Documentation

GQ2.GaussZ.KappaR

The κ⁰ ledger for the Γ_R Gauss residue #

The Γ_R twin of GQ2/GaussZ/FinalGammaA/Kappa.lean (obligation ii.7, supply layer): the x₁-supported section of H¹_{R,word} and the split and ramified Roe wild-value calculations in the concrete κ⁰-extension CentExt (kappa0Cocycle dat hdat).

The gauge swap (docs/orchestration/roe-r31-survey-gaussz.md): the Γ_A normal form is x₀-supported (![0,0,c,0], slot 2; Z¹_w-shape x 1 = x 3 = 0), while the Γ_R normal form is x₁-supported (![0,0,0,d], slot 3; Z¹_{R}-shape x 1 = 0 ∧ x 2 = 0, GQ2/Roe/NormalForms.lean). Accordingly the structural pack here zeroes the x₀-slot (hx0v) where Γ_A zeroed the x₁-slot.

The word collapse. The Roe wild relator r_R = (x₀^σ)⁻¹ · a · x₁² · c (note eq. (1.2) ⟦eq:relators⟧) is much softer than Γ_A's r_A on the x₁-supported gauge: with x₀ ↦ 1 in the extension (zero V-part + trivial head), the first factor dies, the ω₂-word a = (x₀⁻³τ)^{ω₂} collapses to powOmega2 τ = 1 (τ of odd tame-inertia order), and what remains is x₁² · [x₁, x₁^{σ₂}] — the diagonal q(d) plus the V-slice commutator polar b_q(d, σ₂⁻¹d). There is no d₀/h₀ telescope: the ramified evaluation liftMark_kappa0_wildValueR_fib_ramified is the unconditional Wall shape q(d) + b_q(d, σ₂⁻¹·d) — exactly FoxH.QZeroR (GQ2/Roe/Gauss.lean, ⟦eq:QR⟧) — with no ramified hypothesis at all (the Γ_A twin needed htauf/hqg0); the split evaluation is the corollary killing the polar term by the split σ₂-triviality hU.

The generic toolkit (sdSec, liftMark_kappa0_tameValue_fib, sdToWL/sdBaseMarking/ sdOffsets, the m-calculus, kappa0_cc_one, commP_fib_cc_one, …) is imported from GQ2/GaussZ/FinalGammaA/Kappa.lean and reused verbatim, never cloned; only the tame first-component peel is restated at relZPairR (relZPairR_kappa0_fst_eq_zero, a definitional retype of liftMark_kappa0_tameValue_fib — the tame relator is shared).

All std-3; no axioms, no sorries.

The x₁-supported section of H¹_{R,word} (generic marking level) #

The note's "only x₁ varies" gauge (Lemma 4.2's normal forms, ⟦lem:normalforms⟧), as a bijective parametrization V ≃ H¹_{R,word}: membership and bijectivity fall out of the banked shape characterizations (lemma_5_13_split_R / lemma_5_13_ramified_R, GQ2/Roe/NormalForms.lean; normalForm_of_shapes_R / x1Supported_mem_Z1wR_ramified, GQ2/Roe/DualityAssembly.lean). Both regimes provide a unique x₁-supported normal form, so one ∃!-kernel serves both — unlike the Γ_A split case, whose surjectivity had to normalize the σ-row by hand.

theorem GQ2.SectionEight.AffineTLift.x1Supported_mem_Z1wR_split {C : Type u_1} [Group C] [Finite C] {V : Type u_2} [AddCommGroup V] [DistribMulAction C V] (t : Marking C) (ht : t.TameRel) (hw : t.WildRelR) (hV₂ : ∀ (v : V), v + v = 0) (hsimple : FoxH.IsSimpleModTwo C V) [Finite V] (hcore : t.Pro2Core) (htau : ∀ (v : V), t.τ v = v) (hVS : ∀ (v : V), t.σ v = vv = 0) (d : V) :

The x₁-supported tuples are Roe word cocycles (split regime): immediate from the lemma_5_13_split_R Z¹_R-shape (x 1 = 0 ∧ x 2 = 0). No σ₂-tameness hU is needed (one fewer hypothesis than Γ_A's x0Supported_mem_Z1w_split): the Roe wild row carries no σ₂-dependency.

theorem GQ2.SectionEight.AffineTLift.h1wMkR_eq_iff {C : Type u_1} [Group C] [Finite C] {V : Type u_2} [AddCommGroup V] [DistribMulAction C V] {t : Marking C} [Finite V] (x y : (FoxH.Z1wR t)) :
FoxH.h1wMkR t x = FoxH.h1wMkR t y (x - y) FoxH.B1wR t

The H¹_{R,word}-class equality criterion in h1wMkR vocabulary (H1wR is a semireducible def, so the quotient lemmas do not elaborate against it directly — the GaussZLocal.H1mk_eq_iff idiom, as for Γ_A's h1wMk_eq_iff).

theorem GQ2.SectionEight.AffineTLift.x1Section_bijective_split_R {C : Type u_1} [Group C] [Finite C] {V : Type u_2} [AddCommGroup V] [DistribMulAction C V] (t : Marking C) (ht : t.TameRel) (hw : t.WildRelR) (hV₂ : ∀ (v : V), v + v = 0) (hsimple : FoxH.IsSimpleModTwo C V) [Finite V] (hcore : t.Pro2Core) (htau : ∀ (v : V), t.τ v = v) (hVS : ∀ (v : V), t.σ v = vv = 0) :
Function.Bijective fun (v : V) => FoxH.h1wMkR t FoxH.x1Supported v,

The x₁-supported section of H¹_{R,word} is bijective (split regime): the unique normal form comes from the split shapes (normalForm_of_shapes_R at lemma_5_13_split_R). Γ_R twin of x0Section_bijective_split, minus the σ₂-tameness hU.

theorem GQ2.SectionEight.AffineTLift.x1Section_bijective_ramified_R {C : Type u_1} [Group C] [Finite C] {V : Type u_2} [AddCommGroup V] [DistribMulAction C V] (t : Marking C) (ht : t.TameRel) (hw : t.WildRelR) (hV₂ : ∀ (v : V), v + v = 0) [Finite V] (hx0 : ∀ (v : V), t.x₀ v = v) (hx1 : ∀ (v : V), t.x₁ v = v) (htau : ∀ (v : V), t.τ v = vv = 0) (hTodd : ∀ (v : V), powOmega2 t.τ v = v) :
Function.Bijective fun (v : V) => FoxH.h1wMkR t FoxH.x1Supported v,

The x₁-supported section of H¹_{R,word} is bijective (ramified regime): the unique normal form is lemma_5_13_ramified_R directly. Γ_R twin of x0Section_bijective_ramified.

The κ⁰-ledger: the tame peel and the Roe wild values #

The tame relator is shared with Γ_A, so the first-component peel is the imported liftMark_kappa0_tameValue_fib retyped at relZPairR. The Roe wild value collapses on the x₁-supported structural pack with no telescope: see the file docstring.

theorem GQ2.SectionEight.AffineTLift.relZPairR_kappa0_fst_eq_zero {C : Type u_1} {V : Type u_2} [Group C] [AddCommGroup V] [DistribMulAction C V] {q : VZMod 2} (dat : FactorSet C V) (hdat : IsEquivariantFactorSet q dat) [Finite C] [Finite V] (t : Marking (Sd C V)) ( : t.σ.v = 0) ( : t.τ.v = 0) :
(WordCoh2R.relZPairR t (kappa0Cocycle dat hdat)).1 = 0

The A-3 interface form at the Roe relator pair: the FIRST relator-z component vanishes on base-slice σ/τ-slots (the tame relator is shared with Γ_A, so this is the imported liftMark_kappa0_tameValue_fib on the nose).

theorem GQ2.SectionEight.AffineTLift.liftMark_kappa0_wildValueR_fib_ramified {C : Type u_1} {V : Type u_2} [Group C] [AddCommGroup V] [DistribMulAction C V] {q : VZMod 2} (dat : FactorSet C V) (hdat : IsEquivariantFactorSet q dat) [Finite C] [Finite V] (tS : Marking (Sd C V)) (hσv : tS.σ.v = 0) (hτv : tS.τ.v = 0) (hx0v : tS.x₀.v = 0) (hx0cc : tS.x₀.cc = 1) (hx1cc : tS.x₁.cc = 1) (hV₂ : ∀ (w : V), w + w = 0) (hτodd : Odd (orderOf tS.τ.cc)) :

The ramified Roe wild κ⁰-value is the Wall double — unconditionally (note ⟦prop:quadratic⟧/⟦eq:QR⟧): with the x₁-supported structural pack (σ/τ/x₀-slots of zero V-part, wild cc-slots 1, τ.cc of odd order), the lifted Roe wild relator value has fibre

q(x₁.v) + polar q x₁.v (σ₂⁻¹ • x₁.v) (σ₂ = Marking.sigma2 (sdBaseMarking tS)),

i.e. exactly FoxH.QZeroR q σ₂ (x₁.v). Unlike the Γ_A twin (liftMark_kappa0_wildValue_fib_ramified) no ramified hypothesis enters: the word r_R = (x₀^σ)⁻¹ · a · x₁² · c has no d₀/h₀ telescope — x₀ lifts to 1 (killing the first factor and collapsing a to powOmega2 τ = 1 by hτodd), x₁² deposits the diagonal q (one kappa0_cc_one step), and c = [x₁, x₁^{σ₂}] is a V-slice commutator whose fibre is the polar form (commP_fib_cc_one).

theorem GQ2.SectionEight.AffineTLift.liftMark_kappa0_wildValueR_fib_split {C : Type u_1} {V : Type u_2} [Group C] [AddCommGroup V] [DistribMulAction C V] {q : VZMod 2} (dat : FactorSet C V) (hdat : IsEquivariantFactorSet q dat) [Finite C] [Finite V] (tS : Marking (Sd C V)) (hσv : tS.σ.v = 0) (hτv : tS.τ.v = 0) (hx0v : tS.x₀.v = 0) (hx0cc : tS.x₀.cc = 1) (hx1cc : tS.x₁.cc = 1) (hV₂ : ∀ (w : V), w + w = 0) (hU : ∀ (w : V), (sdBaseMarking tS).sigma2 w = w) (hτodd : Odd (orderOf tS.τ.cc)) :

The split Roe wild κ⁰-value is the diagonal q (note ⟦prop:quadratic⟧/⟦eq:QR⟧, T = 1 case): the Wall double of the unconditional evaluation collapses because the split σ₂-triviality hU kills the polar term (b_q(d, d) = 0, the alternating law from f_polar in characteristic 2). Γ_R twin of liftMark_kappa0_wildValue_fib_split — with the whole htau/hU-telescope pack reduced to the single hU.

Paper-tag ledger (Roe note paper/roe-presentation-verification.tex; hand-maintained) #