contradiction: the geometry of structural conflict
when what a system says and what it does cannot be jointly satisfied
abstract
this paper introduces the contradiction trap — a dialectical and game-theoretic mechanism for detecting structural bias, motivated asymmetry, and narrative drift in institutional and algorithmic decision systems. grounded in epistemic game theory, the trap recasts contradiction as a falsifiable evidential event: whenever a system’s stated rationale and observable behaviour cannot be jointly sustained, the resulting inconsistency becomes a measurable signal of underlying deviation. formalised as a one-move, strictly competitive epistemic game in which every admissible response incurs coherence loss, the trap generates informational payoffs that convert contradiction into diagnostic evidence. the framework provides a portable audit instrument for domains that claim impartiality but exhibit asymmetric behaviour — governance, organisational reasoning, and algorithmic architectures alike. by treating inconsistency not as a logical failure but as a data-bearing phenomenon, the contradiction trap establishes the epistemic foundations of the mathematics of integrity: a unified evidential paradigm in which legitimacy is demonstrated not through assertion, but through resistance to structured, adversarial challenge.