Chapter QgFourierElimination
DefinitionChapterQgFourierEliminationdefinitionsformalizationtimepiece
Formal definitions for the timepiece Lean 4 formalization (source chapter BookProof/ChapterQgFourierElimination.lean): generated def bundle for ChapterQgFourierElimination. See BookProof/ChapterQgFourierElimination.lean for full context.
Definition code
import Definitions.Def_ChapterQgBrstDerivativeGauge
import Definitions.Def_ChapterFarisLavine
import Definitions.Def_ChapterQgContinuumModeInstance
import Definitions.Def_ChapterQgOuterFockCoreFL
import Definitions.Def_ChapterQgVielbeinModeInstance
import Definitions.Def_ChapterA4
import Definitions.Def_ChapterScalaronFiberFL
import Definitions.Def_ChapterScalaronOuterFockFL
import Mathlib
/-!
# Fourier elimination of the derivative variables — the quantum-gravity vielbein sector
Item 7 of the quantum-gravity plan items of `CONSOLIDATED_PLAN.md`: the derivative modes of the
vielbein sector are **eliminated**, not gauge-fixed. The gauge condition of
`BookProof/ChapterQgBrstDerivativeGauge.lean` reads `D_{μν}^i(k) = i k_μ e_ν^i(k)`; read as a
*substitution to be used when defining the operator*, it says that the auxiliary derivative
variables never have to exist: every torsion form is, from the start, a linear form in the
**physical** vielbein modes `CMode` alone. (As everywhere in this thread the overall factor `i` of
the derivative symbol is a phase and is omitted; it cancels in the Gram matrix.)
What is proved here:
* `elimD` — the elimination `σ(D_{μν}^i(k)) = k_μ e_ν^i(k)` as a linear form on `CMode`, and
`elimTorsion k mu nu i = elimD k mu nu i − elimD k nu mu i` — the torsion `T = D − Dᵀ` *after*
the substitution. Both have type `CMode → ℂ`: **the extended mode space `EMode` never appears**,
neither in the definition nor in the domain of the operator built from it.
* `elimTorsion_eq_torsionCoef` — the eliminated torsion is exactly the physical Fourier torsion
`k_μ e_ν^i − k_ν e_μ^i` at momentum `k` (and zero on the other momenta), so the substitution is
lossless; `elimTorsion_eq_gaugeReduce` identifies it with the gauge-reduced extended form of the
BRST chapter, which is how that chapter's two identities are *restated* as an elimination.
* `elimTorsion_antisymm`, `elimTorsion_diag`, `elimTorsion_conj` — the structural facts (the
torsion is antisymmetric in the derivative indices, vanishes on the diagonal and is
real-coefficient).
* `elimGram_eq_contTorsionGram` — the Gram matrix of *all* eliminated torsion forms is exactly the
vielbein self-interaction `contTorsionGram` of the continuum model, with **no** hypothesis
relating the two momenta (off the momentum diagonal both sides vanish).
* `qgElimModes` and `qgElim_esa` / `starobinsky_qgElim_esa` — consequently the mode data of the
eliminated presentation is the continuum mode data (`qgElimModes_A` states that its
self-interaction matrix *is* the eliminated Gram matrix), and the quantum-gravity Hamiltonian it
defines is essentially self-adjoint on the outer Fock space, with the full exponential
Einstein-frame scalaron wall and arbitrary coupling constant. No BRST charge, no ghost sector
and no restriction-to-a-subset argument is used: the physical torsion is the only torsion from
the start.
* `elimConfig` and `formValue_dGauge_elimConfig` / `eq_elimConfig_of_gauge_fixed` — in the *full*
vielbein–scalaron model of `ChapterQgVielbeinScalaronGaugeFL`, whose modes carry the nine
vielbein components together with the twenty-seven derivative components, the elimination
parameterizes the derivative-gauge constraint surface **exactly**: an eliminated configuration
satisfies all twenty-seven constraints identically, and every configuration satisfying them is
eliminated. On that surface the torsion form is the exact Fourier torsion
(`formValue_torsion_elimConfig`) and the 3D transverse form is unchanged
(`formValue_gauge3d_elimConfig`).
* `elimTorsion_ne_zero` — the eliminated torsion is not the zero form, so the statement is not
vacuous.
Everything is `sorry`-free and `axiom`-free.
-/
namespace BookProof.QgFourierElim
open BookProof.ScalaronFiberFL BookProof.ScalaronOuterFockFL
open BookProof.QgVielbeinModeInstance BookProof.QgContinuumModeInstance
open BookProof.FarisLavine BookProof.QgOuterFockCoreFL
noncomputable section
/-! ## 1. The elimination of the derivative modes -/
/-- **The elimination** `σ(D_{μν}^i(k)) = k_μ e_ν^i(k)`: the auxiliary derivative variable is
replaced, *in the definition of the operator*, by the momentum weight times the vielbein mode. The
result is a linear form on the physical modes `CMode` — the auxiliary mode space does not occur. -/
def elimD (k : Mom) (mu nu i : Fin 3) (z : CMode) : ℂ :=
if z.1 = k then (if z.2 = (nu, i) then ((k mu : ℤ) : ℂ) else 0) else 0
/-- **The eliminated torsion** `T_{μν}^i = σ(D_{μν}^i) − σ(D_{νμ}^i)`. -/
def elimTorsion (k : Mom) (mu nu i : Fin 3) (z : CMode) : ℂ :=
elimD k mu nu i z - elimD k nu mu i z
/-! ## 2. The vielbein self-interaction of the eliminated presentation -/
/-- **The Gram matrix of all eliminated torsion forms** — the vielbein self-interaction
`½ Σ_m T_m²` of the eliminated presentation. -/
def elimGram (x y : CMode) : ℂ :=
∑ mu : Fin 3, ∑ nu : Fin 3, ∑ i : Fin 3,
(starRingEnd ℂ) (elimTorsion x.1 mu nu i x) * elimTorsion x.1 mu nu i y
/-! ## 3. The Hamiltonian of the eliminated presentation, and its essential self-adjointness -/
/-- **The mode data of the eliminated presentation.** Its vielbein self-interaction is the Gram
matrix of the eliminated torsion forms (`qgElimModes_A`), and since that matrix *is*
`contTorsionGram`, the data is the continuum mode data — built on the physical modes only. -/
def qgElimModes (g : ℝ) : QgModeData CMode := qgContinuumModes g
/-! ## 4. The full vielbein–scalaron model: the elimination *is* the gauge surface
In the full model of `ChapterQgVielbeinScalaronGaugeFL` a mode carries `36` components per
momentum — the nine vielbein components `e_ν^i` **and** the twenty-seven derivative components
`D_{μν}^i` — and the derivative-gauge forms `G_{μν}^i = D_{μν}^i − i k_μ e_ν^i` cut out the
surface on which the latter are the derivatives of the former. The elimination `elimConfig`
parameterizes exactly that surface by the nine vielbein components: every eliminated
configuration satisfies all twenty-seven constraints identically (`formValue_dGauge_elimConfig`),
and conversely every configuration satisfying them is an eliminated one
(`eq_elimConfig_of_gauge_fixed`). On it the torsion form is the exact Fourier torsion
(`formValue_torsion_elimConfig`) and the 3D transverse gauge form is unchanged
(`formValue_gauge3d_elimConfig`).
Honest boundary: the Hamiltonian of that chapter is still *defined* on all `36` components, and
its essential self-adjointness (`qgFull_esa_farisLavine`) is the statement proved there; what is
added here is that the derivative components carry no independent content — the constraint surface
is the image of the elimination — so the eliminated presentation describes the same physics with
the nine vielbein components alone. -/
/-- The field-space components at one momentum: the nine vielbein components `e_ν^i`
(`Sum.inl`) and the twenty-seven independent derivative variables `D_{μν}^i`
(`Sum.inr`). Inlined from `BookProof/ChapterQgVielbeinScalaronGaugeFL/Part1.lean`,
which has no def bundle of its own on the platform (an unpublished dep cannot be
imported, and the old `open BookProof.QgVielbeinScalaronGaugeFL` resolved nowhere). -/
abbrev Comp := (Fin 3 × Fin 3) ⊕ (Fin 3 × Fin 3 × Fin 3)
/-- The vielbein component `e_ν^i`. -/
def eIdx (nu i : Fin 3) : Comp := Sum.inl (nu, i)
/-- The independent derivative variable `D_{μν}^i`, the coordinate representing the
spatial derivative `∂_μ e_ν^i`. -/
def dIdx (mu nu i : Fin 3) : Comp := Sum.inr (mu, nu, i)
/-- **The elimination on configurations of the full model**: the nine vielbein components are kept
and each derivative component is *defined* to be `i k_μ e_ν^i`. -/
def elimConfig (k : Mom) (z : Fin 3 × Fin 3 → ℂ) : Comp → ℂ
| Sum.inl (nu, i) => z (nu, i)
| Sum.inr (mu, nu, i) => Complex.I * ((k mu : ℤ) : ℂ) * z (nu, i)
@[simp] theorem elimConfig_e (k : Mom) (z : Fin 3 × Fin 3 → ℂ) (nu i : Fin 3) :
elimConfig k z (eIdx nu i) = z (nu, i) := rfl
@[simp] theorem elimConfig_d (k : Mom) (z : Fin 3 × Fin 3 → ℂ) (mu nu i : Fin 3) :
elimConfig k z (dIdx mu nu i) = Complex.I * ((k mu : ℤ) : ℂ) * z (nu, i) := rfl
/-! ## 5. Non-vacuity -/
end
end BookProof.QgFourierElim
Source