Peterfalvi §9: inertia lift of linear characters to
DefinitionOddOrder_Peterfalvi_S11_MaximalII_III_IV_InertiaLiftfeit-thompsonfinite-groupsgroup-theoryishida-odd-order
A continuation file for the maximal subgroups of types II, III and IV in the Clifford case (a) analysis. It realizes the single-factor centralizer inside and inside , and lifts linear characters to .
Main declarations: cuSub (The single-factor centralizer C_U(S₀)); cuInHu (C_U(S₀), realized as a subgroup of HU inside ↥M (mirrors cInHu for C)); cuSub_subgroupOf_U_normal (C_U(S₀) ◁ U); cuInHu_normal (C_U(S₀) ◁ U).
The bundle also proves 35 supporting lemmas used by later statements.
Definition code
/-
Copyright (c) 2026 Yawara Ishida. All rights reserved.
Released under Apache 2.0 license as described in the file LICENSE.
Authors: Yawara Ishida
-/
-- From github.com/yawara/odd-order @ 20a6fccbb63c (Apache-2.0; LICENSE at that commit):
-- OddOrder/Peterfalvi/S11_MaximalII_III_IV/InertiaLift.lean
-- Generated for Prove2Me by deleting declarations; the remaining commands are verbatim except:
-- `private` dropped where another file uses the declaration, and a solution's main theorem
-- renamed `_root_.solution` (platform convention).
import Definitions.Def_OddOrder_Peterfalvi_S11_MaximalII_III_IV_CliffordData
namespace OddOrder.Peterfalvi.S11
open OddOrder.GroupTheory
open OddOrder.RepresentationTheory
open OddOrder.Isaacs
open scoped Pointwise
variable {G : Type*} [Group G]
open OddOrder.Isaacs.Ch03 (IsAInvariant isAInvariant_iff_smul_mem)
open OddOrder.Isaacs.Ch03.IsAInvariant (quotientMulAutHom)
variable {M : Subgroup G}
section CuS0
variable {M : Subgroup G} {data : TypesIIIIIIVSetup M} {chief : ChiefFactorData data}
{chars : Section11CharacterData data chief}
/-- **The single-factor centralizer `C_U(S₀)`**, realized as a subgroup of `G` with `C_U(S₀) ≤ U`.
The kernel of the restricted `U`-action `aInvariantRestrictAut caseA.S0_aInvariant` on the order-`p`
factor `S₀`, pushed into `G` along `↥(U.subgroupOf (U ⊔ W₁)) ↪ ↥(U ⊔ W₁) ↪ G` (exactly as `cSub`
pushes `(uActionHom).ker`). By the first isomorphism theorem `|U : C_U(S₀)| = |Ū₁| = caseA.a`
(`index_cuInHu_subgroupOf_uInHu_eq_a`). -/
noncomputable def cuSub (caseA : CliffordCaseAData chars) : Subgroup G :=
((aInvariantRestrictAut caseA.S0_aInvariant).ker.map
(data.typeP.U.subgroupOf (data.typeP.U ⊔ data.typeP.W1)).subtype).map
(data.typeP.U ⊔ data.typeP.W1).subtype
theorem cuSub_le_U (caseA : CliffordCaseAData chars) : cuSub caseA ≤ data.U :=
(Subgroup.map_mono (Subgroup.map_subtype_le _)).trans <| by
rw [Subgroup.subgroupOf_map_subtype]; exact inf_le_left
/-- `|C_U(S₀)| = |ker(aInvariantRestrictAut S₀)|`: the realization is the injective double-image of
the kernel (mirrors `card_cSub_eq_card_ker`). -/
theorem card_cuSub_eq_card_ker (caseA : CliffordCaseAData chars) :
Nat.card ↥(cuSub caseA) = Nat.card ↥(aInvariantRestrictAut caseA.S0_aInvariant).ker := by
change Nat.card ↥(((aInvariantRestrictAut caseA.S0_aInvariant).ker.map
(data.typeP.U.subgroupOf (data.typeP.U ⊔ data.typeP.W1)).subtype).map
(data.typeP.U ⊔ data.typeP.W1).subtype)
= Nat.card ↥(aInvariantRestrictAut caseA.S0_aInvariant).ker
rw [← Nat.card_congr (Subgroup.equivMapOfInjective _ _ (Subgroup.subtype_injective _)).toEquiv,
← Nat.card_congr (Subgroup.equivMapOfInjective _ _ (Subgroup.subtype_injective _)).toEquiv]
/-- `C_U(S₀)`, realized as a subgroup of `HU` inside `↥M` (mirrors `cInHu` for `C`). -/
noncomputable def cuInHu (caseA : CliffordCaseAData chars) : Subgroup ↥(huSub data) :=
((cuSub caseA).subgroupOf M).subgroupOf (huSub data)
theorem cuInHu_le_uInHu (caseA : CliffordCaseAData chars) : cuInHu caseA ≤ uInHu data :=
Subgroup.subgroupOf_mono _ (Subgroup.subgroupOf_mono _ (cuSub_le_U caseA))
/-- `|C_U(S₀)|` realized inside `HU` equals `|C_U(S₀)|` in `G` (mirrors `card_cInHu_eq`). -/
theorem card_cuInHu_eq (caseA : CliffordCaseAData chars) :
Nat.card ↥(cuInHu caseA) = Nat.card ↥(cuSub caseA) := by
have hCsubM : (cuSub caseA).subgroupOf M ≤ huSub data :=
Subgroup.subgroupOf_mono M
((cuSub_le_U caseA).trans (le_sup_right : data.U ≤ data.H ⊔ data.U))
have hCleM : cuSub caseA ≤ M := (cuSub_le_U caseA).trans (U_le_M data)
calc Nat.card ↥(cuInHu caseA)
= Nat.card ↥((cuSub caseA).subgroupOf M) :=
Nat.card_congr (Subgroup.subgroupOfEquivOfLe hCsubM).toEquiv
_ = Nat.card ↥(cuSub caseA) :=
Nat.card_congr (Subgroup.subgroupOfEquivOfLe hCleM).toEquiv
open Subgroup in
/-- **`C_U(S₀) ◁ U`** (mirrors `cSub_subgroupOf_U_normal`): the realization `cuSub` is the `G`-image
of a kernel, so its `subgroupOf U` is normal (kernels are normal, transported by the realization iso
`↥(U.subgroupOf (U ⊔ W₁)) ≃* ↥U`). -/
theorem cuSub_subgroupOf_U_normal (caseA : CliffordCaseAData chars) :
((cuSub caseA).subgroupOf data.U).Normal := by
set e := subgroupOfEquivOfLe (le_sup_left : data.typeP.U ≤ data.typeP.U ⊔ data.typeP.W1) with he
have heq : (cuSub caseA).subgroupOf data.U
= (aInvariantRestrictAut caseA.S0_aInvariant).ker.map e.toMonoidHom := by
ext x
simp only [Subgroup.mem_subgroupOf]
constructor
· intro hx
simp only [cuSub, Subgroup.mem_map] at hx
obtain ⟨z, ⟨y, hy, hyz⟩, hzx⟩ := hx
refine ⟨y, hy, ?_⟩
apply Subtype.ext
rw [MulEquiv.coe_toMonoidHom, he, subgroupOfEquivOfLe_apply_coe, ← hzx, ← hyz]
rfl
· rintro ⟨y, hy, rfl⟩
simp only [cuSub, Subgroup.mem_map]
refine ⟨_, ⟨y, hy, rfl⟩, ?_⟩
rw [MulEquiv.coe_toMonoidHom, he, subgroupOfEquivOfLe_apply_coe]
rfl
rw [heq]
exact (MonoidHom.normal_ker _).map e.toMonoidHom e.surjective
open Subgroup in
set_option backward.isDefEq.respectTransparency false in
/-- **`C_U(S₀) ◁ U`** realized inside `HU` (mirrors `cInHu_normal`): `cuInHu ◁ uInHu`, transported
from `cuSub ◁ U` along `↥uInHu ≃* ↥U`. -/
theorem cuInHu_normal (caseA : CliffordCaseAData chars) :
((cuInHu caseA).subgroupOf (uInHu data)).Normal := by
have hUsubM : data.U.subgroupOf M ≤ huSub data :=
Subgroup.subgroupOf_mono M (le_sup_right : data.U ≤ data.H ⊔ data.U)
set f : ↥(uInHu data) ≃* ↥data.U :=
(subgroupOfEquivOfLe hUsubM).trans (subgroupOfEquivOfLe (U_le_M data)) with hf
have hgval : ∀ x : ↥(uInHu data), ((f x : ↥data.U) : G) = (((x : ↥(huSub data)) : ↥M) : G) := by
intro x
have h1 : (f x : ↥data.U)
= subgroupOfEquivOfLe (U_le_M data) (subgroupOfEquivOfLe hUsubM x) := by rw [hf]; rfl
rw [h1, subgroupOfEquivOfLe_apply_coe, subgroupOfEquivOfLe_apply_coe]
have hcomap : (cuInHu caseA).subgroupOf (uInHu data)
= ((cuSub caseA).subgroupOf data.U).comap f.toMonoidHom := by
ext x
simp only [cuInHu, Subgroup.mem_subgroupOf]
rw [Subgroup.mem_comap, Subgroup.mem_subgroupOf, MulEquiv.coe_toMonoidHom, hgval x]
rw [hcomap]
exact (cuSub_subgroupOf_U_normal caseA).comap f.toMonoidHom
/-- **`H·C_U(S₀) ◁ HU`** (mirrors `hcInHu_normal`): the inertia subgroup of the (9.8.d) source
character is normal, so `Ind_{H·C_U(S₀)}^{HU}` produces an irreducible. From `H ◁ HU`
(`hInHu_normal`), `C_U(S₀) ◁ U` (`cuInHu_normal`), `H ⊔ U = ⊤` (`hInHu_sup_uInHu_eq_top`). -/
theorem hcuInHu_normal (caseA : CliffordCaseAData chars) :
(hInHu data ⊔ cuInHu caseA).Normal :=
haveI := hInHu_normal data
haveI := cuInHu_normal caseA
sup_normal_of_normal_left_of_normal_subgroupOf (cuInHu_le_uInHu caseA)
(hInHu_sup_uInHu_eq_top data)
/-- **`ker(uActionHom) ≤ ker(aInvariantRestrictAut S₀)`**: an element acting trivially on the
*whole*
chief factor `H̄` acts trivially on the summand `S₀ ≤ H̄`. The subgroup inclusion behind
`C = C_U(H̄) ≤ C_U(S₀)`. -/
theorem ker_uActionHom_le_ker_aInvariantRestrictAut (caseA : CliffordCaseAData chars) :
(uActionHom data chief).ker ≤ (aInvariantRestrictAut caseA.S0_aInvariant).ker := by
intro x hx
rw [MonoidHom.mem_ker] at hx ⊢
ext s
rw [MulAut.one_apply, aInvariantRestrictAut_coe, hx, MulAut.one_apply]
/-- **`C = C_U(H̄) ≤ C_U(S₀)`** (`cSub ≤ cuSub`): centralizing the whole chief factor implies
centralizing the summand `S₀`. Both are `G`-images of kernels under the same double-map, and
`ker(uActionHom) ≤ ker(aInvariantRestrictAut S₀)`. -/
theorem cSub_le_cuSub (caseA : CliffordCaseAData chars) : cSub data chief ≤ cuSub caseA :=
Subgroup.map_mono (Subgroup.map_mono (ker_uActionHom_le_ker_aInvariantRestrictAut caseA))
theorem cInHu_le_cuInHu (caseA : CliffordCaseAData chars) : cInHu data chief ≤ cuInHu caseA :=
Subgroup.subgroupOf_mono _ (Subgroup.subgroupOf_mono _ (cSub_le_cuSub caseA))
/-- **`U' ≤ C_U(S₀)`** (`uprimeSub ≤ cuSub`, Peterfalvi (9.8.d)): the derived subgroup `U' = [U,U]`
lies in the single-factor centralizer `C_U(S₀)`, via `U' ≤ C = C_U(H̄) ≤ C_U(S₀)`
(`uprimeSub_le_cSub` then `cSub_le_cuSub`). This is the containment behind the (9.8.d) parameter
`λ ∈ Irr(C_U(S₀)/U')` and the `𝒮(H₀U')`-membership of the induced characters. -/
theorem uprimeSub_le_cuSub [Finite G] (caseA : CliffordCaseAData chars) :
uprimeSub data ≤ cuSub caseA :=
(uprimeSub_le_cSub data chief).trans (cSub_le_cuSub caseA)
/-- **`U' ≤ C_U(S₀)` realized inside `HU`** (`Uprime`/`uprimeSub` ⟶ `cuInHu`). The `HU`-realized
form
of `uprimeSub_le_cuSub`: `(U'.subgroupOf M).subgroupOf HU ≤ cuInHu`. Used to identify `λ` trivial
on
`U'` as a character of `C_U(S₀)/U'` in the (9.8.d) count. -/
theorem uprimeSub_subgroupOf_le_cuInHu [Finite G] (caseA : CliffordCaseAData chars) :
((uprimeSub data).subgroupOf M).subgroupOf (huSub data) ≤ cuInHu caseA :=
Subgroup.subgroupOf_mono _ (Subgroup.subgroupOf_mono _ (uprimeSub_le_cuSub caseA))
/-- **`H ⊓ U ≤ C_U(S₀)`** realized (`hInHu ⊓ uInHu ≤ cuInHu`): an `H ⊓ U` element centralizes the
chief factor `H̄` (`hInHu_inf_uInHu_le_cInHu`), hence `S₀ ≤ H̄` (`cInHu_le_cuInHu`). -/
theorem hInHu_inf_uInHu_le_cuInHu [Finite G] (caseA : CliffordCaseAData chars) :
hInHu data ⊓ uInHu data ≤ cuInHu caseA :=
(hInHu_inf_uInHu_le_cInHu data chief).trans (cInHu_le_cuInHu caseA)
/-- **`H ⊓ C_U(S₀) = ⊥`** realized (`hInHu ⊓ cuInHu = ⊥`). Since `C_U(S₀) ≤ U`, an element of
`hInHu ⊓ cuInHu` has `G`-image in `H ⊓ U = ⊥` (a type-P setup, `typeP_H_inf_U`), so it is trivial.
This is the trivial-intersection input `H ⊓ C_U(S₀) = ⊥` that makes the second isomorphism
`(H·C_U(S₀))/C_U(S₀) ≅ H` (used to build the `θ₁·λ` source character on `H·C_U(S₀)`), mirroring
`hInHu_inf_cInHu_eq_bot` for the (9.8.c) `hcLambdaHom`. -/
theorem hInHu_inf_cuInHu_eq_bot [Finite G] (caseA : CliffordCaseAData chars) :
hInHu data ⊓ cuInHu caseA = ⊥ := by
rw [eq_bot_iff]
intro x hx
obtain ⟨hxH, hxC⟩ := Subgroup.mem_inf.mp hx
have hxU := cuInHu_le_uInHu caseA hxC
rw [Subgroup.mem_bot]
have hxH' : x ∈ (data.H.subgroupOf M).subgroupOf (huSub data) := hxH
have hxU' : x ∈ (data.U.subgroupOf M).subgroupOf (huSub data) := hxU
have keyH : ((x : ↥M) : G) ∈ data.H :=
Subgroup.mem_subgroupOf.mp (Subgroup.mem_subgroupOf.mp hxH')
have keyU : ((x : ↥M) : G) ∈ data.U :=
Subgroup.mem_subgroupOf.mp (Subgroup.mem_subgroupOf.mp hxU')
have key : ((x : ↥M) : G) ∈ data.typeP.H ⊓ data.typeP.U := ⟨keyH, keyU⟩
rw [typeP_H_inf_U data.typeP, Subgroup.mem_bot] at key
exact Subtype.ext (Subtype.ext key)
/-- **`H·C_U(S₀) = C_U(S₀)·H`** (spelling bridge, mirrors `hcRealized_eq_cInHu_sup_hInHu`): the
`sup_comm` reorientation of the (9.8.d) inertia subgroup, used to phrase the `λ`-lift channel
`hcuLambdaHom` via the second isomorphism `(C_U(S₀)·H)/H ≅ C_U(S₀)`. -/
theorem hcuInHu_eq_cuInHu_sup_hInHu (caseA : CliffordCaseAData chars) :
hInHu data ⊔ cuInHu caseA = cuInHu caseA ⊔ hInHu data :=
sup_comm _ _
/-- **`|U| = a · |C_U(S₀)|`** (Peterfalvi (9.7.a)): the order of `U` splits as the Clifford index
`a = [U:C_U(S₀)]` times the centralizer order. Rearranges the first-isomorphism value `|U| = |Ū₁| ·
|C_U(S₀)|` (the `hII` step of `index_cuInHu_subgroupOf_uInHu_eq_a`) using the pin `a = |Ū₁|`
(`a_eq_card_restrictAut_range`). The arithmetic behind the (9.8.d) domain-count identity
`|U|/(a|U'|) = |C_U(S₀):U'|`. -/
theorem card_U_eq_a_mul_card_cuSub [Finite G] (caseA : CliffordCaseAData chars) :
Nat.card ↥data.U = caseA.a * Nat.card ↥(cuSub caseA) := by
rw [caseA.a_eq_card_restrictAut_range]
have h := Subgroup.card_eq_card_quotient_mul_card_subgroup
(aInvariantRestrictAut caseA.S0_aInvariant).ker
rw [Nat.card_congr (QuotientGroup.quotientKerEquivRange
(aInvariantRestrictAut caseA.S0_aInvariant)).toEquiv,
← card_cuSub_eq_card_ker caseA,
Nat.card_congr (Subgroup.subgroupOfEquivOfLe
(le_sup_left : data.typeP.U ≤ data.typeP.U ⊔ data.typeP.W1)).toEquiv] at h
exact h
/-- **`|U|/(a·|U'|) = |C_U(S₀):U'|`** (Peterfalvi (9.8.d) domain-count identity): the count
`((p-1)/a)·(|U|/(a|U'|))` in the (9.8.d) statement equals `((p-1)/a)·|C_U(S₀):U'|`, the number of
`(θ₁,λ)`-pairs divided by the `U`-orbit size `a`. Since `|U| = a·|C_U(S₀)|`
(`card_U_eq_a_mul_card_cuSub`) and `|U'| ∣ |C_U(S₀)|` (`U' ≤ C_U(S₀)`, `uprimeSub_le_cuSub`), the
`a` cancels: `|U|/(a·|U'|) = a·|C_U(S₀)|/(a·|U'|) = |C_U(S₀)|/|U'| = [C_U(S₀):U']`. Here
`[C_U(S₀):U'] = (uprimeSub data).relIndex (cuSub caseA)` (the relative index of `U'` in `C_U(S₀)`,
a genuine subgroup index since `U' ≤ C_U(S₀)`). -/
theorem card_U_div_a_mul_card_Uprime_eq_relIndex [Finite G] (caseA : CliffordCaseAData chars) :
Nat.card ↥data.U / (caseA.a * Nat.card ↥(uprimeSub data))
= (uprimeSub data).relIndex (cuSub caseA) := by
have hUprime_le : uprimeSub data ≤ cuSub caseA := uprimeSub_le_cuSub caseA
-- `[C_U(S₀):U'] · |U'| = |C_U(S₀)|` (via `[K:H]·|H| = |K|` for `H = U'.subgroupOf C_U(S₀)`)
have hrel : (uprimeSub data).relIndex (cuSub caseA) * Nat.card ↥(uprimeSub data)
= Nat.card ↥(cuSub caseA) := by
have h := Subgroup.index_mul_card ((uprimeSub data).subgroupOf (cuSub caseA))
rwa [Nat.card_congr (Subgroup.subgroupOfEquivOfLe hUprime_le).toEquiv] at h
rw [card_U_eq_a_mul_card_cuSub caseA, ← hrel]
-- `a·([C_U(S₀):U']·|U'|) / (a·|U'|) = [C_U(S₀):U']`
rw [show caseA.a * ((uprimeSub data).relIndex (cuSub caseA) * Nat.card ↥(uprimeSub data))
= ((uprimeSub data).relIndex (cuSub caseA)) * (caseA.a * Nat.card ↥(uprimeSub data)) by ring]
exact Nat.mul_div_cancel _ (Nat.mul_pos caseA.a_pos Nat.card_pos)
/-- `C_U(S₀) ≤ H·C_U(S₀)` (mirrors `cInHu_le_hcRealized`): `cuInHu` is contained in the inertia
subgroup `hInHu ⊔ cuInHu`. -/
theorem cuInHu_le_hcuInHu (caseA : CliffordCaseAData chars) :
cuInHu caseA ≤ hInHu data ⊔ cuInHu caseA :=
le_sup_right
/-- **The `λ`-lift `H·C_U(S₀) →* ℂˣ`** of a linear character `λ : C_U(S₀) →* ℂˣ` (the `C`-factor of
the (9.8.d) pair character `θ₁·λ`): the composite `H·C_U(S₀) → H·C_U(S₀)/H ≅ C_U(S₀)/(C_U(S₀) ⊓ H)
= C_U(S₀) —λ→ ℂˣ`, using the trivial intersection `H ⊓ C_U(S₀) = ⊥` (`hInHu_inf_cuInHu_eq_bot`).
Mirrors the (9.9.c) `hcLambdaHom` rewired from `cInHu` to `cuInHu`; `hInHu` is normal in the sup, so
the quotient map is well-defined. Kills `H` (`hcuLambdaHom_eq_one_of_mem_hInHu`) and restricts to
`λ` on `C_U(S₀)` (`hcuLambdaHom_inclusion`). -/
noncomputable def hcuLambdaHom [Finite G] (caseA : CliffordCaseAData chars)
(lam : ↥(cuInHu caseA) →* ℂˣ) :
↥(hInHu data ⊔ cuInHu caseA) →* ℂˣ :=
haveI := hInHu_normal data
letI : ((hInHu data).subgroupOf (cuInHu caseA ⊔ hInHu data)).Normal :=
(hInHu_normal data).subgroupOf _
(QuotientGroup.lift ((hInHu data).subgroupOf (cuInHu caseA)) lam
(fun x hx => by
have hx1 : x = 1 := by
have hmem : (x : ↥(huSub data)) ∈ hInHu data ⊓ cuInHu caseA :=
⟨Subgroup.mem_subgroupOf.mp hx, x.2⟩
rw [hInHu_inf_cuInHu_eq_bot caseA, Subgroup.mem_bot] at hmem
exact Subtype.ext hmem
rw [hx1]
exact lam.ker.one_mem)).comp
((QuotientGroup.quotientInfEquivProdNormalQuotient (cuInHu caseA)
(hInHu data)).symm.toMonoidHom.comp
((QuotientGroup.mk' ((hInHu data).subgroupOf (cuInHu caseA ⊔ hInHu data))).comp
(MulEquiv.subgroupCongr (hcuInHu_eq_cuInHu_sup_hInHu caseA)).toMonoidHom))
/-- **`hcuLambdaHom` kills `H`** (mirrors `hcLambdaHom_eq_one_of_mem_hInHu`): the `λ`-lift is
trivial
on the `H`-part of `H·C_U(S₀)` (the quotient map by `hInHu` kills it). So the pair character
`θ₁·λ` restricts on `hInHu` to the plain seed inflation `θ₀`, and the (9.8.d) inertia lift
`inertia_eq_hcuInHu` applies to the pair unchanged. -/
theorem hcuLambdaHom_eq_one_of_mem_hInHu [Finite G] (caseA : CliffordCaseAData chars)
(lam : ↥(cuInHu caseA) →* ℂˣ)
{x : ↥(hInHu data ⊔ cuInHu caseA)}
(hx : x ∈ (hInHu data).subgroupOf (hInHu data ⊔ cuInHu caseA)) :
hcuLambdaHom caseA lam x = 1 := by
have := hInHu_normal data
let : ((hInHu data).subgroupOf (cuInHu caseA ⊔ hInHu data)).Normal :=
(hInHu_normal data).subgroupOf _
simp only [hcuLambdaHom, MonoidHom.comp_apply, MulEquiv.coe_toMonoidHom,
QuotientGroup.mk'_apply]
have hmem : (MulEquiv.subgroupCongr (hcuInHu_eq_cuInHu_sup_hInHu caseA)) x
∈ (hInHu data).subgroupOf (cuInHu caseA ⊔ hInHu data) :=
Subgroup.mem_subgroupOf.mpr (Subgroup.mem_subgroupOf.mp hx)
rw [(QuotientGroup.eq_one_iff _).mpr hmem, map_one, map_one]
/-- **`hcuLambdaHom` restricts to `λ` on `C_U(S₀)`** (mirrors `hcLambdaHom_inclusion`): on the
inclusion of `c ∈ cuInHu` into `H·C_U(S₀)`, the `λ`-lift returns `λ c`. The second iso sends the
`cuInHu`-class to the `H·C_U(S₀)`-class via inclusion (`hfwd`), so the reversed iso undoes the
quotient map and the lift evaluates `λ`. -/
theorem hcuLambdaHom_inclusion [Finite G] (caseA : CliffordCaseAData chars)
(lam : ↥(cuInHu caseA) →* ℂˣ) (c : ↥(cuInHu caseA)) :
hcuLambdaHom caseA lam (Subgroup.inclusion (cuInHu_le_hcuInHu caseA) c) = lam c := by
have := hInHu_normal data
let : ((hInHu data).subgroupOf (cuInHu caseA ⊔ hInHu data)).Normal :=
(hInHu_normal data).subgroupOf _
have hfwd : (QuotientGroup.quotientInfEquivProdNormalQuotient (cuInHu caseA)
(hInHu data))
(QuotientGroup.mk' _ c)
= QuotientGroup.mk' _ (Subgroup.inclusion le_sup_left c) := by
simp only [QuotientGroup.quotientInfEquivProdNormalQuotient,
QuotientGroup.quotientInfEquivProdNormalizerQuotient, MulEquiv.trans_apply,
QuotientGroup.quotientKerEquivOfSurjective,
QuotientGroup.quotientKerEquivOfRightInverse, MulEquiv.coe_mk]
rfl
simp only [hcuLambdaHom, MonoidHom.comp_apply, MulEquiv.coe_toMonoidHom]
have hcongr : (MulEquiv.subgroupCongr (hcuInHu_eq_cuInHu_sup_hInHu caseA))
(Subgroup.inclusion (cuInHu_le_hcuInHu caseA) c)
= Subgroup.inclusion le_sup_left c := by
apply Subtype.ext
rfl
rw [hcongr, QuotientGroup.mk'_apply, show ((Subgroup.inclusion le_sup_left c :
↥(cuInHu caseA ⊔ hInHu data)) : ↥(cuInHu caseA ⊔ hInHu data)
⧸ (hInHu data).subgroupOf (cuInHu caseA ⊔ hInHu data))
= QuotientGroup.mk' _ (Subgroup.inclusion le_sup_left c) from rfl, ← hfwd,
MulEquiv.symm_apply_apply, QuotientGroup.mk'_apply, QuotientGroup.lift_mk]
/-- **`U ⊓ H·C_U(S₀) = C_U(S₀)`** realized (`uInHu ⊓ (hInHu ⊔ cuInHu) = cuInHu`), the second-iso
input for `[HU : H·C_U(S₀)] = [U : C_U(S₀)]`. Mirrors `uInHu_inf_hcInHu_eq_cInHu`. -/
theorem uInHu_inf_hcuInHu_eq_cuInHu [Finite G] (caseA : CliffordCaseAData chars) :
uInHu data ⊓ (hInHu data ⊔ cuInHu caseA) = cuInHu caseA := by
have := hInHu_normal data
apply le_antisymm
· rintro x ⟨hxU, hxHC⟩
obtain ⟨hh, hhmem, cc, ccmem, rfl⟩ := Subgroup.mem_sup_of_normal_left.mp hxHC
have hcc_u : cc ∈ uInHu data := cuInHu_le_uInHu caseA ccmem
have hh_u : hh ∈ uInHu data := by
have h1 : hh * cc * cc⁻¹ ∈ uInHu data :=
(uInHu data).mul_mem hxU ((uInHu data).inv_mem hcc_u)
rwa [mul_inv_cancel_right] at h1
have hh_c : hh ∈ cuInHu caseA :=
hInHu_inf_uInHu_le_cuInHu caseA (Subgroup.mem_inf.mpr ⟨hhmem, hh_u⟩)
exact (cuInHu caseA).mul_mem hh_c ccmem
· exact le_inf (cuInHu_le_uInHu caseA) le_sup_right
/-- **Second-iso index step: `[HU : H·C_U(S₀)] = [U : C_U(S₀)]`** (realized
`(hInHu ⊔ cuInHu).index = (cuInHu.subgroupOf uInHu).index`). Mirrors
`index_hcInHu_eq_relindex_cInHu`: the second isomorphism theorem for `H·C_U(S₀) ◁ HU` with
`uInHu ⊔ H·C_U(S₀) = ⊤`, and `uInHu ⊓ H·C_U(S₀) = cuInHu`. -/
theorem index_hcuInHu_eq_relindex_cuInHu [Finite G] (caseA : CliffordCaseAData chars) :
(hInHu data ⊔ cuInHu caseA).index
= ((cuInHu caseA).subgroupOf (uInHu data)).index := by
have : (hInHu data ⊔ cuInHu caseA).Normal := hcuInHu_normal caseA
have htop : uInHu data ⊔ (hInHu data ⊔ cuInHu caseA) = ⊤ := by
rw [← sup_assoc, sup_comm (uInHu data) (hInHu data), hInHu_sup_uInHu_eq_top, top_sup_eq]
have he := Nat.card_congr (QuotientGroup.quotientInfEquivProdNormalQuotient
(uInHu data) (hInHu data ⊔ cuInHu caseA)).toEquiv
have hsub : (hInHu data ⊔ cuInHu caseA).subgroupOf (uInHu data)
= (cuInHu caseA).subgroupOf (uInHu data) := by
ext x
simp only [Subgroup.mem_subgroupOf]
constructor
· intro hx
have hxin : (x : ↥(huSub data)) ∈ uInHu data ⊓ (hInHu data ⊔ cuInHu caseA) :=
Subgroup.mem_inf.mpr ⟨x.2, hx⟩
rw [uInHu_inf_hcuInHu_eq_cuInHu caseA] at hxin
exact hxin
· intro hx; exact Subgroup.mem_sup_right hx
rw [hsub] at he
have hsplit := Subgroup.card_eq_card_quotient_mul_card_subgroup
((hInHu data ⊔ cuInHu caseA).subgroupOf (uInHu data ⊔ (hInHu data ⊔ cuInHu caseA)))
rw [Nat.card_congr (Subgroup.subgroupOfEquivOfLe (le_sup_right :
(hInHu data ⊔ cuInHu caseA) ≤ uInHu data ⊔ (hInHu data ⊔ cuInHu caseA))).toEquiv,
← he, ← Subgroup.index_eq_card] at hsplit
have htopcard : Nat.card ↥(uInHu data ⊔ (hInHu data ⊔ cuInHu caseA))
= Nat.card ↥(huSub data) := by
rw [htop]; exact Nat.card_congr Subgroup.topEquiv.toEquiv
rw [htopcard] at hsplit
have hmul := Subgroup.card_mul_index (hInHu data ⊔ cuInHu caseA)
rw [hsplit, mul_comm (((cuInHu caseA).subgroupOf (uInHu data)).index)] at hmul
exact Nat.eq_of_mul_eq_mul_left Nat.card_pos hmul
/-- **`[U : C_U(S₀)] = |Ū₁|`** (realized
`(cuInHu.subgroupOf uInHu).index = |range(aInvariantRestrictAut S₀)|`).
Mirrors `index_cInHu_subgroupOf_uInHu_eq_u`: the first isomorphism theorem for the restricted
`U`-action `aInvariantRestrictAut caseA.S0_aInvariant` on `S₀`, whose image `Ū₁` has order the index
`a` of `C_U(S₀)` in `U`. This is the value `clifford_caseA_data` assigns to
`CliffordCaseAData.a`. -/
theorem index_cuInHu_subgroupOf_uInHu_eq_a [Finite G] (caseA : CliffordCaseAData chars) :
((cuInHu caseA).subgroupOf (uInHu data)).index
= Nat.card ↥(aInvariantRestrictAut caseA.S0_aInvariant).range := by
-- (I): `|C_U(S₀)| · [U:C_U(S₀)] = |U|`.
have hI : Nat.card ↥(cuSub caseA) * ((cuInHu caseA).subgroupOf (uInHu data)).index
= Nat.card ↥data.U := by
have h := Subgroup.card_mul_index ((cuInHu caseA).subgroupOf (uInHu data))
rw [Nat.card_congr (Subgroup.subgroupOfEquivOfLe (cuInHu_le_uInHu caseA)).toEquiv,
card_cuInHu_eq caseA, card_uInHu_eq data] at h
exact h
-- (II): `|U| = |Ū₁| · |C_U(S₀)|` (first iso for the restricted action hom, domain `≃* ↥U`).
have hII : Nat.card ↥data.U
= Nat.card ↥(aInvariantRestrictAut caseA.S0_aInvariant).range * Nat.card ↥(cuSub caseA) := by
have h := Subgroup.card_eq_card_quotient_mul_card_subgroup
(aInvariantRestrictAut caseA.S0_aInvariant).ker
rw [Nat.card_congr (QuotientGroup.quotientKerEquivRange
(aInvariantRestrictAut caseA.S0_aInvariant)).toEquiv,
← card_cuSub_eq_card_ker caseA,
Nat.card_congr (Subgroup.subgroupOfEquivOfLe
(le_sup_left : data.typeP.U ≤ data.typeP.U ⊔ data.typeP.W1)).toEquiv] at h
exact h
have hcancel : Nat.card ↥(cuSub caseA)
* ((cuInHu caseA).subgroupOf (uInHu data)).index
= Nat.card ↥(cuSub caseA) * Nat.card ↥(aInvariantRestrictAut caseA.S0_aInvariant).range := by
rw [hI, hII, mul_comm]
exact Nat.eq_of_mul_eq_mul_left Nat.card_pos hcancel
/-- **`[HU : H·C_U(S₀)] = |Ū₁| = a`** (Peterfalvi (9.8.d) degree index). The inertia subgroup of
the
degree-`qa` source character `θ₁·λ` has index `a` in `HU`, giving the source degree `a` and (after
`Ind_{HU}^M`) the character degree `qa`. Combines the second-iso step
`[HU:H·C_U(S₀)] = [U:C_U(S₀)]`
(`index_hcuInHu_eq_relindex_cuInHu`) with the first-iso value `[U:C_U(S₀)] = |Ū₁|`
(`index_cuInHu_subgroupOf_uInHu_eq_a`). Here `|Ū₁| = Nat.card (aInvariantRestrictAut …).range` is
the
genuine geometric `a = |U:C_U(H₁)|`, the value `clifford_caseA_data` assigns to `caseA.a`. -/
theorem index_hcuInHu_eq_a [Finite G] (caseA : CliffordCaseAData chars) :
(hInHu data ⊔ cuInHu caseA).index
= Nat.card ↥(aInvariantRestrictAut caseA.S0_aInvariant).range :=
(index_hcuInHu_eq_relindex_cuInHu caseA).trans (index_cuInHu_subgroupOf_uInHu_eq_a caseA)
/-- **`[HU : H·C_U(S₀)] = a`** (Peterfalvi (9.8.d), with `a = CliffordCaseAData.a`). The genuine
geometric index `[HU : H·C_U(S₀)] = |Ū₁|` (`index_hcuInHu_eq_a`) equals the carrier's `a`, since `a`
is pinned to `|Ū₁|` (`CliffordCaseAData.a_eq_card_restrictAut_range`). This is the degree of the
(9.8.d) source character `Ind_{H·C_U(S₀)}^{HU}(θ₁·λ)`, whence `Ind_{HU}^M` of it has degree `qa`. -/
theorem index_hcuInHu_eq_caseA_a [Finite G] (caseA : CliffordCaseAData chars) :
(hInHu data ⊔ cuInHu caseA).index = caseA.a := by
rw [index_hcuInHu_eq_a caseA, caseA.a_eq_card_restrictAut_range]
end CuS0
/-- **realized `H₀U' = (realized H₀) ⊔ (realized U')`** (Peterfalvi (9.8.d); mirror of
`realizedH0supCprime_eq_realizedH0_sup_cprimeInHu`): the realized `H₀U'` inside `HU` equals the join
of the realized `H₀` and the realized `U'`. Feeds the `h₀·u'` decomposition of the (9.8.d) pair
character's kernel computation. -/
theorem realizedH0supUprime_eq_realizedH0_sup_uprimeInHu {M : Subgroup G}
{data : TypesIIIIIIVSetup M} (chief : ChiefFactorData data) :
((chief.H0 ⊔ uprimeSub data).subgroupOf M).subgroupOf (huSub data)
= (chief.H0.subgroupOf M).subgroupOf (huSub data)
⊔ ((uprimeSub data).subgroupOf M).subgroupOf (huSub data) := by
have hH0M : chief.H0 ≤ M := chief.H0_lt_H.le.trans (H_le_M data)
have hUM : uprimeSub data ≤ M := (uprimeSub_le_U data).trans (U_le_M data)
rw [Subgroup.subgroupOf_sup hH0M hUM]
have hH0sub : (chief.H0.subgroupOf M) ≤ huSub data :=
Subgroup.subgroupOf_mono M (le_trans chief.H0_lt_H.le le_sup_left)
have hUsub : (uprimeSub data).subgroupOf M ≤ huSub data :=
Subgroup.subgroupOf_mono M (le_trans (uprimeSub_le_U data) le_sup_right)
rw [Subgroup.subgroupOf_sup hH0sub hUsub]
/-- **(9.7) Clifford dimension dichotomy** (the arithmetic heart of (9.7)). Restricting the
`U W₁`-action on the chief factor `H̄ = H/H₀` to `U`, there is a minimal `U`-invariant `S₀ ≠ ⊥` of
order `p^d` with `0 < d` and `d ∣ q`. Since `q` is prime, `d ∈ {1, q}`: the dichotomy of (9.7),
case (a) (`d = 1`, `U` semisimple into order-`p` pieces) vs case (b) (`d = q`, `U` irreducible).
Assembled from the spanning step (`iSup_smul_eq_top_of_irreducible`, on the `U W₁`-irreducible
`H̄`),
the orbit lemmas (each `U W₁`-translate of `S₀` is `U`-irreducible of order `|S₀|`), and the order
step (`card_eq_pow_of_iSup_aInvariant_irreducible`): `|H̄| = |S₀|^k`, while `|H̄| = p^q`, so
`p^q = (p^d)^k`, i.e. `q = d·k`. -/
theorem chiefFactor_clifford_dim_dvd_q [Finite G] {M : Subgroup G}
{data : TypesIIIIIIVSetup M} (chief : ChiefFactorData data) :
∃ (d : ℕ) (S₀ : Subgroup (↥data.H ⧸ chief.N)),
0 < d ∧ d ∣ data.q ∧ S₀ ≠ ⊥ ∧ Nat.card ↥S₀ = chief.p ^ d ∧
IsAInvariant ((typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).φ.comp (typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).U.subtype) S₀ ∧
(∀ J, IsAInvariant ((typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).φ.comp (typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).U.subtype) J → J ≤ S₀ → J = ⊥ ∨ J = S₀) := by
classical
have hU : data.typeP.U ≠ ⊥ := data.nontrivial.1
have : chief.N.Normal := chief.N_normal
have := Fact.mk chief.p_prime
-- The descended `U W₁`-action on the chief factor `H̄` and its restriction to the kernel `U`.
set act := typeP_quotientCoprimeAction data.typeP hU chief.N_aInvariant with hact
set φU := act.φ.comp act.U.subtype with hφU
have hUnorm : act.U.Normal := (typeP_uW1_frobenius data.typeP hU).isNormal
have hKcard : Nat.card (↥data.H ⧸ chief.N) = chief.p ^ data.q := chiefFactor_quotient_card chief
have hq_pos : 0 < data.q := Nat.card_pos
have hKnt : Nontrivial (↥data.H ⧸ chief.N) :=
Finite.one_lt_card_iff_nontrivial.mp (by
rw [hKcard]; exact Nat.one_lt_pow hq_pos.ne' chief.p_prime.one_lt)
-- A minimal nonzero `U`-invariant subgroup `S₀` (the `U`-irreducible Clifford piece).
set T : Set (Subgroup (↥data.H ⧸ chief.N)) := {J | J ≠ ⊥ ∧ IsAInvariant φU J} with hT
obtain ⟨S₀, ⟨hS₀ne, hS₀inv⟩, hS₀min⟩ :=
(Set.toFinite T).exists_minimal ⟨⊤, top_ne_bot, IsAInvariant.top _⟩
have hirr₀ : ∀ J, IsAInvariant φU J → J ≤ S₀ → J = ⊥ ∨ J = S₀ := by
intro J hJinv hJle
by_cases hJ0 : J = ⊥
· exact Or.inl hJ0
· exact Or.inr (le_antisymm hJle (hS₀min ⟨hJ0, hJinv⟩ hJle))
-- `|S₀| = p^d` (subgroup of the elementary abelian `p`-group `H̄`).
obtain ⟨d, hd⟩ := (chief.quotient_elementaryAbelian.to_subgroup S₀).isPGroup.exists_card_eq
-- Spanning step (full action) + order step (restricted `U`-action) give `|H̄| = |S₀|^k`.
have hspan : ⨆ a, act.φ a • S₀ = ⊤ :=
iSup_smul_eq_top_of_irreducible (φ := act.φ) chief.quotient_chiefFactor hS₀ne
obtain ⟨k, hk⟩ := card_eq_pow_of_iSup_aInvariant_irreducible
(φ := φU) (S := fun a => act.φ a • S₀) (n := Nat.card ↥S₀)
(fun x y => chief.quotient_elementaryAbelian.comm x y) hspan
(fun a => isAInvariant_comp_subtype_pointwise_smul hUnorm hS₀inv a)
(fun a J hJinv hJle => forall_aInvariant_le_pointwise_smul hUnorm hirr₀ a J hJinv hJle)
(fun a _ => card_pointwise_smul act.φ a S₀)
-- Arithmetic: `p^q = (p^d)^k = p^{d·k}`, so `q = d·k`.
rw [hd, hKcard, ← pow_mul] at hk
have hdk : data.q = d * k := Nat.pow_right_injective chief.p_prime.two_le hk
have hdpos : 0 < d := by
rcases Nat.eq_zero_or_pos d with h0 | h
· rw [h0, pow_zero] at hd; exact absurd (Subgroup.card_eq_one.mp hd) hS₀ne
· exact h
exact ⟨d, S₀, hdpos, ⟨k, hdk⟩, hS₀ne, hd, hS₀inv, hirr₀⟩
/-- **(9.7) structural dichotomy** (the Clifford case split, read off the chief factor). With
`q = |W₁|` prime, the chief factor `H̄` is, under the restricted `U`-action, *either*
`U`-irreducible (Clifford case (b): every `U`-invariant subgroup is `⊥` or `⊤`) *or* contains a
`U`-invariant subgroup of order `p` (Clifford case (a)).
This is the arithmetic dichotomy `d ∈ {1, q}` of `chiefFactor_clifford_dim_dvd_q` read through the
primality of `q`: the minimal nonzero `U`-invariant piece `S₀` has order `p^d` with `d ∣ q`, so
either `d = q` (`|S₀| = p^q = |H̄|`, hence `S₀ = ⊤`, and minimality forces `U`-irreducibility)
or `d = 1` (`|S₀| = p`). Packaging the two cases into the carriers `CliffordCaseAData` /
`CliffordCaseBData` (the order-`p` factor pullback / the `End_{𝔽ₚ[U]}(H̄)` field model) is the
remaining work of `clifford_dichotomy`. -/
theorem chiefFactor_clifford_U_dichotomy [Finite G] {M : Subgroup G}
{data : TypesIIIIIIVSetup M} (chief : ChiefFactorData data) :
(∀ J : Subgroup (↥data.H ⧸ chief.N),
IsAInvariant ((typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).φ.comp (typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).U.subtype) J → J = ⊥ ∨ J = ⊤) ∨
(∃ S₀ : Subgroup (↥data.H ⧸ chief.N), S₀ ≠ ⊥ ∧
IsAInvariant ((typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).φ.comp (typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).U.subtype) S₀ ∧ Nat.card ↥S₀ = chief.p ∧
∀ J, IsAInvariant ((typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).φ.comp (typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).U.subtype) J → J ≤ S₀ → J = ⊥ ∨ J = S₀) := by
obtain ⟨d, S₀, _hdpos, hdq, hS₀ne, hcard, hS₀inv, hirr₀⟩ :=
chiefFactor_clifford_dim_dvd_q chief
-- `q = |W₁|` is prime and `d ∣ q`, so `d = 1` or `d = q`.
have hq_prime : (data.q).Prime := data.nontrivial.2.1
rcases hq_prime.eq_one_or_self_of_dvd d hdq with hd1 | hdq2
· -- `d = 1`: a `U`-invariant subgroup of order `p`. Clifford case (a).
exact Or.inr ⟨S₀, hS₀ne, hS₀inv, by rw [hcard, hd1, pow_one], hirr₀⟩
· -- `d = q`: `|S₀| = p^q = |H̄|` ⟹ `S₀ = ⊤`; minimality ⟹ `U`-irreducible. Case (b).
have hS₀top : S₀ = ⊤ :=
Subgroup.eq_top_of_card_eq _ (by rw [hcard, hdq2, chiefFactor_quotient_card chief])
refine Or.inl fun J hJinv => ?_
by_cases hJ : J = ⊥
· exact Or.inl hJ
· exact Or.inr
(((hirr₀ J hJinv (le_top.trans_eq hS₀top.symm)).resolve_left hJ).trans hS₀top)
open OddOrder.RepresentationTheory Representation in
open scoped commutatorElement IsMulCommutative in
/-- **Peterfalvi (9.7) case (b), chief-factor Singer conclusion.** When `U` acts irreducibly on
the chief factor `H̄ = H/H₀` (Clifford case (b) — the left branch of
`chiefFactor_clifford_U_dichotomy`), its image `Ū = φ_U(U) ≤ Aut(H̄)` is *cyclic*, of order
dividing `p^q - 1`.
`Ū` is abelian because `[U, U]` centralizes `H` (Peterfalvi (8.5.b),
`typeP_commutator_U_centralizes_H`): a commutator `⁅a, b⁆ ∈ [U, U]` acts trivially on `H̄`, so
`φ_U ⁅a, b⁆ = 1` and the image is commutative. Faithfulness of the inclusion `Ū ↪ Aut(H̄)` and the
irreducibility transferred from the case-(b) hypothesis feed the canonical module-level Singer
mechanism `isCyclic_and_card_dvd_of_faithful_irreducible_comm` (shared `SingerField` leaf) through
the thin adapter `elabRepresentation_isSimpleModule_and_faithful`, with `H̄` the elementary abelian
`p`-group of order `p^q` (`chiefFactor_quotient_card`). This is the structural core behind the
`Coprime u (p-1)` / `u ∣ (p^q-1)/(p-1)` divisibilities of `CliffordCaseBData` (which additionally
pin `u = |Ū|` and use the `W₁`-fixed-point-free refinement). -/
theorem chiefFactor_caseB_image_cyclic [Finite G] {M : Subgroup G}
{data : TypesIIIIIIVSetup M} (chief : ChiefFactorData data)
(hcaseB : ∀ J : Subgroup (↥data.H ⧸ chief.N),
IsAInvariant ((typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).φ.comp (typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).U.subtype) J → J = ⊥ ∨ J = ⊤) :
IsCyclic ↥(((typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).φ.comp (typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).U.subtype).range) ∧
Nat.card ↥(((typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).φ.comp (typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).U.subtype).range) ∣ chief.p ^ data.q - 1 := by
classical
have hU : data.typeP.U ≠ ⊥ := data.nontrivial.1
have : chief.N.Normal := chief.N_normal
have : NeZero chief.p := ⟨chief.p_prime.pos.ne'⟩
set act := typeP_quotientCoprimeAction data.typeP hU chief.N_aInvariant with hact
set φU := act.φ.comp act.U.subtype with hφU
-- `H̄ = ↥H ⧸ N` is finite elementary abelian `p`, nontrivial, of order `p^q`.
have hKcard : Nat.card (↥data.H ⧸ chief.N) = chief.p ^ data.q := chiefFactor_quotient_card chief
have hq_pos : 0 < data.q := Nat.card_pos
have hKnt : Nontrivial (↥data.H ⧸ chief.N) :=
Finite.one_lt_card_iff_nontrivial.mp (by
rw [hKcard]; exact Nat.one_lt_pow hq_pos.ne' chief.p_prime.one_lt)
-- Module instances for the Singer mechanism, from `chief.quotient_elementaryAbelian`.
let : IsMulCommutative (↥data.H ⧸ chief.N) :=
IsMulCommutative.of_comm chief.quotient_elementaryAbelian.comm
let : CommGroup (↥data.H ⧸ chief.N) := inferInstance
let : Module (ZMod chief.p) (Additive (↥data.H ⧸ chief.N)) :=
chief.quotient_elementaryAbelian.zmodModule
-- An element of `U W₁` centralizing `H` acts trivially on `H̄`.
have hcentral_triv : ∀ g : ↥(data.typeP.U ⊔ data.typeP.W1),
(g : G) ∈ Subgroup.centralizer (data.typeP.H : Set G) → act.φ g = 1 := by
intro g hg
have hfix : ∀ x : ↥data.typeP.H, (typeP_conjAction data.typeP g) x = x := by
intro x
apply Subtype.ext
rw [typeP_conjAction_apply]
have hcom : (x : G) * (g : G) = (g : G) * (x : G) :=
(Subgroup.mem_centralizer_iff.mp hg) (x : G) x.2
rw [← hcom, mul_assoc, mul_inv_cancel, mul_one]
ext y
refine QuotientGroup.induction_on y ?_
intro x
change (act.φ g) (QuotientGroup.mk' chief.N x) = QuotientGroup.mk' chief.N x
have hstep : (act.φ g) (QuotientGroup.mk' chief.N x)
= QuotientGroup.mk' chief.N ((typeP_conjAction data.typeP g) x) := rfl
rw [hstep, hfix x]
-- `Ū = φU.range` is abelian: `φU ⁅a, b⁆ = 1` since `⁅a, b⁆` maps into `[U, U] ⊆ C(H)`.
have hComm : ∀ a b : ↥act.U, Commute (φU a) (φU b) := by
intro a b
refine commutatorElement_eq_one_iff_commute.mp ?_
rw [← map_commutatorElement φU a b]
change act.φ (act.U.subtype ⁅a, b⁆) = 1
apply hcentral_triv
change ((data.typeP.U ⊔ data.typeP.W1).subtype.comp act.U.subtype) ⁅a, b⁆
∈ Subgroup.centralizer (data.typeP.H : Set G)
rw [map_commutatorElement]
exact typeP_commutator_U_centralizes_H data.typeP
(Subgroup.commutator_mem_commutator
(Subgroup.mem_subgroupOf.mp a.2) (Subgroup.mem_subgroupOf.mp b.2))
-- Package the data for the Singer mechanism.
have hAcomm : ∀ s t : ↥(φU.range), s * t = t * s := by
rintro ⟨_, a, rfl⟩ ⟨_, b, rfl⟩
exact Subtype.ext (hComm a b)
have hirr : ∀ J : Subgroup (↥data.H ⧸ chief.N),
IsAInvariant (φU.range).subtype J → J = ⊥ ∨ J = ⊤ := by
intro J hJ
apply hcaseB
intro a
exact hJ ⟨φU a, MonoidHom.mem_range.mpr ⟨a, rfl⟩⟩
have hfaith : ∀ a : ↥(φU.range),
(∀ x : (↥data.H ⧸ chief.N), ((φU.range).subtype a) x = x) → a = 1 := by
intro a ha
have hone : (a : MulAut (↥data.H ⧸ chief.N)) = 1 := by
ext x
simpa using ha x
exact Subtype.ext hone
-- Thin adapter into the canonical module-level Singer lemma (shared leaf, issue 9000).
obtain ⟨hcyc, hdvd⟩ := singerAdapter_isCyclic_card_dvd
(A := ↥(φU.range)) (K := ↥data.H ⧸ chief.N) (p := chief.p)
(φ := (φU.range).subtype) hAcomm hKnt hirr hfaith
exact ⟨hcyc, by rwa [hKcard] at hdvd⟩
open OddOrder.RepresentationTheory Representation in
open scoped commutatorElement IsMulCommutative in
/-- **Peterfalvi (9.7) case (b), the fixed-point-free coprimality `Coprime |Ū| (p-1)`.** When `U`
acts irreducibly on the chief factor `H̄ = H/H₀` (Clifford case (b)), the image `Ū = φ_U(U)` has
order coprime to `p - 1`.
This discharges the hard `coprime` hypothesis of `chiefFactor_caseB_image_dvd_norm` from the
Frobenius structure of `U W₁`: a nonidentity `w₀ ∈ W₁` acts on `Ū` fixed-point-freely. Indeed if
`act.φ(w₀)` commutes with `φ_U(u)` then `⁅w₀, u⁆` acts trivially on `H̄`, so `u` lies in a
`w₀`-fixed coset of `C_U(H̄)`; Isaacs Cor 3.28 (`coprime_fixedPoints_quotient`) extracts a
`w₀`-fixed
representative `c ∈ u·C_U(H̄)`, and `c ∈ C_U(w₀) = 1` by the Frobenius condition
(`centralizer_complement_le`, `U ⊓ W₁ = ⊥`), forcing `u ∈ C_U(H̄)`, i.e. `φ_U(u) = 1`. The
canonical module-level `coprime_card_sub_one_of_faithful_irreducible_comm_fpf` (shared
`SingerField` leaf, via `elabRepresentation_isSimpleModule_and_faithful` +
`exists_addEquiv_asModule_fpf`) then gives the coprimality. -/
theorem chiefFactor_caseB_image_coprime [Finite G] {M : Subgroup G}
{data : TypesIIIIIIVSetup M} (chief : ChiefFactorData data)
(hcaseB : ∀ J : Subgroup (↥data.H ⧸ chief.N),
IsAInvariant ((typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).φ.comp (typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).U.subtype) J → J = ⊥ ∨ J = ⊤) :
Nat.Coprime (Nat.card ↥(((typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).φ.comp (typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).U.subtype).range)) (chief.p - 1) := by
classical
have hU : data.typeP.U ≠ ⊥ := data.nontrivial.1
have : chief.N.Normal := chief.N_normal
have : NeZero chief.p := ⟨chief.p_prime.pos.ne'⟩
have : Fact chief.p.Prime := ⟨chief.p_prime⟩
set act := typeP_quotientCoprimeAction data.typeP hU chief.N_aInvariant with hact
set φU := act.φ.comp act.U.subtype with hφU
have hUnorm : act.U.Normal := (typeP_uW1_frobenius data.typeP hU).isNormal
have hKcard : Nat.card (↥data.H ⧸ chief.N) = chief.p ^ data.q := chiefFactor_quotient_card chief
have hq_pos : 0 < data.q := Nat.card_pos
have hKnt : Nontrivial (↥data.H ⧸ chief.N) :=
Finite.one_lt_card_iff_nontrivial.mp (by
rw [hKcard]; exact Nat.one_lt_pow hq_pos.ne' chief.p_prime.one_lt)
let : IsMulCommutative (↥data.H ⧸ chief.N) :=
IsMulCommutative.of_comm chief.quotient_elementaryAbelian.comm
let : CommGroup (↥data.H ⧸ chief.N) := inferInstance
let : Module (ZMod chief.p) (Additive (↥data.H ⧸ chief.N)) :=
chief.quotient_elementaryAbelian.zmodModule
-- Abelianness, irreducibility, faithfulness of `Ū = φU.range` (as in
-- `chiefFactor_caseB_image_cyclic`).
have hcentral_triv : ∀ g : ↥(data.typeP.U ⊔ data.typeP.W1),
(g : G) ∈ Subgroup.centralizer (data.typeP.H : Set G) → act.φ g = 1 := by
intro g hg
have hfix : ∀ x : ↥data.typeP.H, (typeP_conjAction data.typeP g) x = x := by
intro x
apply Subtype.ext
rw [typeP_conjAction_apply]
have hcom : (x : G) * (g : G) = (g : G) * (x : G) :=
(Subgroup.mem_centralizer_iff.mp hg) (x : G) x.2
rw [← hcom, mul_assoc, mul_inv_cancel, mul_one]
ext y
refine QuotientGroup.induction_on y ?_
intro x
change (act.φ g) (QuotientGroup.mk' chief.N x) = QuotientGroup.mk' chief.N x
have hstep : (act.φ g) (QuotientGroup.mk' chief.N x)
= QuotientGroup.mk' chief.N ((typeP_conjAction data.typeP g) x) := rfl
rw [hstep, hfix x]
have hComm : ∀ a b : ↥act.U, Commute (φU a) (φU b) := by
intro a b
refine commutatorElement_eq_one_iff_commute.mp ?_
rw [← map_commutatorElement φU a b]
change act.φ (act.U.subtype ⁅a, b⁆) = 1
apply hcentral_triv
change ((data.typeP.U ⊔ data.typeP.W1).subtype.comp act.U.subtype) ⁅a, b⁆
∈ Subgroup.centralizer (data.typeP.H : Set G)
rw [map_commutatorElement]
exact typeP_commutator_U_centralizes_H data.typeP
(Subgroup.commutator_mem_commutator
(Subgroup.mem_subgroupOf.mp a.2) (Subgroup.mem_subgroupOf.mp b.2))
have hAcomm : ∀ s t : ↥(φU.range), s * t = t * s := by
rintro ⟨_, a, rfl⟩ ⟨_, b, rfl⟩
exact Subtype.ext (hComm a b)
have hirr : ∀ J : Subgroup (↥data.H ⧸ chief.N),
IsAInvariant (φU.range).subtype J → J = ⊥ ∨ J = ⊤ := by
intro J hJ
apply hcaseB
intro a
exact hJ ⟨φU a, MonoidHom.mem_range.mpr ⟨a, rfl⟩⟩
have hfaith : ∀ a : ↥(φU.range),
(∀ x : (↥data.H ⧸ chief.N), ((φU.range).subtype a) x = x) → a = 1 := by
intro a ha
have hone : (a : MulAut (↥data.H ⧸ chief.N)) = 1 := by
ext x
simpa using ha x
exact Subtype.ext hone
-- The fixed-point-free witness: a nonidentity `w₀ ∈ W₁` inside `↥(U ⊔ W₁)`.
obtain ⟨w, hwW1, hwne⟩ :=
(data.typeP.W1.bot_or_exists_ne_one).resolve_left data.typeP.W1_nontrivial
have hwUW1 : w ∈ data.typeP.U ⊔ data.typeP.W1 := Subgroup.mem_sup_right hwW1
set w₀ : ↥(data.typeP.U ⊔ data.typeP.W1) := ⟨w, hwUW1⟩ with hw₀def
have hw₀E : w₀ ∈ act.E := by
change (⟨w, hwUW1⟩ : ↥(data.typeP.U ⊔ data.typeP.W1)) ∈
data.typeP.W1.subgroupOf (data.typeP.U ⊔ data.typeP.W1)
rw [Subgroup.mem_subgroupOf]; exact hwW1
have hw₀ne : w₀ ≠ 1 := fun h => hwne (Subtype.ext_iff.mp h)
-- Conjugation action `ψ` of `U W₁` on the kernel `act.U`, and how `φU` transforms under it.
set ψ : ↥(data.typeP.U ⊔ data.typeP.W1) →* MulAut ↥act.U := MulAut.conjNormal with hψdef
have hψcoe : ∀ (g : ↥(data.typeP.U ⊔ data.typeP.W1)) (y : ↥act.U),
(act.U.subtype (ψ g y) : ↥(data.typeP.U ⊔ data.typeP.W1)) =
(g : ↥(data.typeP.U ⊔ data.typeP.W1)) * act.U.subtype y * g⁻¹ :=
fun g y => MulAut.conjNormal_apply g y
have hφUconj : ∀ (g : ↥(data.typeP.U ⊔ data.typeP.W1)) (y : ↥act.U),
φU (ψ g y) = act.φ g * φU y * (act.φ g)⁻¹ := by
intro g y
have he : φU (ψ g y) = act.φ ((g : ↥(data.typeP.U ⊔ data.typeP.W1)) * act.U.subtype y * g⁻¹) :=
by
change act.φ (act.U.subtype (ψ g y)) = _
rw [hψcoe]
rw [he, map_mul, map_mul, map_inv]; rfl
-- The kernel `N = C_U(H̄)` of `φU`, and its `ψ`-invariance.
set N : Subgroup ↥act.U := φU.ker with hNdef
have hN_inv : IsAInvariant (ψ.comp (Subgroup.zpowers w₀).subtype) N := by
rw [isAInvariant_iff_smul_mem]
intro a y hy
rw [hNdef, MonoidHom.mem_ker] at hy ⊢
change φU (ψ ((Subgroup.zpowers w₀).subtype a) y) = 1
rw [hφUconj, hy, mul_one, mul_inv_cancel]
-- Coprimality and solvability inputs for Isaacs Cor 3.28.
have hCop : Nat.Coprime (Nat.card ↥(Subgroup.zpowers w₀)) (Nat.card ↥act.U) := by
have hkc : Nat.Coprime (Nat.card ↥act.U) (Nat.card ↥act.E) :=
(typeP_uW1_frobenius data.typeP hU).coprime_card_kernel_complement
exact Nat.Coprime.coprime_dvd_left
(Subgroup.card_dvd_of_le (Subgroup.zpowers_le.mpr hw₀E)) hkc.symm
have hSolv : Group.IsSolvable ↥(Subgroup.zpowers w₀) ∨ Group.IsSolvable ↥act.U :=
Or.inl (Group.isSolvable_of_comm fun a b => by
obtain ⟨i, hi⟩ := Subgroup.mem_zpowers_iff.mp a.2
obtain ⟨j, hj⟩ := Subgroup.mem_zpowers_iff.mp b.2
exact Subtype.ext (by rw [Subgroup.coe_mul, Subgroup.coe_mul, ← hi, ← hj]; group))
-- The fixed-point-free hypothesis for `σ = act.φ w₀`, in subgroup terms.
have hfpfσ : ∀ a : ↥(φU.range),
(∀ x : ↥data.H ⧸ chief.N,
(act.φ w₀) (((φU.range).subtype a) x) = ((φU.range).subtype a) ((act.φ w₀) x)) →
a = 1 := by
intro a ha
obtain ⟨u₀, hu₀⟩ := a.2
-- `↑a = φU u₀ = act.φ u₀'`, and `act.φ w₀` commutes with it.
have ha1 : ((φU.range).subtype a : MulAut (↥data.H ⧸ chief.N)) = act.φ (act.U.subtype u₀) :=
hu₀.symm
have hCm : Commute (act.φ w₀) (act.φ (act.U.subtype u₀)) := by
change act.φ w₀ * act.φ (act.U.subtype u₀) = act.φ (act.U.subtype u₀) * act.φ w₀
apply MulEquiv.ext
intro x
rw [MulAut.mul_apply, MulAut.mul_apply, ← ha1]
exact ha x
-- `hg_fix`: each `w₀^k`-conjugate of `u₀` lands in `u₀ · N`.
have hg_fix : ∀ aa : ↥(Subgroup.zpowers w₀),
∃ n ∈ N, (ψ.comp (Subgroup.zpowers w₀).subtype) aa u₀ = u₀ * n := by
intro aa
refine ⟨u₀⁻¹ * (ψ ((Subgroup.zpowers w₀).subtype aa) u₀), ?_, ?_⟩
· rw [hNdef, MonoidHom.mem_ker, map_mul, map_inv, hφUconj]
obtain ⟨k, hk⟩ := Subgroup.mem_zpowers_iff.mp aa.2
have hcaφ : act.φ ((Subgroup.zpowers w₀).subtype aa) = (act.φ w₀) ^ k := by
rw [show (Subgroup.zpowers w₀).subtype aa = w₀ ^ k from hk.symm, map_zpow]
have hcomm_k : Commute (act.φ ((Subgroup.zpowers w₀).subtype aa)) (φU u₀) := by
rw [hcaφ]; exact hCm.zpow_left k
rw [show act.φ ((Subgroup.zpowers w₀).subtype aa) * φU u₀ *
(act.φ ((Subgroup.zpowers w₀).subtype aa))⁻¹ = φU u₀ by
rw [hcomm_k.eq, mul_assoc, mul_inv_cancel, mul_one], inv_mul_cancel]
· rw [MonoidHom.comp_apply, mul_inv_cancel_left]
-- Isaacs Cor 3.28: a `w₀`-fixed representative `c ∈ u₀ · N`.
obtain ⟨c, hc_fix, n, hn, hc_eq⟩ :=
OddOrder.Isaacs.Ch04.coprime_fixedPoints_quotient
(φ := ψ.comp (Subgroup.zpowers w₀).subtype) hCop hSolv hN_inv hg_fix
-- `c` commutes with `w₀`, so `c ∈ C_U(w₀) ⊆ W₁`; as `c ∈ U`, `c = 1`.
have hc_w₀ : ψ w₀ c = c := by
have := hc_fix ⟨w₀, Subgroup.mem_zpowers w₀⟩
rwa [MonoidHom.comp_apply] at this
have hc_comm : (act.U.subtype c : ↥(data.typeP.U ⊔ data.typeP.W1)) * w₀ =
w₀ * act.U.subtype c := by
have h := congrArg act.U.subtype hc_w₀
rw [hψcoe] at h
-- `h : w₀ * subtype c * w₀⁻¹ = subtype c`
rw [mul_inv_eq_iff_eq_mul] at h
exact h.symm
have hc_in_E : (act.U.subtype c : ↥(data.typeP.U ⊔ data.typeP.W1)) ∈ act.E := by
refine (typeP_uW1_frobenius data.typeP hU).centralizer_complement_le w₀ hw₀E hw₀ne ?_
rw [Subgroup.mem_centralizer_singleton_iff]
exact hc_comm
have hc_in_U : (act.U.subtype c : ↥(data.typeP.U ⊔ data.typeP.W1)) ∈ act.U := c.2
have hc1 : (act.U.subtype c : ↥(data.typeP.U ⊔ data.typeP.W1)) = 1 :=
Subgroup.disjoint_def.mp (typeP_uW1_frobenius data.typeP hU).isComplement.disjoint
hc_in_U hc_in_E
have hc0 : c = 1 := act.U.subtype_injective (by rw [hc1, map_one])
-- `1 = c = u₀ · n` ⟹ `u₀ ∈ N` ⟹ `φU u₀ = 1` ⟹ `a = 1`.
rw [hc0] at hc_eq
have hu₀N : u₀ ∈ N := by
have : u₀ = n⁻¹ := by rw [eq_inv_iff_mul_eq_one, ← hc_eq]
rw [this]; exact N.inv_mem hn
apply Subtype.ext
rw [← hu₀, Subgroup.coe_one]
exact (MonoidHom.mem_ker (f := φU)).mp hu₀N
-- Thin adapter into the canonical module-level Singer FPF-coprimality (shared leaf, issue 9000).
exact singerAdapter_coprime_fpf
(A := ↥(φU.range)) (K := ↥data.H ⧸ chief.N) (p := chief.p) (φ := (φU.range).subtype)
hAcomm hKnt hirr hfaith (act.φ w₀) hfpfσ
open OddOrder.RepresentationTheory Representation in
open scoped commutatorElement IsMulCommutative in
/-- **Peterfalvi (9.7) case (b), the `u ∣ (p^q-1)/(p-1)` divisibility** (unconditional).
The Singer divisibility `|Ū| ∣ p^q-1` (`chiefFactor_caseB_image_cyclic`) upgrades to
`|Ū| ∣ (p^q-1)/(p-1)` via the fixed-point-free coprimality `Coprime |Ū| (p-1)`
(`chiefFactor_caseB_image_coprime`): since `(p-1) ∣ (p^q-1)` and `|Ū|` is coprime to `p-1`,
`|Ū|·(p-1) ∣ p^q-1`, hence `|Ū| ∣ (p^q-1)/(p-1)`. This is the second case-(b) divisibility of
`CliffordCaseBData`, now established without any external coprimality hypothesis. -/
theorem chiefFactor_caseB_image_dvd_norm [Finite G] {M : Subgroup G}
{data : TypesIIIIIIVSetup M} (chief : ChiefFactorData data)
(hcaseB : ∀ J : Subgroup (↥data.H ⧸ chief.N),
IsAInvariant ((typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).φ.comp (typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).U.subtype) J → J = ⊥ ∨ J = ⊤) :
Nat.card ↥(((typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).φ.comp (typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).U.subtype).range) ∣ (chief.p ^ data.q - 1) / (chief.p - 1) := by
have hcop := chiefFactor_caseB_image_coprime chief hcaseB
obtain ⟨_, hdvd⟩ := chiefFactor_caseB_image_cyclic chief hcaseB
-- The arithmetic core is the shared `SingerLineBound` lemma (issue 9000 dedup).
exact dvd_div_of_coprime_of_dvd_sub_one chief.p_prime.pos hdvd hcop
open scoped commutatorElement in
/-- **Peterfalvi (9.7) case (b): the `U`-action on `H̄` is fixed-point-free off `C = C_U(H̄)`.**
When `U` acts irreducibly on the chief factor `H̄ = H/H₀` (Clifford case (b)), any `g ∈ U` whose
image `φ_U(g)` is nontrivial (i.e. `g ∉ C`) acts fixed-point-freely on `H̄`: `φ_U(g)·x = x → x = 1`.
This is the structural heart of Peterfalvi (9.9) — `H̄ ⋊ Ū` is a Frobenius group, so a nontrivial
character `θ` of `H̄` has inertia `I(θ) ∩ U = C`, giving the degree `u = |U:C|` of the irreducible
characters of `HU` in `𝒳(H₀C')`. The proof is pure Clifford theory: the image `Ū = φ_U(U)` is
abelian (`hComm`: `⁅a,b⁆ ∈ [U,U] ⊆ C_M(H)`) and acts irreducibly (`hcaseB`), so
`fixedPointFree_of_aInvariant_irreducible_comm` applies; the Singer field model is *not* needed. -/
theorem chiefFactor_caseB_action_fpf [Finite G] {M : Subgroup G}
{data : TypesIIIIIIVSetup M} (chief : ChiefFactorData data)
(hcaseB : ∀ J : Subgroup (↥data.H ⧸ chief.N),
IsAInvariant ((typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).φ.comp (typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).U.subtype) J → J = ⊥ ∨ J = ⊤)
(g : ↥(typeP_quotientCoprimeAction data.typeP data.nontrivial.1 chief.N_aInvariant).U)
(hg : ((typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).φ.comp (typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).U.subtype) g ≠ 1)
(x : ↥data.H ⧸ chief.N)
(hx : ((typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).φ.comp (typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).U.subtype) g x = x) : x = 1 := by
classical
have : chief.N.Normal := chief.N_normal
set act := typeP_quotientCoprimeAction data.typeP data.nontrivial.1 chief.N_aInvariant with hact
set φU := act.φ.comp act.U.subtype with hφU
-- An element of `U W₁` centralizing `H` acts trivially on `H̄`.
have hcentral_triv : ∀ c : ↥(data.typeP.U ⊔ data.typeP.W1),
(c : G) ∈ Subgroup.centralizer (data.typeP.H : Set G) → act.φ c = 1 := by
intro c hc
have hfix : ∀ z : ↥data.typeP.H, (typeP_conjAction data.typeP c) z = z := by
intro z
apply Subtype.ext
rw [typeP_conjAction_apply]
have hcom : (z : G) * (c : G) = (c : G) * (z : G) :=
(Subgroup.mem_centralizer_iff.mp hc) (z : G) z.2
rw [← hcom, mul_assoc, mul_inv_cancel, mul_one]
ext y
refine QuotientGroup.induction_on y ?_
intro z
change (act.φ c) (QuotientGroup.mk' chief.N z) = QuotientGroup.mk' chief.N z
have hstep : (act.φ c) (QuotientGroup.mk' chief.N z)
= QuotientGroup.mk' chief.N ((typeP_conjAction data.typeP c) z) := rfl
rw [hstep, hfix z]
-- `Ū = φU(U)` is abelian: `φU ⁅a, b⁆ = 1` since `⁅a, b⁆` maps into `[U, U] ⊆ C(H)`.
have hComm : ∀ a b : ↥act.U, Commute (φU a) (φU b) := by
intro a b
refine commutatorElement_eq_one_iff_commute.mp ?_
rw [← map_commutatorElement φU a b]
change act.φ (act.U.subtype ⁅a, b⁆) = 1
apply hcentral_triv
change ((data.typeP.U ⊔ data.typeP.W1).subtype.comp act.U.subtype) ⁅a, b⁆
∈ Subgroup.centralizer (data.typeP.H : Set G)
rw [map_commutatorElement]
exact typeP_commutator_U_centralizes_H data.typeP
(Subgroup.commutator_mem_commutator
(Subgroup.mem_subgroupOf.mp a.2) (Subgroup.mem_subgroupOf.mp b.2))
exact fixedPointFree_of_aInvariant_irreducible_comm hComm hcaseB _ hg x hx
/-- **Peterfalvi (9.9): the `U`-action on `Irr(H̄)` is fixed-point-free off `C`.** In Clifford
case (b), if a nontrivial irreducible character `θ` of the chief factor `H̄ = ↥H ⧸ N` is invariant
under the `U`-action `φ_U(g)`, then `g` acts trivially (`φ_U(g) = 1`, i.e. `g ∈ C = C_U(H̄)`).
This is the **character-side inertia** `I_U(θ) ⊆ C` of Peterfalvi (9.9.a), proven
*realization-free*:
the abelian `Irr ↔ Hom` bridge
(`exists_units_monoidHom_of_isIrreducibleCharacter_of_isMulCommutative`)
turns `θ` into a linear character `θ̂`, and a `φ_U(g)`-invariant `θ̂` with `φ_U(g)` fixed-point-free
(`chiefFactor_caseB_action_fpf`, valid for `g ∉ C`) is trivial
(`eq_one_of_invariant_of_fixedPointFree`), contradicting `θ` nontrivial. No realization of `H̄` as
a subgroup is needed; it works on the abstract quotient `↥H ⧸ N` with the abstract action `φ_U`. -/
theorem chiefFactor_caseB_char_inertia [Finite G] {M : Subgroup G}
{data : TypesIIIIIIVSetup M} {chief : ChiefFactorData data}
(hcaseB : ∀ J : Subgroup (↥data.H ⧸ chief.N),
IsAInvariant ((typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).φ.comp (typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).U.subtype) J → J = ⊥ ∨ J = ⊤)
{θ : IrreducibleCharacter (↥data.H ⧸ chief.N)}
(hθ : (θ : ClassFunction (↥data.H ⧸ chief.N) ℂ) ≠ trivialClassFunction _)
(g : ↥(typeP_quotientCoprimeAction data.typeP data.nontrivial.1 chief.N_aInvariant).U)
(hinv : ∀ x, (θ : ClassFunction (↥data.H ⧸ chief.N) ℂ)
(((typeP_quotientCoprimeAction data.typeP data.nontrivial.1 chief.N_aInvariant).φ.comp
(typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).U.subtype) g x)
= (θ : ClassFunction (↥data.H ⧸ chief.N) ℂ) x) :
((typeP_quotientCoprimeAction data.typeP data.nontrivial.1 chief.N_aInvariant).φ.comp
(typeP_quotientCoprimeAction data.typeP data.nontrivial.1 chief.N_aInvariant).U.subtype) g
= 1 := by
classical
have : IsMulCommutative (↥data.H ⧸ chief.N) :=
IsMulCommutative.of_comm chief.quotient_elementaryAbelian.comm
by_contra hne
have hfpf : MonoidHom.FixedPointFree
(((typeP_quotientCoprimeAction data.typeP data.nontrivial.1 chief.N_aInvariant).φ.comp
(typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).U.subtype) g) :=
chiefFactor_caseB_action_fpf chief hcaseB g hne
obtain ⟨θhom, hθhom⟩ :=
exists_units_monoidHom_of_isIrreducibleCharacter_of_isMulCommutative θ.isIrreducible
have hinvhom : ∀ x, θhom (((typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).φ.comp (typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).U.subtype) g x) = θhom x := fun x =>
Units.ext (by rw [hθhom, hθhom]; exact hinv x)
have h1 : θhom = 1 := eq_one_of_invariant_of_fixedPointFree hfpf hinvhom
apply hθ
ext x
have hx := hθhom x
rw [h1] at hx
simpa using hx.symm
/-- **Non-Galois (9.8) core, generic over the factor family.** As `chiefFactor_caseA_char_inertia`
but taking the order-`p`, `U`-invariant, spanning factor family `Hpart` directly (rather than from
`CliffordCaseAData`), so the `W1`-conjugates `{S₀^w}` — which have the same properties but are not
the producer's `caseA.Hpart` family — can drive the inertia argument for the free-`W1`-orbit
character of (9.8.c). -/
theorem chiefFactor_caseA_char_inertia_gen [Finite G] {M : Subgroup G}
{data : TypesIIIIIIVSetup M} {chief : ChiefFactorData data}
{ι : Type*} (Hpart : ι → Subgroup (↥data.H ⧸ chief.N))
(hp_order : ∀ i, Nat.card ↥(Hpart i) = chief.p)
(hspan : ⨆ i, Hpart i = ⊤)
(haInv : ∀ i, IsAInvariant (uActionHom data chief) (Hpart i))
{θ : IrreducibleCharacter (↥data.H ⧸ chief.N)}
(hreg : ∀ i, ∃ x ∈ Hpart i,
(θ : ClassFunction (↥data.H ⧸ chief.N) ℂ) x
≠ (θ : ClassFunction (↥data.H ⧸ chief.N) ℂ) 1)
(g : ↥(typeP_quotientCoprimeAction data.typeP data.nontrivial.1 chief.N_aInvariant).U)
(hinv : ∀ x, (θ : ClassFunction (↥data.H ⧸ chief.N) ℂ) ((uActionHom data chief) g x)
= (θ : ClassFunction (↥data.H ⧸ chief.N) ℂ) x) :
(uActionHom data chief) g = 1 := by
have : IsMulCommutative (↥data.H ⧸ chief.N) :=
IsMulCommutative.of_comm chief.quotient_elementaryAbelian.comm
refine mulAut_eq_one_of_fixes_regular_on_prime_span ((uActionHom data chief) g) Hpart
(fun i => ?_) (fun i x hx => ?_) hspan θ hreg hinv
· rw [hp_order i]; exact chief.p_prime
· exact (haInv i).smul_mem g hx
set_option backward.isDefEq.respectTransparency false in
/-- **Peterfalvi (9.8.d) single-factor char-inertia core**: the single-factor analog of
`chiefFactor_caseA_char_inertia`. For a character `θ₁` nontrivial on the order-`p` orbit generator
`S₀ = H₁` (`hreg` on `caseA.S0`), a `U`-element `g` fixing `θ₁` acts *trivially on `S₀`*, i.e.
`aInvariantRestrictAut caseA.S0_aInvariant g = 1`. This — not `uActionHom g = 1` — is the correct
conclusion for (9.8.d): `θ₁` need only be faithful on the *single* summand `S₀`, so the fixing
element centralizes `S₀` (lands in `C_U(S₀)`), not necessarily all of `H̄`. The pure-algebra heart
is `mulAut_eq_id_on_of_fixes_ne_one_on_prime` (`θ₁` faithful on the prime-order `S₀`), lifted
through
`aInvariantRestrictAut_coe` (which identifies the restricted action with `uActionHom` on `S₀`).
This
gives `I(θ₁) ∩ U ⊆ C_U(S₀)` in the degree-`qa` inertia lift. -/
theorem chiefFactor_caseA_char_inertia_single [Finite G] {M : Subgroup G}
{data : TypesIIIIIIVSetup M} {chief : ChiefFactorData data}
{chars : Section11CharacterData data chief} (caseA : CliffordCaseAData chars)
{θ : IrreducibleCharacter (↥data.H ⧸ chief.N)}
(hreg : ∃ x ∈ caseA.S0,
(θ : ClassFunction (↥data.H ⧸ chief.N) ℂ) x
≠ (θ : ClassFunction (↥data.H ⧸ chief.N) ℂ) 1)
(g : ↥(typeP_quotientCoprimeAction data.typeP data.nontrivial.1 chief.N_aInvariant).U)
(hinv : ∀ x, (θ : ClassFunction (↥data.H ⧸ chief.N) ℂ) ((uActionHom data chief) g x)
= (θ : ClassFunction (↥data.H ⧸ chief.N) ℂ) x) :
aInvariantRestrictAut caseA.S0_aInvariant g = 1 := by
have : IsMulCommutative (↥data.H ⧸ chief.N) :=
IsMulCommutative.of_comm chief.quotient_elementaryAbelian.comm
-- `φg = uActionHom g` acts as the identity on the prime-order summand `S₀` (single-factor core).
have hS0card : Nat.card ↥caseA.S0 = chief.p := by
have h := caseA.Hpart_order ⟨0, data.nontrivial.2.1.pos⟩
rwa [caseA.Hpart_orbit ⟨0, data.nontrivial.2.1.pos⟩, card_pointwise_smul] at h
have hid : ∀ x ∈ caseA.S0, (uActionHom data chief) g x = x :=
mulAut_eq_id_on_of_fixes_ne_one_on_prime ((uActionHom data chief) g) caseA.S0
(by rw [hS0card]; exact chief.p_prime)
(fun x hx => caseA.S0_aInvariant.smul_mem g hx) θ hreg hinv
-- Lift to `aInvariantRestrictAut … g = 1` via the coercion
-- `(restrict g x : H̄) = uActionHom g x`.
ext x
rw [MulAut.one_apply, aInvariantRestrictAut_coe, hid x x.2]
/-- **Inflation equivariance for the chief-factor action.** The inflation map
`compHom (mk' N) : ClassFunction (↥H/N) → ClassFunction ↥H` intertwines the conjugation action
`typeP_conjAction a` on `↥H` (upstairs) with the descended action `quotientMulAutHom a` on the
chief factor `↥H/N` (downstairs): inflating `θ̄` and acting by `a` upstairs equals acting by `a`
downstairs and then inflating. Immediate from `quotientMulAutHom_apply_mk'`
(`mk' N (a · h) = a · (mk' N h)`).
This is the algebraic core that turns the *concrete* conjugation invariance of an inflated
character into the *abstract* `φ_U`-invariance consumed by `chiefFactor_caseB_char_inertia`,
without realizing `H̄` as a subgroup. -/
theorem compHom_typeP_conjAction_inflation [Finite G] {M : Subgroup G}
{data : TypesIIIIIIVSetup M} {chief : ChiefFactorData data}
(a : ↥(data.typeP.U ⊔ data.typeP.W1))
(θbar : ClassFunction (↥data.H ⧸ chief.N) ℂ) :
ClassFunction.compHom (typeP_conjAction data.typeP a).toMonoidHom
(ClassFunction.compHom (QuotientGroup.mk' chief.N) θbar)
= ClassFunction.compHom (QuotientGroup.mk' chief.N)
(ClassFunction.compHom (quotientMulAutHom chief.N_aInvariant a).toMonoidHom θbar) :=
rfl
/-- **Char-inertia inflation, parametrized over the core stabilizer-triviality `hcore`.**
Strips the inflation (`compHom_typeP_conjAction_inflation`, `mk'` injective) from the
`compHom`-fixing
hypothesis to feed the per-element action-invariance to `hcore`. Case (b) supplies `hcore` as
`chiefFactor_caseB_char_inertia` (`U`-irreducible); case (a) via the non-Galois `Hpart` analysis. -/
theorem caseB_char_inertia_inflation_of_core [Finite G] {M : Subgroup G}
{data : TypesIIIIIIVSetup M} {chief : ChiefFactorData data}
{θbar : IrreducibleCharacter (↥data.H ⧸ chief.N)}
(hcore : ∀ (g : ↥(typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).U),
(∀ x, (θbar : ClassFunction (↥data.H ⧸ chief.N) ℂ)
(((typeP_quotientCoprimeAction data.typeP data.nontrivial.1 chief.N_aInvariant).φ.comp
(typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).U.subtype) g x)
= (θbar : ClassFunction (↥data.H ⧸ chief.N) ℂ) x) →
((typeP_quotientCoprimeAction data.typeP data.nontrivial.1 chief.N_aInvariant).φ.comp
(typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).U.subtype) g = 1)
(g : ↥(typeP_quotientCoprimeAction data.typeP data.nontrivial.1 chief.N_aInvariant).U)
(hfix : ClassFunction.compHom (typeP_conjAction data.typeP
((typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).U.subtype g)).toMonoidHom
(ClassFunction.compHom (QuotientGroup.mk' chief.N)
(θbar : ClassFunction (↥data.H ⧸ chief.N) ℂ))
= ClassFunction.compHom (QuotientGroup.mk' chief.N)
(θbar : ClassFunction (↥data.H ⧸ chief.N) ℂ)) :
((typeP_quotientCoprimeAction data.typeP data.nontrivial.1 chief.N_aInvariant).φ.comp
(typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).U.subtype) g = 1 := by
rw [compHom_typeP_conjAction_inflation] at hfix
have hfix2 := ClassFunction.compHom_injective_of_surjective
(QuotientGroup.mk'_surjective chief.N) hfix
apply hcore g
intro x
rw [show ((typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).φ.comp (typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).U.subtype) g
= quotientMulAutHom chief.N_aInvariant
((typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).U.subtype g) from rfl]
exact congrFun (congrArg (fun f : ClassFunction (↥data.H ⧸ chief.N) ℂ =>
(f : (↥data.H ⧸ chief.N) → ℂ)) hfix2) x
theorem caseB_char_inertia_inflation [Finite G] {M : Subgroup G}
{data : TypesIIIIIIVSetup M} {chief : ChiefFactorData data}
(hcaseB : ∀ J : Subgroup (↥data.H ⧸ chief.N),
IsAInvariant ((typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).φ.comp (typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).U.subtype) J → J = ⊥ ∨ J = ⊤)
{θbar : IrreducibleCharacter (↥data.H ⧸ chief.N)}
(hθbar : (θbar : ClassFunction (↥data.H ⧸ chief.N) ℂ) ≠ trivialClassFunction _)
(g : ↥(typeP_quotientCoprimeAction data.typeP data.nontrivial.1 chief.N_aInvariant).U)
(hfix : ClassFunction.compHom (typeP_conjAction data.typeP
((typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).U.subtype g)).toMonoidHom
(ClassFunction.compHom (QuotientGroup.mk' chief.N)
(θbar : ClassFunction (↥data.H ⧸ chief.N) ℂ))
= ClassFunction.compHom (QuotientGroup.mk' chief.N)
(θbar : ClassFunction (↥data.H ⧸ chief.N) ℂ)) :
((typeP_quotientCoprimeAction data.typeP data.nontrivial.1 chief.N_aInvariant).φ.comp
(typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).U.subtype) g = 1 :=
caseB_char_inertia_inflation_of_core (chiefFactor_caseB_char_inertia hcaseB hθbar) g hfix
/-- The realization iso `↥(H-in-HU) ≃* ↥H`: `H` realized inside `HU = H ⊔ U` is `H`, via the
composite of `subgroupOfEquivOfLe (H.subgroupOf M ≤ HU)` and `subgroupOfEquivOfLe (H ≤ M)`. -/
noncomputable def hInHuEquivH {M : Subgroup G} (data : TypesIIIIIIVSetup M) :
↥(hInHu data) ≃* ↥data.H :=
(Subgroup.subgroupOfEquivOfLe
(Subgroup.subgroupOf_mono M (le_sup_left : data.H ≤ data.H ⊔ data.U))).trans
(Subgroup.subgroupOfEquivOfLe (data.typeP.H_le.trans (derivedInG_le_self M)))
/-- The realization iso preserves the underlying `G`-element. -/
theorem hInHuEquivH_coe {M : Subgroup G} (data : TypesIIIIIIVSetup M) (h : ↥(hInHu data)) :
((hInHuEquivH data h : ↥data.H) : G) = (((h : ↥(huSub data)) : ↥M) : G) := rfl
/-- **(9.9.a) realization, conjugation equivariance.** Under the iso
`hInHuEquivH : ↥(hInHu) ≃* ↥H`, the concrete conjugation `conjBy g` in `HU` (for `g ∈ HU`)
corresponds to the abstract conjugation `typeP_conjAction a` on `↥H` (for `a ∈ U W₁` with the same
`G`-image `↑g = ↑a`). Both are conjugation by the same `G`-element, so the equality reduces to
`g·h·g⁻¹ = a·h·a⁻¹` in `G`. -/
theorem conjBy_compHom_hInHuEquivH {M : Subgroup G} (data : TypesIIIIIIVSetup M)
(a : ↥(data.typeP.U ⊔ data.typeP.W1)) (g : ↥(huSub data))
(hag : ((g : ↥M) : G) = (a : G)) (θ : ClassFunction ↥data.H ℂ) :
ClassFunction.conjBy g (ClassFunction.compHom (hInHuEquivH data).toMonoidHom θ)
= ClassFunction.compHom (hInHuEquivH data).toMonoidHom
(ClassFunction.compHom (typeP_conjAction data.typeP a).toMonoidHom θ) := by
ext h
rw [ClassFunction.conjBy_apply, ClassFunction.compHom_apply, ClassFunction.compHom_apply,
ClassFunction.compHom_apply]
refine congrArg _ (Subtype.ext ?_)
simp only [MulEquiv.coe_toMonoidHom, hInHuEquivH_coe, typeP_conjAction_apply,
Subgroup.coe_mul, Subgroup.coe_inv]
rw [hag]
/-- **Realized stabilizer-triviality, parametrized over the char-inertia core `hcharInertia`.**
The case-agnostic transport: a `U`-element `a` realized by `g ∈ HU` fixing `θ₀` (`conjBy` form) is
turned into the `compHom (typeP_conjAction)` form (`conjBy_compHom_hInHuEquivH`, injective
inflation)
and fed to `hcharInertia` to conclude `φ a = 1`. Case (b) supplies `hcharInertia` as
`caseB_char_inertia_inflation`; case (a) via the non-Galois analog. -/
theorem caseB_inertia_realized_of_charInertia [Finite G] {M : Subgroup G}
{data : TypesIIIIIIVSetup M} {chief : ChiefFactorData data}
{θbar : IrreducibleCharacter (↥data.H ⧸ chief.N)}
(hcharInertia : ∀ (g : ↥(typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).U),
ClassFunction.compHom (typeP_conjAction data.typeP
((typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).U.subtype g)).toMonoidHom
(ClassFunction.compHom (QuotientGroup.mk' chief.N)
(θbar : ClassFunction (↥data.H ⧸ chief.N) ℂ))
= ClassFunction.compHom (QuotientGroup.mk' chief.N)
(θbar : ClassFunction (↥data.H ⧸ chief.N) ℂ) →
((typeP_quotientCoprimeAction data.typeP data.nontrivial.1 chief.N_aInvariant).φ.comp
(typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).U.subtype) g = 1)
(a : ↥(typeP_quotientCoprimeAction data.typeP data.nontrivial.1 chief.N_aInvariant).U)
(g : ↥(huSub data))
(hag : ((g : ↥M) : G) =
(((typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).U.subtype a : ↥(data.typeP.U ⊔ data.typeP.W1)) : G))
(hfix : ClassFunction.conjBy g
(ClassFunction.compHom (hInHuEquivH data).toMonoidHom
(ClassFunction.compHom (QuotientGroup.mk' chief.N)
(θbar : ClassFunction (↥data.H ⧸ chief.N) ℂ)))
= ClassFunction.compHom (hInHuEquivH data).toMonoidHom
(ClassFunction.compHom (QuotientGroup.mk' chief.N)
(θbar : ClassFunction (↥data.H ⧸ chief.N) ℂ))) :
((typeP_quotientCoprimeAction data.typeP data.nontrivial.1 chief.N_aInvariant).φ.comp
(typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).U.subtype) a = 1 := by
rw [conjBy_compHom_hInHuEquivH data
((typeP_quotientCoprimeAction data.typeP data.nontrivial.1 chief.N_aInvariant).U.subtype a)
g hag] at hfix
exact hcharInertia a
(ClassFunction.compHom_injective_of_surjective (hInHuEquivH data).surjective hfix)
theorem caseB_inertia_realized [Finite G] {M : Subgroup G}
{data : TypesIIIIIIVSetup M} {chief : ChiefFactorData data}
(hcaseB : ∀ J : Subgroup (↥data.H ⧸ chief.N),
IsAInvariant ((typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).φ.comp (typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).U.subtype) J → J = ⊥ ∨ J = ⊤)
{θbar : IrreducibleCharacter (↥data.H ⧸ chief.N)}
(hθbar : (θbar : ClassFunction (↥data.H ⧸ chief.N) ℂ) ≠ trivialClassFunction _)
(a : ↥(typeP_quotientCoprimeAction data.typeP data.nontrivial.1 chief.N_aInvariant).U)
(g : ↥(huSub data))
(hag : ((g : ↥M) : G) =
(((typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).U.subtype a : ↥(data.typeP.U ⊔ data.typeP.W1)) : G))
(hfix : ClassFunction.conjBy g
(ClassFunction.compHom (hInHuEquivH data).toMonoidHom
(ClassFunction.compHom (QuotientGroup.mk' chief.N)
(θbar : ClassFunction (↥data.H ⧸ chief.N) ℂ)))
= ClassFunction.compHom (hInHuEquivH data).toMonoidHom
(ClassFunction.compHom (QuotientGroup.mk' chief.N)
(θbar : ClassFunction (↥data.H ⧸ chief.N) ℂ))) :
((typeP_quotientCoprimeAction data.typeP data.nontrivial.1 chief.N_aInvariant).φ.comp
(typeP_quotientCoprimeAction data.typeP data.nontrivial.1
chief.N_aInvariant).U.subtype) a = 1 :=
caseB_inertia_realized_of_charInertia (caseB_char_inertia_inflation hcaseB hθbar) a g hag hfix
end OddOrder.Peterfalvi.S11
Source
Yawara Ishida, Lean 4 formalization of the Feit–Thompson (odd order) theorem, github.com/yawara/odd-order @ 20a6fccbb63c (Apache-2.0), https://github.com/yawara/odd-order/blob/20a6fccbb63c0b138acda9f7fafd798c42f9441b/OddOrder/Peterfalvi/S11_MaximalII_III_IV/InertiaLift.lean