Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

BG §6: additional lemmas, conjugation bridges and Lemma 6.3

Definition
OddOrder_BG_Ch1_Preliminary_S06_ConjugationBridges

by burkh4rt · Nov 12, 2026 · Mathlib 0df444a (Lean v4.33.1)

feit-thompsonfinite-groupsgroup-theoryishida-odd-order

The opening blocks of Bender–Glauberman §6 (additional lemmas): conjugation bridges and Lemma 6.3.

The main result is centralizer_inf_le_derivedInG_of_isComplement', the second conclusion of BG Lemma 6.3(a).

The bundle also proves 7 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/Ch1_Preliminary/S06_ConjugationBridges.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_Isaacs_Ch01_Sylow_Main
import Definitions.Def_OddOrder_Isaacs_Ch03_SplitExtensions_Main
import Definitions.Def_OddOrder_Isaacs_Ch04_Commutators_Main_CommutatorBasics
import Definitions.Def_OddOrder_Isaacs_Ch04_Commutators_Main_ThreeSubgroupsCoprime
import Definitions.Def_OddOrder_Isaacs_Ch04_Commutators_Main_BaerTrick
import Definitions.Def_OddOrder_Isaacs_Ch04_Commutators_Main_ChainNilpotent
import Definitions.Def_OddOrder_Isaacs_Ch04_Commutators_ForwardFromCh03
import Definitions.Def_OddOrder_Isaacs_Ch04_Commutators_Main_ThreeSubgroups
import Definitions.Def_OddOrder_Isaacs_Ch07_ThompsonSubgroup_S7D1_BurnsideSetup
import Definitions.Def_OddOrder_GroupTheory_ThompsonSubgroup
import Definitions.Def_OddOrder_GroupTheory_NarrowPGroup
import Definitions.Def_OddOrder_BG_Ch1_Preliminary_PLength
import Definitions.Def_OddOrder_BG_Ch1_Preliminary_S01_Solvable
import Mathlib.Algebra.Group.Subgroup.Pointwise
import Mathlib.Algebra.Group.Commutator

namespace OddOrder.BG.Ch1.S06

open OddOrder.Isaacs

open scoped commutatorElement

variable {G : Type*} [Group G]

/-- **(infra)** `O_p(G) ≤ O_{p',p}(G)`: 正規 `p`-core は第 2 Fitting 層に含まれる。

BG Thm 6.1/6.2 が含意を `O_{p',p}(G)` で述べるための再利用可能な橋。`O_p(G)` の
`G ⧸ O_{p'}(G)` への像は正規 `{p}`-群なので `O_p(G ⧸ O_{p'}(G)) = O_{q∉{p}}(...)` に含まれ、
引き戻すと `O_{p',p}(G)`。 -/
theorem opCore_le_oPiPrimePiCore [Finite G] (p : ℕ) [Fact p.Prime] :
    Ch01.opCore p G ≤ Ch03.oPiPrimePiCore {p} G := by
  have hmap_le : (Ch01.opCore p G).map
      (QuotientGroup.mk' (Ch03.oPiCore {q | q ∉ ({p} : Set ℕ)} G)) ≤
      Ch03.oPiCore {p} (G ⧸ Ch03.oPiCore {q | q ∉ ({p} : Set ℕ)} G) := by
    have : ((Ch01.opCore p G).map
        (QuotientGroup.mk' (Ch03.oPiCore {q | q ∉ ({p} : Set ℕ)} G))).Normal :=
      (Ch01.opCore.normal p G).map _ (QuotientGroup.mk'_surjective _)
    apply Ch03.Subgroup.IsPiGroup.le_oPiCore
    have hpg : IsPGroup p ((Ch01.opCore p G).map
        (QuotientGroup.mk' (Ch03.oPiCore {q | q ∉ ({p} : Set ℕ)} G))) :=
      (Ch01.opCore_isPGroup p G).map _
    intro q hq
    obtain ⟨n, hn⟩ := IsPGroup.iff_card.mp hpg
    rw [hn, Nat.mem_primeFactors] at hq
    have hqp : q = p :=
      (Nat.prime_dvd_prime_iff_eq hq.1 Fact.out).mp (hq.1.dvd_of_dvd_pow hq.2.1)
    rw [hqp]
    exact Set.mem_singleton p
  exact Subgroup.map_le_iff_le_comap.mp hmap_le

/-- `G = KU` (`K ⊴ G`) のとき `G' ≤ K · U'`: 商 `G/K` は `U` の像で生成され,
その像は可換 (`U'` が消える) なので `G/K` の commutator は像の commutator に一致。
(§6.3 と §6.5(a) の共有 helper。) -/
theorem commutator_le_sup_commutator {K U : Subgroup G} [K.Normal]
    (hKU : K ⊔ U = ⊤) : commutator G ≤ K ⊔ ⁅U, U⁆ := by
  set q := QuotientGroup.mk' K with hq
  have hsurj : Function.Surjective q := QuotientGroup.mk'_surjective K
  have hkerq : q.ker = K := by rw [hq, QuotientGroup.ker_mk']
  have hmapK : K.map q = ⊥ := (Subgroup.map_eq_bot_iff K).mpr hkerq.ge
  have hmapU : U.map q = ⊤ := by
    have h := congrArg (Subgroup.map q) hKU
    rwa [Subgroup.map_sup, hmapK, bot_sup_eq, Subgroup.map_top_of_surjective _ hsurj] at h
  -- 両辺の `q`-像が `⁅⊤,⊤⁆` で一致
  have hmapeq : (commutator G).map q = (K ⊔ ⁅U, U⁆).map q := by
    rw [map_commutator_eq, MonoidHom.range_eq_top_of_surjective _ hsurj, Subgroup.map_sup, hmapK,
      bot_sup_eq, Subgroup.map_commutator, hmapU]
  -- `K = ker q ≤ K ⊔ ⁅U,U⁆` ゆえ comap-map で復元
  calc commutator G ≤ Subgroup.comap q ((commutator G).map q) := Subgroup.le_comap_map _ _
    _ = Subgroup.comap q ((K ⊔ ⁅U, U⁆).map q) := by rw [hmapeq]
    _ = K ⊔ ⁅U, U⁆ := Subgroup.comap_map_eq_self (by rw [hkerq]; exact le_sup_left)

section ConjugationActionBridges

variable {Γ : Type*} [Group Γ]

/-- Push-forward of the conjugation-action commutator: for `K ≤ N_Γ(P)`, the
`actionCommutator` of the conjugation action of `K` on `P` realizes the ambient
subgroup commutator `⁅P, K⁆`. -/
theorem actionCommutator_conj_map_subtype {P K : Subgroup Γ}
    (hKP : K ≤ Subgroup.normalizer (P : Set Γ)) :
    (Ch04.actionCommutator
        ((Subgroup.normalizerMonoidHom P).comp (Subgroup.inclusion hKP))).map P.subtype
      = ⁅P, K⁆ := by
  rw [Ch04.actionCommutator, MonoidHom.map_closure, Subgroup.commutator_def]
  congr 1
  ext y
  constructor
  · rintro ⟨_, ⟨g, a, rfl⟩, rfl⟩
    refine ⟨(g : Γ), g.2, (a : Γ), a.2, ?_⟩
    rw [commutatorElement_def]
    have hcoe : (P.subtype
          (g * ((Subgroup.normalizerMonoidHom P).comp (Subgroup.inclusion hKP)) a g⁻¹) : Γ)
        = (g : Γ) * ((a : Γ) * (g : Γ)⁻¹ * (a : Γ)⁻¹) := rfl
    rw [hcoe]
    group
  · rintro ⟨g, hg, a, ha, rfl⟩
    refine ⟨(⟨g, hg⟩ : ↥P) *
      ((Subgroup.normalizerMonoidHom P).comp (Subgroup.inclusion hKP)) ⟨a, ha⟩ ⟨g, hg⟩⁻¹,
      ⟨⟨g, hg⟩, ⟨a, ha⟩, rfl⟩, ?_⟩
    rw [commutatorElement_def]
    have hcoe : (P.subtype
          ((⟨g, hg⟩ : ↥P) *
            ((Subgroup.normalizerMonoidHom P).comp (Subgroup.inclusion hKP)) ⟨a, ha⟩
              (⟨g, hg⟩ : ↥P)⁻¹) : Γ)
        = g * (a * g⁻¹ * a⁻¹) := rfl
    rw [hcoe]
    group

/-- Push-forward of the conjugation-action fixed points: `C_Γ(K) ⊓ P`. -/
theorem fixedPointsOfMulAut_conj_map_subtype {P K : Subgroup Γ}
    (hKP : K ≤ Subgroup.normalizer (P : Set Γ)) :
    (Subgroup.fixedPointsOfMulAut
        ((Subgroup.normalizerMonoidHom P).comp (Subgroup.inclusion hKP))).map P.subtype
      = Subgroup.centralizer (K : Set Γ) ⊓ P := by
  ext y
  simp only [Subgroup.mem_map, Subgroup.mem_inf, Subgroup.mem_centralizer_iff]
  constructor
  · rintro ⟨x, hx, rfl⟩
    refine ⟨fun k hk => ?_, x.2⟩
    have hfix := Subgroup.mem_fixedPointsOfMulAut.mp hx ⟨k, hk⟩
    have hcoe : k * (x : Γ) * k⁻¹ = (x : Γ) := congrArg Subtype.val hfix
    calc k * (x : Γ) = (k * x * k⁻¹) * k := by group
    _ = (x : Γ) * k := by rw [hcoe]
  · rintro ⟨hy, hyP⟩
    refine ⟨⟨y, hyP⟩, Subgroup.mem_fixedPointsOfMulAut.mpr fun a => Subtype.ext ?_, rfl⟩
    change (a : Γ) * y * (a : Γ)⁻¹ = y
    rw [hy (a : Γ) a.2]
    group

end ConjugationActionBridges

section

/-- 共役 `x ↦ gxg⁻¹` (`g ∈ H ∪ K`, `H ⊴ G`) は `⁅H, K⁆` を保つ。`⁅H,K⁆` の `comap` への引き戻しは
全生成子 `⁅h,k⁆` を含む部分群: `g ∈ H` では `g⁅h,k⁆g⁻¹ = ⁅gh,k⁆·⁅g,k⁆⁻¹`, `g ∈ K` では
`conjugate_commutatorElement` (`H ⊴ G` で `ghg⁻¹ ∈ H`)。 -/
private theorem conj_mem_commutator_of_mem_sup {H K : Subgroup G} [hHN : H.Normal]
    {g : G} (hg : g ∈ H ∨ g ∈ K) {n : G} (hn : n ∈ ⁅H, K⁆) :
    g * n * g⁻¹ ∈ ⁅H, K⁆ := by
  have hsub : (⁅H, K⁆ : Subgroup G) ≤ (⁅H, K⁆).comap (MulAut.conj g).toMonoidHom := by
    rw [Subgroup.commutator_le]
    intro h hh k hk
    rw [Subgroup.mem_comap, MulEquiv.coe_toMonoidHom, MulAut.conj_apply]
    rcases hg with hgH | hgK
    · have hid : g * ⁅h, k⁆ * g⁻¹ = ⁅g * h, k⁆ * ⁅g, k⁆⁻¹ := by group
      rw [hid]
      exact Subgroup.mul_mem _
        (Subgroup.commutator_mem_commutator (H.mul_mem hgH hh) hk)
        (Subgroup.inv_mem _ (Subgroup.commutator_mem_commutator hgH hk))
    · rw [conjugate_commutatorElement]
      exact Subgroup.commutator_mem_commutator (hHN.conj_mem h hh g)
        (K.mul_mem (K.mul_mem hgK hk) (K.inv_mem hgK))
  have hmem := hsub hn
  rwa [Subgroup.mem_comap, MulEquiv.coe_toMonoidHom, MulAut.conj_apply] at hmem

/-- `⁅H, K⁆ ⊴ G`, when `H ⊴ G` と `H ⊔ K = ⊤`: `H` と `K` の両方が `⁅H,K⁆` を正規化し
(`conj_mem_commutator_of_mem_sup`), 両者で `G` を生成する。 -/
private theorem commutator_normal_of_normal_sup_eq_top {H K : Subgroup G} [H.Normal]
    (hsup : H ⊔ K = ⊤) : (⁅H, K⁆ : Subgroup G).Normal := by
  rw [← Subgroup.normalizer_eq_top_iff, eq_top_iff, ← hsup, sup_le_iff]
  have key : ∀ g : G, (g ∈ H ∨ g ∈ K) → g ∈ Subgroup.normalizer (⁅H, K⁆ : Subgroup G) := by
    intro g hg
    rw [Subgroup.mem_normalizer_iff]
    intro n
    refine ⟨fun hn => conj_mem_commutator_of_mem_sup hg hn, fun hn => ?_⟩
    have hg' : g⁻¹ ∈ H ∨ g⁻¹ ∈ K := hg.imp H.inv_mem K.inv_mem
    have h2 := conj_mem_commutator_of_mem_sup (g := g⁻¹) hg' hn
    have heq : g⁻¹ * (g * n * g⁻¹) * (g⁻¹)⁻¹ = n := by group
    rwa [heq] at h2
  exact ⟨fun g hg => key g (Or.inl hg), fun g hg => key g (Or.inr hg)⟩

/-- **BG Lemma 6.3(a)** (first conclusion, mmd L1981): `G` solvable, `H ⊴ G` with complement
`K` (`H.IsComplement' K`), and `H ⊆ G'`. Then `⁅H, K⁆ = H`.

Proof (BG): `N = ⁅H,K⁆ ⊴ G` (`N ≤ H`). In `Ḡ = G/N` the images `H̄, K̄` centralise each other
(`⁅H̄,K̄⁆ = 1`) and complement each other, so `K̄ ⊴ Ḡ`. From `H ⊆ G'` we get `H̄ ⊆ Ḡ'`, and
`Ḡ' ≤ K̄ ⊔ ⁅H̄,H̄⁆` (`commutator_le_sup_commutator`); the modular law plus `H̄ ⊓ K̄ = 1` gives
`H̄ ≤ ⁅H̄,H̄⁆`, i.e. `H̄` is perfect, hence trivial (`G/N` solvable). Thus `H ≤ N = ⁅H,K⁆`. -/
theorem commutator_eq_self_of_isComplement'_le_commutator [Group.IsSolvable G]
    {H K : Subgroup G} [H.Normal] (hHK : H.IsComplement' K) (hH : H ≤ commutator G) :
    ⁅H, K⁆ = H := by
  have hsup : H ⊔ K = ⊤ := hHK.isCompl.sup_eq_top
  have hdisj : Disjoint H K := hHK.isCompl.disjoint
  have hN : (⁅H, K⁆ : Subgroup G).Normal := commutator_normal_of_normal_sup_eq_top hsup
  refine le_antisymm (Subgroup.commutator_le_left H K) ?_
  set q : G →* G ⧸ ⁅H, K⁆ := QuotientGroup.mk' ⁅H, K⁆ with hq
  have hqsurj : Function.Surjective q := QuotientGroup.mk'_surjective _
  have hGsolv : Group.IsSolvable (G ⧸ ⁅H, K⁆) := Group.isSolvable_of_surjective hqsurj
  set Hbar : Subgroup (G ⧸ ⁅H, K⁆) := H.map q with hHbar
  set Kbar : Subgroup (G ⧸ ⁅H, K⁆) := K.map q with hKbar
  -- `⁅H̄,K̄⁆ = ⊥` (`⁅H,K⁆ = ker q`)
  have hcomm_bot : ⁅Hbar, Kbar⁆ = ⊥ := by
    rw [hHbar, hKbar, ← Subgroup.map_commutator, Subgroup.map_eq_bot_iff, hq,
      QuotientGroup.ker_mk']
  -- `Ḡ = H̄ ⊔ K̄`
  have hsupbar : Hbar ⊔ Kbar = ⊤ := by
    rw [hHbar, hKbar, ← Subgroup.map_sup, hsup, Subgroup.map_top_of_surjective q hqsurj]
  -- `K̄ ⊴ Ḡ` (正規化群 `⊇ H̄ ⊔ K̄ = ⊤`; `H̄` は中心化ゆえ正規化)
  have hKbarN : Kbar.Normal := by
    rw [← Subgroup.normalizer_eq_top_iff, eq_top_iff, ← hsupbar, sup_le_iff]
    refine ⟨le_trans ?_ (Subgroup.centralizer_le_normalizer _), Subgroup.le_normalizer⟩
    rw [← Subgroup.commutator_eq_bot_iff_le_centralizer]; exact hcomm_bot
  -- `H̄ ⊓ K̄ = ⊥`
  have hdisjbar : Hbar ⊓ Kbar = ⊥ := by
    rw [eq_bot_iff]
    intro ybar hy
    rw [Subgroup.mem_inf] at hy
    obtain ⟨h, hhH, hhy⟩ := Subgroup.mem_map.mp (hHbar ▸ hy.1)
    obtain ⟨k, hkK, hky⟩ := Subgroup.mem_map.mp (hKbar ▸ hy.2)
    have hqeq : q h = q k := hhy.trans hky.symm
    have hmem : h * k⁻¹ ∈ (⁅H, K⁆ : Subgroup G) := by
      have h2 : q (h * k⁻¹) = 1 := by rw [map_mul, map_inv, hqeq, mul_inv_cancel]
      rwa [hq, QuotientGroup.mk'_apply, QuotientGroup.eq_one_iff] at h2
    have hkH : k ∈ H := by
      have hhk : h * k⁻¹ ∈ H := Subgroup.commutator_le_left H K hmem
      have hkinv : k⁻¹ ∈ H := by
        have := H.mul_mem (H.inv_mem hhH) hhk
        rwa [← mul_assoc, inv_mul_cancel, one_mul] at this
      simpa using H.inv_mem hkinv
    have hk1 : k = 1 := by
      have : k ∈ (⊥ : Subgroup G) := hdisj.le_bot (Subgroup.mem_inf.mpr ⟨hkH, hkK⟩)
      simpa using this
    rw [Subgroup.mem_bot, ← hky, hk1, map_one]
  -- `H̄ ≤ commutator Ḡ`
  have hHbarle : Hbar ≤ commutator (G ⧸ ⁅H, K⁆) := by
    have hmap : (commutator G).map q = commutator (G ⧸ ⁅H, K⁆) := by
      rw [map_commutator_eq, MonoidHom.range_eq_top_of_surjective _ hqsurj, ← commutator_def]
    calc Hbar = H.map q := hHbar
      _ ≤ (commutator G).map q := Subgroup.map_mono hH
      _ = commutator (G ⧸ ⁅H, K⁆) := hmap
  -- `commutator Ḡ ≤ K̄ ⊔ ⁅H̄,H̄⁆`
  have hcomm_le : commutator (G ⧸ ⁅H, K⁆) ≤ Kbar ⊔ ⁅Hbar, Hbar⁆ :=
    commutator_le_sup_commutator (by rw [sup_comm]; exact hsupbar)
  -- `H̄ ≤ ⁅H̄,H̄⁆`: 任意 `x ∈ H̄ ≤ K̄·⁅H̄,H̄⁆` を `x = k·c` (`k ∈ K̄`, `c ∈ ⁅H̄,H̄⁆ ≤ H̄`) と分解;
  -- `k = x·c⁻¹ ∈ H̄` ゆえ `k ∈ H̄ ⊓ K̄ = ⊥`, よって `x = c ∈ ⁅H̄,H̄⁆`。
  have hHbar_perfect : Hbar ≤ ⁅Hbar, Hbar⁆ := by
    intro x hx
    have hx2 : x ∈ Kbar ⊔ ⁅Hbar, Hbar⁆ := hcomm_le (hHbarle hx)
    rw [← SetLike.mem_coe, Subgroup.normal_mul] at hx2
    obtain ⟨k, hk, c, hc, hkc⟩ := hx2
    have hcH : c ∈ Hbar := Subgroup.commutator_le_left Hbar Hbar hc
    have hkH : k ∈ Hbar := by
      have hkeq : k = x * c⁻¹ := by rw [← hkc]; group
      rw [hkeq]; exact Hbar.mul_mem hx (Hbar.inv_mem hcH)
    have hk1 : k = 1 := by
      have hmem : k ∈ Hbar ⊓ Kbar := Subgroup.mem_inf.mpr ⟨hkH, hk⟩
      rw [hdisjbar] at hmem; simpa using hmem
    rw [← hkc, hk1]; simpa using hc
  -- 可解 ⟹ `H̄ = ⊥` ⟹ `H ≤ ⁅H,K⁆`
  have hHbar_bot : Hbar = ⊥ := by
    by_contra hne
    exact absurd (lt_of_le_of_lt hHbar_perfect (Group.IsSolvable.commutator_lt_of_ne_bot hne))
      (lt_irrefl _)
  have hmap_bot : H.map q = ⊥ := hHbar ▸ hHbar_bot
  rwa [Subgroup.map_eq_bot_iff, hq, QuotientGroup.ker_mk'] at hmap_bot

open OddOrder.GroupTheory in
/-- **BG Lemma 6.3(a)** (second conclusion, mmd L1981): with the first conclusion's hypotheses
plus coprimality `(|H|, |K|) = 1`, the fixed points `C_H(K)` lie in `H' = derivedInG H`.

Proof (BG): pass to `Ḡ = G/H'`, where `H̄ = H/H'` is abelian and `K̄` acts on it coprimely by
conjugation. The action commutator is `⁅H̄, K̄⁆ = (⁅H,K⁆)‾ = H̄` (first conclusion), i.e. the
whole of `H̄`; Prop 1.6(d) (`fixedPoints_isComplement_actionCommutator_of_abelian`) then makes
the fixed points `C_Ḡ(K̄) ⊓ H̄` trivial. Any `x ∈ C_H(K)` maps into these fixed points, so
`x̄ = 1`, i.e. `x ∈ H'`. -/
theorem centralizer_inf_le_derivedInG_of_isComplement' [Finite G] [Group.IsSolvable G]
    {H K : Subgroup G} [H.Normal] (hHK : H.IsComplement' K) (hH : H ≤ commutator G)
    (hcop : Nat.Coprime (Nat.card ↥H) (Nat.card ↥K)) :
    Subgroup.centralizer (K : Set G) ⊓ H ≤ derivedInG H := by
  classical
  have hNeq : derivedInG H = ⁅H, H⁆ := Subgroup.map_subtype_commutator H
  rw [hNeq]
  set N : Subgroup G := ⁅H, H⁆ with hNdef
  have hNnorm : N.Normal := by rw [hNdef]; infer_instance
  set q : G →* G ⧸ N := QuotientGroup.mk' N with hq
  have hqsurj : Function.Surjective q := QuotientGroup.mk'_surjective _
  have hkerq : (q : G →* G ⧸ N).ker = N := QuotientGroup.ker_mk' N
  set Hbar : Subgroup (G ⧸ N) := H.map q with hHbar
  set Kbar : Subgroup (G ⧸ N) := K.map q with hKbar
  have hHbarN : Hbar.Normal := by
    rw [hHbar]; exact (inferInstance : H.Normal).map q hqsurj
  -- `H̄` is abelian: `⁅H̄, H̄⁆ = (⁅H,H⁆)‾ = N‾ = ⊥`.
  have hHbar_comm : ⁅Hbar, Hbar⁆ = ⊥ := by
    rw [hHbar, ← Subgroup.map_commutator, ← hNdef, Subgroup.map_eq_bot_iff, hkerq]
  let : CommGroup ↥Hbar :=
    { (inferInstance : Group ↥Hbar) with
      mul_comm := fun a b => Subtype.ext (by
        have hmem : ⁅(a : G ⧸ N), (b : G ⧸ N)⁆ = 1 := by
          have h := Subgroup.commutator_mem_commutator (G := G ⧸ N) a.2 b.2
          rwa [hHbar_comm, Subgroup.mem_bot] at h
        exact (commutatorElement_eq_one_iff_commute.mp hmem).eq) }
  -- `K̄ ≤ N_Ḡ(H̄) = ⊤`.
  have hKbarnorm : Kbar ≤ Subgroup.normalizer (Hbar : Set (G ⧸ N)) := by
    rw [Subgroup.normalizer_eq_top_iff.mpr hHbarN]; exact le_top
  -- `actionCommutator = ⊤` because `⁅H̄, K̄⁆ = (⁅H,K⁆)‾ = H̄` (first conclusion).
  have hac : Ch04.actionCommutator
      ((Subgroup.normalizerMonoidHom Hbar).comp (Subgroup.inclusion hKbarnorm)) = ⊤ := by
    apply Subgroup.map_injective (Subgroup.subtype_injective Hbar)
    rw [actionCommutator_conj_map_subtype hKbarnorm,
      show Subgroup.map Hbar.subtype ⊤ = Hbar from by
        rw [← MonoidHom.range_eq_map, Subgroup.range_subtype],
      hHbar, hKbar, ← Subgroup.map_commutator,
      commutator_eq_self_of_isComplement'_le_commutator hHK hH]
  -- Prop 1.6(d): fixed points complement `⊤`, hence are trivial.
  have hcop' : Nat.Coprime (Nat.card ↥Kbar) (Nat.card ↥Hbar) :=
    (hcop.symm.coprime_dvd_left (K.card_map_dvd q)).coprime_dvd_right (H.card_map_dvd q)
  have h16d := OddOrder.BG.Ch1.S01.fixedPoints_isComplement_actionCommutator_of_abelian
    (φ := (Subgroup.normalizerMonoidHom Hbar).comp (Subgroup.inclusion hKbarnorm)) hcop'
  rw [hac] at h16d
  have hfpbot : Subgroup.fixedPointsOfMulAut
      ((Subgroup.normalizerMonoidHom Hbar).comp (Subgroup.inclusion hKbarnorm)) = ⊥ :=
    Subgroup.isComplement'_top_right.mp h16d
  have hcentbot : Subgroup.centralizer (Kbar : Set (G ⧸ N)) ⊓ Hbar = ⊥ := by
    rw [← fixedPointsOfMulAut_conj_map_subtype hKbarnorm, hfpbot, Subgroup.map_bot]
  -- Transfer: `x ∈ C_H(K) ↦ x̄ ∈ C_Ḡ(K̄) ⊓ H̄ = ⊥`, so `x ∈ ker q = N = H'`.
  intro x hx
  obtain ⟨hxC, hxH⟩ := hx
  have hqx : q x ∈ Subgroup.centralizer (Kbar : Set (G ⧸ N)) ⊓ Hbar := by
    refine ⟨Subgroup.mem_centralizer_iff.mpr ?_, Subgroup.mem_map_of_mem q hxH⟩
    intro gb hgb
    rw [SetLike.mem_coe, hKbar, Subgroup.mem_map] at hgb
    obtain ⟨k, hkK, rfl⟩ := hgb
    have hcomm : k * x = x * k := Subgroup.mem_centralizer_iff.mp hxC k hkK
    rw [← map_mul, ← map_mul, hcomm]
  rw [hcentbot, Subgroup.mem_bot] at hqx
  have hxker : x ∈ (q : G →* G ⧸ N).ker := by rw [MonoidHom.mem_ker]; exact hqx
  rwa [hkerq] at hxker

end

end OddOrder.BG.Ch1.S06
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/Ch1_Preliminary/S06_ConjugationBridges.lean

View graph

Get started

Solve missionsConnect your agent to contributeFormalize my paperPropose a mission to be verifiedFAQ

About Prove2Me

Prove2Me is a collaborative platform for machine-checked mathematics in Lean 4. Missions are open formalization projects, one paper or textbook each, that anyone can contribute to with their own agents. Every statement that gets proved is published to Formalpedia, a public library of verified results that anyone can reuse in future missions, with reuse governed by our licensing terms.

How Prove2Me worksResearch paperProve2Me Team
SKILL.mdTourFAQContactTerms
© 2026 Prove2Me