Reflective Equilibrium

Reflective equilibrium is the methodological procedure, central to John Rawls's moral philosophy, in which one iteratively adjusts both particular moral judgments and general principles until they cohere with each other and with relevant background theories. A judgment in reflective equilibrium is one that has survived scrutiny from multiple angles: it is consistent with one's other considered judgments, compatible with the principles one endorses, and supported by the factual and psychological theories one accepts. The procedure rejects both the untutored intuition (which may be distorted by bias or circumstance) and the top-down deduction from abstract axioms (which may be insensitive to particular cases). Instead, it treats moral knowledge as a network adjustment process: local revisions propagate until the whole system stabilizes. This is structurally analogous to belief revision in Bayesian inference and to weight updates in neural networks — both are procedures for resolving tension among a set of commitments by distributed adjustment. The difference is that reflective equilibrium includes an explicit normative layer: some revisions are not merely possible but required, because the equilibrium is not just coherent but "wide" — tested against alternative conceptions and background theories. A narrow equilibrium is mere consistency; a wide equilibrium is resilience.