The Work That Works

This volume gathers, in one continuous manuscript, the work from my 2025–2026 research programme that survives its own audit. The rule of admission is strict and applied uniformly: a result appears here only if it is a proved theorem, a validated empirical claim priced in bits under the frozen Coates PIT meter, or a methodological tool stated honestly, with its limitations in the text rather than left for referees to find. Everything that failed — refuted predictions, overfitted models, claims resting on free parameters or post-hoc selection — is excluded from the body and recorded by name in the Honest Accounting, with the reason for its exclusion and, where possible, its price in bits. A record of what failed is part of what works.

The manuscript is organised into six parts. Part I contains the algebraic foundations: the two boundary theorems singling out c = 2 among order-2 linear recurrences, the inverse-difference identity in SL(2,ℝ), the Klein four-group synthesis at λ = 2, the companion-matrix machinery, the fixed points and period-two orbits of the trace map, the Metallic Thread selection principle, and the Jacobsthal eigenvalue at the photonic band edge. Everything in this part stands on proof alone. Part II gives the complete treatment of the two sequences that carry the boundary in concrete form — the Lichtenberg sequence and its Diamond Lattice geometry, the Book of Jacobsthal in its three parts, and the cyclotomic trace-shadow of the higher-order Jacobsthal family, with its one-way shadow of the Gauss–Wantzel constructibility hierarchy. Part III presents the measurement instruments: the corrected FSRE/CANLE toolkit, the exact octave-coverage computation that prices any 2^a3^b claim at a declared budget and tolerance, the quadratic-ladders reference with its worked null test, and the transparency study that names curve fitting as curve fitting. Part IV reproduces the empirical programme's companion study: the cross-domain test of the P1/P2 partition offsets in Bacteria, in which a qualitative replication, a quantitative non-replication, and a mechanism-dissociating counterexample are reported at the same size. Part V contains the physics: the neutral-scalar spectral-escape theorem for ultracold Kerr–Newman–de Sitter families, the thermodynamic metallic companion with its exact selector and its three negative diagnostics, and the QPO trace-boundary mechanism, retained with an explicit caveat about an unpriced selector. Part VI closes the volume with mathematics as narrative, because explaining a thing plainly is the final test of understanding it.

I am an independent researcher without institutional affiliation or funding. The standards applied to others' work throughout this programme are the same ones applied here to my own; several of my own results were withdrawn, refuted by my own pre-registered tests, or downgraded on audit, and the reader will find each of those outcomes stated rather than hidden — including, in the Honest Accounting, claims of mine priced at a fraction of a bit. What remains is smaller than the corpus that produced it, and that is the point.

This work is dedicated to the memory of Shirley.