# n224 — The glass survives its own frozen gate: 7/7

**Date:** 2026-07-11 · **Gate:** `gates/p3pp-glass.json` (sha `0e52dd78…`), frozen before execution — deliberately built around the test class that executed the crystal (n220's global κ scan). **Script:** `n224_p3pp_gate.py`. **Reproduce:** one command.

## The verdicts

| leg | frozen requirement | result |
|---|---|---|
| G2 | c²₊/c²₊ ∈ [0.9, 1.1] per seed; mean ∈ [0.95, 1.05] at N = 512 | 1.024 / 0.986 / 0.984 — **mean 0.998** |
| G1+G3 (λ_g = 2) | stable family, λ_e ≥ 0, **no interior κ violation** | c² ≈ 2.60, λ_e ≈ 1.19, **min κ = +0.027** — all seeds |
| G1+G3 (λ_g = 4) | same | c² ≈ 1.66, λ_e ≈ 3.27, **min κ = +0.017** — all seeds |
| G4 | seed-spread shrinks with size | 0.0214 → 0.0184 |

The number that matters most is the one that killed the crystal: the global stability scan. On the pinned crystal, κ dipped to **−0.098** and 202 violations executed the theory (n220). On the pinned glass, the minimum over every sampled momentum, every channel, every seed, both family members is **+0.017 to +0.028 — positive with margin.** The amorphous substrate doesn't just pass; it passes the test its predecessor died on, with room.

## Where this leaves the program

- **P3″ — the amorphous two-ledger vacuum — is now a gate-passing candidate**: both forces propagate, the polarizations co-propagate statistically (0.998 ± 0.02), the equal-cone family is stable at scale, the photon kernel is topological, and the disorder that makes it all work is mandated by the theory's own Kibble–Zurek origin rather than assumed for convenience.
- **Pending, exactly as declared in the gate**: Gate D leg 2 — the Coulomb and Newton normalizations must pick the family member, converting the cone, the GW-birefringence amplitude, and the dispersion corrections into no-dial predictions. Plus the survival review's Gates A/E/F (local action, nonlinear mixing, the matter cone) inherited from P3′, and adversarial review of the entire n216–n224 chain.
- **Honest scope**: N = 512, jittered-diamond glass (true amorphous construction still owed), leading-order induction, q-scan at four harmonics. Every number is one command.

## The campaign, in one line each

The one-ledger vacuum died choosing light over gravity. The crystal two-ledger died of the Zener split. The glass two-ledger — imperfect by cosmological necessity — carries both forces, both polarizations, and its first frozen gate. **The theory that survives is the one that stopped pretending to be perfect.**

**Class: [A-model, frozen-gate execution at declared scope]; the candidate status [earned, not adopted — review and leg 2 stand between it and the Action].**