Publication history: 2026-08-27

Summary

New rows TRANSFER.HIDDEN.CIRCULATION.01 and TRANSFER.COHERENT.WILSON.01, both “executed half [E] / selection open [O]”; the exploration probes transfer_ birkhoff_ circulation_ probe.py 42/42 and transfer_ coherent_ wilson_ probe.py 28/28 stay in experiments/tfpt-discovery/ as the discovery record). (1) v976 — the Birkhoff fibre and the entanglement-breaking no-go. The Birkhoff (random-permutation) fibre of the seam single step B (exact sixth root of the v221 kernel, B = \tfrac12 I + \tfrac1{18}P_{12} + \tfrac29 P_{13} + \tfrac29 P_{23} bit-exact) is exactly a one-parameter line t \in [0, 1/18]; the hidden coordinate is the sign-character Fourier coefficient \widehat w_t(\mathrm{sgn}) = 6t, populations are exactly t-free, and no fibre member is chiral (w(C_{123}) = w(C_{132}) forced). The one-step coherence spectrum is \{1, 6t, (2/3)^{\times 2}, (1/3)^{\times 2}\} and the triangle loop current is an exact eigen-observable \Psi_t(A) = 6t\,A — one step plus one current readout measures t without an ancilla. On the contrast qubit the Choi partial transpose has spectrum \{-3t/2,\ 1/6 + 3t/2,\ 1/3 + 3t/2,\ 1/2 - 3t/2\}: the deployed sixth-root spectrum sits exactly on the entanglement-breaking boundary because 2/3 + 1/3 = 1 (mutated spectra miss by \mp 1/12); at the full three-level the family stays NPT at every tested iteration with \lambda_{\min} = -(1/3)(2/3)^n, t-independent. The no-go that sharpens DYN.UNITARY.DILATION.01: the deployed Kraus dilation K_{ij} = \sqrt{T_{ij}}\,|i\rangle\langle j| is measure-and-prepare, hence entanglement breaking with quantum capacity 0 and C_0(T) = 0 exactly — any dilation closing that contract must be a genuinely different object. Shannon KKT certificates: p^\ast = (1/2, 1/2, 0) is exactly capacity-achieving for both B and T (C(T) = 0.0084 bit, C(B) = 0.4890 bit, ratio 1.71\,\%) — the surviving classical message is binary on every scale. The defect–Birkhoff value coincidence (weights (9,1,4,4)/18 vs the v652 ladder g^2/18) is recorded typed [C] with its honest multiplicity audit; the forcing functor is the named open contract TRANSFER.DEFECT.BIRKHOFF.01. (2) v977 — the unistochastic SU(3) lift and the 5-12-13 Wilson code. B is unistochastic: an exact U \in SU(3) with |U_{ij}|^2 = B_{ij} entrywise, angles forced by B (\sin^2\theta_{13} = 2/9, \sin^2\theta_{12} = 1/14, \sin^2\theta_{23} = 2/7), \cos\delta = -4/\sqrt{65} unique, lift set = \{U, U^\ast\} up to rephasing; Jarlskog J = \pm 1/27 exactly, and the real part of every Wilson plaquette is fixed by B through unitarity — the full rephasing-invariant Wilson data is classical up to the \mathbb Z_2 orientation bit \operatorname{sgn} J. The main plaquette is (-5 + 12i)/324, normalized (-5 + 12i)/13 with 5^2 + 12^2 = 13^2 (atoms: 5 = g_{\mathrm{car}}, 12 = \dim\mathfrak g_{\mathrm{SM}}, 13 = \Delta_Q = N(3{+}2i)); all nine plaquettes close on four Gaussian phase types with norms exclusively \{10, 13, 130, 169\}. The center bridge: \operatorname{disc}\chi_A = 65 = \det(A{-}I)^2 + (\det A/2)^2 = 4^2 + 7^2 for the v530 atom matrix, e^{i\delta} = (-4 \pm 7i)/\sqrt{65} = (1{\pm}2i)(2{\pm}3i)/\sqrt{65}, and the v530 sibling quotients fail the identity — the bridge selects the atom quotient. Rationality selection: J_n is rational only at n = 1 (the deployed step hits 3(1 + 2b - 3a^2) = 1 exactly, so J = 1/27); census over 238 valid family points: exactly one full-code point = the deployed (2/3, 1/3) (0.42\,\% \ll the preregistered 10\,\% kill). No fixed microstep: B^2 \neq |U^2|^2 and a Dirichlet recurrence witness (n^\ast = 967909) shows no single closed unitary generates the relaxing sequence — repeated steps require dephasing between steps, the six-step macro-gate U_6 \neq U_1^6 (the coherent sharpening of clock-vs-walk). The PMNS near-miss stays a negative control (240.2551^\circ vs 240^\circ; -1/27 \neq -0.02965; no PMNS derivation). Surfaces. Research-contracts document: the DYN.UNITARY.DILATION.01 section gains the executed lift-structure sharpening, the CCC reduced-channel prose and the two former “exploration-only” citations of rounds DCXCVI/DCXCVII are upgraded to \veri citations; origin_theory gains the honest “what the classical transport does not determine” paragraph next to the v221 recoverability code; tfpt_2 gains the 65 = 4^2 + 7^2 bridge corollary in the v530 center-quotient block. No status marker moves; both contracts stay open; no PMNS/CP/flavor claim.

Summary

The review diagnosis “chiral 1{+}1D seam net \not\Rightarrow local, interacting 3{+}1D quantum world” is implemented as eight new contract IDs in tfpt_research_contracts plus one new load-bearing module. (1) DIMENSION.UPLIFT.FIREWALL.01 (standing rule, in force): no 1{+}1D theorem closes a 3{+}1D claim; every contract closure states the spacetime dimension of hypotheses AND conclusion; audit census recorded (SEAM.EQUIV.01 is 1{+}1D in and out — even fully proved it is a boundary building block, not the 4D theory; the v957 kinematic uplift does not close it; the anomaly-inflow bulk is invertible/topological, never to be silently equated with physical spacetime); the realistic route ordering (seam as boundary module \to genuine 4D Euclidean lattice \to OS package \to chirality/no-mirror \to only then BW/CHM + entanglement gravity \to bulk–seam equivalence) is registered as the programme skeleton. (2) SEAM.BULK4D.RECON.01 [O]: the previously implicit seam\tobulk theorem made explicit (4D local algebras/Schwinger functions; locality/microlocality, Poincaré covariance + spectrum condition, spin-statistics + CPT, both helicities, clustering, nontrivial S-matrix, unique operator dictionary). (3) the dynamics split DYN.MARKOV.EMBED.01 / DYN.UNITARY.DILATION.01 (“three meanings of dynamics”: relaxation T^n / dissipative semigroup e^{t\log T} / unitary e^{-itH}): the finite kernel half of MARKOV.EMBED is executed [E]as v971_markov_embedding_generator.py (15/15, sympy-exact) — Q=\log T of the v221 seam kernel is an EXACT classical Markov generator: e^Q=T by spectral calculus, row sums symbolically zero, off-diagonals exactly \{\ln(9/8),\,2\ln 3,\,2\ln 3\}=\{0.117783,\,2.197225,\,2.197225\}>0 (the reviewer's independently verified numbers), uniformization certificate Q+(4\ln3)I\ge0 entrywise \Rightarrow e^{tQ} positivity-preserving doubly stochastic for all t\ge0, \operatorname{spec}(Q)=-\operatorname{spec}(H_{\mathrm{mod}}) (gap =\Delta=6\ln(3/2)), negative control: second eigenvalue -1/10 gives \det=-1/200<0 — provably NOT embeddable; the field level and the local size-consistent unitary dilation (Stinespring + Lieb–Robinson + OS continuation) stay [O]; binding precision: the shared F_transfer relaxation shape is a universal CONTRACTION CLASS, not one physical clock (v723/v724/v777). (4) QFT4D.OS.RECON.01 [O](the genuine 4D Euclidean lattice OS package) with the binding phrasing rule: the v381 all-orders leg is a local perturbative construction under the stated EG/BRST hypotheses (Duch weak adiabatic limit cited, hypotheses unverified for the concrete model; no confinement / mass gap / asymptotic charged states / nonperturbative Higgs vacuum / full S-matrix follows) — tfpt_2 and tfpt_5 phrasing restricted accordingly. (5) CHIRAL4D.NOMIRROR.01 [O](explicit GW/overlap chiral lattice operator; local gauge-invariant measure, perturbative + global anomaly freedom, no Witten SU(2) anomaly, no mirrors, correct index/charges, stable continuum limit; Mastropietro arXiv:2303.02790 as calibration). (6) GRAV.NONCIRCULAR.01 (standing rule, in force): BW is a theorem WITHIN a local Poincaré-covariant QFT (Mund/Morinelli counterexample landscape cited), never a generator of Lorentz spacetime from a modular flow; every BW/CHM use declares assumption-vs-result; audit census: the entanglement-equilibrium Einstein derivation (v323/v358/v359) is typed ASSUMPTION (conditional on the 4D package, never citable as evidence for it), QG.AMB.01 correctly declared, the seam KMS 2\pi chain dimension-typed, the intrinsic-BW programme the honest result-direction attempt. (7) OBS.TRANSDUCTION.01: the transduction principle (internal eigenvalue + nonvanishing projection B P_r\neq0 + physical coupling = observable signature; without the middle arrow neither non-observation falsifies nor observation confirms; minimal model machine-checked at exploration level in fo01, uniform functional blind to 10^{-14}, character channel recovers 64/729 to 10^{-10}; the ACT+Planck birefringence proximity typed coincidence-level pending the shared-action proof). (8) PRED.JOINTLIKELIHOOD.01 [O]: evidence reorganized from hit-counting into overdetermined relation bundles (flavor sum rule; r=3(1-n_s)^2, \alpha_s=-r/6; one normalized axion action; character-resolved transfer) with the statistical demand (joint likelihood, atom-induced covariance, effective dof, prespecified null + formula grammar, real holdout); the v100 “13/13, <10^{-30}” statement retyped as a look-elsewhere quantifier under its declared null — the 2026-06-09 freeze protects FUTURE measurements, it does not retroactively make historically known data blind. Docs corrected in the same change: docs/OPEN_PROBLEMS.md (the Einstein-derivation overstatement rewritten to the conditional Jacobson-route typing with the explicit does-NOT-follow list; “zero open internal mechanisms” fixed to match the contract scope fence — ledger wins; the all-orders phrasing restricted) and docs/FALSIFICATION.md (the null-model section retitled and rescoped, freeze and kill tests unchanged). Ledger +9 rows; suite 963 modules; bash build.sh audit and the manifest regeneration are the joint exit gate