. OTI/DRC 
notes that the SE*CA team’s September 2005 analysis found that the internet centers were effective 
in supporting the transition. They were difficult to execute, but if they functioned and if they had 
community bulletin boards where they posted the latest news, the centers proved to be very popular 
and useful to disseminate information. However, besides their simple technology, Radio Listening 
Clubs are effective information vehicles, more democratic than internet centers, and provide for
wider rural community involvement.
10. The SE*CA Network and the CONADER follow-on grant to USAID ensure that two OTI
transition support strategies – information flow via community management committees and YES
training to ex-combatants – will continue after OTI closes in July 2006. Neither handover strategy
existed at project inception; both were developed in a timely manner before program end. 
11. By not integrating transition actors and institutions, particularly Congolese political figures, into
its scope of activities, OTI missed a key opportunity to use Kinshasa as a theatre of engagement, one 
with direct access to the political actors and institutions most influential to the outcome of the
transition process. It is true that OTI chose not to work with national actors and institutions, because 
the USAID Mission Office of Democracy and Governance worked with most of the national
institutions in Kinshasa, leading OTI to focus on the “grassroots” issues. Whether this division of
labor (synergy) has been effective remains to be seen as the transition proceeds. 
It should also be noted OTI/DRC feels that its work with CONADER and the Independent Election
Commission (IEC) at the national level is highly significant. Both institutions were involved in
approving and using YES modules. The SE*CA program also influenced CONADER to change to a 
community approach to reintegration of ex-combatants rather than focusing on ex-combatants alone. 
Moreover, OTI designed the national communication campaign for the IEC, and the program’s
Learning Facilitators provided a large percentage of the registration, sensitization, and polling
workers for the IEC during elections registration and the constitutional referendum. OTI also
coordinated closely with the Haute Autorité des Media on media programs.
12. In terms of geographic focus, OTI’s absence from North and South Kivu provinces, the key 
indeterminate variable in the transition, constitutes another significant missed opportunity. Although 
OTI did undertake some actions in South Kivu after security improved in March 2005 (e.g., 
journalist training), it did not place the whole SE*CA program there. Under the IRC follow-on in
South Kivu, OTI has helped launch a program in 47 communities reaching some 3,600 people.
Finally, with regard to program timeline, OTI’s premature departure relative to national elections is 
also unfortunate; the transition is far from over. 
Recommendations
1. In countries whose political transition is protracted and inconclusive, such as the DRC, if OTI is 
not prepared for a longer commitment than the standard 2-3 years, then an effective handover
strategy should be developed as early as possible. OTI/DRC did manage to ensure a short-term 
handover of its CFR capacity-building approach. Moreover, punctual interventions at key junctures –
particularly before, during, and after elections – should be prioritized over sustained interventions
that force a premature withdrawal when funds are exhausted. While OTI/DRC does not feel its
7
departure is premature, this view is not shared by the USAID Mission Director. Since some degree 
of sustained interventions is inevitable, these should be kept as flexible and light as possible to avoid 
onerous recurrent cost commitments. 
2. CFR is appropriate in contexts such as the DRC, where the national DDR program is poorly run. 
Reintegration and reconciliation are community concerns as well, and addressing these issues at the
community level serves the added purpose of mollifying local tensions at the core of the conflict.
When well executed, CFR can promote reconciliation and provide the tools to prevent local conflicts 
from recurring. 
3. The content of the YES training covers three modules dealing with values, conflict resolution,
and democracy/governance, plus two that attempt to inculcate life skills: health/nutrition and
agriculture/income-generation. This long training period reflected a strategy to keep people engaged 
and busy, but expectations were often raised to unrealistic heights and most grants came late (or not 
at all) in this long process. Consequently, the last two modules might better be addressed through
separate follow-on activities (through small grants) to trainees and other community members, thus
shortening the long period of initial ‘classroom’ training to about three months. 
4. Based on the DRC experience, CFR appears to work best when community trainings are
combined with strategic small grants. Since collective rehabilitation projects are best conceived as
the practical application of reconciliation and conflict resolution messages learned in community
capacity-building, OTI should ensure that future CFR programs effectively integrate capacity￾building and small community grants. Communities receiving training but no grants in the DRC
suffered from raised expectations and significant bitterness towards the SE*CA program.
5. Greater integration of the capacity-building, media, and grants components of the program would 
have been desirable and should be carefully considered in future CFR programs in Africa and
elsewhere. This primarily means linking grants more closely to capacity-building and being sure
communities choose and receive what they need. Radio Listening Clubs, moreover, should be open 
to a wider membership among capacity-building graduates. Other activities that should be included
involve practical, not theoretical (as in YES training), income generation and skills development, but 
these activities should remain small in scale and appropriate in technology. The linkage with state￾of-art micro credit programs such as Project Hope in Kisangani should be emulated in future
programs.
OTI Burundi
OTI began working in Burundi in March 2002 in response to significant advances in the peace
process after nearly a decade of civil war. These advances included the Arusha Peace and
Reconciliation Agreement (APRA) in August 2000 and the establishment of a transitional
government in November 2001. 
Phase 2 of the OTI Burundi program, the Community-based Peace and Reconciliation Initiative
(CPRI), began in February 2004 and was to conclude in February 2006, but it has recently been
extended through June 2006. This four-month extension involves all components except the
capacity-building component, which will end as scheduled.
8
Conclusions
1. The CPRI program in Burundi, one of three community-focused reintegration (CFR) projects
launched by USAID/OTI in early 2004, successfully achieved its objectives, strategically followed 
the political transition, and made a significant contribution to the peace and transition process
unfolding in Burundi.
2. The Community-Based Leadership Program (CBLP) was highly innovative, adapted from a
proven model available at the national level, effectively targeted on traditional leaders and persons
of probity and influence, flexible in communal location, and not demanding in time and effort
commitment from community members. High trainer salaries and overlapping organizational
overhead costs rendered this program more expensive than it might have been. 
3. The media component was highly effective in having both national and community-focus, 
remained highly targeted, programmatically flexible, and transition responsive through quarterly
granting, planning, and evaluation with one public and one private radio station partners. Media
content monitoring was also effectively carried out during the elections process and aftermath. 
4. While CPRI was strategically designed to focus several interrelated components on Burundian
communities, these components tended to function largely independently. While this might be
expected of the media component, linkages between community capacity building (CBLP),
community grant initiatives (CI), and vocational skills training (VST) might have been tighter. 
 
5. The community initiatives component proved to be an effective means to bring formerly warring
or feuding ethnic groups, refugees, internally-displaced persons, and persons that had remained in
communities together on construction sites for projects of community interest and ownership. The
choice of specific projects remains problematic, and it is not clear that communities received the
most useful infrastructure for their future well-being. 
6. The program was least successful in integrating a vocational skills training component (VST) into 
its strategic mix. Rather than remaining light and mobile, bringing skills training out to key sites in a 
wide variety of communes, CPRI chose to go deep and fixed in eight locations, engaging in
substantial rehabilitation of former vocational training schools abandoned a decade earlier.
Originally seen as the primary draw for ex-combatants and disaffected youth, demobilization did not 
keep pace with creation of the VST Centers limiting ex-combatant participation. 
7. Both the capacity building and vocational skills training components were expensive, particularly
the latter. The VST Centers entailed not only high investment cost, but imposed a heavy recurrent 
cost burden on the flow of OTI funding. This ‘mortgaging’ effect drained funds away from more
flexible and mobile components, particularly the community initiatives and the media components. 
8. While the VST trainees were clearly mixed by sex, age, ethnic group, and refugee status, the real 
objective of vocational skills training was to launch a process of economic development, without
which the designers of CPRI did not believe future internecine conflict could be prevented. While
this may be true over the long term, OTI prides itself on providing fast, flexible, and short-term 
9
assistance to mitigate conflict and promote political stability. An investment such as the VST
Centers does not correspond to this self image and seriously constrains OTI flexibility and funding,
as it seeks to remain strategically focused on an unfolding political transition process. 
9. From the beginning CPRI chose to concentrate its various program components in only two
provinces, neither of which for security and other reasons was the most appropriate for CPRI
activities. The justification for such narrow focus was that critical synergy could be achieved, but
this has not been borne out. Whether narrow focus led to deep programming or vice versa, it was 
soon impossible for CPRI to spread its benefits to at least two perhaps four more provinces as the 
peace process evolved. 
10. In spite of no fewer than five M&E approaches, ranging from contractor biweekly assessments, 
OTI regular assessments, biannual stakeholder assessments, CBLP biannual assessments, and OTI
mid-term and final evaluations, key impacts on communities remain anecdotal. This is true for all
components, but is particularly problematic for the CBLP and VST components, which together
account for about two-thirds of the program. A baseline assessment was conducted, but follow-up 
biannual assessments appear to have been abandoned after the first follow-up exercise. 
Recommendations
1. The CFR approach used in Burundi contains a mix of components that could serve as a model for 
other countries, if lightened and reduced in cost. Cost reduction should be applied to both the
capacity-building and skills training components, particularly the latter. An income-generating or
skills development component should be included in CFR, but it must be kept as flexible and mobile 
as the capacity-building and community initiatives components. Moreover, all three components
need to be more integrated in communities. Tighter linkage of capacity building, community grants, 
and vocational skills training could be obtained by having the Master Trainers, who actually lived in
the communities, be the center point for all community-level activities. 
2. The income-generating and skills development component could be kept more appropriate to the 
short-term, rapid, flexible, and responsive OTI approach by organizing community enterprises or
associations alongside community reconstruction projects and some type of capacity building for
perceptual and behavior change. In so doing, it would not require anywhere near the level of
physical investment and recurrent cost mortgaging that occurred in the Burundian program.
Moreover, major economic development infrastructure need not be provided, if skills and relevant
local production are supported. Infrastructure that is provided should be oriented to water, health, 
and education needs. 
3. The capacity building component is short and flexible enough to remain highly relevant to a
reconciliation and reintegration process, without posing an undue time burden on participants. It is 
not clear, however, that it is a more effective tool in promoting community reconciliation and
reintegration than community initiatives. Nevertheless, a combination of capacity building of this
type and joint community endeavors for the common good appears far more effective than either one 
alone.
10
4. Since it is an adaptation of the Burundi Leadership Training Program (BLTP), CBLP requires 
significant reinforcement through further interaction, training, and collaborative actions. In future
programs, OTI should make a concerted effort to follow-up on its impacts on communities. Ensuring 
the availability of one or more community initiatives to participating groups is highly recommended.
This means envisaging small-scale activities spread over a much wider area than was done. Some of
these may be income-generation or vocational skills development activities with the new
‘associations’ created by CBLP trainees. 
5. If OTI intends to prioritize community-based capacity building in post-war reintegration efforts, 
it must make a concerted effort to evaluate the real impact of the various capacity building 
experiences and their relative degree of success. It set out to do so in Burundi, but failed to follow up 
on the many issues that were identified in the baseline. While the testimony of community
participants is overwhelmingly positive, it is a little hard to believe that such training can have 
significant and enduring cultural impact if carried out on such small scale and with little continued
interaction. 
6. Given the experience in Burundi, national-level activities should accompany the CFR approach in 
future programs elsewhere. The media component had wide reach and effectively linked the
transition in the capital with reintegration of communities. Institution building of emerging
government structures should be included in future OTI interventions alongside capacity building of
national leaders. All of this allows OTI to intervene in a variety of complementary levels and
sectors, varying its emphasis through time and adjusting its mix as the transition evolves. 
The CFR Approach
A comparison of the CFR approaches used in DRC and Burundi reveals the following:
1. Capacity building of communities in post-conflict situations appears to be very useful for
reconciliation and reintegration, but it must be integrated effectively with a local grants component
or community members will become frustrated or embittered. This disillusionment will certainly be
counter-productive to development efforts undertaken by the next set of donors.
2. The capacity building undertaken in Burundi would appear to be of about the right length for an
OTI-sponsored CFR program; a six-month program of the type employed in DRC seems overly long 
and might be reduced to about half that length with variable selection of modules to fit specific
community needs. The DRC experience revealed that the longer communities engaged in theoretical
training, the more frustrated they became when material rewards failed to arrive to the degree
anticipated. While valuing the content of the modules, community members in DRC felt the training
was primarily a stepping stone to local development projects. Six months with thrice weekly
meetings also entails a substantial opportunity cost for community members. Moreover, OTI/DRC
did shorten and vary YES training in some places and in its follow-on program in Ituri. 
3. Capacity-building of national leaders should also be carried out, if possible, as was done in
Burundi. Some governmental institution-building should also be undertaken. Thus, while CFR may
remain at the heart of a program, it should be linked to national-level transition strengthening or
community achievements will remain in a void.
11
4. Media activities at the national level are also extremely valuable to a transition process and
should be complementary to a CFR program carried out in communities. Training of journalists, 
carried out by OTI briefly in South Kivu and by USAID/DG and other donors unsuccessfully, should 
accompany support to journalist field visits. In the short term, it should be determined why journalist 
training was not successful in Congo. 
5. While a focus on the community level is appropriate in reconciling and reintegrating populations, 
OTI would do best to stick to a mobile, flexible, adaptable, short-term approach and spread the 
benefits as widely as possible. 
6. Vocational skills and micro enterprise development can and should be a part of a CFR support 
package, but these activities should be kept simple, technologically appropriate, and linked to
sustainable credit programs, where possible. 
12
Part A: The OTI Democratic Republic of Congo Program (SE*CA)
I. Background
Struggling to transition from dictatorship to democracy since its return to “multiparty” politics in
April 1990, the Democratic Republic of Congo’s (DRC) path to democracy remains extremely
rocky. The basic democratic requirements of minimum state capacity, order, and disincentives to
violence are still lacking. These absences are compounded by citizens’ distrust of transition actors
and manifest distrust among the political actors themselves. Nevertheless, none of this eclipses an
array of positive indicators for a successful transition, trends for which OTI can claim some
responsibility.
The Inter-Congolese Dialogue resulted in the adoption of a Transition Constitution and of the
“Global and All-Inclusive Agreement on the Transition in the Democratic Republic of Congo,” 
ratified in April 2003. These documents set up the principles and objectives of Congo’s transition,
including reunification, pacification, reconstruction, and restoration of territorial integrity and of
state authority over this territory, the creation of an integrated army, and the organization of
elections. This resulted in the formation in June 2003 of a Congolese power-sharing government
composed of members designated by the delegations attending the Dialogue. The new government 
had a “1 + 4” formula at its core which reserved the presidency for Joseph Kabila (son of late 
Laurent-Désiré Kabila, assassinated in January 2001) and appointed four vice-presidents.1
Together with a 40-strong executive, a parliament of 620 appointed members, and several
“institutions of support to democracy” (the Independent Electoral Commission, High Authority of
the Media, Truth and Reconciliation Commission, Ethics and Anti-Corruption Commission, and
Human Rights National Observatory), the new government was charged with organizing the
transition to democracy over a two-year period. Of particular importance also was the establishment 
of the International Committee for Accompaniment of the Transition (CIAT), composed of members 
of the UN Security Council, other donors, and some African governments, which was charged with
protecting transition institutions and arbitrating disputes. The government was appointed on June 30,
2003, inaugurating a two-year transition period which included among other things: the adoption of
a draft constitution; a law on the principles of nationality; an electoral law; registration of voters; a 
referendum on the draft constitution; and the election of a parliament and president. 
Despite the presence of over 16,000 UN peacekeepers in country, ongoing violence across the 
eastern provinces underscored the fragile state of the transition. Armed groups of disparate
allegiance failed to commit to the peace process. This insecurity limited OTI’s geographic reach into 
key areas. The primary flashpoints, North and South Kivu, remained inaccessible to OTI staff until
April 2005. Continued conflict and transition dysfunction meant that by early 2005, DRC’s 
transitional government had yet to pacify or extend its authority across its territory. National
 
1
 One from the previous government (Yerodia Ndombasi), one from each of the main rebel groups (Jean-Pierre Bemba 
for the Mouvement de Libération du Congo—MLC and Azarias Ruberwa for the Rassemblement Congolais pour la 
Démocratie—RCD-Goma), and one from the unarmed political opposition (Z’Ahidi Ngoma). Long-time political figure, 
Etienne Tshisekedi, leader of UDPS (Union for Democracy and Social Progress) and former Prime Minister under 
Mobutu, was excluded.
13
elections were delayed by six months, prolonging the transition process beyond OTI’s scheduled
departure date of March 2006. 
The Congolese transition has been characterized by a high level of institutional dysfunction and a
paucity of achievements. This led to its formal extension in June 2005 through one of two six-month 
extension periods allowed by the Constitution. Although the government, parliament and military
hierarchy were installed by September 2003, most of the “support institutions to democracy” took 
much longer to become operational, in part because of the government’s failure to adopt the
necessary legal texts. The quest for personal enrichment by the members of the transition institutions 
has triggered widespread reciprocal suspicion among them and made it all but impossible to
establish functional cooperation among the different organs of the state. 
Phase 2 of OTI’s current DRC program, known as Synergie d’Education Communautaire et d’Appui 
à la Transition (SE*CA), began in March 2004, when Chemonics International replaced CARE as 
implementing partner after 15 months of a previous 24-month program. The new phase was reset for 
24 months of implementation. With the national Disarmament, Demobilization, and Reintegration
(DDR) process slow to start, there were few official ex-combatants or spontaneously demobilized
soldiers with whom to work. In spite of this, OTI’s new program design did not drop their
reintegration where possible within its overall strategy of community stabilization and reintegration.
Following trials in Sierra Leone and Liberia, the Community-focused Reintegration (CFR) approach
was applied in the DRC, but it was re-tooled for the Congolese context with a focus on overall 
reintegration and reconciliation to bring about community stabilization.
The SE*CA program consisted of three components: the six-month Youth Education and Skills
(YES) community training component; an information strategy known as Transition Awareness and
Participation (TAP); and community small grants. YES training and community grants were
undertaken in two cycles around field offices in Kisangani and Bunia in Orientale Province and 
Kindu in Maniema Province. Each cycle involved 140 communities. Cycle 1 ran from July 2004 to 
December 2004 and Cycle 2 from April to October 2005. 
TAP activities consisted of information campaigns focusing on DDR, elections, transitional
institutions, and key transition laws (e.g., amnesty and nationality laws). The major vehicle for this
was OTI support to Radio Okapi (in collaboration with MONUC and Fondation Hirondelle) and five 
community radio partnerships. Radio Okapi transmitted an information program known as
“Dialogue entre Congolais.” Quick Response grants established Radio Listening Clubs and funded 
community meetings and other quick-impact informational responses to transition events, such as
youth unrest. 
II. Evaluation Objectives and Methodology
Objectives
In October 2005 OTI/Washington commissioned an external evaluation of its programs in the 
Democratic Republic of Congo and Burundi. The evaluation would consider each program from
three angles:
14
1. Did OTI contribute to the advancement of the transition process in the DRC? If so, how and in 
what ways could it have improved this?
2. Was the program strategic in responding to shifts in the transition process?
3. Did the program meet its stated goals and objectives?
Methodology
The evaluation team consisted of a senior evaluator, Philip Boyle, accompanied by a regional
specialist, Edward Rackley.2
 Both evaluators were supplied by Social Impact, Inc. of Arlington, 
Virginia. From October 31 to November 4, meetings were held in Washington with OTI and 
Chemonics International. Meetings in Kinshasa followed arrival of the team there on November 6, 
with departure to Kisangani on November 7. Work in Isiro began November 11, followed by a
concluding leg in Ituri District from November 14 to November 18. Given logistical constraints, OTI 
DRC decided to limit team field visits to Orientale Province, one of two in which OTI operated
(Maniema was the other). This allowed the possibility of comparison between first and second cycle 
communities. Concentration in Orientale Province also allowed the evaluation team to better
understand the challenges of OTI’s handover in Ituri to CONADER, the national DDR organ funded 
by the Multi-Country Demobilization and Reintegration Program (MDRP). Short field site visits 
around Kisangani, Isiro and Bunia were conducted. One longer visit by plane to Mahagi in northern 
Ituri District concluded the work in DRC. In the field, the team was assisted by Alain Mulumba,
OTI’s specialist for Monitoring and Evaluation. 
The evaluation team used semi-structured interviews (SSI) with a wide variety of program
stakeholders to explore the history, dynamics, successes, and shortcomings of the DRC program.
Beneficiary and participant focus groups were generally assembled in field sites. These groups
represented participants in both program cycles. Interview questions and techniques were similar to
those used in Burundi. They employed a probing approach that proceeds opportunistically from the 
general to the specific. 
III. Findings
OTI/DRC sought to support the reintegration process of community members following long periods 
of violence and anarchy. The program had originally been designed to reintegrate ex-combatants and 
war-affected youth into their home communities, but its focus shifted early on to the reintegration 
and reconciliation of all community members. The program had the following community-level 
objectives: 
Imparting Skills: Preparing youth3
and other community members to participate effectively in civil
society.
 
2
 Rackley served on the Chemonics go-team from March – May 2004 as a curriculum specialist. 
3
 Youth formally came to include adults up to the age of 45. However, YES participants were often even older. 
15
Supporting Reconciliation: Renewing connections between various types of war-affected 
community members, including refugees, displaced persons, victims of sexual and other violence,
ex-combatants, and youth in general. 
Community Rehabilitation and Recovery: Empowering youth and other community members by 
providing opportunities to put their new skills to work for the benefit of the community.
Encouraging Linkages: Fostering connections between communities and formal political structures
at the local and/or provincial level. 
OTI used three operational mechanisms to attain these goals: (1) the Youth Education and Skills 
program (YES) to train war-affected community members, especially youth, in reaffirmation of
traditional values, resolution of conflicts, democracy/good governance, health/nutrition, and
agriculture/income-generating activities; (2) a media program that supported access to information
concerning issues key to the transition; and (3) an in-kind, small-grants program that provided 
materials, vocational training, start-up kits, and paid labor in community rehabilitation projects for
community members in war-affected areas.
At its inception in March 2004, the SE*CA program expected to train 16,240 war-affected youth and 
community members from 280 different communities in Orientale and Maniema Provinces. As of
December 2005, OTI reported having reached 288 communities with trainings and activities that
involved 18,998 participants. Thus, OTI/DRC slightly exceeded its community goal of 280 (+3 %), 
but reached substantially more (+17 %) community members. When the budget was cut after the
first YES cycle, OTI decided to keep the same number of communities to maintain geographic
reach, but reduce the grant budget for each community. The increase in the number of direct 
beneficiaries is due to the popularity of the trainings and the capacity of Learning Facilitators to 
accommodate additional participants. Originally designed to involve about 58 persons per
community, the average number of participants in the YES trainings was 66 per community. 
The Transition Awareness and Participation (TAP) activity supported the development and
dissemination of information focused on: disarmament, demobilization, and reintegration (DDR);
elections; transitional institutions; laws related to the transition (e.g., amnesty and nationality laws);
linkages for information exchange between citizens and decision-makers (e.g., Radio Listening 
Clubs); and quick-response activities (e.g., radio emissions for voter registration) that responded to 
political urgencies. The TAP (media) program is supported by Chemonics’ small grant activities and 
the democracy and governance module of the YES training component. A total of 140 Listening 
Clubs were established, 70 in first cycle communities and 70 in the second cycle. According to the 
database, as of December 2005 the media component comprised 74 of 237 (31%) total grants
allocated, amounting to $1,510,493 (17 % of total grants). 
The small grants component was initially expected to reach every community that participated in the 
YES training. Yet when funding levels for FY 2005 turned out to be less than expected, community 
expectations had to be lowered without undermining the success and trust created by the training
program. Of the 288 communities that received the YES training, 187 (65%) received a community 
small grant for a total of $3,151,394 (37% of overall grants budget). If Radio Listening Clubs are 
counted as community grants, then almost all communities received a small grant. However, each
16
club was limited to 10 members, greatly reducing community impact, although club members were
charged with disseminating information to the wider community. 
Most community members, in fact, consider a community grant to be something significantly 
material, such as a school, health center, or a water point. In this regard, it is clear that communities 
that did not receive infrastructure grants following the YES training became frustrated, even
embittered. On the other hand, the media effort was modest enough to be sustained through budget 
cuts. Radio Listening Clubs were initiated in most YES communities, since they are an easy way to 
maintain a useful presence in those communities and to develop a channel for timely, balanced, and 
accurate information about the peace process and political transition.
The cumulative two-year cost of the SE*CA program is $10.1 million, with $5.1 obligated in FY
2004 and $5.0 in FY 2005.4
 The original ceiling identified was 15.2 million, but at the end of the 
first cycle of YES training the budget was cut. Community grants were frozen until further notice.
The average monthly cost of the full program was $405,000: $55,000 for OTI and $350,000 for 
Chemonics. A total of 237 OTI grants were allocated through December 2005 equaling $8,628,314 
million. During this time there was also support to the Independent Elections Commission (CEI). 
Follow-on activities for CONADER, South Kivu, and the transition quick impact grants brought in
another $2.7 million at the end of FY 2005 and $4.6 million from the World Bank MDRP in FY
2006. 
The grant cost of YES trainings is estimated at $444,498, which includes $350,000 budgeted for 
YES, plus bicycles ($58,598) and motivation payments for the Learning Facilitators ($35,900). 
When Master Trainer salaries and overheads are included, the total cost of the training over the
length of program comes to $1.6 million. When divided by the 18,998 participants, this yields a cost 
of about $84 per trainee. 
1. Did OTI contribute to the advancement of the transition process in the DRC? If so, how 
and in what ways could it have improved this?
OTI contributed to the transition at the local and national levels. Its broad aim was to deliver a 
tangible peace dividend to the Congolese, many of whom were completely isolated from the ‘outside 
world,’ be it nearby towns, provincial capitals, or Kinshasa. Equally important, OTI aimed to ensure 
popular buy-in to the transition process and to prepare the ground for credible national elections.
While clear progress was made in these areas, beneficiaries readily grasped that the program and its 
fruits did not originate within the transitional government. Indeed, the resulting increase in transition
awareness carried a corresponding increase in dissatisfaction with transition actors and continued
poor governance. Nevertheless, the OTI SE*CA program made the following contributions to the
transition process. 
1. Working at the national and local levels, OTI opened up long-isolated communities, reconnecting
and implicating them in national political processes, such as voter registration and the constitutional 
referendum. Many communities reported that reconnecting with the outside world, through the
 
4
 The total amount of OTI funds committed since 1997 in the DRC is $35.4 million.
17
various modes offered by OTI, was the first real peace dividend they had experienced since the
transition began in 2003.
While the Community-focused Reintegration component targeted the eastern provinces, the media
component, known as Transition Awareness and Participation, offered national coverage. Although 
difficult to quantify, in its work with Radio Okapi (Fondation Hirondelle) and the Independent
Electoral Commission (ICE), the media program had the greatest national-level impact. More
consequential at the local community level were OTI’s Youth Education and Skills (YES) trainings, 
in five modules spread over about six months. These proved so popular in parts of Orientale
Province that Learning Facilitators were invited by outlying communities to come and conduct
trainings as their guests. Largely because of this, the YES trainings appear to have had the highest
number of secondary beneficiaries of SE*CA components. Moreover, according to a survey of
training participants conducted by OTI, YES graduates claimed to have trained an average of
another 5-10 community members each in the lessons of the five modules. While this was certainly 
not done as formally as the training extended by Learning Facilitators, it shows how communities
were galvanized by YES. 
2. OTI stabilized and invigorated war-torn communities through group trainings, labor-intensive, 
community-led rehabilitation projects, and radio listening clubs. In contrast to traditional DDR
programming, the CFR approach allowed a focus on entire communities, not just the ‘peace 
spoilers.’ The multi-component CFR approach proved its relevance in the Congolese context, where 
small grants alone are not enough to stabilize communities and increase transition awareness.
Although CFR demands a greater time investment, the combination of YES training and small grants 
resulted in a program strategy that could address the root causes of local conflict and obstacles to 
reconciliation, paving the way for gradual community recovery. 
3. The program contributed to the reintegration of ex-combatants in highly militarized areas, such as 
Ituri and Maniema. This it did by first gaining the trust of communities and by spending significant 
time on site – an instance of OTI process informing product. In areas where ex-combatants or
military deserters were present but not officially recognized (CONADER not yet active), OTI
prepared communities to receive and accept ex-combatants once recognized by CONADER. In areas 
where there were no ex-combatants, the YES training facilitated a different, but equally vital mode 
of reintegration, that between communities and local authorities, whose relations had been marked 
by distrust, mutual exploitation, and fear.
4. OTI facilitated the reconciliation of stigmatized and disenfranchised groups with other community
members. Such reconciliation included: (1) female rape survivors and their husbands, families and
the wider community; (2) the role of women in general in communities; (3) former combatants,
military deserters and child soldiers with their families and the wider community; (4) displaced
persons and refugees with those that had remained; and (5) pygmies with the nearby Bantu
communities. YES training modules laid a solid foundation for conflict resolution without recourse
to violent revenge, a common mode of settlement in rural and peri-urban communities. Such conflict 
resulted from theft, outstanding debt, land disputes, and crop damage due to uncontrolled herd
grazing, all of which are widespread in rural Congo. Other forms of community conflict involved
families of rape victims and perpetrators and between combatants’ families and their victims. 
18
5. Regarding preparations for national elections, including the recent constitutional referendum, OTI 
was instrumental in mobilizing voter registration. Where YES trainings occurred, turnout for voter
registration was 80%; where there was no training or sensitization, people resisted and police were
brought in to force them to register.
6. The program pursued non-discrimination in the selection of beneficiary groups. This was in stark 
contrast to other transition actors, particularly the Church and local authorities, both having played a
highly sectarian and divisive role during the war. By bringing in fellow Congolese to conduct 
trainings and to help develop community-led projects, OTI sent a very positive message in a context 
where there are no positive role models, and the logic of war and an economy of survival have 
undermined collective interest, associative spirit, voluntary effort, and civic identification. 
7. OTI was able to be flexible and increasingly appropriate in local media strategies. Since OTI’s 
internet centers revealed themselves to be technically daunting, resource draining, and less
accessible to the masses, particularly rural populations (although well used in urban areas), OTI
developed community-based Radio Listening Clubs. Internet centers restrict access to only the
wealthy and the literate; they are elitist by design. Besides their simple technology, Radio Listening 
Clubs are effective information vehicles, more democratic (because non-paying and accessible to
illiterates), and provide occasion for further discussion and community involvement.
8. Two seemingly viable structures to ensure durability of impact were launched by the OTI
program. The SE*CA Network and the CONADER follow-on grant to USAID (with continued
implementation by Chemonics) ensure that two OTI transition support strategies – information flow
via Community Integration Management Committees (CGICs) and YES training to ex-combatants –
will continue after OTI closes in July 2006. Neither handoff strategy existed at project inception. 
Both handover mechanisms not only were developed and realized during the course of operations, 
but also by seizing opportunities created, in part, by OTI’s own operational momentum.
In spite of these successes, there were a number of shortcomings to the OTI program in DRC. While 
it should be borne in mind that it was impossible for OTI to be present in all essential areas of Congo
and that it was not feasible for OTI to maintain its presence for the entire length of Congo’s political 
transition, it is nonetheless apparent that OTI underperformed in the following ways: 
(1) By not integrating transition actors and institutions, particularly Congolese political figures, into 
its scope of activities, OTI missed a key opportunity to use Kinshasa as a theatre of engagement, one 
with direct access to the political actors and institutions most influential to the outcome of the 
transition process. Those ultimately responsible for the success of Congo’s transition, its political 
elites, did not figure among the program’s range of beneficiaries, due in large part to the grassroots, 
community bias of the CFR model. OTI maintains that the evaluation should recognize that OTI 
worked with national Congolese political figures where OTI found credible actors. OTI often 
worked with Congolese political figures on the district level, for example in Ituri. Nevertheless, this 
was not a core OTI strategy under SE*CA. 
19
(2) In terms of geographic focus, OTI’s absence from North and South Kivu provinces, the key 
undetermined variable in the transition, constitutes another significant missed opportunity.
5
 It is true
that OTI has been able to train IRC in the use of the YES modules in South Kivu in a follow-on to 
the two-year SE*CA program. OTI/DRC notes that it considers itself to be operational in South Kivu 
after March 2005. 
(3) With regard to program timeline, OTI’s premature departure relative to national elections is 
unfortunate; the transition is far from over. While OTI/DRC does not feel its departure is premature, 
the USAID Mission Director does not share this view. However, although OTI is leaving, its Quick 
Response capacity will continue under the USAID/DG Office in South Kivu and Orientale
provinces. The programs there are also currently working on elections sensitization campaigns. 
(4) Greater synergy with a wider array of actors, including the nascent private sector, could have
unleashed an economic dividend for participating communities. 
These and other less consequential shortcomings are elaborated below:
Limited impact of the CFR approach on national-level transition processes: In the DRC, the chosen 
CFR approach meant going deeply into specific communities with five training modules. The five
modules deal with (1) Reaffirmation of values; (2) Resolution of conflicts; (3) Democracy and
governance; (4) Health and nutrition; and (5) Agriculture and income-generation activities. The last
module was combined from two previous modules. This training entailed a commitment of about six 
months, which in retrospect could have been shortened to improve speed, flexibility, and geographic 
spread.6 OTI did tailor this training in a few areas to reflect response to emergencies, new areas like 
Opienge, use of conflict resolution and democracy/governance modules with youth gangs in
Kisangani, use of the DG module in the referendum sensitization campaign, and combination of the 
values and conflict resolution modules in the Ituri follow-on. However, in the vast majority of
communities the full six-month course was the norm.
In view of the six-month YES training cycle and the vastness of the country, opportunities for
national level impact were limited primarily to media activities of limited quantifiable results. These 
included the national communications campaign of the Independent Elections Commission and 
Radio Okapi emissions on shortwave, the latter constituting the first vehicle to cover the entire
country. National level impact was also achieved by making changes to the national reintegration
strategy for reintegration of ex-combatants.
It is unclear from available strategy documentation whether alternate transition support strategies
were considered. While the Request for Proposals (RFP) for the Phase 2 program did not prescribe 
 
5
 This is illustrated by the fact that with the exception of the announcement of delayed national elections in May 2005, no 
other single event has created violent shockwaves across the country as did the May 2004 invasion of Bukavu by a 
mutinous FARDC commander. Repercussions from Ituri violence, far more vicious by comparison, rarely reach beyond 
Ituri itself. Largely due to fallout from the Rwandan genocide and anti-Tutsi sentiment across the region, the geopolitics 
of the Kivus appear to outweigh those of Ituri in determining the fate of the transition.
6
 The YES training component could be made lighter for quick impact. Its content is not fixed in stone and can be treated 
as a malleable set of modules to be taught following a needs assessment. For instance, not all five modules are equally 
relevant in all contexts. OTI/DRC did vary its length and composition in a few places during SE*CA. 
20
the use of a CFR approach, it did request specific expatriate expertise conformant with the CFR
model (i.e., a curriculum design specialist). Broader, alternative approaches could have made room
for support to transition institutions at the capital level. CFR could allow a mixture of ‘mortgaged’
training plus opportunistic interventions at weak points in the transition process at the national level.
Coordination, not competition: Coordination with other aid agencies, particularly those working in
DDR and infrastructure rehabilitation, is seen by OTI as positive leverage. OTI notes that “on almost 
every project, OTI coordinated with one or more partners and several projects were joint-funded and 
implemented together.” This collaboration included: rehabilitation of schools with MONUC,
UNICEF, and WFP; DFID and OTI collaboration with Radio Okapi; MONUC transport
contributions to projects; coordination with IRC, IFES, and NDI on referendum and registration
sensitization; coordination with IFES, DFID, EU, and UNDP on the IEC elections campaign; and
collaboration with the World Bank, CONADER, the MDRP donors, and (unsuccessfully) with
UNDP on DDR. In Isiro, there is a good example of OTI collaboration with the Diocesan
Development Bureau of the Catholic Church. Another partnership with Project Hope was
successfully implemented in microfinance in Kisangani. 
Obviously, other partnerships could have been pursued. In Ituri, other transition actors were
executing community-led rehabilitation projects: the United Nations Mission to the DRC
(MONUC)’s Quick Impact Projects (QIPs); Agro Action Allemande (AAA); and Italy’s
Cooperazion e Sviluppo (CESVI)). These could have come to OTI YES training sites to implement 
their projects, especially in the second-cycle communities that received no small grants from
SE*CA. On the plus side, OTI apparently did bring pressure to bear on CONADER from December 
2004 to work together on DDR. OTI notes, moreover, that the SE*CA program was based on the 
recommendations in the National DDR Plan, which CONADER executes. There had been many
coordination meetings between OTI and CONADER in 2004. This collaboration ultimately yielded
the October 2005 “handover agreement” between CONADER and the USAID. 
Potential for backlash: Given the huge discrepancy between what OTI and other international 
agencies can deliver and capacity of the state to participate and contribute, it is understandable that 
communities see no ‘transition dividend’ from their government. Indeed, there remains a strong
sense among Congolese that the transition is controlled by outside forces, its outcome foretold and 
beyond their control. This non-credulity undermines popular engagement and ownership of the 
transition process.7 Although there are clear gains in OTI beneficiaries’ sense of implication in the 
transition, the conviction that the West profits from DRC’s chaos (via mineral extraction) is 
widespread. There is relatively little sense that Congolese elites are the primary enablers of this
resource theft. 
A difficult dilemma preoccupies many Congolese today, that between a recognized need for outside 
intervention for a successful transition, and the common belief that outsiders control DRC and profit 
from its dysfunction. The result is a legitimacy deficit regarding foreign donors or international 
agencies, as well as DRC’s political elites. The evaluation team heard a common refrain: “Why vote 
for any of our officials; what have they done to earn our vote?” It is true, however, that where OTI 
had intervened, the evaluation team observed that perceptions of the U.S. government had improved.
 
7
 In fairness, this suspicion is somewhat justified: the CIAT, national elections, Inter-Congolese Dialogue and Sun City, 
MONUC peacekeepers, and DDR agencies – all of this is funded and driven by foreign actors.
21
OTI’s operating assumption is that the transition will succeed and the state will pick up where OTI 
has left off. Unfortunately, this is not likely to happen, and the Congolese know it. However, OTI 
notes that it did involve state actors in the design of the SE*CA program; thus the Ministry of
Education was involved with school projects and the Ministry of Agriculture in the agriculture
component of the YES training. Nevertheless, generally speaking the visibility and frequency of
such contacts were minimal, risking the creation of a legitimacy ‘time bomb’ for the state.
More coordinated synergy: Greater integration of the CFR, media, and grants components of the 
program would have been desirable. Nevertheless, the outline of a common strategy is evident that 
combines the YES training, the media component, and small grants to support a broad, participatory 
effort to inform, educate, and dispel disinformation concerning the transition. Examples of cross￾fertilization between media (transition awareness), small grants, and CFR capacity-building include
projects in Isiro and Mahagi to transmit YES modules concerning democracy and governance over 
community FM radio in support of the voter registration drive. Other activities could have aimed to 
promote income generation. For example, Radio Listening Clubs with two-way radios (now with 
only receivers) might sell messages and transfer money, a service common elsewhere in the country. 
OTI, however, did provide two-way radios for income generation in the Community Resource
Centers, which are connected to the Radio Listening Clubs and form a higher level of the SE*CA
network.
Incomplete activity selection strategy for communities: OTI selected communities based on the
prevalence of desired target groups and strategic factors: ex-combatants, victims of sexual violence, 
returning IDPs and refugees, idle youth, degree of war impact, strategic importance to the transition, 
and geographic dispersion along road axes (for maximizing spread of effects). Grants were selected 
by strategic geographic impact, possibility of sharing between several communities, and community
requests. However, basic needs are glaringly unfilled in the DRC, and such extreme destitution
works to reorient OTI’s loftier transition ambitions into humanitarian work. At the same time,
meeting these basic needs often becomes necessary for community trust and buy-in. In this vein,
YES participants were sometimes mocked for receiving training with no material benefit, and some
communities that received training but no grant became irate. While seeking to advance the political 
transition in DRC, OTI chose to reinforce the self-reliance of communities through capacity building 
and participatory work projects. However, it is a mixed message to communities to argue a
connection between community participation (self-reliance) and the legitimacy of a transitional
government perceived to be absent and dysfunctional. 
Transition awareness deficit not only in the East. The less militarized provinces (Kasaïs, Bandundu)
are an increasing source of volatility and resistance to the transition processes, whose leaders and 
institutions they do not trust or accept. It is mistaken to assume that because an area is pacified that it 
is on board with the transition process. These regions could have benefited from more intentional
media exposure. OTI points out, however, that it first set up a community radio station in Tshikapa 
in the Kasais and another in Bandundu (also in Muanda in Bas-Congo) to become local affiliates of 
Radio Okapi, because OTI saw the need for information. Then, OTI put Radio Okapi on shortwave, 
hitting the whole country. Media campaigns, including those of the Independent Electoral
Commission, also went to the Kasais and Bandundu. 
22
Community agriculture projects (seed multiplication) could have been much improved as
reconciliation tools, including the distribution of improved seeds and tools to all participants. 
Distributing tools to only a handful of participants, while intended to reduce costs and increase
internal community cooperation, only sowed conflict and dissent. With a 60% dropout rate of
participants, the community fields became the property of the local SE*CA management committees 
(CGICs), who then sold the harvests and kept the profit for themselves, instead of redistributing the 
seed for next season’s planting. Project design was not tailored to community needs, but was a top￾down effort conceived and driven by OTI.
2. Was the program strategic in responding to shifts in the transition process?
Questions related to this issue include: Was this program appropriate for OTI in this context? Was 
the DRC ready for OTI’s approach? Did OTI fill a crucial gap to keep the country in the transition 
process? Was OTI strategic in developing a handover plan?
In the DRC, where the duration of the transition process has surpassed the normal span of OTI
engagement and the vastness of the country prohibits OTI activity in many areas of grave need, OTI 
strategy was forced to compromise from its inception, thus limiting and reducing the scope of its
national impact. Further, security concerns in the Kivus initially prevented OTI activities there,
despite their evident strategic role in securing the fate of the transition.
During the design phase, questions were raised about the strategic relevance of Kisangani as rear
base of operations. As a point of operational departure for the CFR model, Kisangani and environs 
offered the same degree of war distress as Ituri or the Kivus, without being in the heat of combat. 
Communities were isolated and cut-off from the surrounding regions and major cities, including the
capital. Despite a relative calm, central authorities had yet to reclaim control; local officials and the 
military were without a master and the subjugation and predation of civilians was their sole source of 
income and sustenance. 
In time, OTI discovered that there were different needs for different regions: Kisangani and environs 
required economic development for stability; Ituri district needed conflict resolution. Overall, a clear 
impression emerged that a promised ‘peace dividend’ was nil without a tangible economic benefit.
In addition, stability was necessary for the program to operate. Ongoing instability in Bunia (Ituri) 
resulted in a high default rate among YES participants and grant beneficiaries. In many volatile areas
of Ituri, participating in the program without compensation or per diem was a luxury residents 
claimed they could not afford. 
To some degree, OTI has been successfully strategic in handing over program elements to USAID or 
other donors. In October 2005, OTI signed an agreement with the World Bank-funded Commission 
Nationale de Désarmement, Démobilisation et Réintegration (CONADER), which will provide $4.7
million to USAID/DRC to fund the Chemonics program to reintegrate 11,200 ex-combatants in Ituri
District. The Ituri Reintegration Program has a total value of $6.4 million, with USAID contributing
$1.7 million. Also benefiting from the program will be 5,040 community members who, like the ex￾combatants, will be provided YES training and then a choice between paid manual labor on
rehabilitation projects or receiving basic vocational training with a start-up kit. This is the first
23
grantee for CONADER and the first time USAID has received World Bank funds to execute a
project. After mapping and identifying ex-combatants, SE*CA deployed 28 Master Trainers to 11
sites to train 426 Learning Facilitators (LFs) from 213 communities. The LFs will, in turn, train
program participants.
In South Kivu Province, the USAID Office of Conflict Management and Mitigation (CMM) has
committed $600,000 towards the reintegration of ex-combatants, implemented by the International 
Rescue Committee as sub-contractor to Chemonics, through June 2006. SE*CA Master Trainers are 
conducting the identification and training of Learning Facilitators, who will in turn train participants
in communities. The program is currently managed by OTI and will be taken up by the USAID
Mission. 
The recently formalized SE*CA Network is comprised of Learning Facilitators, community
management committees (CGICs), Radio Listening Clubs, and regional community centers.
USAID’s Democracy and Governance Office has committed funding to launch the network, which
exists to distribute information on the transition process to areas that have already received YES
trainings. The network will link this community-based information dissemination with Democracy
Resource Centers and anti-corruption committees supported by the DG Office. In all, some 190
communities will be linked in eight provinces to the four Democracy Centers. Costs for this
operation are still being worked out.
Generally, OTI was strategic in its program, but with the following caveats: 
1. The program timeline and geographical coverage did not adapt to two fundamental shifts in the 
transition process: postponement of national elections and the emergence of the Kivus as a wildcard 
in the fate of the transition. Shorter training cycles with grants in each location could have created 
the resources (despite budget cuts) to allow a shift into the Kivus once security permitted (April
2005). The IRC did begin to intervene in South Kivu (Bukavu) using the YES training modules at
this time. 
2. Regarding the voter registration crisis and Independent Electoral Commission (CEI) incapacity,
OTI showed great flexibility and acumen in its support. As 2005 began, the government still had not 
been able to pacify or extend its authority across its territory or to adopt any essential legislation, pay 
regular salaries, or start registering voters. This catastrophic record led to the announcement in
January 2005 by the Independent Electoral Commission that elections would likely not be held
before June 30, 2005. 
One of the merits of the anxiety and apprehensions in advance of June 30, 2005 was a sudden
increase in the legislative and administrative productivity of the transition institutions (prompted by
renewed foreign pressure). After two years of lethargy, a draft constitution was adopted in May, a 
law organizing the constitutional referendum in June, and the registration of voters began in
Kinshasa on June 20. By early November 2005, it was estimated that some 20 million Congolese had 
registered to vote across the country. While OTI’s role in this accomplishment is clear, it is not clear 
who will fill this role in the immediate build-up to national elections – now scheduled for April
2006. Elections oversight is being left in the hands of the USAID Mission, whose director would
have appreciated OTI’s presence well into the post-election transition. He points out strongly that the 
24
political transition is far from over. Nevertheless, the follow-on CFR programs in Ituri and South
Kivu will not end until June 2006. USAID/DG Quick Response grants can also respond to future 
transition needs. 
3. In terms of handover strategy, the DRC program finds itself in surprisingly good stead. The two 
main thrusts of the program, YES training and community outreach through various media, will
continue under different guises, as noted previously. The SE*CA Network and the CONADER grant 
will endeavor to ensure that these two key transition support strategies continue to leave their mark 
on the transition. Although these SE*CA elements are only short-term extensions in most cases, they 
reflect substantial collaborative planning between OTI and the USAID Mission. 
4. The evaluation team found that a number of communities received YES training but no grants.
This was particularly true in the second cycle of community work, while in the first cycle there was 
sometimes considerable delay in the grants component. This left communities with a deep sense of
deception, bitterness and frustration, particularly in areas devastated by violence, such as Ituri.
Community grants are excellent opportunities to put into practice the values learned in the YES
modules, and are the key ‘process’ component of reintegration and reconciliation within a
community. Again, it was not the ‘mortgaging’ of the program that caused these shortages, but
factors beyond the control of OTI/DRC – budget cuts. Still, shortening the CFR training cycle from 
six to three months might have freed up sufficient funds to provide all communities with at least one 
infrastructure grant. Rather than cutting back the program and closing in September 2005 as OTI 
apparently wished to do, the USAID Mission requested that OTI stay until December 2005 and then 
again until March 2006. 
3. Did OTI meet its goals and objectives?
Disengagement and the self-interest of DRC’s political elites combine with their mutual mistrust
produce a dysfunctional transition; it is also a root cause of DRC’s material and social destitution. 
Nor is there a sufficiently robust civil society to act as safeguard against excessive predation by the 
elite or even to mediate the relationship of masses and government officials. DRC’s civil society of 
NGOs and media outlets serves as means of survival rather than as means of holding the state 
accountable. The question remains of how OTI has engaged this dynamic in support of a more
promising transition? 
In setting up and running its operations, OTI appears to have done little to implicate government 
officials at the national or provincial levels, although local chiefs were usually included in the
CGICs. Collaboration with CONADER and the Independent Election Commission has also occurred 
in the operational areas around SE*CA field offices. Beyond this, there is little implication of
government officials. In so doing, OTI stepped directly into the role of service provider, replicating
the acquiescence of other international agencies by responding to essential needs, thereby allowing
the transitional government to continue its infighting and introspection and to disengage from social
service provision, including salaries for health and education workers. Ongoing neglect of the
country’s basic needs fuels the popular sense of abandonment, thus furthering the government’s
legitimacy crisis in the eyes of citizens. Inclusion of local, provincial and national transition actors
and institutions in community work should have been recognized by OTI as a core strategy to
advance and reinforce the transition. 
25
Objective 1: Support the reintegration process between war-affected youth and their host 
communities.
· In highly militarized areas where ex-combatants are common, the reintegration impact is
palpable, despite high frustrations with CONADER. In less militarized areas (e.g., Kisangani
and environs), relations with the local administrative class and national army, the primary
exploiters of the population, have improved. YES training participants say they no longer
fear harassment and abuse from soldiers and officials, and when faced with such harassment, 
they are less content to submit, preferring to engage the antagonists in discussion.
· The program had a positive, if unanticipated, gender impact. The status of women in
communities improved in different ways: (1) overall there is now less discrimination against
women as silent subordinates or ‘secondary citizens’, incapable of negotiation, or using
political reasoning, to resolve local conflicts; (2) ex-combatants and former gang members
(male and female) involved in work projects (road construction) claimed that constructive
activity had reduced household tensions and improved domestic life; (3) domestic 
cooperation is now valued more by men, since, previously, helping one’s wife was
disparaged as male weakness.
· In terms of community participation, OTI’s operational model was not unique. It is a 
common, but not the dominant, mode of community engagement for international aid
agencies. The YES training component with its overt political content was novel and drew
suspicion at the start, requiring high-caliber leadership and sensitization to convince 
communities of its relevance. While the evaluation team did not visit Maniema Province, it 
has been told the program there enjoyed far less community buy-in than in Orientale
Province. The result was that in Maniema very little in in-kind contributions was received
from communities for work projects. Although described as targeting ‘war-affected youth,’
in practice the YES training included people of all ages -- with well over 18,000 participants
claimed. OTI/DRC notes that about 80% of participants were “youth,” although this was
defined as up to 45 years in age. Communities requested that elders be included, so that they 
might have sufficient prestige to serve as teachers to neighboring communities. Both in its
process and in the content of the modules, it clearly contributed to reintegration and
community reconstitution both among community members and vis à vis local authorities.
Work projects provided community members an important peace dividend, as training alone 
‘left them hanging’ and bitter. 
· The CFR program sought to support the political transition in DRC by strengthening
participatory processes at the bottom that would improve community self-reliance/self￾determination. However, there is really little for communities yet to work with. Dire poverty 
and a total absence of capital have spelled utter dependency for many rural communities. 
This is particularly the case in Ituri district, where conflict and displacement still prevent 
continuous farming. When enjoined to pull themselves up by their bootstraps, community
members in Ituri responded: “We have no boots!” Moreover, the government’s incapacity to 
provide the bare minimum of services reinforces the sense of dependency on outside actors
and foments discontent. Such frustrations are a potential source of post-election backlash, 
26
since the bouts of mass looting and vandalism over the last 15 years have been triggered by 
similar sentiment and conditions. 
· Community grants were adequately focused on basic needs (water, schools, and other useful
infrastructure) and brought together disparate groups through the planning and work effort.
Often participants came from different levels of society and would otherwise have avoided
each other. Generally speaking, the YES training served an effective double role in
expanding their perceptions and transition awareness and sensitizing communities to buy-in, 
ownership, and widespread participation in grant projects.
· Working together for community projects was an effective reintegration and reconciliation
tool. Entrepreneurial spirit, a necessary condition for success given the deprived context, is 
not widely or equally shared by community members. Some communities expect SE*CA to 
continue to run things as in the past; others understand the finite relationship and have
developed sustainability strategies on their own. 
· Monitoring and Evaluation took the form of measuring key indicators in SE*CA
communities at start, mid-term and end of intervention. This was done by means of a Quick 
Survey administered by the Master Trainers. The indicators were part of a Performance
Monitoring Plan developed as part of the OTI/DRC Strategic Plan as early as April 2004. 
The PMP appears to have been modified in October 2005. OTI also documented the number 
of people trained and did follow-up surveys with beneficiaries. In this way, OTI sought to 
establish a causal link between the YES training and impact in the community. Community 
grants are also evaluated by Project Development Officers and entered into the OTI grants
database. Other surveys are more subjective, such as the one undertaken by the entire SE*CA 
staff in Bunia in September 2005.8 Additionally, quarterly meetings were held with all OTI
and Chemonics staff and at the end of the first cycle and towards the end of the second cycle 
they also held intensive self-evaluations. Regular monitoring of this type leaves a strong
paper trail to demonstrate outputs, trends, and long-term dynamics, and it serves as one
archive of institutional memory. Evaluation of final performance and impact, however, are
best measured at least a year after program completion. OTI does not, however, engage in
such ‘post-mortem’ impact evaluations.
Objective 2: Reinforce local, provincial, and national awareness in order to foster community 
participation on issues key to the transition process.
· In communities where OTI intervened, identification and buy-in to the transition process is 
palpable. The CFR process of bringing people together for trainings and grants appears to 
have been more catalytic in this regard than what the media program was able to generate 
through its channels. While national media efforts (Radio Okapi, community radio support) 
provide background information on the transition, YES trainings managed to give ordinary
Congolese the sense that the transition, for all its faults, is also their process.
 
8
 “Internal Evaluation of Grants and YES Training Conducted by the Entire SE*CA Staff: Bunia – Week of September 
22-24.” September 2005. OTI. Kinshasa, DRC.
27
· The YES trainings ‘opened minds’, or so many participants claimed. Political support and
votes can no longer be bought; similarly, many youth claim to be less susceptible to political 
manipulation and violence. Yet all ask: “What have these candidates done to earn our votes?”
After nearly two years of OTI investment, the behavior and instincts of the political class are 
unchanged; only voters have evolved. Common sentiment is that the poor are constantly
subjected to bad governance, but are not given the means to change it. While this indeed 
constitutes increased awareness of the transition, it does not come with the desired
component of popular empowerment. Over time this may well degenerate once again into 
violence and anarchy. 
· Inept governance by transition actors and ongoing violence in the Congolese east are the
‘weakest links’ of the transition. In this light, OTI’s definition of ‘transition’ would appear
ill-adapted to the Congolese experience. OTI shifted its end date twice to accommodate
CONADER’s timeline for work in Ituri and the postponed elections – from September 2005 
to December 2005 and then again to March 2006. OTI/DRC notes that the USAID Mission 
will continue management of the program, although the Mission Director clearly felt the
transition was far from over and decried OTI’s “premature” departure. 
· Bypassing opportunities to engage the various transition support institutions and 
‘commissions’ created by government, OTI forfeited a valuable means of reinforcing the
transition by working with the political class at national level. This appears to be due to a 
‘grassroots bias’ in program design, but the weakest link in the transition process is the
political elites, not communities or ex-combatants. Such opportunities were not lost because 
of CFR ‘mortgaging.’ They stem from an unexamined assumption in the CFR program 
approach that communities are more central to a successful transition than the politicians 
themselves. OTI/DRC notes, however, that this assumption was examined and that OTI
chose to go to the “grassroots” level because USAID/DG had already elected to work with
national institutions and actors. Nevertheless, OTI/DRC vaunts its choice of CFR as one of 
several spin-offs from the Sierra Leone experience. 
· This raises the question of possible national-level impact accrued through community level
programming. Do local successes ‘roll up’ to national impact? Such causality would be hard,
but not impossible, to measure. Given that violence and insecurity across the East were
constant during the course of the program and that these battles were the source of much
infighting, distrust and instability within the ‘1+4’ government, one can deduce a direct
causal link between the volatile eastern provinces and the health of the transition in Kinshasa. 
Serious destabilization from two events shook the entire country, and very nearly derailed the 
transition with a return to full-scale civil war. These events were the June 2004 invasion of 
Bukavu by a mutinous commander in the Congolese national army (Forces Armées de la
République Démocratique du Congo/FARDC) and the massacre of 162 Congolese refugees 
in a camp in Burundi by rebel forces based in DRC. That it took events of this magnitude to 
rattle transition actors and galvanize the transition process is significant when considering 
whether OTI activities ‘roll up’ from communities to national-level impact. The country is
simply too vast, the variables determining the success of the transition process too numerous, 
and OTI’s input too remote from the corridors of power, for any direct causality to be
claimed between OTI’s efforts and the consistently volatile course of the transition in DRC. 
28
· OTI had high expectations regarding the portent of the political transition for the mass of the 
population. However, the new government will not be able to meet these expectations. There 
is no substrate for development: no infrastructure, no managerial or administrative systems of 
any kind; nor is there any human capacity to implement such a management system, public
or private, at local or national level. If the administrative and political classes lack these basic 
capacities and instincts, it is highly unlikely that communities themselves can manage basic
investments (schools, information centers, etc.). The DRC government expects the
international community (NGOs, aid agencies) to develop the country. Community buy-in is 
a foreign ideal that is certainly not shared by government. The DRC government does not 
care who provides the pump, the bridge, the school or how it gets there – just that it gets 
done. It is the governmental leaders, more than the people, who need new values and
perspectives. While working with DRC’s inexperienced and self-serving political class may
not be as easy to do as community mobilization and education at the grassroots, it should be 
part of a transition strategy in countries like the DRC. 
IV. Conclusions
1. OTI’s use of the Community-focused Reintegration (CFR) approach allowed a focus on entire
communities, not the ‘peace spoilers’ alone, as do traditional DDR programs. In highly militarized
areas with ex-combatants, OTI contributed to reintegration by first gaining the trust of communities
through its YES trainings. In areas where there were no ex-combatants, the capacity-building and
community grants components of SE*CA achieved their objectives of reintegration and
reconciliation of community members through broad-based, participatory activities. 
2. Working at the national and local levels, OTI opened up long-isolated communities, reconnecting
and implicating them in national political processes, such as voter registration and the constitutional
referendum. Many communities reported that reconnecting with the outside world through the
various modes offered by OTI was the first real peace dividend they had experienced since the
transition began in 2003.
3. As a result of OTI’s intervention, beneficiary communities now grasp that a successful political
transition requires civic engagement and public accountability, actions for which they alone are
responsible. It did not escape them, however, that OTI – not the national government – created
SE*CA and delivered its messages, projects, and hope to the community.
4. OTI stabilized and invigorated war-torn communities through YES group trainings, labor￾intensive, community-led rehabilitation projects, and Radio Listening Clubs. The multi-component 
CFR approach proved its relevance in the Congolese context, where small grants alone are not
enough to stabilize communities and increase transition awareness. 
5. The program contributed to the reintegration of ex-combatants in highly militarized areas, such as 
Ituri and Maniema. This it did by first gaining the trust of communities and by spending significant 
time with them. In areas where ex-combatants or military deserters were present but not officially 
recognized, OTI prepared communities to receive and accept ex-combatants once demobilized. In
areas where there were no ex-combatants, the YES training facilitated a process of reintegration 
29
between communities and local authorities, whose relations had been marked by distrust, mutual 
exploitation, and fear.
6. OTI also facilitated the reconciliation of stigmatized and disenfranchised groups with other 
community members. Such reconciliation included: (1) rape survivors and their husbands, families
and the wider community; and (2) former combatants, military deserters, and child soldiers with
their families and the wider community. YES training modules laid a solid foundation for conflict
resolution without recourse to violent revenge, a common mode of settlement in rural and urban
communities alike. 
7. The six-month YES capacity-building course clearly galvanized community members, who
remained poised for further interaction with SE*CA representatives. Once the training was over,
however, SE*CA moved on to new areas leaving a serious void. This was true even in communities 
that had received some small grants. In Cycle 2 communities, however, budget cuts severely
reduced these projects, leading to severe disappointment and anger. 
8. Regarding preparations for national elections, including the recent constitutional referendum, OTI 
was instrumental in mobilizing voter registration. Where YES trainings occurred, turnout for voter
registration was 80%; where there was no training or sensitization, people resisted and police were
brought in to force them to register.
9. OTI was able to be flexible and increasingly appropriate in local media strategies. Since OTI’s 
internet centers revealed themselves to be ineffective tools to support the transition, it developed
community-based Radio Listening Clubs. This is to say that internet centers were not the most 
effective way to reach the masses, since they were urban-based and required user literacy. OTI/DRC 
notes that the SE*CA team’s September 2005 analysis found that the internet centers were effective 
in supporting the transition. They were difficult to execute, but if they functioned and if they had 
community bulletin boards where they posted the latest news, the centers proved to be very popular 
and useful to disseminate information. However, besides their simple technology, Radio Listening 
Clubs are effective information vehicles, more democratic than internet centers, and provide for
wider rural community involvement.
10. The SE*CA Network and the CONADER follow-on grant to USAID ensure that two OTI
transition support strategies – information flow via community management committees and YES
training to ex-combatants – will continue after OTI closes in March 2006. Neither handover strategy
existed at project inception; both were developed in a timely manner before program end. 
11. By not integrating transition actors and institutions, particularly Congolese political figures,
fully into its scope of activities, OTI missed a key opportunity to use Kinshasa as a theatre of
engagement, one with direct access to the political actors and institutions most influential to the
outcome of the transition process. It is true that OTI chose not to work with national actors and
institutions because the USAID Mission Office of Democracy and Governance worked with most of
the national institutions in Kinshasa, leading OTI to focus on the “grassroots” issues. Whether this
division of labor (synergy) has been effective remains to be seen as the transition proceeds. 
30
It should also be noted that OTI/DRC feels that its work with CONADER and the Independent
Election Commission at the national level is highly significant. Both institutions were involved in
approving and using YES modules. The SE*CA program also influenced CONADER to change to a 
community approach to reintegration of ex-combatants rather than focusing on ex-combatants alone. 
Moreover, OTI designed the national communication campaign for the IEC, and the program’s
Learning Facilitators provided a large percentage of the registration, sensitization, and polling
workers for the IEC during elections registration and the constitutional referendum. OTI also
coordinated closely with the Haute Autorité des Media on media programs.
12. In terms of geographic focus, OTI’s absence from North and South Kivu provinces, the key 
indeterminate variable in the transition, constitutes another significant missed opportunity. Although
OTI did undertake some actions in South Kivu after security improved in March 2005 (e.g., 
journalist training), it did not place the whole SE*CA program there. Under the IRC follow-on in
South Kivu, OTI has helped launch a program in 47 communities reaching some 3,600 people.
Finally, with regard to program timeline, OTI’s premature departure relative to national elections is 
also unfortunate; the transition is far from over. 
V. Recommendations
1. In countries whose political transition is protracted and inconclusive, such as the DRC, if OTI is 
not prepared for a longer commitment than the standard 2-3 years, then an effective handover
strategy should be developed as early as possible. OTI/DRC did manage to ensure a short-term 
handover of its CFR capacity-building approach. Moreover, punctual interventions at key junctures –
particularly before, during, and after elections – should be prioritized over sustained interventions
that force a premature withdrawal when funds are exhausted. While OTI/DRC does not feel its
departure is premature, this view is not shared by the USAID Mission Director. Since some degree 
of sustained interventions is inevitable, these should be kept as flexible and light as possible to avoid 
onerous recurrent cost commitments. 
2. CFR is appropriate in contexts such as the DRC, where the national DDR program is poorly run. 
Reintegration and reconciliation are community concerns as well, and addressing these issues at the
community level serves the added purpose of mollifying local tensions at the core of the conflict.
When well executed, CFR can promote reconciliation and provide the tools to prevent local conflicts 
from recurring. 
3. The content of the YES training covers three modules dealing with values, conflict resolution,
and democracy/governance, plus two that attempt to inculcate life skills: health/nutrition and
agriculture/income-generation. This long training period reflected a strategy to keep people engaged
and busy, but expectations were often raised to unrealistic heights and most grants came late (or not 
at all) in this long process. Consequently, the last two modules might better be addressed through
separate follow-on activities (through small grants) to trainees and other community members, thus
shortening the long period of initial ‘classroom’ training to about three months. 
4. Based on the DRC experience, CFR appears to work best when community trainings are
combined with strategic small grants. Since collective rehabilitation projects are best conceived as
the practical application of reconciliation and conflict resolution messages learned in community
31
capacity-building, OTI should ensure that future CFR programs effectively integrate capacity￾building and small community grants. Communities receiving training but no grants in the DRC
suffered from raised expectations and significant bitterness towards the SE*CA program.
5. Greater integration of the capacity building, media, and grants components of the program would 
have been desirable and should be carefully considered in future CFR programs in Africa and
elsewhere. This primarily means linking grants more closely to capacity building and being sure
communities choose and receive what they need. Radio Listening Clubs, moreover, should be open 
to a wider membership among capacity-building graduates. Other activities that should be included
involve practical, not theoretical (as in YES training), income generation and skills development, but 
these activities should remain small in scale and appropriate in technology. The linkage with state￾of-art micro credit programs such as Project Hope in Kisangani should be emulated in future
programs. 
32
Part B: The OTI Burundi Program (CPRI)
I. Background
OTI began working in Burundi in March 2002 in response to significant advances in the peace
process after nearly a decade of civil war. These advances included the Arusha Peace and
Reconciliation Agreement (APRA) in August 2000 and the establishment of a transitional 
government in November 2001. 
Phase 1 of the Burundi program began in May 2002 and was entitled the Burundi Initiative for Peace 
(BIP). It was administered by International Foundation for Election Systems (IFES), National
Democratic Institute (NDI), and International Republican Institute (IRI). From May 2002 through
December 2003, the principal grantees awarded 80 subgrants to local organizations in Burundi. OTI 
disbursed some $2.5 million over the 20-month program period. However, the BIP was stopped 
short of its two-year authorized life and $3 million budget to make way for a reoriented Phase 2 
program. 
Phase 2 of the OTI Burundi program was named the Community-based Peace and Reconciliation
Initiative (CPRI). It was to run from February 2004 through February 2006 but has recently been
extended through June 2006. This four-month extension involves all components except the
Community-based Leadership Program, which will end as scheduled.
II. Evaluation Objectives and Methodology
Objectives
The second phase of the Burundi program is one of three community-focused reintegration (CFR) 
programs launched by OTI in early 2004 in Burundi, the DRC, and Liberia. These programs
employed a community-based, capacity-building approach that built on lessons learned from OTI’s
earlier program in Sierra Leone: the Youth Reintegration Training and Education for Peace Program 
(YRTEP). Since the Burundi and DRC programs were ending at about the same time, OTI decided 
to evaluate them in tandem. This twin evaluation has been designed in part to evaluate the use of the 
CFR approach in African countries beyond Sierra Leone. 
Identical to those for the DRC, the specific objectives of the Burundi program evaluation are: 
1. Did OTI contribute to the advancement of the transition process in Burundi? 
 If so, how and in what ways could it have improved this?
2. Was the Burundi program strategic in responding to shifts in the transition process?
3. Did the program meet its stated goals and objectives?
33
Methodology 
The two-person evaluation team consisted of Philip Boyle and Edward Rackley, supplied by Social
Impact, Inc. of Arlington, Virginia. The two evaluators carried out preliminary interviews
concerning the Burundi program in person or by telephone in Washington, D.C. from October 31 
through November 4. Following its time in the DRC, the team was in Burundi from November 19 to 
December 1. The visit to Burundi began with briefings and interviews in Bujumbura, then moved its 
focus to Gitega and Ruyigi provinces for three days of interviews and site visits in each. Visits were 
made to activities or accomplishments of all four program components, including six of the eight
Vocational Skills Training (VST) Centers. In the field the team was accompanied alternatively by
Joseph Bigirumwami of ASI and Emmanuel Niyonkuru of OTI, who provided translation and
explanation assistance. A series of debriefings was held with OTI and Planning and Development
Collaborative International/African Strategic Initiative (PADCO/ASI), the U.S. Embassy, and
Regional Economic Development Support Office / East and Southern African (REDSO/ESA) prior 
to the team’s final departure from Bujumbura. 
The evaluation team used semi-structured interviews (SSI) with a wide variety of program
stakeholders to explore the history, dynamics, successes, and shortcomings of the Burundi program.
Beneficiary and participant focus groups were generally assembled in field sites. Interview questions 
and techniques were similar to those used in the DRC. They employed a probing approach that
proceeds opportunistically from the general to the specific. 
III. Findings
1. Did OTI contribute to the advancement of the transition process in Burundi? 
If so, how and in what ways could it have improved this?
A number of transition processes can be identified in Burundi, all but one of which were
successfully addressed by the OTI program. Only the reintegration of demobilized combatants failed 
to achieve its expectations, but this was largely beyond OTI’s control. Burundi transition processes 
that OTI sought to support include: 
1. Return of refugees and internally-displaced persons (IDPs) to their homes and farms. 
2. Reintegration of demobilized combatants into their home communities.
3. Reconciliation of ethnic groups, refugees, IDPs, and remainees in their home communities.
4. Building a foundation of conflict mitigation and resolution in communities. 
5. Effective information flow between the capital and communities of the interior. 
6. Local and national elections process.
CPRI program components and activities were generally well conceived, appropriate, well timed, 
and certainly well managed, but the level of investment, operational cost, and inflexibility of the
VST Centers must be seriously questioned. The keystone of the Burundi CFR approach was the
community capacity-building component known as the Community-based Leadership Program
(CBLP). The program component was an adaptation of the nationally-focused Burundi Leadership
Training Program (BLTP), developed and implemented by the Woodrow Wilson Center for
International Scholars (WWICS). The CBLP, while relatively expensive, was nevertheless 
34
innovative, well timed, and appropriately adapted to support the return of refugees and IDPs,
reintegration and reconciliation of ethnic groups, and the establishment of a new culture of
community-level conflict mitigation and resolution. Community leaders that have completed this
two-month training are equipped with the skills and confidence necessary to act as mediators in
nascent community conflicts. The hope is that they will be able to head off many conflicts that might 
degenerate again into murderous internecine strife. Trainees also take pride in providing mediation
services freely, as opposed to corrupt or easily corruptible services provided by traditional conflict 
resolvers known as bashingantahe. 
The media component was effective in suppressing rumors in rural communities, bridging the
communication gap between the capital and the interior, and focusing popular attention on the key
milestones of the political transition and elections process in 2005 – constitutional referendum on
February 28, communal elections on June 3, legislative elections on July 4, and election of the
President on August 19 by the National Assembly and Senate. Because of its obvious success, many 
stakeholders felt that the media had a crucial continuing role to play in a transition process to
effective governance that is only just beginning. 
The community initiatives (CI) component was effective in integrating ethnic groups, returnees, and
remainees by bringing them together to choose, plan, construct, and manage community structures,
such as meeting halls, sports facilities, schools, and cooperative workplaces. In addition to
reconciliation and reintegration objectives, CI also laid a foundation for community participation in
local planning and outright ownership of social infrastructure. However, the appropriateness of some 
of this new infrastructure is open to doubt, given its high cost, uncertain future use, high
maintenance need over time, and potential for promoting social divisiveness as management
committees seek to control these structures for income generation. This is particularly true of the
Peace Centers (community meeting and event halls) and sports facilities. 
The vocational skills training (VST) component, while well intentioned, became far more complex
and costly than it needed to be to support the reintegration and reconciliation processes. It became 
coupled with an economic development objective whose potential attainment extended well beyond
OTI’s short-term mandate. Vocational skills training was carried out as an act of faith, without a
great deal of regard for the severely deprived economic context in which the eight VST Centers 
operated. Several of these centers, moreover, are found in rather marginal, rural locations. 
The timing of the CPRI program components was generally effective, but off with respect to
reintegrating demobilized combatants. Disarmament, demobilization, and reinsertion (DDR) are
handled by the National Commission for Demobilization, Reinsertion, and Reintegration (CNDRR),
which although it got under way in December 2004, only began to process out significant numbers 
of demobilized ex-combatants in the final months of CPRI activity. At the time of evaluation, there 
were only 84 official ex-combatants in the VST Centers in Ruyigi. This does not mean that CPRI did 
not attempt to include unofficial or self-demobilized ex-combatants from the beginning, nor that it
does not have close working relationships with the CNDRR. The VST Centers particularly have
brought in ex-combatants whenever possible during all three VST cycles. 
Nevertheless, by the force of circumstances community-based reintegration and reconciliation
generally focused on bringing together refugees and internally-displaced persons (IDPs) of various
35
ethnic groups with populations that had remained in their communities. This is still an important
function, given the ethnic focus that political conflict quickly took on in the early 1990’s. 
The targeting of beneficiary groups important to the peace and political transition process was also 
quite good. Although relatively few officially demobilized combatants were available, returnees 
from refugee camps in Tanzania (Hutus) and returnees or inhabitants of IDP camps in Burundi
(Tutsis) were clearly mixed in OTI program components with those that had remained in their
communities. The latter generally looked down on those that had fled, and land conflicts had 
proliferated. OTI also took pains to integrate women, youth, and the Twa pariah ethnic group into
beneficiary populations. 
Geographic targeting of OTI activities to former “hot spots” of ethnic conflict and civil war was
deliberately pursued in the CBLP and CI components. However, the CPRI program was only
implemented in two of Burundi’s 17 provinces, and the two provinces selected were not the most 
conflictive at time of program start-up. U.S. Embassy security concerns and OTI programmatic 
interests directed CPRI to operate in Gitega and Ruyigi provinces, where there had certainly been a 
history of ethnic conflict. There was a high number of IDPs in both provinces, and Ruyigi has had 
the highest number of returning refugees from camps in Tanzania. Large numbers of ex-combatants 
were expected to resettle in both provinces. Gitega was chosen largely for its central location and 
political importance; it had been the traditional capital under the Burundian kings and colonial
regimes and the seat of the major Hutu opposition party in the 1990s. 
Nevertheless, the U.S Embassy has not allowed expatriates to reside in the field. This encouraged the 
selection of provinces fairly well linked to Bujumbura. Some of the peripheral provinces to the north 
and east were more insecure than Gitega, but getting started was of greater priority for OTI than 
waiting for optimal targeting. This is a common trade-off issue in OTI programming. An early plan
to replicate the program in at least two more provinces never materialized. Moving into two or more 
additional provinces would have meant reaching more of the Burundian population as security
progressively improved, but this would have required a much lighter investment in program
components, particularly the VSTs. 
The VST component, in fact, chose to rehabilitate pre-existing vocational schools put in place by a 
World Bank project in the early 1980s. By choosing rehabilitation, OTI committed itself to
immobility with respect to unfolding events of ethnic return and reintegration. While the centers did 
serve to bring together various population elements, the capital outlay and operational costs of these 
eight centers far outweigh their community reintegration or economic impact. 
In June 2004, OTI and its implementing partners (PADCO, ASI, and WWICS) conducted a baseline 
assessment for CPRI in all 18 communes of the two program provinces.
9
 This was to be the first of 
biannual follow-up assessments over the course of CPRI, but this was done only once more in 
December 2004 and no overall report was assembled. This is unfortunate, since the information
gathered could have provided some glimpse into unfolding impact. A stocktaking exercise was
conducted by USAID in November 2004, but the report covers the whole USAID program.
10
 
 
9
 USAID/OTI: PADCO, ASI, WWICS. “Baseline Assessment: Gitega and Ruyigi Provinces.” June 2004.
10 USAID. “USAID-Burundi Stock-taking.” December 2004. 
36
The baseline assessment relied on focus groups in all program communes to collect basic awareness 
and attitudinal information on nine major areas of interest: the Arusha Accord; elections and the
political transition; civic education; leadership and conflict resolution; reconciliation and justice;
democracy and citizen’s rights and responsibilities; strengths and weaknesses of communities;
community development and NGO assistance; and the media and sources of information.
Although the baseline assessment was carried out after CPRI was designed and under way, it draws 
conclusions that validate previous programming choices. It also billed itself not “as a primary tool
for data collection to measure progress toward the program’s objectives” but rather as a contribution 
to the “ability to determine if the program is achieving its intended impact.” 
In the absence of planned semi-annual updates, the usefulness of the baseline assessment was
unfortunately diminished as a means to monitor and steer the CPRI program. There is no evidence 
that any attempt was made over the life of the program to track the key issues discussed in the
baseline document. Only the Mid-term Assessment 11 returns to key concepts of the CPRI program. 
Nevertheless, the baseline document outlined an interesting set of key issues (trends) that future
assessments should address. 
The evaluation team certainly found ample anecdotal evidence in communities to confirm baseline 
program hypotheses. Communities draw a clear relationship between the peace process, the political
process, and the presence of CPRI activities in their communities, although they may not always
know that CPRI is behind media activities. Community members also generally know of the VST
Centers and of their relationship to CPRI, although they referred to the program generally as
“Padico,” their version of PADCO, the principal contractor. Community members also appear to be 
more confident of their own powers and of their ability to influence the behavior of their informal
and formal leaders. Finally, as predicted at baseline, there were unintended media benefits beyond 
those contributing to conflict mitigation by providing reliable information. These had to do with
raising public consciousness about the electoral process and nipping numerous cases of electoral
fraud in the bud. 
How might OTI have improved its performance in the Burundi Phase 2 program? In this regard, it is 
difficult to see how CBLP content or training could have been designed differently to improve 
community impact. The amount and length of training – eight 2-4 hour training sessions spread over 
six to eight weeks – appears to have achieved its purpose cost-effectively. Nine cycles of training
were carried out in all 18 communes, and ten occurred in one commune. However, salary costs
associated with the 20 Master Trainers posted to the 18 communes and eight VST Centers appear 
unreasonably high (about $1,000 per month per trainer). Nevertheless, beyond salaries and some
materials, CBLP remained quite light and flexible, eschewing building training structures or even
distributing travel allowances or per diems to trainees. The CBLP itself has been managed by ASI as 
sub-contractor to PADCO, resulting in a costly superimposition of overhead structures. This 
component’s implementation structure could and should have been lightened.
The CPRI media component generally seems to have been both effective and cost-effective. In 
retrospect, however, OTI could have funded the start-up and distribution of a newspaper to rival the 
government and party-dominated newspapers that rarely circulate beyond Bujumbura. This could
 
11 Briggs, J. and J. Rigby. “OTI Burundi Mid-term Assessment Report.” February 9, 2005. 
37
have been launched and maintained for two years with reasonable cost. Setting up and supporting a 
newspaper would not have cost as much as one VST Center. The target audience, while obviously 
more limited than that of radio broadcasting, nevertheless would have included youth in schools,
local politicians and administrators, and a variety of literate persons capable of exerting influence on 
the transition to democratic governance. 
The community initiatives component focused on very visible, highly integrative, and relatively
large community projects. While this is certainly an OTI preference, there is a dilemma in doing this 
in extremely poor countries, such as the DRC, Burundi, and many others in Africa. The poverty
context is such that fundamental infrastructure, such as potable water sources, small bridges, health
posts, roads, crop storage sites, market structures, and primary schools do not yet exist or have been 
destroyed by war. Given the great community need for basic services infrastructure, it seems more 
reasonable to spread a greater number of small community projects over a wider area than to
concentrate on a smaller number of large projects in a more restricted area. If the objective of these 
community projects was to bring people together around common needs, small infrastructure would
not only have been equally effective, but also more useful to the communities over the longer term. 
Post-construction management issues would also be more tractable, more likely to be within
community experience, and less likely to lead to conflict over control. 
The VST component could have been simplified considerably from the beginning. Running at nearly 
$90,000 a month ($540,000 per cycle) and involving sunk costs of some $450,000, the centers are a 
wonderful, but inappropriate, accomplishment with unclear beneficiary outcomes. According to OTI, 
these costs have been reduced to $500,000 for the current (4th) VST cycle, which primarily reflects a 
reduction in course length. This is surely not the most cost-effective way to reintegrate ethnic
groups, refugees, IDPs, and former combatants, while at the same time providing them with basic 
income-generating skills. A more effective means to achieve this dual objective would have been to 
expand community mobilization activities to include support to existing or potential economic
associations. This was a desire expressed repeatedly by both graduates of CBLP training and
community members involved in large community projects. Moreover, some non-VST economic
associations have been supported by CPRI already. The rather lavish support to the weavers and
soap makers of Nyabitsinda is a case in point, but building two craft centers in the same small 
community is certainly not cost-effective. Using technology appropriate to the economic
environment, trainers could have equipped existing or new associations with skills, tools, and even 
start-up kits necessary for launching activities, much as promoters are now doing with graduates of
the VST Centers. By not launching too many similar associations in the same communal zones,
over-supply relative to local demand and purchasing power may be avoided. There is certainly risk 
of this among graduates of the VSTs. 
2. Was the Burundi program strategic in responding to shifts in the transition processes?
The CPRI program was generally strategic in its approach. This is true in a number of ways. First, 
the program began as a strategic shift away from a national-level approach toward a community￾level focused program (except for media). By September 2003, the OTI Country Representative in
Burundi set out to swing the program around to reflect new priorities on the ground in rural
communities. The peace process at the national level seemed on course and supported by the
Burundian population. A culture of peace and justice at the national level seemed attainable, but
38
avoiding a return to the bloody ethnic massacres of the preceding decade required a shift of focus to 
the community. As the Mid-term Assessment points out, “OTI also concluded that it would be best 
to approach the reintegration effort as much from the point of view of the disposition and capacity of 
the receiving communities as from the perspective of citizens returning to their homes.” 12
A full strategic plan was produced for CPRI by the end of 2004, but it built on an earlier M&E
Performance Monitoring Plan (June 2004) drawn up close to program inception. The strategic plan
covers the background, rationale, and structure of CPRI and presents input, output, and
outcome/impact indicators for the three program objectives. There is no evidence that data was
systematically collected and published on these indicators, although the data must exist in various
places. A brief section of the plan describes how the various program components fit together to
achieve the program goal. 
The CPRI program certainly had broad vision, clearly defined objectives, logically interrelated
components, and did not through its life ‘lose the forest for the trees.’ In other words, there is clear 
evidence of strategic intent in the formulation and management of CPRI. Project design drew
strategically on the successes of a previous model in Sierra Leone and the availability of a conflict￾resolution training program in Burundi (BLTP). It was to be an internally consistent (holistic)
approach to reintegration: conflict resolution/mitigation training of leaders for maximal spread;
targeted community grants to bring people together and to involve CBLP graduates in community
mobilization and participation; and vocational skills training to jump-start economic opportunities
for returnees and remainees. As it turned out, program components worked largely separate from one 
another, even in the case of the two community-focused components (CBLP and CI). OTI/Burundi 
points out, however, that Master Trainers (CBLP), Program Officers (CI), Business, Numeracy, and 
Literacy facilitators (VSTs), and reporters (media) all were focused on the same communities.13
 
While the choice of Gitega and Ruyigi provinces were not the most strategically appropriate for
reaching CPRI objectives, security concerns restricted OTI to these safer provinces. Within those
constraints and the need to move forward rapidly, these two provinces do not seem badly chosen. Of 
greater concern is the decision to concentrate all project resources (except media) in just two
provinces, when four or even six might have achieved greater impact with costs controlled by
maintaining a lighter and more mobile mix of interventions. 
 
The degree to which program components were able to shift along with the political transition and 
reintegration process varies by component. CBLP and media activities were able to shift focus onto 
the referendum and elections process at the appropriate times. This is particularly true of media 
activities, but elections and their relevance to community peace and conflict reduction were also 
focused on during CBLP training sessions. 
At the request of the President and the United Nations Mission (ONUB), CBLP content and Master 
Trainers were also used in the run-up to the legislative elections in June and July 2005. Trainers
 
12 Briggs, J. and J. Rigby. “OTI Burundi Mid-term Assessment Report.” February 9, 2005. 
13 Master Trainers from CBLP identified communities where community initiatives were implemented. Master Trainers 
participated in meetings and mediated conflicts on community initiative grants. VST Business, Numeracy and Literacy 
(BNL) facilitators trained all community initiative association members. VST schools supplied labor and filled orders for 
community initiative grants. Finally, journalists under the media component covered stories on CPRI grants.
39
fanned out to 34 communes nationwide and gave rapid training using a condensed version of the
three modules to 1,000 political party candidates and other leaders. Elected officials were also
targeted for training after the elections. The 34 communes were considered to be the hottest and 
potentially most divisive during the elections period. A good deal more of this flexible, mobile
programming might have been possible, had the Burundi program not ‘mortgaged’ itself so deeply in
two provinces. 
Community initiatives do not seem to have been modified in content or strategic importance until
recently. The Mid-term Assessment recommended that CI should receive far more attention and
another $500,000. Under the recently approved cost extension, community initiatives will be
stressed, since it is obvious that they could have been more effective in mobilizing communities to 
work on projects of common interest and clearer usefulness. CI could have been used more
effectively strategically, if focused on a wider area with smaller projects. Reintegration of refugees
and IDPs, reconciliation of ethnic groups, promotion of local democracy, community participation in
development, and demonstrating the dividends of peace could have had greater geographic coverage 
through a wider spread of projects focused on basic population needs. 
It should be noted that a local population may indeed request projects that do little to address its
basic needs. Does this represent proto-democracy in action, and should OTI adopt a position that 
whatever community members want, they should automatically get? This does raise an ethical issue, 
since the trade-off for community recreation centers and sports courts in the Burundian context may
well be potable water, health, and education infrastructure that saves lives and forms the next
generation. Strategically speaking, if this kind of infrastructure can unite and reconcile communities
as effectively as larger, less utilitarian structures, it should be given priority in future OTI programs
in the least developed countries. 
There should have been a strategic decision from the beginning not to invest so heavily in the VST
Centers, since doing so clearly drained resources from CI. Given this decision to proceed with the 
centers, OTI should have turned its strategic attention from the beginning to the politically-sensitive 
issue of how and to whom to hand over the VST Centers. Only in the last few months has serious 
thought been given to the handover issue, which has not yet involved the other program components. 
Moreover, the issue of how to keep the VSTs going seems as much related to avoiding the
embarrassment of future abandonment than how to keep a worthy benefit stream flowing. If costs 
can be reduced considerably, it is possible these centers may yet prove to be cost-effective. 
3. Did the program meet its stated goals and objectives?
A final version of the CPRI Strategic and Performance Monitoring Plan dates from January 2005, 
although the Performance Monitoring Plan (PMP) was modified in July 2005. The objectives
correspond well to program components and have not changed since they were first formalized in
late 2004. According to these documents, the CPRI program has the following goal and objectives: 
Goal: Strengthen local capacities to benefit from and contribute to the peace process.
Objective 1: Strengthen local-level cooperation for conflict mitigation.
Objective 2: Help groups incorporating diverse population segments generate non-farm income.
Objective 3: Increase knowledge and participation in public dialogue on current issues.
40
Grant Activities in the OTI/Burundi Database
The OTI/Burundi database contains 211 grants that have been cleared, completed, or closed. The 
total value of these grants is $4,510,170. However, the database focus areas (except media)
correspond neither to program objectives nor to what we have termed program components. The 
focus areas break out as follows. 
Focus Area Number of Grants Value of Grants Percent of Total
Civil Society 
Organizational Support 
5 $125,304 2.8%
Community Impact 
Activities
38 $829,899 18.4%
Conflict Management 119 $2,629,374 58.3%
Election Processes 2 $65,413 1.5%
Justice/Human Rights 2 $25,169 0.6%
Media 41 $802,262 17.8%
Transparency/Good 
Governance 
4 $32,750 0.7%
Total 211 $4,510,170 100%
While these categories undoubtedly have their own logic, they make it difficult to break out the grant 
budget according to the objectives and components below. It should be noted that about half (2/3 by 
funding) of what are commonly called Community Initiatives are found under the Community
Impact focus area and another half (1/3 by funding) are located under Conflict Management. The
Community Impact and Conflict Management focus areas also contain the grants involving VST and 
CBLP (or CBLP-related) activities. All media activities appear under the media focus area.
The other four grant focus areas collectively count 13 grants for $248,456, only 6.2% of total grants 
and 5.5% of total grant funding. Thus, Civil Society Organizational Support contains five grants, of
which all but $12,000 went to the Burundi Leadership Training Program. Election Processes consists 
of two grants of which the lion’s share ($55,174) went to the BLTP for training of political party
leaders. Justice/Human Rights contains two grants equaling $25,469 for two conferences, and
Transparency/Good Governance has 4 grants totaling $32,750, of which $28,097 is for provincial
governmental offices (Gitega and Ruyigi). 
Objective 1: Strengthen local-level cooperation for conflict mitigation (CBLP, CI)
Both the Community-based Leadership Program (CBLP) and community initiatives (CI) component 
served Objective 1 and sought to repair the social fabric and cohesion of previously warring
communities. The CBLP component falls under the category of community capacity-building, 
similar to the processes employed in the Community-focused Reintegration activities carried out in
Sierra Leone, Liberia, and the DRC. It was not adapted directly from the Sierra Leone program,
however, as in the DRC. 
On the other hand, the community initiatives component (formerly known as small grants), 
represents a common mechanism by which OTI operates in political transition environments. While 
CBLP sought to change behaviors and perspectives by learning transfer in an educational mode, CI 
41
endeavored to change habits, behaviors, and preconceptions by bringing formerly mutually hostile 
people together in undertakings for the collective good, primarily construction activities. Community 
initiatives also establish a framework for future participatory democracy in communities where it is 
employed. 
CBLP Training
The CBLP drew its inspiration and content from the Burundi Leadership Training Program (BLTP)
implemented at the national level since 2002 by the Woodrow Wilson Center for International
Scholars. The national-level training modules were specifically adapted to the rural context and
targeted largely illiterate, non-governmental local leaders. While the BLTP training entails six full
days, the CBLP training is about half as long, and spread over two months. A training manual was 
produced by the Woodrow Wilson Center and is dated April/May 2004. CBLP training consists of 
three community capacity-building modules, plus a module for trainer development: (1) Self￾management and Our Own Role in Conflicts; (2) Stimulating Dialogue; (3) Managing Conflicts
Effectively; (4) Facilitation.
The Woodrow Wilson Center also trained some 40 Master Trainer candidates, from which 20 were 
finally selected for CBLP work in the field. One Master Trainer was assigned to live in each of the 
18 CPRI communes, and two more have worked only with the eight VST Centers. Following
development of the curriculum and selection of Master Trainers, the Burundian NGO African
Strategic Initiative took over management of the CBLP program. 
Training of community groups is spread over six to eight weeks and involves eight 2-4 hour
sessions. This has entailed only one or two training sessions per week, which should not have posed
an undue burden on participants’ time. Moreover, some absenteeism was tolerated, so that trainees 
received a graduation certificate if they attended at least six of the eight sessions. 
There have been nine groups trained to date in each of the 18 CPRI communes, while one commune 
received a tenth training. The number of total groups trained is 163 and involves some 6,850
participants. Group size generally ranged from 30 to 50 participants, with an average number of 42. 
Of this total number of trainees, some 46 % are female. Upon graduation, participants are given a T￾shirt, an umbrella, and a certificate. No other rewards are given. 
Operational costs of the CBLP component are said to be about $175,000 per three-month cycle
($58,000 per month), but this includes monthly follow-up refresher meetings for graduates. The nine 
training cycles carried out to date must have cost about $1,575,000, a figure very close to the
operational cost of three cycles of VST training (cf. Objective 2 below). However, upfront costs are 
obviously considerably less for CBLP than for the VST component. Given the large number of
persons trained, the cost per trainee is about $230. 
Most of this relatively high operational cost for such a “soft” training component is explained by
salary load, primarily for Master Trainers. At about $1,000 per month per trainer, the 20 Master
Trainers entail a joint monthly wage cost of $20,000, some $240,000 per year. Much of the balance 
must be explained by salaries and overheads of two Provincial CBLP Coordinators, the ASI director 
42
and deputy director, the PADCO CBLP component manager, and the Woodrow Wilson Center
through which the training is actually financed. 
CBLP groups were organized and facilitated by Master Trainers working in teams of two, although 
only one lived in each of the 18 communes. Participant groups were drawn from all local
administrative units (hillsides) surrounding the training sites and quite deliberately brought together
the three formerly antagonistic ethnic groups: Tutsi, Hutu, and Twa. They also brought in a mix of
remainees, returnees, and persons still living in nearby IDP camps, where present. 
Master Trainers conducted the CBLP selection process on the hillsides by seeking out those
considered by community members as leaders (imbodedze), that is, persons of influence and
confidence who tend to guide others. Leaders were focused on to enhance the likelihood of rapid and 
effective conflict mitigation or resolution and to encourage the spread of changed perspectives and 
behaviors through their own high status and example. This appears to be a sound decision. 
The overt anti-corruption message of the CBLP was that participants would receive no travel 
allowance or per diem, not even refreshments. This has also been the practice in the VST Centers. 
No other donor in Burundi insists on this lack of incentive, and it remains to be seen what impact 
this may have on future donor-beneficiary relations. Moreover, conflict mitigation services provided 
by graduates are to be given free of charge, which is also in contradistinction to common practice at 
all levels of Burundian society. It seems this morality message has been understood by participants, 
although there is still some mocking from others. While OTI remains strong in its conviction that
this is the “right step for positive engagement,” trainees in general expressed strong dissatisfaction
with this practice. Moreover, it is at least possible that the poorest community members might find it 
more difficult to leave subsistence (or survival) activities to attend meetings. This would have been
far less a problem in the Burundi than the DRC CFR program. 
At first, traditional conflict resolvers (bashingantahe) and local government officials felt threatened
by the creation of a credentialed set of new community leaders and conflict mediators, but many of
the bashingantahe were simply co-opted into later training sessions. However, communal
government officials were not added to CBLP trainings until they were actually elected (June 2005). 
CBLP training has clearly impressed and galvanized participants, who claim now to be actively 
acting as cost-free mediators at the first-level of conflicts, leaving more complex issues to the 
bashingantahe or, in serious cases, to communal or provincial authorities that make binding judicial 
decisions. Moreover, CBLP graduates claim to be in high demand for their services. The 
bashingantahe declare there is no conflict of interest between the two groups, since the CBLP
graduates simply filter out the less complex cases before they reach the next level. 
Each commune now has two to three Community Facilitators (CFs) who have been selected by
Master Trainers to assist them in their work and spread the principles and lessons of the CBLP to 
surrounding areas. Although they have been provided with a bicycle, raincoat, a radio and boots, it 
does not seem likely that Community Facilitators will long be able to replicate the training or
maintain the enthusiasm of graduates with the end of OTI support. 
43
In sum, OTI’s objective of strengthening local-level cooperation to enhance and improve social
mechanisms for conflict mitigation and resolution was effectively addressed through the CBLP. By 
focusing on leaders, OTI improved the chances of impact and its potential spread. It is likely this 
training will remain as a substantial legacy to OTI activity, but the degree to which it will spread to 
new communities is unclear. If it does spread, it will probably do so informally through the influence 
and leadership of graduates. 
In this regard, a number of these graduates have already been elected to positions of authority as 
zonal chiefs or communal councilors. In Gitega, out of a total of 1,325 elected representative to the 
new hillside councils, 242 (18.3%) are CBLP-trained leaders, including 44 women. Some 45 were 
elected chef de colline (hillside chief), including 1 woman. In Ruyigi 323 CBLP leaders were
elected, including 33 women, and 37 CBLP-trained candidates were elected chef de colline.
Community Initiatives
Community initiatives focused on Objective 1 by bringing together mixed ethnic groups, as well as 
returnees, remainees, and IDPs, on construction sites and into management committees for structures 
once completed. Working together for community projects was an effective reintegration and
reconciliation tool, and it certainly promotes local-level cooperation that mitigates renewed
outbreaks of ethnic conflict. Organizing management committees around completed structures not
only provided sustainability of local cooperation, but promoted an example of grassroots democracy 
as the elections process unfolded and ran its course in 2005. 
The CI component is managed in the field by program officers, two of which divide each province. 
The program officers are the interface with communities with respect to generating small project
ideas, although Master Trainers often are involved in assisting communities to generate project
ideas. 
An examination of the OTI/Burundi database reveals that some $937,112 has been spent to date on 
what can be called community initiatives, most of this on infrastructure projects. This includes the
sizeable amounts of money spent on building two new schools to replace those displaced by the
rehabilitation of VSTs. The construction of the Giheta secondary school, for example, cost $103,722. 
Some $600,411 of these CI grants are found in the database under Community Impact activities, 
where they account for about 72 % of monies spent under that focus area. The remaining $336,701
that are clearly centered on community initiatives, as opposed to VST support, training communal
leaders, and other non-community support activities, are found under Conflict Mitigation, where
they account for about 13% of monies spent under that focus area. 
All together, the $937,112 spent on what are clearly non-VST and non-CBLP community initiatives 
constitute 36 of 211 grants (17.1 %) and 20.8 % of total grant funding ($4,510,170). The average CI 
grant size is $26,031. However, grants range in size from $103,722 to $1,806. 
There appear to have been only 11 water supply projects among community initiatives, and these 
generally varied in size from $3,500 to $30,000. Eight of these 11 projects, however, supplied water 
to VST Centers, with additional supply for the surrounding communities. These eight water projects 
totaled $110,569 and may be counted among CI activities. On the other hand, they could also be 
44
included in the numerous grants focused on the VST Centers, since they would probably not have 
been done without the need for water in the VSTs. 
The construction of three large community recreation centers (Peace Centers) was costly ($113,713) 
with an average cost of about $37,900, but they had the advantage of pulling together a very large 
number of participants on one site. These participants worked in shifts for remuneration once the
foundation was laid as a community contribution. However, these Peace Centers may result in
heightened community tensions, because of the power for income-generation their management
conveys. While creating local power centers related to managing community infrastructure is
definitely not the intention of this CPRI component, it might be an unforeseen consequence. An 
interesting parallel may be drawn from a quote from an archaeologist excavating Stonehenge: 
“Building Stonehenge might seem an absolutely crazy thing to do, but involving lots of people in a
construction project is a way of bringing people together [for] creating and seizing status.”14
Other large construction expenses appear to center primarily on primary and secondary schools,
although the Mubira Sport and Water construction grant at $62,281 (primarily water supply) is well 
above the cost of the recreation centers ($43,184 for the most expensive in Buhevyi) and more
expensive than all but one of the schools ($79,786 for Gisiru Primary School). Nonetheless, these
projects had the advantage of concentrating people on work sites and by doing so, rapidly worked 
toward community reconciliation and reintegration after years of conflict. 
Objective 2: Help groups incorporating diverse population segments generate non-farm 
income (VST).
From the beginning, OTI pushed PADCO to complete three training cycles by end of program. This 
has been attained, in spite of a slow start in mid-2004. The VST Centers have produced some 2,760 
graduates, of which about 32% are women. As many as 250 income-generation associations have 
been created to launch graduates into sustainable small business activities. Once under way, the VST 
component has demonstrated a high degree of operational capacity and efficiency that surely
exceeds what one would expect of a normal development project. 
While non-farm income generation is the main objective of the vocational skills training component, 
the VST Centers are also an effective mechanism to reconcile ethnic groups and various categories 
of displaced persons, although not always in their communities of origin. The VST Centers are in 
fixed sites that serve the communes in which they are located, but trainees may come from a wider 
area. There is high and increasing demand for entry by young men and women. 
After successfully completing the six-month course, VST graduates maintain ethnic mix in the small
enterprises (associations) they form and spin off into surrounding communes. Associations number
from eight to 16 people and receive a start-up kit worth from $400 to $800. Kits consist of enough 
basic tools (e.g., sewing machines or saws) and raw materials to allow associations to continue
functioning as a group. 
When functioning at full capacity, each VST has six trainers and 11 assistant trainers for the six
trades taught: bread making, brick making/tile making, carpentry, furniture making, masonry, and 
 
14 Smithsonian. August 2005. Volume 36, Number 5. Smithsonian Institution. Washington, D.C.
45
sewing. Brick making/tile making and carpentry have been discontinued in most centers and the
usual six-month cycle has been trimmed to five months. 
Given the low level of education of most VST trainees, classes in business, numeracy, and literacy
(BNL) also form a part of their education while at the Centers. This includes time spent on civic
education, human rights, and elections education. Beyond this, the full CBLP course is given to
trainees by two Master Trainers who focus only on the centers. 
The VST Centers also contain a Business, Numeracy, and Literacy (BNL) instructor, a BNL
facilitator (outreach to associations), an accountant, and a center director. Counting a large number 
of support and maintenance staff, each VST has 27 to 28 employees. Oversight for the VST
component in each province is provided by a coordinator for VST training and a coordinator for 
BNL. Overall component management is carried out from Bujumbura. It should be noted that the 
core staff for both the CBLP and CI components is larger than that involved in the VST Centers. 
The cost of running the VST Centers is about $66,000 per cycle per center. This entails a total bill of 
some $528,000 for a six-month cycle, or approximately $88,000 per month. The annual cost of
operating these centers must be about $1,056,000, when running at full capacity. The cost per
graduate over the three cycles of activity is about $574. Adding each graduate’s share of the start-up 
kit would add another $50, bringing the overall total to $624. This does not factor in salaries for the 
national VST component coordinator, the four provincial VST and BNL coordinators, or the two
Master Trainers assigned to VST Centers. Moreover, this operational cost follows upon sunk costs of 
$400,000 to $500,000 made in 2004, which will have to be depreciated and replaced at reasonable 
intervals in the future. 
A survey is now being conducted on the income-generation of the 1,850 graduates of the first two 
VST cycles, whose results were not available at the time of evaluation. One factor is clear: 
graduates tend not to work at their trade full time, but combine it with traditional agricultural
activities. With regard to their current income position, they are all still likely to be using the tools 
and materials supplied with their start-up kits. This is all the more true since graduates of the first 
cycle only received their full kits in November 2005. At present, it is unlikely these graduates and 
associations are generating much original income, but many may do so over the long term. A better 
time to do a follow-up survey of these graduates would be about two years after graduation. 
While significant non-farm income generation over the lifetime of these graduates seems likely, OTI 
is a short-term, bridging mechanism. This program component, while impressive and effectively
managed by PADCO staff, does not fit with the CBLP, CI, and media components, since it is
designed to provide income in an economic context that cannot support so many graduates except
over the long term. The VST Centers, in many ways, have fallen into the typical development trap of 
producing an overly costly, unsustainable, and ultimately inappropriate solution to the problem they
are attempting to alleviate. 
OTI/Burundi notes that the VST program was not conceived primarily to address economic issues, 
but rather served as a mechanism to bring together diverse groups. Moreover, OTI feels that this was 
the most successful component at population mixing. Although figures were not kept on ethnic
identity in the VST Centers, it is possible that they achieved a greater mix of ethnic groups with 
46
youth and self-demobilized ex-combatants than the very localized CBLP trainings and community
initiatives. It is nonetheless true that this is an over-elaborate means by which to achieve population 
mixing. 
OTI notes further that the graduates, because they have remained together for six months, received 
the CBLP training, discussed evolving political issues derived from media programs, and developed 
saleable skills, constitute the beneficiary population most affected by CPRI activities. This is a
reasonable assertion, although the significance of this is unclear. OTI also maintains that VST
graduates are “perhaps more adept at CBLP mechanisms than many of the community facilitators.”
This seems unlikely. 
On the other hand, the VST Centers by their very size have generated a high level of visibility for 
USAID and the U.S. government, and they have found great favor with provincial governors and
national politicians and functionaries. However, in many ways these structures stand out too much,
raising the specter of U.S. Embassy embarrassment if they cannot be sustained. The VST Centers are 
considered to be a great success, but they may well become monuments to emptiness. 
Handing over an operation costing $88,000 a month and containing sunk costs of about $450,000 is 
no easy task, as OTI has come to realize in recent months. If the centers are taken over sometime 
next summer, their personnel and operational costs will need to be reduced drastically. They must 
certainly become schools where trainees are trained on appropriate local tools. Moreover, continuing 
to provide technical assistance to graduates’ enterprises does not seem cost effective, since it is
likely they can be employed more effectively by existing or future private sector entrepreneurs. One 
case was noted by the evaluation team of a former instructor in a VST who left the school with a 
large part of his class to create an enterprise. The latter has already received a large number of
carpentry orders for the reconstruction effort funded by the United Nations. 
Objective 3: Increase knowledge and participation in public dialogue on current issues
(Media). 
The media program bridged the national and community level effectively. It did this by assisting the 
state-run Radio Télévision Nationale Burundaise (RTNB) and privately-held Radio RSF Bonesha to 
broadcast nationwide. It also provided relay structures for Radio Isanganiro (supported by the NGO
Search for Common Ground). The media program also followed the transition process, by focusing
on the referendum (February 2005) and the various elections as they occurred later in the year. Also 
impressive was the grant support given for media monitoring to the Organisation des Média de
l’Afrique Centrale (OMAC). This monitoring covered the six months from June to December 2005
and focused on the elections process and aftermath. It involved reviewing and analyzing newspaper
coverage and email sites, as well as radio broadcasts. 
Altogether, 41 grants have been given out to date by OTI under the media component for a total of 
$802,262. Average grant size was $19,567. The media component thus constitutes about 18% of
total grants given out under CPRI ($4,510,170). 
47
RTNB and Radio RSF Bonesha15 began active programming collaboration with CPRI in middle to 
late 2004, following a media assessment (March 2004) that indicated that radio was the predominant 
mass medium in the country.16 The assessment indicated that OTI/PADCO had previously upgraded 
the RTNB and Bonesha FM technical equipment due to the “obsolete or declining technical quality 
of their original studios, production and transmitting equipment.” New equipment was supplied later
to both stations. 
Both Radio Bonesha and RTNB are FM, but with relay points they are able to cover the whole 
country. The principal objective of OTI support to radio programming has been to fill the
communications gap between the capital and the 95% of the Burundian population that live outside 
Bujumbura. A specific objective was simply to dispel rumors. These had caused countless
internecine struggles, even massacres, during the nearly 10 years of civil war. 
Quarterly programming plans were developed by the PADCO media advisor in collaboration with
both RTNB and Radio Bonesha. At the same time, an evaluation was conducted of the previous 
quarter’s programming. Quarterly plans were intentionally developed to respond to the shifting
needs for information in the rural areas, and grants to the radio station partners were made each
quarter for specific activities and content. They varied in amount but were generally around $40,000
to each partner. In this way, the content of radio programming closely followed the peace and
political transition process. This use of short-term programming grants, while somewhat labor
intensive, nevertheless ensures a high degree of component flexibility and targeting. 
One or twice a week journalists from both stations have gone into communities served by OTI in the 
two program provinces, but each month journalists are also sent to another province. Approximately 
one of every four visits by journalists has been to non-program provinces. In this way, most other 
provinces have also been served to some degree by CPRI. Journalists are provided with vehicles, 
fuel, and per diems to penetrate far into rural areas to do public interest segments that the common 
people can listen to later. Local populations are encouraged to sound off on issues of the day. Visits 
were keyed to the elections process, among other topics. 
Both Radio Bonesha and RTNB directors claim that local and national authorities and decision
makers now listen to what the population is expressing through journalist visits and are undertaking 
visits themselves to converse directly with rural populations. Moreover, the directors take credit for
ensuring free communal and legislative elections by placing journalists on the ground on election
day. In their opinion, without radio reporters in the field the Independent National Electoral 
Commission (CENI) would not have been able to ensure honest and fair elections in June and July
2005. These directors declare that “freedom of the press has now been confirmed as a power
amongst others” as democracy moves forward in Burundi. Five years earlier, they state, the media 
“could not have criticized ministers as they now do.” 
The principal radio stations, including Radio Bonesha and RTNB, joined together to create a “Media 
Synergy” to follow the elections process throughout Burundi. This collaboration resulted in
extremely fair elections by exposing cheating and electoral fraud on the ground. The stations
carefully watched the counting of ballots. 
 
15 RSF = Radio Sans Frontières. 
16 PADCO. March 2004. “Media Baseline Assessment.”
48
The media program seems cost-effective for its reach and message, but it could also have launched 
some type of written medium for circulation through schools and communities. Radio broadcasts are 
very short-term; the printed word can also be powerful and is longer-lasting. It can be passed from 
one person to another and can act as a stimulant to youth schooling and adult literacy. 
IV. Conclusions
1. The CPRI program in Burundi, one of three community-focused reintegration (CFR) projects 
launched by USAID/OTI in early 2004, successfully achieved its objectives, strategically followed
the political transition, and made a significant contribution to the peace and transition process
unfolding in Burundi.
2. The Community-based Leadership Program (CBLP) was highly innovative, adapted from a
proven model available at the national level, effectively targeted on traditional leaders and persons
of probity and influence, flexible in communal location, and not demanding in time and effort
commitment from community members. High trainer salaries and overlapping organizational
overhead costs rendered this program more expensive than it might have been. 
3. The media component was highly effective in having both national and community-focus, 
remained highly targeted, programmatically flexible, and transition responsive through quarterly
granting, planning, and evaluation with one public and one private radio station partners. Media
content monitoring was also effectively carried out during the elections process and aftermath. 
4. While CPRI was strategically designed to focus several interrelated components on Burundian
communities, these components tended to function largely independently. While this might be 
expected of the media component, linkages between community capacity building (CBLP), 
community grant initiatives (CI), and vocational skills training (VST) might have been tighter. 
5. The community initiatives component proved to be an effective means to bring formerly warring
or feuding ethnic groups, refugees, internally-displaced persons, and persons that had remained in
communities together on construction sites for projects of community interest and ownership. The 
choice of specific projects remains problematic, and it is not clear that communities received the
most useful infrastructure for their future well-being. 
6. The program was least successful in integrating a vocational skills training component (VST) into 
its strategic mix. Rather than remaining light and mobile, bringing skills training out to key sites in a 
wide variety of communes, CPRI chose to go deep and fixed in eight locations, engaging in
substantial rehabilitation of former vocational training schools abandoned a decade earlier.
Originally seen as the primary draw for ex-combatants and disaffected youth, demobilization did not 
keep pace with creation of the VST Centers limiting ex-combatant participation. 
7. Both the capacity building and vocational skills training components were expensive, particularly
the latter. The VST Centers entailed not only high investment cost, but imposed a heavy recurrent 
cost burden on the flow of OTI funding. This ‘mortgaging’ effect drained funds away from more
flexible and mobile components, particularly the community initiatives and the media components. 
49
8. While the VST trainees were clearly mixed by sex, age, ethnic group, and refugee status, the real 
objective of vocational skills training was to launch a process of economic development, without
which the designers of CPRI did not believe future internecine conflict could be prevented. While 
this may be true over the long term, OTI prides itself on providing fast, flexible, and short-term 
assistance to mitigate conflict and promote political stability. An investment such as the VST
Centers does not correspond to this self image and seriously constrains OTI flexibility and funding
as it seeks to remain strategically focused on an unfolding political transition process. 
9. From the beginning CPRI chose to concentrate its various program components in only two
provinces, neither of which for security and other reasons was the most appropriate for CPRI
activities. The justification for such narrow focus was that critical synergy could be achieved, but 
this has not been borne out. Whether narrow focus led to deep programming or vice versa, it was 
soon impossible for CPRI to spread its benefits to at least two and perhaps four more provinces as 
the peace process evolved. 
10. In spite of no fewer than five M&E approaches, ranging from contractor biweekly assessments, 
OTI regular assessments, biannual stakeholder assessments, CBLP biannual assessments, and OTI
midterm and final evaluations, key impacts on communities remain anecdotal. This is true for all
components, but is particularly problematic for the CBLP and VST components, which together
account for about two-thirds of the program. A baseline assessment was conducted, but follow-up 
biannual assessments appear to have been abandoned after the first follow-up exercise. 
V. Recommendations
1. The CFR approach used in Burundi contains a mix of components that could serve as a model for 
other countries, if lightened and reduced in cost. Cost reduction should be applied to both the
capacity building and skills training components, particularly the latter. An income-generating or 
skills development component should be included in CFR, but it must be kept as flexible and mobile 
as the capacity-building and community initiatives components. Moreover, all three components
need to be more integrated in communities. Tighter linkage of capacity building, community grants, 
and vocational skills training could be obtained by having the Master Trainers, who actually lived in
the communities, be the center point for all activities. 
2. The income-generating and skills development component could be kept more appropriate to the 
short-term, rapid, flexible, and responsive OTI approach by organizing community enterprises or
associations alongside community reconstruction projects and some type of capacity building for
perceptual and behavior change. In so doing, it would not require anywhere near the level of
physical investment and recurrent cost mortgaging that occurred in the Burundian program.
Moreover, major economic development infrastructure need not be provided, if skills and relevant 
local production are supported. Infrastructure that is provided should be oriented to water, health, 
and education needs. 
3. The capacity building component is short and flexible enough to remain highly relevant to a
reconciliation and reintegration process, without posing an undue time burden on participants. It is
not clear, however, that it is a more effective tool in promoting community reconciliation and
50
reintegration than community initiatives. Nevertheless, a combination of capacity building of this
type and joint community endeavors for the common good appears far more effective than either one 
alone. 
4. Since it is an adaptation of the BLTP training, CBLP requires significant reinforcement through
further interaction, training, and collaborative actions. In future programs, OTI should make a
concerted effort to follow-up on its impacts on communities. Ensuring the availability of one or
more community initiatives to participating groups is highly recommended. Some of these may be 
income-generation or vocational skills development activities with the new ‘associations’ created by
CBLP trainees. 
5. If OTI intends to prioritize community-based capacity building in post-war reintegration efforts, 
it must make a concerted effort to evaluate the real impact of the various capacity-building 
experiences and their relative degree of success. It set out to do so in Burundi, but failed to follow up 
on the many issues that were identified in the baseline. While the testimony of community
participants is overwhelmingly positive, it is a little hard to believe that such training can have
significant and enduring cultural impact if carried out on such small scale and with little continued
interaction. 
6. Given the experience in Burundi, national-level activities should accompany the CFR approach in 
future programs elsewhere. The media component had wide reach and effectively linked the
transition in the capital with reintegration of communities. Institution-building of emerging
government structures should be included in future OTI interventions alongside capacity building of
national leaders. All of this allows OTI to intervene in a variety of complementary levels and sectors, 
varying its emphasis through time and adjusting its mix as the transition evolves. 
51
Part C: The OTI Burundi and DRC Programs Compared
1. Capacity building component: The Burundi Community-based Leadership Program involved
some eight 3-hour sessions spread over six to eight weeks, plus periodic refresher training, while the 
DRC YES training required far more frequent contact over six months. While the former focused on 
training community leaders in conflict mitigation and resolution, the YES training covered three
political modules and two involving life skills. The objectives of the DRC training go well beyond 
those of the CBLP and certainly involve far greater opportunity cost. The immediate results of the 
YES training appear to include some of those reached through community initiatives and vocational
skills training in the Burundi program. On the other hand, the relative lack of community grants
(initiatives) in DRC led to severe disappointment among community members. It would seem that
future CFR capacity-building should be briefer than in the DRC, but possibly broader in scope than
in Burundi, although this would depend on specific training objectives. If reconciliation,
reintegration, and future dispute resolution among community members are the objectives, the
Burundi program would appear to be sufficient. If, in addition, educating communities about the
political transition and building a base for future community development are considered necessary
to mend communities torn by war, then more and longer training can be envisaged.
2. Training vs. the community grants component: There was a trade-off in both the Burundi and 
DRC programs between monies and time spent on capacity-building and on community grant
initiatives. Both components were designed to support the reintegration and reconciliation of
disparate community members, particularly displaced persons, ex-combatants, sexually-victimized 
women, and warring or marginalized ethnic groups. Capacity building sought to empower
participants with skills that could be used quickly and well into the future to prevent further 
outbreaks of violence. Community projects sought to bring various groups together in common,
community-oriented endeavors that would also provide useful infrastructure for future community
development. In the DRC so much emphasis was placed on the training component that monies for 
community projects had to be reduced substantially from planned levels in the second cycle of
SE*CA activities. This has resulted in loss of confidence among community members whose
expectations, raised by the training, have now been largely dashed. While in Burundi there was a
better balance between capacity building and community initiatives, the amount and spread of the
latter were reduced by what appears to be too great emphasis on large, show-piece community
structures. 
3. Depth vs. breadth and fixed vs. flexible in overall approach: The Burundi program went deep 
in two of 17 provinces and much of its investment was fixed and inflexible in VST centers, while in
DRC the program remained more geographically widespread and generally avoided large
infrastructure. On the other hand, the heavy emphasis and investment in a six-month training cycle 
meant the DRC program was inflexible in content and ability to shift to areas where community 
reintegration might be more appropriately pursued, such as in the two Kivu provinces. OTI/DRC 
notes that the YES training did not always require all modules and six months, but in the vast
majority of communities the six-month cycle was followed. Moreover, activity in South Kivu after
April 2005 was limited essentially to IRC use of the YES training modules. Concentration of
resources through long cycles meant that other program components had to be greatly reduced as 
budget pressures increased. 
52
4. Integration of program components: In both programs, training, media, and grants components
were seen to complement each other, but the degree to which they actually did so varied. The
Burundi program was more successful in maintaining an appropriate balance between components,
but they still tended to operate largely independently. In Burundi the media component was well
integrated into the overall community-focused reintegration activities. In DRC, the media
component seems to have been left hanging, while the Radio Listening Clubs appear rigidly
organized and uncertain in their objectives and outcomes. 
5. Geographic and beneficiary targeting and timing: Geographic targeting in both Burundi and
DRC seems to have depended far more on security concerns and U.S. Embassy permission than on 
focusing activities on the most strategic areas for OTI activities. Nevertheless, there is probably not 
much that OTI can do in this regard, if speed to get a presence on the ground remains an OTI
watchword. On the other hand, beneficiary targeting seems to have been carried out well, with the 
exception of significant numbers of officially demobilized ex-combatants, whose demobilization
lagged far behind OTI community-focused activities in both DRC and Burundi. Here again, there is 
a potential trade-off between CFR and DDR that should be borne in mind. However, there were 
numerous self-demobilized Mayi-Mayi in Maniema and central Orientale provinces that were
included in community-focused capacity building. Generally speaking, CFR will almost certainly
continue to be focused on a majority of non-combatants, if speed of entry remains an OTI value, as it 
probably should. 
6. Spread and durability of impact: In both programs formal, or organized, spread of beneficiary
impacts is likely to be minimal, in spite of attempts to empower local (community) facilitators with
the skills, motivation, and the immediate means to spread the training to other areas. In the absence 
of such formal spread, the experience of participants may diffuse informally by example, especially
the values and conflict resolution techniques in both training programs. Focusing on leaders and
persons of influence, as in Burundi, is probably a better way to ensure spread and continued use of
the training experience, but in the DRC program local chiefs were also included to enhance the
likelihood of broad acceptance and sustainability. The sustainability of OTI community
reconciliation and confliction mitigation/resolution activities does not depend on whether they can
be maintained formally, but rather on whether participants continue to impress others with what they
have learned. 
7. Handover strategy or issues: A handover strategy was pursued more vigorously by OTI/DRC
than OTI/Burundi. In DRC, the CFR program will be continued in Ituri district for another six
months, using the same contractor as in the OTI program. In the Kivus, the International Rescue
Committee will use the OTI training modules in its CFR activities. Both programs end in July 2006,
and USAID/DG Quick Response funds can also be used to meet exceptional transition needs. 
OTI/DRC negotiated a handover over several months and had a signed memo in place some months 
prior to its departure date. OTI/DRC also worked for over one year to secure funding from
CONADER to ensure program continuation in Ituri. In Burundi, handover concerns only involve the 
VST component. Sustaining this component beyond the extended OTI program is highly
problematic and seems to be motivated more by a desire not to be embarrassed than an appreciation 
of its intrinsic value. 
53
8. Community-wide reintegration vs. reinsertion of ex-combatants: OTI is rightfully proud of the 
speed with which it launches operations in a new country context. However, in situations of
insecurity that are likely to prevail after war, OTI programs may suffer from lack of optimal
geographic targeting. This occurred in both DRC and Burundi. Since the CFR approach was largely 
predicated on reintegrating demobilized ex-combatants into their communities, the actual CFR
approach employed in both programs has dealt considerably more with reintegration and
reconciliation between non-combatants. This is not without value, but it means that OTI will need to 
be aware of the fact that reinsertion and reintegration of ex-combatants may have to be left to
agencies that set up operations much later than OTI. 
9. Basic needs vs. democratic process: The community participatory approach used by OTI in
DRC and Burundi is designed to generate activities that unite and reconcile disparate community
members in endeavors for the common good. In both countries communities expressed their desire
for development projects. In Burundi community members were clearer that these community
initiatives had the objective of reconciling ethnic groups, refugees, IDPs, and community remainees. 
In DRC, where community reintegration was generally not a complex process, small grant initiatives 
were seen far more as development oriented. Given the extremely basic level of community needs, if 
small projects can unite and reconcile people about as well as large, show-piece endeavors, OTI 
should fund basic-needs activities. The usefulness to communities of water, health, and educational
infrastructure is far greater than community recreation centers or youth sports facilities. This is all
the more true when ex-combatants are not yet a major returnee population, as was the case in both 
the DRC and Burundi programs. 
10. The Community-focused Reintegration (CFR) approach: Both CFR-based programs
discovered that ensuring participation in community projects was difficult without prior training,
which served to sensitize community members to the participatory approach. Both programs resisted 
compensating beneficiaries or paying kick-backs to local officials. Yet some compensation (per
diem, meals) is justifiable, since it may discriminate against the very poor, who are unable to spend 
significant time away from subsistence activities. This is more likely to have been the case in the
DRC program with its six-month capacity building activity. OTI staff assumed that an absence of
compensation is a necessary condition of successful training activities, community participation in
grant activities, and long-term self-management of projects by local steering committees. It is the
degree and kind of compensation that matters, not whether something is given to participants.
Low number of ex-combatants: True of both OTI programs, this was due to the administrative torpor 
of national DDR programs, which made few official ex-combatants available to participate in OTI
programs. This gap was not without positive consequences, since it led to a more holistic approach
to community reintegration. 
Positive consequences of a holistic focus on war-torn communities: Lacking ex-combatants as the
core beneficiary target group, CFR programs were able to address tensions at the source of the
conflict – combatants after all are the symptoms or outgrowth of latent tensions, not the root causes 
of conflict. With this bottom-up approach focused on communities as a whole, OTI addressed a
number of root causes of dissension: land conflicts, inequality and exclusion of marginalized groups 
(women in general, female survivors of sexual violence, pygmies, displaced persons), and
exploitative governance and predation by the political class. It is evident in both Burundi and DRC 
54
that the CFR approach offers superior reinsertion, reintegration and reconciliation dividends
compared to traditional DDR.
Content and structure of CBLP vs. YES modules: The three CBLP modules had a short, flexible 
framework, while the YES training with its five (originally six) modules was considerably more
substantive and implied far more and longer commitment. As a result, CBLP was easier to adapt to 
the changing political landscape of the political transition process. Lessons on reconciliation, the
constitutional referendum, and the elections process were examples of curriculum additions made at 
strategic moments during the Burundi program. The six-month cycle of YES training made it less
adaptable to shifts in the political landscape, particularly its potential use in North and South Kivu,
where security restrictions were removed by April 2005. 
Little inter-agency coordination with similar donor programs: In the Burundi programs, CFR
activities were undertaken with little or no involvement with other donors for synergy and multiplier
effects. Perhaps the most glaring example of this is the apparent duplication of OTI and GTZ
vocational schools in Ruyigi province of Burundi. While it is not clear what OTI could have done to 
mitigate this overlap or duplication, this donor coordination issue should be raised for future
projects. On the other hand, there appears to have been considerably more donor collaboration in the 
OTI/DRC program. 
Community-driven programming is hazardous. Balancing top-down or donor-driven projects with
expressed community needs is difficult. Examples of OTI success and failure abound in the Burundi
and Congo programs. In both training programs, for instance, beneficiaries were at first skeptical and 
dismissive of political training. Political and internecine conflict and their resolution later emerged
as the most relevant and appreciated element of capacity-building. In many community initiatives,
requests by community members for large-scale, show pieces, such as the community activity
(peace) centers in Burundi, may have represented their short-term interests, but over the longer term 
the lack of appropriate water, health, and educational infrastructure will be sorely missed. Even in
the community schools it built in Burundi, OTI acceded to student requests for drumming and
dancing materials, when textbooks and small libraries were the real need. 
11. Pertinence to the transition process and responsiveness to critical shifts: The Burundi 
program was well-timed to national political developments, particularly the tiered election process.
OTI/DRC reports that it did shift its programming to accommodate delayed elections and a lagging
DDR program, but this does not stand out as much as in Burundi. The USAID Mission will continue 
supporting the transition, but the Mission Director lamented what he termed an OTI premature
departure while the transition was still under way. Moreover, full OTI/DRC operations did not reach 
the region posing the greatest threat to the transition – North and South Kivu provinces. OTI reports 
a significant follow-on activity for South Kivu, however. 
12. Strategic involvement of the political class: Efforts to integrate the political class into program 
conception, life, and close-out were nominal in both countries. OTI/Burundi obtained consent and
buy-in at national and provincial levels for program design, but authorities were absent from actual
operations. At no level in Burundi were authorities aware of OTI’s impending departure. OTI/DRC
reports that national and provincial administrators were involved in SE*CA and that OTI worked
with both national and district-level political figures when they appeared to be credible actors.
55
OTI/DRC also reports extensive work at the national level with both CONADER and the
Independent Elections Commission. In both programs community project management committees
often included chiefs and notables. In both Burundi and DRC, politicians and administrators
generally behaved like other beneficiaries, rather than as governing authorities.
13. Reconciliation in the absence of a judiciary process: Reconciliation among a variety of
formerly warring or feuding ethnic groups or factions was advanced in both programs through 
capacity building and community projects requiring participation and in-kind contributions from
disparate community members. In Burundi these activities included an appropriate mix of different
ethnic groups, refugees, internally-displaced persons, and community remainees. However, in DRC,
program areas around Kisangani did not require significant reintegration or reconciliation of
community members. In both the Burundi and DRC programs, the process of disarmament and
demobilization generally lagged well behind the community-focused reintegration program, robbing
it of part of its original importance. In Maniema province, however, spontaneous demobilization was 
a key issue, and the OTI program was on target to deal with it. Nevertheless, in spite of community 
statements to the contrary, lasting reconciliation in such contexts is impossible to achieve within the 
course of a two-year program.
14. Making political engagement safe for civilians: In both countries open discussion of politics 
has long been taboo, because of its direct association with violence. In its immediate areas of
operation, OTI programming managed to destigmatize political discussion and demystify the
political process for beneficiaries. Use of the media was also successful in raising national awareness 
of the peace and political transition process, although this was far more successful in the much
smaller Burundian context. Generally speaking, there is no nation-wide impact on these obstacles to 
democracy, and stakeholders in both countries stated that the political transition is far from having
run its course. 
15. Gender involvement and impact: Both programs actively included women with men in their 
activities and had positive, possibly lasting impact on gender equity. Women were not only included 
in large numbers in community capacity-building and grant initiatives, but were also placed in
responsible positions in various management committees. The status of women in communities
certainly improved overall, but especially through their increased visibility and voice in community
affairs and their capacity to solve local conflicts through negotiation and mediation. 
16. Future self-management of community projects: In both the Burundi and DRC programs, 
large community projects, such as schools, information centers, and community activity (peace)
centers that require long-term management by a local steering committee, risk falling prey to internal
divisions. Such large assets, if neglected or abandoned through conflict over their use or control,
would reflect poorly on OTI and USAID. While short-term reconciliation and peace-building 
objectives may have been achieved, these structures could turn counter-productive to U.S. 
government interests. Steering committees and voluntary associations charged with managing these
facilities do not, on the whole, appear capable of ensuring the equitable use and appropriate
maintenance of these structures. 
17. Program design and structure: Overall program design was more complex in Burundi than in
DRC, but the former also suffered from lack of ‘brand’ recognition. This was in part due to the
56
multiplicity of implementing entities in Burundi (OTI, ASI, PADCO, WWCIS) that led to
overlapping, circuitous, and top-heavy management, when a more streamlined approach was in
order. In Burundi, this implementation structure also resulted in a swollen national staff with 
unrealistically high salaries. A more streamlined and less costly national staff would have
substantially reduced program costs. This meant reducing or combining national staff positions and 
salaries, rather than replacing them by even more costly expatriates. However, the absence of
expatriate in-the-field supervision in Burundi may have created inefficiencies and opportunities for
mismanagement, although none were discovered by the evaluation team. OTI/DRC maintained more 
hands-on oversight and control by the main contractor. 
18. Role and importance of the media component: Media projects in both countries were the most 
flexible in responding to shifts in the transition process. In Burundi this was an adept tool, incisive at 
many major junctures in the transition, most notably the constitutional referendum and the
communal and legislative elections. In DRC, the only real critical transition benchmark to which
media seems to have responded was voter registration. While OTI/DRC reports having conducted a 
constitutional referendum sensitization campaign, the evaluation team found that communities were
unaware of the content of the new constitution. However, OTI/DRC did respond to General 
Nkunda’s invasion of Bukavu in 2004 and its violent national fall-out by bringing together
journalists from South Kivu to train them on reporting issues. Additionally, OTI/DRC will be 
present for national elections in April 2006 or post-election developments, and the CONADER
project in Ituri and the IRC work in South Kivu will be on-going. 
19. Relations with USAID: Neither program appears to have liaised sufficiently with its logical
handover partner – USAID/DRC and REDSO Nairobi – in the planning and implementation phases 
of the program. OTI/DRC, however, reports that the Phase 2 program was designed in collaboration 
with the USAID Mission and that the Country Representative coordinated frequently with the
USAID/DG Office and others. Although OTI/DRC began discussing program handover in January
2005, it will only be short-term. OTI/Burundi is now playing catch-up, but previous failure to
maintain closer collaboration is now seriously complicating handover. If handover were not an issue 
for OTI, this lack of planning would be of little concern. 
20. Monitoring and Evaluation: In spite of PMP and indicator development, it does not appear 
that M&E was very effective in either the DRC or the Burundi program. This is especially true of the 
Burundi program, where a set of objectives and indicators were designed early on in the program but 
never effectively monitored or measured. One way to improve monitoring would be to issue semi￾annual progress reports, which would not replace the hot topics or bi-weekly reports, but rather
synthesize the experience in a way that would allow effective higher-level tracking of program
activities. At the very least, it would allow a summing up of outputs on a reasonable schedule. 
In the DRC, M&E was considerably more developed, but an analysis of impact has yet to be
produced. Nor was any substantial mid-term assessment undertaken as in Burundi. A final
evaluation by Chemonics may yield some data on indicators that were tracked from mid-2004. 
However, a nice handbook was prepared by OTI/DRC that pulls together the various M&E materials 
employed under SE*CA.17 These materials include: the Performance Monitoring Plan 2004-2006; a 
baseline questionnaire (a baseline report was also produced in July 2004); a Quick Survey
 
17 OTI/Chemonics International. ND. “Monitoring and Evaluation Booklet for SE*CA.” Kinshasa, DRC. 
57
questionnaire for indicator follow-up; a YES participant survey; a grant proposal evaluation guide;
and a final grant evaluation guide. It seems a good deal was done in attempting to track impact and 
program quality, but the somewhat disparate materials produced as results do not present a very clear 
picture. The evaluation team is unaware of any final report on evolving indicator values and their
analysis. Consequently, impact on communities through the various SE*CA components remains
largely unclear, anecdotal, or subjective. In this vein, a very informative, but rather subjective,
evaluation was the “Internal Evaluation of Grants and YES Training” conducted in Bunia in
September 2005 by the entire SE*CA staff. While the OTI/DRC material is commendable, it needs 
to be pulled together in some kind of final report on the impact of CFR on SE*CA communities. 
Perhaps this is pending. 
58
Annex A: List of Persons Interviewed
OTI / Washington
Michele Amatangelo Program Manager
Bronwyn Bruton Former Program Manager for the DRC
Konrad Huber Africa Team Leader
Galeeb Kachra Former Program Manager for the DRC
Donna Kerner Program Manager for Burundi
Robert Jenkins Deputy Director
John Langlois Media Advisor
Carlisle Levine Former Program Manager for Burundi
John Rigby Senior Field Advisor (Bullpen)
Wendy Henning Program Manager for the DRC
Cyndi Scarlett Former OTI Representative in Burundi 
Mary Stewart Former M&E Specialist
Rick Swanson Program Manager, Outreach and Public Affairs
Patrick Wingate Former Program Manager for the DRC
OTI / DRC
Anaia Bewa Program Manager (Media)
Stacia George OTI Country Representative in the DRC
Clement Kashala Logistics/Administration Manager
Ines Krauth Deputy Country Representative
Alain Mulumba M&E Specialist
Simon Mutala Program Manager
OTI / Burundi
Radegonde Bijeje Supervisor, CBLP Component
Tyrone Gaston OTI Country Representative in Burundi
Alexis Mvuyekure Supervisor, VST Component
Josephine Ndikumama Supervisor, Community Initiatives Component
Justine Nkurunziza Supervisor, Media Component
USAID / DRC
Bob Hellyer Mission Director
Deko Moleka-Bekangba Program Manager, Democracy and Governance
59
USAID REDSO / ESA
Rob Luneberg Representative in Burundi
Laura Pavlovic Program Manager for Burundi
Andy Sisson Director
Chemonics / Washington
Jennifer Boggs Program Manager for Africa
Andrea Dykman Director for Africa
Bob Kramer Director, Conflict Prevention and Recovery
Marilisa Miller Former Director for Africa 
Chemonics / Congo
Jack Asiyo Field Supervisor in Bunia
Marthe Balunda Master Trainer in Kisangani
Honorine Bantikoko Master Trainer in Kindu
Yves Bawa Project Development Officer (grants manager)
Lambert Biyempo Project Assistant for SE*CA
Rhett Gurian Field Supervisor in Kisangani
Gerry Holdrinet Chief of Party for SE*CA
Benoit Kalugurha Master Trainer in Bunia
Marie Kanyobayo Master Trainer in Bunia
Martine Kasongo Program Assistant for Transition Awareness/Participation
Seraphine Kokere Master Trainer in Bunia
Monique Manduara Office Manager for SE*CA in Bunia
Ron Mininger Program Manager, CONADER project (follow-on)
Bruno Longela Master Trainer for SE*CA (Kisangani)
Tom Nasongo YES Component Manager 
Congo (Others)
Soeur Anuarite Diocesan Development Bureau, Isiro
Dieu-donné Mongu SE*CA Local Facilitator in Isiro
PADCO / Washington
Tom Bayer Senior Advisor, Humanitarian Response/Reconstruction
Joy Benn Senior Contracts Administrator
Ted Bratrud Director, Humanitarian Response/Reconstruction
April Peetz Project Manager for Burundi
60
PADCO / Burundi
Vital Bancako Coordinator for BNL/Associations, Gitega
Leanne Bayer Chief of Party
Adolphe Bigirimana Head of Logistics, Ruyigi
Godefroid Bigirimana Administratration/Finances Officer, Gitega 
Henry-Paul Bolap Media Coordinator
Jean-Claude Gahungu Program Officer, Gitega
Cécile Gakima Program Assistant, Gitega
Juan Bosco Hakizimana Program Officer, Ruyigi
Jill Morris Community Initiatives Coordinator
Sebastien Mtirampeba VST Coordinator, Ruyigi Province
Paul Mzozejo Agro-forestry Coordinator, Gitega
Louis Ndababonye Construction Engineer
Roger Ngendabanyikwa Construction Engineer, Gitega
Bonsecuture Niyonizigiye VST Coordinator, Gitega Province
Emmanuel Niyonkuru Assistant Media Coordinator
Bernard Nsabimana BNL Coordinator, Ruyigi
Aline Rivuzimana Program Officer, Ruyigi
Frédéric Rugemintwaza Program Officer, Gitega
Greg Seelhorst VST Coordinator
Elysée Sindayigaya Administration/Finances Officer, Ruyigi
African Strategic Initiative (ASI)
Pierre Bahata Master Trainer, Ruyigi
Guerrier Iryabauyeyi Master Trainer, Ruyigi
Louise Nahayo Master Trainer, Gitega
Jimmy Naneceiw CBLP Coordinator, Gitega Province
Egide Nderagakura Master Trainer, Ruyigi
Alice Nduwimana Master Trainer, Ruyigi
Etionnette Nshimirimana Master Trainer, Ruyigi
Lydia Nyahoza Master Trainer (VST), Ruyigi
Donatien Nzokira Master Trainer, Gitega
Thierry Rwagatore Executive Director
U.S. Embassy Burundi
Ann Breiter Deputy Chief of Mission
61
Burundi (Others)
Déo Barutwanayo Head of M&E, Ruyigi Provincial Bureau, CNDDR
Moise Bukumi Governor of Ruyigi Province
Philibert Buturo Manager, Ruyigi Provincial Bureau, CNDRR
Godefroid Hakizimana Former Minister, Ministry of Handicrafts
Léonidas Hakizimana Director General, Radio Television Nationale Burundaise 
Denis Hynes Resident Representative for Africare
Steve McDonald Woodrow Wilson International Center for Scholars
Michelle Mummy Country Program Manager for IFES
Onésime Nduwimana Former Minister of Communications
Corneille Nibaruta Director, RSF Bonesha
Innocent Nsabimana Supervisor of Media Monitoring Center, OMAC
Joseph Ntakirutimana Minister, Ministry of Good Governance
Mossi Sélémani Governor of Gitega Province
62
Annex B: Selected Bibliography
Adelski, Elizabeth and St. George, E. 2003. “Findings and Recommendations for OTI’s 
Community-based Reintegration and Non-formal Education Program in Burundi.” Creative
Associates International, Inc. March 31, 2003. Washington, D.C. 
Briggs, Jeanne and J. Rigby. 2005. “OTI Burundi Mid-term Assessment Report.” February 9,
2005. Washington, D.C. 
Brewer, Sara and K. Huber, R. Swanson. 2005. “Memorandum Re DRC Program Performance Review.” 
May 4, 2005. USAID/OTI. Washington, D.C. 
Chemonics International. 2004. “Chemonics International/OTI Design Report: Community Focused 
Reintegration Program for the Democratic Republic of Congo. May 31, 2004. Washington, D.C.
Chemonics International. 2004. “Baseline Report.” Synergie d’Education Communautaire et
d’Appui a la Transition (SE*CA). July 2004. Kinshasa, DR Congo.
Chemonics International. 2005. “SE*CA Annual Report.” March 2005. Chemonics International. 
Washington, D.C. 
Chemonics International. 2005. “Monitoring and Evaluation Booklet for SE*CA.” Kinshasa, DRC. 
Cobham, Kury. 2004. “OTI/DRC Response to Changing Political Environment.” Internal Memo. 
June 29, 2004. OTI/DR Congo. Kinshasa, DR Congo. 
Flood, Holly and S. George, G. Kachra. 2005. “Program Performance Review: OTI – Democratic 
Republic of Congo (DRC).” April 11, 2005. Washington, D.C. 
Langlois, John. 2004. “National and Regional Media Activity Plan.” Congo TDY: October 6 to 
16, 2004. October 2004. Washington, D.C. 
McClintock, Elizabeth et al. 2004. “Community-Based Leadership Program CBLP au Burundi.” 
Guide de formation. Woodrow Wilson International Center for Scholars and CM Partners. 
April/May 2004. Bujumbura, Burundi. 
Mulumba, Alain. 2005. “Internal Evaluation of Grants and YES Training Conducted by the Entire 
SE*CA Staff – Bunia - Week of September 22-24, 2005. October 2005. OTI/DR Congo. Kinshasa, 
DR Congo. 
OTI/Burundi et al. 2004. “Baseline Assessment: Gitega and Ruyigi Provinces.” USAID/OTI: 
PADCO, ASI, WWICS. June 2004. Bujumbura, Burundi. 
OTI/DRC. 2005. “DRC Performance Monitoring Plan: 2004-2006.” October 5, 2005. OTI/DRC. 
Kinshasa, DR Congo. 
63
OTI/DRC. 2005. “Internal Evaluation of Grants and YES Training Conducted by the Entire SE*CA 
Staff: Bunia – Week of September 22-24.” OTI/DRC. Kinshasa, DRC. 
OTI/Washington. 2003. “A New Program for USAID/OTI in Burundi.” December 2, 2003. 
Washington, D.C. 
OTI/Washington. 2005. “Burundi Strategic Plan Revised.” January 11, 2005. Washington, D.C. 
OTI/Washington. 2005. “OTI Burundi’s CPRI Monitoring and Evaluation System. April 27, 2005. 
Washington, D.C. 
OTI/Washington. 2005. “Prioritizing Community-based Capacity-Building in Post-War 
Reintegration Efforts: The OTI Experience.” September 18, 2005. Washington, D.C. 
OTI/Washington. 2005. “CPRI Feedback for USAID Annual Report 2005.” October 2005. 
Washington, D.C. 
Martin, Angela. 2005. “DRC Program Performance Review.” Memorandum. July 8, 2005. 
USAID/OTI. Washington, D.C. 
PADCO. 2004. “Media Baseline Assessment.” Bujumbura, Burundi. 
Sommers, Marc. 2005. “Field Evaluation Mission to Burundi.” Mission #4: July-August 2005.
Woodrow Wilson International Center for Scholars (WWICS). September 12, 2005. Washington, 
D.C. 
Stone, Richard. 2005. “Mystery Man of Stonehenge.” Smithsonian. Volume 36, Number 5. 
August 2005. Smithsonian Institution. Washington, D.C. 
USAID/Washington. 2004. “USAID-Burundi Stock-Taking.” USAID/AFR and USAID/DCHA. 
December 2004. Washington, D.C. 
Uvin, Peter and S. Campbell. 2004. “The Burundi Leadership Training Program: A Prospective 
Assessment.” Woodrow Wilson International Center for Scholars. July 23, 2004. Bujumbura, 
Burundi. 
Wolpe, Howard et al. 2004. “Rebuilding Peace and State Capacity in War-torn Burundi.” 
Woodrow Wilson International Center for Scholars. March 1, 2004. Washington, D.C. 
64
Annex C: Evaluation Scope of Work
Introduction
In early 2004, USAID’s Office of Transition Initiatives (OTI) launched programs in Burundi, the
Democratic Republic of Congo (DRC) and Liberia, with each addressing the post-conflict challenge 
of integrating ex-combatants into civilian life. For each of these reintegration programs, OTI chose a 
community-based, capacity-building approach that built on lessons learned from OTI’s earlier
program in Sierra Leone, the Youth Reintegration Training and Education for Peace Program
(YRTEP). The OTI/DRC program is scheduled to end in December 2005, and the OTI/Burundi
program is scheduled to end in February 2006. The OTI/Liberia program is currently planned to run 
until 2007. Due to the fact that both DRC and Burundi are closing within a few months of each
other, and the potential for leveraging lessons learned across two programs that have similar goals
but have arrived at them through different approaches, OTI believes it would be useful to field one 
evaluation team that will examine both programs.
Background
Congo
Following the conclusion of the Pretoria peace agreement in July 2002 and the growing momentum
for peace leading to the Sun City accords of 2003, OTI renewed its engagement in the DRC in April 
2002 through the Congo in Action for Peace (CAP) program. In early 2004, the program was re￾tooled with a new implementing partner, Chemonics International Inc., as SE*CA (Synergie 
d’Education Communautaire et d’Appui a la Transition). The program aims to i) respond to the need 
to re-integrate ex-combatants, specifically war-affected youth and victims of rape, and ii) support the 
advancement of the overall transition through the strengthening of awareness and participation
around key transition issues. 
Although the program has been fully operational for only one year and without the benefit of a
programmatic evaluation, it is largely believed that SE*CA is achieving these two goals. The
program consists of three main components (1) basic life skills and vocational training (Youth
Education and Skills/YES) provided by Master Trainers; (2) community-driven small grants to
participating communities, providing a chance to put learned skills into practice; and (3) Transition
Awareness and Participation (TAP) grants intended to reinforce YES messages through media￾focused activities such as information campaigns.
Burundi
The Community-based Peace and Reconciliation Initiative (CPRI) was launched by OTI in February
2004 to strengthen local capacities to benefit from and contribute to the peace process. PADCO, 
Inc., the Woodrow Wilson International Center for Scholars (WWICS) and the national NGO
African Strategic Impact (ASI) implement CPRI and manage field offices in the provinces of Gitega 
and Ruyigi. Through CPRI’s Community-based Leadership Program (CBLP), vocational skills
training (VST), community initiatives grants, and media support activities, OTI and its three
implementing partners encourage and enable: (1) local-level cooperation for mutual problem
solving, especially related to reconciliation in communities during the period of reintegration of
65
refugees, internally displaced persons, and demobilized combatants; (2) provision of opportunities
for generation of new non-farm income; and (3) the dissemination of timely and balanced
information that encourages broad participation in discussions related to the current peace and
reconciliation process in Burundi.
Purpose
As OTI prepares to handover (or end) program activities in Burundi and DRC to the respective
USAID Missions and other donors, we would like to explore through an independent evaluation the 
following questions:
ß Did OTI contribute to the advancement of transition processes in Burundi and DRC? If so, 
how and in what ways could it have improved this?
ß Were the programs strategic in responding to shifts in the transition processes?
ß Did the programs meet their stated goals and objectives?
Objectives
The objective of this task order is to provide technical assistance to OTI for the following:
1. To recruit and field a 2-person final evaluation team to DRC and Burundi;
2. To evaluate the performance and impact of the OTI/DRC and OTI/Burundi programs; 
3. To determine, together with OTI, and appropriate methodology for the evaluation;
4. To document, in final evaluation reports, findings, conclusions and lessons learned from the
programs, as well as recommendations for the future; 
5. To provide out-briefings in DRC and Burundi and an official presentation in Washington on
the above. 
A substantial amount of information and documentation exists on both programs. Therefore, the
requirement for fieldwork is not extensive. OTI would like the evaluation team to conduct a desk
study of existing documentation on each program, and then conduct a field mission that will include 
semi-structured interviews with key stakeholders in Kinshasa and Bujumbura, and a limited number
of site visits.
Team Composition
1. One senior level evaluator with extensive experience designing and conducting evaluations. 
The senior level evaluator will serve as team leader and be responsible for the draft and final
evaluation reports and for debriefs in DRC, Burundi, and Washington, DC.
2. One senior or mid-level evaluator with experience designing and conducting evaluations. 
The second evaluator will focus on the desk study component of the evaluation, and also
conduct field work.
At least one of the two evaluators should have experience in DRC and Burundi and be fluent in
French. OTI would also like to provide two OTI staff members as resources on the evaluation team: 
one senior field advisor, and one local staff member from another OTI program in Africa.
66
Deliverables
The deliverables expected from the technical assistance provided are the following:
1. Brief outline of methodological approach for assessments, including proposed itinerary for
field work and identification of all logistical support needs;
2. Draft evaluation reports, not to exceed 20 pages per country program, plus additional
annexes;
3. Brief Power Point presentations summarizing conclusions and recommendations, deliverable
at the same time as the draft report;
4. Final evaluation reports, deliverable no later than two weeks after receipt of all comments on
first draft.
Proposed Time frame
November – December 2005
Level of Effort
Desk review of program documents_______________ _______________ 10 days
Phone interviews__________________________________________________3 days
Preparation of proposed methodology and detailed schedule of field work 3 days
Field work and outbriefing (DRC, Burundi)___________ ___ 40 days (20 each)
Preparation of draft reports _____ 10 days
Presentation of findings, recommendations and discussion of draft (DC)_____ _2 days
Preparation of final report s 10 days
Note on logistics: Depending on when the evaluation begins, the order of travel for field work
will likely be to Congo and then Burundi. Contingent on field specifications, a possible itinerary
is Kinshasa – Kisangani – Bunia – Kampala – Bujumbura. Internal travel within Congo and
Burundi may be arranged through OTI and implementing partners using UN or Airserv flights. 
Other logistical support may be provided by OTI and partners as feasible. Costs for OTI
personnel accompanying the team will be covered by OTI.