A Unified Framework for Moral Epistemology under Uncertainty

Thesis Abstract

This study examines how moral epistemology can be unified under conditions of epistemic uncertainty, addressing the persistent fragmentation between foundationalist, virtue-epistemic, and contextualist accounts in ethical judgment and moral reasoning. The central problem is to develop a coherent framework that integrates epistemic virtues, evidence-sensitive moral reasoning, and pragmatic constraints to explain how agents form morally credible beliefs when sources of moral evidence are incomplete, conflicting, or probabilistic. The aim is to formulate a unified framework that delineates principled criteria for moral justification, belief revision, and normative guidance under uncertainty, and to test its coherence against competing theories. The objectives are (1) to articulate a formal model that combines elements from Bayesian epistemology, virtue theory, and moral rationalism into a single evaluative apparatus; (2) to identify the role of epistemic virtues (such as intellectual humility, conservatism, and open-mindedness) in moral belief formation under uncertainty; (3) to specify conditions under which moral beliefs should be revised in light of new, uncertain evidence; (4) to operationalize the framework through a set of normative decision criteria and decision-procedure rules; and (5) to evaluate the framework against empirical data on moral judgments and policy recommendations in uncertain scenarios. The methodology adopts a mixed-methods design. The theoretical component constructs a formal model drawing on Bayesian updating, justification theory, and virtue-epistemic norms, to yield a calculable framework for moral epistemic status under uncertainty. An empirical component deploys a cross-sectional survey of 420 adult participants, stratified by age, education, and cultural background, complemented by 30 semi-structured interviews with philosophers, ethicists, and cognitive scientists to calibrate the framework to professional practice. Data collection employs a standardized moral dilemma battery, uncertainty salience manipulations, and measures of epistemic virtues, cognitive styles, and moral confidence. Instrument validity is established through expert review and pilot testing (n=60), with reliability assessed via Cronbach’s alpha for multi-item scales (target ? ? .80) and test-retest reliability (r ? .75). For the empirical analysis, Bayesian hierarchical regression will be used to model the relationship between uncertainty, epistemic virtues, and moral confidence, while thematic analysis will be applied to interview transcripts to extract normative criteria and calibration factors. Model specification includes a latent variable for moral justification strength and a pathway model linking uncertainty, evidence quality, epistemic virtues, and moral action likelihood. Robustness checks will involve sensitivity analyses with alternative priors and cross-validation. Expected findings indicate that incorporating epistemic virtues into a Bayesian moral framework yields higher predictive accuracy for moral judgments under uncertainty than baseline deontological or consequentialist models. It is anticipated that intellectual humility and evidence sensitivity will significantly modulate belief revision in response to uncertain moral evidence, reducing error rates in moral judgment during ambiguous dilemmas by 12–18% relative to non-virtue-informed models. The study also expects context-dependent shifts in justification strength, with professional normativity (as expressed by ethicists and cognitive scientists) strengthening convergence toward the unified framework in high-uncertainty domains such as bioethics and public policy. Contributions to knowledge include (i) a formal, integrative model of moral epistemology under uncertainty that synthesizes Bayesian reasoning with virtue-epistemic norms; (ii) an empirically validated set of criteria and procedures for moral justification and belief revision under uncertain evidence; and (iii) methodological guidance for deploying mixed-methods research in moral epistemology, including instrument design and analytical pipelines for Bayesian and thematic analyses. The main conclusion anticipated is that moral justification under uncertainty is best achieved by a unified framework that specifies how epistemic virtues interact with evidential quality to govern belief formation and revision, producing more reliable and normatively defensible moral judgments. Practical recommendations include training programs to cultivate epistemic virtues among policymakers and ethicists, and the adoption of the framework as a decision-support tool in ethically uncertain professional contexts, such as biomedical research governance and climate ethics policy.