Peterfalvi (8.15) claim 1: support for the book-literal type set
DefinitionOddOrder_Peterfalvi_S10_TypePSupportA0feit-thompsonfinite-groupsgroup-theoryishida-odd-order
Verifies the Dade (2.2) support hypotheses for the type- support (Peterfalvi, Character Theory for the Odd Order Theorem, §8, (8.10)/(8.15), pp. 47-48). The bundle consists of lemmas only.
The bundle also proves 18 supporting lemmas used by later statements.
Definition code
/-
Copyright (c) 2026 Yawara Ishida. All rights reserved.
Released under Apache 2.0 license as described in the file LICENSE.
Authors: Yawara Ishida
-/
-- From github.com/yawara/odd-order @ 20a6fccbb63c (Apache-2.0; LICENSE at that commit):
-- OddOrder/Peterfalvi/S10_TypePSupportA0.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_TheoremIIPackaging
import Definitions.Def_OddOrder_Peterfalvi_S10_MinimalSimpleStructure
namespace OddOrder.Peterfalvi.S10
open OddOrder.GroupTheory
variable {G : Type*} [Group G]
/-- `A(S) ⊆ A₀(S)`: the honest type-`P₂` support is contained in its `A₀`-completion. -/
theorem typePACore_subset_A0Set {M : Subgroup G} (data : TypePData M) :
S10.typePACore M ⊆ S10.typePACore0 M data :=
Set.subset_union_left
/-- `W ≤ M` for type-`P` data: the cyclic factor `W = W₁ ⊔ W₂` lands in `M`
(`W₁ ≤ M`; `W₂ ≤ H ⊓ M'' ≤ H ≤ M' ≤ M`). -/
theorem typePData_W_le_M {M : Subgroup G} (data : TypePData M) :
(data.W : Subgroup G) ≤ M := by
rw [data.W_eq]
exact sup_le data.W1_le
(data.W2_le.trans (le_trans inf_le_left
(data.H_le.trans (Subgroup.map_subtype_le _))))
/-- `V_S ⊆ M`: the exceptional regular set lands in `M` (it is `⊆ W ≤ M`). -/
theorem typePV_subset_M {M : Subgroup G} (data : TypePData M) :
typePV M data ⊆ (M : Set G) := fun _ hv =>
typePData_W_le_M data (((Set.mem_sdiff _).mp hv).1)
/-- **`V^M`-points lie in BG's `σ`-saturation `hatMsigma M`** (issue 9076 piece 4c-2b″, step 2a of
the `S10.typePACore0 ⊆ A0Set` bridge). A `V = typePV`-point `v = a·b` (`a ∈ W₁`, `b ∈ W₂`, the
`W = W₁ ⊔ W₂` factorization) has a **nontrivial** `W₂`-component `b ≠ 1` (else `v = a ∈ W₁`), and
`W₂ ≤ H = M_F ≤ M_σ` (`maxNilpotentNormalHall_le_Msigma`); since `W` is cyclic (abelian) `v`
commutes
with `b`, so `b ∈ M_σ ⊓ C_G(v)` witnesses `M_σ ⊓ C_G(v) ≠ ⊥`. This is the `V`-side half of showing
`A₀(S) ⊆ hatMsigma M`, feeding the BG §16 Theorem-II tame conjugation. -/
theorem typePV_subset_hatMsigma [Finite G] (hG : OddOrder.BG.IsMinimalSimpleOdd G)
{M : Subgroup G} (hM : M ∈ maximalSubgroups G) (data : TypePData M) :
typePV M data ⊆ OddOrder.BG.Ch4.S16.hatMsigma M := by
intro v hv
simp only [typePV, Set.mem_sdiff, Set.mem_union, SetLike.mem_coe, not_or] at hv
obtain ⟨hvW, hvnW1, _hvnW2⟩ := hv
have hcyc : IsCyclic ↥data.W := data.W_cyclic
let : CommGroup ↥data.W := hcyc.commGroup
have hW1le : data.W1 ≤ data.W := data.W_eq ▸ le_sup_left
have hW2le : data.W2 ≤ data.W := data.W_eq ▸ le_sup_right
have hWM : (data.W : Subgroup G) ≤ M := by
rw [data.W_eq]
exact sup_le data.W1_le
(data.W2_le.trans (inf_le_left.trans (data.H_le.trans (Subgroup.map_subtype_le _))))
-- `v = a·b` with `a ∈ W₁`, `b ∈ W₂` (cyclic `W = W₁ ⊔ W₂`).
have hsup : data.W1.subgroupOf data.W ⊔ data.W2.subgroupOf data.W = ⊤ := by
rw [← Subgroup.subgroupOf_sup hW1le hW2le, ← data.W_eq, Subgroup.subgroupOf_self]
have hvmem : (⟨v, hvW⟩ : ↥data.W) ∈
data.W1.subgroupOf data.W ⊔ data.W2.subgroupOf data.W := by
rw [hsup]; exact Subgroup.mem_top _
rw [Subgroup.mem_sup] at hvmem
obtain ⟨a, ha, b, hb, hab⟩ := hvmem
have haW1 : ((a : ↥data.W) : G) ∈ data.W1 := Subgroup.mem_subgroupOf.mp ha
have hbW2 : ((b : ↥data.W) : G) ∈ data.W2 := Subgroup.mem_subgroupOf.mp hb
have habG : ((a : ↥data.W) : G) * ((b : ↥data.W) : G) = v := by
have := congrArg (Subtype.val) hab; simpa using this
-- `b ≠ 1` (else `v = a ∈ W₁`).
have hb1 : ((b : ↥data.W) : G) ≠ 1 := by
intro h; exact hvnW1 (by rw [← habG, h, mul_one]; exact haW1)
-- `b ∈ M_σ` (`W₂ ≤ H = M_F ≤ M_σ`).
have hbMσ : ((b : ↥data.W) : G) ∈ OddOrder.BG.Ch3.S10.Msigma M := by
refine (data.W2_le.trans (inf_le_left.trans ?_)) hbW2
rw [data.H_eq]; exact OddOrder.BG.Ch4.S15.maxNilpotentNormalHall_le_Msigma hG hM
-- `v` commutes with `b` (both in the abelian `W`).
have hcomm : v * ((b : ↥data.W) : G) = ((b : ↥data.W) : G) * v := by
have h := congrArg Subtype.val (mul_comm (⟨v, hvW⟩ : ↥data.W) b)
simpa using h
refine ⟨hWM hvW, fun hbot => hb1 ?_⟩
have hbInf : ((b : ↥data.W) : G) ∈
OddOrder.BG.Ch3.S10.Msigma M ⊓ Subgroup.centralizer ({v} : Set G) :=
⟨hbMσ, Subgroup.mem_centralizer_singleton_iff.mpr hcomm.symm⟩
rw [hbot, Subgroup.mem_bot] at hbInf; exact hbInf
/-- `A₀(S) ⊆ M`: both the `A(S)`-part (`S10.typePACore_subset`) and the `V^S`-part
(`conjClassSetIn` of the `⊆ M` regular set) land in `M`. -/
theorem typePACore0_subset {M : Subgroup G} (data : TypePData M) :
S10.typePACore0 M data ⊆ (M : Set G) := by
rintro x (hx | hx)
· exact S10.typePACore_subset hx
· exact conjClassSetIn_subset (typePV_subset_M data) hx
/-- A `V^M`-element is nonidentity: `V_S` avoids `1` (it is off `W₁ ∋ 1`), preserved under
`M`-conjugation. -/
theorem conjClassSetIn_typePV_one_not_mem {M : Subgroup G} (data : TypePData M) :
(1 : G) ∉ conjClassSetIn M (typePV M data) := by
rintro ⟨v, hv, h, -, hconj⟩
have hv1 : v ≠ 1 := by
rintro rfl
exact ((Set.mem_sdiff _).mp hv).2 (Or.inl (Subgroup.one_mem data.W1))
exact hv1 (by
have : v = h⁻¹ * (h * v * h⁻¹) * h := by group
rw [this, hconj]; group)
/-- `1 ∉ A₀(S)`: both parts avoid the identity (`S10.typePACore_one_not_mem` and
`conjClassSetIn_typePV_one_not_mem`). -/
theorem typePACore0_one_not_mem {M : Subgroup G} (data : TypePData M) :
(1 : G) ∉ S10.typePACore0 M data := by
rintro (h | h)
· exact S10.typePACore_one_not_mem h
· exact conjClassSetIn_typePV_one_not_mem data h
/-- **`A₀(S)` is `M`-conjugation invariant**: both `A(S)` (`S10.typePACore_conj_mem`) and `V^S`
(`mem_conjClassSetIn_conj_iff`) are stable under conjugation by `m ∈ M`. -/
theorem typePACore0_conj_mem [Finite G] {M : Subgroup G} (data : TypePData M) {m : G}
(hm : m ∈ M) {x : G} :
m * x * m⁻¹ ∈ S10.typePACore0 M data ↔ x ∈ S10.typePACore0 M data := by
simp only [S10.typePACore0, Set.mem_union]
constructor
· rintro (h | h)
· exact Or.inl (by
have := S10.typePACore_conj_mem (inv_mem hm) h
rwa [show m⁻¹ * (m * x * m⁻¹) * m⁻¹⁻¹ = x from by group] at this)
· exact Or.inr ((mem_conjClassSetIn_conj_iff hm x).mp h)
· rintro (h | h)
· exact Or.inl (S10.typePACore_conj_mem hm h)
· exact Or.inr ((mem_conjClassSetIn_conj_iff hm x).mpr h)
/-- `∀ x ∈ A₀(S), x ≠ 1` — the `≠ 1` form of `typePACore0_one_not_mem` (the shape the
`σ`-decomposition Dade engine's `hXsharp` obligation takes). -/
theorem typePACore0_ne_one {M : Subgroup G} (data : TypePData M) :
∀ x ∈ S10.typePACore0 M data, x ≠ (1 : G) :=
fun _ hx h => typePACore0_one_not_mem data (h ▸ hx)
/-- **A `V^S`-point does not escape `M`**: its centralizer lies in `M`. For `b = h·v·h⁻¹`
(`v ∈ V_S`, `h ∈ M`), `C_G(b) = h·C_G(v)·h⁻¹ ≤ h·M·h⁻¹ = M` (`centralizer_typePV_le_M`, `h ∈ M`).
This is what makes the `V`-part of the Dade engine's escaping-`σ`-sharp obligation **vacuous**. -/
theorem conjClassSetIn_typePV_centralizer_le_M {M : Subgroup G} (data : TypePData M)
{b : G} (hb : b ∈ conjClassSetIn M (typePV M data)) :
Subgroup.centralizer ({b} : Set G) ≤ M := by
obtain ⟨v, hv, h, hhM, rfl⟩ := hb
intro x hx
have hcomm : x * (h * v * h⁻¹) = (h * v * h⁻¹) * x :=
Subgroup.mem_centralizer_singleton_iff.mp hx
-- `h⁻¹ x h` commutes with `v`, hence lies in `C_G(v) ≤ M`.
have hcv : (h⁻¹ * x * h) ∈ Subgroup.centralizer ({v} : Set G) := by
rw [Subgroup.mem_centralizer_singleton_iff]
calc (h⁻¹ * x * h) * v
= h⁻¹ * (x * (h * v * h⁻¹)) * h := by group
_ = h⁻¹ * ((h * v * h⁻¹) * x) * h := by rw [hcomm]
_ = v * (h⁻¹ * x * h) := by group
have hxM : (h⁻¹ * x * h) ∈ M :=
OddOrder.Peterfalvi.S10.centralizer_typePV_le_M data hv hcv
have hxeq : x = h * (h⁻¹ * x * h) * h⁻¹ := by group
rw [hxeq]
exact M.mul_mem (M.mul_mem hhM hxM) (M.inv_mem hhM)
/-- **Escaping `A₀(S)`-points come from the `A(S)`-part**: since `V^S`-points do not escape
(`conjClassSetIn_typePV_centralizer_le_M`), any escaping point of `A₀(S)` lies in `A(S)` and escapes
there too. This reduces the Dade engine's escaping-`σ`-sharp and coprimality obligations for
`A₀(S)` to the already-established `S10.typePACore` ones. -/
theorem escaping_typePACore0_mem_typePACore {M : Subgroup G} (data : TypePData M)
{a : G} (ha : a ∈ escapingCentralizerSet M (S10.typePACore0 M data)) :
a ∈ escapingCentralizerSet M (S10.typePACore M) := by
obtain ⟨haA0, haesc⟩ := ha
rcases haA0 with hpa | hva
· exact ⟨hpa, haesc⟩
· exact absurd (conjClassSetIn_typePV_centralizer_le_M data hva) haesc
/-- **`W₁ ⊓ W₂ = ⊥`** for a `TypePData`: the two cyclic factors intersect trivially. `W₂ ≤ M'`
(`W2_le`/`H_le`) while `W₁` complements `M' = derivedInG M` in `M` (`M_complement`), so
`W₁ ⊓ W₂ ≤ W₁ ⊓ M' = ⊥`. (Ambient-subgroup form of the disjoint half of `M_complement`, mirroring
`TypePData.fitting_inf_U_eq_bot`.) -/
theorem typePData_W1_inf_W2_eq_bot {M : Subgroup G} (data : TypePData M) :
data.W1 ⊓ data.W2 = ⊥ := by
rw [eq_bot_iff]
intro x hx
rw [Subgroup.mem_inf] at hx
obtain ⟨hxW1, hxW2⟩ := hx
have hxM : x ∈ M := data.W1_le hxW1
have hxD : x ∈ derivedInG M := data.H_le (Subgroup.mem_inf.mp (data.W2_le hxW2)).1
have hmem : (⟨x, hxM⟩ : ↥M) ∈
((derivedInG M).subgroupOf M) ⊓ (data.W1.subgroupOf M) :=
Subgroup.mem_inf.mpr ⟨Subgroup.mem_subgroupOf.mpr hxD, Subgroup.mem_subgroupOf.mpr hxW1⟩
have hd := data.M_complement.disjoint
rw [disjoint_iff] at hd
rw [hd, Subgroup.mem_bot] at hmem
rw [Subgroup.mem_bot]
exact Subtype.ext_iff.mp hmem
/-- **A `V^S`-point's order is divisible by a `σ`-prime** (issue 9076 piece 4c): a
`V = typePV`-point `v = a·b` (`a ∈ W₁`, `b ∈ W₂`, cyclic `W = W₁ ⊔ W₂`) has a nontrivial
`W₂`-component `b ≠ 1` (else
`v = a ∈ W₁`). Since `W₁ ⊓ W₂ = ⊥` (`typePData_W1_inf_W2_eq_bot`) and `a, b` commute (abelian `W`),
`orderOf b ∣ orderOf v`; and `b ∈ W₂ ≤ M_σ` is a `σ`-element (`isPiElement_sigma_of_mem_Msigma`),
so some `σ`-prime divides `orderOf b ∣ orderOf v`. This is the `V`-side "order carries a `σ`-prime"
input that excludes `V^S` from the pure-`κ` set `𝒞_G(K₀#)`. -/
theorem exists_sigma_prime_dvd_orderOf_typePV [Finite G]
(hG : OddOrder.BG.IsMinimalSimpleOdd G) {M : Subgroup G} (hM : M ∈ maximalSubgroups G)
(data : TypePData M) {v : G} (hv : v ∈ typePV M data) :
∃ p ∈ (orderOf v).primeFactors,
p ∈ OddOrder.BG.Ch3.S10.sigma M ∧ p ∣ Nat.card ↥data.W2 := by
simp only [typePV, Set.mem_sdiff, Set.mem_union, SetLike.mem_coe, not_or] at hv
obtain ⟨hvW, hvnW1, _hvnW2⟩ := hv
have hcyc : IsCyclic ↥data.W := data.W_cyclic
let : CommGroup ↥data.W := hcyc.commGroup
have hW1le : data.W1 ≤ data.W := data.W_eq ▸ le_sup_left
have hW2le : data.W2 ≤ data.W := data.W_eq ▸ le_sup_right
have hsup : data.W1.subgroupOf data.W ⊔ data.W2.subgroupOf data.W = ⊤ := by
rw [← Subgroup.subgroupOf_sup hW1le hW2le, ← data.W_eq, Subgroup.subgroupOf_self]
have hvmem : (⟨v, hvW⟩ : ↥data.W) ∈
data.W1.subgroupOf data.W ⊔ data.W2.subgroupOf data.W := by
rw [hsup]; exact Subgroup.mem_top _
rw [Subgroup.mem_sup] at hvmem
obtain ⟨a, ha, b, hb, hab⟩ := hvmem
have haW1 : ((a : ↥data.W) : G) ∈ data.W1 := Subgroup.mem_subgroupOf.mp ha
have hbW2 : ((b : ↥data.W) : G) ∈ data.W2 := Subgroup.mem_subgroupOf.mp hb
have habG : ((a : ↥data.W) : G) * ((b : ↥data.W) : G) = v := by
have := congrArg (Subtype.val) hab; simpa using this
have hb1 : ((b : ↥data.W) : G) ≠ 1 := by
intro h; exact hvnW1 (by rw [← habG, h, mul_one]; exact haW1)
have hcomm : Commute ((a : ↥data.W) : G) ((b : ↥data.W) : G) := by
have h : ((a : ↥data.W) : G) * ((b : ↥data.W) : G)
= ((b : ↥data.W) : G) * ((a : ↥data.W) : G) := by
have h0 := congrArg Subtype.val (mul_comm a b); simpa using h0
exact h
-- `orderOf b ∣ orderOf v` (commuting factors of a cyclic `W = W₁ × W₂`).
have hbdvd : orderOf ((b : ↥data.W) : G) ∣ orderOf v := by
have hab_n : ((a : ↥data.W) : G) ^ orderOf v * ((b : ↥data.W) : G) ^ orderOf v = 1 := by
rw [← hcomm.mul_pow, habG]; exact pow_orderOf_eq_one v
have hanW1 : ((a : ↥data.W) : G) ^ orderOf v ∈ data.W1 := pow_mem haW1 _
have hbnW2 : ((b : ↥data.W) : G) ^ orderOf v ∈ data.W2 := pow_mem hbW2 _
have han_eq : ((a : ↥data.W) : G) ^ orderOf v
= (((b : ↥data.W) : G) ^ orderOf v)⁻¹ := mul_eq_one_iff_eq_inv.mp hab_n
have han_in_W2 : ((a : ↥data.W) : G) ^ orderOf v ∈ data.W2 := by
rw [han_eq]; exact inv_mem hbnW2
have han_bot : ((a : ↥data.W) : G) ^ orderOf v ∈ data.W1 ⊓ data.W2 :=
Subgroup.mem_inf.mpr ⟨hanW1, han_in_W2⟩
rw [typePData_W1_inf_W2_eq_bot data, Subgroup.mem_bot] at han_bot
have hbn1 : ((b : ↥data.W) : G) ^ orderOf v = 1 := by
rw [han_bot, one_mul] at hab_n; exact hab_n
exact orderOf_dvd_of_pow_eq_one hbn1
-- `b ∈ M_σ` (`W₂ ≤ H = M_F ≤ M_σ`), so a `σ`-prime divides `orderOf b`.
have hbMσ : ((b : ↥data.W) : G) ∈ OddOrder.BG.Ch3.S10.Msigma M := by
refine (data.W2_le.trans (inf_le_left.trans ?_)) hbW2
rw [data.H_eq]; exact OddOrder.BG.Ch4.S15.maxNilpotentNormalHall_le_Msigma hG hM
have hbσ : IsPiElement (OddOrder.BG.Ch3.S10.sigma M) ((b : ↥data.W) : G) :=
OddOrder.BG.Ch4.S14.isPiElement_sigma_of_mem_Msigma hbMσ
have hbord1 : orderOf ((b : ↥data.W) : G) ≠ 1 := fun h => hb1 (orderOf_eq_one_iff.mp h)
obtain ⟨p, hpp, hpdvdb⟩ := (orderOf ((b : ↥data.W) : G)).exists_prime_and_dvd hbord1
refine ⟨p, Nat.mem_primeFactors.mpr ⟨hpp, hpdvdb.trans hbdvd, (orderOf_pos v).ne'⟩, ?_, ?_⟩
· exact hbσ p (Nat.mem_primeFactors.mpr ⟨hpp, hpdvdb, (orderOf_pos _).ne'⟩)
· exact hpdvdb.trans (data.W2.orderOf_dvd_natCard hbW2)
/-- **`V^S ⊆ A0Set M K₀`** (issue 9076 piece 4c): the `M`-conjugacy closure of the exceptional
regular set `V = typePV` embeds into the `A_0(M)`-set. `V ⊆ hatMsigma M`
(`typePV_subset_hatMsigma`) extends to the closure because `hatMsigma M` is `M`-conjugation
invariant (`M_σ ◁ M`); and a `V`-point carries a `σ`-prime
(`exists_sigma_prime_dvd_orderOf_typePV`), `σ ⊆ κ′` (`kappa_subset_sigmaCompl`), so it is not a
`κ`-element, hence off the pure-`κ` set `𝒞_G(K₀#)` (conjugation-invariant,
`mem_conjClassSet_conj_iff`). -/
theorem conjClassSetIn_typePV_subset_A0Set [Finite G]
(hG : OddOrder.BG.IsMinimalSimpleOdd G) {M K₀ : Subgroup G} (hM : M ∈ maximalSubgroups G)
(hKM : K₀ ≤ M)
(hK : OddOrder.Isaacs.Ch03.IsHallSubgroup (OddOrder.BG.Ch4.S14.kappa M) (K₀.subgroupOf M))
(data : TypePData M) :
conjClassSetIn M (typePV M data) ⊆ OddOrder.BG.Ch4.S16.A0Set M K₀ := by
have hM_le_NMσ : M ≤ Subgroup.normalizer (OddOrder.BG.Ch3.S10.Msigma M : Set G) := by
rw [OddOrder.BG.Ch3.S10.Msigma]
exact le_normalizer_opiCoreInG (OddOrder.BG.Ch3.S10.sigma M) M
have hMσ_norm : ((OddOrder.BG.Ch3.S10.Msigma M).subgroupOf M).Normal :=
(Subgroup.normal_subgroupOf_iff_le_normalizer
(OddOrder.BG.Ch3.S10.Msigma_le M)).mpr hM_le_NMσ
rintro _ ⟨v, hv, m, hmM, rfl⟩
have hvhat : v ∈ OddOrder.BG.Ch4.S16.hatMsigma M := typePV_subset_hatMsigma hG hM data hv
simp only [OddOrder.BG.Ch4.S16.A0Set, Set.mem_sdiff, OddOrder.BG.Ch4.S16.hatMsigma,
Set.mem_ofPred_eq]
refine ⟨⟨M.mul_mem (M.mul_mem hmM hvhat.1) (M.inv_mem hmM), ?_⟩, ?_⟩
· -- `M_σ ⊓ C_G(m·v·m⁻¹) ≠ ⊥`: conjugate the `M_σ`-centralizing witness of `v`.
obtain ⟨w, hw1⟩ := Subgroup.ne_bot_iff_exists_ne_one.mp hvhat.2
obtain ⟨hwMσ, hwC⟩ := Subgroup.mem_inf.mp w.2
have hw0C : (w : G) * v = v * (w : G) := Subgroup.mem_centralizer_singleton_iff.mp hwC
have hwne : (w : G) ≠ 1 := fun h => hw1 (Subtype.ext h)
have hconjMσ : m * (w : G) * m⁻¹ ∈ OddOrder.BG.Ch3.S10.Msigma M := by
have h := hMσ_norm.conj_mem ⟨(w : G), OddOrder.BG.Ch3.S10.Msigma_le M hwMσ⟩
(Subgroup.mem_subgroupOf.mpr hwMσ) ⟨m, hmM⟩
rw [Subgroup.mem_subgroupOf] at h
simpa using h
have hconjC : m * (w : G) * m⁻¹ ∈ Subgroup.centralizer ({m * v * m⁻¹} : Set G) := by
rw [Subgroup.mem_centralizer_singleton_iff]
calc (m * (w : G) * m⁻¹) * (m * v * m⁻¹)
= m * ((w : G) * v) * m⁻¹ := by group
_ = m * (v * (w : G)) * m⁻¹ := by rw [hw0C]
_ = (m * v * m⁻¹) * (m * (w : G) * m⁻¹) := by group
have hconjne : m * (w : G) * m⁻¹ ≠ 1 := by
intro h
refine hwne ?_
have hww : (w : G) = m⁻¹ * (m * (w : G) * m⁻¹) * m := by group
rw [hww, h]; group
rw [Subgroup.ne_bot_iff_exists_ne_one]
exact ⟨⟨m * (w : G) * m⁻¹, Subgroup.mem_inf.mpr ⟨hconjMσ, hconjC⟩⟩,
fun h => hconjne (Subtype.ext_iff.mp h)⟩
· -- `m·v·m⁻¹ ∉ 𝒞_G(K₀#)`: reduce to `v` (conj-invariant), which carries a `σ ⊆ κ′`-prime.
intro hconj
rw [mem_conjClassSet_conj_iff] at hconj
obtain ⟨hvκ, -⟩ :=
OddOrder.BG.Ch4.S16.kappaHall_conjClassSet_isPiElement hKM hK hconj
obtain ⟨p, hpf, hpσ, -⟩ := exists_sigma_prime_dvd_orderOf_typePV hG hM data hv
exact OddOrder.BG.Ch4.S14.kappa_subset_sigmaCompl (hvκ p hpf) hpσ
/-- **`A₀(S) ⊆ A0Set M K₀`** (issue 9076 piece 4c): the honest type-`P₂` support embeds into BG's
Theorem-E set. `A(S)`-part via `S10.typePACore_subset_ASet` + `aSet_subset_A0Set`; `V^S`-part via
`conjClassSetIn_typePV_subset_A0Set`. This is the bridge feeding `theoremII_tame_embedding` for
`typePACore0_tame_conj`. -/
theorem typePACore0_subset_A0Set [Finite G]
(hG : OddOrder.BG.IsMinimalSimpleOdd G) {M K₀ U₀ : Subgroup G} (hM : M ∈ maximalSubgroups G)
(hKM : K₀ ≤ M) (hUM : U₀ ≤ M) (hKne : K₀ ≠ ⊥)
(hK : OddOrder.Isaacs.Ch03.IsHallSubgroup (OddOrder.BG.Ch4.S14.kappa M) (K₀.subgroupOf M))
(hU : OddOrder.Isaacs.Ch03.IsHallSubgroup
((OddOrder.BG.Ch4.S14.kappa M ∪ OddOrder.BG.Ch3.S10.sigma M)ᶜ) (U₀.subgroupOf M))
(data : TypePData M) :
S10.typePACore0 M data ⊆ OddOrder.BG.Ch4.S16.A0Set M K₀ := by
apply Set.union_subset
· exact (S10.typePACore_subset_ASet hG hM hKM hUM hKne hK hU).trans
(OddOrder.BG.Ch4.S16.aSet_subset_A0Set hG hM hKM hUM hK hU)
· exact conjClassSetIn_typePV_subset_A0Set hG hM hKM hK data
/-- **Tame conjugation for the honest type-`P₂` `A₀`-support** (BG §16 Theorem II): two
`G`-conjugate elements of `A₀(M) = A(M) ∪ V^M` are already `M`-conjugate. This is the first
conjunct of BG Theorem II (`OddOrder.BG.Ch4.S16.theoremII_tame_embedding` with `X = A0Set M K`): for
the tame embedding, a `G`-fusion of support points is controlled by `N_G(M) = M`.
Discharge route (issue 9076 piece 4c): bridge the honest support `S10.typePACore0 M data` into
BG's
`A0Set M K = hatMsigma M ∖ 𝒞_G(K#)` — the `A(M)`-part via `S10.typePACore_subset_hatMsigma`, the
`V^M`-part via `typePV ⊆ hatMsigma` plus the order argument `V^M ∩ 𝒞_G(K#) = ∅` (a `V`-point has
order divisible by `pq`, a `K#`-point only by `q`) — and produce the `κ`-Hall `K` / `(κ∪σ)′`-Hall
`U`
of the type-`P` maximal `M`. With `a, b ∈ A0Set M K`, `theoremII_tame_embedding`'s first conjunct
supplies the `M`-conjugator. (Reduces the earlier "genuine deep FT-support geometry" pin to this
concrete BG-support bridge.) -/
theorem typePACore0_tame_conj [Finite G]
(hG : OddOrder.BG.IsMinimalSimpleOdd G) {M : Subgroup G} (hM : M ∈ maximalSubgroups G)
(data : TypePData M) {K₀ U₀ : Subgroup G} (hKM : K₀ ≤ M) (hUM : U₀ ≤ M) (hKne : K₀ ≠ ⊥)
(hK : OddOrder.Isaacs.Ch03.IsHallSubgroup (OddOrder.BG.Ch4.S14.kappa M) (K₀.subgroupOf M))
(hU : OddOrder.Isaacs.Ch03.IsHallSubgroup
((OddOrder.BG.Ch4.S14.kappa M ∪ OddOrder.BG.Ch3.S10.sigma M)ᶜ) (U₀.subgroupOf M))
{a b : G}
(ha : a ∈ S10.typePACore0 M data) (hb : b ∈ S10.typePACore0 M data)
(hab : IsConj a b) : ∃ m ∈ M, b = m * a * m⁻¹ := by
have hsub := typePACore0_subset_A0Set hG hM hKM hUM hKne hK hU data
have hII := OddOrder.BG.Ch4.S16.theoremII_tame_embedding hG hM hKM hUM hK hU
(X := OddOrder.BG.Ch4.S16.A0Set M K₀) (Or.inr rfl)
obtain ⟨g, hg⟩ := isConj_iff.mp hab
obtain ⟨m, hmM, hmb⟩ := hII.1 a (hsub ha) b (hsub hb) ⟨g, hg.symm⟩
exact ⟨m, hmM, hmb⟩
/-- **(8.13.a), the mixed `A(S)`–`V^S` case is vacuous**: an `A(S)`-point is never `G`-conjugate to
a `V^S`-point. An `A(S) = S10.typePACore` element lies in `S' = derivedInG S`, while a
`V^S`-point lies **outside** `S'` (`typePData_typePV_not_mem_derived`, the nontrivial
`W₁`-component).
Proved (issue 9076 piece 4c) via the honest-support tame conjugation `typePACore0_tame_conj`
(BG §16 Theorem II): if `a` and `b` were `G`-conjugate, they would be **`M`-conjugate**
(`b = m·a·m⁻¹`, `m ∈ M`); but `M' = derivedInG M` is normal in `M`, so `M`-conjugation preserves
`M'`-membership, forcing `b ∈ M'` from `a ∈ M'` — contradicting `b ∉ M'`. This replaces the earlier
circular `normedTI`-based argument with the direct FT-support-geometry route (Coq `FTsupp0`). -/
theorem not_isConj_typePACore_typePV [Finite G]
(hG : OddOrder.BG.IsMinimalSimpleOdd G) {M : Subgroup G} (hM : M ∈ maximalSubgroups G)
(data : TypePData M) {K₀ U₀ : Subgroup G} (hKM : K₀ ≤ M) (hUM : U₀ ≤ M) (hKne : K₀ ≠ ⊥)
(hK : OddOrder.Isaacs.Ch03.IsHallSubgroup (OddOrder.BG.Ch4.S14.kappa M) (K₀.subgroupOf M))
(hU : OddOrder.Isaacs.Ch03.IsHallSubgroup
((OddOrder.BG.Ch4.S14.kappa M ∪ OddOrder.BG.Ch3.S10.sigma M)ᶜ) (U₀.subgroupOf M))
{a b : G}
(ha : a ∈ S10.typePACore M) (hb : b ∈ conjClassSetIn M (typePV M data))
(hab : IsConj a b) : False := by
-- `M' ⊴ M`. (In the file this was extracted from, the instance came in transitively through the
-- §13 import closure; here it is derived on the spot — `M'.subgroupOf M` is the comap of a
-- `map` along the injective `M.subtype`.)
have : ((derivedInG M).subgroupOf M).Normal := by
rw [derivedInG, Subgroup.subgroupOf,
Subgroup.comap_map_eq_self_of_injective M.subtype_injective]
infer_instance
-- `M`-conjugation preserves `M' = derivedInG M`-membership (`M' ⊴ M`).
have hM'le : derivedInG M ≤ M := Subgroup.map_subtype_le _
have hconj_derived : ∀ x w : G, x ∈ M → w ∈ derivedInG M → x * w * x⁻¹ ∈ derivedInG M := by
intro x w hxM hwD
have hconj := (inferInstance : ((derivedInG M).subgroupOf M).Normal).conj_mem
⟨w, hM'le hwD⟩ (Subgroup.mem_subgroupOf.mpr hwD) ⟨x, hxM⟩
rw [Subgroup.mem_subgroupOf] at hconj
simpa using hconj
-- `a ∈ M'`, and (BG tame) `b ∈ A₀(M)` is `M`-conjugate to `a ∈ A(M) ⊆ A₀(M)`.
have haD : a ∈ derivedInG M := (S10.mem_typePACore.mp ha).1
have hbA0 : b ∈ S10.typePACore0 M data := Set.mem_union_right _ hb
obtain ⟨v, hv, h, hhM, hhvb⟩ := hb
-- `b ∉ M'`: `b = h·v·h⁻¹`, `v ∈ V ⊄ M'`, `h ∈ M`, and `M' ⊴ M`.
have hbnD : b ∉ derivedInG M := by
intro hbD
refine OddOrder.Peterfalvi.S10.typePData_typePV_not_mem_derived data hv ?_
have hvconj : h⁻¹ * b * (h⁻¹)⁻¹ ∈ derivedInG M :=
hconj_derived h⁻¹ b (M.inv_mem hhM) hbD
rwa [show h⁻¹ * b * (h⁻¹)⁻¹ = v from by rw [← hhvb]; group] at hvconj
-- BG §16 Theorem II tame conjugation ⟹ `b = m·a·m⁻¹ ∈ M'`, contradiction.
obtain ⟨m, hmM, hmc⟩ :=
typePACore0_tame_conj hG hM data hKM hUM hKne hK hU
(typePACore_subset_A0Set data ha) hbA0 hab
exact hbnD (hmc ▸ hconj_derived m a hmM haD)
/-- **Peterfalvi (8.15) for the honest type-`P₂` `A₀`-support**: the Dade (2.2) support hypotheses
hold for `A₀(S) = A(S) ∪ V^S`. Assembled through the `σ`-decomposition engine
`dadeSupportHypothesisData_of_subset_escaping_sigmaSharp`:
* set-facts (`⊆ M`, `≠ 1`, `M`-conjugation-invariant, nonempty) — the `typePACore0_*` facts;
* escaping-`σ`-sharp and coprimality — reduced to the `A(S)`-part
(`escaping_typePACore0_mem_typePACore`, since `V^S` does not escape) plus the generic
`V`-part coprimality (`coprime_FT_signalizer_centralizerIn_typePV`);
* the `isConj → M`-conjugate obligation — the `A`–`A` and `V`–`V` cases are
`S10.typePACore_isConj_conj_in_M` / `conjClassSetIn_typePV_isConj_conj_in_M`, and the mixed case
is the vacuity `not_isConj_typePACore_typePV` (the one deep `'A0`-`normedTI` pin).
This is the `S`-side Dade datum the (13.18) row-`0` cross-relation `τ_S(μ_{0j} − μ_{01}) =
η_{0j} − η_{01}` actually needs (the `μ`-differences are `A₀(S)`-supported, not `A(S)`-supported).
-/
theorem dadeSupportHypothesisData_typePACore0 [Fintype G] [Finite G]
(hG : OddOrder.BG.IsMinimalSimpleOdd G) {M : Subgroup G} (hM : M ∈ maximalSubgroups G)
(hTP : OddOrder.BG.Ch4.S14.IsTypeP M) (data : TypePData M) :
Nonempty (OddOrder.Peterfalvi.S10.DadeSupportHypothesisData M (S10.typePACore0 M data)) := by
classical
obtain ⟨K₀, U₀, hKM, hUM, hKne, hK, hU⟩ :=
OddOrder.Peterfalvi.S10.typeP_exists_kappa_hall_pair hG hM hTP
refine OddOrder.Peterfalvi.S10.dadeSupportHypothesisData_of_subset_escaping_sigmaSharp hG hM
(typePACore0_subset data) (typePACore0_ne_one data)
(fun a ha => S10.escaping_typePACore_mem_sigmaSharp hG hM hKM hUM hKne hK hU
(escaping_typePACore0_mem_typePACore data ha))
?_ ?_ ?_ ?_
· -- `isConj → M`-conjugate: A–A / A–V (vacuous) / V–A (vacuous) / V–V.
intro a ha b hb hab
rcases ha with hpa | hva
· rcases hb with hpb | hvb
· exact S10.typePACore_isConj_conj_in_M hG hM hKM hUM hKne hK hU hpa hpb hab
· exact (not_isConj_typePACore_typePV hG hM data hKM hUM hKne hK hU hpa hvb hab).elim
· rcases hb with hpb | hvb
· exact (not_isConj_typePACore_typePV hG hM data hKM hUM hKne hK hU hpb hva
hab.symm).elim
· exact OddOrder.Peterfalvi.S10.conjClassSetIn_typePV_isConj_conj_in_M data hva hvb hab
· -- coprimality: the escaping point is in `A(S)`; `b` is in `A(S)` or `V^S`.
intro a ha b hb
have haA := escaping_typePACore0_mem_typePACore data ha
have haσ := S10.escaping_typePACore_mem_sigmaSharp hG hM hKM hUM hKne hK hU haA
rcases hb with hpb | hvb
· exact S10.coprime_FT_signalizer_centralizerIn_typePACore hG hM haσ ha.2 hpb
· exact OddOrder.Peterfalvi.S10.coprime_FT_signalizer_centralizerIn_typePV hG hM data haσ
ha.2 hvb
· -- nonempty: `M_σ^# ⊆ A(S) ⊆ A₀(S)`.
obtain ⟨a, ha1⟩ :=
Subgroup.ne_bot_iff_exists_ne_one.mp (OddOrder.BG.Ch3.S10.Msigma_ne_bot hG hM)
have ha1' : (a : G) ≠ 1 := fun h => ha1 (Subtype.ext h)
have haMσ : (a : G) ∈ OddOrder.BG.Ch3.S10.Msigma M := a.2
have haM' : (a : G) ∈ derivedInG M := OddOrder.BG.Ch3.S10.Msigma_le_derived hG hM haMσ
exact ⟨a.1, Or.inl ⟨haM', ha1', a.1,
(Set.mem_sdiff _).mpr ⟨SetLike.mem_coe.mpr haMσ, fun h => ha1' (Set.mem_singleton_iff.mp h)⟩,
Subgroup.mem_centralizer_singleton_iff.mpr rfl⟩⟩
· -- `M`-conjugation invariance.
intro m x hm
exact typePACore0_conj_mem data hm
end OddOrder.Peterfalvi.S10
Source
Yawara Ishida, Lean 4 formalization of the Feit–Thompson (odd order) theorem, github.com/yawara/odd-order @ 20a6fccbb63c (Apache-2.0), https://github.com/yawara/odd-order/blob/20a6fccbb63c0b138acda9f7fafd798c42f9441b/OddOrder/Peterfalvi/S10_TypePSupportA0.lean