What should we do when we disagree?

Abstract

The epistemology of peer disagreement is characterized by a tension between nonconformist and conformist views. Nonconformism encounters the “One against Many” problem, as it fails to explain why a large number of independent dissenting peers should eventually necessitate doxastic revision. Conversely, conformism faces the “Many against One” problem, granting disproportionate epistemic weight to a single dissenting peer even when numerous other peers agree. A justificationist account resolves these difficulties by asserting that the required degree of doxastic revision tracks the level of justified confidence already possessed by the agent’s belief. In cases where a belief is highly justified, an agent’s personal information regarding their own cognitive state—such as awareness of their own sobriety or mental focus—functions as a symmetry breaker that permits the downgrading of a dissenting peer’s epistemic status. This framework also accounts for “protected” beliefs, wherein a highly justified auxiliary belief can insulate a specific, less justified target belief from the impact of disagreement. Ultimately, the epistemic significance of disagreement is variable rather than fixed, depending on the original justificatory resources available to the agent. This approach provides a unified explanation for the rational requirements of disagreement across diverse domains, from simple calculations to complex medical and theological disputes. – AI-generated abstract.