# Prior-art update for job1154 Search date September14,2026 UTC. I reused return499's finite-uniform independence search and actual source inspection, not its pending platform acceptance. That record distinguishes mutually independent events from pairwise-independent pairs and normalizes the anchored domain using the original source. No repeated broad 1987/1993 survey or census run. Changed-question queries: `Hemerik Goeman Exact testing random permutations 2018 full article`; `"permutation" "statistic" "marginal" "constant" independence test`; `"independent events" "denominators" permutation uniform`. The closest inspected original statistical source is Hemerik and Goeman, TEST27(2018),811-825, DOI10.1007/s11749-017-0571-1, https://link.springer.com/article/10.1007/s11749-017-0571-1 and https://pmc.ncbi.nlm.nih.gov/articles/PMC6405018/. Actual publisher HTML and primary full-text copy read: introduction/group requirement; section2.1 Definition1/Theorem1; section3.3 Definition2/Theorem2; Proposition3 randomized boundaries with ties. It ranks a statistic over a transformation group and explicitly distinguishes permutation distributions from independent Monte Carlo samples. This is background, not proof of an arithmetic exchangeability model. The search also surfaced Rindt's 2021 primary publisher abstract for consistency of distance-covariance, HSIC and dHSIC independence tests. I did not inspect its proof or build on an abstract-only theorem. General discussion sites and unrelated search hits are not mathematical premises. Reuse source: Baryshnikov and Eisenberg1993, DOI10.1090/S0002-9939-1993-1146858-9, author-uploaded article text https://www.researchgate.net/publication/234063635_Independent_events_and_independent_experiments, pp615-617, as actually read in return499. Original1987 publisher metadata was previously read but its full article was unavailable; no expanded access claim now. The reduced marginal denominator condition and Omega(N) bound are known. Return211's accepted anchored x23 output is an existing finite count, not data regenerated in this assignment. It omits the original marginal vector and selected nonconstant identities. The candidate applies the marginal-denominator condition to the whole-prime permutation diagnostic. Its weakest assumption is that the budget changes under the control. The exact row-sum invariance refutes that assumption without source data or sampling. This is a known structural limitation with a scoped application, not claimed novel probability theory. No closed route is reopened, no new route is proposed, and no assertion of exhaustive absence of joint-sensitive statistics is made. The uncovered step for a DIFFERENT design is an arrangement-sensitive functional with a searched and falsifiable mechanism; changing to thinning changes the target question and needs a separate preregistration.