Room to Maneuver: Rebel Organization and Ideological Flexibility in Insurgencies

This dissertation examines why some rebel groups adjust their ideological claims during civil conflict while others remain firmly committed. Ideology is a central component of armed organizations: it presents their worldview, identifies a constituency, and shapes the goals and means they endorse. Changes to ideology carry consequences for patterns of recruitment, civilian targeting, prospects for negotiated settlements, and the direction of postwar political transitions. In contrast to existing instrumental and normative approaches that conceptualize ideology as either highly mutable or sticky, respectively, I theorize that flexibility will vary based on rebel organizational structure.

I argue that a rebel group's capacity for ideological change is determined by the interaction of two organizational features: the level of leadership personalization, and the degree to which it has socialized its members into a shared ideology. Personalization captures whether a single leader holds unconstrained decision-making authority or whether power is distributed across councils, coalitions, and veto players. Socialization describes the investment a group has made in indoctrinating members through political education, recruitment screening, and organizational rituals. Together, these variables generate a typology of four rebel organizational types. Cults of personality combine one-person rule with high socialization and tend toward ideological lock-in. Warlords and charismatic mercenaries have personalized leadership but low socialization and can shift claims opportunistically. Constrained ideologues combine distributed authority with high socialization and can achieve genuine change, but only through slow internal deliberation and public legitimation. Ambiguous generalists combine distributed authority with low socialization and tend to produce competing ideological currents or stasis.

To examine these dynamics, I conduct longitudinal case studies combining qualitative process tracing of internal ideological debates with computational text analysis of each group's public-facing statements. The three primary cases — Sendero Luminoso in Peru, the SPLM/A in Sudan, and Hezbollah in Lebanon — trace an archetypal cult of personality, warlord-type group, and constrained ideologue respectively. A fourth empirical chapter examines four smaller cases through paired comparisons (LURD and the RUF in West Africa, and the OPM and Fretilin in Indonesia), illustrating how ambiguous generalist groups produce fragmentation or stasis, and how shifts in organizational structure open or foreclose the possibility of ideological revision.

This dissertation makes three primary contributions. First, it develops an original theory of rebel ideological flexibility centered on organizational structure, thus helping to explain why the same external pressure produces divergent ideological responses across groups. Second, it brings research on institutions and regime type into conversation with scholarship on civil wars, showing how the structures around rebel leaders shape dynamics of factional competition, legitimation, and institutional path dependency that in turn impact ideology. Third, it introduces a computational pipeline using large language models to measure ideological stability across six dimensions, which can be replicated with other organizations that produce public-facing statements over time.