✶ The Phenomenology of Certainty: Why Felt Sureness Is Not Evidence

The Credencing framework evaluates whether assigned confidence answers to evidence. Its existing papers distinguish Objective Evidence, E0, Perceived Evidence, EP, Assigned Credence, CA, and Deep Rationality, SD; separate evidence-processing failures from belief-integrity failures; defend warranted uncertainty; diagnose rationalization; scale credencing to institutions and AI-mediated inquiry; distinguish raw from excess gaps; make posteriors answerable to evidential history; credential frames and categories; distinguish compatibility from confirmation through likelihood literacy; discipline priors through base-rate literacy; separate evidential support from explanatory satisfaction; model epistemic akrasia as costly update lag; and analyze assertion as a socially pressured interface between private confidence and public speech. This paper adds a missing phenomenological layer: felt certainty. Agents do not merely weigh evidence, compare likelihoods, consult priors, prefer satisfying explanations, resist costly updates, or overstate claims in public. They also undergo experiences of sureness, obviousness, closure, settledness, inevitability, and "of course." These experiences can be useful metacognitive signals. But they can also function as pseudo-evidential signals when fluency, narrative closure, identity fit, emotional relief, or social reinforcement is misread as support for the proposition itself. The lacuna is therefore the underformalized distinction between warranted confidence and the felt experience of certainty. This paper formalizes felt certainty, certainty-source profiles, evidence-trace quality, certainty misattribution, pseudo-evidential certainty, the phenomenological-evidential gap, certainty-based confidence inflation, warranted phenomenological slack, diagnostic confidence, source-ablation tests, calibration checks, responsibility assessment, and repair. It argues that certainty should be treated as data about an agent's cognitive and affective state before it is treated as data about the world. The practical result is a certainty audit: state the claim, reconstruct the evidence trace, estimate warranted confidence, estimate felt certainty, identify the source profile of the feeling, test whether the feeling is being entered as evidence, compare live rivals, vary non-evidential sources, check social reinforcement and calibration, grant humane slack, and repair by separating "I feel certain" from "the evidence warrants certainty." Credencing thereby gains a missing discipline: the force with which a proposition feels settled is not itself what settles it.