Find the full issue of Horn of Africa Bulletin here


Review article: Community Perceptions of Violent Extremism in Kenya

Review article: Buchanan-Clarke S., Humphrey A., Villa-Vicencio C., 2016, Community Perceptions of Violent Extremism in Kenya, Occasional Paper 21, Justice and Reconciliation in Africa, Institute for Justice and Reconciliation [i]

The “lack of definitional consensus [on violent extremism] often stems from a scarcity of empirical evidence on the assumed root causes and drivers of violent extremism”. Indeed, locally-informed empirical research on violent extremism remains thin, although violent extremism is increasingly put as a priority on the decision-makers’ agenda at the national, regional and international levels, and has become a focus of peacebuilding programming in the Horn of Africa under the emerging Countering Violent Extremism (CVE) approach.

This gap was emphasised by the authors of the research report Community Perceptions of Violent Extremism in Kenya, the product of a three-month study led by researchers from the Institute for Justice and Reconciliation and Georgetown University, in consultation with the Life & Peace Institute.  The study aimed to inform CVE approaches in Kenya and promote context-sensitive CVE programming in the country, by documenting local perceptions of violent extremism and echoing rarely considered grassroots voices.

The following article provides a summary of the research findings and assessment of its contribution to academic, programmatic and policy conversations on violent extremism in Kenya.

Research Scope

Community Perceptions of Violent Extremism in Kenya articulates key insights on the way diverse communities from urban Kenya understand and experience violent extremism. Local communities from four locations were targeted by the research, namely Majengo and Eastleigh in Nairobi, Garissa Township, and Majengo in Mombasa. All locations were selected based on the criterion that they are all affected by violent extremism. Specifically, these areas have allegedly been used as recruitment pools by armed groups labelled as extremist, in particular al-Shabab, and consequently targeted by governmental and civil society CVE initiatives.

Based on data collected through focus group discussions and key informant interviews, the report sheds light on local communities’ perceptions on four aspects, namely how the term violent extremism is understood locally; insecurity; local drivers of violent extremism; and dynamics of recruitment by groups like al-Shabab. The report is divided along these aspects.

Although this study is not unusual in terms of its purpose and methodology[ii], it is significant as it relies on field research and primary data. The analysis of perceptions was contextualised to the four research locations at the time of data collection, thereby allowing comparisons between differently affected locales. This inherent effort of contextualisation throughout the research also prevented it from drawing any generalising conclusion, which would have contradicted the essential study purpose. Indeed, the report demonstrates that the dynamics of violent extremism differ by community, and even among individuals within the same community. Thus, any generalised theory of violent extremism would likely be unfounded.

 CVE Terminology: A Fundamental Challenge

“There exists no universally accepted definition of violent extremism” (p.4), thus begins the report’s terminology section. Consistently with most studies on violent extremism, the research team emphasizes the lack of semantic consensus on violent extremism. However, instead of offering their own definition of the phenomenon, the authors summarise a myriad of definitions. They notably highlight the fact that “some definitions emphasise a group’s ideological or religious objectives as foundations for violent extremism, while others place more focus on the particular tactics employed by a group.” (p.4). This lack of consensus stresses the relativity of the concept of violent extremism and the fact that it “inevitably means different things to different people” (p.5), to the extent that some could argue violent extremism is devoid of meaning.

The implications of this lack of commonly shared definition are detrimental to any academic, policy or programmatic attempt to address violent extremism in a sensitive and effective manner. The emerging CVE field can be seen as a way to overcome this definitional challenge, by emphasizing a “more comprehensive and contextualised” understanding of violent extremism (p.5) and to forward measures to prevent violent extremism based on this more complete understanding. This disparate non-coercive set of measures addressing “structural drivers (including push and pull factors) that fuel grievances and may entice individuals to support violent extremist groups” (p.5) would constitute an adequate approach to address the root causes and drivers of violent extremism in a sustainable and context-sensitive manner. Yet, as the authors highlight, “CVE initiatives in Kenya have almost exclusively focused on Muslim communities”, thus suggesting an inherent bias and persistent reductionist prism.

Violent Extremism: A Snapshot of Local Understandings

When asking respondents about their understanding of violent extremism, the research team realised that this concept, increasingly used in policy-making and Western media, cannot be literally translated in Kiswahili or Somali languages. This illustrates the lack of a “common understanding […] within and between the communities interviewed” (p.6). Interestingly, respondents’ definitions reflected their daily experiences of insecurity and “immediate security concerns regardless of the actor involved or their motive for violence” (p.6). In a mainstreamed effort to nuance and contextualise the research findings, the authors provide multiple examples of understandings per location, with respondents from Garissa including “violence perpetrated as a result of conflict between sub-clans in their definitions of violent extremism” (p.9) and respondents from Eastleigh differentiating “between violent extremist actors and local gangs not by ideology, as is common in civil society and academic circles, but by the weapons they use.” (p.9)

This zooming in on communities’ understanding of violent extremism in diverse communities has implications for CVE initiatives. Findings reveal that communities at the grassroot level understand violent extremism based on their primary security concerns, and do not perceive armed extremist groups’ actions as mainly driven by ideology. This implies an incompatibility between local perceptions and more “conventional” definitions of violent extremism disseminated by higher-level arena (policymakers, media, donors or implementing civil society organisations (CSOs)). This incompatibility may lead to ineffective top-down or non-locally-owned bottom-up initiatives aimed to address the communities’ security challenges.

Lived Experiences of Insecurity

Consistent with the findings mentioned above, the authors re-place violent extremism in a broader “complex web of drivers of insecurity” perceived as “mutually influencing and/or reinforcing” (p.ix). Interestingly, respondents rarely identified al-Shabab as their main source of insecurity. Instead, the security sector, street gangs and other forms or criminality were identified as key insecurity drivers. Regarding the former (the role of the security actors), “police harassment, corruption and extortion […] alleged cases of forced disappearances and extra-judicial killings by security forces.” (p.10) were reported by respondents across locations, in particular in Nairobi, as a main source of insecurity.

These findings therefore question the dominant narratives and the increasing focus on violent extremism among policy-makers, donors and CSOs. Violent extremism should be analysed along with other obstacles to peace and security in Kenya and the Greater Horn at large. Concealing and overlooking – intentionally or not – other, more localised sources of insecurity will prevent progress towards a sustainable peace and stability in the region.

While the report identifies the mentioned phenomena as drivers or sources of insecurity, it seems to be critical to acknowledge these dynamics are also symptoms of socio-economic and political marginalisation of these areas. The following findings focus on the drivers of violent extremism and start elucidating perceived linkages between insecurity root causes, drivers and symptoms.

Perceived Violent Extremism Drivers

This section maps drivers that trigger, escalate or sustain “the emergence and spread of violent extremism in Kenya” by referring to “the alignment of various structural, socio-cultural and individual ‘push’ and ‘pull’ factors”, a commonly used analytical framework. Push factors are defined as “the structural and socio-political conditions which favour the rise and spread of armed extremist groups, and those sometimes used by these groups to create propaganda narratives”. Pull factors have “a more direct influence on the individual and are associated with the personal rewards an individual may gain through membership in an extremist group”.   In order to distinguish further these factors, the authors consider the push and pull factors according to three analytical levels, at the structural/macro level (push factors), meso level and micro level (pull factors). Cross-cutting dynamics and dynamics that are specific to one or more area-s are explored.

At the structural/macro level, respondents perceived socio-economic marginalisation as pushing individuals to join armed groups. This entails unemployment and poverty – particularly affecting the youth across locations; underdevelopment – in particular in Majengo, Mombasa and Garissa; and an entrenched sense of alienation. Harassment, extortion and extrajudicial killings by security forces were also seen as a structural factor driving individuals to join al-Shabab, as they nurture a sense of injustice and desire for revenge. While it is not specified in this section, this last cluster of drivers echoes the previously mentioned systemic nature of insecurity, with police brutality and harassment seen both as one of the main direct insecurity sources and as a driver of violent extremism.

At the meso and micro levels, eroding family structures; financial compensation; quest for status and sense of belonging; coercion; and religious messaging (“by offering the individual a sense of empowerment, duty and potential reward”, p.20) were perceived as pull factors. Interestingly, only a few respondents, mostly in Mombasa, identified religious messaging as a driver of VE. They insisted on the need to distance their religion from being associated with non-state armed groups labelled as extremist, and affirmed that al-Shabab’s references to religion are opportunistic and their interpretations fallacious. This purposeful detachment was interpreted by the research team as a reactive attitude to the increasing profiling of Muslim Kenyans as sympathetic to extremism.

Despite an increasing available literature on the drivers of violent extremism, one of the main methodological challenges of the study seems to be the articulation of these factors in a nuanced enough way to inform specific programming. While this multi-factorial analysis provides insights into local perceptions of what drives violent extremism, the push and pull framework remains restrictive. This typology is indeed increasingly questioned: one, push and pull factors tend to overlap making any distinction effort superficial; two, this framework does not help understand why individuals, experiencing similar socio-economic and insecurity issues, do not join armed groups labelled as extremist.[iii] To strategically inform relevant CVE programming, and peacebuilding programming, other typologies can be applied. A typology developed by the Royal United Services Institute (RUSI), for instance, focuses on structural motivators (e.g. social marginalisation), individual incentives (e.g. financial compensation) and enabling factors (e.g. eroding family structures).[iv]

Perceived Recruitment Strategies  

With the aim to understand further the recruitment processes, local perceptions of recruitment strategies, deployed in particular by al-Shabab, were collected across the research collections. False promises (“promises of a better, more rewarding life”, p.22) and coercive misguidance were identified by the respondents across all research locations as a leading recruitment tool employed by armed groups labelled as extremist. Respondents also mentioned the rhetoric used by the recruiting groups who capitalise on identity-based feelings of humiliation and oppression (“the exploitation of corruption in state and society; the incitement of religious or ethnic sentiments and feelings or marginalisation”, p.22).

Consistently with previous findings, only a few respondents, in particular from Garissa and Mombasa, cited the role of extremist religious discourse as a recruitment strategy. Diverse underground channels used to circulate extreme ideas were identified, including social networks and person-to-person discussions. Respondents insisted on the fact open discussions on al-Shabab are increasingly rare, with communities fearing to be associated with them.

“So what?” Conclusions from a Peacebuilding Perspective

This research builds on and strengthens the evidentiary basis informing CVE programming and policy, thus contributing to sharpening existing knowledge on a highly sensitive and complex phenomenon. While violent extremism is not exclusive to any region, nationality or system of belief, this research’s findings suggest that a systematic, locally-owned, context-sensitive and multi-sectoral approach is necessary to effectively address the complex web of drivers and diverse recruitment strategies that lead individuals to join armed groups labelled as extremist.

This research underlines the importance of conceptualizing violent extremism within a broader peacebuilding and human security framework. The gap between dominant narratives, seeing violent extremism as the most pressing issue, and local perceptions of violent extremism, seeing it as one among other security concerns, implies that violent extremism will not be sustainably addressed without tackling communities’ broader security concerns.

While violent extremism is primarily experienced at the individual and local levels, it remains a transnational phenomenon. A comparative analysis of local perceptions of violent extremism in the Greater Horn region would inform CVE approaches further and would contribute to promoting sensitive and effective initiatives at the regional level.


Flavie Bertouille is LPI’s Regional Programme Learning Advisor covering Kenya, Somalia and Sudan


[i] Accessible online

[ii] See Shetret L., Schwartz M. and Cotter D., 2013, Mapping Perceptions of Violent Extremism. Pilot Study of Community Attitudes in Kenya and Somaliland. Center on Global Counterterrorism Cooperation. Accessible online.

[iii] Search for Common Ground, 2017, Transforming Violent Extremism. A Peacebuilders’ Guide. Accessible online.

[iv] Ris L. and Ernstorfer A., March 2017, Borrowing a Wheel, Briefing Paper, p.13. Accessible online.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.