The silent transmission belt: testing safety bullying's indirect effects on construction incident learning failures via voice suppression

Purpose This study aims to conduct a theoretical reconstruction of the phenomenon of failure to learn from incidents (FLFI) in construction projects by conceptualizing safety bullying (SB) not as episodic interpersonal misconduct, but as an institutionalized governance mechanism that structurally suppresses voice. We examine how project participant silence (PPS) operates as the transmission channel through which coercive control cultures translate managerial pressure into learning breakdowns, thereby challenging dominant explanations that attribute learning failure primarily to weak implementation or individual error. Design/methodology/approach Data from 215 valid questionnaires were analyzed using partial least squares structural equation modeling (PLS-SEM). The structural model's stability was verified via a bootstrapping procedure with 3,000 resamples. Findings PPS exerts a significant and substantial effect on FLFI. SB shows no direct impact on learning outcomes but operates entirely through PPS, indicating a full mediation effect. This indirect-only pattern reveals that learning failures emerge when governance-induced silence distorts feedback loops and fragments safety knowledge flows. Originality/value The study advances theory by reframing learning failure as a systemic managerial pathology rather than an implementation deficit. By identifying PPS as a structural inhibitor embedded in temporary project governance, it bridges organizational silence research with resilience engineering and extends high-reliability organization theory into project-based contexts. Practically, it shifts managerial focus from correcting frontline errors to redesigning silence-producing governance structures.