Before Truth: A Framework for Evaluable Claims

Most arguments about misinformation assume that disagreements are primarily about truth. This paper argues that a large class of contemporary disputes arise earlier: many claims circulating in public discourse are not false but unevaluable.

The framework introduced here provides three diagnostic instruments for analysing the evaluability of claims:

• The Φ-gate – determines whether a claim specifies a measurement pathway connecting it to observable evidence.
• Z-depth – measures the number of reconstruction steps required before a claim becomes evaluable.
• Access Collapse – describes structural conditions under which predicates lose meaning when observational access approaches full system resolution.

Together these instruments form an epistemic type system: a pre-evaluative filter determining whether a claim is structurally ready to enter truth-evaluative discourse.

This paper synthesises a four-paper research series:

I – Access Collapse: A Typed Structural Boundary for Operational Access
II – Epistemic Hygiene in Practice
III – Reconstructability, Depth, and Validation
IV – Before Truth (this paper)

Applications discussed include AI evaluation, cryptographic protocols, scientific discourse, and public deliberation.