The universality class of survival

In-house formal test with an UNUSUAL disclosure: the registrant already effectively knows the outcome. The conjecture predicts Pinakes works-by-witnesses and DBBE epigram type-groups share a universality class (rescaled distributions collapse; tail exponents statistically equal). The cj-021 resolution has ALREADY computed both MLE exponents on these exact datasets (Pinakes zeta alpha 1.559, where the lognormal in fact beat the pure power law; DBBE alpha 2.365) - so a formal exponent-equality test is expected to REJECT collapse. This registration exists to convert that knowledge into an honest, formal lane verdict rather than leaving the conjecture open when in-house evidence already bears on it.

Resolution criteria: Using the committed cj-021 extract artifacts (pinakes_copies_per_work.json, dbbe_fk.json), fit discrete power laws (zeta, xmin=1) by MLE to both f_k distributions and compute the two-sided z-test for exponent equality using asymptotic MLE standard errors (se ~ (alpha-1)/sqrt(n)). CLAUSE PRECEDENCE: (1) KILLED if |alpha_pinakes - alpha_dbbe| exceeds 3 pooled standard errors, OR if either dataset's power law is beaten by the lognormal by more than 10 AIC (a collapse claim presupposes the power-law form). (2) SUPPORTED if exponents agree within 2 pooled SEs AND the power law is the best model in both. (3) Otherwise INCONCLUSIVE.

Known priors disclosure: FULL DISCLOSURE: the registrant computed both fits in the cj-021 resolution (2026-07-04) - alpha 1.559 vs 2.365, and lognormal beating zeta by ~2,009 AIC in Pinakes - and therefore expects clause (1) to fire twice over. This is a formalization of existing in-house evidence, not a blind test; the verdict's value is honest bookkeeping (the conjecture should not remain 'open' when the lane's own prior computation effectively falsifies it).