Audit Methods
P/A Institute treats philosophical concepts, classifications, evaluations, and inferential systems as revisable objects of analysis. The following methods are applied across the Institute’s research programs.
Hyperflat Analysis Audit
Places researchers, humans, AI systems, instruments, laboratories, corporations, states, electrical infrastructures, cities, resources, and ecosystems on the same audit plane — without erasing consequential differences. Audit symmetry does not entail ontological symmetry.
Boundary Audit
Audit of the boundaries drawn by a configuration: what is included, what excluded, under what conditions the boundary holds, and who controls its placement. Examines how boundary decisions distribute visibility, access, accountability, and analytical reach.
Ontological Deletability Audit
If deleting a realist commitment leaves prediction, intervention, error attribution, reproducibility, and model revision unchanged, that commitment is operationally deletable for the specified task.
Explanatory Added-Value Test
A commitment may be provisionally retained if it produces independently testable gains unavailable from configuration-constrained accounts alone: novel risky predictions, new counterfactual discriminations, improved intervention design, discovery of previously inaccessible phenomena. Elegance, simplicity, or unification alone is insufficient.
FED-DLR
Core operations: contradiction detection, determinate negation, directional justification, free generation. Evaluates whether configuration transitions expand or restrict possibilities of freedom. FED-DLR does not produce a single score; it produces a structured profile of transition direction across multiple dimensions.
Generative Rate Family
These rates measure distinct forms of generation, maintenance, revision, freedom, reproduction, and transformation. They are not collapsed into a single aggregate score.