Kill Switches

Seven switches, Claim / Test / Status / Recovery, identifiers pending MKSR registration.

KS-CCC.1 — Forced structure

Claim. Every structural choice entering the commitment — count, placement, boundary, magnitude — is either forced from the axiom and locked results, or is a disclosed selection with its eliminations on the public record (PROV-CCC.1–6). This switch binds the corpus’s claim, not its author. The author is bound by nothing here — not this paper, not this corpus, not this switch: exploration, revision, and new derivation remain his without condition, the proton, neutron, and deuteron included. What the fence holds is the flag: the zero-fitted-parameters status of the G result.

Test. The standing in canon of any refinement of the commitment whose selection depended on comparison with measured G and whose selection history is undisclosed.

Status. LIVE — METHODOLOGICAL. It binds the claim.

Recovery. The flag retires for the refined result — it may stand as work, but not as parameter-free. The original 1×ε result, with its disclosed history, is unaffected.

KS-CCC.2 — Mass-side drift

Claim. No mass is dressed by the commitment: matter configurations’ commitments are already booked as mass (AP29; §4, Corollary), and m_p/m_e carries no commitment signature above experimental uncertainty (±0.0174 ppb, CODATA 2022).

Test. Any derivation, anywhere in the corpus, that applies a commitment factor to a mass or exhibits a commitment residue in a mass ratio.

Status. LIVE — DISCIPLINE.

Recovery. The arena reading fails, and with it the paper: a mass-side or channel-scaled replacement is excluded by AP30 at eight orders of magnitude, so the proposal reverts to unpriced — AP28 §2’s sentence stands, the commitment withdrawn.

KS-CCC.3 — Existence, direction, and the migration fork

Claim. A commitment of order ε exists; the structural value exceeds the realised value; and — the fork — as apparatus systematics are resolved, the adjusted (consensus) value of G migrates downward, by roughly 300 ppm from its current centre, toward the realised 6.672 × 10⁻¹¹. The null (no commitment; AP28 coincidental) predicts it remains near 6.6743. This is the one forward prediction that separates the commitment from the null, and it is slow, dated, and out of the author’s hands.

Test. Successive CODATA adjustments and replication-grade determinations, beginning with the first datum after this switch was written: NIST-26, which moved downward. Convergence upward toward 6.7206 × 10⁻¹¹ kills the commitment and revives Candidate 1 in full. Migration downward toward 6.672 × 10⁻¹¹ supports the commitment against the null; stability at 6.6743 as systematics close supports the null. Individual determinations may sit on either side of the realised value within their uncertainties (LENS-14’s centre sits marginally below it); the fork is about the adjusted value, not any single experiment.

Status. LIVE — EMPIRICAL. The switch tests existence and direction, not magnitude: by §10, the admissible window is seven and a half per cent of ε wide, and convergence cannot distinguish 1×ε from nearby fractions. No claim of magnitude confirmation will be made from it.

Recovery. If metrology converges toward the structural value, the commitment is withdrawn and the 0.69% returns, whole, to Candidates 1 and 2.

KS-CCC.4a — Gap-scale exclusion

Claim. Apparatus corrections are not of gap scale: no physical characterisation of the published apparatuses supports a common leakage term of order 6,900 ppm.

Test. The Appendix C meta-analysis — executed twice at Tier 1, independently, on the same day (WP-CCC.1; the reviewer-thread execution), outcome N in both — and Tier 2’s blind coding of absolute leakage capacity.

Status. LIVE — EMPIRICAL. Tier 1 did not fire it; the pass is exclusion of Candidate 1’s scale, not confirmation of the commitment.

Recovery. Gap-scale corrections confirm Candidate 1 and withdraw the commitment; the realised formula reverts to AP28’s.

KS-CCC.4b — Scatter structure

Claim. Residual boundary-dependence exists in the inter-laboratory scatter at AP06’s honest size — order 10² ppm.

Test. Tier 2’s blind boundary coding of the scatter, under a stated limit: implied leakage is affine in G, so class-correlation statistics are identical under every constant-true-value hypothesis — 4b tests AP06’s residual-leakage reading of the scatter, and cannot by itself discriminate the candidates’ central values.

Status. LIVE — EMPIRICAL. Tier 1 found the structure at best marginal (p ≈ 0.06– 0.2).

Recovery. If no boundary structure exists at any scale, AP06’s reading of the scatter weakens; the commitment’s central claim is unaffected either way.

KS-CCC.5 — Held, not shed

Claim. A constitution instant’s ε is held as the persistence commitment (Proposition 2), and in particular the arena’s constitution ε is held.

Test. Exhibit a mechanism by which a constitution ε — the arena’s above all — is shed while the constitution record is held open by other means.

Status. LIVE — STRUCTURAL.

Recovery. Proposition 2 fails; the dichotomy collapses to leakage-only; the commitment is withdrawn. AP06 is unaffected — the leakage never depended on this paper.

KS-CCC.6 — Single entry

Claim. The puncture factor and the commitment are independent readings of the unpaired element (Proposition 6, claimed as an extension of AP24’s licence).

Test. The completed puncture-topology normalisation owed under KS-R.9, if it is shown to already contain the persistence provision.

Status. LIVE — STRUCTURAL, coupled to KS-R.9.

Recovery. The correction double-counts and is removed; the puncture factor absorbs the commitment under a reinterpreted normalisation; the realised formula reverts to AP28’s.