BG §16: Theorems A–E (continuation)
DefinitionOddOrder_BG_Ch4_FamilyOfMaximal_S16_MainResults_TheoremsAEfeit-thompsonfinite-groupsgroup-theoryishida-odd-order
A continuation of the Theorems A–E development of §16 of Bender–Glauberman, Local Analysis for the Odd Order Theorem. It consists of lemmas only, including theoremC_paired_structure, msigma_fusion_control, matched -Hall pairs for type and conjugate-of- statements.
The bundle consists of lemmas only.
The bundle also proves 12 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/BG/Ch4_FamilyOfMaximal/S16_MainResults/TheoremsAE.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_BG_Ch4_FamilyOfMaximal_S16_MainResults_Notation
import Definitions.Def_OddOrder_BG_Ch4_FamilyOfMaximal_S16_MainResults_TheoremAB
namespace OddOrder.BG.Ch4.S16
open OddOrder.GroupTheory
open OddOrder.Isaacs
open OddOrder.BG.Ch3.S12
open OddOrder.BG.Ch4.S14
open OddOrder.BG.Ch4.S15
open scoped Pointwise
variable {G : Type*} [Group G]
/-- **BG Theorem C(9), structural inclusion** (mmd L4385; schematic proof: Proposition 14.2(d) +
Theorem A(3),(5)): every element of `A_0(M) − A(M)` is `M`-conjugate to an element of
`Ẑ = Z − (K ∪ K*)`. This is the `⊆` half of BG's identity `A_0(M) − A(M) = 𝒞_M(Ẑ)`, and together
with the TI-ness of `Ẑ` (Theorem 14.7(e), `N_G(Ẑ) = K ⊔ K* ≤ M`) it is exactly what the TI claim of
Theorem C(9) needs, via the transport lemma `IsTISubset.of_subset_conj_of_isTISubset`.
Proof: let `a ∈ A_0(M) − A(M)`, so `a ∈ M`, `M_σ ⊓ C_G(a) ≠ 1`, `a ∉ 𝒞_G(K#)`, and `a ∉ U M_σ`.
Take the `κ`/`κ'`-decomposition `a = a_κ · a_{κ'}` (`exists_isPiElement_mul`: commuting powers of
`a`, `a_κ` a `κ`-element, `a_{κ'}` a `κ'`-element). Conjugate by some `w ∈ M` so that `a_κ ∈ K`
(`exists_conj_smul_le_isHall_kappa`, as `⟨a_κ⟩` is a `κ`-subgroup of `M`). Then:
* `a_κ ≠ 1`: otherwise `a = a_{κ'}` is a `κ'`-element, hence lies in the normal Hall `κ'`-subgroup
`M' = U M_σ` (Theorem C(3)), contradicting `a ∉ U M_σ`.
* `a_{κ'} ≠ 1`: otherwise `a` is `G`-conjugate to `a_κ ∈ K#`, contradicting `a ∉ 𝒞_G(K#)`.
* `a_{κ'} ∈ K*`: `a_{κ'} ∈ M' = U ⋊ M_σ` commutes with `a_κ ∈ K#`; writing `a_{κ'} = u·t`
(`u ∈ U`, `t ∈ M_σ`), `K`-invariance of `U` (Proposition 14.2(a)) and of `M_σ` plus uniqueness of
the `U ⋊ M_σ` factorization force `u ∈ C_U(a_κ) = 1` (Theorem A(4)) and `t ∈ C_{M_σ}(a_κ) = K*`
(Theorem A(5)), so `a_{κ'} = t ∈ K*`.
Hence (after the conjugation) `a = a_κ · a_{κ'} ∈ K·K*` with both factors nontrivial, i.e.
`a ∈ Z − (K ∪ K*) = Ẑ`. -/
theorem a0_minus_a_subset_conj_zTilde [Finite G] (hG : OddOrder.BG.IsMinimalSimpleOdd G)
{M K Kstar U : Subgroup G} (hM : M ∈ maximalSubgroups G) (hP : S14.IsTypeP M)
(hKM : K ≤ M) (hUM : U ≤ M)
(hK : Ch03.IsHallSubgroup (S14.kappa M) (K.subgroupOf M))
(hKstar : Kstar = OddOrder.BG.Ch3.S10.Msigma M ⊓ Subgroup.centralizer (K : Set G))
(hU : Ch03.IsHallSubgroup ((S14.kappa M ∪ OddOrder.BG.Ch3.S10.sigma M)ᶜ) (U.subgroupOf M)) :
∀ a ∈ A0Set M K \ ASet M U,
∃ m ∈ M, ∃ t ∈ S14.zTilde K Kstar, a = m * t * m⁻¹ := by
classical
have hsolv : Group.IsSolvable ↥M := hG.isSolvable_of_mem_maximalSubgroups hM
have hKne : K ≠ ⊥ := fun h => card_kappaHall_ne_one hP hKM hK (by rw [h, Subgroup.card_bot])
have hKcyc : IsCyclic ↥K := (typeP_auxiliary_structure hG hM hKM hUM hK hKstar hU).2.1
have hMσM : OddOrder.BG.Ch3.S10.Msigma M ≤ M := OddOrder.BG.Ch3.S10.Msigma_le M
have hM'M : (U ⊔ OddOrder.BG.Ch3.S10.Msigma M : Subgroup G) ≤ M := sup_le hUM hMσM
have hKstarMσ : Kstar ≤ OddOrder.BG.Ch3.S10.Msigma M := by rw [hKstar]; exact inf_le_left
-- `M' = U ⊔ M_σ` (Theorem C(3) / Lemma 15.1(b)).
have hM'eq : derivedInG M = U ⊔ OddOrder.BG.Ch3.S10.Msigma M :=
(typeP_hall_derived_eq_and_abelian hG hM hKM hUM hKne hK hU).1
-- **Helper A**: every `κ(M)'`-element of `M` lies in the normal Hall `κ'`-subgroup
-- `M' = U ⊔ M_σ`.
-- (`M'` complements the Hall `κ`-subgroup `K` in `M`, so `[M : M'] = |K|` is a `κ`-number;
-- a `κ'`-element has image of order dividing both `[M : M']` and its own order, hence trivial.)
have hkappaComplMem : ∀ x : G, x ∈ M → IsPiElement (S14.kappa M)ᶜ x →
x ∈ (U ⊔ OddOrder.BG.Ch3.S10.Msigma M : Subgroup G) := by
intro x hxM hxπ'
rw [← hM'eq]
have hNnorm : ((derivedInG M).subgroupOf M).Normal :=
Subgroup.normal_subgroupOf_of_le_normalizer (OddOrder.BG.Ch3.S10.le_normalizer_derivedInG M)
set N := (derivedInG M).subgroupOf M with hNdef
set x' : ↥M := ⟨x, hxM⟩ with hx'def
-- `N.index = |K|` is a `κ`-number (`M'` complements the Hall `κ`-subgroup `K`).
have hidxκ : ∀ p ∈ N.index.primeFactors, p ∈ S14.kappa M := by
have hcompl := (typeP_duality hG hM hP hKM hK hKstar).1
have hidxeq : N.index = Nat.card ↥(K.subgroupOf M) := hcompl.symm.index_eq_card
rw [hidxeq]; exact hK.1
-- `orderOf x'` is a `κ'`-number, hence coprime to `N.index`.
have hordx' : orderOf x' = orderOf x :=
(orderOf_injective M.subtype M.subtype_injective x').symm
have hcop : Nat.Coprime (orderOf x') N.index := by
rw [Nat.coprime_iff_gcd_eq_one]
by_contra hne
obtain ⟨p, hpp, hpdvd⟩ := Nat.exists_prime_and_dvd hne
rw [Nat.dvd_gcd_iff] at hpdvd
have hpord : p ∈ (orderOf x).primeFactors :=
Nat.mem_primeFactors.mpr ⟨hpp, hordx' ▸ hpdvd.1, (orderOf_pos x).ne'⟩
exact (hxπ' p hpord)
(hidxκ p (Nat.mem_primeFactors.mpr ⟨hpp, hpdvd.2, Subgroup.index_ne_zero_of_finite⟩))
-- The image of `x'` in `↥M ⧸ N` has order dividing both `orderOf x'` and `N.index`, hence `1`.
have hord1 : orderOf (QuotientGroup.mk' N x') = 1 := by
have h1 : orderOf (QuotientGroup.mk' N x') ∣ orderOf x' :=
orderOf_dvd_of_pow_eq_one (by rw [← map_pow, pow_orderOf_eq_one, map_one])
have h2 : orderOf (QuotientGroup.mk' N x') ∣ N.index := orderOf_dvd_natCard _
exact Nat.dvd_one.mp (hcop ▸ Nat.dvd_gcd h1 h2)
have hx'N : x' ∈ N := by
have h := orderOf_eq_one_iff.mp hord1
rwa [QuotientGroup.mk'_apply, QuotientGroup.eq_one_iff] at h
exact Subgroup.mem_subgroupOf.mp hx'N
-- **Helper B**: `M' ⊓ (K ⊔ K*) = K*` (direct: `x = k·k* ∈ M'`, `k* ∈ K* ≤ M'`, so
-- `k ∈ K ⊓ M' = 1`).
have hMeetKKstar : (U ⊔ OddOrder.BG.Ch3.S10.Msigma M : Subgroup G) ⊓ (K ⊔ Kstar) = Kstar := by
refine le_antisymm ?_ (le_inf (hKstarMσ.trans le_sup_right) le_sup_right)
-- `K` is normal in `K ⊔ K*` (`K*` centralizes it), so an element `x` of `K ⊔ K*` is `a·b`
-- with `a ∈ K`, `b ∈ K*`. If also `x ∈ M'`, then `a = x·b⁻¹ ∈ K ⊓ M' = ⊥`, so `x = b ∈ K*`.
have hKstarC : Kstar ≤ Subgroup.centralizer (K : Set G) := by rw [hKstar]; exact inf_le_right
have hKnorm : (K ⊔ Kstar : Subgroup G) ≤ Subgroup.normalizer (K : Set G) :=
sup_le Subgroup.le_normalizer (hKstarC.trans (Subgroup.centralizer_le_normalizer _))
have : (K.subgroupOf (K ⊔ Kstar)).Normal :=
Subgroup.normal_subgroupOf_of_le_normalizer hKnorm
have hsuptop : (K.subgroupOf (K ⊔ Kstar)) ⊔ (Kstar.subgroupOf (K ⊔ Kstar)) = ⊤ := by
rw [← Subgroup.subgroupOf_sup le_sup_left le_sup_right, Subgroup.subgroupOf_self]
have hKM'bot : K ⊓ (U ⊔ OddOrder.BG.Ch3.S10.Msigma M : Subgroup G) = ⊥ := by
have hcompl := (typeP_duality hG hM hP hKM hK hKstar).1
have hdisj : Disjoint ((derivedInG M).subgroupOf M) (K.subgroupOf M) := hcompl.disjoint
rw [← hM'eq, eq_bot_iff]
intro y hy
rw [Subgroup.mem_inf] at hy
have hmemMinf : (⟨y, hKM hy.1⟩ : ↥M) ∈ (K.subgroupOf M) ⊓ ((derivedInG M).subgroupOf M) :=
Subgroup.mem_inf.mpr ⟨Subgroup.mem_subgroupOf.mpr hy.1, Subgroup.mem_subgroupOf.mpr hy.2⟩
rw [disjoint_iff.mp hdisj.symm, Subgroup.mem_bot] at hmemMinf
rw [Subgroup.mem_bot]; exact Subtype.ext_iff.mp hmemMinf
intro x hx
rw [Subgroup.mem_inf] at hx
obtain ⟨hxM', hxKK⟩ := hx
obtain ⟨a, ha, b, hb, hab⟩ := Subgroup.mem_sup_of_normal_left.mp
(hsuptop ▸ Subgroup.mem_top (⟨x, hxKK⟩ : ↥(K ⊔ Kstar)))
have haK : (a : G) ∈ K := Subgroup.mem_subgroupOf.mp ha
have hbKstar : (b : G) ∈ Kstar := Subgroup.mem_subgroupOf.mp hb
have hab' : (a : G) * (b : G) = x := by have := congrArg Subtype.val hab; simpa using this
have haM' : (a : G) ∈ (U ⊔ OddOrder.BG.Ch3.S10.Msigma M : Subgroup G) := by
have heq : (a : G) = x * (b : G)⁻¹ := by rw [← hab']; group
rw [heq]
exact Subgroup.mul_mem _ hxM'
(Subgroup.inv_mem _ ((hKstarMσ.trans le_sup_right) hbKstar))
have ha1 : (a : G) = 1 := Subgroup.mem_bot.mp (hKM'bot ▸ Subgroup.mem_inf.mpr ⟨haK, haM'⟩)
rw [← hab', ha1, one_mul]; exact hbKstar
-- ## Main argument.
intro a ha
have haM : a ∈ M := ha.1.1.1
have hanc : a ∉ conjClassSet (sharpSubgroup K) := ha.1.2
have haUMσ : a ∉ ((U ⊔ OddOrder.BG.Ch3.S10.Msigma M : Subgroup G) : Set G) :=
fun hmem => ha.2 ⟨ha.1.1, hmem⟩
-- `κ`/`κ'`-decomposition `a = a_κ · a_{κ'}`.
obtain ⟨aκ, aκ', hmul, hcomm, hκ, hκ', hazκ, hazκ'⟩ := exists_isPiElement_mul (S14.kappa M) a
have hzpaM : Subgroup.zpowers a ≤ M := Subgroup.zpowers_le.mpr haM
have haκM : aκ ∈ M := hzpaM hazκ
have haκ'M : aκ' ∈ M := hzpaM hazκ'
-- `a_κ ≠ 1`: else `a = a_{κ'}` is a `κ'`-element, so `a ∈ M'`, contradicting `a ∉ U M_σ`.
have haκne : aκ ≠ 1 := by
intro h
refine haUMσ ?_
have : a = aκ' := by rw [← hmul, h, one_mul]
rw [this]; exact hkappaComplMem aκ' haκ'M hκ'
-- Conjugate `⟨a_κ⟩` into `K`.
have hXpi : Ch03.Subgroup.IsPiGroup (S14.kappa M) ((Subgroup.zpowers aκ).subgroupOf M) := by
intro p hp
rw [Nat.card_congr (Subgroup.subgroupOfEquivOfLe (Subgroup.zpowers_le.mpr haκM)).toEquiv,
Nat.card_zpowers] at hp
exact hκ p hp
obtain ⟨w, hwM, hwle⟩ :=
exists_conj_smul_le_isHall_kappa hG hM hKM hK (Subgroup.zpowers_le.mpr haκM) hXpi
-- `b := w·a·w⁻¹ = b_κ·b_{κ'}` with `b_κ = w·a_κ·w⁻¹ ∈ K`.
have hbκK : w * aκ * w⁻¹ ∈ K := by
have hmem : (MulAut.conj w) • aκ ∈ MulAut.conj w • Subgroup.zpowers aκ :=
Subgroup.smul_mem_pointwise_smul aκ (MulAut.conj w) _ (Subgroup.mem_zpowers aκ)
have heq : (MulAut.conj w) • aκ = w * aκ * w⁻¹ := rfl
rw [heq] at hmem; exact hwle hmem
-- `b_κ' := w·a_{κ'}·w⁻¹ ∈ M'` (conjugate of a `κ'`-element, `M' ◁ M`, `w ∈ M`).
have hbκ'M' : w * aκ' * w⁻¹ ∈ (U ⊔ OddOrder.BG.Ch3.S10.Msigma M : Subgroup G) := by
have hwnorm : w ∈ Subgroup.normalizer (derivedInG M) :=
OddOrder.BG.Ch3.S10.le_normalizer_derivedInG M hwM
have h1 : aκ' ∈ derivedInG M := by rw [hM'eq]; exact hkappaComplMem aκ' haκ'M hκ'
have hconjM' : w * aκ' * w⁻¹ ∈ derivedInG M :=
(Subgroup.mem_normalizer_iff.mp hwnorm aκ').mp h1
rwa [hM'eq] at hconjM'
-- `b_κ ≠ 1` and `b_κ' ≠ 1`.
have hbκne : w * aκ * w⁻¹ ≠ 1 := by
intro h
exact haκne (by
have := congrArg (fun y => w⁻¹ * y * w) h; simpa [mul_assoc] using this)
have hbκ'ne : w * aκ' * w⁻¹ ≠ 1 := by
intro h
-- `b_κ' = 1 ⟹ a_{κ'} = 1 ⟹ a = a_κ`, and `a = w⁻¹·b_κ·w` is `G`-conjugate to `b_κ ∈ K#`.
have haκ'1 : aκ' = 1 := by
have h2 := congrArg (fun y => w⁻¹ * y * w) h; simpa [mul_assoc] using h2
have haaκ : a = aκ := by rw [← hmul, haκ'1, mul_one]
exact hanc ⟨w * aκ * w⁻¹, ⟨hbκK, hbκne⟩, w⁻¹, by rw [haaκ]; group⟩
-- `b_κ' ∈ K*`: it lies in `M' ⊓ C_M(b_κ) = M' ⊓ (K ⊔ K*) = K*`.
have hbκ'comm : Commute (w * aκ * w⁻¹) (w * aκ' * w⁻¹) := by
change w * aκ * w⁻¹ * (w * aκ' * w⁻¹) = w * aκ' * w⁻¹ * (w * aκ * w⁻¹)
calc w * aκ * w⁻¹ * (w * aκ' * w⁻¹) = w * (aκ * aκ') * w⁻¹ := by group
_ = w * (aκ' * aκ) * w⁻¹ := by rw [hcomm.eq]
_ = w * aκ' * w⁻¹ * (w * aκ * w⁻¹) := by group
have hbκ'CM : w * aκ' * w⁻¹ ∈ Subgroup.centralizer ({w * aκ * w⁻¹} : Set G) := by
rw [Subgroup.mem_centralizer_iff]
intro g hg
rw [Set.mem_singleton_iff] at hg; subst hg
exact hbκ'comm.eq
have hbκ'Kstar : w * aκ' * w⁻¹ ∈ Kstar := by
have hCMeq : M ⊓ Subgroup.centralizer ({w * aκ * w⁻¹} : Set G) = K ⊔ Kstar :=
typeP_centralizer_kappaElement_eq hG hM hP hKM hK hKstar hU _ hbκK hbκne
have hbκ'KK : w * aκ' * w⁻¹ ∈ K ⊔ Kstar := by
rw [← hCMeq, Subgroup.mem_inf]
exact ⟨hM'M hbκ'M', hbκ'CM⟩
rw [← hMeetKKstar, Subgroup.mem_inf]
exact ⟨hbκ'M', hbκ'KK⟩
-- Assemble: `b := w·a·w⁻¹ = b_κ·b_{κ'} ∈ K·K* ⊆ K ⊔ K*`, and `b ∉ K`, `b ∉ K*`, so `b ∈ Ẑ`.
refine ⟨w⁻¹, M.inv_mem hwM, w * a * w⁻¹, ?_, by group⟩
have hbeq : w * a * w⁻¹ = (w * aκ * w⁻¹) * (w * aκ' * w⁻¹) := by
rw [← hmul]; group
refine ⟨?_, ?_⟩
· -- `b ∈ K ⊔ K*`.
rw [hbeq]
exact Subgroup.mul_mem _ (Subgroup.mem_sup_left hbκK) (Subgroup.mem_sup_right hbκ'Kstar)
· -- `b ∉ K ∪ K*`.
rw [Set.mem_union]
push Not
refine ⟨fun hbK => ?_, fun hbKstar => ?_⟩
· -- `b ∈ K ⟹ b_κ' = b_κ⁻¹·b ∈ K ⊓ K* = 1`, contra `b_κ' ≠ 1`.
have hbκ'K : w * aκ' * w⁻¹ ∈ K := by
have : w * aκ' * w⁻¹ = (w * aκ * w⁻¹)⁻¹ * (w * a * w⁻¹) := by rw [hbeq]; group
rw [this]; exact Subgroup.mul_mem _ (Subgroup.inv_mem _ hbκK) hbK
exact hbκ'ne (Subgroup.mem_bot.mp
(kappaHall_inf_Kstar_eq_bot hKM hK hKstar ▸ Subgroup.mem_inf.mpr ⟨hbκ'K, hbκ'Kstar⟩))
· -- `b ∈ K* ⟹ b_κ = b·b_κ'⁻¹ ∈ K ⊓ K* = 1`, contra `b_κ ≠ 1`.
have hbκK' : w * aκ * w⁻¹ ∈ Kstar := by
have : w * aκ * w⁻¹ = (w * a * w⁻¹) * (w * aκ' * w⁻¹)⁻¹ := by rw [hbeq]; group
rw [this]; exact Subgroup.mul_mem _ hbKstar (Subgroup.inv_mem _ hbκ'Kstar)
exact hbκne (Subgroup.mem_bot.mp
(kappaHall_inf_Kstar_eq_bot hKM hK hKstar ▸ Subgroup.mem_inf.mpr ⟨hbκK, hbκK'⟩))
/-- **Matched `κ`-Hall / `(κ∪σ)'`-Hall pair for a type-`P₂` maximal subgroup**
(BG `kappa_complement`;
the Frobenius complement `E = K ⋉ U` of Proposition 14.2(a), with `U = E₂E₃` the `M_σ`-complement's
derived subgroup). There is a `κ(M)`-Hall `K₀` and a nontrivial abelian `(κ(M)∪σ(M))'`-Hall `U₀`,
both `≤ M`, with `K₀ ≤ N_G(U₀)` — since `U₀ = E₂E₃ ◁ E` (`E23_normal`) and `K₀ = E₁ ≤ E`.
This is exactly the matched pair consumed by Corollary 14.12 (`typeP2_neighbor_is_typeF`) via its
`hKNU : K ≤ N(U)` hypothesis. As all `(κ∪σ)'`-Hall subgroups of `M` are `M`-conjugate, the
arbitrary `U` of Theorem C is `M`-conjugate to `U₀`, which transports `N_G(U) ⊄ M` between them.
Construction: take a §12 `E`-setup (`E = E₁E₂E₃`, `E` a `σ(M)'`-complement of `M_σ`). Type-`P₂`
excludes the degenerate cases (`κ ∩ τ₃ ≠ ∅`, or `E₂E₃ = 1`) which both make `E` a `κ`-group, hence
`κ(M) = π(M)∖σ(M)` (`kappa_eq_sigmaComplementPrimes_of_isPiGroup_card_E`), i.e. type-`P₁`. In the
remaining case `κ ⊆ τ₁`, `E₁` is the `κ`-Hall (`mem_kappa_of_mem_primeFactors_card_E1`) and
`E₂E₃ = [E : E₁]` is the `(κ∪σ)'`-Hall (its prime divisors avoid `κ` as the index of the `κ`-Hall
`E₁`, and `σ` as `E` is a `σ'`-group). -/
theorem typeP2_matched_kappa_hall_pair_of_esetup [Finite G]
(hG : OddOrder.BG.IsMinimalSimpleOdd G) {M E E₁ E₂ E₃ : Subgroup G}
(hM : M ∈ maximalSubgroups G) (hP2 : S14.IsTypeP2 M)
(hsetup : OddOrder.BG.Ch3.S12.SubgroupESetup M E E₁ E₂ E₃) :
E₁ ≤ M ∧ E₂ ⊔ E₃ ≤ E ∧ E₂ ⊔ E₃ ≠ ⊥ ∧
Ch03.IsHallSubgroup (S14.kappa M) (E₁.subgroupOf M) ∧
Ch03.IsHallSubgroup ((S14.kappa M ∪ OddOrder.BG.Ch3.S10.sigma M)ᶜ)
((E₂ ⊔ E₃).subgroupOf M) ∧
IsMulCommutative ↥(E₂ ⊔ E₃) ∧
E₁ ≤ Subgroup.normalizer ((E₂ ⊔ E₃ : Subgroup G) : Set G) := by
classical
have hP : S14.IsTypeP M := hP2.1
-- `E` is a `σ(M)'`-group.
have hEσ' : Ch03.Subgroup.IsPiGroup (OddOrder.BG.Ch3.S10.sigma M)ᶜ E :=
hsetup.isPiGroup_sigma_compl hG
-- `κ(M) ∩ τ₃(M) = ∅` (else `E` is a `κ`-group, giving type-`P₁`, contradicting `hP2`).
have hτ3empty : ¬ (S14.kappa M ∩ tau3 M).Nonempty := by
rintro ⟨p, hpκ, hpτ3⟩
have hp : p.Prime := S14.prime_of_mem_kappa hpκ
obtain ⟨hE3ne, hreg⟩ := S14.E3_not_regular_of_mem_kappa_tau3 hG hsetup hp hpκ hpτ3
obtain ⟨_, _, hEprime, _⟩ := OddOrder.BG.Ch3.S13.E3_not_regular_consequences hG hsetup hE3ne
hreg
obtain ⟨x, hxE3, hxne, hxC⟩ : ∃ x ∈ E₃, x ≠ 1 ∧
OddOrder.BG.Ch3.S10.Msigma M ⊓ Subgroup.centralizer ({x} : Set G) ≠ ⊥ := by
by_contra hcon; push Not at hcon; exact hreg fun y hy hy1 => hcon y hy hy1
exact hP2.2 (S14.kappa_eq_sigmaComplementPrimes_of_isPiGroup_card_E hG hsetup
(fun q hq => S14.mem_kappa_of_mem_primeFactors_card_E hG hsetup hEprime hxE3 hxne hxC hq))
have hκτ1 : ∀ p ∈ S14.kappa M, p ∈ tau1 M := fun p hpκ =>
(S14.kappa_subset_tau1_union_tau3 hpκ).resolve_right (fun hpτ3 => hτ3empty ⟨p, hpκ, hpτ3⟩)
-- `E₁` acts prime/non-regularly on `M_σ`; `E₁` is the `κ`-Hall of `E`, hence of `M`.
obtain ⟨p₀, hp₀κ⟩ := hP
obtain ⟨hE1ne, hE1nonreg⟩ := S14.E1_not_regular_of_mem_kappa_tau1 hG hsetup
(S14.prime_of_mem_kappa hp₀κ) hp₀κ (hκτ1 p₀ hp₀κ)
have hE1prime := OddOrder.BG.Ch3.S13.E1_actsPrime hG hsetup hE1ne
have hCE1 : OddOrder.BG.Ch3.S10.Msigma M ⊓ Subgroup.centralizer (E₁ : Set G) ≠ ⊥ :=
S14.Msigma_inf_centralizer_E_ne_bot_of_actsPrime_nonregular hE1prime (le_refl E₁) hE1nonreg
have hE1HallκE : Ch03.IsHallSubgroup (S14.kappa M) (E₁.subgroupOf E) :=
⟨fun p hp => S14.mem_kappa_of_mem_primeFactors_card_E1 hG hsetup hE1prime hCE1
(by rwa [Nat.card_congr (Subgroup.subgroupOfEquivOfLe hsetup.E₁_le).toEquiv] at hp),
fun p hp hpκ => hsetup.E₁_hall.2 p hp (hκτ1 p hpκ)⟩
have hK₀ : Ch03.IsHallSubgroup (S14.kappa M) (E₁.subgroupOf M) :=
hallPiece_isHall_in_M hG hsetup hsetup.E₁_le hE1HallκE S14.kappa_subset_sigmaCompl
set U₀ : Subgroup G := E₂ ⊔ E₃ with hU₀def
have hU₀leE : U₀ ≤ E := sup_le hsetup.E₂_le hsetup.E₃_le
-- `|E| = |E₁| · |U₀|` (the `τ₁` vs `τ₂∪τ₃` factorisation; `E₁ ⊓ U₀ = 1` by coprimality).
have hc1E : Nat.card ↥(E₁.subgroupOf E) = Nat.card ↥E₁ :=
Nat.card_congr (Subgroup.subgroupOfEquivOfLe hsetup.E₁_le).toEquiv
have hc2E : Nat.card ↥(E₂.subgroupOf E) = Nat.card ↥E₂ :=
Nat.card_congr (Subgroup.subgroupOfEquivOfLe hsetup.E₂_le).toEquiv
have hc3E : Nat.card ↥(E₃.subgroupOf E) = Nat.card ↥E₃ :=
Nat.card_congr (Subgroup.subgroupOfEquivOfLe hsetup.E₃_le).toEquiv
have hcardE : Nat.card ↥E = Nat.card ↥E₁ * Nat.card ↥U₀ := by
have hEsup : E = E₁ ⊔ E₂ ⊔ E₃ := (subgroupE_basic hG hsetup).2.2.2.2.1.1
have hEnormE3 : E ≤ Subgroup.normalizer (E₃ : Set G) := (subgroupE_basic hG hsetup).2.1.2
have hcop2 : Nat.Coprime (Nat.card ↥E₁) (Nat.card ↥E₂) := by
by_contra hnc
obtain ⟨s, hs, hsm, hsn⟩ := Nat.Prime.not_coprime_iff_dvd.mp hnc
have := tau1_pRank_eq_one (hsetup.E₁_hall.1 s
(hc1E ▸ Nat.mem_primeFactors.mpr ⟨hs, hsm, Nat.card_pos.ne'⟩))
have := tau2_pRank_eq_two (hsetup.E₂_hall.1 s
(hc2E ▸ Nat.mem_primeFactors.mpr ⟨hs, hsn, Nat.card_pos.ne'⟩))
omega
have hcop3 : Nat.Coprime (Nat.card ↥E₁) (Nat.card ↥E₃) := by
by_contra hnc
obtain ⟨s, hs, hsm, hsn⟩ := Nat.Prime.not_coprime_iff_dvd.mp hnc
exact not_mem_tau3_of_mem_tau1
(hsetup.E₁_hall.1 s (hc1E ▸ Nat.mem_primeFactors.mpr ⟨hs, hsm, Nat.card_pos.ne'⟩))
(hsetup.E₃_hall.1 s (hc3E ▸ Nat.mem_primeFactors.mpr ⟨hs, hsn, Nat.card_pos.ne'⟩))
have hE23disj : E₂ ⊓ E₃ = ⊥ := by
have hd1 : Nat.card ↥(E₂ ⊓ E₃) ∣ Nat.card ↥E₂ := Subgroup.card_dvd_of_le inf_le_left
have hd2 : Nat.card ↥(E₂ ⊓ E₃) ∣ Nat.card ↥E₃ := Subgroup.card_dvd_of_le inf_le_right
have hcop23 : Nat.Coprime (Nat.card ↥E₂) (Nat.card ↥E₃) := by
by_contra hnc
obtain ⟨s, hs, hsm, hsn⟩ := Nat.Prime.not_coprime_iff_dvd.mp hnc
have := tau2_pRank_eq_two (hsetup.E₂_hall.1 s
(hc2E ▸ Nat.mem_primeFactors.mpr ⟨hs, hsm, Nat.card_pos.ne'⟩))
have := tau3_pRank_eq_one (hsetup.E₃_hall.1 s
(hc3E ▸ Nat.mem_primeFactors.mpr ⟨hs, hsn, Nat.card_pos.ne'⟩))
omega
exact Subgroup.card_eq_one.mp (Nat.dvd_one.mp (hcop23 ▸ Nat.dvd_gcd hd1 hd2))
have hcard23 : Nat.card ↥U₀ = Nat.card ↥E₂ * Nat.card ↥E₃ :=
card_sup_eq_mul_of_le_normalizer_of_disjoint (hsetup.E₂_le.trans hEnormE3) hE23disj
have hE1_23_disj : E₁ ⊓ U₀ = ⊥ := by
have hd1 : Nat.card ↥(E₁ ⊓ U₀) ∣ Nat.card ↥E₁ := Subgroup.card_dvd_of_le inf_le_left
have hd2 : Nat.card ↥(E₁ ⊓ U₀) ∣ Nat.card ↥U₀ := Subgroup.card_dvd_of_le inf_le_right
have hcop : Nat.Coprime (Nat.card ↥E₁) (Nat.card ↥U₀) :=
hcard23 ▸ Nat.Coprime.mul_right hcop2 hcop3
exact Subgroup.card_eq_one.mp (Nat.dvd_one.mp (hcop ▸ Nat.dvd_gcd hd1 hd2))
have hE1normU0 : E₁ ≤ Subgroup.normalizer (U₀ : Set G) :=
hsetup.E₁_le.trans (hsetup.E23_normal hG)
have hsupEq : E₁ ⊔ U₀ = E := by rw [hU₀def, ← sup_assoc, ← hEsup]
rw [← hsupEq, card_sup_eq_mul_of_le_normalizer_of_disjoint hE1normU0 hE1_23_disj]
-- index facts `[E:E₁] = |U₀|`, `[E:U₀] = |E₁|`.
have hidxE1 : (E₁.subgroupOf E).index = Nat.card ↥U₀ := by
have hmi := Subgroup.card_mul_index (E₁.subgroupOf E)
rw [hc1E] at hmi
exact Nat.eq_of_mul_eq_mul_left Nat.card_pos (by rw [hmi, hcardE])
have hidxU0 : (U₀.subgroupOf E).index = Nat.card ↥E₁ := by
have hmi := Subgroup.card_mul_index (U₀.subgroupOf E)
rw [Nat.card_congr (Subgroup.subgroupOfEquivOfLe hU₀leE).toEquiv] at hmi
exact Nat.eq_of_mul_eq_mul_left Nat.card_pos (by rw [hmi, hcardE]; ring)
-- `U₀` is a `(κ∪σ)'`-Hall of `E`, hence of `M`.
have hU₀HallE : Ch03.IsHallSubgroup ((S14.kappa M ∪ OddOrder.BG.Ch3.S10.sigma M)ᶜ)
(U₀.subgroupOf E) := by
constructor
· intro p hp
have hpU₀ : p ∈ (Nat.card ↥U₀).primeFactors := by
rwa [Nat.card_congr (Subgroup.subgroupOfEquivOfLe hU₀leE).toEquiv] at hp
rw [Set.mem_compl_iff, Set.mem_union, not_or]
refine ⟨fun hpκ => hE1HallκE.2 p (by rw [hidxE1]; exact hpU₀) hpκ, ?_⟩
exact hEσ' p (Nat.primeFactors_mono (Subgroup.card_dvd_of_le hU₀leE) Nat.card_pos.ne' hpU₀)
· intro p hp
rw [hidxU0] at hp
exact fun hc => hc (Set.mem_union_left _ (hE1HallκE.1 p (by rwa [hc1E])))
have hU₀ : Ch03.IsHallSubgroup ((S14.kappa M ∪ OddOrder.BG.Ch3.S10.sigma M)ᶜ) (U₀.subgroupOf M) :=
hallPiece_isHall_in_M hG hsetup hU₀leE hU₀HallE
(fun p hp hpσ => hp (Or.inr hpσ))
-- `U₀ ≠ ⊥` (else `E = E₁` is a `κ`-Hall covering all `σ'`-primes, giving type-`P₁`).
have hU₀ne : U₀ ≠ ⊥ := by
intro hU₀bot
refine hP2.2 (S14.kappa_eq_sigmaComplementPrimes_of_isPiGroup_card_E hG hsetup (fun q hq => ?_))
rw [hcardE, hU₀bot, Subgroup.card_bot, mul_one] at hq
exact hE1HallκE.1 q (by rwa [hc1E])
-- `U₀` abelian (Lemma 15.1(b)).
have hU₀ab : IsMulCommutative ↥U₀ :=
(S15.typeP_hall_derived_eq_and_abelian hG hM hsetup.E1_le_M (hU₀leE.trans hsetup.E_le)
hE1ne hK₀ hU₀).2
have hK₀NU₀ : E₁ ≤ Subgroup.normalizer (U₀ : Set G) :=
hsetup.E₁_le.trans (hsetup.E23_normal hG)
exact ⟨hsetup.E1_le_M, hU₀leE, hU₀ne, hK₀, hU₀, hU₀ab, hK₀NU₀⟩
/-- **Matched `κ`-Hall / `(κ∪σ)'`-Hall pair for a type-`P₂` maximal subgroup**
(BG `kappa_complement`;
existential form). Obtains an arbitrary `E`-setup (`exists_subgroupESetup`) and reads off the pair
via `typeP2_matched_kappa_hall_pair_of_esetup`. Consumers needing the pair inside a *prescribed*
`σ(M)'`-complement `E` (e.g. `E = M ⊓ N` in Corollary 15.9, so a Sylow of `U₀` lands in `M`) call
the
`_of_esetup` core with their own setup. -/
theorem typeP2_exists_matched_kappa_hall_pair [Finite G]
(hG : OddOrder.BG.IsMinimalSimpleOdd G) {M : Subgroup G}
(hM : M ∈ maximalSubgroups G) (hP2 : S14.IsTypeP2 M) :
∃ K₀ U₀ : Subgroup G, K₀ ≤ M ∧ U₀ ≤ M ∧ U₀ ≠ ⊥ ∧
Ch03.IsHallSubgroup (S14.kappa M) (K₀.subgroupOf M) ∧
Ch03.IsHallSubgroup ((S14.kappa M ∪ OddOrder.BG.Ch3.S10.sigma M)ᶜ) (U₀.subgroupOf M) ∧
IsMulCommutative ↥U₀ ∧
K₀ ≤ Subgroup.normalizer (U₀ : Set G) := by
obtain ⟨E, E₁, E₂, E₃, hsetup⟩ := exists_subgroupESetup hG hM
obtain ⟨h1, h2, h3, h4, h5, h6, h7⟩ :=
typeP2_matched_kappa_hall_pair_of_esetup hG hM hP2 hsetup
exact ⟨E₁, E₂ ⊔ E₃, h1, h2.trans hsetup.E_le, h3, h4, h5, h6, h7⟩
/-- **BG Theorem C** (mmd L4303): when `K ≠ 1`, `M` has a paired maximal
subgroup `Mstar`, the cyclic product `Z = K Kstar`, a TI set `Z_tilde`, and the
associated type-P duality.
**Faithfulness fixes (2026-06-21, lane F).**
* **Hypotheses `hKM : K ≤ M`, `hUM : U ≤ M`.** In BG, `K` is the cyclic Hall
`κ(M)`-subgroup *of* `M` and `U` the Hall `(κ(M) ∪ σ(M))'`-subgroup *of* `M`
(`M = K U M_σ`), so `K, U ≤ M`. The `…subgroupOf M` Hall hypotheses constrain
only `K ⊓ M`, `U ⊓ M`, so they do **not** force `K, U ≤ M`; without them conjunct 7
(`M' = U ⊔ M_σ`) is false for `U ⊄ M`. The landed §14/§15 lemmas
(`typeP_duality`, `typeP_kstar_in_mf`, `typeP_hall_derived_eq_and_abelian`) all
take these explicitly, so they are added here and threaded through the only
callers (`theoremII_tame_embedding{,_of_inputs}`).
* **The `∃! Mstar` clause is strengthened to the faithful `typeP_duality` predicate.**
The previous scaffold pinned `Mstar` only by `maximal ∧ type-P ∧ ¬conj M Mstar ∧
(type-P₂ M ∨ type-P₂ Mstar)`; since every `G`-conjugate of the partner satisfies
that predicate (and `N_G(M*) = M* ⊊ G`), the conjunction had **no unique** witness —
the `∃!` was false as stated. BG Theorem C(4)-(7),(11) pins `Mstar` by the dual
Hall datum `K* ≤ M*`, `K*` Hall-`κ(M*)`, `K = M*_σ ⊓ C_G(K*)`, which is exactly
Theorem 14.7 (`typeP_duality`); that is the predicate used here. The sole callers
(`theoremII_tame_embedding_of_inputs`) project out only the TI conjunct, so the
strengthening does not affect them.
Proof: conjuncts 1,7 from Lemma 15.1(b) (`typeP_hall_derived_eq_and_abelian`),
conjuncts 3-6,8 from Corollary 15.6 (`typeP_kstar_in_mf`), conjunct 9 from
Theorem 14.7 (`typeP_duality`). The residual conjuncts are the genuinely deep
§14/§16 endpoints not yet available as standalone lemmas: conjunct 2 (`N_G(U) ⊄ M`,
BG Theorem C(1) / Corollary 14.12), conjunct 10 (`A_0(M) - A(M)` TI, BG Theorem A(3),(5)
/ Proposition 14.2(d)), conjunct 11 (`U ≠ 1 → |K|` prime ∧ `F(M)` TI, BG Theorem C(10)
= Proposition 14.2(g) + Theorem 15.7(a)), and conjunct 12 (`U = 1 → |K*|` prime, BG
Theorem C(8), pending the `IsTypeP1`-from-`U = ⊥` bridge). -/
theorem theoremC_paired_structure [Finite G]
(hG : OddOrder.BG.IsMinimalSimpleOdd G) {M K Kstar U : Subgroup G}
(hM : M ∈ maximalSubgroups G) (hKne : K ≠ ⊥) (hKM : K ≤ M) (hUM : U ≤ M)
(hK : Ch03.IsHallSubgroup (S14.kappa M) (K.subgroupOf M))
(hKstar : Kstar = OddOrder.BG.Ch3.S10.Msigma M ⊓ Subgroup.centralizer (K : Set G))
(hU : Ch03.IsHallSubgroup ((S14.kappa M ∪ OddOrder.BG.Ch3.S10.sigma M)ᶜ)
(U.subgroupOf M)) :
IsMulCommutative ↥U ∧ ¬ Subgroup.normalizer (U : Set G) ≤ M ∧
Kstar ≠ ⊥ ∧ IsCyclic ↥Kstar ∧ Kstar ≤ S15.MF M ∧ ¬ IsCyclic ↥(S15.MF M) ∧
derivedInG M = U ⊔ OddOrder.BG.Ch3.S10.Msigma M ∧
Kstar ≤ derivedInG (derivedInG M) ∧
(∃! Mstar : Subgroup G,
Mstar ∈ maximalSubgroups G ∧ S14.IsTypeP Mstar ∧
¬ S14.IsConjugateSubgroup M Mstar ∧
(Kstar ≤ Mstar ∧ Ch03.IsHallSubgroup (S14.kappa Mstar) (Kstar.subgroupOf Mstar) ∧
K = OddOrder.BG.Ch3.S10.Msigma Mstar ⊓ Subgroup.centralizer (Kstar : Set G)) ∧
IsCyclic ↥(K ⊔ Kstar) ∧ IsTISubset (S14.zTilde K Kstar) (K ⊔ Kstar) ∧
(S14.IsTypeP2 M ∨ S14.IsTypeP2 Mstar) ∧
(∀ H : Subgroup G, H ∈ maximalSubgroups G → S14.IsTypeP H →
S14.IsConjugateSubgroup H M ∨ S14.IsConjugateSubgroup H Mstar)) ∧
IsTISubset (A0Set M K \ ASet M U) M ∧
(U ≠ ⊥ → ∃ p : ℕ, p.Prime ∧ Nat.card ↥K = p ∧ S15.FittingIsTI M) ∧
(U = ⊥ → ∃ q : ℕ, q.Prime ∧ Nat.card ↥Kstar = q) := by
classical
-- `IsTypeP M` from the nonempty `κ`-Hall factor `K`.
have hKofne : K.subgroupOf M ≠ ⊥ := by
rw [ne_eq, Subgroup.subgroupOf_eq_bot]
exact fun hd => hKne (hd.eq_bot_of_le hKM)
have hP : S14.IsTypeP M := isTypeP_of_isHall_kappa_subgroupOf_ne_bot hK hKofne
-- Conjuncts 1, 7 (Lemma 15.1(b)): `U` abelian and `M' = U ⊔ M_σ`.
obtain ⟨hM'eq, hUab⟩ := typeP_hall_derived_eq_and_abelian hG hM hKM hUM hKne hK hU
-- Conjuncts 3,4,5,8,6 (Corollary 15.6): `K*` nontrivial cyclic `≤ M_F`, `K* ≤ M''`, `M_F` not
-- cyclic.
obtain ⟨hKsne, hKscyc, hKsMF, hKsdd, hMFnc⟩ := typeP_kstar_in_mf hG hM hP hKM hK hKstar
-- Conjunct 9 (Theorem 14.7): the unique non-conjugate type-`P` partner `M*`.
have hdual := (typeP_duality hG hM hP hKM hK hKstar).2.2
refine ⟨hUab, ?_, hKsne, hKscyc, hKsMF, hMFnc, hM'eq, hKsdd, hdual, ?_, ?_, ?_⟩
· -- Conjunct 2 (BG Theorem C(1) / Corollary 14.12): `N_G(U) ⊄ M`.
by_cases hUbot : U = ⊥
· -- `U = ⊥` (type-`P₁`): `N_G(⊥) = ⊤ ⊄ M` since `M` is a proper (maximal) subgroup.
subst hUbot
intro hle
have htop : (⊤ : Subgroup G) ≤ Subgroup.normalizer ((⊥ : Subgroup G) : Set G) := by
intro g _
rw [Subgroup.mem_normalizer_iff]
intro h
simp [Subgroup.mem_bot]
exact (mem_maximalSubgroups.mp hM).1 (top_le_iff.mp (htop.trans hle))
· -- `U ≠ ⊥` (type-`P₂`): `N_G(U) ⊄ M` is BG Corollary 14.12. We produce a *matched*
-- `κ`-Hall / `(κ∪σ)'`-Hall pair `(K₀, U₀)` (with `K₀ ≤ N_G(U₀)`, the hypothesis Corollary
-- 14.12 consumes), apply 14.12 to get `N_G(U₀) ⊄ M`, then transport along the `M`-conjugacy
-- `U₀ ~ U` of `(κ∪σ)'`-Hall subgroups.
have hUne' : U.subgroupOf M ≠ ⊥ := by
rw [ne_eq, Subgroup.subgroupOf_eq_bot]
exact fun hd => hUbot (hd.eq_bot_of_le hUM)
have hP2 : S14.IsTypeP2 M := isTypeP2_of_hall_subgroupOf_ne_bot hP hU hUne'
obtain ⟨K₀, U₀, hK₀M, hU₀M, hU₀ne, hK₀, hU₀, hU₀ab, hK₀NU₀⟩ :=
typeP2_exists_matched_kappa_hall_pair hG hM hP2
have : Group.IsSolvable ↥M := hG.isSolvable_of_mem_maximalSubgroups hM
have : Group.IsSolvable ↥U₀ :=
Group.isSolvable_of_isSolvable_injective (Subgroup.inclusion_injective hU₀M)
-- a prime `r ∣ |U₀|` and a Sylow (= Hall `{r}`) subgroup `R₀ ≤ U₀`.
have hU₀card : Nat.card ↥U₀ ≠ 1 := fun h => hU₀ne (Subgroup.card_eq_one.mp h)
obtain ⟨r, hr⟩ : (Nat.card ↥U₀).primeFactors.Nonempty :=
Nat.nonempty_primeFactors.mpr
(lt_of_le_of_ne (Nat.one_le_iff_ne_zero.mpr Nat.card_pos.ne') (Ne.symm hU₀card))
obtain ⟨R', hR'⟩ := Ch03.hall_E_exists (G := ↥U₀) ({r} : Set ℕ)
have hR₀sub : ((R'.map U₀.subtype).subgroupOf U₀) = R' :=
Subgroup.comap_map_eq_self_of_injective U₀.subtype_injective R'
have hR₀ : Ch03.IsHallSubgroup ({r} : Set ℕ) ((R'.map U₀.subtype).subgroupOf U₀) := by
rw [hR₀sub]; exact hR'
have hU₀ab' : ∀ a ∈ U₀, ∀ b ∈ U₀, a * b = b * a := fun a ha b hb =>
congrArg Subtype.val (isMulCommutative_iff.mp hU₀ab (⟨a, ha⟩ : ↥U₀) ⟨b, hb⟩)
obtain ⟨H, _, _, _, _, hHNU₀⟩ :=
S14.typeP2_neighbor_is_typeF hG hM hP2 hK₀M hU₀M hK₀ hU₀ hU₀ab' hr (Subgroup.map_subtype_le
_)
hR₀ hK₀NU₀
-- Transport `N_G(U) ⊄ M` (assumed `≤`, for contradiction) to `N_G(U₀) ≤ M`.
intro hle
apply hHNU₀
obtain ⟨w, hwM, hw⟩ := OddOrder.BG.Ch1.S06.exists_conj_eq_of_isHall_subgroupOf
inferInstance hU₀M hUM hU₀ hU
refine le_trans inf_le_right ?_
intro n hn
have hn_fix : MulAut.conj n • U₀ = U₀ := conj_smul_eq_self_of_mem_normalizer hn
have hwinv : MulAut.conj w⁻¹ • U = U₀ := by
rw [← hw, ← mul_smul, ← map_mul, inv_mul_cancel, map_one, one_smul]
have hfix : MulAut.conj (w * n * w⁻¹) • U = U := by
rw [map_mul, map_mul, mul_smul, mul_smul, hwinv, hn_fix, hw]
have hwnwM : w * n * w⁻¹ ∈ M := hle (mem_normalizer_of_conj_smul_eq_self hfix)
have hn' : n = w⁻¹ * (w * n * w⁻¹) * w := by group
rw [hn']
exact M.mul_mem (M.mul_mem (M.inv_mem hwM) hwnwM) hwM
· -- Conjunct 10 (BG Theorem C(9)): `A_0(M) - A(M) = 𝒞_M(Ẑ)` is a TI-subset of `M`. Reduce to the
-- structural inclusion `A_0(M) - A(M) ⊆ 𝒞_M(Ẑ)` (`a0_minus_a_subset_conj_zTilde`) plus the
-- TI-ness of `Ẑ` (`typeP_duality`; `N_G(Ẑ) = K ⊔ K* ≤ M`), via the transport lemma.
obtain ⟨Mstar, hMstarP⟩ := hdual.exists
have hZti : OddOrder.GroupTheory.IsTISubset (S14.zTilde K Kstar) (K ⊔ Kstar) :=
hMstarP.2.2.2.2.2.1
have hKstarMσ : Kstar ≤ OddOrder.BG.Ch3.S10.Msigma M := by rw [hKstar]; exact inf_le_left
have hZM : (K ⊔ Kstar : Subgroup G) ≤ M :=
sup_le hKM (hKstarMσ.trans (OddOrder.BG.Ch3.S10.Msigma_le M))
exact hZti.of_subset_conj_of_isTISubset hZM
(a0_minus_a_subset_conj_zTilde hG hM hP hKM hUM hK hKstar hU)
· -- Conjunct 11 (BG Theorem C(10) = Prop 14.2(g) + Theorem 15.7(a)):
-- `U ≠ ⊥ → |K|` prime ∧ `F(M)` TI.
-- `U ≠ ⊥` makes `M` type-`P₂`; `|K| = q` prime is Prop 14.2's type-`P₂` clause. The
-- `FittingIsTI M` conjunct (BG Theorem C(10), via Theorem 15.7(a)) is the residual.
intro hUne
have hUne' : U.subgroupOf M ≠ ⊥ := by
rw [ne_eq, Subgroup.subgroupOf_eq_bot]
exact fun hd => hUne (hd.eq_bot_of_le hUM)
have hP2 : S14.IsTypeP2 M := isTypeP2_of_hall_subgroupOf_ne_bot hP hU hUne'
obtain ⟨q, hq, hKq, _⟩ := ((S14.typeP_structure hG hM hP hKM hK hKstar hU).2.2.2.2.1 hP2).2
-- `FittingIsTI M` for type-`P₂`: BG Theorem 15.7(a) (`fittingIsTI_of_isTypeP2`).
exact ⟨q, hq, hKq, S15.fittingIsTI_of_isTypeP2 hG hM hP2⟩
· -- Conjunct 12 (BG Theorem C(8) = `kstar_card_prime_of_inputs`): `U = ⊥ → |K*|` prime.
-- `U = ⊥` makes `M` type-`P₁` (`κ(M) = σ(M)'` via the trivial Hall complement), then `|K*|`
-- prime.
intro hUbot
have hUbot' : U.subgroupOf M = ⊥ := by simp [hUbot]
have hP1 : S14.IsTypeP1 M :=
⟨hP, kappa_eq_sigmaComplementPrimes_of_hall_subgroupOf_eq_bot hU hUbot'⟩
exact ⟨Nat.card ↥Kstar, kstar_card_prime_of_inputs hG hM hP1 hKM hK hKstar, rfl⟩
/-- **BG Theorem D(1), `M_σ`-fusion control** (mmd L4317, the first conjunct of
`theoremD_msigma_conjugacy_and_centralizers`): two elements of `M_σ` conjugate in `G` are already
conjugate by an element of `M`. This is the one Theorem-D conjunct that is unconditional given
Corollary 15.3: applying `mf_hall_centralizer_control` (Cor 15.3(b), sorry-free) to the trivial
Hall subgroup `H := M_σ` of `M_σ` gives `N_G(M_σ)`-fusion, and `N_G(M_σ) = M`
(`normalizer_Msigma_eq_self`) upgrades it to `M`-fusion. Standalone and reusable; the remaining
Theorem-D conjuncts D(2) (Lemma 12.17) / D(3)(4) (the `R(x)` signalizer normal complement, deep) are
isolated in `theoremD_msigma_conjugacy_and_centralizers_of_inputs` (issue 8019). -/
theorem msigma_fusion_control [Finite G] (hG : OddOrder.BG.IsMinimalSimpleOdd G)
{M : Subgroup G} (hM : M ∈ maximalSubgroups G) :
∀ x ∈ OddOrder.BG.Ch3.S10.Msigma M, ∀ y ∈ OddOrder.BG.Ch3.S10.Msigma M,
(∃ g : G, y = g * x * g⁻¹) → ∃ m ∈ M, y = m * x * m⁻¹ := by
-- `M_σ` is (trivially) a `piSet`-Hall subgroup of itself.
have hHall : Ch03.IsHallSubgroup (S14.piSet (OddOrder.BG.Ch3.S10.Msigma M))
((OddOrder.BG.Ch3.S10.Msigma M).subgroupOf (OddOrder.BG.Ch3.S10.Msigma M)) := by
rw [Subgroup.subgroupOf_self, Ch03.IsHallSubgroup.top_iff]
exact fun p hp => hp
intro x hx y hy hconj
-- Corollary 15.3(b) at `H := M_σ`: `N_G(M_σ)`-fusion; then `N_G(M_σ) = M`.
obtain ⟨n, hnN, hnconj⟩ :=
(mf_hall_centralizer_control hG hM le_rfl hHall (OddOrder.BG.Ch3.S10.Msigma_ne_bot hG hM)).2
x hx y hy hconj
rw [normalizer_Msigma_eq_self hG hM] at hnN
exact ⟨n, hnN, hnconj⟩
/-- **Dichotomy `|𝓜_σ(x)| ≤ 1 ⟹ C_G(x) ≤ M`** (the converse of
`maximalSigmaSubgroupsOfElement_eq_singleton_of_centralizer_le`, Coq Theorem 14.4's `not_sCX_M`
direction): for `x ∈ M_σ^#` with at most one `σ`-maximal, `C_G(x) ≤ M`. If some
`c ∈ C_G(x)` escaped
`M`, then `Mᶜ` would be a second `σ`-maximal of `x` — `x ∈ Mᶜ` (as `c⁻¹xc = x ∈ M`), so `Mᶜ` is a
maximal conjugate containing `x`, hence in `𝓜_σ(x)` (`maximalConjugatesContaining_eq_maximalSigma`),
and `Mᶜ ≠ M` (else `c ∈ N_G(M) = M`) — contradicting `|𝓜_σ(x)| ≤ 1`. Much shallower than the Coq
proof's `C(X)`-orbit count: this direction needs only `N_G(M) = M` (maximal self-normalizing,
inlined). -/
theorem centralizer_le_of_maximalSigma_le_one [Finite G] (hG : OddOrder.BG.IsMinimalSimpleOdd G)
{M : Subgroup G} (hM : M ∈ maximalSubgroups G) {x : G}
(hxMσ : x ∈ OddOrder.BG.Ch3.S10.Msigma M) (hx1 : x ≠ 1)
(hle : (S14.maximalSigmaSubgroupsOfElement x).ncard ≤ 1) :
Subgroup.centralizer ({x} : Set G) ≤ M := by
have hMne : M ≠ ⊥ := fun h =>
OddOrder.BG.Ch3.S10.Msigma_ne_bot hG hM (le_bot_iff.mp (h ▸ OddOrder.BG.Ch3.S10.Msigma_le M))
-- `N_G(M) = M`: `M ≤ N_G(M)`, and a proper inclusion forces `M ◁ G` (maximality), excluded by
-- simplicity (`M ≠ ⊥, ⊤`).
have hNM : Subgroup.normalizer (M : Set G) = M := by
refine le_antisymm ?_ Subgroup.le_normalizer
rcases eq_or_lt_of_le (Subgroup.le_normalizer (H := M)) with heq | hlt
· exact le_of_eq heq.symm
· have hnorm : M.Normal :=
Subgroup.normalizer_eq_top_iff.mp ((mem_maximalSubgroups.mp hM).2 _ hlt)
rcases hG.simple.eq_bot_or_eq_top_of_normal _ hnorm with hbot | htop
· exact absurd hbot hMne
· exact absurd htop (mem_maximalSubgroups.mp hM).1
-- `conj c • M = M → c ∈ N_G(M)`.
have hconj_mem : ∀ c : G, MulAut.conj c • M = M → c ∈ Subgroup.normalizer (M : Set G) := by
intro c hcfix
rw [Subgroup.mem_normalizer_iff]
intro n
have key : MulAut.conj c • n ∈ MulAut.conj c • M ↔ n ∈ M :=
Subgroup.smul_mem_pointwise_smul_iff
rw [hcfix, show (MulAut.conj c • n) = (MulAut.conj c) n from rfl, MulAut.conj_apply] at key
exact key.symm
intro c hc
by_contra hcM
have hMmem : M ∈ S14.maximalSigmaSubgroupsOfElement x := ⟨hM, hxMσ⟩
-- `Mᶜ ∈ 𝓜_σ(x)`: `x ∈ Mᶜ` since `c⁻¹xc = x ∈ M`.
have hxcM : x ∈ MulAut.conj c • M := by
rw [Subgroup.mem_pointwise_smul_iff_inv_smul_mem]
have hcalc : (MulAut.conj c)⁻¹ • x = c⁻¹ * x * c := by
change (MulAut.conj c).symm x = _
rw [MulAut.conj_symm_apply]
rw [hcalc]
have hcx : c * x = x * c := Subgroup.mem_centralizer_singleton_iff.mp hc
have hxconj : c⁻¹ * x * c = x := by rw [mul_assoc, ← hcx]; group
rw [hxconj]
exact OddOrder.BG.Ch3.S10.Msigma_le M hxMσ
have hconjmem : MulAut.conj c • M ∈ S14.maximalSigmaSubgroupsOfElement x := by
rw [← maximalConjugatesContaining_eq_maximalSigma hG hM hxMσ hx1]
exact ⟨c, rfl, hxcM⟩
-- two members of a `≤ 1`-element set coincide, so `c ∈ N_G(M) = M`.
have heq : MulAut.conj c • M = M :=
(Set.ncard_le_one (Set.toFinite _)).mp hle _ hconjmem _ hMmem
exact hcM (hNM ▸ hconj_mem c heq)
/-- **BG Theorem D(3), full `hD3`** (`∀ x ∈ M_σ^#, ∃ R, RData M x R`): combine the two branches of
the `|𝓜_σ(x)|` dichotomy. When `|𝓜_σ(x)| > 1`, `RData_of_gt_one` supplies the signalizer `R(x)`;
when `|𝓜_σ(x)| ≤ 1`, the dichotomy gives `C_G(x) ≤ M`, whereupon `R = ⊥` works: `C_M(x) = C_G(x)`
is all of `C_G(x)` (conjunct 1 trivial, conjunct 3 is `⊤ ⋊ ⊥`), `⊥ ◁ C_G(x)`, and `𝓜_σ(x) = {M}`
makes sharp transitivity vacuous. -/
theorem exists_RData_of_mem_sigmaSharp [Finite G] (hG : OddOrder.BG.IsMinimalSimpleOdd G)
{M : Subgroup G} (hM : M ∈ maximalSubgroups G) :
∀ x : G, x ∈ S14.sigmaSharp M → ∃ R : Subgroup G, RData M x R := by
intro x hx
have hxMσ : x ∈ OddOrder.BG.Ch3.S10.Msigma M := hx.1
have hx1 : x ≠ 1 := hx.2
by_cases hgt : 1 < (S14.maximalSigmaSubgroupsOfElement x).ncard
· exact RData_of_gt_one hG hM hx hgt
· have hCxM : Subgroup.centralizer ({x} : Set G) ≤ M :=
centralizer_le_of_maximalSigma_le_one hG hM hxMσ hx1 (not_lt.mp hgt)
have hMCx : M ⊓ Subgroup.centralizer ({x} : Set G) = Subgroup.centralizer ({x} : Set G) :=
inf_eq_right.mpr hCxM
have hc1 : Nat.Coprime
(Nat.card ↥((M ⊓ Subgroup.centralizer ({x} : Set G)).subgroupOf
(Subgroup.centralizer ({x} : Set G))))
((M ⊓ Subgroup.centralizer ({x} : Set G)).subgroupOf
(Subgroup.centralizer ({x} : Set G))).index := by
rw [hMCx, Subgroup.subgroupOf_self, Subgroup.index_top]; exact Nat.coprime_one_right _
have hc2 : ((⊥ : Subgroup G).subgroupOf (Subgroup.centralizer ({x} : Set G))).Normal := by
rw [Subgroup.bot_subgroupOf]; infer_instance
have hc3 : Subgroup.IsComplement'
((M ⊓ Subgroup.centralizer ({x} : Set G)).subgroupOf (Subgroup.centralizer ({x} : Set G)))
((⊥ : Subgroup G).subgroupOf (Subgroup.centralizer ({x} : Set G))) := by
rw [hMCx, Subgroup.subgroupOf_self, Subgroup.bot_subgroupOf]
exact Subgroup.isComplement'_top_bot
have hc4 : ConjSharplyTransitiveOn (⊥ : Subgroup G) (maximalConjugatesContaining M x) := by
have hsingle : maximalConjugatesContaining M x = {M} := by
rw [maximalConjugatesContaining_eq_maximalSigma hG hM hxMσ hx1]
exact maximalSigmaSubgroupsOfElement_eq_singleton_of_centralizer_le hG hM hxMσ hx1 hCxM
rw [hsingle]
intro A hA B hB
rw [Set.mem_singleton_iff] at hA hB
subst hA; subst hB
exact ⟨1, ⟨Subgroup.mem_bot.mpr rfl, by rw [map_one, one_smul]⟩,
fun r hr => Subgroup.mem_bot.mp hr.1⟩
exact ⟨⊥, hc1, hc2, hc3, hc4⟩
/-- **BG Theorem E, `σ(Mᵢ)`-disjointness conjunct** (mmd L4370): distinct
representatives of the conjugacy classes of maximal subgroups have disjoint
`σ`-sets. This is the `σ(Mᵢ)` "disjoint union" piece of Theorem E, and it is
**already unconditional**: distinct class representatives are non-conjugate (forced
by the `∃!` uniqueness in `hreps` — both `Mᵢ` and `Mⱼ` would be *the* representative
of `Mᵢ`'s class), so BG Theorem 13.9
(`OddOrder.BG.Ch3.S13.sigma_disjoint_of_nonconjugate`, landed sorry-free) applies
directly. `hrepsMax` records that the representatives are themselves maximal
subgroups (the hypothesis Theorem 13.9 needs); it holds for any genuine set of
conjugacy-class representatives. -/
theorem sigma_reps_pairwise_disjoint [Finite G] (hG : OddOrder.BG.IsMinimalSimpleOdd G)
{reps : Set (Subgroup G)}
(hrepsMax : ∀ Mi ∈ reps, Mi ∈ maximalSubgroups G)
(hreps : ∀ H : Subgroup G, H ∈ maximalSubgroups G →
∃! Mi : Subgroup G, Mi ∈ reps ∧ S14.IsConjugateSubgroup H Mi)
{Mi Mj : Subgroup G} (hMi : Mi ∈ reps) (hMj : Mj ∈ reps) (hne : Mi ≠ Mj) :
OddOrder.BG.Ch3.S10.sigma Mi ∩ OddOrder.BG.Ch3.S10.sigma Mj = ∅ := by
rw [← Set.disjoint_iff_inter_eq_empty]
refine OddOrder.BG.Ch3.S13.sigma_disjoint_of_nonconjugate hG
(hrepsMax Mi hMi) (hrepsMax Mj hMj) ?_
intro hconj
obtain ⟨Z₀, _, huniq⟩ := hreps Mi (hrepsMax Mi hMi)
exact hne ((huniq Mi ⟨hMi, S14.IsConjugateSubgroup.refl Mi⟩).trans
(huniq Mj ⟨hMj, hconj⟩).symm)
/-- **BG Theorem E, the `π(G)` cover** (mmd L4370, clause (a1)): for a system of conjugacy-class
representatives `reps` of the maximal subgroups, a prime `p` divides `|G|` iff `p ∈ σ(Mᵢ)` for some
representative `Mᵢ`. Forward: a prime `p ∣ |G|` is a `σ`-prime of *some* maximal subgroup `M`
(`exists_mem_sigma_of_prime_dvd_card`, BG §1 — a Sylow `p`-subgroup is non-normal, so its normalizer
lies in a maximal `M` with `p ∈ σ(M)`), and `σ` is conjugation-equivariant (`sigma_conj`), so the
representative `Mᵢ` conjugate to `M` also has `p ∈ σ(Mᵢ)`. Reverse: `σ(Mᵢ) ⊆ π(|Mᵢ|)`
(`mem_sigma_iff`) and `|Mᵢ| ∣ |G|` (Lagrange). Together with `sigma_reps_pairwise_disjoint`
(clause (a2)) this is the full `π(G)` partition by the `σ(Mᵢ)` — the partition core of BG Theorem E
(`theoremE_sigma_partition_and_counting`, issue 8019). -/
theorem sigma_reps_prime_cover [Finite G] (hG : OddOrder.BG.IsMinimalSimpleOdd G)
{reps : Set (Subgroup G)}
(_hrepsMax : ∀ Mi ∈ reps, Mi ∈ maximalSubgroups G)
(hreps : ∀ H : Subgroup G, H ∈ maximalSubgroups G →
∃! Mi : Subgroup G, Mi ∈ reps ∧ S14.IsConjugateSubgroup H Mi)
(p : ℕ) :
p ∈ (Nat.card G).primeFactors ↔ ∃ Mi : Subgroup G, Mi ∈ reps ∧ p ∈ OddOrder.BG.Ch3.S10.sigma
Mi := by
constructor
· -- `p ∣ |G|` ⟹ `p ∈ σ(M)` for some maximal `M` ⟹ `p ∈ σ(Mᵢ)` for its representative.
intro hp
have : Fact p.Prime := ⟨Nat.prime_of_mem_primeFactors hp⟩
obtain ⟨M, hMmax, hpM⟩ :=
S14.exists_mem_sigma_of_prime_dvd_card hG (Nat.dvd_of_mem_primeFactors hp)
obtain ⟨Mi, ⟨hMirep, g, hg⟩, _⟩ := hreps M hMmax
exact ⟨Mi, hMirep, hg ▸ OddOrder.BG.Ch3.S10.sigma_conj g hpM⟩
· -- `p ∈ σ(Mᵢ) ⊆ π(|Mᵢ|)` and `|Mᵢ| ∣ |G|`.
rintro ⟨Mi, hMirep, hpMi⟩
have hpMiPF : p ∈ (Nat.card ↥Mi).primeFactors :=
((OddOrder.BG.Ch3.S10.mem_sigma_iff Mi p).mp hpMi).1
exact Nat.mem_primeFactors.mpr ⟨Nat.prime_of_mem_primeFactors hpMiPF,
(Nat.dvd_of_mem_primeFactors hpMiPF).trans (Subgroup.card_subgroup_dvd_card Mi),
(Nat.card_pos).ne'⟩
/-- **`π(M_σ) = σ(M)`** (the prime-factor content of `M_σ` being the `σ(M)`-Hall subgroup,
BG (8.11) /
`Msigma_isHall`): a prime divides `|M_σ|` iff it is a `σ`-prime of `M`. Forward is the `π`-part of
the Hall property (`Msigma_isHall.1`); backward, a `σ`-prime `p` divides `|M|` (`mem_sigma_iff`)
hence
`|G|`, but not the `σ′`-index `[G : M_σ]` (`Msigma_isHall.2`), so by `|G| = |M_σ|·[G : M_σ]` it
divides `|M_σ|`. Bridges the `σ`-stated partition (`sigma_reps_prime_cover`) to the
`π(mainSubgroup ·)` form of `BGTheoremECoverData` via Peterfalvi (8.10)
(`M_s = M_σ`, issue 8020). -/
theorem primeFactors_Msigma_eq_sigma [Finite G] (hG : OddOrder.BG.IsMinimalSimpleOdd G)
{M : Subgroup G} (hM : M ∈ maximalSubgroups G) (p : ℕ) :
p ∈ (Nat.card ↥(OddOrder.BG.Ch3.S10.Msigma M)).primeFactors ↔
p ∈ OddOrder.BG.Ch3.S10.sigma M := by
have hHall := OddOrder.BG.Ch3.S10.Msigma_isHall hG hM
refine ⟨fun hp => hHall.1 p hp, fun hp => ?_⟩
have hpM : p ∈ (Nat.card ↥M).primeFactors := ((OddOrder.BG.Ch3.S10.mem_sigma_iff M p).mp hp).1
have hpprime : p.Prime := Nat.prime_of_mem_primeFactors hpM
have hpG : p ∣ Nat.card G :=
(Nat.dvd_of_mem_primeFactors hpM).trans (Subgroup.card_subgroup_dvd_card M)
have hpnidx : ¬ p ∣ (OddOrder.BG.Ch3.S10.Msigma M).index := fun hdvd =>
hHall.2 p (Nat.mem_primeFactors.mpr ⟨hpprime, hdvd, Subgroup.index_ne_zero_of_finite⟩) hp
rw [← Subgroup.card_mul_index (OddOrder.BG.Ch3.S10.Msigma M)] at hpG
exact Nat.mem_primeFactors.mpr
⟨hpprime, (hpprime.dvd_mul.mp hpG).resolve_right hpnidx, Nat.card_pos.ne'⟩
/-- **A system of conjugacy-class representatives of the maximal subgroups exists** (the `reps`
hypothesis of BG Theorem E, issue 8019). Subgroup conjugacy `IsConjugateSubgroup` is an equivalence
relation (`isConjugateSubgroup_equivalence`) on the finite type of maximal subgroups; taking the
quotient and its canonical representatives `Quotient.out` yields a set `reps` of maximal subgroups
such that every maximal subgroup is conjugate to a *unique* representative. Discharges the `hreps`
input of `sigma_reps_prime_cover` / `sigma_reps_pairwise_disjoint` (and of
`theoremE_sigma_partition_and_counting` / `bgTheoremE_cover_data`), making the `π(G)` partition
unconditional (`exists_reps_sigma_partition`). -/
theorem exists_maximal_conjugacy_reps [Finite G] :
∃ reps : Set (Subgroup G),
(∀ Mi ∈ reps, Mi ∈ maximalSubgroups G) ∧
∀ H : Subgroup G, H ∈ maximalSubgroups G →
∃! Mi : Subgroup G, Mi ∈ reps ∧ S14.IsConjugateSubgroup H Mi := by
classical
-- Conjugacy as a setoid on the (finite) subtype of maximal subgroups.
let S := {M : Subgroup G // M ∈ maximalSubgroups G}
let σ : Setoid S :=
⟨fun a b => S14.IsConjugateSubgroup a.1 b.1,
⟨fun a => S14.IsConjugateSubgroup.refl a.1, fun h => S14.IsConjugateSubgroup.symm h,
fun h₁ h₂ => S14.IsConjugateSubgroup.trans h₁ h₂⟩⟩
-- `reps` = the canonical representatives `Quotient.out` of each conjugacy class.
refine ⟨Set.range (fun c : Quotient σ => (c.out : Subgroup G)), ?_, ?_⟩
· rintro Mi ⟨c, rfl⟩; exact c.out.2
· intro H hH
refine ⟨((Quotient.mk σ ⟨H, hH⟩).out : Subgroup G), ⟨⟨Quotient.mk σ ⟨H, hH⟩, rfl⟩, ?_⟩, ?_⟩
· -- `H` is conjugate to its representative.
have h : S14.IsConjugateSubgroup ((Quotient.mk σ ⟨H, hH⟩).out : Subgroup G) H :=
Quotient.exact (Quotient.out_eq (Quotient.mk σ ⟨H, hH⟩))
exact h.symm
· -- Any representative conjugate to `H` is *the* representative of `H`'s class.
rintro Mi' ⟨⟨c, rfl⟩, hconj'⟩
have hrel : @Setoid.r S σ ⟨H, hH⟩ c.out := hconj'
have heq : Quotient.mk σ ⟨H, hH⟩ = c := by
rw [Quotient.sound hrel]; exact Quotient.out_eq c
rw [heq]
/-- The `𝒞_G(M̃)` cover taken over *all* maximal subgroups equals the cover taken over a set of
conjugacy representatives `reps` (`M̃` is conjugation-equivariant, `𝒞_G` is conjugation-invariant).
The `←` direction needs `reps ⊆ maximalSubgroups`. -/
theorem mem_conjClassSet_Mtilde_maximals_iff_reps [Finite G]
(hG : OddOrder.BG.IsMinimalSimpleOdd G) {reps : Set (Subgroup G)}
(hreps_max : ∀ Mi ∈ reps, Mi ∈ maximalSubgroups G)
(hreps : ∀ H ∈ maximalSubgroups G, ∃ Mi ∈ reps, S14.IsConjugateSubgroup H Mi) (g : G) :
(∃ M' ∈ maximalSubgroups G,
g ∈ conjClassSet (S14.Mtilde hG (S14.genuineSigmaDecomposition hG) M')) ↔
(∃ Mi ∈ reps,
g ∈ conjClassSet (S14.Mtilde hG (S14.genuineSigmaDecomposition hG) Mi)) := by
refine ⟨fun ⟨M', hM'max, hg⟩ => ?_, fun ⟨Mi, hMirep, hg⟩ => ⟨Mi, hreps_max Mi hMirep, hg⟩⟩
obtain ⟨Mi, hMirep, c, hc⟩ := hreps M' hM'max
exact ⟨Mi, hMirep, by
rw [← hc, ← S14.Mtilde_conj_smul, S14.conjClassSet_conj_smul]; exact hg⟩
end OddOrder.BG.Ch4.S16
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/BG/Ch4_FamilyOfMaximal/S16_MainResults/TheoremsAE.lean