The Vulnerable Offline/Online and VEO's in Kenya


April 12, 2022

Executive Summary

As inherently social creatures, humans are natural storytellers, utilizing narratives to imbue meaning into their surroundings—a profound tool for the persuasion process. Narratives weave intricate connections, effectively shaping and solidifying power structures within both offline and online realms. These narratives possess potent mental constructs that influence diverse facets and tiers of human interaction. Moreover, narratives are context-dependent, serving a dual purpose of cultural organization and sense-making. Notably, they function as instruments, consciously or unconsciously employed to construct social orders in the context of violent extremism and terrorism.

Over time, both practitioners and academia have consistently linked narratives influencing violent extremism to religious ideology. However, this approach has inadvertently sidestepped the significance of religious interpretation, which serves as a conveyance mechanism for seamlessly transmitting information, effectively linking local issues to global discourses on violent extremism (VE). From an interpretive standpoint, violent extremists have harnessed religion as a self-contained system, offering an uncomplicated explanation for societal social, economic, and political processes. This entails the manipulation of specific texts, subtly endorsing violence as a legitimate action, contextualizing distinct discourses, and prescribing a mandate for action. Consequently, such interpretations also establish a division of roles, identification of actors, and validation of their actions.

While local spaces furnish the context and content for factual accumulation, religious interpretation functions as the conduit linking and associating these narratives to global discourses. As elucidated in the forthcoming review of this study, the Neglected Duty pamphlet encapsulates global discourses, justifying Jihad in its violent interpretation as a duty for Muslims. The texts highlighted in the report as the genesis of violent-extremist narrative framing processes ensure that conversations and actions within the “Muslim” community align with specific requisites.

The internet has emerged as a platform for narrative dissemination. However, it is often misconstrued as a driver of violent extremism rather than a medium. The ease of access to social media spaces has led to widespread usage in the Horn of Africa, although this platform’s utilization extends beyond connecting people and sharing thoughts; it also propagates false information. Privacy rules have further facilitated niche closure, amplifying the targeting of vulnerable individuals. Though these privacy rules are welcomed, they present challenges for prevention analysis.

It is worth noting that the increased utilization of the internet may partly result from private sector initiatives aiming to penetrate specific demographics through free services and associated benefits, intertwined with privacy. Social platforms have become increasingly accessible, user-friendly, and often free. When an account is deactivated or banned, users can swiftly migrate content to other platforms or accounts at minimal to no cost. The COVID-19 crisis has exacerbated the misuse of social media.

This report meticulously documents an analysis of radicalization and recruitment processes through a narrative review lens. Focused on modeling, interpretation, and the propagation of narratives across Mombasa, Kwale, Lamu, Tana River, Garissa, Isiolo, Nakuru, Kisumu, Kakamega, and Nairobi counties, the study reveals commonalities in narrative development, application, and dissemination, despite varying local realities and susceptibility to VE.

The study, qualitatively conducted, employed a purposive sample and snowballing technique to target diverse respondents: Maskani dwellers, returnees, direct victims, mothers of those who traveled to Somalia, university students, and those incarcerated due to VE-related incidents. A total of 224 participants were interviewed, with 70% being male and 30% female. Among the participants, 67% identified as Muslim, 32% as Christian, and 1% under the “other” category, spanning the ten counties. Additionally, the study drew from a comprehensive desktop review of existing research on narrative development and diffusion from the global sphere to local Kenyan communities.

The online dimension of the study focused on data collected from Twitter over a six-month research period, amassing five million tweets. These tweets were curated through keywords and hashtags linked to violent extremism and terrorism. While the study anticipated data collection from Facebook and Telegram, API restrictions prevented its completion. The Twitter data analysis employed second-level random sampling, selecting an average of 1,800 tweets for Social Network Analysis, Word-Cloud generation, and Sentiment Analysis each week for three days over the six-month period.

Key findings underscore that narrative formation hinges on three core attributes: a credible storyline, actionable plans for recipients, and the incorporation of religious cover. This religious aspect provides support to the persuasion process and bolsters the global connection. The persuasion process is particularly effective in online platforms or audiences, while contextual factors within communities enhance narrative credibility and relatability. Each narrative mandates a call to action and the designation of actors as either adversaries or allies. The role of religious interpretation then supplies the cover that links the provider to the recipient within the given context.

Furthermore, narrative sharing operates within goal-oriented frameworks structured around hierarchically ordered preferences. This study’s findings confirm that all narratives are imbued with intentions and objectives, their persuasiveness varying based on the targeted vulnerable individual and their degree of socialization.

Across the ten counties, four distinct narrative goals emerged: Recruitment narratives emphasize the benefits of affiliation, particularly economic gains; Sympathizer-targeted narratives leverage empathy and concern for the Muslim Ummah; Morale-building narratives employ audiovisual mediums to showcase heroism; and Jihad and Hijra narratives engage individuals already within the fold and willing to take action.

Moreover, narratives can be categorized for more targeted responses. This analysis led to the identification of four additional categorizations based on rewards offered to vulnerable individuals, encompassing the Takfiri model of self-reinvention, justification of engagement via perceived retribution and marginalization, and the call to establish the Islamic State.

Narratives of Jihad exhibit limited online linkages, suggesting that such conversations predominantly occur in person, with higher levels of socialization. Notably, narratives associated with Al-Shabaab show extensive networks, suggesting multiple nodes transmitting information to numerous individuals. Other Violent Extremist Organizations (VEOs) exhibit single-node linkages, indicating comparatively lower interest. Overall, narratives linked to terrorism showcase limited linkages, stemming from the lack of consensus on specific actions constituting terrorism.

In summation, responses from Garissa and Isiolo revealed that narratives fall broadly into three categories: Retribution-focused narratives, Reward-oriented narratives, and Self-determination narratives. In counties with “matured” conversations and heightened socialization processes—such as Mombasa, Lamu, Kwale, and Nairobi—the messaging largely centers on religious duty interpretation, justification for caliphate establishment, and the need for violent engagement. Counties like Kakamega and Kisumu presented limited narrative responses, possibly due to a higher Christian respondent ratio. Nakuru, relatively isolated from violent extremist attacks, features in the discourse as a transit point, possibly driven by political pressures.

In response, this report advocates for a framework for identifying and countering violent extremism. It underscores that:

  1. All developed narratives target vulnerabilities, often originating at higher contextual levels. Religious interpretation plays a pivotal role in linking narratives to global discourses, often facilitated by online platforms.
  2. As targeting shifts from high to low levels, the reliance on religious cover diminishes while contextual resonance factors become more prominent.
  3. Narratives, as they progress horizontally, encompass varying degrees of socialization, moving from low-level tasks aimed at proving loyalty and engagement to more intense actions in the context of higher socialization.
  4. While the messenger remains vital in the response process, alternative narratives require a comprehensive understanding of their formation, goals, and categorization.
  5. Across the medium spectrum, greater reliance on social media corresponds to higher targeting levels, while lesser reliance involves more in-person interactions with vulnerable individuals. As socialization advances, in-person outreach can occur in groups within social media spaces.
  6. Successful alternative narratives entail contextual facts, actionable guidelines, and relatable actors, pivotal for effective communication and goal achievement. Absence of these attributes renders narratives ineffective, failing to achieve desired outcomes.