Iteration 24 — the transfer lemma falls

Status: Two proof lenses on LEM-A1‴-T (T1 the direct transfer bound; T2 the closed-form Gram analysis refereeing the iteration-23 conditioning surprise), a binding assembling referee (independent cold re-derivation + machine verification), and light watch-sweep #4 (window since 07-11 + two-week rescan: 0 strong, 0 in-window items at all). Outcome: PROOF_ASSEMBLES — LEM-A1‴-T is [ESTABLISHED] (conditional only on its own explicit spectral data, which (a′) supports at [INFERENCE, high]). Verdict UNCHANGED — 23rd consecutive confirmation; hedge HELD at R7 by the house rule (R8 requires the FULL free-field closure; LEM-A1⁗ — the transversality crux — is now the single remaining blocker). Last updated: 2026-07-11 Iteration: 24

1. The theorem `[ESTABLISHED — referee re-derived cold; machine-verified to 1.4×10⁻¹⁵]`

LEM-A1‴-T (proved). For every fixed c > 1, the transverse-pencil eigenvector family {(a_n, b_n)}, n ∈ ℤ — a_n = cos(s_nη) + i·tanh(πτ₀)·sin(s_nη), b_n = cos(s_nη), s_n = n + iτ₀ — is a Riesz basis of L²((0,π); ℂ²), with explicit frame bounds 2π·min_η λ_min(K) and 2π·max_η λ_max(K), K = D M*M D.

The proof, in three elementary steps (no Birkhoff–Langer, no Mennicken–Möller, no Lunyov–Malamud — not even Levin–Golovin):

Weighted Fourier. Since Im s_n = τ₀ is constant, E_n = e^{is_nx} = e^{−τ₀x}e^{inx} on (0, 2π) is a bounded-invertible weight times the Fourier orthonormal basis — a Riesz basis with exact bounds [2πe^{−4πτ₀}, 2π].
Unitary folding. (Uf)(η) = (f(η), f(2π−η)) is unitary from L²(0,2π) to L²((0,π);ℂ²), and (UE_n)(η) = (e^{is_nη}, q·e^{−is_nη}) with q = e^{2πis_n} = e^{−2πτ₀} independent of n — the miracle: integer real parts + constant imaginary part. This n-independence is exactly what the 4×4 repeated-root reduction obscured.
Constant intertwiner. The constant matrix M = [v₊ | v₋/q] (v± the fixed coefficient vectors; det M = (t/2)e^{2πτ₀} ≠ 0) satisfies F_n = M(UE_n) exactly for all n (verified to 1.4×10⁻¹⁵ at c = 1.02, 1.5, 4, 25). A bounded-invertible operator carries a Riesz basis to a Riesz basis, completeness included. ∎

Falsification passed: the explicit frame bounds bracket the measured Gram spectra at all twelve (c, N) probe points, with measured values converging toward the bounds as N grows.

Honest conditionality (flagged by the referee, inside the lemma): the proof consumes the explicit spectral data — the exact lattice s_n = n + iτ₀ and n-independent t — as established by (a′); if the true pencil spectrum were only asymptotically lattice-like, a standard-but-unwritten Bari quadratic-closeness supplement would be required (risk graded nil per (a′)'s simple separated zeros). And no c-uniform bound exists at either end of (1, ∞) — the first thing iteration 25 must determine is whether LEM-A1⁗ needs c-uniformity.

2. The conditioning surprise dissolved — including the integrator's own misreading

T2 derived the Gram matrix in closed form — G = Toeplitz(n−m) + (πC/2)·J exactly (J the index flip; verified entrywise to ~10⁻¹⁵) — and the assembler confirmed both degeneracy laws:

c → 1⁺: λ_min ~ (c−1)/2 (pair collision a_n → b_n as t → 0; ratio measured/predicted → 1: 0.993 at c = 1.005). The iteration-23 reading that "c → 1.02 looks cleanest" was a misread: its N-stability is the rank-local pair mechanism saturating by N ≈ 3, not health. Recorded per house discipline — the referee corrected the integrator's own numerics interpretation.
c → ∞: λ_min ~ ln(4c)/(2c²) (the e^{4πτ₀} envelope lopsidedness; log-slow in N via the Toeplitz jump symbol, Böttcher–Silbermann).
Bounds are uniform precisely on compact [c₁, c₂] ⊂ (1, ∞); sweet spot near c ≈ 1.75.

3. Ledger effect

LEM-A1‴-T: [ESTABLISHED] — the first of the two named lemmas falls; the Riesz basis unlocks the Feshbach/Keldysh channel machinery, the local Pontryagin structure, and the H1 mode decomposition with explicit tools (the intertwiner T = M∘U; the exact frame bounds; the closed-form Gram).
LEM-A1⁗ is the sole remaining blocker of (E_O)-free-field and the R7 → R8 hedge move. Iteration-25 designated first move: push the transversality pairing through the folding coordinates (where the mode decomposition is diagonal), and determine first whether quadprime needs per-fixed-c bounds (available) or c-uniformity (not available). Secondary: the one-page Bari supplement discharging the exact-lattice conditionality.
Watch-sweep #4: 0 strong, 0 in-window items (math.OA and math.SP July listings fully scanned live).
Hedge: HYP-CKV-VACUITY HELD at R7 (the rule requires both lemmas). Verdict unchanged — 23rd consecutive confirmation; count FIVE; no numeric ID consumed.