The inflation gate — Gap-08: the standing bet
This is the page where the framework puts a number on the sky’s table. Its surviving inflaton candidate — the breathing mode of the internal geometry — stakes a pre-registered tensor band, r ∈ [3.5, 36]×10⁻³, every branch of it already sitting strictly under today’s measured ceiling (r < 0.036). Its scalar-tilt window, [0.9643, 0.9679], already brackets Planck’s measured 0.9649 ± 0.0042. Both routes agree; the measurement is pending. LiteBIRD, around 2030, gets the deciding vote — and it can rule the candidate out outright.
The gate, as the ledger grades it
On the gate board, Gap-08 asks: does the framework owe an early-inflation prediction? The answer is no, and the gate is RESOLVED +0 · DISSOLVED-GIVEN-root (full dossier): the frozen 13-dimensional shape does not force an inflationary episode, and asking a fixed, timeless geometry to dictate the universe’s specific early history is a category mismatch — two universes could share the exact same shape and have different histories. No extra early-expansion chapter is added unless a finite record forces it, and none does: no primordial B-mode signal has been detected that would demand an inflationary explanation. The inflaton sector is a declared-excluded input, not a derived output — which is exactly what makes the band below a bet the framework chose to place, not a debt it owed.
The measured CMB anchors — the tilt, the amplitude, the tensor bound — stay genuine, measured facts throughout, and ordinary post-Big-Bang expansion is untouched by any of this.
1. The bet: r ∈ [3.5, 36]×10⁻³, and a named satellite
The candidate’s tensor-to-scalar ratio band is pre-registered as the union of four normalization branches, r ∈ [3.5, 36]×10⁻³, with the operative branch at [3.5, 10]×10⁻³ — shown openly, not collapsed into a single number the geometry has not earned. Every branch of the band sits strictly under the current BICEP/Keck ceiling, r < 0.036 (95% CL). On the scoreboard this is Test 08, and its badge says exactly what it is: routes agree — measurement pending.
If LiteBIRD lands outside the band
The breathing-mode candidate is falsified — regardless of which normalization branch turns out right. A real, pre-registered, binary exposure.
If LiteBIRD lands inside the band
The candidate survives its sharpest available test, and the standing bet converts to a confirmed agreement on the scoreboard.
Until the satellite flies, the row stays indeterminate — that is the discipline, not a doubt. Standing behind the band is real; having the sky agree is a moment that hasn’t arrived.
2. Already on the board: the tilt brackets Planck
The candidate’s scalar spectral index lands in the window n_s ∈ [0.9643, 0.9679]; Planck measures 0.9649 ± 0.0042, squarely inside it. On the scoreboard this is Test 07, badged Agrees — with its caveats printed on its face: the window comes from running the exponential-plateau candidate through standard slow roll with the historical λ² = 1/6 slope, carried strictly as a phenomenological fit, not a forced prediction, and the row is marked consistency check — shared inputs. A match, not yet a lock: proving the roads had to meet there is the work still ahead.
3. Two candidate inflatons: one dead, one alive
The ρ-inflaton — dead
Too stiff to inflate: its internal restoring force is too strong for the slow roll inflation requires. Disqualified in kind, recorded, retired. It is not being revived.
The σ-inflaton — the surviving candidate
The breathing mode of the internal geometry produces an exponential-plateau potential — the same α-attractor-like family that independently fits the observed CMB well. This is the candidate behind the bet: the tilt window above, and the tensor band it stakes.
4. The slope, told straight
One number in this story was checked and retired, and this program said so in public. The plateau’s
normalization denominator K_σσ = 24 is genuinely forced by the geometry (the breathing-mode count
d(d+2)/2 with d = 6). The slope λ² = 1/6 built on it was a declared convention — and when the
frozen geometry’s own curvature term was computed directly, the genuine canonical slopes came out
8/3, 4, and 22/9 for the three ways of counting the breathing directions:
none of them 1/6. The in-scope slope-forcing question is CLOSED-NEGATIVE, the 1/6 convention is
retired, and no replacement slope is asserted in its place. That finding is why the bet is a band rather
than a point — and publishing it, instead of patching it, is the same discipline that makes the scoreboard’s
30 agreements — out of 44 tests, with 14 indeterminate and 0 disagreements — worth believing.
The band survives the finding by construction: it was pre-registered wide enough to absorb the full range of canonical slopes and normalization branches, which is exactly why it is branch-independent — the falsifiable test does not care which convention wins.
5. How this loses — and what stands either way
There is nothing hedged about the exposure. A LiteBIRD measurement outside [3.5, 36]×10⁻³ rules out the breathing-mode candidate — no normalization branch rescues it, and this program’s standing rules bar widening a band after the sky votes. What stands either way is the gate: the framework never owed an inflation prediction (the dissolution above), and the measured CMB record — tilt, amplitude, tensor bound — remains anchored regardless. The bet is a strength precisely because it is volunteered: a sharp, dated, falsifiable prediction — a strength, not a miss.
6. Bottom line
Gap-08 is RESOLVED +0 on the gate board: no inflation prediction is owed, and the demand for one dissolves against the framework’s own roots. What the framework placed on the table anyway is the best kind of claim a theory can make about an unmeasured number: a pre-registered band, every branch under today’s ceiling, a tilt window that already brackets the measured value, and a named, dated experiment with the deciding vote. Test 07 says Agrees; Test 08 says routes agree — measurement pending; LiteBIRD says the rest, around 2030.
See the standing bets on the scoreboard → Read the Gap-08 dossier → ← Early & Distant Universe overview