The USIP Learning Agenda: An Evidence Review

Preventing and Countering

Violent Extremism:
Assessing Missteps
and Promising
Community Approaches

Lauren Van Metre

Director, Peace, Climate and Demo­cratic Resilience, National Demo­cratic Institute

Thomas Leo Scherer

Research Director, FP21

march 2023
Preventing and Countering Violent Extremism

The United States Institute of Peace seeks to advance the field of peacebuilding by evaluating
the evidence base supporting its core practices, such as dialogue and conflict analy­sis, engage-
ment with religious leaders, and the prevention and countering of violent extremism. ­These
systematic reviews identify effective programming and new approaches for further explora-
tion. This evidence review paper evaluates the evidence and practice of an evolving approach
to preventing and countering violent extremism: understanding and strengthening community
resilience.
Preventing and countering violent extremism (P/CVE) is unique among peacebuilding ar-
eas. The field was initially ­shaped and influenced by a frenzied national security response to a
perceived imminent threat from a global religious radical movement that sought the destruc-
tion of the West and its secular governments. Thus, the prob­lem of violent extremism and its
countering strategy ­were neatly encapsulated in an ideological paradigm that facilitated crisis
decision-­making rather than purposeful action in support of an evidence-­based policy and
practice. ­Today, in promoting a community resilience approach to P/CVE, it is critical to steer
away from ­earlier ideologically influenced forms of community engagement by acknowledging
that ideological remnants persist and continue to do harm to frontline communities; ­these
forms of community engagement scapegoat communities for attracting violent-extremist net-
works and target them as “threats” for security force responses. Instead, the P/CVE field needs
to adopt a radically dif­fer­ent resilience approach that presumes and strengthens a communi-
ty’s capacity to resist violent extremism.

Research Questions and Methodology

Violent extremism generally spreads through localized conflict in which extremist groups ma-
nipulate community, group, or individual grievances to gain position and traction. Thus, P/CVE
­will depend on how communities redeploy core capacities (information gathering, communica-
tion, leadership) to shape community responses to meet the threat—­responses that require
significant levels of local trust and community cohesion. In thinking about how to assess and
support community resilience—­including how to characterize and evaluate resilience capacity—
it is critical to clearly define what resilience to violent extremism is and how it is mea­sured.
Resilience is defined quite differently by dif­fer­ent fields. In the field of engineering, resil-
ience means the capacity to maintain the status quo or return to an original state ­after experi-
encing a shock. Ecologists study the resilience of complex, adaptive systems, and it is this

2 | Preventing and Countering Violent Extremism | USIP.ORG

conceptualization that has the most promise in its applicability to social systems and vio­lence.
In the context of vio­lence, resilience is the ability of a community, ­people, state, or region to
prevent, mitigate, or recover from vio­lence by adopting new pro­cesses, norms, and strategies
for conducting their lives and adapting societal relationships in response to a violent shock or
an uptick in aggression.
Violent extremism is dif­fer­ent from other forms of vio­lence: it is, at its core, an ideologi-
cal strug­gle, unlike ethnic, secessionist, or criminal extremism, which involve contestations
over land, resources, or access to po­liti­cal power. An ideology is an all-­encompassing world-
view that presupposes its own po­liti­cal and social truth. An ideological extremist worldview
cannot coexist with another ideologically based system of belief, nor can it accept or show
tolerance ­toward its adherents. It seeks to eradicate the opposing system, and its supporters,
through coercion and vio­lence.
A community’s resilience to violent extremism resides in its capacity to ­counter ideologi-
cally driven groups that seek dominion over ­people and power. Jihadist extremism is rooted in
a profound fear of erasure among ­people of Muslim faith, orchestrated by a liberal or secular
order that has persecuted and oppressed them; examples include Ataturk’s outlawing of Sufi
­orders in Turkey and Egypt’s outlawing of the Muslim Brotherhood. The goals and tactics of
the Islamic State of Iraq and Syria, al-­Qaeda, and Boko Haram might differ: they might collabo-
rate with local groups or demand their fealty, or they might establish a religious kingdom on
Earth or seek it in the afterlife. But each group sees Islam as locked in an existential strug­gle
with secular states that employ po­liti­cal and systemic vio­lence, in par­tic­u­lar, against followers
of Islam. The grievances espoused by violent-­extremist groups regarding corrupt, venal state
actors and institutions and their persecution can be legitimate. Community resilience to vio-
lent extremism depends, in many cases, as much on the ability of local leaders to manage and
counteract the effects of state predation—or accounting for state weaknesses in the local de-
livery of security, rule of law, and services—as on directly countering or preventing violent-­
extremist operations within the community. In other words, the ability of a community to
acknowledge and respond to citizens’ legitimately held grievances regarding poor, weak, ex-
clusionary, or predatory governance appears central to preventing and countering the pull of
violent-­extremist ideology.
This evidence review paper pre­sents the findings of an evidence review that aimed to

• evaluate the evidence base for identifying and assessing community resilience capacities
in the context of violent extremism;
• evaluate the evidence base regarding the effectiveness of programs that are designed to
specifically support and strengthen community resilience;
• examine the evolution of core fields of P/CVE practice—­youth, gender, and religion—to
determine what, if anything, they say about community resilience;

USIP.ORG | Preventing and Countering Violent Extremism | 3

 • explore how current programs and interventions that seek to ­counter terrorism and vio-
lent extremism may, in fact, be reinforcing the “dark” social capital (the systems of
­vio­lence and discrimination) that reinforces the d
­ rivers of violent extremism; and
• review two areas of peacebuilding practice—local peace committees (LPCs) and hybrid
governance—­and how they might contribute to community resilience to violent extremism.

T­ hese dif­fer­ent lines of inquiry invited two fundamental research questions: What are
the ­factors that underlie community resilience to violent extremism, and what is the evidence
base that supports them? How strong is the evidence base for community resilience practices
that address violent-­extremist threats, and what does that mean for implementing community
resilience programming in P/CVE policy and practice?
The mixed-­method evidence review included the following, in the order given:

• a quantitative analy­sis of the quality of evidence on community resilience programming

and research that has identified actionable resilience ­factors, which nonetheless could be
improved in terms of data transparency and reproducibility;
• key in­for­mant interviews with P/CVE community leaders on youth, gender, and religious
engagement, who outline needed evolutions for their sectors and their relationship to
community resilience;
• a critical review of the lit­er­a­ture on community resilience to violent extremism that ex-
plores the types of social cohesion that are critical for preventing or resisting vio­lence;
how and what types of bad governance weaken resilience to vio­lence; and the systems of
community dark capital (existing systems or patterns of vio­lence) that are the vectors
through which extremist groups enter a community; and
• two qualitative, evaluative case studies that assess the effectiveness and nature of two
peacebuilding practices that may have promise for strengthening community resilience
to extremist vio­lence: LPCs and hybrid governance.

This study makes key recommendations on how the policy, programming, and research
communities can validate and strengthen community resilience approaches in the prevention
and countering of violent extremism.

Community Resilience: A Policy and Programming Plea

Over recent de­cades, research has widened the lens for understanding the root ­causes of
­violent extremism, reframing its threat as an interplay of complex dynamics rather than a
­composite list of risk ­factors, such as youth unemployment or po­liti­cal exclusion. A number of
seminal studies and policy and program reviews have moved the field beyond the radicaliza-
tion of deviant youth as a key ­factor for the spread of extremist groups and vio­lence. For

4 | Preventing and Countering Violent Extremism | USIP.ORG

example, research examining the way socioecological ­factors contribute to youth vio­lence
has pinpointed state-­sponsored discriminatory vio­lence, in par­tic­u­lar against young men, as a
strong determinant of involvement with violent-­extremist groups.1 In more traditional socie­
ties, youth are excluded from community leadership roles and the ability to achieve status as
adults in many ways. Research shows, for example, that youth cannot afford the bride-­price
for wives, though marriage and ­children are the path to leadership in their communities; that
age brackets for youth are extended even beyond the age of forty, consigning the very young
to de­cades of disregard; and that traditional leaders reestablish ever-­higher barriers to adult-
hood in rewriting the rules of the power game.2
A seminal study by Mercy Corps has established that lack of status in the community, not
lack of employment, is the stronger determining ­factor for youth affinity with violent-­extremist
groups, underscoring the potency of social-­political barriers and discrimination as a ­factor in
youth vulnerability. In many re­spects, when youth do radicalize, their motivation is rooted in a
universal desire for community belonging and meaning that has been systematically, unfairly,
or brutally suppressed.3 More attention and research are needed on the community socio­
ecological ­factors creating risk among youth, particularly governance strategies (both informal
and formal) and policies that systematically discriminate against them. The perpetrators, pat-
terns, and systems that oppress youth need further study to understand their impact in creat-
ing youth vulnerability to violent extremism. Is it the direct experience of state-­perpetrated
vio­lence by an individual or peer? Is it a young person’s immediate and shocking loss of com-
munity status (loss of land, access to employment) caused by power­ful po­liti­cal forces that
leave that person ­little recourse? Does prolonged, systematic discrimination against young
persons based on their ethnicity or religion and punctuated by a formative event mobilize a
peer group, driving them to join violent-­extremist groups?
It is critical that any community resilience approach to P/CVE not return to a failed school
of thought on youth radicalization—­one that associates their deviance with the qualities and
characteristics of certain groups and communities. Instead, it must focus on the deeper, more
hidden issues, enumerated above, at the intersection of youth and community resilience.
What social and governance relationships and policies at play in communities have been able
to protect their youth from violent-­extremist recruitment, and how can they be replicated?
Perhaps an even more pertinent reframing would explore how and why youth overwhelmingly
choose the path of peace and nonviolence in the face of the broad-­based, systematic discrimi-
nation they face in many countries and locations around the world. What explains youth resil-
ience, and how can it be reinforced and replicated in the communities in which they live?
In similar ways, research has shifted the ­earlier ideologically driven perspective of com-
munities as fertile grounds of radical recruitment. The new framing of communities as effec-
tive first-­line responders (that is, more effective than top-­down, state interventions) was
reflected in the United States’ adoption of community-­oriented approaches to countering ex-
tremist recruitment and vio­lence in 2015.4 Early socioecological research on youth protective

USIP.ORG | Preventing and Countering Violent Extremism | 5

 and vulnerability ­factors considered broader community patterns, although the research often
focused exclusively on collective deficits rather than group resiliencies and strengths. While
­these early studies moved the field beyond a focus on individual deviance, they did feed into
the “suspect community” construct by placing the responsibility for youth vulnerability to rad-
icalization squarely on the communities in which they lived.5 Other research critiqued state-­
based interventions in Muslim communities, calling out the lack of evidence ­behind ­these
policies while documenting the negative effects.6 This second generation of community-­based
research often applied frameworks for evaluating community capacity—­evolutionary ­because
they moved beyond the exclusive focus on community weakness, although the analytical
framing was often grounded in Western po­liti­cal theory. (That is, ­whether the framing was
applicable to immigrant communities or developing and fragile states was never tested or
validated.)
The most promising research, however, has explored community agency and re­sis­tance
to violent actors through rigorous comparative and inductive research methods. Ashutosh
Varshney, Ami Carpenter, Oliver Kaplan, the Berghof Foundation, and Lauren Van Metre
compared communities that tipped into vio­lence with ­those that resisted.7 Their research
established emergent qualities and patterns in nonviolent communities that did not exist in
communities that engaged in vio­lence. This research has had a formidable impact by acknowl-
edging that many frontline communities have developed strategies and capacities to resist vio-
lent actors nonviolently. ­These studies not only validate community agency but also establish
community authority on nonviolent resistance—­a way of thinking that aligns with a major shift
in the peacebuilding field; the locally led peacebuilding by Pamina Firchow and Roger Mc-
Ginty’s Everyday Peace Indicators; Severine Autesserre’s Peaceland; and both Peace Direct and
Conciliation Resources, which support local communities at the forefront of peacebuilding
work through small grants and participatory program design.
This paper seeks to advance the field of P/CVE by calling for a focus on building the resil-
ience of frontline communities to prevent their operational, po­liti­cal, and economic capture
by violent-­extremist groups; the establishment of recruitment to violent extremism and the
supporting financial networks and safe havens in communities; and the weakening of commu-
nity resilience sources and capacities by corrupt, predatory state actors and international
interventions that reinforce community dark capital (that is, the systems of vio­lence that feed
the grievances driving community and youth support for violent-­extremist governance, opera-
tions, and recruitment).
­These systems of dark capital are often strengthened and enabled by international
counterterrorism policies, leading to a pernicious feedback loop in which counterterrorism
operations reinforce and strengthen the very local grievances that violent-­extremist groups
exploit, as well as undermine the community systems of resilience. Thus, international coun-
terterrorism strategies and operations hit communities from both sides, increasing risk and
tearing apart their resiliencies to violent extremism. We do not purport to advance community

6 | Preventing and Countering Violent Extremism | USIP.ORG

resilience as a developed P/CVE approach. To do so would require a concerted effort to direct
research, evaluation, practice, and policy reviews to answer critical questions on the viability
of community resilience as an approach, such as the following:

• Are community resilience capacities similar across cultural and conflict contexts and, if
understood and validated, could this lead to an evidence-­based prevention practice? Or is
resilience highly localized and contextualized, requiring the rapid deployment or expan-
sion of research and assessment teams to assess community resilience—­where it exists,
how it is being degraded, and how it should be strengthened in the light of increased risk
or presence of violent extremism?
• Can international actors build, support, or strengthen community resilience capacities
without ­doing harm or delegitimizing them? If so, how?
• What strategies have communities ­adopted that degrade their resilience to violent-­
extremist groups? For example, community-­based armed groups, ethnic militias, and so
on? Erosion of local conflict resolution, balancing mechanisms?
• How can government presence in and support to frontline communities be managed to
enhance resilience and reduce risk? Can this be accomplished by the international com-
munity? By strengthening demo­cratic norms, pro­cesses, and institutions? And if so, how?
Through security sector reform? And what types?
• An impor­tant aspect of community resilience is social capital, but how does that capital
need to be configured (interlocking networks, bridging, bonding) or enacted (trust, col-
laboration, collective action, coordination) to resist violent extremism most effectively?

Fi­nally, while we assert that community resilience is a promising P/CVE approach that
should be methodically and systematically explored, the effort must be sustained through in-
ternational and national commitments to frontline communities and a willingness to support
and advocate their agency, expertise, and voice in the development of re­sis­tance strategies. It
must also recognize and reject previous ideologically motivated approaches to community en-
gagement that stigmatized communities as suspect—as hotbeds of youth religious radicaliza-
tion that need to be targeted for social reprogramming and security interventions. This policy
and the programs that supported it have been debunked by research and evidence, as noted
­earlier in this paper. However, community stigmatization still drives many state-­led P/CVE strat-
egies in countries such as Burkina Faso, Mali, the Philippines, and Mozambique and drives
­these communities to embrace violent-­extremist groups. To reinforce the new framing—­that
communities have critical capacities, networks, and strategies with which to manage and pre-
vent vio­lence—­this paper examines two areas of community, bottom-up practice: LPCs and
hybrid governance structures and actors. It also describes how, when successful, the practices
prevent vio­lence and violent extremism by shoring up the systems of social capital that are the
foundation of community resilience.

USIP.ORG | Preventing and Countering Violent Extremism | 7

Assessing the Evidence

This mixed-­method review of the quality of evidence on community resilience—­its ­factors and
capacities and its supporting programming—­seeks to quantitatively assess the research base
and, in the pro­cess, offer academics, policymakers, and implementers a framework for under-
standing and evaluating existing evidence and gaps. To determine the state of P/CVE practice
on other core lines of effort—­youth, gender, and religion—we staged in­for­mant interviews of
leading experts who have worked at the intersection of policy, practice, and research. The pur-
pose of ­these interviews was to assess the effectiveness of their fields in mitigating the spread
of violent extremism in comparison to community resilience approaches and to determine
­whether and how ­these fields might feed into resilience practice. The other pieces of this re-
search methodology included an in-­depth review of the lit­er­a­ture for what it says about resil-
ience and effective programming (articles that ­were coded and assessed in the quantitative
review on research quality) and two case studies on promising practices in fostering community
resilience to violent extremism—LPCs and hybrid governance—­where an effort was made to in-
clude the work of local researchers and evaluators.

REVIEW OF EVIDENCE QUALITY

We identified relevant documents through keyword searches and a review of recent docu-
ments from relevant forums.8 We especially sought out documents that ­were highlighted in
the field for their quality. This collection is not a systematic or representative sample of P/CVE
community resilience documents; instead, we focused on collecting the highest quality and
most-­often cited documents to highlight exemplary work and identify weaknesses in even the
most informative work. For primary documents (research articles and program evaluations),
we coded multiple indicators of quality that ­were informed by other evidence assessments
and our own experience.9 The main indicators are summarized in appendix A.10
This endeavor was ­limited in some impor­tant ways. First, the scoring criteria ­were not
completely objective. Second, most documents ­were reviewed by only one person, so scores
may vary by each person’s interpretation of the code­book. Third, as the P/CVE community is
small and connected, bias may have affected the scoring of some documents. We make any
personal connections explicit for the documents highlighted within this paper. Fourth, we
scored only program evaluations and research articles; ­after early attempts, we de­cided that
primary, secondary, and conceptual pieces serve dif­fer­ent purposes and would be difficult to
compare. We also found that resource guides and toolkits ­were especially difficult to evaluate,
as they rarely cited sources to support their recommendations.
We examined thirty-­two documents on community resilience and evaluated the evi-
dence quality of the twenty-­four research articles and program evaluations. (See appendix B

8 | Preventing and Countering Violent Extremism | USIP.ORG

­Figure 1. Ratings of P/CVE Documents on Community Resilience across Quality Dimensions

Number of Documents 20

15

10

0
Internal
Validity
Explanatory

Explanatory
Actionability

Discussion

Discussion
Transparency

Reproducibility
Explanatory

Considerations
Research
Relevance
Explained

Program
Proximity
Variable

Variable
Variable

Power

Ethical
Data

Indicator of Quality

Score NA 1 2 3

for the list of research studies.) Their scores are shown in figure 1. We found that low scores
­were actually rare in some topics, especially the actionability and proximity of the explanatory
variable and the explained variable relevance, and in program and research discussion. We also
found that some types of analy­sis in community resilience research often ­adopted methods that
did not allow for mea­sures of internal validity and explanatory power. Thus, we found that ­there
is still significant room for improvement in terms of data transparency and reproducibility.
We highlight Resisting War: How Communities Protect Themselves by Oliver Kaplan for
its substantive nature and for his methods and transparency.11 Kaplan’s proj­ect combines
case studies of five towns and interviews with multiple sources of observational data. He de-
scribes the methodology in detail, including transparency in the case se­lection and how he
deals with ambiguity and contrasting accounts. While he does not make all of his data pub-
licly available, he provides multiple views of the data. He also includes out-­of-­sample cases,
which ­were not used in the formation of the theories of community re­sis­tance as a test of
­those theories.
As part of this evidence review paper, we conducted interviews with community-­of-­practice
experts (in other words, experts who had conducted major meta-­evaluations of their field;
managed communities of practice made up of researchers, prac­ti­tion­ers, or policymakers; or

USIP.ORG | Preventing and Countering Violent Extremism | 9

 spoke prominently about the state of their field based on their own extensive research, prac-
tice, and policy experience). (See appendix C.) ­These key in­for­mant interviews indicate that
core areas of P/CVE practice are evolving, or need to evolve, in ways that support whole-­of-­
community, or community resilience, approaches. Youth, gender, and religious P/CVE pro-
grams could thereby reinforce and strengthen the resilience of frontline and at-­risk communities
in new and innovative ways.

Youth
A major effort of the US government, which has had promising results for youth P/CVE pro-
gramming, is a recently completed comprehensive evidence review on positive youth develop-
ment funded by the National Resource Council and conducted by the Interagency Working
Group on Youth Programs.12 Positive youth development is rooted in prevention and recog-
nizes that previous efforts to address youth prob­lems, such as delinquency, self-­harm, and
radicalization, ­were single-­issue and risk-­focused endeavors. It is a more comprehensive ap-
proach that emphasizes youth resiliency—­how protective ­factors in a youth’s environment
and community can be strengthened to prevent young ­people from falling into adversity. The
recently conducted evidence review identified multiple domains of youth programming that
enhanced their positive development: strengthening youth assets and their agency through
skills-­building; creating opportunities for youth to contribute to their communities and engage
with leaders and decision makers; and supporting a healthy enabling environment for youth
development, including positive relationships with adults and peers, safe spaces tailored for
youth, a sense of belonging, exposure to positive norms and expectations, and youth-­friendly
ser­vices.13
The applicability of positive youth development to P/CVE and the issue of youth radical-
ization was examined in the study through interviews with young ­people who had joined Boko
Haram. In many ways, positive youth development is a pro­cess of reradicalization: redirecting
youth agency and instilling the desire for status and belonging ­toward positive outcomes by
strengthening their skills and their networks of support. The researchers acknowledged that
not enough is known or documented regarding effective positive youth programming for the
most marginalized—an area of pos­si­ble exploration for the P/CVE field and communities en-
gaged in positive youth development. In addition, more research and evaluation need to be
done by sector. For example, we know that youth employed in the agricultural sector in the
Sahel may be exposed to violent conflict, food insecurity, and climate shocks. What does
the programming of positive youth development look like for youth in this sector, in compari-
son with that of young urban entrepreneurs? This includes determining how positive youth
development approaches would work in highly securitized communities, where governments
are not necessarily constructive partners and ­little trust exists between youth and po­liti­cal and
security actors.

10 | Preventing and Countering Violent Extremism | USIP.ORG

Gender
To advance the field of gender in relation to the prevention of violent extremism requires the
policy community to look beyond ste­reo­typical ideas of ­women’s roles and its reflexive nod to
gender inclusion. Terrorist groups incorporate and elevate gendered roles as part of their re-
cruitment and retention efforts. The ways that ­these gendered roles are manipulated, abused,
and implemented are critical to terrorists’ orga­nizational structure, their po­liti­cal and social
worldviews, and their operational and strategic effectiveness. Yet gender roles are vastly un-
derstudied, given their importance: how terrorist groups structure concepts of masculinity
and femininity among members is crucial to their legitimation of vio­lence and how they main-
tain social cohesion.
Understanding intersectionality—­the way gendered norms dictated by terrorist world-
views traverse socioeconomic, ethnic, and religious identities—is also critical. To break ­free
of current P/CVE programming that ste­reo­types ­women and their gendered roles also re-
quires a much more sophisticated understanding of the roles ­women play in terrorist groups
and the roles they assume when their communities are captured by extremist groups. Much
of the groundbreaking thinking on ­women and combatant groups suggests that ­women play
power­ful noncombatant roles in affirming male participation, recruiting, and so on. Too
much analy­sis of groups such as the Islamic State has focused on ­women as abused, as vic-
tims. ­There is not much understanding of how ­women respond when confronted by violent
actors: Do they resist in small, everyday acts? In community collective action? The work of
Julia Zulver on high-­risk feminism would be highly instructive to the field of P/CVE; it resists
the binary view that ­women are ­either peacebuilders or victims to examine how ­women
navigate environments of systematic vio­lence with a strategic mix of collaboration and
re­sis­tance.
This nuanced perspective of gendered identities and actions is critical for understanding
terrorist operations and how communities respond to them in identity-­ and gender-­based
ways. It is also critical for programming and policy success in the response to violent extrem-
ism, for how extremist networks and operations can be resisted, how ­women and men are
reintegrated into communities, and how safe spaces can be created for recovery of individual
genders in communities that have been occupied or targeted by extremist groups. This means
abandoning the current generic approaches to gender and P/CVE for more localized, nuanced
interventions that reflect the ways in which ­women exercise influence differently in diverse
community settings and reveal variations in ­women’s risk to and participation in violent-­
extremist groups. While community responses and violent-­extremist roles may be highly gen-
dered, individual responses to vio­lence or the threat of vio­lence are a complex calculation of
risk, survival, principled response, and personal interest.
Transitional justice programs in Colombia have introduced the idea of complex victim-
hood: that victims have been perpetrators and collaborators and perpetrators have been

USIP.ORG | Preventing and Countering Violent Extremism | 11

 victimized. To deny recognition and compensation of complex victims is to exclude large
segments of populations caught in violent conflict—­and often results in their failure to tran-
sition to peace and nonviolence. ­Women and men, particularly members of the lesbian,
gay, bisexual, transsexual, queer/questioning, and intersex community, experience violent
extremism in highly gendered ways; however, their choices within the constraints and op-
portunities associated with gender are complex. Moving beyond binary perceptions of men
(as perpetrators, victims), w­ omen (as peacebuilders, victims), and other sexual orientations
(as demonization, marginalization) is imperative to preventing recruitment, gender-­based
vio­lence, acts of sexual hate, harassment, and repression and to more effectively reinte-
grate members of violent-­extremist groups back into their communities. This would mean
adopting mea­sures that safeguard against the repetition of the violent events suffered,
which means recognizing the victimhood of collaborators and some perpetrators of violent
extremism and tackling the gendered ste­reo­types that may prevent their full integration or
right to partial or full reparation in order to prevent the social risk of their continued cleav-
ing to vio­lence and violent groups.

Religion
The religious engagement advocated by the peacebuilding field rests on an understanding
of the multiple roles religious leaders play in their communities. Religious leaders are educa-
tors, counselors, judges, governance actors, community leaders, and role models. Program-
ming that engages religious leaders in ­these diverse capacities advances peace and nonviolence
along multiple lines within a community. For example, in Pakistan, the International Center
for Religion and Diplomacy has sponsored programming focused on strengthening the
critical-­thinking skills of religious educators to incorporate into their teachings, recognizing
how such skills are a protective ­factor against radicalization.14 Exposure to vio­lence is an-
other significant risk ­factor for youth radicalization. In their role as counselors and confi-
dants, imams have opportunities to address violence-­related trauma in youth with training
and support. For example, the US Institute of Peace is building the capacity of religious lead-
ers in trauma healing, which can safeguard at-­risk youth from engaging in vio­lence or joining
violent groups.15
In the end, however, any P/CVE programming that involves religious leaders must ask
what it is that the field is trying to change in terms of the spiritual dimensions of religious
thought and be­hav­ior. This requires better religious literacy among prac­ti­tion­ers and policy-
makers engaged in the field and a deeper understanding of the local, community context and
the multiple roles of religious leaders in strengthening (or weakening) social cohesion, address-
ing collective and individual trauma, and mediating (or driving) local grievances—­capacities that
are critical to community resilience.

12 | Preventing and Countering Violent Extremism | USIP.ORG

Community Resilience to Violent Extremism

As an approach to countering violent extremism, community resilience is not a default middle-­

of-­the-­road option, situated between top-­down state interventions and individual radicaliza-
tion prevention programs, which have been largely discredited. Community-­led efforts, in fact,
make a lot of sense. First, communities are the locus of violent-­extremist activity: they are
where the recruitment networks of violent extremists are embedded and their operational
cells, financial sources, and systems are located. Second, extremist groups exploit local griev-
ances to gain a foothold in communities, leading to a conflict that is so localized that only
communities themselves understand the under­lying dynamics and appropriate responses. Fi­
nally, a growing research and programmatic focus on community activism and agency amid
vio­lence, or the risk of vio­lence, illuminates a set of capacities and strategies that communities
successfully deploy. What are the most effective local capacities and strategies to resist com-
munity capture by violent-­extremist groups?
­There is a distinct pattern to violent-­extremist infiltration of communities, characterized
by a medley of overlapping ties (vectors that allow violent actors to enter a community); com-
munity fracture; youth exploitation and marginalization; and predatory, abusive, or exclusion-
ary governance actors. Mapping the systemic vio­lence that extremists exploit is critical to
analyzing how and ­whether communities can resist them. De­cades of research have validated
the importance of social capital (the networks of relationships among ­people who live and
work in a par­tic­u­lar society, enabling that society to function effectively) in preventing vio­lence
and in postconflict recovery. As Ashutosh Varshney describes in his seminal study on commu-
nity resilience to urban riots in India, ­these social networks facilitate learning, trust, informa-
tion exchange, and collective decision-­making—­the capacities needed to effectively respond to
the threat of vio­lence.16 In the same way, in any community, ­there is also dark social capital:
the networks and relationships that enable venal, violent systems to function effectively and
maintain resilience. ­These social and dark-­capital networks can exist in parallel or even rein-
force each other. In communities that are resilient to violent extremism, strong networks of
social capital can counteract, or even break down, dark capital.

Dark Capital and Extremist Vio­lence

OVERLAPPING TIES: LINKING INTERNAL AND

EXTERNAL NETWORKS
In urban settings, extremists—­and their ideas—­enter communities through overlapping
ties, or established vectors by which external actors can connect to groups internally. Ami

USIP.ORG | Preventing and Countering Violent Extremism | 13

 Carpenter, in her book on community resilience to sectarian vio­lence in Iraq, observes that
rural-­urban tribal connections ­were the conduit through which extremist groups entered
Baghdad communities.17 In communities globally, in countries such as Kosovo, Niger, Nigeria,
North Macedonia, ­Kenya, Mozambique, and Côte d’Ivoire, the conduit for violent extremists
has been itinerant imams who enter through community mosques. Mosques are not only sites
of youth recruitment, but also significant revenue channels that provide financing to extremist
groups and their leaders.18 In Kosovo, for ­people who traveled to live or fight with the Islamic
State, the recruitment effort was conducted in mosques by a specific group of imams from
Macedonia with ties to Kosovo. Their religious education tied them to networks in the ­Middle
East and Gulf countries.19
While the roots and affiliation of the growing extremist threat in Mozambique’s Cabo Del-
gado Province are not completely clear, locals affirm that extremist leaders espouse the ideas of
the deceased ­Kenyan cleric Aboud Rogo Mohammed, whose sermons in Swahili have spread
rapidly through Mozambique’s Swahili-­speaking North.20 Niger’s Diffa region has maintained
deep historical ties with Nigeria’s Northeast, created when the two regions ­were unified ­under a
traditional po­liti­cal administration, Kanem-­Bornu, which existed for more than ten centuries be-
fore falling to colonialism. Maiduguri, in Nigeria, is a regional center of Islamic education and a
conduit through which Boko Haram’s teachings quickly spread to Niger’s Southeast: Nigerien
youth—­who heard Mohammad Yusuf’s teachings and attended his mosque in neighboring
Maiduguri—­formed a movement within the central mosque in Diffa, Niger, condemned the Izala
Society for criticizing state corruption while maintaining links with the government, and trans-
lated Yusuf’s teachings into the local Kanuri language.21 In a statistical analy­sis of how social
capital can strengthen group cohesion around vio­lence, The Dark Side of Social Capital: A Cross-­
National Examination of the Social Relationship between Social Capital and Vio­lence in Africa,
Ludovico Alcorta and colleagues note that when individuals are connected to larger, global
networks, leaders can more readily mobilize them to collective action by exploiting the ideas
around shared, global grievances that are rapidly diffusing throughout the network.22 Further-
more, if ­these individuals associate with a singular identity, ethnic or religious, and not with
multiple identities—­including, above all, with a national identity—­then they more intensely feel
collective grievances and are more easily mobilized to collective action. Thus, ­these local-­global
connections of marginalized youth (who are stripped of identities other than that for which they
are being excluded) through itinerant imams are potent systems of dark social capital, according
to this study on what types of social networks are more prone to vio­lence.23

THE FRACTIONALIZATION OF COMMUNITIES

Social vectors channel violent extremists and their ideas, injecting them into a highly fractured
religious tableau that, for some communities, makes it difficult to mount a coherent counter-
response. In Kosovo, the fractionalization of the Muslim community began in 2008, when the

14 | Preventing and Countering Violent Extremism | USIP.ORG

country’s Islamic community—­which managed local mosques, selected imams, and ordered
the themes for Friday prayers throughout the country—­had a crisis of legitimacy. Following
the 1999 war, two schools of thought emerged within the association: one that Kosovo should
promote its own brand of Hanefi Islam stemming from the Ottoman Empire and another that
wanted to open up the country to external Islamic influences. When the Kosovo Hanefi faction
moved to dominate the association, it lost control of the mosque system; illegal mosques and
breakaway imams emerged throughout Kosovo and began to recruit youth.24
Mozambique’s Muslim community is fractured along national and local lines, with ten-
sions between Moroccan Muslims in the South and African Muslims in the North and locally, in
the North, between Sunni Sufist communities and Wahhabist Salafist communities. Beginning
in 2008, the government’s external financial and religious engagement grew, including issuing
visas for foreign missionary organ­izations to reside in Mozambique and for its youth to attend
religious schools throughout Africa and the ­Middle East. With this expanded engagement, in-
cidents of vio­lence and disrespect against state-­sanctioned mosques grew, and fi­nally, in 2017,
so did incidents of extremist vio­lence.25
In Kisauni, ­Kenya, the site of infamous police raids on the Masjid Swafaa mosque in 2014,
the community has historically had a Shia-­Sunni divide, but Muslim groups have further
split, according to residents: “Even if ­people differed in small aspects, like how to perform
prayers, or their stand on jihadism, they opt to form their own sect rather than resolving the
issue in a scholarly way.”26
In-­group bonding, as validated by many studies, affirms and strengthens a group collec-
tive identity that is not moderated by interaction with other groups and leads to a sense
of superiority and bias against ­others.27 When confronted with an existential or real threat,
bonded groups are easily mobilized to collective action and vio­lence. The effects of strong in-­
group relationships can be attenuated by bridging ties, engaging everyday associations, or
working with members of other groups.28 ­These intergroup relationships break down bias
based on group identity, provide ave­nues of communication and reassurance when tensions
rise between groups, and establish networks of trust that can be tapped when elites manipu-
late group perceptions about the other. The fractionalization of the Muslim community can
lead to in-­group bonding based on strict adherence to religious beliefs and prevent intragroup
interactions with other sectarian groups that could erect their collective re­sis­tance to extrem-
ist inroads into the community. Extremism does not flourish in communities ­because they are
potent incubators of radicalism; it flourishes ­because the networks that can be used to strat-
egize and mobilize against it are broken.

THE DESTRUCTION OF COGNITIVE CAPITAL

A troubling aspect of any analy­sis of ­drivers of violent extremism is the use of the passive
voice—­a pernicious rhetorical device used to describe acts of vio­lence in ways that render

USIP.ORG | Preventing and Countering Violent Extremism | 15

 perpetrators invisible and shift the focus onto the victim, such as “the victim was murdered.”
Yet, in lit­er­a­ture on preventing or countering violent extremism, ­there is an even more perva-
sive tendency to identify the perpetrator as a vague, ambiguous phenomenon, as implied in
the statement, “individual radicalism is driven by local grievances.” A recent report by the
United Nations (UN) Special Rapporteur on the Promotion and Protection of ­Human Rights
and Fundamental Freedoms while Countering Terrorism has noted that “the link with interna-
tional jihadism is more tenuous on the ground than the global rhe­toric suggests. Study ­after
study reveals that the experience or perception of abuse and violations by government au-
thorities are determining ­factors that contribute to the level of vulnerability to violent extrem-
ism, or resilience thereto.”29 This finding is validated by a UN Development Programme survey
on extremism in Africa, which finds that 71 ­percent of ­those polled identified government ac-
tion as the critical ­factor that drove them to join a violent-­extremist group. The passive voice,
therefore, directs attention away from the power­ful po­liti­cal actors and state institutions that
drive youth to violent extremism while holding out in front its victims—­the communities and
youth that are exploited by them.

­KENYA CASE
In Lauren Van Metre’s research of six communities in ­Kenya, it was not the effects of long-­term
structural systems of injustice that participants attributed to violent-­extremist risk: it was di-
rect acts by corrupt, venal po­liti­cal actors (so-­called demo­cratic representatives) that stripped
local youth of status. For example, in ­Kenya’s Coast Province, politicians exploited communal-­
land tenure systems, whereby communities for generations had lived and worked the land
without formal proprietary rights. Politicians used their access to the state bureaucracy to is-
sue titles for ­these communal lands and assign them to po­liti­cal loyalists. Stripped of their
land, many youth lost access to employment and economic and po­liti­cal status. In addition,
the influx to the province of wealthier up-­country ­Kenyans, who ­were able to corner local land
and business markets, was also disrupting networks of community associative trust and disen-
franchising youth. ­Family elders ­were capitalizing on the higher real estate prices, selling land
that had been in families for generations—­housing complexes that had been rented to families
of dif­fer­ent ethnicities, who lived side by side. When ­these communal complexes ­were sold,
interethnic associations ­were disrupted, and the next generation, which had been raised with
the expectation of taking over as landlords, was displaced. Not only was the situation increas-
ing intergenerational tensions, but the influx of up-­country ­Kenyans was perceived as a po­liti­
cal power move within a highly polarized po­liti­cal landscape. Up-­country ­Kenyans ­were
affiliated with the (perceived Christian/anti-­Muslim) party in power, while the Coast Province
was aligned with the opposition.
Since in­de­pen­dence, ­Kenyan politicians have employed youth militias and gangs in
the lead-up to and during elections to disrupt opposition rallies, intimidate opposition voters,

16 | Preventing and Countering Violent Extremism | USIP.ORG

 seize communal and rivals’ land to distribute to po­liti­cal loyalists, and, most egregiously, drive
rival ethnic groups from their land in advance of the vote.30 Economic and po­liti­cal reforms,
including structural adjustments, decentralization, and po­liti­cal party reform, have improved
and strengthened this collusion.31 The current illicit triangle between politicians, gangs, and
police controls ser­vice delivery—­electricity, security, transportation—to the poorest of ­Kenya’s
urban neighborhoods. ­These sustained systems of corruption fuel po­liti­cal candidates’ cam-
paigns popularly and financially, provide gangs with steady income streams outside of elec-
toral cycles, and attract police protection rackets. At the same time, gangs provide politicians
po­liti­cal cover. They are scapegoated in a cynical po­liti­cal game whereby politicians blame
gang vio­lence for the nondelivery of government ser­vices and po­liti­cal promises even as they
collaborate with them.32 In a masterful stroke of po­liti­cal deflection, the corrupt po­liti­cal class
blames communities for the moral breakdown that allows gangs to exist and thrive. For ­those
gangs that seek greater autonomy from the po­liti­cal class, or rival gangs that challenge a politi-
cian’s turf, extrajudicial actions by the police help politicians manage the temerity of youth
gangs.
The links between state-­sanctioned killing, torture, disappearance, and po­liti­cal impris-
onment and violent extremism are well established, notably in a study of government vio­lence
in 159 countries over twenty years. The study shows strong empirical evidence that increased
levels of state vio­lence lead to increased levels of violent extremism by already existing violent-­
extremist groups. If a major group of this sort did not yet exist, above-­average levels of state-­
sponsored vio­lence in a country doubled the risk that one would emerge.33 The repression of
dissent has a similar effect: it increases the chance of group vio­lence in that state with a clear
sense of grievance ­toward the government.34 In ­Kenya, po­liti­cal vio­lence perpetrated by secu-
rity ser­vices against young men is a legacy of colonial times and occurs systematically and with
impunity. In fact, ­Kenya’s highly securitized counterterrorism operations strengthened ­these
institutions of state-­sponsored vio­lence, which regularly targeted ethnic Somali and Muslim
communities in Coast Province, driving more and more youth, in turn, into violent extremism.
Globally, however, state predation takes more direct forms: elites exploit the institutions
that are meant to address collective needs and instead use them to extract community wealth
and resources. Many po­liti­cal party representatives do not engage with constituents; instead,
they prey on them. To protect their criminal activities, they co-­opt or starve official security
forces or create private militias, generally made up of local youth, to protect their economic
rents. In other words, the very defining features of democracy—­political parties, relationships
between representatives and constituents, and institutions of justice and security—­become
the mechanisms for elite predation, in turn undermining democracy and the state’s legitimacy.
Alcorta and his colleagues’ study discusses the way cognitive social capital—­peoples’ identity,
norms, and values—­affect their willingness or reluctance to use vio­lence. ­Those who believed
using vio­lence was justified to achieve po­liti­cal goals (a small 3 ­percent of ­those surveyed)
­were most likely to have strong attachments to a singular identity, ­either ethnic or national

USIP.ORG | Preventing and Countering Violent Extremism | 17

 (most likely an attachment to a dominant ethnic or racial group associated with the state).
­Those with multiple identities—­ethnic and national—­were most likely to oppose the use of
vio­lence, believing it was never justified. Heavy-­handed and predatory state vio­lence under-
mines trust in and identification with the nation among citizens and youth (who are its most
likely targets), in the pro­cess weakening social cohesion wrought by a shared identity with citi-
zens from other ethnic, social, and religious groups.

DARK CAPITAL AND RURAL COMMUNITIES

Rural neglect—­the retreat of the state in areas that need governance of resources—in coun-
tries such as Kosovo, Mali, Niger, and ­Kenya has led to the intrusion of violent-­extremist
groups in par­tic­u­lar circumstances where traditional balancing mechanisms have weakened.
In Kosovo, following the war, the UN administration, and subsequently the national govern-
ment, largely ignored rural community economic and social development. As a result, Saudi
charities proliferated in the Kosovar countryside, providing education and health and food ser­
vices normally provided by the state. Exploiting rural poverty and fragmentation, the Saudi
Joint Committee for the Relief of Kosovo and Chechnya had a captive community for its funda-
mentalist ideology that rejected the Islam traditionally found in the Kosovar countryside—­a
religious practice that combined Islamic traditions and promoted tolerance and inclusion.35
In Mali, the retreat of the state administration in the North has been ongoing since 2012,
when, with the collapse of the regime in Libya, a Tuareg armed group returned to Mali and
ignited an armed rebellion. The state presence is primarily security focused, with minimal civil-
ian administrative offices. For example, by 2019, only 23 ­percent of civil administrators ­were
employed in duty stations throughout northern Mali. Where civil servants are pre­sent, they
are perceived as a threat to the burgeoning illicit economy through which many armed
groups are funded, in addition to taxing local populations. As state actors have retreated, al-
ternative actors have taken on local governance responsibilities, but they have not been able
to unify all ethnic groups, as competition for ­water has increased with climate change. Access
to ­water resources is now the purview of the tribal or armed group that controls that territory
and no longer a collective resource, as it was ­under state territorial administration. When land
changes hands owing to armed strug­gle, contestations become more intense.
In some cases, where single ethnic groups have had a long-­held, historical claim to the
land and ­water, customary leaders have been able to adjudicate use. In areas where multiple
ethic groups reside, customary leaders strug­gle to adjudicate use of ­water without reinforce-
ment from the state. In ­these cases, extremist groups move in to control access to ­water
through ­these fractures in community cohesion. In other cases, such as in the regions of Kedal
and Menaka, the government has sided with a par­tic­u­lar conflict actor, the Imghad tribe,
which has thrown the Menaka region into a yearslong competition between progovernment
and proautonomy forces. Since the French intervened in 2013, Kedal and Menaka have been

18 | Preventing and Countering Violent Extremism | USIP.ORG

 caught in multiple protracted conflicts within and among groups, complicated by the presence
of international and radical armed groups. T­ hese conflicts have destroyed the social trust that
previously balanced tribal relations in ­these areas. As a result, local communities have created
atomized community-­based armed groups.36
Niger has experienced a similar pattern of dark social capital that sustains extremist
power and infiltration. When customary balancing mechanisms of tribes rely on the marginal-
ization of one group over another, alliances with external actors drive support for extremist
groups and for recruitment. Black Tuaregs (a subcaste within the Tuareg tribe) and some
Fulani pastoralist groups are now perceived by state and international forces as a threat based
on their perceived or ­actual marginalization; they are cast as fully complicit with violent-­extremist
groups, but in the context of their relationship with the state. Thus, as in Mali, the po­liti­cal and
security opportunities presented by international forces or militant groups have subsumed
traditional, customary balancing mechanisms in Niger. As security has unraveled, ethnic groups
that are lower in the power hierarchy have formed community-­based armed groups. Both
community-­based armed groups and extremist groups have engaged in “chief hunting,” fur-
ther destroying the communal cohesion that solidified the economic interdependencies that
prevented intercommunal vio­lence. As state security forces crack down on the transhumance
corridors that Fulani pastoralists depend on and farmer communities continue to expand, Fu-
lani (and Black Tuareg) youth, faced with an existential threat to their status and livelihood, are
vulnerable to violent-­extremist recruitment and messaging and join in significant numbers.37

Community Resilience to Violent Extremism:

Community Capacities

Most strategies to prevent violent extremism focus exclusively on the symptoms of the prob­
lem, ­either at the broad, societal level—­service delivery, job creation, education curriculum
(critical thinking, civic education)—or at the individual level—­youth marginalization and radi-
calization. If community resilience approaches are brought into play, it is generally to address
issues of social cohesion that are a result of ­these breakdowns; the approaches aim to smooth
over community fault lines (community cleanups, sports) rather than address the root ­causes
of the breakage. This is often the criticism that local organ­izations and prac­ti­tion­ers on the
ground have of P/CVE practice: that it is designed to pacify communities with targeted, narrow
interventions that work in the short term ­because they reduce symptomatic effects. Job train-
ing for youth is a case in point. Providing young ­people with ­viable skills may resolve issues of
unemployment. However, at heart, the issue is not unoccupied youth; the issue in many socie­
ties in which violent extremism thrives is the systematic exclusion of young ­people from power.
While the lit­er­a­ture on prevention of violent extremism often cites local grievances or youth
marginalization as significant ­drivers of violent extremism, it rarely seeks to shift the po­liti­cal

USIP.ORG | Preventing and Countering Violent Extremism | 19

 power dynamics—­the dark capital ­drivers—­behind them or work with communities to strengthen
their resilience to violent extremism by strengthening community cohesion.
What does research say about community resilience to vio­lence, and how might that in-
troduce new areas of P/CVE practice? Several studies focused on community resilience to
armed groups have shared findings regarding how communities resist them. In Community
Resilience to Sectarian Vio­lence in Iraq, Ami Carpenter notes that the old Baghdad communi-
ties ­were more resilient to sectarian vio­lence ­because rural-­urban tribal connections, conduits
for violent-­extremist groups entering Baghdad, had been severed for de­cades.38 Thus, moni-
toring and disrupting ­these conduits of recruitment to violent extremism are critical for com-
munity resilience.
In ­Kenya, communities that tapped into existing informal security networks to report to
traditional chiefs on stranger activity ­were able to identify suspicious actors and activity. In one
neighborhood, a community had formed networks to monitor electoral vio­lence that ­were
also used to observe and report on other types of vio­lence, including extremist networks. In
another community in Mombasa, a group of ­women had ­earlier formed an alliance to stop a
serial rapist and continued their security monitoring to prevent other types of criminal and
violent activity, believing that it also increased the risk of rape for ­women. ­These local, infor-
mal security networks, however, had to have certain characteristics to maintain community
cohesion and not feed into destructive, or dark, capital. First, they had to be defensive, to
protect and prevent, unlike offensive vigilante groups that might invite intergroup vio­lence
or attract violent actors from outside the community.39 (Carpenter’s research reinforces this
point; in Baghdad, community security networks that monitored and prevented vio­lence ­were
more effective than ­those that joined the fight and saw vio­lence escalate.) Second, ­there
needed to be multiple, interlocking informal networks: a lone network could be manipulated—­
turned into an in­for­mant network, where community members might provide false informa-
tion for payment or retribution. Multiple networks, on the other hand, could validate one
another’s findings and provide broad, diverse perspectives on community goings-on. Multiple
networks could also provide anonymity to community members reporting suspicious violent-­
extremist activity, which was critical, as in­for­mants ­were often targeted by ­either security
forces or violent-­extremist groups.
Oliver Kaplan, in his multimethod study on how communities resist armed groups in Co-
lombia, provides some insights into how rural communities might ­counter violent-­extremist
groups.40 According to Kaplan, ­those communities that successfully repelled armed group oc-
cupation or vio­lence had three critical characteristics: They ­were apo­liti­cal and did not ally
with any par­tic­u­lar armed or po­liti­cal movement. They had existing decision-­making mecha-
nisms in place before the conflict. And they demonstrated collective re­sis­tance to armed
groups, which forced ­those groups to decide ­whether to punish the entire community or leave
it in­de­pen­dent.

20 | Preventing and Countering Violent Extremism | USIP.ORG

 Kaplan’s research reinforces the vulnerabilities of communities in the Sahel, which have
been forced to make po­liti­cal alliances with ­either international forces or violent-­extremist
groups at the same time that their decision-­making systems (that is, their customary leaders)
are weakening. Moreover, owing to ethnic or tribal cleavages or the marginalization of
­subgroups within the community, they are unable to mount collective action against violent-­
extremist groups.
Resilient communities also have bridging mechanisms across community divisions that
help manage or prevent vio­lence. In his Ethnic Conflict and Civic Life, Ashutosh Varshney
notes that ­these cross-­community groups ­were LPCs or civil society groups. In ­Kenya, they
might be local business or religious leaders who could reach out across conflict lines to pro-
vide reassurance when the risk of vio­lence ­rose or could help end cycles of retributive vio­
lence ­after community bombings or attacks. Following a market attack in Kisauni by violent
extremists, for example, Christian leaders reminded their constituents, who ­were mobilizing
to conduct retributive vio­lence against Muslim neighbors, that Muslims ­were also victims in
the attacks. Communities can share information, provide reassurances, and dismiss inflam-
matory rumors through ­these all-­important bridging relationships. They can also or­ga­nize
collective re­sis­tance and support the development of effective community strategies to
­counter violent-­extremist incursions—­strategies that work for all residents in the
community.
Fi­nally, communities that ­were able to manage the state or substitute for the functions of
the state ­were resilient to violent extremism. In contexts of violent extremism, the state has
­either ­adopted extralegal or abusive practices in support of counterterrorism strategies or
become itself a conflict actor. Po­liti­cal actors may have preyed on communities to collect illicit
rents, undercutting the legitimacy of the state as a protective force. Communities that can
contain or limit exposure to predatory state actors—­for example, by reassuring security forces
that they have the capacity to ­counter extremist groups and limiting security force operations—­
are more resilient. Heavy-­handed security operations not only increase youth vulnerability to
violent-­extremist messaging, but they also undermine the community’s ability to share infor-
mation and develop countering strategies to extremism. Afraid of inviting unwanted security
force assaults should they name the violent-­extremist threat or identify ­those at risk, commu-
nities go ­silent and cannot engage in collective action. In other words, security force abuse hits
communities doubly by increasing their risk and undermining their resilience. Thus, as demon-
strated in multiple research studies, constructive engagement with the state, ­whether through
managing state abuse or working in partnership, is crucial for resisting or preventing violent
extremism.41
In rural communities, hybrid governance models can play a critical role in countering vio-
lent extremism in areas where the state has ­limited or no presence or where the government
has ceded its authorities to traditional or informal leadership. The county of Wajir, in ­Kenya, is

USIP.ORG | Preventing and Countering Violent Extremism | 21

 located in the northeast near the Somali border, where al-­Shabaab has been increasingly ac-
tive, even organ­izing one of its most deadly attacks on ­Kenyan soil in the neighboring county of
Garissa, killing 179 students and teachers at Garissa University. In 1992, following postelection
vio­lence, a group of ­women from a local market or­ga­nized a peace-­and-­dialogue group to stop
further cycles of vio­lence. The Wajir ­Women for Peace Group soon expanded to form the
Wajir Peace Group, inclusive of all tribes in the country. ­Women’s groups and local peacebuilders
or­ga­nized a larger, all-­clan dialogue that resulted, in 1994, in an all-­clan agreement, the Al-­
Fatah Declaration, to end the vio­lence and establish a permanent council of clan elders that
established an early warning system and engaged in local peacebuilding and conflict resolu-
tion forums. The tribal council, to this day, continues to work with the county commissioner
through a local government directorate of peace and cohesion—­a form of community-­
government collaboration that is seen as a potent barrier to extremist vio­lence, especially in
light of the county’s proximity to Somalia and al-­Shabaab networks.42
According to local communities, Niger’s High Authority for the Consolidation of Peace—­a
national organ­ization established in the 1990s following a series of Tuareg rebellions from
1990 to 1995 in both Mali and Niger—is playing a constructive role amid the increasing ex-
tremist vio­lence on the country’s borders with both Mali and Nigeria.43 While dif­fer­ent in
scope and scale from the local Wajir mechanisms, the social cohesion patterns for ensuring
rural resilience are the same: inclusive networks of diverse tribal actors (­women, civil society,
youth) connected to the government whose longevity and experience with dif­fer­ent forms of
vio­lence demonstrate their ability to adapt and learn. If they are to resist violent-­extremist
recruitment or capture, it is critical that ­these models of hybrid governance are not exclusion-
ary or repressive against certain classes or groups. (It is through the most marginalized that
violent-­extremist groups can enter communities.)44

How Counterterrorism Strategies Reinforce Dark Capital

YOUTH CAPTURE
Community resilience practice must take into account not only the patterns of social cohesion
that strengthen community capacity to prevent or recover from vio­lence (bonding, bridging,
vectors, vertical) but also the dark social capital that reinforces systems of vio­lence at the
community level. It is not enough to reinforce relationships of resilience without mitigating or
dismantling the systems of vio­lence that violent-­extremist groups exploit.45 Counterterrorism
strategies promoted by the international community and the United States that reinforce and
sustain ­these dark networks compromise this effort, which is why governments and po­liti­cal
leaders so readily adopt (co-­opt) them. Especially in cases where po­liti­cal predation and cor-
ruption fuel the systemic vio­lence, counterterrorism strategies secure the very pinnacles of
dark capital power.

22 | Preventing and Countering Violent Extremism | USIP.ORG

 For example, the manipulation of youth by a country’s po­liti­cal actors—­organ­izing youth
into gang, vigilante, and criminal groups to protect and increase politicians’ economic rents—
is a global phenomenon.46 How do counterterrorism approaches sustain and strengthen ­these
pernicious forms of malgovernance? ­Kenya’s antiterrorism laws, for example, have given lati-
tude for heavy-­handed police action, reinforcing already existing collusion between po­liti­cal
actors and the police in the recruitment and management of youth militia and gangs. In their
justification for ­these laws, po­liti­cal leaders divert blame from their role, or that of the state,
in driving violent extremism while scapegoating disaffected youth and their communities—­
reinforcing a pernicious narrative that shields entrenched po­liti­cal rentier systems. Fi­nally, anti-­
terrorism operations against violent-­extremist recruitment networks and operations now look
like attacks by the corrupt po­liti­cal class on a rival system of youth capture—­that employed by
violent-­extremist groups. Especially for ­those youth who have directly experienced the dark
forms of po­liti­cal corruption and predation in the guise of demo­cratic politics, violent-­extremist
groups do not look particularly dif­fer­ent, except that they offer opportunities for the dispossessed.
Given the sclerotic, exclusionary leadership in many fragile states and the lack of opportunities
for professional advancement, the appeal of youthful, ambitious violent-­extremist leaders is a
potent promise of mobility and agency, as opposed to the alternative—­membership in po­liti­
cally sponsored gangs and militias that the po­liti­cal class ­will cynically abandon or scapegoat
for their own po­liti­cal advantage.

YOUTH MARGINALIZATION
The field of neuropsychology has opened up new thinking on youth radicalization, which points
to behavioral and normative relationships between marginalization, sacred values, and violent
extremism. Individuals who experience abrupt, acute acts of marginalization and loss of status
become strongly bonded to their in-­group: they may be willing to fight and die in defense of
that group and its values. The rigidification of their group identity and their extreme loyalty
to that group can be mapped in neural scans, which show strong activity in the rules-­based
part of the brain, compared with activity in that part of the brain that involves judgment,
weighing and calculating when values must be protected or ­whether other considerations are
more impor­tant. Strong in-­group bonding is also strongly associated with a person’s willing-
ness to sacrifice for the group and engage in unethical, violent be­hav­ior on its behalf. Margin-
alization also has been shown to transform nonsacred values to sacred values, so that
individuals accumulate more and more values that require protection.47
In Kosovo, Mali, Niger, and ­Kenya, profound systems of marginalization are often the re-
sult of abuse of international policies or corrupt national government, at times sanctioned by
counterterrorism policies. In Kosovo, many ­people who traveled to fight with the Islamic State
­were former combatants in the Kosovo Liberation Army, whose status ­after the war was sharply
diminished and for whom ­there was ­little opportunity, owing to the lack of international

USIP.ORG | Preventing and Countering Violent Extremism | 23

 support for the countryside. In ­Kenya, Muslim youth who had lost access to communal lands
owing to po­liti­cal corruption or through parents’ or grandparents’ sale of ­family holdings (their
inheritance) to up-­country ­Kenyans ­were perceived as highly vulnerable to violent-­extremist
recruitment. Also vulnerable ­were ­Kenyan youth who had experienced abuse at the hands of
the security sector, who had increased latitude to target Muslim communities ­under new
counterterrorism laws. With the escalation of vio­lence following the involvement of interna-
tional forces in the Sahel, Fulani and Tuareg pastoralists in Mali, Niger, and Burkina Faso, long
marginalized by the state, have chosen to align with violent-­extremist groups. In addition, both
groups continue to support slavery, even ­after its prohibition and, in some countries, criminal-
ization. Black Tuareg youth, the slave cast in Tuareg society, and Fulani youth—­the most mar-
ginalized of the marginalized—­make up most of the Islamic State and al-­Qaeda recruits in the
Sahel.
The dark capital of systemic exclusion of youth is intensified by bad governance, such
as the selling of communal lands for personal gain or retraction of state presence or ser­vices
and counterterrorism strategies and operations that target groups or individuals through
state-­led discrimination or vio­lence. The already marginalized have strong in-­group affilia-
tions; punctuated moments of acute exclusion by governance actors then cause even lesser
values to become sacred, strengthening an individual’s willingness to sacrifice and commit
vio­lence on behalf of the group. Research has shown that when ­these individuals’ identities
become singular (that is, as they abandon broader national or tribal identities, owing to the
exclusionary acts of governance actors, which can be sanctioned by counterterrorism strate-
gies and operations) and, at the same time, they are exposed to external ideologies and
ideas from international networks, vio­lence becomes acceptable to them. Thus, counterter-
rorism strategies and operations that further scapegoat already marginalized groups and
individuals expand their set of sacred values and their willingness to sacrifice for their in-­
group by committing acts of terrorism and vio­lence to protect that group, among other
­things.

Community Resilience to Vio­lence and to

Violent Extremism: The Research Base

Resilience research studies have incorporated science-­based research methods, including

clearly articulated hypotheses and methods that test them and validate or triangulate the
findings, awareness of and methods for managing bias, exploration of alternative explana-
tions, and incorporation of ethical approaches. The primary focus ­here is on ­those studies
that address the resilience of urban communities in three areas: networks of social cohesion,
security, and engagement with the state. In general, each of ­these studies concludes that

24 | Preventing and Countering Violent Extremism | USIP.ORG

 communities with ­little or no violent extremism had developed strategies and exhibited signifi-
cant agency in addressing the threat. They had similar vulnerabilities to communities with
active violent-­extremist recruitment and operational cells, but they had been able to self-­
organize. However, community activism, adaptation, and agency ­were not enough for any of
­those communities. Their relationship with the state was critical; in fact, in many ways, the
state was a determining ­factor. This addresses head-on what has often been a critique of the
community resilience approach to vio­lence and violent extremism: that it can be a strategy by
states to relieve them of local responsibilities by highlighting the effectiveness of community
agency and leadership. In fact, vertical cohesion, between communities and governance ac-
tors, is a critical aspect of any system of community resilience.

Social Cohesion, Social Capital, and Resilience:

The Research Base

The first generation of community research studies emerged out of the individual radicalization
thread that dominated the field of counterterrorism and countering violent extremism in the
initial years. Some researchers ­were consumed with deciphering what, in an individual’s envi-
ronment, was driving radicalization. Other studies focused on the pos­si­ble role that commu-
nity organ­izations and leaders could play in countering radicalization networks and mitigating
youth risk to recruitment. This initial phase of research was tainted somewhat by the “sus-
pect” community approaches prevalent at the time: that weaknesses in the socioecol­ogy of
communities ­were responsible for the infiltration of recruitment networks and the vulnerabil-
ity of youth and that instrumentalizing community leaders and organ­izations in the fight
against violent extremism would shore up youth re­sis­tance to recruitment.
A second generation of studies on community resilience focused on community
­responsibility—in many ways, a ­counter to the suspect community approaches. ­Here, many
researchers emphasized ­either the lack of evidence or faulty assumptions regarding commu-
nity ­factors driving youth vulnerability to violent extremism. For example, the Eu­ro­pean Insti-
tute of Peace’s forward-­looking survey of Molenbeek-­Saint-­Jean, a community in Brussels with
known numbers of ­people who traveled to live or fight with the Islamic State, shows that the
community, in fact, rejected fundamentalism and its appeal to youth, sought better relation-
ships with the police, whom they trusted, and ­were open to building better relationships with
communities outside their predominantly Moroccan immigrant base. ­Others noted that it was
national po­liti­cal scapegoating and targeting of communities and individuals within them that
entrenched their vulnerability and weakened their collective response—­that is, their resil-
ience. This second generation of research challenged the assumptions of the first generation
that the risk ­factors for violent extremism resided within communities themselves and expanded

USIP.ORG | Preventing and Countering Violent Extremism | 25

 the scope of research to consider impor­tant externalities, such as the role of the state, sys-
temic discrimination and marginalization, and so on.
A third generation of research is rooted in the growing influence of the peacebuilding
field on preventing and countering violent extremism and its focus on the prevention of vio­
lence in general. This research is primarily inductive; it makes no assumptions regarding com-
munity risk or resilience to vio­lence; rather, it investigates community comparative cases to
explain why among communities with similar risk, some fall into vio­lence and ­others do not.
­There is also the assumption that communities have agency and authority as frontline actors
in resisting violent extremism and should not be interpreted through the values and interests
of external actors. This third generation of research has identified a number of resilience
­factors across dif­fer­ent communities and contexts of vio­lence, including defensive security
postures; interlocking, informal security networks; the monitoring and shutting down of vul-
nerable vectors; bridging social cohesion; and constructive engagement with state actors that
keeps the state out, in the case of conflict actors.
However, community resilience research has not evolved to a level where it can inform
practice, for a number of reasons: The research describes community resilience networks at
high levels but not how they are governed, how they manage membership, or how trust was
established. It also does not establish how communities learn, adapt, strategize, and imple-
ment their re­sis­tance strategies. Fi­nally, it does not establish ­whether resilience is an organic
pro­cess, ­whether at-­risk communities can learn from other communities, and ­whether inter-
national actors can strengthen or even create resilience networks in communities without
delegitimizing them.
A number of studies have begun to look specifically at the nature of social cohesion as it
contributes to community resilience to vio­lence. This is a critical area for further research,
given its role in strengthening resilience. A recent Mercy Corps study on the Tillabéri region in
Niger has established that working trust was the only cohesion ­factor that prevented commu-
nity vio­lence. Collaboration, shared understanding, coordination, and collective action, as
types of social cohesion, did not yield vio­lence prevention outcomes. Meanwhile, Oliver Ka-
plan’s study of communities in Colombia that resisted armed groups found that collective ac-
tion was the determining ­factor that contributed to their nonviolent re­sis­tance. What types of
social cohesion work (trust building, collective action, collaboration) for vio­lence prevention
and ­under what circumstances must be better understood, including how vio­lence, state re-
pression, and security sector abuse unravel systems of community cohesion. In addition, more
research needs to be conducted on the role of international and national actors in programs
focused on building and strengthening community cohesion: Can external actors fund, lead,
and participate in ­these efforts without delegitimizing them? What does effective external
support look like, and, if cohesion is built over time, how long must t­ hese pro­cesses be sus-
tained? Fi­nally, what are other international and national actors ­doing in the same space: Are
they reinforcing the systems of dark, violent cohesion or weakening community resilience

26 | Preventing and Countering Violent Extremism | USIP.ORG

in ways that undercut the very pro­cesses they are trying to build and strengthen in a
community?
It is broadly acknowledged that local grievances provide opportunity for violent-­
extremist exploitation, but ­there has been ­little study of which grievances are of par­tic­u­lar
consequence, ­whether they are systemic or proximate, and how they interact with youth,
who are the primary targets of such recruitment. Anecdotes from ­Kenya and the Sahel dem-
onstrate that it is a combination of the systemic marginalization of a class of land use and an
acute act of exclusion and loss of status caused by governance actors (elected politicians,
customary leaders) that creates an opening for radicalization. More attention certainly needs
to be paid to the role of land tenure and violent extremism, the impact of state po­liti­cal
­corruption and security abuse at the local level, and, fi­nally, how systems of dark capital—­
sustained networks of vio­lence—­are created or reinforced by the state and international actors
in ways that invite extremist vio­lence. It is also clear that engagement with and ­management
of the state—­civil servants, local government officials, governors, po­liti­cal representatives—­are
critical ­factors in community resilience. State actors can back or facilitate locally negotiated
agreements; provide security, health, and education resources; and translate local needs to
national authorities to ­counter and prevent violent extremism. Communities can only achieve
resilience with a dispassionate, unbiased, responsive state or hybrid governance system; they
are at risk when that state or system ­favors certain actors on the ground or participates in the
marginalization of one group by another. More research and evaluation need to be done on
governance and the prevention of violent extremism, research that addresses the following
questions:

• How can the impact of corrupt po­liti­cal actors on community risk of violent extremism be
mitigated? What is the relationship between po­liti­cal corruption and youth recruitment—
is it a sustained system of exclusion? Punctuated incidents? How do the effects of corrup-
tion interact with other violent-­extremist recruitment risks?
• If the state cannot sustain a presence locally, given security risks, how can effective gov-
ernance support for communities be implemented? What are the critical state capacities
that communities need to sustain their networks of resilience?
• How can state-­sanctioned marginalization of ethnic, tribal, or rural groups, which in-
creases vulnerability to violent extremism, be resolved?
• How can international actors incentivize good governance while respecting state sover-
eignty? How can international actors refrain from contributing to the bad governance
prob­lem?

Fi­nally, in many ways, the established research on the role of social cohesion and the
prevention of vio­lence still rests on the old paradigm of the foundations of ethnic vio­lence:
ethnic bonding, ethnic bridging, and elite manipulation. ­There is a critical need to conduct

USIP.ORG | Preventing and Countering Violent Extremism | 27

 more research across communities and dif­fer­ent violent-­extremist contexts to validate the so-
cial cohesion dynamics that prevent or enhance violent extremism. This research is critical, as
the international community and national governments are feeding into or participating in the
systems of dark capital that violent-­extremist groups are exploiting. ­These systems are unseen
for many reasons, including the per­sis­tence of ideological assumptions, narratives perpetu-
ated by power­ful state actors to strengthen or hide their illicit networks, or the perennial im-
pulse to blame the victim to protect existing power structures.

Promising Approaches to Community Resilience and

the Prevention of Violent Extremism

This effort to examine the evidence on preventing and countering violent extremism and to
identify promising new approaches to stanching its spread has centered on community re-
silience. Two promising areas of community resilience practice are LPCs and hybrid
governance.
Local peace committees have largely been or­ga­nized as a grassroots response to com-
munal and ethnic vio­lence, where the state has been largely absent. In the 1990s, the com-
mittees ­were ­either incorporated into or established as a result of formal peace settlements,
such as in South Africa and Burundi. They have not been systematically considered or used
for countering or preventing violent extremism. This section explores the challenges and op-
portunities LPCs would pre­sent if they ­were incorporated as a P/CVE practice. Hybrid gover-
nance is a bit of a dif­fer­ent study. The breakdown or weakening of hybrid governance structures
has provided opportunities for violent-­extremist groups to proliferate in many countries. The
question is how and ­whether the two practices can be reconstituted as a frontline force
for P/CVE.

LOCAL PEACE COMMITTEES

Local peace committees arose in the 1990s in response to a confluence of several ­factors: evo-
lutions in thinking in the peacebuilding field on the importance of localization and local cul-
tures of peace, an explosion of civil society and community efforts to prevent and resolve
burgeoning intrastate wars in Africa and around the world, and the recognition that elite-­
negotiated peace deals alone could not deliver peace. In 2013, the movement became for-
malized as a peacebuilding approach with the publication of Andries Odendaal’s A Crucial Link:
Local Peace Committees and National Peacebuilding, published by the US Institute of Peace.48
The book, which focuses exclusively on formal LPCs (­those that ­were mandated in formal
peace agreements in South Africa, ­Kenya, and Burundi), led to a community of practice around
infrastructures for peace (I4P) that included the following:

28 | Preventing and Countering Violent Extremism | USIP.ORG

• robust research focused on cross-­national comparative case studies to tease out the
conditions of success for LPCs that contributed to the building of I4P;
• funding provided to LPCs in Libya, Mali, Nigeria, Niger, Iraq, ­Kenya, and ­others through
practices initiated by the UN Development Programme and UN Peacebuilding Fund;
and
• the Infrastructures for Peace initiative, formed by the Global Partnership for the
­Prevention of Armed Conflict, creating a learning hub for civil society networks to forge
peace infrastructures in their countries.

This evidence review focuses on the burst of I4P research and practice between 2014
and 2018. While the initiative tapered off ­after about 2016, the UN Development Programme
is currently funding some LPC work in Iraq. The Global Partnership hosted an I4P regional
conference in West Africa in 2020, although the conference report reads somewhat like a
plea, or an advocacy campaign, to recognize the value of peacebuilding efforts by funding
them.49

Local Peace Committees: The Research Base

Most research on LPCs focuses on committees established by formal peace pro­cesses, such as
­those in South Africa and Northern Ireland, where a peace agreement created an architecture
that structured community participation. However, in other cases, such as Burundi, LPCs ex-
isted before the peace agreement and ­were officially incorporated into it.50 Emerging out of
this comparative case-­study analy­sis is a consensus on what formal LPCs can and cannot do,
how they should be structured and resourced for success, and their impact on social cohesion
and community resilience. More research is needed on informal LPCs, as they hold impor­tant
insights on indigenous conflict-­resolution practices, the way communities negotiate inclusion,
and the relationship between LPCs and community resilience to vio­lence. This focus on LPCs
grew into a larger research interest in established I4P and relationships between communities
and governments within the system.
Local peace committees have existed as organic, bottom-up institutions in Africa for de­
cades. Their value was recognized in South Africa when negotiators observed that black com-
munities ­were unlikely to engage with the state and government institutions that had overseen
apartheid. To gain popu­lar buy-in for the peace pro­cess and citizen engagement in peace im-
plementation, the settlement institutionalized LPCs and mandated them to conduct local
dialogue, build tolerance, and engage the community in problem-­solving exercises. The post-­
apartheid transition in South Africa was extremely volatile, and the LPCs ­were credited with
mitigating its most violent effects. In 2013, Andries Odendaal’s comparative analy­sis of LPCs in
South Africa, Sierra Leone, Northern Ireland, and ­Kenya launched the I4P community of prac-
tice, identifying the capacities, and limitations, of formal LPCs. According to Odendaal, formal

USIP.ORG | Preventing and Countering Violent Extremism | 29

 LPCs excel at resolving the everyday conflicts that communities experience—­disagreements
that, in times of escalation or recovery from vio­lence, become acute. They use dialogue, prob­
lem solving, and information sharing in the effort to prevent vio­lence. They cannot enforce
peace when armed groups are promoting vio­lence, nor can they address vio­lence’s structural
­causes, especially when they are rooted in the policies of the state.51
Research indicates that LPCs do not only facilitate resolution of community conflicts,
but also transform be­hav­iors and norms, helping community members to gain confidence in
nonviolent methods for resolving conflict and to learn to trust one another.52 They also con-
tribute to issue transformation, which can then lead to structural transformations over the
long term. Dialogue fundamentally reframes conflict issues, as participants increase their
knowledge and gain multiple perspectives, potentially reaching compromise. And as compro-
mises are reached, power dynamics in the communities can shift over time, leading to struc-
tural change.53
While peace committees can have significant transformative effects locally, their affilia-
tion with a national structure is a distinctly mixed bag. Where LPCs have been externally im-
posed, they must focus first and foremost on establishing their legitimacy by clarifying their
roles and responsibilities vis-­à-­vis other local institutions, confronting community concerns
and suspicions regarding its activity and demonstrating complete transparency in its activities
and mandate. When national actors and institutions push their own agendas on LPCs or put pres-
sure on them to exercise position and authority, they weaken committees’ fundamental
power—to provide a space for consensus and compromise. The relationship between LPCs
and local governance actors can also be both problematic and promising. The committees can
strengthen the social cohesion that makes local governance pos­si­ble, overriding the rule of
armed actors. They can drive local agreements that allow elections to occur peaceably and
armed actors to reintegrate. Where local government has been devastated by armed conflict,
LPCs can insert themselves into the po­liti­cal system and supersede the authority of local gov-
ernment officials.
Much of the research on informal LPCs is rooted in the African experience, where peace
committees have been plentiful as states have withdrawn from conflict-­affected areas, such as
Wajir, ­Kenya, and throughout Burundi in the 1990s.54 ­These committees often drew on the
customary conflict-­resolution practices of local institutions, such as the bashingantahe
­­(a widespread informal justice mechanism) in Burundi.55 Their roots in customary practice,
however, have had mixed effects on community peace. As they ­were designed to reestablish
contact between individuals and groups that had been in conflict, their goal was community
harmony and reconciliation: inclusive, win-­win solutions achieved through dialogue and joint
prob­lem solving.56 But ­these informal LPCs also reflected the power dynamics in their commu-
nities. If a community was male-­dominated, so ­were the LPCs. If age was associated with
power in a community, the LPCs would be populated with the older generations. Committees

30 | Preventing and Countering Violent Extremism | USIP.ORG

in highly patriarchal communities, in ­these cases, often ended up reinforcing, rather than dis-
mantling, the systems of exclusion that undergirded local vio­lence.57
A cross-­case study analy­sis of informal LPCs, including cases in the Philippines, Colombia,
­Kenya, and Somaliland, describes successful committees as being rooted in communities that
have significant resilience—­“social networks, existing structures and development-­oriented
social norms and values . . . [​ that] became building blocks for their peacebuilding.”58 They also
shared another resilience capacity: the ability of the community to come together to cre-
atively advance their collective economic well-­being. Individuals in the community expressed
confidence in their own agency and the unique role they played in contributing to the com-
munity’s health and development. In ­these resilient communities, citizens would take
owner­ship of a peace pro­cess, in the same way that they would own other community needs—­
development, security, and so on. The capacities that they harnessed for development ­were
used for local peace processes—­the very definition of resilience. ­These resilience capacities—­
social trust, individual agency, inclusive networks—in the initial response to vio­lence allowed
the community to come together “to ensure a common concern and vision and they charted a
better and preferred peaceful f­ uture.”59
The committees explored in the study differed significantly from formal LPCs engaged in
the implementation of a peace settlement. Whereas ­these formal LPCs ­were more successful
when engaged in everyday conflict resolution, informal LPCs addressed significant conflict and
vio­lence head-on. A meta-­evaluation of LPCs in conflict-­affected communities conducted by
Peace Direct shows that they are highly effective in situations of long-­term chronic conflict,
where state involvement is often missing. They engage in local dispute resolution for serious
conflicts—­such as farmer-­herder ­battles, escalatory vio­lence, electoral vio­lence, and engage-
ment with armed militias. Furthermore, and highly relevant to P/CVE, LPCs, when they engage
with youth, are a highly impor­tant protective ­factor. Youth involvement in LPCs, especially in
socie­ties where youth are excluded from formal po­liti­cal, social, and economic structures, are
less vulnerable to recruitment in violent-­extremist and other armed groups. When participat-
ing in LPC events and dialogue, young ­people ­were able to create their own vision of peace
and work actively to realize it. This exercise, according to the evidence, prevented youth en-
gagement in vio­lence, increased their participation in demo­cratic pro­cesses and institutions,
and improved their perceptions of other groups within the community.60
Comparing the research findings on formal and informal LPCs raises some critical issues
and gaps that should be explored before LPCs are drawn into P/CVE work. First, many coun-
tries in the Sahel and the Horn of Africa are developing infrastructures for P/CVE, approving
national security strategies that call for community engagement and support. How ­these LPCs
are incorporated into that infrastructure is critical for the types of vio­lence they can address:
committees associated with the state seem to have less authority to resolve major incidents
to vio­lence, whereas local committees in highly resilient communities (where social trust and

USIP.ORG | Preventing and Countering Violent Extremism | 31

 individual agency are pre­sent) have significant capacity and authority to deal with violent ac-
tors and core community conflict. How LPCs establish their authority in the community and
the impact on their capacity to resolve conflict needs to be further researched and
understood.
Moreover, more research needs to be done on failed LPCs, and on LPCs that initially
failed and then succeeded, to ensure that the conditions for success for P/CVE are understood
and implemented.
Fi­nally, a critical f­ actor for LPC success or failure appears to be how and in what commu-
nity power context committees exercise inclusion. Studying ­those dynamics is critical to the
effectiveness of LPCs within P/CVE: for example, an LPC pushing for inclusion in a highly exclu-
sionary society may help prevent youth recruitment to violent extremism but risks a loss in
authority with community leaders for resolving other community conflicts.

Local Peace Committees in Practice

Several program evaluations of LPCs in Burundi, the Central African Republic, Côte d’Ivoire,
Niger, and Senegal yield in­ter­est­ing best practices that should be considered for ­future pro-
gramming.61 Each program had an extensive, multifaceted evaluation phase and focused on a
unique aspect of LPCs.62 In 2005, Search for Common Ground worked with LPCs in ten Côte
d’Ivoire villages that had experienced significant levels of vio­lence in the north, west, and cen-
ter of the country. The proj­ect used multiple, mutually reinforcing efforts to build local support
for peace that then legitimated the work of the LPCs. ­These efforts included enhanced train-
ing for LPC facilitators and expanded training to include new facilitator cohorts of youth and
­women. The proj­ect also recognized the importance of strengthening a local culture of peace
to enhance the LPC’s work. It included participatory theater and radio programming to edu-
cate community members on the importance of nonviolent approaches for resolving conflict.
While the evaluation did not provide information on the outcomes in social cohesion and vio­
lence prevention for ­these communities, it did reinforce the point that building a community
culture of peace cannot depend solely on LPC dialogue. Committee efforts must be reinforced
through a variety of ave­nues that communicate to the community more broadly the social
norms and be­hav­iors associated with peace and what makes it a ­viable alternative to
vio­lence.
A UN Peacebuilding Fund effort in Niger from 2015 to 2018 focused primarily on strength-
ening the position of youth and ­women in the community through the work of international
organ­izations in support of LPCs throughout the country. The cross-­programmatic evaluation
noted that it was not enough to implement programs specifically on ­women and youth’s inclu-
sion. All programming designed to support local peacebuilding committees—­economic, devel-
opment, health, security, conflict resolution, and the like—­had to be linked together with the
common goal of strengthening inclusion to fully realize cumulative community social change.

32 | Preventing and Countering Violent Extremism | USIP.ORG

A critical component of this intensive work was a robust intersectionality analy­sis on youth
and ­women and their relationship to vio­lence and violent extremism through the multiple
roles they played in a community. The inclusion work succeeded when it did not assume or
treat ­either youth or ­women as a homogeneous group. For example, certain subgroups of
youth and ­women ­were more at risk for vio­lence than ­others, depending on their socioeco-
nomic, po­liti­cal, cultural, religious, or ethnic identities. Guided by this risk analy­sis, the pro-
grams strove to meet beneficiaries and recipients where they ­were. In par­tic­u­lar, programs
had to be specially designed to ensure the participation of the most marginalized ­women
and youth, including by providing childcare, enhancing their learning and participation
through preparatory activities, securing travel, recruiting them through their trusted lead-
ers, and so on.
A US Agency for International Development–­funded proj­ect in the Central African Re-
public from 2016 to 2019 sought to establish LPCs in eight neighborhoods in the city of Bangui
and six in mining areas in the southeast region of the country. The evaluation of this effort re-
inforced the need for LPCs to be fully integrated with civil society and local government in
­these communities, which required a much greater level of technical skills development of LPC
members to give them the confidence and capacity to form better partnerships. In addition,
the proj­ect needed to fund more and better joint meetings between LPCs, local government,
and civil society partners to ensure agreement regarding local conflict resolution strategies
and roles. Fi­nally, thought had to be given to how to make the committees self-­sustaining—­
that is, how they could make money for their organ­izations in ways that did not undermine
their local legitimacy. This final point regarding donor de­pen­dency and self-­sustainment was
reinforced by an assessment of LPCs in Casamance, Senegal, which showed the harmful ef-
fects of donor-­led efforts. The commodification of LPCs in Senegal’s competitive donor market
for civil-­society funding had inverted LPC incentives: the committees predominantly responded
to donor demand rather than local community input regarding the issues they addressed and
the methods they used. The result was that, in general, the committees in Casamance ad-
dressed the symptoms of the conflict, not the root ­causes, and ­were perceived by locals as
partial actors.
The final evaluation of LPC work in Burundi resulted in two tools that ­were designed to
enhance local community engagement in the work of the committees and international sup-
port for them. The evaluators brought together civil society organ­izations, international do-
nors, members of LPCs, and traditional leaders to brainstorm ideas based on their existing
experience working with such committees. The proj­ect resulted in a set of dialogue questions
to enhance community engagement and better outcomes from LPC dialogue on the nature of
peace the community wanted and the means for getting ­there. The questions ­were accompa-
nied by a set of visual prompts—­images that would spur discussion in illiterate communities
and lay the foundation for a more in-­depth conversation on the following questions:63

USIP.ORG | Preventing and Countering Violent Extremism | 33

 • How should we deal with the past?
• Is our purpose to stop vio­lence or find responses to structural ­causes?
• Are we working ­toward grassroots peace or comprehensive peace?
• Should we deal with peace directly or indirectly through development?
• What degree of vertical collaboration do we seek?
• Who should be included in our efforts?
• What are our sources of inspiration—­external or internal?
• What is the role of traditions and customs or customary institutions?
• What relation should we establish with state institutions?
• ­Will this be a permanent or transitional structure?
In addition, videos ­were developed to assist international organ­izations by laying out the
complex decisions that communities have to make, with LPC mediation, to integrate the re-
sponses to t­ hese questions into local peacebuilding strategies.

Local Peace Committees and Community Resilience to Violent Extremism

This evidence review hypothesizes that the slowing of the I4P initiative may owe in some part
to the layering of counterterrorism operations onto many of ­today’s current conflicts. This
leads to what must be a fundamental question: Can LPCs strengthen community resilience to
violent extremism and become effective mechanisms for ­either preventing or countering vio-
lent extremism? Several policy-­level dilemmas must be addressed before exploring the evi-
dence and programmatic base for LPCs and what this means for preventing and countering
violent extremism.
As noted, LPCs fall into two categories—­formal and informal. As many countries adopt
UN national counterterrorism action plans or national security strategies for countering ter-
rorism, it is pos­si­ble that LPCs could be part of a state-­community infrastructure for prevent-
ing and countering violent terrorism, where ­these strategies call for community or civil
society engagement.64 In most countries that have currently ­adopted LPCs—­Côte d’Ivoire,
­Kenya, Niger—LPCs are already tackling other manifestations of vio­lence, such as electoral or
secessionist vio­lence. In this case, could currently existing LPCs also work to prevent or
­counter extremist vio­lence? The fundamental issue is ­whether an I4P for addressing one type
of vio­lence can be used for addressing another, especially in the context of violent extrem-
ism, where the infrastructure is dedicated to an entirely dif­fer­ent objective, namely, winning
the war on terror.
­There is an increasing effort in West Africa, ­Kenya, and other countries affected by vio-
lent extremism to engage communities constructively in national counterterrorism strategies,
such as ­Kenya’s county action plans. ­These efforts pose fundamental policy contradictions
around the legitimate use of vio­lence, the internationalization of local conflicts, and the role of
the state that could potentially complicate the work of LPCs. The war on terror and the

34 | Preventing and Countering Violent Extremism | USIP.ORG

 national counterterrorism strategies that support it incorporate a rigid morality that grossly
undermines peacebuilding efforts, as seen in Af­ghan­i­stan, Mali, ­Kenya, among other places.
The only option in counterterrorism is to defeat terrorist groups and any group associated with
them; negotiated outcomes are not acceptable.
The conflict binary (win/lose) imposed by counterterrorism has severely complicated lo-
cal peacebuilding efforts in several ways. First, governments seize on ­these internationalized
narratives to demonize entire groups that have historically challenged the state (through
­secession, autonomy, insurgency) when associated factions or individuals align with violent
extremists. In many cases, violent extremism allows the state to shift the conflict narrative, ­either
to rationalize historical discrimination by the state or to direct attention away from the role of
the state in ­these localized conflicts. The conflict narrative shifts from local grievances to the
international spread of jihadism, limiting the ability of LPCs to deal with the historical root
­causes of the conflict. This deflection also eliminates a core component of an I4P: the state,
which has now diverted attention from its role in and responsibilities for the conflict or is sim-
ply no longer a presence on the ground.
Second, the internationalization of the conflict fundamentally transforms the localized
conflict as conflict actors realign along new resource and incentive structures. The changing
conflict ecosystem is just such a case in central Mali, where Fulani tribes are now associated
with JNIM (Jama’at Nasr al-­Islam wal Muslimin) and where Dan Nam Ambassagou (which
draws its members from the primarily non-­Muslim Dogon communities) have been aligned
with Mali security forces. What conflict would local LPCs be solving when extremist vio­lence
becomes intertwined with localized vio­lence? And do the committees have the capacity to ad-
dress increasingly complex, dynamic systems of vio­lence? A recent report by the RESOLVE
network on the border conflicts in Niger, where the team had conducted extensive interviews,
notes a phenomenon that was impeding community resilience to vio­lence: the community
calls the condition psychosis and describes it as a community-­wide paralysis (part physical and
part ­mental) in the face of vio­lence, as conflict becomes too complex for communities to iden-
tify the entry points for its resolution.65
Another critical issue for considering the formation of, or partnering with, informal
LPCs to prevent or ­counter violent extremism is the princi­ple Do No Harm. The committees
­will need to protect against the risk of infiltration by violent-­extremist actors. In the Sahel,
Syria, and Af­ghan­i­stan, extremist groups target I4P to infiltrate and control communities.
The assassination of traditional governance leaders, community police, and civil society and
moderate religious leaders is the first line of any violent-­extremist offensive; violent-­
extremist groups understand their role in community resilience and local I4P. Communities
at risk for violent extremism have sometimes contributed to that risk with their extreme
marginalization of youth by traditional or local government leaders. (For example, youth in
the lowest castes of the Tuareg ethnic group’s caste system are joining violent-­extremist
groups in significant numbers in Mali and Niger.) Thus, LPCs in ­those communities might be

USIP.ORG | Preventing and Countering Violent Extremism | 35

 prone to elite capture or reflect community power dynamics, exacerbating the systems of
marginalization that already exist or experiencing significant po­liti­cal backlash if they ­were
to press for more inclusion. Fi­nally, informal, community-­centered LPCs are challenged by
the issue of scale and violent-­extremist mobility, especially in communities located on na-
tional or regional borders. A strong community peace response to violent extremism may
push ­these groups to neighboring communities or across borders, where they can become
stronger and return. Without a cohesive strategy, sustained support, and a basic infrastruc-
ture to ground their community efforts, informal LPCs would have ­little chance of long-­term
resilience.
What, then, are the conditions for engaging LPCs for community resilience to violent ex-
tremism? First, mapping the power and presence of LPCs in rural communities at risk of, but
not yet subject to, violent extremism can and should be incorporated into any prevention
strategy to determine how successful they are in working with governance actors; engaging
all groups in the community; and resolving everyday, historic, systemic, and electoral
vio­lence.
­These committees should be educated on the dynamics of violent extremism, how to
conduct an analy­sis of the community’s risk for violent extremism, and how to develop a pre-
vention strategy with community groups (especially ­those most likely to be affected by violent-­
extremist recruitment and operations) and with the local government and security actors. This
should include a Do No Harm analy­sis so that LPCs can understand the risk to themselves and
to their members should violent-­extremist groups increase their presence and operations lo-
cally. The committees should also be briefed on the contradictions between the country’s
counterterrorism and peacebuilding strategies so that they have agency and choice in navigat-
ing the contradictions between t­ hese conflicting agendas.
Where LPCs are connected to a national-­local I4P, such as around electoral vio­lence and
po­liti­cal reconciliation in ­Kenya and Côte d’Ivoire or as part of a peace pro­cess in Niger, it is
critical that leaders be briefed on ­these infrastructures, at both national and local levels, and
on how electoral vio­lence intersects with extremist vio­lence. For example, extremist groups
now exploit the po­liti­cal polarization around elections to disenfranchise Muslim voters from
secular states or piggyback on opposition critiques of the state to reinforce their recruitment
narratives on state corruption and predation as reasons for youth to join their extremist ranks.
Extremist groups also exploit historical grievances in communities that have been subject to
state discrimination and marginalization, such as par­tic­u­lar tribes in Burkina Faso and Niger.
I4P that have been put in place to address issues of historical and systematic exclusion should
be educated on how t­ hose issues play into extremist hands in order to advocate for prioritiza-
tion in terms of resources and capacity. This is not to say that ­these I4P should be harnessed
by P/CVE and counterterrorism strategies and programs. It is to say that supporting them to
fulfill their mandates ­will likely have P/CVE effects, and ­these should be monitored, analyzed,
and mea­sured.

36 | Preventing and Countering Violent Extremism | USIP.ORG

HYBRID GOVERNANCE
The Role of Hybrid Governance in Conflict Management
It is frequently noted that violent extremism spreads in regions where the administrative ap-
paratus of the state is weak or non­ex­is­tent. But the real­ity is much dif­fer­ent. Governance exists
in all regions, towns, and communities: if institutions of the state are not pre­sent, civil society,
armed groups, or traditional leaders assume governance roles, providing ser­vices, security
and justice, and rule of law. In many communities around the globe, it is traditional or
­customary leaders who interact with residuals of state administration locally to provide local
governance.
This relation between formal government, local state administrators (governors, civil ser-
vants, mayors), and traditional leaders, who derive their authority from the customs and tradi-
tions of a par­tic­u­lar area, has been a common form of governance throughout many areas of
the world, including Africa, Asia, Central Asia, the Near East and the ­Middle East. Thus, the
idea that violent-­extremist groups proliferate where the state is not pre­sent cannot be sup-
ported. Even the concept of hybridity, which implies a blending of formal state and informal
customary systems, is not entirely accurate. Hybridity is often a complex constellation of for-
mal and informal institutions that engage in the governance space, continuously vying for le-
gitimacy and power. This institutional multiplicity can lead to a continual and dynamic pro­cess
of reordering and renegotiation. Many violent-­extremist groups proliferate in areas where the
systems of hybrid governance have collapsed or weakened or where a constellation of institu-
tions engage in violent conflict to assert their power and legitimacy.
To assess the usefulness of hybrid governance in preventing the proliferation of violent-­
extremist groups, it is critical to understand how the deterioration of traditional governance as
a dynamic balancing mechanism has contributed to the spread of violent-­extremist groups.
Several ­factors explain the crisis of hybrid governance in the Sahel. In northern Mali, the au-
thority of customary leaders has rested on a symbiotic relationship with state authorities in
the administration of governance, especially since 2012, when the government devolved sig-
nificant power to local entities. When the state retreated following the rise in extremist and
armed group vio­lence and the failure of the peace accords, the authority, legitimacy, and ca-
pacity of customary leaders at first ­rose. They attempted to step into the governance gap to
adjudicate community access to critical resources, such as ­water and land, but only achieved
mixed results.
When a state administers ­these resources, they are perceived for the most part as a col-
lective good. When customary leaders associated with par­tic­u­lar tribes step into resource ad-
ministration, ­these resources are increasingly perceived as tribally owned. Thus, the legitimacy
of traditional leaders as equitable adjudicators of resources is not collectively understood
or exercised in real­ity. In regions where ethnic or tribal identity in the Sahel has not been
contested, resources have been managed peacefully. In regions where multiple tribes exist,

USIP.ORG | Preventing and Countering Violent Extremism | 37

 contestations over resources have further fractured tribal groups, leading to ever proliferating
conflict. In moments when the Malian state has intervened, the state has been perceived as
partial to colonial-­established tribal alliances, further exacerbating local resource conflicts.
As conflict has escalated, traditional leaders have had to govern in partnership with
armed groups—­state security forces, community-­based armed groups, local vigilante groups,
or violent extremists. Traditional leaders have discussed how the proliferation of armed
groups has greatly undercut their authority and legitimacy. As community grievances and con-
flicts have proliferated, violent-­extremist groups have entered the picture, further weakening
the governance and conflict-­resolution authority of traditional leaders. As violent-­extremist
groups have looked to consolidate local control, they have targeted traditional leaders in as-
sassination campaigns, especially ­those who have or are perceived to have collaborated with
the state. They have also recruited from tribal groups that have suffered the most from a
long-­standing colonial hangover: the state’s discriminatory, divide-­and-­conquer rule. Tribal
groups like the Fulani, Tuareg, and Peul, which have borne the brunt of state antagonism and
predation, have accepted jihadist rule as a protection against the state: “On the one hand,
villa­gers suffer jihadists’ constraints on religious practice and tradition. . . . B
​ ut some Peul in
par­tic­u­lar see the Islamic State affiliate as a necessary bulwark against a state that has preyed
upon them.”66
The other ­factor that has contributed to the breaking apart of hybrid governance, and its
capacity for balancing competing community interests, has been international actors. In the
Sahel, the UN peacekeeping mission, the involvement of international development actors,
and the regional counterterrorism intervention introduced incentive structures that shifted
community alliances in ways that have further undermined customary authority and its ability
to manage community conflict dynamics.67
Thus, the Sahel case prefaces the ways in which hybrid governance systems can work
in traditional, rural socie­ties where the state has not invested in a robust administration. First,
­there is often an implicit or explicit division of responsibilities between informal and formal
governance systems, which provide legitimacy to both. Weak local state administration is
bolstered by its interaction with informal governance systems in terms of information flow,
prioritization of citizen needs, ave­nues to push state policies to communities and provide ser­
vices, and so on. Customary leaders enjoy enhanced legitimacy, by being designated as inte-
gral to official governance, and enhanced status, owing to their ability to communicate and
deliver on community needs and interests. However, as seen in the Sahel, the mutual legitima-
cies and authorities are both weakened when the state is seen as a partial, conflict actor or when
traditional leaders use this relationship to extract benefits—in this case, security—­only for cer-
tain groups. Hybrid leadership is also weakened when the state or system of customary leader-
ship fails to balance the needs and interests of the multiplicity of institutions that inhabit the
governance space and hence becomes part of the competition for legitimacy and power.

38 | Preventing and Countering Violent Extremism | USIP.ORG

 A significant issue that must be raised in the Sahel regarding governance, in addition to
control of natu­ral resources, is land tenure. Traditional governance systems alone do not seem
to manage contested land tenure well, and access to land and dispossession from it is a major
driver of violent extremism, both in the Sahel and in other countries, such as Af­ghan­i­stan and
Iraq. The US Institute of Peace initiated programming in Af­ghan­i­stan to address this weakness:
customary leaders had the responsibility of deciding local land-­tenure issues that ­were then
registered in a state-­managed database to ensure fidelity to the ruling. In many instances, es-
pecially in conflict-­affected countries with significant numbers of displaced persons, the oral
rulings of customary leaders become lost or misremembered, igniting another round of local
conflict. While the issues in the Sahel around land tenure are both similar and dif­fer­ent (access
to communal lands by herder-­pastoralist communities, the shrinking of arable land owing to
climate change), strengthening hybrid governance capacity on resource management and
land tenure in at-­risk communities seems to be a prevention strategy for preventing and coun-
tering violent extremism.
A final insight from the Sahel regarding hybrid governance and its capacity to manage the
community conflict dynamics through which violent-­extremist groups infiltrate is that the in-
ternational community must have greater awareness of the impact of its activities and actions
on hybrid governance structures. Its overly securitized approaches contribute to the rise in
armed groups as communities self-­protect from now-­multiple conflicts—­international, state-­
local, and local. As armed actors become more power­ful, the legitimacy of customary leaders
declines. Where development work does occur, in many cases the funding has been captured
by elites, both in capitals and locally; ­these elites are then incentivized to respond to the inter-
ests of donors rather than t­ hose of citizens.68

Strengthening Systems of Hybrid Governance: Policy Impediments

The research and practice base on reconstituting or strengthening systems of hybrid gover-
nance to recover from or prevent vio­lence was substantial in the wake of the 1990s ethnic
conflicts and the nation-­building efforts in Iraq and Af­ghan­i­stan following the 9/11 attacks.
However, many of ­these studies and policies ­were not without significant institutional bias.
First, the state-­building interventions in Af­ghan­i­stan and Iraq ­were primarily focused on
­extending formal state administrative systems into rural communities or “mimicking” custom-
ary structures and integrating them into national governance structures, such as the National
Solidarity Program in Af­ghan­is­ tan. Preserving, strengthening, and integrating customary sys-
tems is often more of a puzzle and rarely a governance priority of international institutions,
donors, or the development arms of other national governments. State-­centric institutions are
wired to fund proj­ects that support Western conceptions of the state. Second, customary in-
stitutions are generally exclusive, patriarchal systems that do not mesh with the demo­cratic
institutions, norms, and pro­cesses that many governance donors prefer. They are often seen

USIP.ORG | Preventing and Countering Violent Extremism | 39

 as an impediment to demo­cratic governance rather than institutions that can be reformed or
engaged with. Governments in fragile states can view traditional and customary systems with
some suspicion—as competitors for citizen loyalties, impediments to state rule, and historical
enemies of the national government, especially if that government is dominated by a par­tic­u­
lar ethnic or tribal group. Fi­nally, in the area of peace pro­cesses, the field of hybridity discusses
local and customary actors as ­those who subvert or bastardize national peacebuilding efforts
to maintain local power and control over local conflict dynamics.
The focus of strengthening hybrid governance to manage vio­lence and serve citizen in-
terests should not be on forging institutional relationships and capacities but on understand-
ing the po­liti­cal outcomes of such governance. In some cases, ­those po­liti­cal outcomes ­will
build and strengthen community resilience to vio­lence and violent extremism. In other cases,
the po­liti­cal machinations of hybrid systems ­will reinforce dark social capital, resulting in sys-
temic vio­lence against groups and citizens—­the vulnerability that extremist groups exploit.

Hybrid Systems of Land Management

In the early years of the twenty-­first ­century, many African countries, with support from inter-
national financial institutions such as the World Bank, embarked on rural land reform, recog-
nizing that competition over land was becoming extremely fierce and even violent and that
customary systems ­were ill equipped to resolve the growing competitions.69 The weaknesses
in customary leaders’ management of land tenure w ­ ere multifold and included the following:

• increasing demand for natu­ral resources;

• increasingly massive relocations of populations seeking land, such as the flight of livestock
producers owing to drought and resource degradation, the return of emigrants from
neighboring conflict-­affected countries, the pursuit of new agricultural pioneering ven-
tures such as cotton production and government-­sponsored development proj­ects, and
the eradication of river blindness;70
• tension between customary and state land tenure rulings; and
• an independence-­era national ­legal system that sought to eliminate customary leaders’
authority on land tenure.

The reforms ­were also billed as an opportunity to create economic capital and an entre-
preneurial base in rural communities by creating a market in land. Much of the land reform
efforts ­were based on neoliberal concepts of the privatization and securitization of land owner­
ship in order to drive economic development and growth. ­There was also recognition that
previous attempts to eradicate customary land management systems had been counterpro-
ductive. Thus, ­there was a significant nod to recognizing and integrating rural customary
systems, such as the documentation and registration of customary land users; assigning
land rights to customary users and collectives; and physically surveying bound­aries to make

40 | Preventing and Countering Violent Extremism | USIP.ORG

customary tenure clear.71 In the effort to create a rural land market, many countries in Africa
attempted to deregulate land sales; in cases where customary systems dominated, this of-
ten meant the state devolving ­legal authority to local entities, as was the case in Burkina
Faso.72 ­There was also a simultaneous reregulation of land as countries sought to register cus-
tomary decisions on land owner­ship in state-­run databases to provide greater security to land
markets.73
­These blended, hybridized land management reforms, which ­were rooted in neoliberal
traditions but incorporated customary system realities, have had some perverse effects that
may have some relationship to the spread of violent extremism—­although this relationship
must be researched and tested. The first unintended consequence of ­these reforms is that
they have actually made it easier for wealthy urbanites, international corporations, and investors
to acquire communal lands ­because their bound­aries are clearly delineated, as are the terms
of owner­ship. The issue of the sale of ­these large-­scale landholdings to investors and corpo-
rations must be examined as a potential driver of violent extremism. With the insertion of the
state into local land tenure issues and the integration of customary leaders into state sys-
tems, clashes have occurred locally between national elites and their brokers at the local
level.74 Wealthy, nationally po­liti­cally connected individuals at the local level have used the
hybrid system to challenge the authority of customary leaders ­there. ­These local “hot shots”
pretend to represent national po­liti­cal interests to adjudicate land deals, while lining their
pockets.
One such issue is the sale of large-­scale communal and state landholdings to interna-
tional and national investors with the promise of creating local jobs, stimulating the local econ-
omy, and providing access to new agricultural technologies. Some research suggests that ­these
large land sales have numerous negative effects: increased local food insecurity, environmen-
tal degradation, inadequate compensation for the land sale, and increased land conflicts.75
While the link to large-­scale sales of land has not been directly studied, several studies on
community dynamics and violent extremism have pinpointed communal land sales by po­liti­cal
officials and traditional leaders—­and the immediate ­triple dispossession of youth from com-
munity, employment, and status—as a key driver of youth membership in violent-­extremist
groups.76 This is an area that needs to be further researched and studied as a critical hybrid-­
governance P/CVE issue.
As ­these new, hybridized land tenure systems have taken hold, a critical issue has been
the shifting structure of landholdings. A hope of the new rural land market system had been a
bottom-up consolidation of landholdings as some rural farmers increased their economic
wealth. Instead, the system has generally resulted in increasingly smaller rural farmer land-
holdings, as the competition for arable land grows and urban investors take advantage of the
land-­titling system to acquire rural lands. In real­ity, the idea that better resourced landowners
would buy land from the unproductive rural poor (thus creating capital for the poor and po-
tentially jobs) has in many cases not occurred. Poor ­people have sold at prices lower than the

USIP.ORG | Preventing and Countering Violent Extremism | 41

 value of their land, as a result of distress sales; and land-­grabbing local elites, working with
state officials, have taken poor ­people’s lands in areas where customary tenure does not
­offer statutory protection.77 As the new land reforms dispossess the poorest of the poor
­either through corruption or through market forces, the link between further marginaliza-
tion of the most marginalized in rural communities and the risk for violent extremism should
be studied.
While the links between land tenure, po­liti­cal corruption, marginalization, and youth dis-
possession and violent extremism have been discussed anecdotally by communities in key in­
for­mant interviews and focus group discussions, this issue needs to be researched further. The
relationships intuitively make sense, and if they are proved and validated, they introduce a
cautionary note regarding hybrid governance, not as an approach for P/CVE, but as a driver of
it. In any case, the current hybridization of land reform has introduced a corrupt po­liti­cal class
into local rural communities and has caused a further impoverishment of some rural poor and
potentially their dispossession from the land. In other words, in terms of community resilience
to vio­lence and violent extremism, it has introduced or strengthened systems of dark capital
in some rural communities, even as it has introduced a market in rural land that is at times
more transparent and secure—­factors that are critical to reducing local strug­gles over land
owner­ship.

Hybrid Security Systems

Local, customary security organ­izations have long been a presence in African communities.
Fostered and supported by colonial administrations as part of their dividing and conquering of
colonies by elevating certain ethnic groups over ­others as a mechanism of po­liti­cal control,
ethnic militias and vigilante groups ­were used in conjunction with colonial security forces to
monitor and put down other ethnic groups that ­were being excluded by the state.78 The Ma-
lian state has used a similar system of ethnic alliances, hybrid governance, and community-­
based armed groups in Central Mali by enlisting the Dan Nam Ambassagou, an ethnic militia
made up primarily of non-­Muslim Dogon fighters, to assist its security forces in battling ex-
tremist groups and their perceived local Fulani tribal associates. The result in Central Mali has
been the proliferation of armed and violent-­extremist groups and a significant deterioration in
the resilience of local communities and their ability to manage localized conflicts.79
Thus, the ­Kenyan government’s incorporation of customary security norms and institu-
tions into a nationwide effort to prevent and c­ ounter violent extremism is an informative case
for this study. The government of ­Kenya established the Nyumba Kumi initiative ­after the
Westgate Mall terrorist attacks. It is a community policing model based on Tanzania’s custom-
ary policing, which itself is based on Tanzanian communal values and the African ubuntu phi-
losophy of social connectedness.80 The premise ­behind Nyumba Kumi is that neighborhoods
(in this case, or­ga­nized in units of ten blocks) share a common interest in a safe and prosperous

42 | Preventing and Countering Violent Extremism | USIP.ORG

community and that, if or­ga­nized, community members ­will work with one another to get to
know neighbors, monitor activities in their neighborhood, and report any suspicious activities to
local and customary leaders.81 The intent of this strategy was also to improve community rela-
tions with the police, who ­were generally seen as predatory and abusive actors. Local leaders
could report community “intelligence” to security officials without exposing members to police
harassment and abuse. With this buffer in place, police would receive better intelligence and
perhaps grow to trust and improve relations with communities, as they began to perceive them
as partners in countering crime and terrorism. In form, Nyumba Kumi has all the characteristics
of a resilient system: engaging horizontally across community lines and vertically with traditional
and security leaders and having the capacity to manage predatory governance actors.
Both the Tanzania and ­Kenya cases are insightful ­because they are instances where the
state has attempted to harness or integrate customary systems and norms into national policy.
Communities in Tanzania have long engaged in a customary local policing practice called sun-
gusungu vigilantism, whereby organ­izations ­were founded by communities in the 1980s to
stem the rising rates of violent ­cattle raiding to which the state was not adequately respond-
ing. The sungusungu ­were governed by a group of village-­level leaders publicly selected by the
community and funded by villa­gers with a small remuneration in ­either cash or food. All able-­
bodied men ­were required to arm themselves to participate in defending the village in a
raid. The sungusungu system also administered justice: ­cattle raiders ­were whipped and tor-
tured.82 The ruling party of Tanzania, the Chama Cha Mapinduzi, recognized the practical ben-
efits of this growing crime prevention movement. In addition, it was extremely complementary
to the party’s ideological support for socialism and grassroots mobilization.
The Chama Cha Mapinduzi promoted and incorporated the sungusungu with local ad-
ministration and designated them an official organ­ization of the state in 1989, with the amend-
ment of the ­People’s Militia Act to give the sungusungu the equivalent powers of a peace
constable. Their authorities and responsibilities expanded ­under the Chama Cha Mapinduzi to
include state administration functions such as collecting taxes and enforcing citizen partici-
pation in volunteer nation-­building proj­ects.83 When the party’s fortunes diminished, and
multiparty rule was instituted in Tanzania in 1992, the power of the sungusungu in urban
communities declined, although their presence in rural communities, which did not depend
on po­liti­cal party patronage, remained, as did their cultural prominence in Tanzanian society.
In 2006, the Tanzanian government introduced a new community policing strategy called
ulinzi shirikishi (collaborative security), whereby citizens ­were required to form security com-
mittees that would conduct night patrols, investigate reported crimes, and resolve disputes.
With this new form of collaborative security, the government intended to draw on society’s
cultural comfort with community policing structures such as the sungusungu, although in both
form and substance ­these reforms ­were tangibly dif­fer­ent. A first-­order difference was that the
ulinzi shirikishi ­were to collaborate with police, while the sungusungu had long operated in­de­
pen­dently from the police. Their relationship was largely antagonistic.84

USIP.ORG | Preventing and Countering Violent Extremism | 43

 A research study of the ulinzi shirikishi system in Tanzania that looks at the po­liti­cal out-
comes from the blending of customary and official police security institutions indicates why it
was problematic. First, a reliance on customary, community-­based systems should not be the
cheap alternative to the provision of security. Even customary systems require resourcing,
which in the system instituted in Tanzania fell on the shoulders of community members. In
fact, the largest burden fell on the poorest of citizens, who, if they could not provide remu-
neration and payment, ­were required to serve greater amounts of time. Second, community-­
based systems suffer from bad governance as well. In many communities in Tanzania, ­there
was ­little accounting for the funds that ­were spent on security. And when security becomes
commercialized, wealthier residents, who can pay more for additional ser­vices, receive the
bulk of security rather than their poorer counter­parts, who are often more at risk of being
victims of a crime.85
While modeled on the Tanzania community-­based policing system, ­there ­were prob­lems
grafting a system from another country and culture into the ­Kenya context. The first issue was
related to constitutionality. ­Kenya’s criminal code contains stipulations for citizen arrests that
­were assumed in Nyumba Kumi, leading frequently to mob justice or the abuse of suspected
criminals, especially in the absence of alternative dispute resolution mechanisms. The second
issue was the lack of guidelines for establishing community security systems ­under Nyumba
Kumi, leading to a confusion of roles and responsibilities and a lack of transparency and
oversight. Often, Nyumba Kumi organ­izations ­were the victims of elite capture by community
members who saw them as a source of revenue or by local criminals who sought to dominate
them. Fi­nally, Nyumba Kumi organ­izations tended to exclude youth, even though they are
most frequently the victim of crimes or engaged in criminal activity. Often, the exclusion was
at the hands of community elders. Community polling demonstrated support for greater youth
involvement to give youth a source of employment, to include them in community decisions
on security that affect them, and to give them more leadership roles and responsibilities.86
Thus, in the absence of state guidelines for implementing Nyumba Kumi, the local systems of
dark capital asserted themselves in many community initiatives, including the systematic dis-
crimination against youth and elite capture of community resources and assets.

Hybridity
In the early years of this ­century, the World Bank conducted a series of major studies on cus-
tomary justice systems, investigating how they interacted with formal rule-­of-­law systems and
what more could be done to strengthen the legitimacy and complementarity of both. No such
major study has been conducted on traditional, informal governance and the state: how the
two interact, where each might have more legitimacy, and how hybrid systems succeed and
how they fail. Yet for prac­ti­tion­ers involved in preventing and countering violent extremism,
the issue of hybridity is critical, as violent-­extremist groups proliferate in areas where this
structure of governance is the norm. Hybrid governance space is best defined not as the

44 | Preventing and Countering Violent Extremism | USIP.ORG

juncture between formal and informal systems but as a place of governance fluidity, where
state and customary actors and institutions create governance linkages and straddle multiple
bound­aries. Their negotiated relationships fill governance gaps in good and bad ways, some-
times driving ­toward good governance outcomes and sometimes promoting their parochial
interests. While good at preventing conflict by balancing a multitude of actor and group inter-
ests, “­these ­orders are not structured to manage complex emergencies” that require collec-
tive, sustained response.87 Thus, strengthening hybrid governance may be best suited for
preventing violent extremism and communities at pos­si­ble or imminent risk for violent ex-
tremism. In terms of countering violent extremism and hybrid governance, the focus must be
on the princi­ple of Do No Harm, recognizing that violent shocks, such as extremist group op-
erations and international responses to them, may quickly unravel the negotiated relation-
ships that underpin hybrid governance systems.
The research cases for this section focus on the Sahel and the Somali-­Kenyan border,
where violent groups increasingly dominated the governance landscape but where, in the
case of ­Kenya, hybrid governance succeeded in preventing vio­lence. In the other case, the
Sahel, hybrid governance structures fell like dominoes. In between both cases lie power­ful les-
sons learned regarding how hybrid governance models can prevent vio­lence, but as such vio­
lence increases, they contribute to the “unraveling.” Ken Menkhaus’s research on the mediated
state and the peacebuilding efforts of ­women in Wajir, ­Kenya, is a seminal study on hybridity
and peace.88 In the early 1990s, as land and resources became scarce, three Somali clans that
inhabited Wajir began to increasingly compete for economic and po­liti­cal dominance. During
the 1992 elections, each clan felt threatened by the outcome, anticipating a loss of status and
resources should the other clan win. With so much at stake, vio­lence around the elections in
Wajir was significant.
A group of ­women in Wajir intervened to stop the postelection vio­lence and formed a
peace committee that grew to include members of the local professional class. Together they
convened a meeting of clan leaders who negotiated and committed to a set of princi­ples to gov-
ern peaceful relationships among them. Local businesspeople then raised money to fund com-
munity peace activities, and a new district commissioner enjoined the peace committee and
local leaders to establish an early-­warning network to stop vio­lence before it happened or esca-
lated. This meant resolving everyday disputes and punishing local crimes, using customary So-
mali practice rather than formal ­Kenyan rule-­of-­law systems. The entire set of relationships was
formalized in the establishment of the Wajir Peace and Development Committee, which was
chaired by the district commissioner and formalized government-­civic collaboration in support
of peace and nonviolence. Three impor­tant lessons regarding hybrid governance and peace can
be taken from this experience: such efforts must find ways to bridge competition between civil
society and traditional leadership to support local peace efforts, to act quickly to resolve local
disputes and committed crimes before they escalate, and to cement civic-­governance partner-
ships to prevent vio­lence, recognizing that neither entity can go it alone.

USIP.ORG | Preventing and Countering Violent Extremism | 45

 T­ here are several other insights that Menkhaus does not note but which are relevant to
both Wajir and the Sahel. First, the state did not attempt to undercut local, grassroots peace
efforts but instead joined and further legitimated them. Often in moments of shock and crisis,
state actors in hybrid systems ­will inject themselves as conflict actors to assert their power and
control, further exacerbating the crisis, as has been seen in the Sahel as the Malian state has
sided with dif­fer­ent armed factions. Second, the Wajir peace committee recognized that con-
flict fundamentally changes local power dynamics and transforms incentives for peace and
vio­lence. The previous balance of interests that had preserved the peace, and on which hybrid
systems depend, was gone. The politics of hybrid governance needed to be immediately rene-
gotiated to establish new balancing mechanisms, such as the 1993 Al Fatah peace declaration
negotiated by the clans. Third, once vio­lence erupted, a culture of peace needed to be rees-
tablished through concerted efforts to bring the community together around peace activities,
dialogue, and so on.
How rapidly hybrid systems can unravel in the face of vio­lence—as evident in the case of
the Sahel—is quite instructive and an argument for strong, intense interventions to prevent
violent extremism by strengthening hybrid governance systems in at-­risk regions. As armed
groups proliferate in the region, traditional leaders at the local level are rapidly losing author-
ity and legitimacy owing to the withdrawal of state institutions that formalized and supported
their governance authorities. They are ­under increased pressure by external groups to gain
access to their territory and by internal groups to provide them access to resources; the mi-
crodynamics of fragmentation is leading to increased intercommunal vio­lence. This intercom-
munal breaking apart is allowing violent-­extremist groups to create ethnic alliances with
certain communities, which the state counteracts with its own alliances, and thus the frag-
mentation spreads at the intermediate level. The international intervention to prop up the
Malian state has led to counteractions by jihadist groups to seize more and more territory.89
Without the glue of institutions, norms, and pro­cesses, ­these systems of governance, based
on tactical negotiations and dynamic interplay, cannot reconstitute their po­liti­cal alliances in
the face of rapidly escalating vio­lence, and the incentive structures of hybrid actors are thereby
fundamentally changed. Their propensity to balance po­liti­cal relationships is now based on a
new calculus of vio­lence. ­There is no institutional drag to prevent this rapid spiraling and
fragmentation.

Establishing a Research and Practice Base: What Is the Best

Governance Model for the Prevention of Violent Extremism?

The P/CVE field has strug­gled with the issue of governance and what model is appropriate for
managing the proliferation of violent-­extremist groups, especially in gray areas where the state
is minimally pre­sent or largely absent. The material that follows examines a number of recent

46 | Preventing and Countering Violent Extremism | USIP.ORG

studies of the Sahel—­conference reports, research, programming summaries—­that address
this issue, and the recommendations are distinctly mixed. Some studies recommend hybrid
governance structures where customary systems, in fact, play a prominent role in reconstitut-
ing the state. ­Others are distinctly statist and put forth a model where a “healthy” state (that
is, one guided by demo­cratic princi­ples) is established in rural communities with a supporting
architecture that is inclusive, transparent, and accountable. Fi­nally, a research study con-
ducted by a local organ­ization in Cameroon recommends po­liti­cal decentralization as the
preferred governance approach for preventing violent extremism. We may be drawing more
distinctions in this section than the research warrants, but ­these studies do raise an impor­tant
issue that must be addressed regarding governance and the prevention and countering of vio-
lent extremism: If poor governance is a key driver of violent extremism, is the issue structural
or qualitative? That is, does the structure of governance ­matter for reestablishing the legiti-
macy of governance in regions suffering from upticks in vio­lence and violent extremism, or is
it a m
­ atter of improving the ability of the system to govern regardless of that system?
With the proliferation of violent-­extremist groups in the Sahel, several regional confer-
ences have been or­ga­nized to look at the breakdown in governance in rural communities in
Mali, Niger, and Burkina Faso. ­These exchanges have been impressive in their scale, participa-
tion, and expertise. The first, a 2017 regional meeting or­ga­nized by the International Institute
for Peace, the UN, and the Swiss Federal Department for Federal Foreign Affairs in N’Djamena,
included 100 participants from northern, central, and western Africa looking specifically at is-
sues of governance and rising vio­lence in the Lake Chad Basin.90 Participants broke into three
working groups to assess state-­citizen relations, po­liti­cal participation, and inclusive dialogue.
The crosscutting issue was clearly the necessity to control and manage external interventions
to prevent them from undercutting the development of local and sustainable solutions to pre-
venting vio­lence, including how to balance the plethora of state and international security,
po­liti­cal, socioeconomic, and development responses and the linkages between them at the
local, national, and international levels.

THE HYBRID GOVERNANCE MODEL

The recommendations called for a central role to be played by traditional leaders—­first and
foremost, that state, regional, and international actors should consolidate customary gover-
nance structures where they have local legitimacy. According to participants, the state must
play a leading role in incentivizing ­these empowered traditional governance structures to exer-
cise greater inclusivity in communities by creating national and local frameworks for the par-
ticipation and funding of youth and ­women and by promoting initiatives by youth and ­women
within ­these nationally led efforts. This approach should, above all, prioritize and ­favor local
solutions to local conflict dynamics through the dialogue, public-­private partnerships, and
community coexistence facilitated by traditional ethnic and religious leaders. The issue of

USIP.ORG | Preventing and Countering Violent Extremism | 47

 v­ iable local governance structures was prioritized by participants over the current excessively
securitized approaches to vio­lence in the Lake Chad Basin. Participants demanded that secu-
rity forces, deployed locally, reflect the ethnic and religious makeup of the communities in
which they serve and that security actors no longer be allowed to provide humanitarian as-
sistance in conflict-­affected areas, which politicizes assistance. Instead, civil society should be
the primary deliverers of humanitarian aid, even in areas with high levels of vio­lence. Regional
organ­izations should develop a standard set of princi­ples and training on ­human rights and
international humanitarian law for all security ser­vices so that they operate in the same way
according to t­ hese norms.
The call for the strengthening of hybrid structures of governance was also reinforced in a
research study by the Clingendael Institute and a development approach implemented by the
UN Development Programme in fifteen countries in Africa from 2016 to 2019. In a research
study on hybrid governance systems and conflict resolution in Mali (Kidal and Ménaka), Niger
(Tahoua region), and Libya (Fezzan), Clingendael researchers conducted 323 in-­depth inter-
views and thirty-­four discussion groups with citizens, customary leaders, and local and state
governance actors, asking how customary leaders maintained their legitimacy while engaged
in local governance in highly fragile contexts where state presence is ­limited.91 The results of
the interviews ­were then briefed to local research teams, academic experts, nongovernmental
organ­izations, and members of the international community. The recommendations from this
study ­were based on the research finding that customary leaders in ­these regions ­were largely
seen as legitimate representatives of state institutions.
Given that legitimacy, customary leaders should be engaged by national and interna-
tional actors as programmatic allies for reinforcing governance, stability, and security in areas
with ­limited state presence. Their customs should be codified and standardized to solidify and
make clear the relationships among customary leader authorities, governance, and judicial
decisions. This would improve local perceptions of governance and rule-­of-­law objectivity, and
it would also make pos­si­ble a comparison of local customary law and international norms of
­human rights. Customary leaders should also receive training to ensure their neutrality and
objectivity in their governance and judicial decisions. To further enhance the leaders’ standing
as governance actors, the Clingendael study recommends holding a region-wide dialogue
on the place of customary leaders in con­temporary demo­cratic governance systems in the
region.
The study notes that most customary leaders form alliances with armed groups not for
ideological or po­liti­cal reasons but out of survival. International and national actors should not
assume that ­these alliances subvert the ability of customary leaders to mediate local conflicts
with the goal of achieving a base level of stability that can then be supported in other “devel-
opment” ways. Customary leaders should be encouraged to engage in local mediation to help
prevent the spread and escalation of local conflicts before they become broader ethnic con-
flicts or are co-­opted by violent-­extremist groups.92 The UN Development Programme’s efforts

48 | Preventing and Countering Violent Extremism | USIP.ORG

have also reinforced a central role for customary leaders in a multiyear, multicountry, multilevel
effort to prevent violent extremism. The programming divided its fifteen beneficiary countries
into dif­fer­ent categories of risk: violent-­extremism epicenters (Libya, Mali, Nigeria, and Somalia);
secondary countries (Cameroon, ­Kenya, Mauritania, Niger, Chad, and Tunisia); and countries at
risk (Morocco, Uganda, the Central African Republic, Sudan, and Tanzania). The programming
was implemented in six areas—­socio­economics, law and security, demobilization and reintegra-
tion, media and technology, community resilience, and gender initiatives—at the regional level,
working with the African Union, and also on national and local levels. At the local level, work with
traditional leaders was seen as a critical ­factor in reinforcing the linkages within communities
that reinforced their resilience. In certain contexts, the traditional leaders ­were asked to estab-
lish a committee of elders to conduct this resilience work.93

THE STATIZATION MODEL: STRENGTHENING THE

SOCIAL CONTRACT
A second conference—­sponsored by the government of Cameroon; the African Center for In-
ternational, Diplomatic, Economic, and Strategic Studies; the Swiss Federal Department of
Foreign Affairs; the UN Regional Office for Central Africa; and the UN Regional Office for Cen-
tral Western Africa and the Sahel—­was held in Yaoundé on November 27 and 28, 2017.94 The
meeting included representatives of po­liti­cal, civil society, and international organ­izations
from the Lake Chad Basin and western and central Africa. The recommendations coming out
of this meeting also focused primarily on governance and the role of the state, encouraging
countries to adopt the prevention of violent extremism as an approach that could support
improved governance and conflict-­resolution outcomes. However, participants suggested that
state authority, and not traditional and customary leadership, should be reasserted in rural,
marginalized, and abandoned regions—­the strategic goal being to establish a new rapport
between the state and citizenry based on the paradigm of the utility of the state. It was critical
that the reinstatement of state administration in rural communities be reinforced through the
establishment of a governance architecture at the national, regional, and local levels that pro-
vided venues for citizens to participate in state decision-­making on issues that impact them.95
Participants also agreed that the state, in the pro­cess of reestablishing its authority in
­these areas, needed to federalize and support any and all citizen initiatives that promoted col-
lective peace and national belonging as a broad-­based attempt to restore the state-­citizen so-
cial contract. Fi­nally, participants believed that economic and peacebuilding programming
should support this restatization, including a more fair and equitable distribution of wealth;
education reform to include values of peace and nonviolence; and local dialogue among state
officials, security institutions, civil society, and citizens. That the role of hybrid governance as a
P/CVE ­factor was contested between the two gatherings of regional elite suggests that for
some regions and communities, restatization was the preferred outcome.

USIP.ORG | Preventing and Countering Violent Extremism | 49

 The gathering of officials in Yaoundé also emphasized the role that prevention of violent
extremism could play in creating a unifying strategy for local and national governments, secu-
rity forces, media, civil society, and regional organ­izations, working from a robust research
agenda that would mobilize researchers and experts to deliver usable empirically based
studies and recommendations of strategic value to decision makers.96 This would mean sup-
porting and sustaining countries’ research centers and think tanks to focus their work on un-
derstanding the deep-­rooted ­causes of violent extremism, initiating studies on the roles of
­women in violent extremism, and proposing policy and programmatic options based on that
research. The participants recommended that the analytical frameworks on preventing vio-
lent extremism, developed as a result of this research, be used to structure exchanges on
preventive efforts among dif­fer­ent relevant stakeholders. They suggested that civil society
and media cohorts at all levels—­national, regional, and local—be trained on the ­causes and
promising remedies of violent extremism and that civil society and media be mandated, in
turn, to educate the general public in local languages on how to diagnose and peaceably
­respond to threats of violent extremism. This idea of prevention as a strategic organ­izing
framework extended to regional organ­izations as well: Conference participants demanded a
regional strategy for sharing information and research and communicating on preventive
tactics to pool knowledge, experience, and approaches, with an aim of reinforcing the ca-
pacities of dif­fer­ent actors in addressing violent extremism. Participants also took a much
harder line on the issue of security force involvement, demanding the condemnation of
­human rights abuses by the security sector and their adherence to a strategy of preventing
violent extremism.

THE DECENTRALIZATION MODEL

Yet another research study, by Mvondo Hervé, has recommended that decentralization is
the appropriate response to violent extremism.97 In fact, he asserts that the participation of
populations in the management of local affairs should be the first component of a P/CVE re-
sponse. Communities most at risk for violent extremism have generally been pushed aside by
governance actors; reinforcing the communities’ own local governance capacity, especially
when accompanied with economic development support, has resulted in their solidarity with
national efforts to eradicate violent extremism. This research shows that communities with
decentralized governance authorities are more likely to engage violent-­extremist groups and
to report violent-­extremist supporters to local governance and security officials.
However, to be effective in the prevention of violent extremism, ­these decentralized po­
liti­cal functions must be inclusive and demo­cratic, and they must reinforce demo­cratic pro­
cesses and norms. To bolster local, decentralized demo­cratic governance, the state must play
a role. It must commit to the full disclosure of information on violent-­extremist threats posed
to a community and the mea­sures being taken by the state to combat them. The state must

50 | Preventing and Countering Violent Extremism | USIP.ORG

also guarantee the right of local communities to self-­government while, at the same time,
helping ­these local governments, which are ­under the strong influence of violent-­extremist
groups, to provide basic ser­vices. This would have an immediate impact of lowering youth re-
cruitment levels, as would the expansion of youth programming to include employment
opportunities.

EVIDENCE
We recently evaluated the quality of research and monitoring and evaluation in major fields
of P/CVE practice and found that the field overall suffers from a deficit of evidence to sup-
port its youth, gender, and religious engagement programming. This was not the case with
research and evidence around community resilience practice, which was of a slightly higher
quality than other sectors. That said, to better support communities on the frontlines of vio-
lent extremism, or at significant risk, we recommend funding a “next generation” of re-
search that (1) includes a cross-­study comparison to identify and validate a set of community
resilience capacities across cultural and conflict contexts that ­will set the stage for a robust
preventing violent extremism practice; (2) studies the interplay between community resil-
ience capacities and external interventions to determine if and how they can strengthen
­those capacities without ­doing harm or delegitimizing them; (3) identifies and learns from
failed community strategies to resist or prevent extremist recruitment and vio­lence; (4) ex-
amines the inter­active effects of predatory, corrupt governance and community resilience
to vio­lence; and (5) explores the types and configurations of community social capital (over-
lapping, bridging, and bonding) and their enactment (trust, collaboration, collective action,
coordination) that contribute specifically to community resilience to violent extremism, as
opposed to ethnic, communal, or sectarian vio­lence. This last point is particularly impor­tant.
The bulk of research on social capital and vio­lence focuses on ethnic vio­lence and how elites
mobilized “bonded” groups, and how the absence of conflict was often explained by “bridg-
ing” capital (that is, organ­izations that could mediate between ethnic or religious groups to
prevent or de-­escalate vio­lence). The social capital networks around violent extremism look
very dif­fer­ent in terms of the conflict lines extremists exploit and the interactive networks
that communities mobilize for peace. Understanding ­these dif­fer­ent social dynamics around
extremist vio­lence and validating new, evidence-­based social capital models is critical to
advancing community resilience programming. Fi­nally, in embarking on a robust research
agenda around community resilience, a first step should be convening prac­ti­tion­ers working
on communities to reduce extremist recruitment and risk. All of our interviewees empha-
sized that while ­there is a troublesome lack of evidence in the field of P/CVE, prac­ti­tion­ers,
who rarely have time to reflect and write on programming successes, “know what works”
and what has failed. Gathering that knowledge could establish a solid evidence base and
generate a new set of research questions.

USIP.ORG | Preventing and Countering Violent Extremism | 51

 PROGRAMMING
The fields of counterterrorism and countering violent extremism, largely driven by national
security and ideological imperatives, have ignored some hard lessons learned by the peace-­
building, development, and governance fields around playing into local conflict dynamics
and even sometimes escalating and broadening vio­lence. ­Earlier approaches to engaging
communities have instrumentalized or scapegoated them, increasing the risk of violent ex-
tremism, inviting heavy-­handed state interventions, and exacerbating existing patterns of
marginalization and exclusion or structural vio­lence. Any new approach to community engage-
ment, even one intended for good to build resilience, must incorporate contextual and conflict
analy­sis and Do No Harm princi­ples to avoid the worst practices of the past and to not invite
new forms of vio­lence and conflict. This means analyzing the impact of well-­intended actions
and policies on communities, especially counterterrorism operations by the state and interna-
tional actors, as well as how they are affecting community resilience capacities and strategies
and the systems of vio­lence that exacerbate local conflicts and grievances that drive commu-
nity and individual support for violent extremist groups. Once ­these conflict dynamics are un-
derstood, programs must be continuously adapted to, above all, preserve community capacities
and strategies to prevent and c­ ounter violent extremism.
Program implementers must commit to support for and integration of better research
and evidence, especially around understanding how communities mobilize and respond to vi-
olent extremism. Promising research has been done around informal security and early warn-
ing systems, community strategy development, and the characteristics of successful bridging
organ­izations; this research must be expanded on and the findings must be incorporated into
community resilience programming. Many of the major programmatic lines of effort around
P/CVE, such as youth, gender, and religious engagement, have employed new thinking and evi-
dence that support and strengthen community resilience approaches. ­These efforts have ex-
plored positive youth development and socioecological strengthening; ­women’s community
roles and how they interact with extremist recruitment, operations, and group cohesion; and
religious leaders’ diverse community roles and the prevention of vio­lence. Again, supporting
a robust research agenda to accompany community resilience programming does not have to
mean committing exorbitant amounts of funding; small experiments could be built into pro-
gramming to test new approaches and strengthen analytical frameworks. Carefully adopting
and implementing approaches from other fields, such as the peacebuilding community’s LPCs
and the governance field’s hybrid governance programs, could support promising ave­nues for
building community resilience in new conflict contexts.

POLICY
A focus on building community resilience to extremist vio­lence and recruitment, however, re-
quires the P/CVE field to abandon and ­counter disproved ideologically based paradigms that

52 | Preventing and Countering Violent Extremism | USIP.ORG

reinforce community and individual deviance framing. The scapegoating of individuals, groups,
and/or communities, which continues to this day, allows predatory and corrupt state actors to
maintain their control by manipulating long-­standing historical divisions in society that have
marginalized and discriminated against them and justified state vio­lence historically and ­under
newly ­adopted national counterterrorism strategies often legitimated by international actors.
A strong antidote to state-­enforced community fragility is to center P/CVE programs in peace-
building and democracy and governance and to adapt counterterrorism and P/CVE policies
accordingly. Po­liti­cal and social transformation should be the focus of P/CVE policy and not
security imperatives. Knowing that local actors adapt governance and peacebuilding interven-
tions to secure their own po­liti­cal power and statues—­strategies of hybridity—­all community
resilience programming must keep an eye on how efforts to build bridging capital among
groups and create informal security and early warning networks might be ripe for elite cap-
ture. Policies on building community resilience must be adaptive and take into account the
very real possibility of co-­optation and capture by elites and the networks of dark capital that
continue to exist in communities. A key ­factor established in research and evaluations of com-
munity resilience programming is vertical relations with the state in addition to horizontal
capital across community social groups. Policy frameworks on community resilience and coun-
tering violent extremism must include governance at their very core; they must focus on
strengthening hybrid governance systems, establishing better connections between communi-
ties and national-­level institutions, and combatting po­liti­cal corruption. Without a focus on the
role of the state in driving violent extremism, community resilience programming could easily
revert to the old policies of scapegoating communities or framing them as suspect and morally
delinquent, a framing that often results in approaches that further degrade communities’ re-
silience and increase their risk to violent extremism.

USIP.ORG | Preventing and Countering Violent Extremism | 53

Appendix A

INDICATORS OF RESEARCH QUALITY

Variable Description

Explanatory variable: Relevance How relevant is the explanatory variable for P/CVE work?

Explanatory variable: Actionability How actionable is this explanatory variable for P/CVE
work?

Explained variable: Type Is the explained variable based on observations or

responses? If responses, are ­there efforts to mitigate
response bias?

Explained variable: Relevance How relevant is the explained variable to P/CVE?

Internal validity Does the piece adopt methods to better determine the
link between the explanatory and explained value and minimize
the possibility of outside variables having a confounding
effect?

Explanatory power Is ­there a discussion of how well the researchers explain the
outcome?

Data transparency Is the data publicly available?

Reproducibility Is the methodology publicly available? (qualitative notes,

survey questionnaire, quantitative software)

Research framing Does the piece situate itself within existing research,
engage with its own limitations, and identify areas for ­future
research?

Policy and practice cogency Are the piece’s policy and practice recommendations
specific, actionable, and logically consistent with the data and
analy­sis?

Ethics consideration Does the piece actively engage with the ethics and cultural
sensitivities of the intervention and data collection?

Funding transparency Does the piece acknowledge its funders and potential conflicts
of interest?

Document type: What type of document is it? We sorted documents into eight dif­fer­ent types. Research
articles and program evaluations are primary sources that collect, analyze, and pre­sent data. Lit­er­a­ture re-
views (systematic or nonsystematic), resources guides, and meta-­reviews are secondary sources that sum-
marize and interrogate primary work. Theoretical pieces and toolkits are conceptual documents that focus
entirely on theoretical concepts with l­ittle empirical data.

54 | Preventing and Countering Violent Extremism | USIP.ORG

Major theme: What was the major thematic area of the document? We collected documents focused on
four dif­fer­ent themes: youth, gender, religion, and community resilience.
Explanatory variable: What exactly is the input or explanatory variable that is being mea­sured in the docu-
ment? This is sometimes called the program, treatment, intervention, or in­de­pen­dent variable. When mul-
tiple explanatory variables are reported, we tracked the one that seemed most relevant to P/CVE.
Explanatory variable, actionability: How actionable is the explanatory variable on a scale from 1 (low) to
3 (high)? Explanatory variables that include a program such as job training or mentorship are highly actionable
since additional programs can be created. Explanatory variables that include immutable characteristics such
as a subject’s age and gender are less actionable. Research with low actionability explanatory variables may
still be valuable to our understanding of how to apply programs, similar to how a medi­cation’s efficacy varies
with a patient’s age.
Explanatory variable, proximity: How proximate is this explanatory variable to P/CVE work on a scale from
1 (low) to 3 (high)? A high proximity explanatory variable is closer to the violent-­extremism pro­cess and
more likely to have a direct and meaningful effect, such as enrollment in vocational training and cash trans-
fers for youth.98 A low proximity explanatory variable is something less likely to have a direct and meaning-
ful effect, such as w
­ hether a vocational training program is pre­sent in the subject’s country.
Explained variable: What exactly is the output or the explained variable (sometimes called the outcome,
result, or dependent variable) mea­sured in the document?
Explained variable, type: What type of data is being explained; is it based on observations or responses?
If based on a response, are ­there efforts to mitigate response bias, such as a survey experiment (for ex-
ample, list experiment) or a game (for example, dictator game) to observe be­hav­ior?99
Explained variable, relevance: On a scale from 1 (low) to 3 (high), how relevant is this explained variable
to P/CVE work? A high relevance explained variable is more likely to be related to violent-­extremist be­hav­
ior, such as mea­sur­ing ­whether a subject committed vio­lence. A low relevance explained variable is less
directly related to violent-­extremist be­hav­ior, such as tracking ­whether a job training program increased a
subject’s skills.
Data type: What type of data is used in the document? Is it qualitative data, such as described in case stud-
ies, key in­for­mant interviews, or focus groups? Is it quantitative data, such as descriptive statistics or cross-­
section, time-­series, or panel data? Or is it some other type of data, such as ethnography, systems analy­sis,
or meta-­analysis?
Internal validity: On a scale from 1 (low) to 3 (high), what level of internal validity is pos­si­ble given the meth-
ods that are used in the document? Methods that minimize the possibility of outside variables confounding
the result, such as a randomized control trial, allow for a high level of internal validity. Methods that fail to
control for outside variables may have results driven by other unobservable ­factors, leading to false positives
where a relationship is found where none actually exists.100
Explanatory power: On a scale from 1 (low) to 3 (high), how much discussion is ­there about how well the
analy­sis or the model explains the outcome? A document with a high level of discussion of explanatory
power may include out-­of-­sample predictions. A document with a low level may report a goodness-­of-­fit
statistic such as the R² but does not discuss it. Considering the explanatory power of an analy­sis is
impor­tant ­because inputs can have greatly varying effect sizes even if they have the same statistical
significance.101 Out-­of-­sample analy­sis is useful to prevent overfitting where a model or analy­sis gets so
good at explaining the available data that it fails to generalize to out-­of-­sample data. For example, the
models developed by the Po­liti­cal Instability Task Force perform well on the in-­sample time period, but

USIP.ORG | Preventing and Countering Violent Extremism | 55

per­for­mance decreases as we move to recent times, suggesting that the ­drivers of instability are changing
over time.102
Data transparency: On a scale from 1 (unavailable) to 3 (directly available), how available is the data
analyzed in this document. For example, the document “Reducing Crime and ­Vio­lence: Experimental Evi-
dence from Cognitive Behavioral Therapy in Liberia” has the data directly available on the publication
website.103 Data transparency and reproducibility (discussed below) are impor­tant for allowing ­others to
review and attempt to reproduce the research with the same data. Data analy­sis can include several
seemingly inconsequential decisions that may inadvertently lead to greatly varying results.104 Data trans-
parency allows for dif­fer­ent teams to conduct their own analy­sis and identify what small assumptions
may lead to dif­fer­ent conclusions.105 ­These efforts can also catch minor errors and ­dissuade data fabrica-
tion.106 Data transparency is impor­tant for both quantitative and qualitative data.107
Research framing: On a scale from 1 (poorly) to 3 (strongly), how well does the document engage with
existing research, its own research limitations, and areas for ­future research? As an example of strong
engagement, in “ ‘If Youth Are Given the Chance’: Effects of Education and Civic Engagement on Somali
Youth Support of Po­liti­cal Vio­lence,” the authors ­review the existing lit­er­a­ture on education/civic en-
gagement and po­liti­cal vio­lence, are up front about the limitations of their research design, and raise
several questions for f­ uture research.108
Policy and practice cogency: How suitable are the document’s recommendations for policy and practice
on a scale from 1 (low) to 3 (highly)? Are they specific, actionable, and connected to the data and analy­sis?
For example, the report “Does youth employment build stability? Evidence from an impact evaluation of
vocational training in Af­ghan­i­stan” recommends decoupling employment and stabilization interventions ­after
finding that a program that successfully impacted economic outcomes for participants had almost no impact
on outcomes theoretically related to po­liti­cal vio­lence.109
Reproducibility: On a scale from 1 (unavailable) to 3 (directly available), how available is the methodology
for the document so that someone ­else could reproduce the analy­sis or attempt to replicate the study? For
a survey, this would include the questionnaire and the sampling methods. For data analy­sis, it would in-
clude the software code used to prepare the data and create the results and figures. For example, the study
“Can Social Contact Reduce Prejudice and Discrimination?” includes an online appendix with information on
the case se­lection, sampling, survey questions, and be­hav­ior games, as well as a Dataverse Proj­ect repository
with data and analy­sis files for replication.110 Reproducibility is impor­tant for allowing ­others to reproduce the
study (repeat the research with the same data) and replicate the study (repeat the research with new data,
often in a dif­fer­ent context).111 This goes for qualitative and quantitative methods.112 Replication may be es-
pecially impor­tant for P/CVE when ­there is a par­tic­u­lar concern that programs or interventions that work in
one context may not transfer to another.113
Ethical considerations: On a scale from 1 (low) to 3 (high), how much does the document engage with
ethics and cultural sensitivities of the content? Research is not without its risks, both to the researchers
and the research subjects. This is especially true for P/CVE research, which often focuses on sensitive sub-
jects, traumatic experiences, and harmful actors. In the United States, academic research has government
guidelines, but program evaluations often fall into a gray area.114 Regardless of government standards, all
research and evidence reviews should consider the risks involved, how they can be mitigated, and ­whether
they are worth the benefits.
Funding: Who funded the document’s creation? Are ­there any other relevant financial interests? We only
listed “no funders” if ­there was an explicit declaration that ­there was no funding; other­wise, we listed
“unknown.”

56 | Preventing and Countering Violent Extremism | USIP.ORG

Appendix B

RESEARCH AND EVALUATIONS ASSESSED IN THIS STUDY

Ludovico Alcorta, Jeroen Smits, Haley J. Swedlund, and Eelke de Jong, “The ‘Dark Side’ of Social Capital:
A Cross-­National Examination of the Relationship between Social Capital and Vio­lence in Africa,” Social
Indicators Research 1, no. 49 (2020): 445–65, https://­link​.­springer​.­com​/­article​/­10​.­1007​/­s11205​-­019​
-­02264​-z­ .
Ami C. Carpenter, Community Resilience to Sectarian Vio­lence in Baghdad, 1st ed. (New York: Springer,
2014).
Katherine Casey, Rachel Glennerster, and Edward Miguel, “Reshaping Institutions: Evidence on External
Aid and Local Collective Action,” International Growth Center, April 26, 2011, www​.­theigc​.­org​/­publication​
/­reshaping​-­institutions​-­evidence​-­on​-­external​-a­ id​-­and​-­local​-­collective​-­action​-­working​-­paper​/­.
Nat J. Colletta and Michelle L. Cullen, “The Nexus between Violent Conflict, Social Capital, and Social
Cohesion: Case Studies from Cambodia and Rwanda,” Social Capital Initiative Working Paper 23, World
Bank, September 30, 2000, https://­documents​.­worldbank​.o ­ rg​/­en​/p
­ ublication​/­documents​-­reports​
/­documentdetail​/­318921468743671186​/­the​-n ­ exus​-­between​-­violent​-­conflict​-­social​-­capital​-­and​-­social​
-­cohesion​-­case​-­studies​-­from​-­cambodia​-­and​-­rwanda.
Nat J. Colletta and Michelle L. Cullen, Violent Conflict and the Transformation of Social Capital: Lessons from
Cambodia, Rwanda, Guatemala, and Somalia (Washington, DC: World Bank, 2000).
“Concepts and Dilemmas of State Building in Fragile Situations: From Fragility to Resilience,” OECD
Journal on Development 9, no. 3 (2009): 61–148, doi​.­org​/­10​.­1787​/­journal​_­dev​-­v9​-­art27​-­en.
Anja Dalgaard-­Nielsen and Patrick Schack, “Community Resilience to Militant Islamism: Who and What?
An Explorative Study of Resilience in Three Danish Communities,” Democracy and Security 12, no. 4
(2016): 309–27, www​.­tandfonline​.­com​/­doi​/­abs​/1
­ 0​.1
­ 080​/­17419166​.2
­ 016​.­1236691.
B. Heidi Ellis and Saida Abdi, “Building Community Resilience to Violent Extremism through Genuine
Partnerships,” American Psychologist 72, no. 3 (2017): 289–300, https://­pubmed​.­ncbi​.­nlm​.­nih​.­gov​
/­28383981​/­.
Moroina Engjellushe, Beatrix Austin, Tim Jan Roetman, and Véronique Dudouet, “Community Perspectives
on Preventing Violent Extremism: Lessons Learned from the Western Balkans,” Berghof Foundation, April 24,
2019, https://­berghof​-­foundation​.­org​/­library​/­community​-­perspectives​-­on​-­preventing​-­violent​-­extremism​
-­lessons​-l­earned​-­from​-­the​-­western​-­balkans.
“From High Risk to Resilient: Reducing Vulnerability to Violent Extremism in K­ enya through Social and
Economic Interventions,” Mercy Corps, March 10, 2022, www​.­mercycorps​.o ­ rg​/­research​-­resources​/f­ rom​
-­high​-­risk​-­to​-­resilient.
Michele Grossman, Kristin Hadfield, Philip Jefferies, Vivian Gerrand, and Michael Ungar, “Youth Resilience
to Violent Extremism: Development and Validation of the BRAVE Mea­sure, Terrorism, and Po­liti­cal Vio­
lence,” Terrorism and Po­liti­cal Vio­lence 34, no. 3 (2020): 468–88, doi: 10.1080/09546553.2019.1705283.
Nete Sloth Hansen-­Nord, “Reducing Vio­lence in Poor Urban Areas of Honduras by Building Community
Resilience through Community-­Based Interventions,” International Journal of Public Health 61, no. 8
(2016): 935–43, https://­pubmed​.­ncbi​.­nlm​.­nih​.­gov​/2
­ 7431688​/­.

USIP.ORG | Preventing and Countering Violent Extremism | 57

Remadji Hoinathy and Teniola Tayo, “Lake Chad Basin: Socio-­Economic Resilience in the Shadow of Boko
Haram,” Institute for Security Studies, March 29, 2022, https://­issafrica​.o ­ rg​/­research​/­west​-­africa​-­report​
/­lake​-­chad​-­basin​-­socio​-­economic​-­resilience​-­in​-t­ he​-­shadow​-o
­ f​-­boko​-h
­ aram.
Iffat Idris, “Building Social Cohesion in Post-­Conflict Situations,” GSDRC Helpdesk Research Report 1332,
GSDRC, University of Birmingham, February 2, 2016, https://­gsdrc​.o ­ rg​/­publications​/b
­ uilding​-­social​
-­cohesion​-­in​-­post​-­conflict​-­situations​/­.
International Initiative for Impact Evaluation, “In Fragile Situations, Which Interventions Strengthen Inter-
group Social Cohesion?” YouTube, October 30, 2020, www​.­youtube​.­com​/w ­ atch​?v­ ​=O
­ QvNHalkuis.
Rudine Jakupi and Garentina Kraja, “Accounting for the Difference: Vulnerability and Resilience to Violent
Extremism in Kosovo,” Country Case Study 3, Berghof Foundation, October 18, 2018, https://­berghof​
-­foundation​.o
­ rg​/­library​/­accounting​-­for​-t­ he​-d
­ ifference​-­vulnerability​-­and​-r­ esilience​-­to​-­violent​-­extremism​-­in​
-­kosovo.
Oliver Kaplan, Resisting War: How Communities Protect Themselves (Cambridge: Cambridge University
Press, 2017).
Eelco Kessels, Priscilla Nzabanita, Alistair Millar, Liat Shetret, Naureen Chowdhury Fink, and Jason Ipe,
“Countering Violent Extremism and Promoting Community Resilience in the Greater Horn of Africa: An
Action Agenda,” Global Center on Cooperative Security, May 2015, www​.­globalcenter​.­org​/­resource​
/­countering​-­violent​-­extremism​-­and​-­promoting​-­community​-­resilience​-­in​-t­ he​-­greater​-­horn​-­of​-­africa​/­.
Jeeyon Kim, Ryan Sheely, and Carly Schmidt, “Social Capital and Social Cohesion Mea­sure­ment Toolkit for
Community-­Driven Development Operations,” Mercy Corps, February 2020, www​.­hks​.­harvard​.­edu​
/­centers​/­mrcbg​/­programs​/­growthpolicy​/­social​-­capital​-a­ nd​-­social​-c­ ohesion​-­measurement​-­toolkit.
Karina Korostalina, Neighborhood Resilience and Urban Conflict (Oxfordshire: Routledge Press, 2022).
Augustin Loada and Peter Romaniuk, “Preventing Violent Extremism in Burkina Faso: T­ oward National
Resilience amid Regional Insecurity,” Global Center on Cooperative Security, June 2014, www​.­globalcenter​
.­org​/­resource​/p
­ reventing​-­violent​-­extremism​-­in​-­burkina​-­faso​-­prevention​-­de​-­lextremisme​-­violent​-­au​-­burkina​
-­faso​/­.
“Molenbeek and Violent Radicalisation: A Social Mapping,” Eu­ro­pean Institute of Peace, June 2017.
Redion Qirjazi and Romario Shehu, “Community Perspectives on Preventing Violent Extremism in Alba-
nia,” Country Case Study 4, Berghof Foundation, October 18, 2018, https://­berghof​-­foundation​.­org​
/­library​/­community​-­perspectives​-­on​-­preventing​-­violent​-e­ xtremism​-­in​-a­ lbania.
“Rethinking Resilience: Prioritizing Gender Integration to Enhance House­hold and Community Resilience
to Food Insecurity in the Sahel,” Mercy Corps, 2014, www​.­mercycorps​.o
­ rg​/­sites​/­default​/­files​/2
­ 019​-­12​
/­Rethinking​_­Resilience​_G
­ ender​_­Integration​.p
­ df.
Filip Stojkovski and Natasia Kalajdziovski, “Community Perspectives on the Prevention of Violent Extrem-
ism in Macedonia,” Country Case Study 1, Berghof Foundation, October 18, 2018, https://­berghof​
-­foundation​.o
­ rg​/­library​/­community​-­perspectives​-­on​-t­ he​-­prevention​-­of​-­violent​-e­ xtremism​-­in​-­macedonia.
“Understanding the Links between Social Cohesion and Vio­lence: Evidence from Niger,” Mercy Corps, March 17,
2021, www​.­mercycorps​.­org​/­research​-­resources​/­understanding​-­social​-­cohesion​-­violence.
Lauren Van Metre, “Community Resilience to Violent Extremism in K ­ enya,” US Institute of Peace, October 7, 2016,
www​.­usip​.­org​/­publications​/­2016​/­10​/­community​-­resilience​-­violent​-­extremism​-­kenya.

58 | Preventing and Countering Violent Extremism | USIP.ORG

Floris Vermeulen, “Suspect Communities—­Targeting Violent Extremism at the Local Level: Policies of
Engagement in Amsterdam, Berlin, and London,” Terrorism and Po­liti­cal Vio­lence 26, no. 2 (2014):
286–306, doi: 10.1080/09546553.2012.705254.
Maarten J. Voors and Erwin H. Bulte, “Conflict and the Evolution of Institutions: Unbundling Institutions at
the Local Level in Burundi,” Journal of Peace Research 51, no. 4 (2014): 455–69, www​.­jstor​.­org​/­stable​
/­24557473.
“Vulnerability and Resilience Assessment Initiative to C
­ ounter Violent Extremism (VRAI): Final Synthesis
Report,” Mercy Corps, March 2018, www​.­mercycorps​.­org​/­sites​/­default​/­files​/­2019​-­11​/V
­ RAI Niger Final
Report.pdf.
Stevan Weine and Osman Ahmed, “Building Resilience to Violent Extremism among Somali-­Americans
in Minneapolis-­St. Paul,” Final Report to H ­ uman Factors/Behavioral Sciences D ­ ivision, Science and
Technology Directorate, U.S. Department of Homeland Security, START, A ­ ugust 2012, www​.­dhs​.­gov​
/­publication​/­st​-­building​-­resilience​-­violent​-­extremism​-­among​-­somali​-­americans​-­minneapolis​-­st​-p
­ aul.
Stevan Weine, Schuyler Henderson, Stephen Shanfield, Rupinder Legha, and Jerrold Post, “Building
Community Resilience to C ­ ounter Violent Extremism,” Democracy and Security 9, no. 4 (2013): 327–33,
www​.j­stor​.o
­ rg​/­stable​/­48602748.

USIP.ORG | Preventing and Countering Violent Extremism | 59

Appendix C

PRACTICE LEADER INTERVIEWEES

Jessica Baumgardner-­Zuzik, se­nior director for Learning and Evaluation, Alliance for Peacebuilding
Anais Caput, design and learning specialist, Search for Common Ground
Chantal de Jonge Oudraat, member of the board of directors, W
­ omen in International Security, Wilson
Center Fellow
Cassandra Jessee, director of YouthPower Learning
Peter Mandaville, professor of government and politics, Schar School of Policy and Government, George
Mason University
Hilary Matfess, assistant professor, Josef Korbel School of International Studies, University of Denver
Andrew McDonnell, National Endowment for Democracy, author of four International Center for Religion
and Diplomacy studies: “A New Vision for Countering Violent Extremism in Tunisia,” “Religious Actors
Addressing Violent Extremism,” “Countering Violent Religious Extremism in Pakistan,” and “Engaging
Salafi Religious Actors in Morocco”
Nyambura Mundia, founding director of Usawa Inc. and lead convener of W
­ omen Leaders’ Hangout, K
­ enya
Melissa Nozell, program officer for Religion and Inclusive Socie­ties, US Institute of Peace
Franziska Praxl-­Tabuchi, Global Center on Cooperative Security
Saji Prelis, director of C
­ hildren and Youth Programs, Search for Common Ground
Rachel Walsh Taza, program man­ag­er for ­Children and Youth, Search for Common Ground

Acknowl­edgments

This report benefited greatly from the research assistance of Analise Schmidt, who conducted and cata­
loged the lit­er­a­ture review in French and En­glish, analyzed and coded relevant research, and generated the
bibliography and endnotes.

60 | Preventing and Countering Violent Extremism | USIP.ORG

Notes

1. Marc Sommers, “Youth and the Field of Countering Violent Extremism,” Promundo-­US, 2019, www​
.­equimundo​.­org​/­resources​/­youth​-­and​-­the​-­field​- ­of​- ­countering​-­violent​- ­extremism​/­.
2. Valerie M. Hudson and Hilary Matfess, “In Plain Sight: The Neglected Linkage between Brideprice
and Violent Conflict,” International Security 42, no. 1 (2017): 7–40; Jennifer Glassco and Lina
Holguin, “Youth and In­equality: Time to Support Youth as Agents of Their Own F­ uture,” Oxfam
Briefing Paper, August 12, 2016, https://­policy​-­practice​.­oxfam​.­org​/­resources​/­youth​-­and​-­inequality​
-­time​-­to​-­support​-­youth​-­as​-­agents​- ­of​-­their​- ­own​-­future​- ­618006​/­; Lauren Van Metre, “Community
Resilience to Violent Extremism in ­Kenya,” US Institute of Peace, October 7, 2016, www​.­usip​.­org​
/­publications​/­2016​/­10​/­community​-­resilience​-­violent​- ­extremism​-­kenya​.­
3. “Thought Leadership on Peace and Conflict,” Mercy Corps, June 22, 2021, www​.­mercycorps​.­org​
/­research​-­resources​/­thought​-­leadership​-­peace​- ­conflict.
4. “Fact Sheet: The White House Summit on Countering Violent Extremism,” The White House,
February 18, 2015, https://­obamawhitehouse​.­archives​.­gov​/­the​-­press​- ­office​/­2015​/­02​/­18​/­fact​-­sheet​
-­white​-­house​-­summit​- ­countering​-­violent​- ­extremism​.­
5. Stevan Weine and Osman Ahmed, “Building Resilience to Violent Extremism among Somali-­
Americans in Minneapolis-­St. Paul,” Final Report to H ­ uman Factors/Behavioral Sciences Division,
Science and Technology Directorate, U.S. Department of Homeland Security, START, August 2012,
www​.­dhs​.­gov​/­publication​/­st​-­building​-­resilience​-­violent​- ­extremism​-­among​-­somali​-­americans​
-­minneapolis​-­st​-­paul.
6. Antonia Ward, “To Ensure Deradicalisation Programmes Are Effective, Better Evaluation Practices
Must First Be Implemented,” The RAND Blog, March 4, 2019, www​.­rand​.­org​/­blog​/­2019​/­03​/­to​
-­ensure​- ­deradicalisation​-­programmes​-­are​- ­effective​.­html.
7. Ashutosh Varshney, Ethnic Conflict and Civic Life: Hindus and Muslims in India, 2nd ed. (New Haven:
Yale University Press, 2003); Ami C. Carpenter, Community Resilience to Sectarian Vio­lence in
Baghdad, 1st ed. (New York: Springer, 2014); Oliver Kaplan, Resisting War: How Communities
Protect Themselves (Cambridge: Cambridge University Press, 2017), available at ProQuest Ebook
Central, https://­ebookcentral​.­proquest​.­com​/­lib​/­g wu​/­detail​.­action​?­docID​= ­4 812248. Full disclosure:
We have personal ties to Oliver Kaplan, including as a result of working at the US Institute of Peace
while Kaplan was a fellow t­ here; Moroina Engjellushe, Beatrix Austin, Tim Jan Roetman, and Véro-
nique Dudouet, “Community Perspectives on Preventing Violent Extremism: Lessons Learned from
the Western Balkans,” Berghof Foundation, April 24, 2019, https://­berghof​-­foundation​.­org​/­library​
/­community​-­perspectives​- ­on​-­preventing​-­violent​- ­extremism​-­lessons​-­learned​-­from​-­the​-­western​
-­balkans; Van Metre, “Community Resilience to Violent Extremism in ­Kenya.”
8. The forums include the African Centre for the Study and Research on Terrorism; the Berghof
Foundation; Hedayah; the International Centre for Counter-­Terrorism; the International Organ­
ization for Migration; Mercy Corps; Centre of Excellence Defence against Terrorism; the Organisa-
tion for Economic Co-­operation and Development; the Organ­ization for Security and Cooperation
in Eu­rope; the RAND Corporation; the RESOLVE Network; the Royal United Ser­vices Institute for
Defence and Security Studies; the Stabilisation Unit of the government of the United Kingdom; the

USIP.ORG | Preventing and Countering Violent Extremism | 61

 University of Mary­land’s Study of Terrorism and Responses to Terrorism consortium; the United
Nations, including the United Nations Development Programme and United Nations Office of
Counter-­Terrorism; and the United States Agency for International Development.
9. Claudia Wallner, “The Contested Relationship between Youth and Violent Extremism: Assessing the
Evidence Base in Relation to P/CVE Interventions,” Royal United Ser­vices Institute for Defence and
Security Studies, February 8, 2021, https://­rusi​.­org​/­explore​-­our​-­research​/­publications​/­occasional​
-­papers​/­contested​-­relationship​-­between​-­youth​-­and​-­violent​-­extremism​-­assessing​-­evidence​-­base​
-­relation​-­pcve; “How to Note: Assessing the Strength of Evidence,” United Kingdom Department for
International Development, March 19, 2014, www​.­gov​.­uk​/­government​/­publications​/­how​-­to​-­note​
-­assessing​-­the​-­strength​-­of​-­evidence; Georgia Holmer, Peter Bauman, and Kateira Aryaeinejad, “Mea­
sur­ing Up: Evaluating the Impact of P/CVE Programs,” US Institute of Peace, September 6, 2018, www​
.­usip​.­org​/­publications​/­2018​/­09​/­measuring​-­monitoring​-­and​-­evaluating​-­pcve​-­programs; “Preventing
Violent Extremism through Education: Effective Activities and Impact,” United Nations Educational,
Scientific, and Cultural Organ­ization, 2018, https://­unesdoc​.­unesco​.­org​/­ark:​/­48223​/­pf0000266105;
Beverley J. Shea, Barnaby C. Reeves, George Wells, Micere Thuku, Candyce Hamel, Julian Moran, David
Moher, Peter Tugwell, Vivian Welch, Elizabeth Kristjansson, and David A Henry, “AMSTAR 2: A Critical
Appraisal Tool for Systematic Reviews That Include Randomised or Non-­Randomised Studies of Health-
care Interventions, or Both,” BMJ 358 (2017), www​.­bmj​.­com​/­content​/­358​/­bmj​.­j4008.
10. The full code­book, along with the codings, is available at the report’s Github Repository.
11. Kaplan, Resisting War.
12. “Positive Youth Development,” Youth​.­gov, https://­youth​.­gov​/­youth​-­topics​/­positive​-­youth​
-­development.
13. “Positive Youth Development (PYD) Framework,” YouthPower2, www​.­youthpower​.­org​/­positive​
-­youth​- ­development​-­pyd​-­framework.
14. Interview with Melissa Nozell, United States Institute of Peace, November 5, 2021.
15. Interview with Peter Mandaville, United States Institute of Peace, November 5, 2021.
16. Varshney, Ethnic Conflict and Civic Life. https://­ebookcentral​.­proquest​.­com​/­lib​/­g wu​/­detail​.­action​
?­docID​=­3420042
17. Carpenter, Community Resilience to Sectarian Vio­lence in Baghdad.
18. Van Metre, “Community Resilience to Violent Extremism in ­Kenya.”
19. Rudine Jakupi and Garentina Kraja, “Accounting for the Difference: Vulnerability and Resilience to
Violent Extremism in Kosovo,” Country Case Study 3, Berghof Foundation, October 18, 2018,
https://­berghof​-­foundation​.­org​/­library​/­accounting​-­for​-­the​- ­difference​-­vulnerability​-­and​-­resilience​
-­to​-­violent​- ­extremism​-­in​-­kosovo.
20. Dorina A. Bekoe, Stephanie M. Burchard, and Sarah A. Daly, “Extremism in Mozambique: Interpreting
Group Tactics and the Role of the Government’s Response in the Crisis in Cabo Delgado,” Institute
for Defense Analyses, March 2020, www​.­ida​.­org​/­research​-­and​-­publications​/­publications​/­all​/­e​/­ex​
/­extremism​-­in​-­mozambique​-­interpreting​-­group​-­tactics​-­and​-­the​-­role​-­of​-­the​-­governments​-­response​.­
21. “Niger and Boko Haram: Beyond Counter-­Insurgency,” International Crisis Group, February 27, 2017,
www​.­crisisgroup​.­org​/­africa​/­west​-­africa​/­niger​/­245​-­niger​-­and​-­boko​-­haram​-­beyond​- ­counter​
-­insurgency.

62 | Preventing and Countering Violent Extremism | USIP.ORG

22. Ludovico Alcorta, Jeroen Smits, Haley J. Swedlund, and Eelke de Jong, “The ‘Dark Side’ of Social
Capital: A Cross-­National Examination of the Relationship between Social Capital and Vio­lence in
Africa,” Social Indicators Research 149, no. 2 (2020): 445–65, https://­link​.­springer​.­com​/­article​/­10​
.­1007​/­s11205​- ­019​- ­02264​-­z.
23. Alcorta et al., “The ‘Dark Side’ of Social Capital.”
24. Jakupi and Kraja, “Accounting for the Difference.”
25. Bekoe, Burchard, and Daly, “Extremism in Mozambique.”
26. Van Metre, “Community Resilience to Violent Extremism in ­Kenya.”
27. For example, see Varshney, Ethnic Conflict and Civic Life.
28. Carpenter, Community Resilience to Sectarian Vio­lence in Baghdad; Van Metre, “Community
Resilience to Violent Extremism in ­Kenya”; Varshney, Ethnic Conflict and Civic Life.
29. “­Human Rights Impact of Policies and Practices Aimed at Preventing and Countering Violent
Extremism,” report of the Special Rapporteur on the Promotion and Protection of H ­ uman Rights
and Fundamental Freedoms while Countering Terrorism, 2020, 8, https://­digitallibrary​.­un​.­org​
/­record​/­3872336​?­ln​= ­en.
30. Simone Haysom and Ken Opala, “The Politics of Crime: ­Kenya’s Gang Phenomenon,” Global Initia-
tive against Transnational Or­ga­nized Crime, November 18, 2020, 10, https://­globalinitiative​.­net​
/­analysis​/­gangs​/­.
31. Haysom and Opala, “The Politics of Crime,” 16.
32. Haysom and Opala, “The Politics of Crime,” 44.
33. Sommers, “Youth and the Field of Countering Violent Extremism.”
34. United Nations Development Programme, Journey to Extremism in Africa: D
­ rivers, Incentives, and
the Tipping Point for Recruitment (New York: UNDP, 2017).
35. Isa Blumi, “Po­liti­cal Islam among the Albanians: Are the Taliban Coming to the Balkans?,”
Kosovar Institute for Policy Research and Development, June 2005, www​.­kipred​.­org​/­repository​
/­docs​/­Political​_ ­Islam​_ ­Among​_­the​_ ­Albanians​-­​_­Are​_­the​_­Taliban​_­coming​_­to​_­the​_ ­Balkans​_­137060​
.­pdf.
36. Rida Lyammouri and Noufal Abboud, “Traditional Authorities and Governance in Mali,” Sahel
memo, written for the National Demo­cratic Institute internal report, 2019.
37. Rahmane Idrissa and Bethany McGann, “Mistrust and Imbalance: The Collapse of Intercommunal
Relations and the Rise of Armed Community Mobilization on the Niger-­Mali Border,” RESOLVE
Network, April 21, 2021, www​.­resolvenet​.­org​/­research​/­mistrust​-­and​-­imbalance​- ­collapse​
-­intercommunal​-­relations​-­and​-­rise​-­armed​- ­community.
38. Carpenter, Community Resilience to Sectarian Vio­lence in Baghdad.
39. Carpenter, Community Resilience to Sectarian Vio­lence in Baghdad; Van Metre, “Community
Resilience to Violent Extremism in ­Kenya.”
40. Kaplan, Resisting War.
41. Van Metre, “Community Resilience to Violent Extremism in ­Kenya”; Jakupi and Kraja, “Accounting
for the Difference”; Lyammouri and Abboud, “Traditional Authorities and Governance in Mali.”

USIP.ORG | Preventing and Countering Violent Extremism | 63

42. Ngala Chome, “Violent Extremism and Clan Dynamics in K ­ enya,” US Institute of Peace, October 31,
2016, www​.­usip​.­org​/­publications​/­2016​/­10​/­violent​- ­extremism​-­and​- ­clan​- ­dynamics​-­kenya​.­
43. Apprendre a vivre ensemble. “Niger: A Forum on Peace and Stabilization ­Will Be Held in Niamey,”
March 2021.
44. Idrissa and McGann, “Mistrust and Imbalance.”
45. Bonding social capital occurs within a group or community, whereas bridging social capital occurs
between social groups, according to social class, race, religion, or other impor­tant sociodemo-
graphic or socioeconomic characteristics. In terms of social capital and vio­lence, without “bridging”
social capital, “bonding” groups can become isolated and disenfranchised from the rest of society,
and the strengthening of insular ties can lead to a variety of effects, such as ethnic marginalization
or social isolation. In extreme cases, ethnic cleansing may result if the relationship between dif­fer­
ent groups is so strongly negative.
46. On Côte d’Ivoire, Latin Amer­i­ca, and Nigeria, see Johannes Harnischfeger, “The Bakassi Boys:
Fighting Crime in Nigeria,” Journal of Modern African Studies 41, no. 1 (2003): 23–49, www​.­jstor​.­org​
/­stable​/­3876188; on Pakistan, see Laurent Gayer, “Guns, Slums, and ‘Yellow Dev­ils’: A Genealogy of
Urban Conflicts in Karachi, Pakistan,” Modern Asian Studies 41, no. 3 (2007): 515–44, www​.­jstor​.­org​
/­stable​/­4 499791.
47. Clara Pretus, Nafees Hamid, Hammad Sheikh, Jeremy Ginges, Adolf Tobeña, Richard Davis, Oscar
Vilarroya, and Scott Atran, “Neural and Behavioral Correlates of Sacred Values and Vulnerability to
Violent Extremism,” Frontiers in Psy­chol­ogy 9 (2018): 24–62, doi​.­org​/­10​.­3389​/­fpsyg​.­2018​.­02462.
48. Andries Odendaal, “An Architecture for Building Peace at the Local Level: A Comparative Study of
Local Peace Committees,” UN Development Programme, December 2010, www​.­un​.­org​/­en​/­land​
-­natural​-­resources​- ­conflict​/­pdfs​/­UNDP​_ ­Local Peace Committees_2011.pdf.
49. “Building Infrastructures for Peace in Africa,” Global Partnership for the Prevention of Armed
Conflict, July 23, 2020, www​.­gppac​.­net​/­resources​/­building​-­infrastructures​-­peace​-­africa.
50. Rene Claude Niyonkuru, “Building the Peace Architecture from Bottom-­up: The Experience of Local
Peace Committees in Burundi,” Peace Building Research Proj­ect, Occasional Paper 5, Novem-
ber 2012, https://­cdn​.­future​.­edu​/­wp​- ­content​/­uploads​/­2018​/­0 6​/­Burundi​-­Bottom​-­up​-­Peace​
-­Architecture​-­and​-­Local​-­Peace​- ­Committees​.­pdf.
51. Odendaal, “An Architecture for Building Peace at the Local Level.”
52. Samuel Majekodunmi, “Infrastructure for Peace: The African Experience,” African Research Review
11, no. 2 (2017): 25–41, www​.­ajol​.­info​/­index​.­php​/­afrrev​/­article​/­view​/­157108.
53. Majekodunmi, “Infrastructure for Peace”; Niyonkuru, “Building the Peace Architecture from
Bottom-up.”
54. Fritz Nganje, “Local Peace Committees and Grassroots Peacebuilding in Africa,” in The State of
Peacebuilding in Africa, ed. Terence McNamee and Monde Myangwa (London: Palgrave Macmillan,
2021).
55. Niyonkuru, “Building the Peace Architecture from Bottom-up.” The bashingantahe is a traditional
justice mechanism in Uganda made up of a council of wise elders who represent all ethnic and
social identities in the community. They s­ ettle property disputes and mediate community conflicts
in order to safeguard peace and strengthen social harmony based on their moral authority.

64 | Preventing and Countering Violent Extremism | USIP.ORG

56. Abdul Karim Issifum, “Local Peace Committees in Africa: The Unseen Role in Conflict Resolution and
Peacebuilding,” Journal of Pan African Studies 9, no. 1 (2016): 141–58.
57. Mediel Hove and Geoff Harris, eds., Infrastructures for Peace in Sub-­Saharan Africa (Cham, Switzer-
land: Springer, 2019), 10.
58. Hove and Harris, Infrastructures for Peace in Sub-Saharan Africa, 10
59. Hove and Harris, Infrastructures for Peace in Sub-Saharan Africa, 10
60. Peace Direct.
61. On Burundi, see ZOA International “Guide: Instructions à accompagner les vidéos et la boite à
images pour discute des questions stratégiques dans la mise en llace des Comités Locaux de Paix et
leur renforcement,” 2016. On the Central African Republic, see United States Agency for Interna-
tional Development, Rapport d’evaluation participative comités locaux de Paix et de Réconciliation
(CLPR) et Des Antennes Locales de Suivi Du Pro­cessus de Kimberley (ALS) Dans La Zone Conforme de
Berberati,” US Agency for International Development, April 2019. On Côte d’Ivoire, see Jean Bosson
Kouadia, Evariste Sadie, and Djamel Kasmi, “Rapport d’évaluation du projet Search for Common
Ground: Renforcement des capacités locales à réduire les conflits et cacilitation uu renforcement de
la cohésion sociale et de la réconciliation,” Search for Common Ground, June 2007. On Niger, see
Salif Nimaga, Exercice évaluatif Niger: Rapport final; Fond pour la consolidation de la paix (PBF,
2020). On Senegal, see Alvar Jones Sanchez, “Les comités de paix pour la résolution de conflits en
Casamance: De l’illusion populaire au déni politique,” Fondation Croix-­Rouge française, March 2018.
62. In Burundi, tools w
­ ere tested during two experimental workshops with team members from
national and international development organ­izations, as well as during ten workshops delivered
in-­country for members of local peace committees and other local actors, including administrative
representatives, customary leaders, and civil society. In the Central African Republic, peace-­and-­
reconciliation committees and ALSs w ­ ere assembled for a two-­day period to gather members’
perspectives on the strengths and weaknesses of the bodies through guided discussions. In Côte
d’Ivoire, Search for Common Ground distributed three surveys: one evaluative survey for each of
the three programs they implemented in the country. In Niger, two cycles of individual and group
reflections ­were conducted; an online survey was sent to members of the country’s UN team,
government partners, and technical and financial partners; and a document review was conducted.
For Senegal, an evaluative overview of the scope of the peace committee strategy was conducted.
63. More detailed subquestions ­were also created and can be found ­here.
64. ­There has been a coordinated effort in the Sahel to develop national security strategies in Mali,
Burkina Faso, Niger, and Nigeria to address the cross-­border threats of violent extremist groups.
­These strategies also center their efforts around community engagement, although it remains to
be seen how committed they are to this approach in their implementation.
65. Idrissa and McGann, “Mistrust and Imbalance.”
66. “Sidelining the Islamic State in Niger’s Tillaberie,” International Crisis Group, June 3, 2020, 46,
www​.­crisisgroup​.­org​/­africa​/­sahel​/­niger​/­289​-­sidelining​-­islamic​-­state​-­nigers​-­tillabery.
67. Idrissa and McGann, “Mistrust and Imbalance.”
68. Bernardo Venturi, ed., Governance and Security in the Sahel: Tackling Mobility, Demography, and
Climate Change (Brussels: Foundation for Eu­ro­pean Progressive Studies and Istituto Affari Internazi-
onali, April 2019).

USIP.ORG | Preventing and Countering Violent Extremism | 65

69. Kent Elbow, “Burkina Faso’s Ambitious Experiment in Participatory Land Tenure Reform,” Focus on
Land in Africa, August 2013, https://­gatesopenresearch​.­org​/­documents​/­3​- ­856.
70. Kent Elbow, “Emerging Lessons from MCC/MCA-­Sponsored Initiatives to Formalize Customary Land
Tenure Rights and Local Land Management Practices in Benin, Burkina Faso and Senegal,” paper
presented at the Annual World Bank Conference on Land and Poverty, 2012.
71. Admos Chimhowu, “The ‘New’ African Customary Land Tenure: Characteristics, Features, and
Policy Implications of a New Paradigm,” Land Use Policy 81 (2019): 897–903, doi​.­org​/­10​.­1016​/­j​
.­landusepol​.­2018​.­0 4​.­014.
72. Elbow, “Burkina Faso’s Ambitious Experiment in Participatory Land Tenure Reform.”
73. Chimhowu, “The ‘New’ African Customary Land Tenure.”
74. Chimhowu, “The ‘New’ African Customary Land Tenure.”
75. Paul Hufe and Daniel F. Heuermann, “The Local Impacts of Large-­Scale Land Acquisitions: A Review
of Case Study Evidence from Sub-­Saharan Africa,” Journal of Con­temporary African Studies 35, no.2
(2017): 168–89, www​.­tandfonline​.­com​/­doi​/­abs​/­10​.­1080​/­02589001​.­2017​.­1307505.
76. Van Metre, “Community Resilience to Violent Extremism in ­Kenya”; Thomas Nyagah, James
Mwangi, and Larry Attree, “Inside K ­ enya’s War on Terror: The Case of Lamu,” SaferWorld, May 15,
2019, www​.­saferworld​.­org​.­uk​/­long​-­reads​/­inside​-­kenyaas​-­war​- ­on​-­terror​-­the​- ­case​- ­of​-­lamu.
77. Chimhowu, “The ‘New’ African Customary Land Tenure.”
78. Daniel Agbiboa, Origins of Hybrid Governance and Armed Community Mobilization in Sub-­Saharan
Africa (Washington, DC: RESOLVE Network, 2019), doi​.­org​/­10​.­37805​/­cbags2019​.­2.
79. Community resilience studies by Van Metre and Carpenter have found that local security networks
are critical for their resilience to vio­lence—­only, however, when they are defensive and protective
in nature. When they become offensive, they invite retributive vio­lence.
80. Wamaitha Ndono, Nzioka John Muthama, and Kariuki Muigua, “Effectiveness of the Nyumba Kumi
Community Policing Initiative in K ­ enya,” Journal of Sustainability, Environment and Peace 1, no. 2
(2019): 63–67, http://­jsep​.­uonbi​.­ac​.­ke​/­ojs​/­index​.­php​/­jsep​/­article​/­view​/­203.
81. Peterson Munene Njagi, “Nyumba Kumi Initiative: A Critical Analy­sis,” in Securitization of the
Everyday, ed. Giuditta Fontana (Colchester, UK: Eu­ro­pean Consortium for Po­liti­cal Research,
2020).
82. Charlotte Cross, “Ulinzi Shirikishi: Popu­lar Experiences of Hybrid Security in Tanzania,” Development
and Change 7, no. 5 (2016): 1102–24, https://­onlinelibrary​.­wiley​.­com​/­doi​/­abs​/­10​.­1111​/­dech​.­12261.
83. Cross, “Ulinzi Shirikishi.”
84. Cross, “Ulinzi Shirikishi.”
85. Cross, “Ulinzi Shirikishi.”
86. Ndono, Muthama, and Muigua, “Effectiveness of the Nyumba Kumi Community Policing Initiative in
­Kenya.”
87. Bethany L. McGann, “Hybridity and Fragmentation: Implications for Regional Security Policy in the
Sahel and Beyond,” in Extremisms in Africa, vol. 3, ed. Stephen Buchanan-­Clarke, Lloyd Coutts,
Susan Russell, Alan Yschudin, and Craig Moffat (Johannesburg: Bookstorm, 2020).

66 | Preventing and Countering Violent Extremism | USIP.ORG

88. Ken Menkhaus, “The Rise of the Mediated State in K ­ enya: The Wajir Story and Its Implications for
State-­Building,” Afrika Focus 21, no. 2 (2008): 23–38.
89. McGann, “Hybridity and Fragmentation.”
90. International Peace Institute, “Investir dans la paix et la prévention de la vio­lence dans le Sahel-­
Sahara: Deuxièmes conversations régionales,” in Investir dans la paix et la prévention de la vio­lence
dans le Sahel-­Sahara (N’Djamena, Chad: International Peace Institute, August 2017).
91. Fransje Molenaar, Jonathan Tossell, Anna Schmauder, Abdourahmane Idrissa, and Rida Lyammouri,
“La fin du statu quo? La égitimité des autorités coutumières dans les zones de gouvernance hybride
au Mali, au Niger, et en Libye,” Clingendael Institute, 2020.
92. Molenaar et al., “La fin du statu quo?”
93. “Une nouvelle approche pour prévenir l’extrémisme violent,” United Nations Development Pro-
gramme, 2019, www​.­undp​.­org​/­fr​/­blog​/­une​-­nouvelle​-­approche​-­pour​-­prevenir​-­lextremisme​-­violent.
94. Centre africain d’Études Internationales Diplomatiques, Économiques, et Stratégiques, Séminaire
régional pour la prévention de l’extrémisme violent en Afrique Centrale et dans le Bassin du Lac
Tchad, Yaoudé, les 27 et 28 novembre, 2017 (Yaoundé, Cameroon: United Nations Regional Office
for Central Africa and Swiss Confederation, 2017).
95. This is an emerging governance area of practice called radical transparency—­the idea that the bulk
of policymaking is hidden from constituents and that it should be made increasingly transparent
and participatory.
96. African Center for International, Diplomatic, Economic, and Strategic Studies, Séminaire régional
pour la prévention de l’extrémisme violent.
97. Mvondo Hervé, “La Gouvernance Locale face au défi de Sécurité en Afrique Centrale: Une analyse
de la Décentralisation comme moyen de la lutte Contre l’Extrémisme Violent au Cameroun,” Revue
Africaine sur le Terrorisme 7, no. 2 (2019): 109–22.
98. Jon Kurtz, Beza Tesfaye, and Rebecca Wolfe, Can Economic Interventions Reduce Vio­lence? Impacts
of Vocational Training and Cash Transfers on Youth Support for Po­liti­cal Vio­lence in Af­ghan­i­stan
(Washington, DC: Mercy Corps, February 2018).
99. For an example of a dictator game, see Alexandra Scacco and Shana S. Warren, “Can Social Contact
Reduce Prejudice and Discrimination? Evidence from a Field Experiment in Nigeria,” American
Po­liti­cal Science Review 112, no. 3 (2018): 654–77, doi​.­org​/­10​.­1017​/­S 0003055418000151.
100. Stephen Chaudoin, Jude Hays, and Raymond Hicks, “Do We R ­ eally Know the WTO Cures Cancer?”
British Journal of Po­liti­cal Science 48, no. 4 (2018): 903–28, doi​.­org​/­10​.­1017​/­S 000712341600034X.
101. Michael D. Ward, Brian D. Greenhill, and Kristin M. Bakke, “The Perils of Policy by P-­Value: Predict-
ing Civil Conflicts,” Journal of Peace Research 47, no. 4 (2010): 363–75, doi​.­org​/­10​.­1177​
/­0 022343309356491.
102. Drew Bowlsby, Erica Chenoweth, Cullen Hendrix, and Jonathan D. Moyer, “The F­ uture Is a Moving
Target: Predicting Po­liti­cal Instability,” British Journal of Po­liti­cal Science 50, no. 4 (2020): 1405–17,
doi​.­org​/­10​.­1017​/­S 0007123418000443.
103. Christopher Blattman, Julian C. Jamison, and Margaret Sheridan, “Reducing Crime and Vio­lence:
Experimental Evidence from Cognitive Behavioral Therapy in Liberia,” American Economic Review
107, no. 4 (2017): 1165–1206, www​.­aeaweb​.­org​/­articles​?­id​=­10​.­1257​/­aer​.­20150503.

USIP.ORG | Preventing and Countering Violent Extremism | 67

104. Rex Douglass and Kristen Harkness demonstrate this phenomenon in their study of vio­lence in the
Mau Mau rebellion by showing how small changes in how event data are aggregated can greatly
affect results; see Rex W. Douglass and Kristen A. Harkness, “Mea­sur­ing the Landscape of Civil War:
Evaluating Geographic Coding Decisions with Historic Data from the Mau Mau Rebellion,” Journal
of Peace Research 55, no. 2 (2018): 190–205, doi​.­org​/­10​.­1177​/­0 022343318754959.
105. A recent preprint study gave seventy-­three teams the same data and hypothesis. About 60 ­percent
of the teams found no statistical significant relationship, 25 ­percent found a negative relationship,
and 15 ­percent found a positive relationship. See Nate Breznau, Eike Mark Rinke, Alexander Wuttke,
Hung H. V. Nguyen, Muna Adem, Jule Adriaans, Amalia Alvarez-­Benjumea, Henrik K. Andersen,
et al., “Observing Many Researchers Using the Same Data and Hypothesis Reveals a Hidden Uni-
verse of Uncertainty,” MetaArXiv preprint, March 2021.
106. See the Reinhart-­Rogoff spreadsheet error and the recent case of LaCour and Green’s Science
article based on fabricated survey data.
107. Colin Elman, Diana Kapiszewski, and Arthur Lupia, “Transparent Social Inquiry: Implications for
Po­liti­cal Science,” Annual Review of Po­liti­cal Science 21, no. 1 (2018): 29–47, doi​.­org​/­10​.­1146​
/­annurev​-­polisci​- ­091515​- ­025429; Andrew Moravcsik, “Trust, but Verify: The Transparency Revolu-
tion and Qualitative International Relations,” Security Studies 23, no. 4 (2014): 663–88, www​
.­tandfonline​.­com​/­doi​/­abs​/­10​.­1080​/­0 9636412​.­2014​.­970846​?­journalCode​= ­fsst20.
108. Beza Tesfaye, Topher McDougal, Beth Maclin, and Andy Blum, “If Youth Are Given the Chance:
Effects of Education and Civic Engagement on Somali Youth Support of Po­liti­cal Vio­lence,” Mercy
Corps, May 16, 2018, www​.­mercycorps​.­org​/­research​-­resources​/­youth​- ­chance​- ­education​-­somalia.
109. “Does Youth Employment Build Stability? Evidence from an Impact Evaluation of Vocational Training
in Af­ghan­i­stan,” Mercy Corps, January 2015, www​.­mercycorps​.­org​/­research​-­resources​/­youth​
-­employment​-­stability.
110. Scacco and Warren, “Can Social Contact Reduce Prejudice and Discrimination?” Also see their
online appendix and Dataverse Proj­ect repository.
111. Jeremy Freese and David Peterson, “Replication in Social Science,” Annual Review of Sociology 43,
no. 1 (2017): 147–65, doi​.­org​/­10​.­1146​/­annurev​-­soc​- ­0 60116​- ­053450.
112. Megan Becker et al., “Replicating the Resource Curse: A Qualitative Replication of Ross 2004,”
preprint, MetaArXiv, 2021.
113. See Aila M. Matanock, “Experiments in Post-­Conflict Contexts,” in Advances in Experimental
Po­liti­cal Science, ed. James N. Druckman and Donald P. Green (Cambridge: Cambridge University
Press, 2021), 562–91.
114. We score a document that has Institutional Review Board (IRB) oversight but no other discussion of
ethics as a 2 (medium).

68 | Preventing and Countering Violent Extremism | USIP.ORG