A Constraint-Modulated Rate Law at the Molecular Scale: Viscosity-to-Dielectric Geometry Transfer in Glass-Forming Liquids

This dataset reports a controlled cross-observable test of the constraint-modulated rate law (CPA + C) in glass-forming liquids. The central question is whether a constraint modulation inferred from salol viscosity carries predictive content for independent dielectric alpha-relaxation. The primary salol viscosity analysis treats the 35 published Laughlin–Uhlmann viscosity symbols as equal inferential and bootstrap units. The fit gives T_lock = 176.97 K and T_ref = 234.01 K. These quantities and the resulting constraint modulation C_eta(T) are frozen for transfer to dielectric relaxation under Q2b, with dielectric-specific coupling coefficients. Against a denominator-only control carrying the same frozen T_lock, Q2b gives ΔAIC = 16.32, with ordinary- and alternative-count ΔAICc margins of 14.04 and 12.32. Residual error falls from 0.090 to 0.036 in log10 relaxation time. The stricter single-scale transfer, Q2a, does not outperform the same control, so the supported relationship is transferable constraint modulation rather than identical flow-reorientation scaling. Across 1,000 joint evaluations propagating uncertainty in the viscosity-derived T_lock and T_ref, Q2b is favored by AIC, ordinary AICc, and BIC in every case, and in 998 of 1,000 under an alternative parameter count that includes the residual variance. Direct dielectric fits place CPA + C first in the broad discriminating window for salol (n = 21, ΔAIC = 27.5 over Avramov–Milchev). The design is the contribution: a constraint modulation estimated from one observable is frozen and tested on another observable of the same liquid, against a control carrying the same denominator, which isolates the transferred modulation from the shared divergence structure. No precedent for this controlled architecture was identified in the accompanying literature search. Exploratory dielectric-only comparisons for glycerol, propylene carbonate, and benzophenone are preserved in the deposit with their data, code, and outputs; they supply no matched viscosity data and are not interpreted in the accompanying note. The cross-observable test was identified in version 6 of Dynamic Present Theory I: Foundations of Continuous Present Actualization (https://doi.org/10.5281/zenodo.20798793) before the transfer results were known. The corrected inferential and resampling unit was specified, frozen, and hash-recorded on August 3, 2026, before the 35-symbol execution. The earlier 95-coordinate analysis had already been executed and independently validated. The salol analysis was independently reproduced by Dr. Christian Eike Precker and validated with qualifications. The 35-row primary viscosity input is his independent digitization of the source figure, performed with the deposited coordinate record withheld, and the frozen input is byte-identical to that artifact by SHA-256. The record includes the original independent validation report, the signed 35-symbol addendum, and its machine-readable comparison file. It also contains the manuscript and LaTeX source, data, code, complete bootstrap and joint-propagation outputs, figures, protocols, provenance records, reproducibility witness run, and SHA-256 manifest needed to reproduce or challenge the reported results.