Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

BG §16: Proposition 16.1, the local taxonomy predicates

Definition
OddOrder_BG_Ch4_FamilyOfMaximal_S16_MainResults_LocalTaxonomy

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

feit-thompsonfinite-groupsgroup-theoryishida-odd-order

The shared type I–V predicates and the Proposition 16.1 taxonomy layer of Bender–Glauberman, Local Analysis for the Odd Order Theorem, §16, consumed by the Theorem I/II output file. It consists of lemmas only, including proposition_type_classification_of_inputs.

The bundle consists of lemmas only.

The bundle also proves 16 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/LocalTaxonomy.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_TypeBridges

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]

/-- **§14/§15-independent assembly engine for BG Proposition 16.1** (mmd L4478; the source proof
runs over Theorems A(8)/B(1)(2)(3)(4)/C(1)(2)(3)(10)/D(1), Theorem 15.2(a), and Theorem 15.7(c)).
Proposition 16.1 is the bridge from the BG-local `κ`/`σ`/`M_F` taxonomy to the shared, bundled
Type I--V predicates; the genuinely gated content is the *construction* of each `TypeXData`
structure from the local classification.  This engine isolates those constructions (and the few
structural facts the source proof uses to combine them) as named hypotheses and discharges the full
six-clause conjunction `sorry`-free.  The §15--§16 structural theory has since landed, and the
wrapper `proposition_type_classification` cites it and applies this skeleton (the gated-endpoint
pattern, cf. `theoremD_msigma_conjugacy_and_centralizers_of_inputs`).

The named obligations, with their BG sources:

* the four **forward bridges** `hFI`/`hP2II`/`hP1neIIIIV`/`hP1eqV` — construct the Type
  I/II/III--IV/V data from the local classification (Theorem A(8)+B(1)(2)(3)+15.7(c) for I;
  C(1)(10)+B(1)(4)+A(8) for II; A(8)+Frattini for III/IV; 15.7(c) for V).  These are exactly the
  directions that `theoremI_nilpotentHall_conjugacy_and_type_dichotomy` and
  `theoremII_tame_embedding` consume;
* the four **reverse classifications** `hIF`/`hIIP2`/`hIIIIVP1`/`hVP1` — read off the local type
  from the Peterfalvi data (the `π(W₁) ⊆ κ(M)` argument for `→ M_P`, plus the `κ`/`M_F` refinement;
  Theorem C(2) for `I → M_F`);
* `hP_derived` (**Theorem C(3)**: `M' = U M_σ` for `M ∈ M_P`) and `hF_not_derived` (**Theorem
  A(3)**: `M = U M_σ ⊋ M'` for `M ∈ M_F`), which power clause (e);
* `h152a` (**Theorem 15.2(a)**: `M_F ≠ M_σ ⟹ M ∈ M_P₁`), used for clause (f).

The genuinely *derived* content (not a renamed hypothesis) is clauses (e) and (f), assembled from
the `κ`-trichotomy (`isTypeP_iff_isTypeP1_or_isTypeP2`, `isTypeF_iff_not_isTypeP`,
`not_isTypeP1_and_isTypeP2`) together with the bridges. -/
theorem proposition_type_classification_of_inputs {M : Subgroup G}
    (hFI : S14.IsTypeF M → OddOrder.GroupTheory.IsTypeI M)
    (hP2II : S14.IsTypeP2 M → OddOrder.GroupTheory.IsTypeII M)
    (hP1neIIIIV : S14.IsTypeP1 M → S15.MF M ≠ OddOrder.BG.Ch3.S10.Msigma M →
      OddOrder.GroupTheory.IsTypeIII M ∨ OddOrder.GroupTheory.IsTypeIV M)
    (hP1eqV : S14.IsTypeP1 M → S15.MF M = OddOrder.BG.Ch3.S10.Msigma M →
      OddOrder.GroupTheory.IsTypeV M)
    (hIF : OddOrder.GroupTheory.IsTypeI M → S14.IsTypeF M)
    (hIIP2 : OddOrder.GroupTheory.IsTypeII M → S14.IsTypeP2 M)
    (hIIIIVP1 : (OddOrder.GroupTheory.IsTypeIII M ∨ OddOrder.GroupTheory.IsTypeIV M) →
      S14.IsTypeP1 M ∧ S15.MF M ≠ OddOrder.BG.Ch3.S10.Msigma M)
    (hVP1 : OddOrder.GroupTheory.IsTypeV M →
      S14.IsTypeP1 M ∧ S15.MF M = OddOrder.BG.Ch3.S10.Msigma M)
    (hP_derived : S14.IsTypeP M →
      ∃ U : Subgroup G,
        Ch03.IsHallSubgroup ((S14.kappa M ∪ OddOrder.BG.Ch3.S10.sigma M)ᶜ) (U.subgroupOf M) ∧
        derivedInG M = U ⊔ OddOrder.BG.Ch3.S10.Msigma M)
    (hF_not_derived : S14.IsTypeF M →
      ¬ ∃ U : Subgroup G,
        Ch03.IsHallSubgroup ((S14.kappa M ∪ OddOrder.BG.Ch3.S10.sigma M)ᶜ) (U.subgroupOf M) ∧
        derivedInG M = U ⊔ OddOrder.BG.Ch3.S10.Msigma M)
    (h152a : S15.MF M ≠ OddOrder.BG.Ch3.S10.Msigma M → S14.IsTypeP1 M) :
    (OddOrder.GroupTheory.IsTypeI M ↔ S14.IsTypeF M) ∧
      (OddOrder.GroupTheory.IsTypeII M ↔ S14.IsTypeP2 M) ∧
      ((OddOrder.GroupTheory.IsTypeIII M ∨ OddOrder.GroupTheory.IsTypeIV M) ↔
        S14.IsTypeP1 M ∧ S15.MF M ≠ OddOrder.BG.Ch3.S10.Msigma M) ∧
      (OddOrder.GroupTheory.IsTypeV M ↔
        S14.IsTypeP1 M ∧ S15.MF M = OddOrder.BG.Ch3.S10.Msigma M) ∧
      ((∃ U : Subgroup G,
        Ch03.IsHallSubgroup ((S14.kappa M ∪ OddOrder.BG.Ch3.S10.sigma M)ᶜ)
          (U.subgroupOf M) ∧
        derivedInG M = U ⊔ OddOrder.BG.Ch3.S10.Msigma M) ↔
          ¬ OddOrder.GroupTheory.IsTypeI M) ∧
      (S15.MF M = OddOrder.BG.Ch3.S10.Msigma M ↔
        OddOrder.GroupTheory.IsTypeI M ∨ OddOrder.GroupTheory.IsTypeII M ∨
          OddOrder.GroupTheory.IsTypeV M) := by
  -- `¬ M_P₁ ⟹ M_F = M_σ`, the contrapositive of Theorem 15.2(a).
  have mf_eq_of_not_typeP1 :
      ¬ S14.IsTypeP1 M → S15.MF M = OddOrder.BG.Ch3.S10.Msigma M := by
    intro hnP1
    by_contra hne
    exact hnP1 (h152a hne)
  refine ⟨⟨hIF, hFI⟩, ⟨hIIP2, hP2II⟩, ⟨hIIIIVP1, fun h => hP1neIIIIV h.1 h.2⟩,
    ⟨hVP1, fun h => hP1eqV h.1 h.2⟩, ?_, ?_⟩
  · -- **(e)** `M' = U M_σ ⟺ ¬ Type I`.  Via (a) (`Type I ⟺ M_F`), this is `(∃U …) ⟺ M_P`.
    constructor
    · -- `→`: a Type I `M` would be `M_F` (`hIF`), contradicting Theorem A(3) (`hF_not_derived`).
      intro hex hI
      exact hF_not_derived (hIF hI) hex
    · -- `←`: `¬ Type I ⟹ ¬ M_F ⟹ M_P`, and Theorem C(3) (`hP_derived`) supplies the decomposition.
      intro hnI
      have hnF : ¬ S14.IsTypeF M := fun hF => hnI (hFI hF)
      have hP : S14.IsTypeP M := not_not.mp (by rwa [S14.isTypeF_iff_not_isTypeP] at hnF)
      exact hP_derived hP
  · -- **(f)** `M_F = M_σ ⟺ M` is Type I, II, or V.
    constructor
    · -- `→`: case on the `κ`-trichotomy.  `M_F` (`κ = ∅`) ⟹ I; `M_P₂` ⟹ II; `M_P₁`+`M_F = M_σ` ⟹ V.
      intro heq
      by_cases hP : S14.IsTypeP M
      · rcases S14.isTypeP_iff_isTypeP1_or_isTypeP2.mp hP with hP1 | hP2
        · exact Or.inr (Or.inr (hP1eqV hP1 heq))
        · exact Or.inr (Or.inl (hP2II hP2))
      · exact Or.inl (hFI (S14.isTypeF_iff_not_isTypeP.mpr hP))
    · -- `←`: Type I ⟹ `M_F` ⟹ `¬ M_P₁` ⟹ `M_F = M_σ`; Type II ⟹ `M_P₂` ⟹ `¬ M_P₁` ⟹ `M_F = M_σ`;
      -- Type V carries `M_F = M_σ` directly (`hVP1`).
      rintro (hI | hII | hV)
      · have hnP : ¬ S14.IsTypeP M := S14.isTypeF_iff_not_isTypeP.mp (hIF hI)
        exact mf_eq_of_not_typeP1 (fun hP1 => hnP (S14.isTypeP_of_isTypeP1 hP1))
      · have hP2 := hIIP2 hII
        exact mf_eq_of_not_typeP1 (fun hP1 => S14.not_isTypeP1_and_isTypeP2 ⟨hP1, hP2⟩)
      · exact (hVP1 hV).2

/-- **Proposition 16.1 input `hF_not_derived` / BG Theorem A(3) contrapositive** (mmd L4290): a
type-`F` maximal subgroup `M` (`κ(M) = ∅`) has **no** `(κ ∪ σ)'`-Hall `U` with `M' = U M_σ`.
For type-`F`, `M = U M_σ` for *every* such `U` (`typeP_maximal_eq_kappaHall_sup_U_sup_Msigma` with
`K = ⊥`: the `⊥`-`κ`-Hall witness exists since `κ(M) = ∅`), so `M' = U M_σ` would force `M' = M`,
contradicting the proper derived subgroup `M' < M` of the nontrivial solvable `M`
(`Group.IsSolvable.commutator_lt_top_of_nontrivial`).  This is the `M ∈ ℳ_𝓕 ⟹ M' ⊊ M = U M_σ` half
powering Proposition 16.1 clause (e) (`M' = U M_σ ⟺ ¬ Type I`). -/
theorem typeF_not_exists_hall_derived_eq [Finite G]
    (hG : OddOrder.BG.IsMinimalSimpleOdd G) {M : Subgroup G} (hM : M ∈ maximalSubgroups G)
    (hF : S14.IsTypeF M) :
    ¬ ∃ U : Subgroup G,
      Ch03.IsHallSubgroup ((S14.kappa M ∪ OddOrder.BG.Ch3.S10.sigma M)ᶜ) (U.subgroupOf M) ∧
      derivedInG M = U ⊔ OddOrder.BG.Ch3.S10.Msigma M := by
  rintro ⟨U, hUhall, hM'eq⟩
  have hkappa : S14.kappa M = ∅ := hF
  have : Group.IsSolvable ↥M := hG.isSolvable_of_mem_maximalSubgroups hM
  -- `U ≤ M' ≤ M`.
  have hUM : U ≤ M := by
    have h : U ≤ derivedInG M := by rw [hM'eq]; exact le_sup_left
    exact h.trans (Subgroup.map_subtype_le _)
  -- `⊥` is a `κ(M)`-Hall subgroup of `M` (type-`F`: `κ(M) = ∅`).
  have hK_bot : Ch03.IsHallSubgroup (S14.kappa M) ((⊥ : Subgroup G).subgroupOf M) := by
    rw [Subgroup.bot_subgroupOf, Ch03.IsHallSubgroup.bot_iff]
    intro p _
    rw [hkappa]; exact Set.notMem_empty p
  -- type-`F` decomposition `M = ⊥ ⊔ U ⊔ M_σ = U ⊔ M_σ`.
  have hMeq : M = U ⊔ OddOrder.BG.Ch3.S10.Msigma M := by
    have h := typeP_maximal_eq_kappaHall_sup_U_sup_Msigma hG hM bot_le hUM hK_bot hUhall
    rwa [bot_sup_eq] at h
  -- but `M' < M` (proper derived subgroup of the nontrivial solvable `M`).
  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))
  have : Nontrivial ↥M := (Subgroup.nontrivial_iff_ne_bot M).mpr hMne
  have hlt : derivedInG M < M := by
    rw [derivedInG]
    conv_rhs => rw [← Subgroup.range_subtype M, MonoidHom.range_eq_map]
    rw [Subgroup.map_lt_map_iff_of_injective M.subtype_injective]
    exact Group.IsSolvable.commutator_lt_top_of_nontrivial (G := ↥M)
  exact (ne_of_lt hlt) (hM'eq.trans hMeq.symm)

/-- **Proposition 16.1 input `hP_derived` / BG Theorem C(3)** (mmd L4307): a type-`P` maximal
subgroup `M` has a `(κ ∪ σ)'`-Hall `U` with `M' = U M_σ`.  Construct a `κ(M)`-Hall `K` and a
`(κ ∪ σ)'`-Hall `U` of the solvable `M` (Hall's theorem); `K ≠ ⊥` since `M` is type-`P`
(`isTypeF_of_isHall_kappa_eq_bot` would force `κ(M) = ∅`); then `typeP_hall_derived_eq_and_abelian`
(Theorem 14.7(h) + the three-Hall partition) gives `M' = U M_σ`.  Together with
`typeF_not_exists_hall_derived_eq` this supplies both directions of Proposition 16.1 clause (e)
(`M' = U M_σ ⟺ ¬ Type I`). -/
theorem typeP_exists_hall_derived_eq [Finite G]
    (hG : OddOrder.BG.IsMinimalSimpleOdd G) {M : Subgroup G} (hM : M ∈ maximalSubgroups G)
    (hP : S14.IsTypeP M) :
    ∃ U : Subgroup G,
      Ch03.IsHallSubgroup ((S14.kappa M ∪ OddOrder.BG.Ch3.S10.sigma M)ᶜ) (U.subgroupOf M) ∧
      derivedInG M = U ⊔ OddOrder.BG.Ch3.S10.Msigma M := by
  classical
  have : Group.IsSolvable ↥M := hG.isSolvable_of_mem_maximalSubgroups hM
  -- A `κ(M)`-Hall subgroup `K` of `M` (Hall's theorem in the solvable `M`).
  obtain ⟨K', hK'⟩ := Ch03.hall_E_exists (G := ↥M) (S14.kappa M)
  set K : Subgroup G := K'.map M.subtype with hKdef
  have hKM : K ≤ M := hKdef ▸ Subgroup.map_subtype_le K'
  have hKeq : K.subgroupOf M = K' :=
    hKdef ▸ Subgroup.comap_map_eq_self_of_injective M.subtype_injective K'
  have hK : Ch03.IsHallSubgroup (S14.kappa M) (K.subgroupOf M) := hKeq ▸ hK'
  -- A `(κ ∪ σ)'`-Hall subgroup `U` of `M`.
  obtain ⟨U', hU'⟩ := Ch03.hall_E_exists (G := ↥M)
    ((S14.kappa M ∪ OddOrder.BG.Ch3.S10.sigma M)ᶜ)
  set U : Subgroup G := U'.map M.subtype with hUdef
  have hUM : U ≤ M := hUdef ▸ Subgroup.map_subtype_le U'
  have hUeq : U.subgroupOf M = U' :=
    hUdef ▸ Subgroup.comap_map_eq_self_of_injective M.subtype_injective U'
  have hU : Ch03.IsHallSubgroup ((S14.kappa M ∪ OddOrder.BG.Ch3.S10.sigma M)ᶜ) (U.subgroupOf M) :=
    hUeq ▸ hU'
  -- `K ≠ ⊥`: else `M` would be type-`F` (`κ(M) = ∅`), contradicting `IsTypeP M`.
  have hKne : K ≠ ⊥ := fun h =>
    (S14.isTypeF_iff_not_isTypeP.mp (isTypeF_of_isHall_kappa_eq_bot hKM hK h)) hP
  exact ⟨U, hU, (typeP_hall_derived_eq_and_abelian hG hM hKM hUM hKne hK hU).1⟩

/--
**Type-`F` Frobenius FPF against a `U₀`-element** (mmd L4486, the engine of `isTypeF_of_isTypeI`,
following the Coq `BGsection16` argument): a nontrivial `X ≤ U₀` of the Frobenius complement has
trivial `M_F`-centralizer, `M_F ⊓ C_G(X) = ⊥`. Any `M_F`-element `y` centralizing some
`x ∈ X# ⊆ U₀#`
lifts to `↥(M_F ⊔ U₀)`, where `frobenius_HU0` (kernel `M_F = H`, complement `U₀`) and
`centralizer_complement_le` place it in `U₀`; then `y ∈ M_F ⊓ U₀ = ⊥` (the `complement` field with
`U₀ ≤ U`). This is the `C_H(K) = 1` half of the BG argument, applied to a `U₀`-element `X ⊆ K`
rather
than to the `κ`-Hall `K` itself (which need not lie in `H ⊔ U₀`). -/
theorem typeFData_fitting_inf_centralizer_eq_bot [Finite G]
    {M : Subgroup G} (td : OddOrder.GroupTheory.TypeFData M) {X : Subgroup G}
    (hXU0 : X ≤ td.U0) (hXne : X ≠ ⊥) :
    td.H ⊓ Subgroup.centralizer (X : Set G) = ⊥ := by
  classical
  -- `M_F ⊓ U₀ = ⊥` from the complement (`U₀ ≤ U`, `M_F.subgroupOf M ⊓ U.subgroupOf M = ⊥`).
  have hHU0 : td.H ⊓ td.U0 = ⊥ := by
    rw [eq_bot_iff]
    intro z hz
    obtain ⟨hzH, hzU0⟩ := Subgroup.mem_inf.mp hz
    have hzM : z ∈ M := td.H_le hzH
    have hmem : (⟨z, hzM⟩ : ↥M) ∈ (td.H.subgroupOf M) ⊓ (td.U.subgroupOf M) :=
      Subgroup.mem_inf.mpr ⟨(Subgroup.mem_subgroupOf).mpr hzH,
        (Subgroup.mem_subgroupOf).mpr (td.U0_le hzU0)⟩
    rw [td.complement.disjoint.eq_bot, Subgroup.mem_bot] at hmem
    rw [Subgroup.mem_bot]
    simpa using congrArg Subtype.val hmem
  rw [eq_bot_iff]
  intro y hy
  obtain ⟨hyH, hyC⟩ := Subgroup.mem_inf.mp hy
  obtain ⟨x, hxX, hxne⟩ : ∃ x ∈ X, x ≠ 1 := by
    by_contra hc
    push Not at hc
    exact hXne (eq_bot_iff.mpr fun z hz => Subgroup.mem_bot.mpr (hc z hz))
  have hxU0 : x ∈ td.U0 := hXU0 hxX
  have hyHU0 : y ∈ td.H ⊔ td.U0 := Subgroup.mem_sup_left hyH
  have hxHU0 : x ∈ td.H ⊔ td.U0 := Subgroup.mem_sup_right hxU0
  -- `y` and `x` commute in `G` (`y ∈ C_G(X)`, `x ∈ X`).
  have hcomm : y * x = x * y := (Subgroup.mem_centralizer_iff.mp hyC x hxX).symm
  -- lift to `↥(M_F ⊔ U₀)` and apply `centralizer_complement_le`.
  have hxmem : (⟨x, hxHU0⟩ : ↥(td.H ⊔ td.U0)) ∈ (td.U0).subgroupOf (td.H ⊔ td.U0) :=
    (Subgroup.mem_subgroupOf).mpr hxU0
  have hxne' : (⟨x, hxHU0⟩ : ↥(td.H ⊔ td.U0)) ≠ 1 := by
    rw [ne_eq, Subtype.ext_iff]; simpa using hxne
  have hymem : (⟨y, hyHU0⟩ : ↥(td.H ⊔ td.U0)) ∈
      Subgroup.centralizer ({(⟨x, hxHU0⟩ : ↥(td.H ⊔ td.U0))} : Set ↥(td.H ⊔ td.U0)) := by
    rw [Subgroup.mem_centralizer_singleton_iff]
    apply Subtype.ext
    simp only [Subgroup.coe_mul]
    exact hcomm
  have hyU0' := td.frobenius_HU0.centralizer_complement_le _ hxmem hxne' hymem
  have hyU0 : y ∈ td.U0 := (Subgroup.mem_subgroupOf).mp hyU0'
  rw [← hHU0]
  exact Subgroup.mem_inf.mpr ⟨hyH, hyU0⟩

/-- **Type-`F` `κ`-element placement** (mmd L4486, the Coq `Hall_superset` + `kappa_pi` step; the
last residual of `isTypeF_of_isTypeI`): from `p ∈ κ(M)` and the type-`F` datum, produce a nontrivial
`p`-subgroup `X ≤ U₀` (inside the Frobenius complement, where
`typeFData_fitting_inf_centralizer_eq_bot` applies) together with a `κ(M)`-Hall `K ⊇ X`.

The construction (Coq `BGsection16.v:1031`): `p ∈ κ(M) ⟹ p ∉ σ(M)` (`kappa_subset_sigmaCompl`) and
`p ∈ π(M)`, so `p ∤ |M_F|` (`M_F ⊆ M_σ`) and `p ∣ |U| = [M : M_F]`; since
`exponent U₀ = exponent U`,
`p ∈ π(U₀)`, giving a Sylow `p`-subgroup `X ≤ U₀`, `X ≠ ⊥`.  Then `X` is a `κ`-group, so Hall's
theorem in the solvable `M` (`hall_E_exists` + Hall conjugacy) places it in a `κ(M)`-Hall `K`.  This
was the last residual of `isTypeF_of_isTypeI`, which is now `sorry`-free throughout. -/
theorem typeFData_exists_kappaElement_le_kappaHall [Finite G]
    (hG : OddOrder.BG.IsMinimalSimpleOdd G) {M : Subgroup G}
    (hM : M ∈ maximalSubgroups G) (td : OddOrder.GroupTheory.TypeFData M)
    {p : ℕ} (hp : p ∈ S14.kappa M) :
    ∃ X K : Subgroup G, X ≤ td.U0 ∧ X ≠ ⊥ ∧ X ≤ K ∧ K ≤ M ∧
      Ch03.IsHallSubgroup (S14.kappa M) (K.subgroupOf M) := by
  classical
  have : Group.IsSolvable ↥M := hG.isSolvable_of_mem_maximalSubgroups hM
  obtain ⟨hp_prime, hp_tau, P, hP_elem, hP_le, hP_centr⟩ := hp
  have : Fact p.Prime := ⟨hp_prime⟩
  -- `p ∈ π(M)`: `|P| = p` and `P ≤ M`.
  obtain ⟨_, hPcard⟩ := mem_elemAbelianOfRank.mp hP_elem
  rw [pow_one] at hPcard
  have hp_dvd_M : p ∣ Nat.card ↥M := hPcard ▸ Subgroup.card_dvd_of_le hP_le
  -- `p ∉ σ(M)`.
  have hp_not_sigma : p ∉ OddOrder.BG.Ch3.S10.sigma M :=
    hp_tau.elim (fun h => tau1_subset_sigma_compl M h) (fun h => tau3_subset_sigma_compl M h)
  -- `p ∤ |M_F|` (`M_F ≤ M_σ`, and `M_σ` is `σ`-Hall).
  have hMFMσ : td.H ≤ OddOrder.BG.Ch3.S10.Msigma M :=
    td.H_eq ▸ maxNilpotentNormalHall_le_Msigma hG hM
  have hp_not_dvd_MF : ¬ p ∣ Nat.card ↥td.H := fun hdvd =>
    hp_not_sigma ((OddOrder.BG.Ch3.S10.Msigma_isHall hG hM).primeFactors_card_subset p
      (Nat.mem_primeFactors.mpr ⟨hp_prime, hdvd.trans (Subgroup.card_dvd_of_le hMFMσ),
        Nat.card_pos.ne'⟩))
  -- `p ∣ |U|`: `|M_F| · |U| = |M|`.
  have hcard : Nat.card ↥td.H * Nat.card ↥td.U = Nat.card ↥M := by
    have h := td.complement.card_mul_card
    rwa [Nat.card_congr (Subgroup.subgroupOfEquivOfLe td.H_le).toEquiv,
      Nat.card_congr (Subgroup.subgroupOfEquivOfLe td.U_le).toEquiv] at h
  have hp_dvd_U : p ∣ Nat.card ↥td.U :=
    (hp_prime.dvd_mul.mp (hcard ▸ hp_dvd_M)).resolve_left hp_not_dvd_MF
  -- `p ∣ |U₀|`: a `p`-element of `U` gives `p ∣ exponent U = exponent U₀ ∣ |U₀|`.
  have hp_dvd_U0 : p ∣ Nat.card ↥td.U0 := by
    have : Fintype ↥td.U := Fintype.ofFinite _
    obtain ⟨g, hg⟩ := exists_prime_orderOf_dvd_card (G := ↥td.U) p
      (by rwa [Nat.card_eq_fintype_card] at hp_dvd_U)
    exact (td.exponent_eq ▸ (hg ▸ Monoid.order_dvd_exponent g : p ∣ Monoid.exponent ↥td.U)).trans
      Group.exponent_dvd_nat_card
  -- A `p`-element `g ∈ U₀` generates a nontrivial `p`-subgroup `X = ⟨g⟩ ≤ U₀`.
  have : Fintype ↥td.U0 := Fintype.ofFinite _
  obtain ⟨g, hg⟩ := exists_prime_orderOf_dvd_card (G := ↥td.U0) p
    (by rwa [Nat.card_eq_fintype_card] at hp_dvd_U0)
  have hXMle : (Subgroup.zpowers g).map td.U0.subtype ≤ M :=
    (Subgroup.map_subtype_le _).trans (td.U0_le.trans td.U_le)
  have hXcard : Nat.card ↥((Subgroup.zpowers g).map td.U0.subtype) = p := by
    rw [Nat.card_congr (Subgroup.equivMapOfInjective _ _ td.U0.subtype_injective).symm.toEquiv,
      Nat.card_zpowers, hg]
  -- `X` is a `κ(M)`-group (`|X| = p ∈ κ`), so Hall-D places it in a `κ(M)`-Hall `K`.
  have hX_kappa :
      ∀ q ∈ (Nat.card ↥(((Subgroup.zpowers g).map td.U0.subtype).subgroupOf M)).primeFactors,
      q ∈ S14.kappa M := by
    intro q hq
    rw [Nat.card_congr (Subgroup.subgroupOfEquivOfLe hXMle).toEquiv, hXcard,
      hp_prime.primeFactors, Finset.mem_singleton] at hq
    rw [hq]; exact ⟨hp_prime, hp_tau, P, hP_elem, hP_le, hP_centr⟩
  obtain ⟨K', hK'_hall, hXK'⟩ := Ch03.hall_D (G := ↥M) hX_kappa
  refine ⟨(Subgroup.zpowers g).map td.U0.subtype, K'.map M.subtype,
    Subgroup.map_subtype_le _, ?_, ?_, Subgroup.map_subtype_le _, ?_⟩
  · -- `X ≠ ⊥` (`|X| = p ≠ 1`).
    intro hbot
    rw [hbot, Subgroup.card_bot] at hXcard
    exact hp_prime.ne_one hXcard.symm
  · -- `X ≤ K'.map subtype` from `X.subgroupOf M ≤ K'`.
    rw [← Subgroup.map_subgroupOf_eq_of_le hXMle]
    exact Subgroup.map_mono hXK'
  · -- `IsHallSubgroup κ ((K'.map subtype).subgroupOf M)` reduces to `hK'_hall` on `K'`.
    change Ch03.IsHallSubgroup (S14.kappa M) ((K'.map M.subtype).comap M.subtype)
    rw [Subgroup.comap_map_eq_self_of_injective M.subtype_injective]
    exact hK'_hall

/-- **Proposition 16.1(a), reverse direction — type I ⟹ type `F`** (mmd L4486): a Type I maximal
subgroup `M` has `κ(M) = ∅`.  This is the `hIF` bridge of `proposition_type_classification`, and
together with `isTypeP_of_isTypeNonI` it is everything the FT-critical `not_isTypeI_of_isTypeNonI`
consumes (it lets a non-Type-I `M` be placed in `ℳ_𝓟`, which a Type I `M` cannot also be).

**Proof** (BG L4486, by contradiction): suppose `κ(M) ≠ ∅`, witnessed by `p ∈ κ(M)`.
`typeFData_exists_kappaElement_le_kappaHall` produces a nontrivial `p`-subgroup `X ≤ U₀` and a
`κ(M)`-Hall `K ⊇ X`.  Theorem C(2) (`theoremC_paired_structure`) gives `K* = M_σ ⊓ C_G(K) ≠ ⊥` with
`K* ≤ M_F`.  But `typeFData_fitting_inf_centralizer_eq_bot` forces `M_F ⊓ C_G(X) = ⊥`, and since
`X ≤ K`, `K* ≤ C_G(K) ≤ C_G(X)` together with `K* ≤ M_F` place `K* ≤ M_F ⊓ C_G(X) = ⊥`,
contradicting
`K* ≠ ⊥`. -/
theorem isTypeF_of_isTypeI [Finite G]
    (hG : OddOrder.BG.IsMinimalSimpleOdd G) {M : Subgroup G}
    (hM : M ∈ maximalSubgroups G) (hI : OddOrder.GroupTheory.IsTypeI M) :
    S14.IsTypeF M := by
  rw [S14.isTypeF_iff_not_isTypeP]
  intro hP
  classical
  have : Group.IsSolvable ↥M := hG.isSolvable_of_mem_maximalSubgroups hM
  obtain ⟨td⟩ := hI
  obtain ⟨p, hp⟩ := hP
  -- A nontrivial `p`-subgroup `X ≤ U₀` and a `κ(M)`-Hall `K ⊇ X`.
  obtain ⟨X, K, hXU0, hXne, hXK, hKM, hK⟩ :=
    typeFData_exists_kappaElement_le_kappaHall hG hM td.typeF hp
  -- A `(κ ∪ σ)'`-Hall subgroup `U` of `M`.
  obtain ⟨U', hU'⟩ := Ch03.hall_E_exists (G := ↥M)
    ((S14.kappa M ∪ OddOrder.BG.Ch3.S10.sigma M)ᶜ)
  set U : Subgroup G := U'.map M.subtype with hUdef
  have hUM : U ≤ M := hUdef ▸ Subgroup.map_subtype_le U'
  have hUeq : U.subgroupOf M = U' :=
    hUdef ▸ Subgroup.comap_map_eq_self_of_injective M.subtype_injective U'
  have hU : Ch03.IsHallSubgroup ((S14.kappa M ∪ OddOrder.BG.Ch3.S10.sigma M)ᶜ) (U.subgroupOf M) :=
    hUeq ▸ hU'
  -- `K ≠ ⊥` (since `X ≠ ⊥` and `X ≤ K`).
  have hKne : K ≠ ⊥ := fun h => hXne (le_bot_iff.mp (hXK.trans h.le))
  -- Theorem C(2): `K* = M_σ ⊓ C_G(K)` is nonempty and contained in `M_F`.
  obtain ⟨_, _, hKsne, _, hKsMF, _, _, _, _, _, _, _⟩ :=
    theoremC_paired_structure hG hM hKne hKM hUM hK rfl hU
  -- The type-`F` Frobenius structure gives `M_F ⊓ C_G(X) = ⊥`; with `X ≤ K`, `K* ≤ M_F ⊓ C_G(X)`.
  have hAX := typeFData_fitting_inf_centralizer_eq_bot td.typeF hXU0 hXne
  apply hKsne
  rw [← le_bot_iff, ← hAX, td.typeF.H_eq]
  refine le_inf hKsMF (inf_le_right.trans ?_)
  intro g hg
  rw [Subgroup.mem_centralizer_iff] at hg ⊢
  exact fun x hx => hg x (hXK hx)

/-- **BG Proposition 16.1** (mmd L4478): the §14--§15 local families are exactly
the shared Type I--V maximal-subgroup predicates consumed downstream by Peterfalvi.
Six clauses = mmd (a)-(f): (a) Type I ⟺ `M ∈ ℳ_𝓕`, (b) Type II ⟺ `M ∈ ℳ_𝓟₂`,
(c) Type III/IV ⟺ `M ∈ ℳ_𝓟₁ ∧ M_F ≠ M_σ`, (d) Type V ⟺ `M ∈ ℳ_𝓟₁ ∧ M_F = M_σ`,
(e) `M' = U M_σ ⟺ M` not Type I, (f) `M_F = M_σ ⟺ M` Type I/II/V. -/
theorem proposition_type_classification [Finite G]
    (hG : OddOrder.BG.IsMinimalSimpleOdd G) {M : Subgroup G}
    (hM : M ∈ maximalSubgroups G) :
    (OddOrder.GroupTheory.IsTypeI M ↔ S14.IsTypeF M) ∧
      (OddOrder.GroupTheory.IsTypeII M ↔ S14.IsTypeP2 M) ∧
      ((OddOrder.GroupTheory.IsTypeIII M ∨ OddOrder.GroupTheory.IsTypeIV M) ↔
        S14.IsTypeP1 M ∧ S15.MF M ≠ OddOrder.BG.Ch3.S10.Msigma M) ∧
      (OddOrder.GroupTheory.IsTypeV M ↔
        S14.IsTypeP1 M ∧ S15.MF M = OddOrder.BG.Ch3.S10.Msigma M) ∧
      ((∃ U : Subgroup G,
        Ch03.IsHallSubgroup ((S14.kappa M ∪ OddOrder.BG.Ch3.S10.sigma M)ᶜ)
          (U.subgroupOf M) ∧
        derivedInG M = U ⊔ OddOrder.BG.Ch3.S10.Msigma M) ↔
          ¬ OddOrder.GroupTheory.IsTypeI M) ∧
      (S15.MF M = OddOrder.BG.Ch3.S10.Msigma M ↔
        OddOrder.GroupTheory.IsTypeI M ∨ OddOrder.GroupTheory.IsTypeII M ∨
          OddOrder.GroupTheory.IsTypeV M) := by
  -- Apply the `§14`/`§15`-independent assembly engine.  The proved inputs: `hFI` =
  -- `isTypeI_of_isTypeF` (axiom-clean), `hP2II` = `isTypeII_of_isTypeP2` (axiom-clean), `hIF` =
  -- `isTypeF_of_isTypeI` (BG L4486 reverse, modulo the Frobenius FPF crux), `hP_derived` /
  -- `hF_not_derived` = Theorem C(3)/A(3), `h152a` = Theorem 15.2(a).  The 5 residual bridges
  -- (issue 8015) bottom out on the carrier `W₁`/`U`-Hall characterization (reverse `hIIP2` /
  -- `hIIIIVP1` / `hVP1`) or the type-`P₁` data construction (`hP1neIIIIV` / `hP1eqV` = Peterfalvi
  -- (8.3)/(8.8)).
  refine proposition_type_classification_of_inputs
    ?hFI (fun hP2 => isTypeII_of_isTypeP2 hG hM hP2) ?hP1neIIIIV ?hP1eqV ?hIF ?hIIP2 ?hIIIIVP1 ?hVP1
    (typeP_exists_hall_derived_eq hG hM) (typeF_not_exists_hall_derived_eq hG hM)
    (fun hne => isTypeP1_of_mf_ne_msigma hG hM hne)
  -- `hFI` (Type F ⟹ Type I): the `TypeFData` is built (`isTypeF_groupTheory_of_isTypeF`) and the
  -- `alternative` TI case is proved; only the `¬FittingIsTI` trichotomy (BG 15.7(e)) is residual.
  case hFI => exact isTypeI_of_isTypeF hG hM
  -- `hP1neIIIIV` (Type P₁, `M_F ≠ M_σ` ⟹ Type III/IV): the `TypePData` is fully constructed
  -- (`typePData_of_isTypeP1_mf_ne_msigma`, BG Cor 15.5, `U ≠ ⊥` nilpotent); the sole residual is
  -- the
  -- Peterfalvi (8.7) normalizer `N_G(U) ⊆ M` (isolated in
  -- `isTypeIII_or_IV_of_isTypeP1_mf_ne_msigma`).
  case hP1neIIIIV => exact isTypeIII_or_IV_of_isTypeP1_mf_ne_msigma hG hM
  -- `hP1eqV` (Type P₁, `M_F = M_σ` ⟹ Type V): the type-V `TypePData` is fully constructed
  -- (`typePData_of_isTypeP1_mf_eq_msigma`, `U = ⊥`); the sole residual is the Peterfalvi (8.8)
  -- trichotomy on `M_F` (isolated in `isTypeV_of_isTypeP1_mf_eq_msigma`).
  case hP1eqV => exact isTypeV_of_isTypeP1_mf_eq_msigma hG hM
  -- `hIF` (Type I ⟹ Type F): `isTypeF_of_isTypeI` (BG L4486 reverse direction), modulo the
  -- type-`F` Frobenius FPF crux.
  case hIF => exact isTypeF_of_isTypeI hG hM
  -- `hIIP2` (Type II ⟹ Type P₂): Type II is non-Type-I, so `IsTypeP` (`= P₁ ∨ P₂`,
  -- `isTypeP_of_isTypeNonI`).  The `P₁` branch is excluded: with `M_F = M_σ` it is Type V
  -- (`hP1eqV`, contradicting II via `not_isTypeII_of_isTypeV`); with `M_F ≠ M_σ` it is Type III/IV
  -- (`hP1neIIIIV`, contradicting II via `not_isTypeII_of_isTypeIII_or_IV`).  Hence `P₂`.
  case hIIP2 =>
    intro hII
    rcases isTypeP_iff_isTypeP1_or_isTypeP2.mp (isTypeP_of_isTypeNonI hG hM (Or.inl hII))
      with hP1 | hP2
    · by_cases hmf : S15.MF M = OddOrder.BG.Ch3.S10.Msigma M
      · exact absurd hII
          (not_isTypeII_of_isTypeV (isTypeV_of_isTypeP1_mf_eq_msigma hG hM hP1 hmf))
      · exact absurd hII (not_isTypeII_of_isTypeIII_or_IV hG hM
          (isTypeIII_or_IV_of_isTypeP1_mf_ne_msigma hG hM hP1 hmf))
    · exact hP2
  -- `hIIIIVP1` (Type III/IV ⟹ Type P₁ ∧ `M_F ≠ M_σ`): Type III/IV is non-Type-I, so `IsTypeP`.
  -- The `P₂` branch is excluded (`P₂ ⟹ II`, contradicting III/IV via
  -- `not_isTypeII_of_isTypeIII_or_IV`), giving `P₁`; and `M_F = M_σ` would make it Type V
  -- (`hP1eqV`,
  -- contradicting III/IV via `not_isTypeV_of_isTypeIII_or_IV`), so `M_F ≠ M_σ`.
  case hIIIIVP1 =>
    intro h34
    have hP : S14.IsTypeP M := isTypeP_of_isTypeNonI hG hM
      (h34.elim (fun h => Or.inr (Or.inl h)) (fun h => Or.inr (Or.inr (Or.inl h))))
    have hP1 : S14.IsTypeP1 M := by
      rcases isTypeP_iff_isTypeP1_or_isTypeP2.mp hP with hP1 | hP2
      · exact hP1
      · exact absurd (isTypeII_of_isTypeP2 hG hM hP2) (not_isTypeII_of_isTypeIII_or_IV hG hM h34)
    exact ⟨hP1, fun hmf =>
      not_isTypeV_of_isTypeIII_or_IV h34 (isTypeV_of_isTypeP1_mf_eq_msigma hG hM hP1 hmf)⟩
  -- `hVP1` (Type V ⟹ Type P₁ ∧ `M_F = M_σ`): `M_F = M_σ` from `U = ⊥`
  -- (`mf_eq_msigma_of_typePData_U_eq_bot`); `IsTypeP1` (not `P₂`) because `P₂ ⟹ Type II`
  -- (`isTypeII_of_isTypeP2`) contradicts `Type V` (`not_isTypeII_of_isTypeV`).
  case hVP1 =>
    intro hV
    obtain ⟨dV⟩ := hV
    refine ⟨?_, mf_eq_msigma_of_typePData_U_eq_bot hG hM dV.typeP dV.U_eq_bot⟩
    rcases isTypeP_iff_isTypeP1_or_isTypeP2.mp (isTypeP_of_isTypeV hG hM ⟨dV⟩) with h1 | h2
    · exact h1
    · exact absurd (isTypeII_of_isTypeP2 hG hM h2) (not_isTypeII_of_isTypeV ⟨dV⟩)

/-- **Peterfalvi (8.10)/(8.11), the full `M_s = M_σ` identity** (mmd 04.10:123: "M_s is the group
denoted by M_σ in [BG]"): for a maximal subgroup `M` of its classified Peterfalvi type `τ`, the
"main subgroup" `M_s` (`mainSubgroup`, `= M_F` for I/II/V, `= M'` for III/IV) coincides with BG's
σ-Hall subgroup `M_σ`. Assembled from `proposition_type_classification` (BG Prop 16.1): for I/II/V
it
is clause (f) (`M_F = M_σ ⟺ τ ∈ {I,II,V}`); for III/IV it is clause (c) (`τ ∈ {III,IV} ⟹ M` is type
`P₁`) followed by `isTypeP1_derivedInG_eq_Msigma` (`M' = M_σ`).  The linchpin bridge turning BG's
`M_σ`-stated Theorem E (`sigmaConjugacySaturation_Mtilde_ncard`, `sigma_reps_prime_cover`) into the
`mainSubgroup`-stated `BGTheoremECoverData` (issue 8020). -/
theorem mainSubgroup_eq_Msigma [Finite G]
    (hG : OddOrder.BG.IsMinimalSimpleOdd G) {M : Subgroup G}
    (hM : M ∈ maximalSubgroups G) {tau : PeterfalviType} (htau : HasPeterfalviType tau M) :
    mainSubgroup M tau = OddOrder.BG.Ch3.S10.Msigma M := by
  have hcls := proposition_type_classification hG hM
  cases tau with
  | I => exact hcls.2.2.2.2.2.mpr (Or.inl htau)
  | II => exact hcls.2.2.2.2.2.mpr (Or.inr (Or.inl htau))
  | V => exact hcls.2.2.2.2.2.mpr (Or.inr (Or.inr htau))
  | III => exact isTypeP1_derivedInG_eq_Msigma hG hM (hcls.2.2.1.mp (Or.inl htau)).1
  | IV => exact isTypeP1_derivedInG_eq_Msigma hG hM (hcls.2.2.1.mp (Or.inr htau)).1

/-- **Every maximal subgroup has a Peterfalvi type** (exhaustiveness of the I–V classification): a
maximal subgroup `M` of a minimal simple group of odd order has `HasPeterfalviType τ M` for some
`τ ∈ {I,II,III,IV,V}`. Reads off `proposition_type_classification` (BG Prop 16.1) over the
exhaustive
BG trichotomy `F`/`P₁`/`P₂` (`isTypeF_iff_not_isTypeP`, `isTypeP_iff_isTypeP1_or_isTypeP2`): type
`F`
is I (clause a), `P₂` is II (clause b), `P₁` splits as V if `M_F = M_σ` (clause d) else III/IV
(clause c).  Supplies the `tau`/`typed` fields of `BGTheoremECoverData` (issue 8020). -/
theorem exists_peterfalviType [Finite G] (hG : OddOrder.BG.IsMinimalSimpleOdd G)
    {M : Subgroup G} (hM : M ∈ maximalSubgroups G) :
    ∃ tau : PeterfalviType, HasPeterfalviType tau M := by
  have hcls := proposition_type_classification hG hM
  by_cases hF : S14.IsTypeF M
  · exact ⟨.I, hcls.1.mpr hF⟩
  · have hP : S14.IsTypeP M := by
      by_contra hnP
      exact hF (S14.isTypeF_iff_not_isTypeP.mpr hnP)
    rcases S14.isTypeP_iff_isTypeP1_or_isTypeP2.mp hP with hP1 | hP2
    · by_cases hmf : S15.MF M = OddOrder.BG.Ch3.S10.Msigma M
      · exact ⟨.V, hcls.2.2.2.1.mpr ⟨hP1, hmf⟩⟩
      · rcases hcls.2.2.1.mpr ⟨hP1, hmf⟩ with hIII | hIV
        · exact ⟨.III, hIII⟩
        · exact ⟨.IV, hIV⟩
    · exact ⟨.II, hcls.2.1.mpr hP2⟩

/-- **The signalizer maximal's Fitting equals its `σ`-core** (`M_F = M_σ` for type-`F`/`P₂`
maximals): a maximal subgroup that is BG type `F` or `P₂` has `maxNilpotentNormalHall N = M_σ(N)`.
Type `F` is Peterfalvi I and type `P₂` is Peterfalvi II (`proposition_type_classification` clauses
(a)/(b)), both of which satisfy `M_F = M_σ` (clause (f)).  This is exactly the identity making
Peterfalvi's (8.14) signalizer `R(x) = C_{(N[x])_F}(x)` (Coq `FTsignalizer`, Fitting of the
signalizer
maximal `N[x]`, which is type `F`/`P₂` by `signalizer_structure_of_mem_sigmaSharp`) coincide with
BG's
`R(x) = (N[x])_σ ⊓ C_G(x)` (`Rsub`): `(N[x])_F = (N[x])_σ` (issue 8020). -/
theorem maxNilpotentNormalHall_eq_Msigma_of_isTypeF_or_isTypeP2 [Finite G]
    (hG : OddOrder.BG.IsMinimalSimpleOdd G) {N : Subgroup G} (hN : N ∈ maximalSubgroups G)
    (h : S14.IsTypeF N ∨ S14.IsTypeP2 N) :
    maxNilpotentNormalHall N = OddOrder.BG.Ch3.S10.Msigma N := by
  have hcls := proposition_type_classification hG hN
  rcases h with hF | hP2
  · exact hcls.2.2.2.2.2.mpr (Or.inl (hcls.1.mpr hF))
  · exact hcls.2.2.2.2.2.mpr (Or.inr (Or.inl (hcls.2.1.mpr hP2)))

/-- **Type `F` with no `τ₂`-primes is Frobenius over `M_σ`** (the `∃ U` conclusion of BG
Lemma 14.13(a)): if `κ(M) = ∅`, no prime lies in `τ₂(M)`, and some prime of `π(M)` is
outside `σ(M)` (so the complement is nontrivial), then `M = M_σ ⋊ E` is a Frobenius group.

A fixed point `n ∈ M_σ^#` of `e ∈ E^#` would give a prime `r` of `orderOf e` — necessarily
in `τ₁(M) ∪ τ₃(M)` by the `π(E)`-partition (Lemma 12.1) and `τ₂`-freeness — a rank-one
subgroup `X ≤ ⟨e⟩` with `C_{M_σ}(X) ≠ 1`, i.e. `r ∈ κ(M)` — contradiction. -/
theorem typeF_frobenius_of_esetup [Finite G]
    (hG : OddOrder.BG.IsMinimalSimpleOdd G) {M E E₁ E₂ E₃ : Subgroup G}
    (hM : M ∈ maximalSubgroups G) (hF : S14.IsTypeF M)
    (ht2 : ∀ p : ℕ, p.Prime → p ∉ tau2 M)
    (hsetup : OddOrder.BG.Ch3.S12.SubgroupESetup M E E₁ E₂ E₃)
    (hEne : E.subgroupOf M ≠ ⊥) :
    Subgroup.IsComplement' ((OddOrder.BG.Ch3.S10.Msigma M).subgroupOf M) (E.subgroupOf M) ∧
      OddOrder.Isaacs.Ch06.IsFrobeniusGroup ↥M
        ((OddOrder.BG.Ch3.S10.Msigma M).subgroupOf M) (E.subgroupOf M) := by
  classical
  have hcompl : ((OddOrder.BG.Ch3.S10.Msigma M).subgroupOf M).IsComplement'
      (E.subgroupOf M) := hsetup.isComplement'_subgroupOf
  have hMσhall : Ch03.IsHallSubgroup (OddOrder.BG.Ch3.S10.sigma M)
      ((OddOrder.BG.Ch3.S10.Msigma M).subgroupOf M) :=
    OddOrder.BG.Ch3.S10.Msigma_subgroupOf_isHall_of_isHall
      (OddOrder.BG.Ch3.S10.Msigma_isHall hG hM)
  have hcardE : Nat.card ↥(E.subgroupOf M) =
      ((OddOrder.BG.Ch3.S10.Msigma M).subgroupOf M).index := (hcompl.symm.index_eq_card).symm
  -- `π(E) ⊆ σ(M)ᶜ` (the complement realizes the `σ'`-index).
  have hEpi : ∀ r ∈ (Nat.card ↥(E.subgroupOf M)).primeFactors,
      r ∉ OddOrder.BG.Ch3.S10.sigma M := by
    intro r hr
    rw [hcardE] at hr
    exact hMσhall.2 r hr
  refine ⟨hcompl, ?_⟩
  refine
    { isNormal := ?_
      isComplement := hcompl
      ne_bot_kernel := ?_
      ne_bot_complement := hEne
      conj_frobenius := ?_ }
  · rw [OddOrder.BG.Ch3.S10.Msigma_subgroupOf]; infer_instance
  · intro hbot
    have hMσne : OddOrder.BG.Ch3.S10.Msigma M ≠ ⊥ :=
      OddOrder.BG.Ch3.S10.Msigma_ne_bot hG hM
    exact hMσne (by
      have := congrArg (Subgroup.map M.subtype) hbot
      rwa [Subgroup.map_subgroupOf_eq_of_le (OddOrder.BG.Ch3.S10.Msigma_le M),
        Subgroup.map_bot] at this)
  · -- Frobenius action: no nontrivial fixed points.
    intro a haE ha1 n hnMσ hn1 hfix
    -- `n` commutes with `a` (as elements of `↥M`, hence of `G`).
    have hcomm : Commute (a : G) (n : G) := by
      have hcM : a * n = n * a := mul_inv_eq_iff_eq_mul.mp hfix
      have h2 := congrArg (fun z : ↥M => (z : G)) hcM
      simpa [commute_iff_eq] using h2
    -- a prime `r ∣ orderOf (a : G)`, with an order-`r` power `c` of `a`.
    have haG1 : (a : G) ≠ 1 := fun h => ha1 (by
      apply Subtype.ext
      simpa using h)
    have hordne : orderOf (a : G) ≠ 1 := fun h => haG1 (orderOf_eq_one_iff.mp h)
    obtain ⟨r, hr_prime, hr_dvd⟩ := (orderOf (a : G)).exists_prime_and_dvd hordne
    have : Fact r.Prime := ⟨hr_prime⟩
    have hrcard : r ∣ Nat.card ↥(Subgroup.zpowers (a : G)) := by
      rwa [Nat.card_zpowers]
    obtain ⟨c, hc_ord⟩ := exists_prime_orderOf_dvd_card' (G := ↥(Subgroup.zpowers (a : G))) r
      hrcard
    have hcz : (c : G) ∈ Subgroup.zpowers (a : G) := c.2
    have hc_ordG : orderOf (c : G) = r := by
      rw [← hc_ord]
      exact (orderOf_injective (Subgroup.zpowers (a : G)).subtype
        (Subgroup.zpowers (a : G)).subtype_injective c).symm ▸ rfl
    -- `X = ⟨c⟩ ∈ ℰ_r¹(M)`.
    set X : Subgroup G := Subgroup.zpowers (c : G) with hXdef
    have hXcard : Nat.card ↥X = r := by rw [hXdef, Nat.card_zpowers, hc_ordG]
    have hXelem : X ∈ elemAbelianOfRank G r 1 :=
      mem_elemAbelianOfRank.mpr
        ⟨Subgroup.IsElementaryAbelian.of_card_prime hXcard, by rw [hXcard, pow_one]⟩
    have haM : (a : G) ∈ E := by
      have := haE
      rwa [Subgroup.mem_subgroupOf] at this
    have hXE : X ≤ E := by
      rw [hXdef, Subgroup.zpowers_le]
      exact (Subgroup.zpowers_le.mpr haM) hcz
    have hXM : X ≤ M := hXE.trans hsetup.E_le
    -- `r ∈ π(E) ∖ σ(M) ∖ τ₂(M) ⊆ τ₁(M) ∪ τ₃(M)`.
    have hrE : r ∈ (Nat.card ↥E).primeFactors := by
      refine Nat.mem_primeFactors.mpr ⟨hr_prime, ?_, Nat.card_pos.ne'⟩
      calc r = Nat.card ↥X := hXcard.symm
        _ ∣ Nat.card ↥E := Subgroup.card_dvd_of_le hXE
    have hrτ : r ∈ tau1 M ∪ tau2 M ∪ tau3 M :=
      hsetup.mem_tau_union_of_mem_primeFactors hG hrE
    have hrτ13 : r ∈ tau1 M ∪ tau3 M := by
      rcases hrτ with (h1 | h2) | h3
      · exact Or.inl h1
      · exact absurd h2 (ht2 r hr_prime)
      · exact Or.inr h3
    -- `n` centralizes `X` and lies in `M_σ`, so `r ∈ κ(M)` — against type `F`.
    have hnG : (n : G) ∈ OddOrder.BG.Ch3.S10.Msigma M := by
      have := hnMσ
      rwa [Subgroup.mem_subgroupOf] at this
    have hcn : Commute (c : G) (n : G) := by
      obtain ⟨k, hk⟩ := Subgroup.mem_zpowers_iff.mp hcz
      rw [← hk]; exact hcomm.zpow_left k
    have hnX : (n : G) ∈ Subgroup.centralizer (X : Set G) := by
      rw [Subgroup.mem_centralizer_iff]
      intro y hy
      rw [hXdef] at hy
      obtain ⟨m, rfl⟩ := Subgroup.mem_zpowers_iff.mp hy
      exact (hcn.zpow_left m).eq
    have hne : OddOrder.BG.Ch3.S10.Msigma M ⊓ Subgroup.centralizer (X : Set G) ≠ ⊥ := by
      intro hbot
      have hnG1 : (n : G) ≠ 1 := fun h => hn1 (by
        apply Subtype.ext
        simpa using h)
      exact hnG1 (Subgroup.mem_bot.mp (hbot ▸ Subgroup.mem_inf.mpr ⟨hnG, hnX⟩))
    have hrκ : r ∈ S14.kappa M := ⟨hr_prime, hrτ13, X, hXelem, hXM, hne⟩
    rw [hF] at hrκ
    exact Set.notMem_empty r hrκ

/-- **BG Corollary 15.9** (mmd L4240; Coq `nonFtype_signalizer_base`, BGsection15.v:1399, parts
(a)(b), due to Sibley and Feit--Thompson): the final local landing point for a centralizer escaping
`M`.  Given `x ∈ M_σ^#` with `C_G(x) ⊄ M` and the signalizer neighbour `N ⊇ C_G(x)` that is *not*
type-`F`, then `M` is type-`F`, `F(M)` is not `TI`, `N` is type-`P₂`, and `M = M_σ ⋊ E` is a
Frobenius group with cyclic `σ(M)′`-Hall complement `E`.  The Sibley/Feit--Thompson package used
by §16.

⚠ **Statement corrected (2026-07-07, issue 9017 更新 #19)**: an earlier draft over-specified an
`∃ r prime ∈ τ₂(N), N_G(⟨x⟩) ≤ E ⊓ N` localization conjunct.  That is **unsound** — `x ∈ ⟨x⟩ ≤
N_G(⟨x⟩)` but `x ∈ M_σ` is disjoint from the complement `E` (`M_σ ⊓ E = 1`), so `N_G(⟨x⟩) ⊄ E` —
and it is neither part of Coq 15.9(a)(b) nor consumed downstream (`exists_RData_escape_structure`
discards it).  Dropped.

**Proof plan** (Coq `nonFtype_signalizer_base`): `IsTypeP2 N` from the signalizer dichotomy
`IsTypeF N ∨ IsTypeP2 N` (`signalizer_structure_of_mem_sigmaSharp`) with `¬IsTypeF N`.  Fix a prime
`r ∣ ord(x)` (so `r ∈ τ₂(N)`, signalizer conjunct); the matched `κ(N)` / `(κ∪σ)(N)′`-Hall pair
`K, U` (`typeP2_exists_matched_kappa_hall_pair`) has an `r`-Sylow `R ≤ U` of `r`-rank 2 (noncyclic),
and `N_G(R) ≤ M` (`norm_noncyclic_sigma`, since `r ∈ σ(M)` via `τ₂(N) ∩ π(N) ⊆ σ(M)`).  Then
`typeP2_neighbor_is_typeF_of_mem` (Coq `P2type_signalizer`) with `H := M` gives `IsTypeF M`;
`tau2_transfer_constraint` (Thm 15.8) forces `τ₂(M) = ∅` (else its first conjunct gives `τ₂(N) = ∅`,
contradicting `r ∈ τ₂(N)`); `typeF_frobenius_of_tau2_prime_free` builds the Frobenius `E`; and
`¬FittingIsTI M` follows from `not_fittingIsTI_of_mem_fittingSharp_of_centralizer_not_le`
(`x ∈ M_σ^# ⊆ F(M)^#`, `M_σ` nilpotent for type-`F`). -/
theorem centralizer_escape_final_local [Finite G]
    (hG : OddOrder.BG.IsMinimalSimpleOdd G) {M N : Subgroup G}
    (hM : M ∈ maximalSubgroups G) (hN : N ∈ maximalSubgroups G) {x : G}
    (hx : x ∈ sigmaSharp M) (hesc : ¬ Subgroup.centralizer ({x} : Set G) ≤ M)
    (hNmem : N ∈ maximalSubgroupsContaining (Subgroup.centralizer ({x} : Set G)))
    (hNnotF : ¬ S14.IsTypeF N) :
    S14.IsTypeF M ∧ ¬ FittingIsTI M ∧ S14.IsTypeP2 N ∧
      ∃ E : Subgroup G,
        E ≤ M ∧ Subgroup.IsComplement' ((OddOrder.BG.Ch3.S10.Msigma M).subgroupOf M)
          (E.subgroupOf M) ∧ IsCyclic ↥E ∧
        OddOrder.Isaacs.Ch06.IsFrobeniusGroup ↥M
          ((OddOrder.BG.Ch3.S10.Msigma M).subgroupOf M) (E.subgroupOf M) := by
  classical
  have hx1 : x ≠ 1 := hx.2
  have hxMσ : x ∈ OddOrder.BG.Ch3.S10.Msigma M := hx.1
  -- The escape gives `1 < |𝓜_σ(x)|`, and the signalizer structure yields the unique neighbour `N'`.
  have hgt : 1 < (S14.maximalSigmaSubgroupsOfElement x).ncard := by
    by_contra h
    push Not at h
    exact hesc (centralizer_le_of_maximalSigma_le_one hG hM hxMσ hx1 h)
  obtain ⟨N', hNstruct, -⟩ := signalizer_structure_of_mem_sigmaSharp hG hM hx hgt
  obtain ⟨hNmax', hCN', hRne', hRhall', hxtau2', hNtype', hforall'⟩ := hNstruct
  have hxN' : x ∈ N' := hCN' (Subgroup.mem_centralizer_iff.mpr
    (fun y hy => by rw [Set.mem_singleton_iff.mp hy]))
  -- Our given `N ⊇ C_G(x)` is that unique neighbour (`ℳ(C_G(x)) = {N'}`).
  have hNeq : N = N' := Set.mem_singleton_iff.mp
    ((maximalContaining_centralizer_eq_singleton_of_tau2_element hG hNmax' hxN' hx1 hxtau2' hRne')
      ▸ hNmem)
  subst hNeq
  -- `N` is type-`P₂` (signalizer dichotomy `IsTypeF N ∨ IsTypeP2 N` with `¬IsTypeF N`).
  have hP2N : S14.IsTypeP2 N := hNtype'.resolve_left hNnotF
  -- `IsTypeF M` via `typeP2_neighbor_is_typeF_of_mem` (Coq `P2type_signalizer`); issue 9017 #19.
  -- **Hoisted R-localization setup** (shared by `hFM`, `τ₂(M)=∅`, and the cyclic complement):
  -- `M ∈ 𝓜_σ(x)`; the signalizer structure gives `M ⊓ N` as a `σ(N)'`-complement of `N_σ`.
  have hMσx : M ∈ S14.maximalSigmaSubgroupsOfElement x := ⟨hM, hxMσ⟩
  obtain ⟨-, -, hcompl, -⟩ := hforall' M hMσx
  have hEleN : M ⊓ N ≤ N := inf_le_right
  have hinf : OddOrder.BG.Ch3.S10.Msigma N ⊓ (M ⊓ N) = ⊥ := by
    have hd : Disjoint (OddOrder.BG.Ch3.S10.Msigma N ⊓ (M ⊓ N)) N := by
      rw [← Subgroup.subgroupOf_eq_bot]
      change (OddOrder.BG.Ch3.S10.Msigma N ⊓ (M ⊓ N)).comap N.subtype = ⊥
      rw [Subgroup.comap_inf]; exact disjoint_iff.mp hcompl.disjoint
    have hle : OddOrder.BG.Ch3.S10.Msigma N ⊓ (M ⊓ N) ≤ N := inf_le_right.trans inf_le_right
    rw [← inf_of_le_left hle]; exact disjoint_iff.mp hd
  have hsup : OddOrder.BG.Ch3.S10.Msigma N ⊔ (M ⊓ N) = N := by
    refine le_antisymm (sup_le (OddOrder.BG.Ch3.S10.Msigma_le N) inf_le_right) ?_
    rw [← Subgroup.subgroupOf_eq_top,
      Subgroup.subgroupOf_sup (OddOrder.BG.Ch3.S10.Msigma_le N) inf_le_right]
    exact hcompl.sup_eq_top
  obtain ⟨E₁, E₂, E₃, hsetup⟩ := subgroupESetup_of_complement hG hN hEleN hinf hsup
  obtain ⟨hE1N, hU0E, -, hK₀, hU₀, hU₀ab, hK₀NU₀⟩ :=
    typeP2_matched_kappa_hall_pair_of_esetup hG hN hP2N hsetup
  set U₀ : Subgroup G := E₂ ⊔ E₃ with hU₀def
  have hU0N : U₀ ≤ N := hU0E.trans hEleN
  have hU0M : U₀ ≤ M := hU0E.trans inf_le_left
  -- A prime `r ∣ ord(x)` is a `τ₂(N)`-element (signalizer conjunct) and a `σ(M)`-prime.
  have hcard_eq : Nat.card ↥(Subgroup.closure ({x} : Set G)) = orderOf x := by
    rw [← Subgroup.zpowers_eq_closure, Nat.card_zpowers]
  have hclosne : Nat.card ↥(Subgroup.closure ({x} : Set G)) ≠ 1 := by
    rw [hcard_eq]; exact fun h => hx1 (orderOf_eq_one_iff.mp h)
  obtain ⟨r, hrp, hrdvd⟩ := Nat.exists_prime_and_dvd hclosne
  have : Fact r.Prime := ⟨hrp⟩
  have hr_pi : r ∈ S14.piSet (Subgroup.closure ({x} : Set G)) :=
    Nat.mem_primeFactors.mpr ⟨hrp, hrdvd, Nat.card_pos.ne'⟩
  have hrτ2 : r ∈ tau2 N := hxtau2' r hr_pi
  have hrσM : r ∈ OddOrder.BG.Ch3.S10.sigma M :=
    S14.isPiElement_sigma_of_mem_Msigma hxMσ r
      (by rw [← hcard_eq]; exact Nat.mem_primeFactors.mpr ⟨hrp, hrdvd, Nat.card_pos.ne'⟩)
  have hr2 : pRank ↥N r = 2 := tau2_pRank_eq_two hrτ2
  have hrσ'N : r ∉ OddOrder.BG.Ch3.S10.sigma N := hrτ2.1
  have hrκ'N : r ∉ S14.kappa N := by
    intro hrκ
    have hru : r ∈ tau1 N ∨ r ∈ tau3 N := S14.kappa_subset_tau1_union_tau3 hrκ
    rcases hru with h | h
    · exact absurd (hr2.symm.trans (tau1_pRank_eq_one h)) (by norm_num)
    · exact absurd (hr2.symm.trans (tau3_pRank_eq_one h)) (by norm_num)
  have hrκσ'N : r ∈ (S14.kappa N ∪ OddOrder.BG.Ch3.S10.sigma N)ᶜ := by
    rw [Set.mem_compl_iff, Set.mem_union, not_or]; exact ⟨hrκ'N, hrσ'N⟩
  -- `r ∤ [N : U₀]`, `r ∣ |N|` ⟹ `r ∣ |U₀|` and `pRank_{U₀} r = pRank_N r = 2`.
  have hridx : ¬ r ∣ (U₀.subgroupOf N).index := fun hd =>
    hU₀.2 r (Nat.mem_primeFactors.mpr ⟨hrp, hd, Subgroup.index_ne_zero_of_finite⟩) hrκσ'N
  have hrN : r ∣ Nat.card ↥N :=
    hrdvd.trans (Subgroup.card_dvd_of_le
      (by rw [← Subgroup.zpowers_eq_closure]; exact Subgroup.zpowers_le.mpr hxN'))
  have hrU₀ : r ∈ S14.piSet U₀ := by
    have hlag : Nat.card ↥(U₀.subgroupOf N) * (U₀.subgroupOf N).index = Nat.card ↥N :=
      Subgroup.card_mul_index _
    have hrsub : r ∣ Nat.card ↥(U₀.subgroupOf N) :=
      ((Nat.Prime.dvd_mul hrp).mp (hlag ▸ hrN)).resolve_right hridx
    refine Nat.mem_primeFactors.mpr ⟨hrp, ?_, Nat.card_pos.ne'⟩
    rwa [Nat.card_congr (Subgroup.subgroupOfEquivOfLe hU0N).toEquiv] at hrsub
  -- `R := O_r(U₀)`, a Sylow `r`-subgroup of `U₀`: `≤ M`, an `r`-group, noncyclic (`pRank = 2`).
  have := hU₀ab
  have : Group.IsSolvable ↥N := hG.isSolvable_of_mem_maximalSubgroups hN
  have : Group.IsSolvable ↥U₀ :=
    Group.isSolvable_of_isSolvable_injective (Subgroup.inclusion_injective hU0N)
  obtain ⟨R', hR'⟩ := Ch03.hall_E_exists (G := ↥U₀) ({r} : Set ℕ)
  set R : Subgroup G := R'.map U₀.subtype with hRdef
  have hRU₀ : R ≤ U₀ := Subgroup.map_subtype_le _
  have hRsubOf : R.subgroupOf U₀ = R' := by
    rw [hRdef]; exact Subgroup.comap_map_eq_self_of_injective U₀.subtype_injective R'
  have hRhall : Ch03.IsHallSubgroup ({r} : Set ℕ) (R.subgroupOf U₀) := by
    rw [hRsubOf]; exact hR'
  have hRM : R ≤ M := hRU₀.trans hU0M
  have hRpi : Subgroup.IsPiSubgroup ({r} : Set ℕ) R := by
    intro p hp
    rw [hRdef, Subgroup.card_map_of_injective U₀.subtype_injective] at hp
    exact hR'.1 p hp
  have hRpg : IsPGroup r ↥R := isPGroup_of_isPiSubgroup_singleton hRpi
  have hridxR : ¬ r ∣ (R.subgroupOf U₀).index := fun hd =>
    hRhall.2 r (Nat.mem_primeFactors.mpr ⟨hrp, hd, Subgroup.index_ne_zero_of_finite⟩) rfl
  have hpR2 : pRank ↥R r = 2 := by
    rw [OddOrder.GroupTheory.pRank_eq_of_le_of_not_dvd_index hRU₀ hridxR,
      OddOrder.GroupTheory.pRank_eq_of_le_of_not_dvd_index hU0N hridx, hr2]
  have hRnc : ¬ IsCyclic ↥R := by
    obtain ⟨A, -, hAnc⟩ :=
      exists_isElementaryAbelian_not_isCyclic_of_two_le_pRank (G := ↥R) (p := r) hpR2.ge
    exact fun hRcyc => hAnc (by have := hRcyc; infer_instance)
  have hNRM : Subgroup.normalizer (R : Set G) ≤ M :=
    norm_noncyclic_sigma hG hM hrσM hRpg hRM hRnc
  have hHmem : M ∈ maximalSubgroupsContaining (Subgroup.normalizer (R : Set G)) :=
    mem_maximalSubgroupsContaining.mpr ⟨mem_maximalSubgroups.mp hM, hNRM⟩
  -- Corollary 14.12 (`P2type_signalizer`) with the `P₂` neighbour `N` and `H := M`.
  have hUab : ∀ a ∈ U₀, ∀ b ∈ U₀, a * b = b * a := fun a ha b hb =>
    congrArg Subtype.val (mul_comm' (⟨a, ha⟩ : ↥U₀) (⟨b, hb⟩ : ↥U₀))
  have hFM : S14.IsTypeF M :=
    (S14.typeP2_neighbor_is_typeF_of_mem hG hN hP2N hE1N hU0N hK₀ hU₀ hUab
      hrU₀ hRU₀ hRhall hK₀NU₀ hHmem).1
  -- `¬FittingIsTI M`: `x ∈ M_σ^# ⊆ F(M)^#` (type-`F` ⟹ `M_σ = M_F` nilpotent normal `⊆ F(M)`).
  have hnotTI : ¬ FittingIsTI M := by
    have : Group.IsNilpotent ↥(OddOrder.BG.Ch3.S10.Msigma M) :=
      (maxNilpotentNormalHall_eq_Msigma_iff_isNilpotent hG hM).mp
        (maxNilpotentNormalHall_eq_Msigma_of_isTypeF_or_isTypeP2 hG hM (Or.inl hFM))
    have hMσF : OddOrder.BG.Ch3.S10.Msigma M ≤ fittingInAmbient M :=
      le_fittingInAmbient_of_subgroupOf_normal_of_isNilpotent
        (OddOrder.BG.Ch3.S10.Msigma_le M)
        ((Subgroup.normal_subgroupOf_iff_le_normalizer (OddOrder.BG.Ch3.S10.Msigma_le M)).mpr
          (OddOrder.GroupTheory.le_normalizer_opiCoreInG (OddOrder.BG.Ch3.S10.sigma M) M))
    exact not_fittingIsTI_of_mem_fittingSharp_of_centralizer_not_le hG hM ⟨hMσF hx.1, hx.2⟩ hesc
  -- **`τ₂(M) = ∅`** (Theorem 15.8, `tau2_transfer_constraint` with `H := M`): a prime `p ∈ τ₂(M)`
  -- forces `τ₂(N) = ∅`, contradicting `r ∈ τ₂(N)`.  `Mstar ∈ 𝓜(C_G(E₁))` exists (`E₁ ≠ ⊥`).
  have hE1ne : E₁ ≠ ⊥ := fun h =>
    S14.card_kappaHall_ne_one hP2N.1 hE1N hK₀ (by rw [h, Subgroup.card_bot])
  obtain ⟨Mstar, hMstarmem⟩ :
      (maximalSubgroupsContaining (Subgroup.centralizer (E₁ : Set G))).Nonempty := by
    have : Nontrivial ↥E₁ := (Subgroup.nontrivial_iff_ne_bot E₁).mpr hE1ne
    obtain ⟨⟨e, he⟩, heNe⟩ := exists_ne (1 : ↥E₁)
    have heNe1 : e ≠ 1 := fun h => heNe (by apply Subtype.ext; simpa using h)
    have hlt : Subgroup.centralizer (E₁ : Set G) < ⊤ :=
      lt_of_le_of_lt (Subgroup.centralizer_le (Set.singleton_subset_iff.mpr he))
        (OddOrder.BG.Ch2.S09.centralizer_singleton_lt_top hG heNe1)
    obtain ⟨Mstar, hco, hle⟩ := (eq_top_or_exists_le_coatom _).resolve_left hlt.ne
    exact ⟨Mstar, mem_maximalSubgroupsContaining.mpr ⟨hco, hle⟩⟩
  have htau2M : ∀ p : ℕ, p.Prime → p ∉ tau2 M := fun p hpp hpM =>
    (S15.tau2_transfer_constraint hG hN hP2N hE1N hU0N hK₀ hU₀ hUab hMstarmem
      hrU₀ hRU₀ hRhall hK₀NU₀ hHmem ⟨p, hpp, hpM⟩).1 r hrp hrτ2
  refine ⟨hFM, hnotTI, hP2N, ?_⟩
  -- **Cyclic Frobenius complement** `M = M_σ ⋊ E`, `E = E₁` cyclic: `τ₂(M)=∅ ⟹ E₂=1`
  -- (`E2_eq_bot_of_tau2_eq_empty`), `¬FittingIsTI ⟹ E₃=1` (`E3_eq_bot_of_not_fittingIsTI`),
  -- `E₁` cyclic (`E1_isCyclic`); `E ≠ ⊥` is free (`SubgroupESetup.E_ne_bot`), Frobenius via
  -- `typeF_frobenius_of_esetup`.  BG Corollary 15.9(b) (Coq `nonFtype_signalizer_base` `cycE`).
  obtain ⟨EM, EM₁, EM₂, EM₃, hsetupM⟩ := exists_subgroupESetup hG hM
  have hEM2 : EM₂ = ⊥ := S15.E2_eq_bot_of_tau2_eq_empty hsetupM htau2M
  have hEM3 : EM₃ = ⊥ := S15.E3_eq_bot_of_not_fittingIsTI hG hM hnotTI hsetupM
  have hEMeq : EM = EM₁ := by rw [hsetupM.eq_sup hG, hEM2, hEM3, sup_bot_eq, sup_bot_eq]
  have hEMcyc : IsCyclic ↥EM := by rw [hEMeq]; exact hsetupM.E1_isCyclic hG
  have hEMne : EM.subgroupOf M ≠ ⊥ := fun hbot =>
    OddOrder.BG.Ch3.S12.SubgroupESetup.E_ne_bot hG hsetupM
      (by rw [← inf_of_le_left hsetupM.E_le]
          exact disjoint_iff.mp (Subgroup.subgroupOf_eq_bot.mp hbot))
  obtain ⟨hIsCompl, hFrob⟩ := typeF_frobenius_of_esetup hG hM hFM htau2M hsetupM hEMne
  exact ⟨EM, hsetupM.E_le, hIsCompl, hEMcyc, hFrob⟩

/-- **A `τ₂(N)`-element lies outside `N_σ`.**  Every prime of `π(⟨x⟩)` is a `τ₂(N)`-prime, hence a
`σ(N)′`-prime (`mem_tau2_iff`), while `N_σ` is a `σ(N)`-group.  Extracted from the Theorem D(4)
proof, which needs it for the `∖ N_σ` half of its fourth component. -/
theorem notMem_Msigma_of_forall_mem_tau2 [Finite G] {N : Subgroup G} {x : G} (hx1 : x ≠ 1)
    (hxtau2 : ∀ p ∈ S14.piSet (Subgroup.closure ({x} : Set G)), p ∈ tau2 N) :
    x ∉ (OddOrder.BG.Ch3.S10.Msigma N : Set G) := by
  intro hxMσN
  obtain ⟨p, hpp, hpdvd⟩ := (orderOf x).exists_prime_and_dvd
    (fun h => hx1 (orderOf_eq_one_iff.mp h))
  have hpπx : p ∈ S14.piSet (Subgroup.closure ({x} : Set G)) := by
    rw [S14.piSet, ← Subgroup.zpowers_eq_closure, Nat.card_zpowers]
    exact Nat.mem_primeFactors.mpr ⟨hpp, hpdvd, (orderOf_pos x).ne'⟩
  have hpτ2 : p ∈ tau2 N := hxtau2 p hpπx
  have hpσN : p ∈ OddOrder.BG.Ch3.S10.sigma N :=
    OddOrder.BG.Ch3.S10.Msigma_isPiGroup N p (Nat.mem_primeFactors.mpr ⟨hpp,
      hpdvd.trans ((OddOrder.BG.Ch3.S10.Msigma N).orderOf_dvd_natCard
        (SetLike.mem_coe.mp hxMσN)), Nat.card_pos.ne'⟩)
  exact ((mem_tau2_iff N p).mp hpτ2).1 hpσN

/-- **BG Theorem D(4), the escape structure** (the `hD4` conjunct of
`theoremD_msigma_conjugacy_and_centralizers`): for `x ∈ M_σ^#` whose centralizer escapes `M`, the
normal-complement data `R(x)` exists and is attached to a *unique* maximal `N ⊇ C_G(x)` of type `F`
or `P₂`.  Assembled from the signalizer structure (`signalizer_structure_of_mem_sigmaSharp`), the
`τ₂`-element centralizer uniqueness (`maximalContaining_centralizer_eq_singleton_of_tau2_element`,
giving `ℳ(C_G(x)) = {N}`), `RData_of_inputs`,
`maxNilpotentNormalHall_eq_Msigma_of_isTypeF_or_isTypeP2`,
and the `P₂`-escape package (`centralizer_escape_final_local`). -/
theorem exists_RData_escape_structure [Finite G]
    (hG : OddOrder.BG.IsMinimalSimpleOdd G) {M : Subgroup G} (hM : M ∈ maximalSubgroups G)
    {x : G} (hx : x ∈ sigmaSharp M) (hesc : ¬ Subgroup.centralizer ({x} : Set G) ≤ M) :
    ∃ R : Subgroup G,
      RData M x R ∧
      ∃! N : Subgroup G,
        N ∈ maximalSubgroupsContaining (Subgroup.centralizer ({x} : Set G)) ∧
        R = OddOrder.BG.Ch3.S10.Msigma N ⊓ Subgroup.centralizer ({x} : Set G) ∧
        S15.MF N = OddOrder.BG.Ch3.S10.Msigma N ∧
        x ∈ ASet N ⊤ \ (OddOrder.BG.Ch3.S10.Msigma N : Set G) ∧
        (S14.IsTypeF N ∨ S14.IsTypeP2 N) ∧
        Subgroup.IsComplement' ((M ⊓ N).subgroupOf N)
          ((OddOrder.BG.Ch3.S10.Msigma N).subgroupOf N) ∧
        (S14.IsTypeP2 N →
          S14.IsTypeF M ∧ ¬ S15.FittingIsTI M ∧
            ∃ E : Subgroup G,
              E ≤ M ∧ IsCyclic ↥E ∧
              Subgroup.IsComplement' ((OddOrder.BG.Ch3.S10.Msigma M).subgroupOf M)
                (E.subgroupOf M) ∧
              OddOrder.Isaacs.Ch06.IsFrobeniusGroup ↥M
                ((OddOrder.BG.Ch3.S10.Msigma M).subgroupOf M) (E.subgroupOf M)) := by
  have hx1 : x ≠ 1 := hx.2
  have hxMσ : x ∈ OddOrder.BG.Ch3.S10.Msigma M := hx.1
  -- `|𝓜_σ(x)| > 1` from the escape.
  have hgt : 1 < (S14.maximalSigmaSubgroupsOfElement x).ncard := by
    by_contra h
    push Not at h
    exact hesc (centralizer_le_of_maximalSigma_le_one hG hM hxMσ hx1 h)
  -- The signalizer neighbour `N`.
  obtain ⟨N, hNstruct, -⟩ := signalizer_structure_of_mem_sigmaSharp hG hM hx hgt
  obtain ⟨hNmax, hCN, hRne, hRhall, hxtau2, hNtype, hforall⟩ := hNstruct
  have hxN : x ∈ N := hCN (Subgroup.mem_centralizer_iff.mpr
    (fun y hy => by rw [Set.mem_singleton_iff.mp hy]))
  -- Uniqueness: `ℳ(C_G(x)) = {N}` (Cor 14.3 τ₂-uniqueness applied to `N`).
  have hMC : maximalSubgroupsContaining (Subgroup.centralizer ({x} : Set G)) = {N} :=
    maximalContaining_centralizer_eq_singleton_of_tau2_element hG hNmax hxN hx1 hxtau2 hRne
  have hxM_mem : x ∈ M := OddOrder.BG.Ch3.S10.Msigma_le M hxMσ
  obtain ⟨-, -, hMcompl, hMsharp⟩ := hforall M ⟨hM, hxMσ⟩
  -- `x ∉ M_σ(N)` since `x` is a `τ₂(N)`-element (`σ(N)′`).
  have hxnotMσN : x ∉ (OddOrder.BG.Ch3.S10.Msigma N : Set G) :=
    notMem_Msigma_of_forall_mem_tau2 hx1 hxtau2
  refine ⟨OddOrder.BG.Ch3.S10.Msigma N ⊓ Subgroup.centralizer ({x} : Set G),
    RData_of_inputs hG hM hxMσ hx1 hCN hMsharp hRhall
      (signalizer_centralizer_isComplement hMcompl hCN hxM_mem),
    N, ⟨?_, rfl, ?_, ⟨?_, hxnotMσN⟩, hNtype, hMcompl.symm, ?_⟩, ?_⟩
  · rw [hMC]; rfl
  · exact maxNilpotentNormalHall_eq_Msigma_of_isTypeF_or_isTypeP2 hG hNmax hNtype
  · -- `x ∈ ASet N ⊤ = hatMsigma N`.
    show x ∈ ASet N ⊤
    refine ⟨⟨hxN, hRne⟩, ?_⟩
    simp
  · -- `IsTypeP2 N → IsTypeF M ∧ ¬FittingIsTI M ∧ Frobenius`.
    intro hP2
    have hNnotF : ¬ S14.IsTypeF N := fun hF => S14.not_isTypeP_and_isTypeF ⟨hP2.1, hF⟩
    have hNmem : N ∈ maximalSubgroupsContaining (Subgroup.centralizer ({x} : Set G)) := by
      rw [hMC]; rfl
    obtain ⟨hFM, hnotTI, -, E, hEM, hEcompl, hEcyc, hEfrob⟩ :=
      centralizer_escape_final_local hG hM hNmax hx hesc hNmem hNnotF
    exact ⟨hFM, hnotTI, E, hEM, hEcyc, hEcompl, hEfrob⟩
  · rintro N' ⟨hN'mem, -⟩
    rw [hMC] at hN'mem
    exact hN'mem

/-- **BG Theorem D** (mmd L4317, recovered tail L4368): conjugacy and centralizer
control for `M_sigma`, including the `R(x)` normal complement, its sharply transitive
action, and the unique maximal subgroup attached to escaping centralizers.

The final conjunct is the recovered BG D(4) tail: `M ∩ N` complements `N_sigma` in
`N`, and if `N` is type `P2` then `M` is type `F`, Frobenius with cyclic complement,
and `M_F` is not TI. -/
theorem theoremD_msigma_conjugacy_and_centralizers [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) ∧
      (∀ g : G, g ∉ M → IsCyclic ↥(OddOrder.BG.Ch3.S10.Msigma M ⊓ (MulAut.conj g • M))) ∧
      (∀ x : G, x ∈ sigmaSharp M → ∃ R : Subgroup G, RData M x R) ∧
      (∀ x : G, x ∈ sigmaSharp M → ¬ Subgroup.centralizer ({x} : Set G) ≤ M →
        ∃ R : Subgroup G,
          RData M x R ∧
          ∃! N : Subgroup G,
            N ∈ maximalSubgroupsContaining (Subgroup.centralizer ({x} : Set G)) ∧
            R = OddOrder.BG.Ch3.S10.Msigma N ⊓ Subgroup.centralizer ({x} : Set G) ∧
            S15.MF N = OddOrder.BG.Ch3.S10.Msigma N ∧
            x ∈ ASet N ⊤ \ (OddOrder.BG.Ch3.S10.Msigma N : Set G) ∧
            (S14.IsTypeF N ∨ S14.IsTypeP2 N) ∧
            Subgroup.IsComplement' ((M ⊓ N).subgroupOf N)
              ((OddOrder.BG.Ch3.S10.Msigma N).subgroupOf N) ∧
            (S14.IsTypeP2 N →
              S14.IsTypeF M ∧ ¬ S15.FittingIsTI M ∧
                ∃ E : Subgroup G,
                  E ≤ M ∧ IsCyclic ↥E ∧
                  Subgroup.IsComplement' ((OddOrder.BG.Ch3.S10.Msigma M).subgroupOf M)
                    (E.subgroupOf M) ∧
                  OddOrder.Isaacs.Ch06.IsFrobeniusGroup ↥M
                    ((OddOrder.BG.Ch3.S10.Msigma M).subgroupOf M) (E.subgroupOf M))) := by
  refine ⟨msigma_fusion_control hG hM, fun g hg => Msigma_inf_conj_isCyclic hG hM hg, ?_, ?_⟩
  · exact exists_RData_of_mem_sigmaSharp hG hM
  · exact fun x hx hesc => exists_RData_escape_structure hG hM hx hesc

/-- **Type I and non-Type-I are mutually exclusive** (corollary of Proposition
16.1(a)–(d)).  A maximal subgroup of a minimal simple group of odd order that is
Type I cannot also be one of Types II–V.

The proof reads the type dictionary of `proposition_type_classification`: clause
(a) says Type I `⟺ M ∈ ℳ_𝓕` (`S14.IsTypeF`), while clauses (b)–(d) place each of
Types II–V in `ℳ_𝓟` (`S14.IsTypeP`) — Type II in `ℳ_𝓟₂`, Types III/IV/V in
`ℳ_𝓟₁`.  Since `ℳ_𝓕` and `ℳ_𝓟` are complementary
(`S14.isTypeF_iff_not_isTypeP`), the two cannot coincide.

Used by `OddOrder.section16MaximalPair_of_isMinimalSimpleOdd` to discharge the
all-Type-I branch of Peterfalvi (8.8): the case-(b) witness of (12.17) is a
non-Type-I maximal subgroup, contradicting "every maximal subgroup is Type I". -/
theorem not_isTypeI_of_isTypeNonI [Finite G] (hG : OddOrder.BG.IsMinimalSimpleOdd G)
    {M : Subgroup G} (hM : M ∈ maximalSubgroups G)
    (hNonI : OddOrder.GroupTheory.IsTypeNonI M) :
    ¬ OddOrder.GroupTheory.IsTypeI M := by
  intro hI
  -- Type I forces `κ(M) = ∅` (`isTypeF_of_isTypeI`), i.e. `M ∉ ℳ_𝓟`; but every non-Type-I `M` is
  -- type `P` (`isTypeP_of_isTypeNonI`).  This routes around the other five
  -- `proposition_type_classification` bridges, leaving the FT-critical surface gated on the single
  -- `hIF` crux (`typeF_mf_inf_centralizer_kappaHall_eq_bot`).
  exact (S14.isTypeF_iff_not_isTypeP.mp (isTypeF_of_isTypeI hG hM hI))
    (isTypeP_of_isTypeNonI hG hM hNonI)

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/LocalTaxonomy.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