# Gap-08 — The Inflation Spectrum

## What this gate must establish

This gate asks whether the frozen 13D geometry, run forward into the early universe, produces a primordial-fluctuation spectrum compatible with what the cosmic microwave background measures — concretely, a scalar spectral tilt n_s and a tensor-to-scalar ratio r in the observed range. It is the gate where the geometry first touches cosmology. **Critically, cosmology is a declared out-of-scope sector for the programme** (named under the claim boundary, SG-10 §9.3.2), so this gate is conditional and diagnostic by construction: it may *exhibit* a consistency, but it can never close, and it can never be used to support any in-scope claim.

## The geometry's role (load-bearing, not validation)

If an inflationary mode exists at all here, it is not added by hand: the candidate is the geometry's own "breathing" modulus rolling down a potential, and the one number that would be a genuine output — the plateau slope λ² = 4/K_σσ = 1/6 — is read off the way the thirteen dimensions split as n = (6, 2, 1), with no dial available to place it. That is the load-bearing point: the tilt is geometry-fixed, not fitted. But the forcing layer **fails closed**: of the two candidate inflatons the geometry offers, one (ρ) is a computed DEAD verdict (KK-stiff, qualitative grade), and the survivor (σ) reaches a number only after a normalization convention and two pending countersigns are supplied — so the geometry narrows the field but does not deliver a forced spectrum.

## Current honest status

**Badge: OPEN (cascade). Conditional / diagnostic; excluded sector. No status was ever upgraded.**

- **ρ-inflaton: DEAD** at qualitative grade (computed verdict; countersign slot pending).
- **σ-inflaton: SURVIVES as a CANDIDATE only.** The corrected (v10) regime is an exponential plateau giving **n_s ∈ [0.9643, 0.9679]** (Planck-centered) and **r ∈ [3.5, 10]×10⁻³** on the frozen operative branch. These are explicitly **PROJECTED outputs compared against MEASURED comparators** — under the corpus-wide no tuning to the known answer rule, Planck n_s, the r bound, and N_eff are comparison targets, never inputs.
- **Blockers / open legs (all genuinely open):**
  1. **K13-2, the σ-kinetic normalization convention.** The D.1 kinetic denominator (D−2)=11 exceeds the textbook reduction (D₄−2)=2 by exactly 81/22. The breathing axis is source-resolved (uniform across n=(6,2,1)); the normalization axis is an **owner ruling** retaining D.1 as intentional — recorded as authorial, *not* a theorem forced by the source. Four convention branches (T6/T_u/D6/D_u) span r from ~3.5 to ~36×10⁻³.
  2. **Pending countersigns** for σ-advancement, the plateau outputs, and the second upgrade — none cleared.
  3. **c_loop dependency** — inherited from the still-open moduli/loop-coefficient cascade elsewhere in the programme.
  4. **A_s (absolute amplitude): BLOCKED / conditional** — no unconditional amplitude.

Because any one open leg keeps the gate open, the gate is OPEN.

## What would close it

There is no in-scope closure: cosmology is excluded, so even a perfect n_s/r match would remain a diagnostic consistency, not a derivation. Within the diagnostic lane, advancing the candidate requires (a) resolving K13-2 from the source rather than by owner ruling, (b) clearing the pending σ countersigns, and (c) discharging the upstream c_loop dependency — and then computing the amplitude A_s unconditionally. **The decisive external adjudicator is observation:** the union falsifier is a LiteBIRD tensor-to-scalar measurement falling outside r ∈ [3.5, 36]×10⁻³ (all branches), which would kill the surviving candidate. Until then this is a candidate consistency, pre-registered to die, not a result.

## Sources

- Gaps & Walls Register (programme-wide honest status): `…/Fable_Version/rendered/TOE/GAPS_AND_WALLS_REGISTER.md` — claim-boundary and excluded-sector framing.
- Scoped TOE main statement, §6.2 "Inflation branches and K13-2 (Gap 08)" and §7 truth-table: `…/Fable_Version/rendered/TOE/Scoped_TOE_ST1_main_statement.md` — the ρ-DEAD / σ-CANDIDATE split, the four-branch table, the no tuning to the known answer rule, and the LiteBIRD falsifier.
- Claim-boundary dossier (SG-10), §9.3.2: `…/Fable_Version/rendered/TOE/PER_GATE_DOSSIERS/SG10_COMPLETION_HANDOFF/01_DOSSIER.md` — full cosmology (inflation, reheating, CMB, structure) named as out-of-scope.
- Paper IV formal register (K13-2 source anchor, L10268): published manuscript at https://physics.magflowmeters.com/articles/TOE.html.

**Honest ceiling: serious candidate / partial unification — NOT validated.** A given spectrum is not a derived spectrum; the geometry fixes a tilt and fails the rest closed; the sector is excluded and the gate is OPEN.