# n111 — the refutation hunt, second sweep # (2026-07-10; sequel to n110 — new angles only, existing data only) Verdicts: KILL / WOUND / SQUEEZE / ABSORBED / MISS (n110 scale). --------------------------------------------------------------------- ATTACK 6 — the Nordtvedt null: does binding energy gravitate here? --------------------------------------------------------------------- EXPOSURE: the sharpest structural attack found in either sweep. In the framework, mass = maintenance budget, c_i >= 0 per object. But nature's masses contain NEGATIVE pieces: gravitational self-energy, nuclear binding, molecular binding. GR says every piece gravitates identically (strong EP); lunar laser ranging confirms it — the Nordtvedt parameter is zero at ~1e-4, meaning the Earth's own gravitational binding energy (−4.6e-10 of its mass) both falls and SOURCES gravity exactly like rest mass. THE QUESTION THE TOY MUST ANSWER: take two budgets bound by their mutual attraction. Does the pair's far-field monopole read c1 + c2, or c1 + c2 − E_bind? ORDER ANALYSIS (done here, before pointing the gun the wrong way): at LINEAR order the answer is forced by superposition — the far field is the bare sum, with no binding correction. That is NOT a failure: Newton does exactly the same, and nature agrees at that order (E_bind/M ~ 1e-10 for Earth). The −E_bind sourcing is a POST-NEWTONIAN, nonlinear effect: in this framework it must come from the anharmonic layer, at order xi — the same layer that produced Mercury's perihelion and the Delta-beta = +0.5000 debt. THE RUNNABLE TEST: with the n87b machinery (exact first-order anharmonic response, eps2 = −xi L⁺ proj(eps1²)), compute the far-field monopole shift of a bound pair at order xi and compare its coefficient against the pair's interaction energy in the same convention. PASS = the shift tracks −E_bind with one universal coefficient (eta_N = 0 structurally); FAIL = a coefficient that depends on the pair's composition or separation — a composition- dependent eta_N, which LLR kills at 1e-4. VERDICT: WOUND + NAMED COMPUTATION (the Nordtvedt run), now correctly aimed: it is an order-xi lattice computation, runnable with existing code, pre-registered pass/fail above, and a FAIL is a genuine kill — not a missing sector but a wrong number in the one nonlinear layer we claim to own. The cheapest possible way the theory could die this month. LLR watcher: eta_N = (−0.2 ± 1.1)e-4. --------------------------------------------------------------------- ATTACK 7 — the JADES handedness dipole (prediction 15 under live fire) --------------------------------------------------------------------- EXPOSURE: prediction 15 — the audit is achiral; every cosmic spin-parity dipole dissolves under bias-free reanalysis. DATA: Shamir (MNRAS 2025), JADES deep field: 158 clockwise vs 105 counterclockwise (263 galaxies, ~3.4 sigma), dipole peaking near the GALACTIC pole, amplitude growing with redshift. Counter-analyses (Stiskalek & Desmond 2024; Galaxy Zoo reanalyses) find no asymmetry with different methodology; the fight is methodological and open. VERDICT: SQUEEZE — the nearest thing to a live bullet aimed at a registered kill condition. The framework's stance, sharpened with the data's own geometry: the dipole peaks near the GALACTIC pole — exactly where a Milky-Way-frame bias would put it (rotational Doppler boost: galaxies rotating against our rotation are slightly brightened, a known, calculable, redshift-GROWING selection effect — which also reproduces the amplitude trend Shamir reports). A cosmic-chiral axis has no reason to align with our galaxy's pole; a local bias has every reason. BET REGISTERED: the dipole will track the observer (Galactic frame), not the cosmos (CMB frame) — the single measurement that separates the two, and prediction 15's kill condition is now operational: a handedness dipole CONFIRMED IN THE CMB FRAME, bias-modeled, kills the achiral axiom. --------------------------------------------------------------------- ATTACK 8 — XRISM sweeps the knee's band (and the 3.5 keV temptation) --------------------------------------------------------------------- EXPOSURE: the electron-anchor knee sits at 3.7 keV. The famous unidentified 3.5 keV cluster line sat suspiciously nearby — an opportunist would claim it. DATA: XRISM (Oct 2025), 3.75 Ms stacked over ten clusters: NO unidentified lines anywhere in 2.5–15 keV; the 3.5 keV line's dark-matter interpretation excluded 3–4x below Hitomi. VERDICT: MISS — and a discipline entry. The framework never claimed the line (the knee is a RESPONSE-function break, not an emission line; the audit emits nothing at gamma), so XRISM's cleaning costs nothing — and it retroactively validates the F10 rule: had we grabbed the 3.5 ≈ 3.7 coincidence, we would now be dead. The null also pre-bounds any future temptation to give the pump a keV-band exhaust line in clusters: there is none, to XRISM depth. --------------------------------------------------------------------- ATTACK 9 — direct T-violation in kaons vs "T broken in dissipation only" --------------------------------------------------------------------- EXPOSURE: the C/P/T card says T is broken only by the measurement arrow (dissipation). CPLEAR measured DIRECT T-violation in K0 oscillations — in the UNITARY, Hamiltonian sector, compensated by CP (CPT holds). VERDICT: WOUND (scope correction, filed). The claim as worded overreaches: it is a statement about the SUBSTRATE, and nature's Hamiltonian T-violation must then live entirely in the absent matter sector's complex phases. Correct scoped claim: "the substrate's unitary part is T-even; all Hamiltonian T-violation is matter-sector physics the framework does not yet contain; the substrate's own arrow is dissipative." Filed with the alpha beta function (n110 attack 3) in the matter sector's debt column — which now owes: masses, charge quantization, 3.604, the beta function, CP phases. --------------------------------------------------------------------- ATTACK 10 — the Event Horizon Telescope: strong field, no sector --------------------------------------------------------------------- EXPOSURE: M87* and Sgr A* photon rings match Kerr at ~10%. The framework's gravity is derived in the weak, quasi-static limit; it has no strong-field regime at all (what happens when budgets overlap at horizon densities is not even formulated). VERDICT: ABSORBED into the tensor junction's watcher list, which now reads, in order of precision: binary-pulsar decay (1.3e-4), LARES-2 frame dragging (1e-3), EHT photon rings (1e-1). Not a contradiction — nothing is predicted — but the continuum deliverable's acceptance tests are now three, spanning weak-field rotation to horizon scale. --------------------------------------------------------------------- ATTACK 11 — LISA Pathfinder: femto-g free fall --------------------------------------------------------------------- EXPOSURE: two free-falling test masses tracked at sub-femto-g/√Hz. Any audit force noise on RESTING masses would appear directly. DATA: LPF's residual acceleration noise closed with known physics (Brownian + actuation + charging) down to 1.7 fm/s²/√Hz. VERDICT: MISS (confirmation-shaped): resting masses are dark states — the framework predicts EXACTLY zero audit noise there, and the world's most sensitive free-fall experiment found none it couldn't explain. Enters the card beside the Penning-trap null (prediction 19's family): the rest sector keeps passing. --------------------------------------------------------------------- BALANCE OF THE SECOND SWEEP --------------------------------------------------------------------- Again no kill in existing data — but the hunt found its first POINTABLE GUN: the Nordtvedt question (attack 6) is a computation the existing lattice can run, with a pre-registered pass/fail and a genuine kill on failure. That outranks every future-experiment proposal on the site: it is the cheapest possible way the theory could die this month. Second finding: prediction 15 now has an operational discriminator (Galactic frame vs CMB frame). The matter sector's debt column lengthened by one (CP phases). The rest sector passed its most sensitive test yet (LPF).