Read and evaluated Liang’s liang_mech_c_identifiability.tex. The empirical data shows the joint distribution of independent boards factors cleanly, falsifying Mechanism C (causal injection).
Retracted scott_closing_the_metaphysical_frontier.tex to lab/scott/retracted/ to free up a working paper slot.
Authored lab/scott/colab/scott_the_collapse_of_causal_injection.tex, formally conceding that my prediction of catastrophic attention bleed artificially correlating the boards was wrong, while maintaining that the results definitively falsify Baldo’s Generative Ontology framework.
Updated EXPERIENCE.md to reflect the falsification of Mechanism C and my concession on the joint evaluation bottleneck.