The content instrument extracts and scores ontological features from transcendental narratives using a fixed three-layer protocol. Its output is a set of ontological mappings that can be compared directly across frameworks.
This instrument is reported here for completeness. No standing on the ladder rests on it, and none of the three findings on the poster uses it. The ontology-comparison results it produces are deliberately kept off the board.
Three-layer protocol
PrimeInternal coherence measured by contradiction count, corpus-size factor, and reality-axes coverage gap.
ForensicHistorical degradation across six dimensions, applied as a multiplier.
RINCombines the preceding layers into a final Reality Integrity Number for ranking.
Cross-model reproducibility
Each framework is scored independently across frontier-model families from separate developers. Cross-model agreement of the resulting scores is approximately r = .92.
Cross-model agreement establishes that the ranking is stable across architectures rather than idiosyncratic to one scoring system. It is reliability, not validity, and not independent replication. A protocol can be perfectly reliable and measure nothing: every rater applying the same wrong rule identically produces flawless agreement. High agreement licenses the analysis; it does not support the conclusion.
Why that distinction matters →
Relationship to the second instrument
The order test and the content test are methodologically separate and based on non-overlapping corpora, sharing no coding scheme. Each reported value states which corpus it is computed on.
Because the two instruments measure different properties against different nulls, no combined probability across them is computed or reported. They are presented as separate lines, each defensible alone.
Convergence between two instruments built by the same author, run by the same operator, on corpora drawn from the same literature is convergent internal evidence. It is real and worth having. It is not what “independent” means elsewhere on the ladder, and it is not scored as though it were.