Empirical Intelligibility in Physically Non Selective Theories

This paper analyzes empirical intelligibility as a necessary condition for physical theory, distinct from empirical adequacy, predictive success, or mathematical consistency. It shows that a theory may be empirically non-selective—compatible with all possible observations—yet fail to be empirically intelligible, and therefore fail to qualify as an admissible physical description. Within a closed admissibility framework, intelligibility requires that empirical outcomes can be meaningfully related to the theory’s representational structure in a standing-preserving way. When a theory licenses every outcome indiscriminately, no observation can fix reference, discriminate claims, or constrain description. In such cases, empirical success becomes vacuous: observations no longer inform, test, or refine the theory because nothing is excluded. The analysis is eliminative. No alternative theory class is proposed, and no empirical criteria are introduced. Instead, the paper establishes a boundary condition: physical theories must be empirically intelligible, not merely empirically compatible. Non-selectivity is shown to undermine intelligibility by dissolving the link between observation and representational commitment. By separating intelligibility from adequacy, the paper dissolves a class of debates in which unfalsifiable or maximally permissive theories are defended on the basis of observational consistency alone. It closes these debates by showing that empirical non-selectivity is not a virtue or a limit case, but a violation of the conditions under which empirical description is possible at all.