Constraint-Driven Non-Absorbability in Bounded Transformation Systems: Separating Capacity Pressure, Encoding Failure, and Class-Relative Structural Distortion

This paper introduces a formal diagnostic framework for distinguishing capacity pressure, encoding failure, and class-relative structural non-absorbability in Bounded Transformation Systems (BTS). The framework models transformation as a constrained family of encoding operators. A system constraint vector K = (K_{\mathrm{rep}}, K_{\mathrm{comp}}, K_{\mathrm{audit}}) induces an admissible transformation space within the unrestricted operator universe. Structural preservation is evaluated through a normalized multi-invariant structural-loss function, while nominal capacity pressure is represented independently by the capacity pressure ratio \gamma = C_X/K_{\mathrm{rep}}. The paper establishes two central propositions. Proposition 1.1 demonstrates the logical separability of capacity pressure and structural non-absorbability: capacity pressure does not necessarily imply structural loss, and structural non-absorbability does not necessarily imply capacity overload. Proposition 1.2 establishes that non-absorbability is system-relative, depending on the declared constraint vector and admissible transformation class rather than being an intrinsic property of the source phenomenon alone. The resulting diagnostic framework defines four states, Q1–Q4, corresponding to the combinations of nominal capacity adequacy or pressure and structural preservation or non-preservation. An explicit INFEASIBLE state is additionally defined when the admissible transformation space is empty. A finite executable directed-cycle model and an accompanying PyTest verification suite provide Level 2 model-level computational verification of the formal specification, including the four diagnostic quadrants and the infeasibility mapping. The work explicitly maintains an epistemic boundary between mathematical consistency, computational verification, and empirical validation. The present artifact establishes Levels 1 and 2 only. It does not constitute empirical validation, causal identification, or evidence that the proposed diagnostic states necessarily occur in any particular real-world institution, organization, or artificial intelligence system. Frozen Specification v1.1.0DOI: 10.5281/zenodo.22019066