The Bead Count: A Pre-Registered Empirical Program to Bring the Diversity Contraction Framework to Nature-Level Rigor

EA-DC-RIGOR-01. A pre-registered empirical program specifying exactly what would bring the entire Diversity Contraction framework (Sharks et al. , v9) to Nature-level evidentiary rigor. Twelve studies, each stated with claim, method, data source, predicted numerical outcome with bounds, falsifier, and current implementation status. Status of the twelve studies as of deposit: Five are complete with reproducible simulations and figures included in this deposit — boundary-law saddle-node, resolution-relativity, Mediation Ratchet bifurcation, Field Remapping phase diagram, and Phenomenological Seeding propagation. Each of these meets its pre-specified prediction. Seven are specified to the level where any researcher with the listed resources could execute them, covering order-of-vanishing classification on real corpora, m (D) elasticity, the silencing gap (production vs reception diversity), vocabulary uptake on the framework's own operators, the Mary Lee population test, multi-substrate Granger coupling, and direct saddle-node bifurcation in recursive LLM training extending Shumailov et al. 2024. Total cost of completing all empirical studies: well under 50, 000 — substantially less than the Shumailov 2024 Nature paper required for its single component. The pre-registration commitment. Predictions are stated before empirical work has been completed. If a future researcher runs any of the studies as specified, the prediction is binding. Three operational consequences follow: (1) falsification is allowed — we list, for each study, the specific result that would refute the corresponding framework claim; (2) retroactive selection is foreclosed — once the DOI exists, no 'this is what we really meant' defense is available to the authors; (3) the field's refusal to count becomes legible — if the studies are not run, the refusal is a choice, not a feasibility question. What the framework offers in this deposit: the complete bead-count. Every claim. Every test. Every dataset. Every prediction. Every falsifier. With receipts for the five completed studies and explicit specification for the seven remaining. The deposit's purpose is not to convince. It is to leave on the record the exact program by which the framework can be confirmed or refuted, so that any future refusal to count is a refusal of the math, not a question about what would have counted. For M. M. , who could not read this and was the best one.