Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Peterfalvi (6.2)/(6.3): the standalone general (6.1) coherence theorems

Definition
OddOrder_Peterfalvi_S08_Theorem62_63_Standalone

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

feit-thompsonfinite-groupsgroup-theoryishida-odd-order

Peterfalvi, Character Theory for the Odd Order Theorem, §6, (6.1)-(6.3). Earlier files proved the coherence theorems (6.2)/(6.3) only in the Sibley–Dade capstone setup, where the induced-character kernel KKK equals the nilpotent Frobenius complement HHH; this file states them for the general (6.1) hypothesis with KKK solvable.

Main results: six_two_general (the general (6.2) form), inducedMember_re_le_general (the (6.2) θ\thetaθ-bound for an induced member of a general solvable kernel) and six_three_index_bound_general (the (6.3) per-step index bound), plus subset (coherence is inherited by subsets).

The bundle also proves 3 supporting lemmas used by later statements.

Definition code
/-
Copyright (c) 2026 The Odd Order Project. 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/S08_Theorem62_63_Standalone.lean
-- Generated for Prove2Me by deleting declarations; the remaining commands are verbatim except:
-- `private` dropped where another file uses the declaration, and a solution's main theorem
-- renamed `_root_.solution` (platform convention).
import Definitions.Def_OddOrder_Peterfalvi_S08_CoherenceCorePart1

namespace OddOrder.Peterfalvi.S07

open OddOrder.RepresentationTheory

variable {L G : Type*} [Group L] [Group G]

variable [Fintype L] [Fintype G] [Invertible (Nat.card L : ℂ)] [Invertible (Nat.card G : ℂ)]

namespace IsCoherent

variable {τ : IntegralCharacterMap L G} {S S' : Set (ClassFunction L ℂ)} {A : Set L}

/-- **Coherence is inherited by subsets.**  If `(S, A, τ)` is coherent and `S' ⊆ S` still admits a
nonzero `A`-supported element of `ℤ[S']`, then `(S', A, τ)` is coherent.

The *same* coherent extension works: `ℤ[S'] ≤ ℤ[S]` (`Submodule.span_mono`) so the lattice isometry
(`extension_inner_eq`) and the `ℤ[Irr G]`-codomain condition (`extension_mem_ZIrr`) restrict, and
`zSupportedSpan S' A ⊆ zSupportedSpan S A` (`zSupportedSpan_mono_left`) so agreement with `τ` on the
supported lattice (`extends_on_supported`) restricts.  Only the nondegeneracy witness `nonzero` has
to be re-supplied — it is genuinely set-dependent — hence the `hne` hypothesis.

This is the general monotonicity of Peterfalvi's (5.1) coherence predicate.  In the (6.2) coherence
dichotomy a smaller coherent set's witness `φ ∈ zSupportedSpan S(A) A` transports up by
`zSupportedSpan_mono_left`, so `hne` is always available when `S' ⊇` a known coherent set. -/
noncomputable def subset (hτ : IsCoherent τ S A) (hsub : S' ⊆ S)
    (hne : ∃ φ : ClassFunction L ℂ, φ ∈ zSupportedSpan (L := L) S' A ∧ φ ≠ 0) :
    IsCoherent τ S' A where
  nonzero := hne
  extension := hτ.extension
  extension_inner_eq := fun φ ψ hφ hψ =>
    hτ.extension_inner_eq φ ψ (Submodule.span_mono hsub hφ) (Submodule.span_mono hsub hψ)
  extends_on_supported := fun φ hφ =>
    hτ.extends_on_supported φ (zSupportedSpan_mono_left hsub hφ)
  extension_mem_ZIrr := fun φ hφ =>
    hτ.extension_mem_ZIrr φ (Submodule.span_mono hsub hφ)

end IsCoherent

end OddOrder.Peterfalvi.S07

namespace OddOrder.Peterfalvi.S08

open OddOrder.RepresentationTheory

variable {G : Type*} [Group G] [Fintype G] [Invertible (Nat.card G : ℂ)]

variable {L : Subgroup G} [Fintype ↥L] [Invertible (Nat.card ↥L : ℂ)]

/-- **Peterfalvi (6.3), general Hypothesis (6.1) form** (the minimal-`A` descent).

Let `K` be a solvable normal subgroup of `L` (the induced-family kernel `S = {Ind_K^L θ}`), and let
`M ≤ H₁ < H ≤ K` be normal subgroups with `H` nilpotent (Peterfalvi states `H/M` nilpotent; the
§11 consumer has `M = 1`, `H = HC` nilpotent).  Writing `S(·)` for the coherence functor `SOf`, if
`S(H₁)` is coherent and `|H : H₁| > 4|L:K|² + 1`, then `S(M)` is coherent.

The character-theoretic content — Peterfalvi (6.2) in index form, `|H:H₁| ≤ 4|L:K|² + 1` whenever a
proper section `B ⊆ A ⊆ H₁` has `A/B` central in `H/B`, `S(A)` coherent and `S(B)` not — is the
explicit **oracle** `h62`.  (In the Sibley case `h62` is `six_three_index_bound`; for a general
solvable `K` it bottoms out in the (5.6) bound for reducible induced members.)  Everything else is
proved here, mirroring `S08.six_three` but with `K` and `H` separated:

* the minimal coherent `A ∈ [M, H₁]` (`Set.Finite.exists_minimalFor`);
* the maximal proper normal `B` below it (`exists_maximal_normal_between`);
* `S(B)` not coherent (minimality of `A`);
* `A/B ⊆ Z(H/B)` (`normal_central_of_maximal_normal_below`, `H` nilpotent);
* `[K/A, K/A] ≠ ⊤` from `K` solvable and `A < K` (the degree-`|L:K|` anchor input of (6.2)).

This is the producer cited by `S13.coherent_S_of_coherent_SH0C` (Peterfalvi (11.3), `(6.3)` applied
with `(L,K,M,H,H₁) = (M, M', 1, HC, H₀C)`). -/
theorem six_three_descent
    {K H M H₁ : Subgroup ↥L} [Group.IsSolvable ↥K]
    [M.Normal] [H₁.Normal]
    [Group.IsNilpotent (↥H ⧸ M.subgroupOf H)] (hHnorm : H.Normal)
    (hMH₁ : M ≤ H₁) (hH₁H : H₁ < H) (hHK : H ≤ K)
    (τ : OddOrder.Peterfalvi.S07.IntegralCharacterMap ↥L G) (A0 : Set ↥L)
    (SOf : Subgroup ↥L → Set (ClassFunction ↥L ℂ))
    (h62 : ∀ (A B : Subgroup ↥L) [A.Normal] [B.Normal], B ≤ A → A ≤ H₁ →
        _root_.commutator (↥K ⧸ A.subgroupOf K) ≠ ⊤ →
        (A.subgroupOf H).map (QuotientGroup.mk' (B.subgroupOf H)) ≤
          Subgroup.center (↥H ⧸ B.subgroupOf H) →
        Nonempty (OddOrder.Peterfalvi.S07.IsCoherent τ (SOf A) A0) →
        ¬ Nonempty (OddOrder.Peterfalvi.S07.IsCoherent τ (SOf B) A0) →
        Nat.card (↥H ⧸ H₁.subgroupOf H) ≤ 4 * K.index ^ 2 + 1)
    (hcoh : Nonempty (OddOrder.Peterfalvi.S07.IsCoherent τ (SOf H₁) A0))
    (hbound : 4 * K.index ^ 2 + 1 < Nat.card (↥H ⧸ H₁.subgroupOf H)) :
    Nonempty (OddOrder.Peterfalvi.S07.IsCoherent τ (SOf M) A0) := by
  classical
  have : Finite (Subgroup ↥L) := Finite.of_injective (fun K : Subgroup ↥L => (K : Set ↥L))
    (fun _ _ h => SetLike.coe_injective h)
  set s : Set (Subgroup ↥L) := {A | A.Normal ∧ M ≤ A ∧ A ≤ H₁ ∧
    Nonempty (OddOrder.Peterfalvi.S07.IsCoherent τ (SOf A) A0)} with hs_def
  have hH₁s : H₁ ∈ s := by
    simp only [hs_def, Set.mem_ofPred_eq]; exact ⟨‹H₁.Normal›, hMH₁, le_refl _, hcoh⟩
  obtain ⟨A, hAmem, hAmin⟩ :=
    Set.Finite.exists_minimalFor (id : Subgroup ↥L → Subgroup ↥L) s (Set.toFinite _) ⟨H₁, hH₁s⟩
  simp only [hs_def, Set.mem_ofPred_eq] at hAmem
  obtain ⟨hAnorm, hMA, hAH₁, hAcoh⟩ := hAmem
  have : A.Normal := hAnorm
  have hAeqM : A = M := by
    by_contra hne
    have hMltA : M < A := lt_of_le_of_ne hMA (Ne.symm hne)
    obtain ⟨B, hBnorm, hMB, hBltA, hBmaxl⟩ := exists_maximal_normal_between hMltA
    have : B.Normal := hBnorm
    have hAltH : A < H := lt_of_le_of_lt hAH₁ hH₁H
    have hAltK : A < K := lt_of_lt_of_le hAltH hHK
    -- `[K/A, K/A] ≠ ⊤`: `K` solvable, `K/A` nontrivial (`A < K`).
    have hAcomm : _root_.commutator (↥K ⧸ A.subgroupOf K) ≠ ⊤ := by
      have : (A.subgroupOf K).Normal := (‹A.Normal›).subgroupOf K
      have : Nontrivial (↥K ⧸ A.subgroupOf K) := by
        rw [QuotientGroup.nontrivial_iff]
        intro htop
        exact hAltK.ne' (le_antisymm (Subgroup.subgroupOf_eq_top.mp htop) hAltK.le)
      exact (Group.IsSolvable.commutator_lt_top_of_nontrivial (↥K ⧸ A.subgroupOf K)).ne
    -- `A/B ⊆ Z(H/B)`: `H/B` nilpotent (a quotient of the nilpotent `H/M`, as `M ≤ B`), and
    -- `B` is maximal below `A`.
    have hnilB : Group.IsNilpotent (↥H ⧸ B.subgroupOf H) :=
      Group.nilpotent_of_surjective
        (QuotientGroup.map (M.subgroupOf H) (B.subgroupOf H) (MonoidHom.id ↥H)
          (by simpa using Subgroup.subgroupOf_mono H hMB))
        (by
          intro q
          obtain ⟨x, rfl⟩ := QuotientGroup.mk_surjective (s := B.subgroupOf H) q
          exact ⟨QuotientGroup.mk x, by simp⟩)
    have hcentral := normal_central_of_maximal_normal_below (H := H) (A := A) (B := B)
      hHnorm hAltH.le hBltA hBmaxl
    -- `S(B)` not coherent: else `B` beats the minimality of `A`.
    have hSBncoh : ¬ Nonempty (OddOrder.Peterfalvi.S07.IsCoherent τ (SOf B) A0) := by
      intro hBcoh
      have hBs : B ∈ s := by
        simp only [hs_def, Set.mem_ofPred_eq]
        exact ⟨hBnorm, hMB, hBltA.le.trans hAH₁, hBcoh⟩
      exact lt_irrefl _ (hBltA.trans_le (hAmin hBs hBltA.le))
    have hbnd := h62 A B hBltA.le hAH₁ hAcomm hcentral hAcoh hSBncoh
    omega
  rw [← hAeqM]; exact hAcoh

set_option linter.unusedFintypeInType false in
omit [Fintype G] [Invertible (Nat.card G : ℂ)] [Invertible (Nat.card ↥L : ℂ)] in
/-- **Peterfalvi (6.3) per-step index bound, general (6.1) form** (`K` solvable, `H ≤ K`, `K ≠ H`).

The arithmetic glue from the (6.2) bound to the (6.3) index bound, separating `K` (the
induced-family kernel) from `H` (the nilpotent middle group).  The Sibley `six_three_index_bound`
has `K = H` (`|K:H| = 1`); here `K ≠ H` is allowed.  Given the (6.2) consequence for the section
`C = H, D = A` — `|K:A| − 1 ≤ 2|L:H|·√|H:A|` (`h62`) — the index `|H:H₁|` is at most `4|L:K|² + 1`.

Rewrites `|K:A| = |H:A|·|K:H|` (`relindex_mul_relindex`, `A ≤ H ≤ K`) and `|L:H| = |K:H|·|L:K|`
(`relindex_mul_index`, `H ≤ K`) into the `|K:H|·|H:A| − 1 ≤ 2|L:K|·|K:H|·√|H:A|` form consumed by
the arithmetic core `six_three_HH1_le` (`LK = |L:K|`, `KH = |K:H|`, `HA = |H:A|`).

Composing this with `six_three_descent` (supply `h62 := fun A B _ _ _ _ hA hB =>
six_three_index_bound_general … (six_two A B … hA hB)`) leaves the single remaining gate the general
`six_two` (the (6.2) bound for the possibly-reducible induced members of a solvable `K`). -/
theorem six_three_index_bound_general
    {K H A H₁ : Subgroup ↥L} (hAH₁ : A ≤ H₁) (hAH : A ≤ H) (hHK : H ≤ K)
    (h62 : (Nat.card (↥K ⧸ A.subgroupOf K) : ℝ) - 1 ≤
        2 * (H.index : ℝ) * Real.sqrt (Nat.card (↥H ⧸ A.subgroupOf H) : ℝ)) :
    Nat.card (↥H ⧸ H₁.subgroupOf H) ≤ 4 * K.index ^ 2 + 1 := by
  -- `relindex`/`Nat.card`-quotient identities (`relindex H K = (H.subgroupOf K).index`).
  have hcardKA : Nat.card (↥K ⧸ A.subgroupOf K) = A.relIndex K := (Subgroup.index_eq_card _).symm
  have hcardHA : Nat.card (↥H ⧸ A.subgroupOf H) = A.relIndex H := (Subgroup.index_eq_card _).symm
  have hcardHH1 : Nat.card (↥H ⧸ H₁.subgroupOf H) = H₁.relIndex H := (Subgroup.index_eq_card _).symm
  -- tower index multiplicativity: `|K:A| = |H:A|·|K:H|` and `|L:H| = |K:H|·|L:K|`.
  have key1 : A.relIndex K = A.relIndex H * H.relIndex K :=
    (Subgroup.relIndex_mul_relIndex A H K hAH hHK).symm
  have key2 : H.index = H.relIndex K * K.index := (Subgroup.relIndex_mul_index hHK).symm
  -- `1 ≤ |K:H|`.
  have hKHpos : 0 < H.relIndex K := by
    rw [show H.relIndex K = Nat.card (↥K ⧸ H.subgroupOf K) from (Subgroup.index_eq_card _).symm]
    exact Nat.card_pos
  -- `|H:H₁| ≤ |H:A|` (from `A ≤ H₁`, `A∩H ≤ H₁∩H`).
  have hHH1le : H₁.relIndex H ≤ A.relIndex H :=
    Nat.le_of_dvd Nat.card_pos (Subgroup.index_dvd_of_le (Subgroup.subgroupOf_mono H hAH₁))
  rw [hcardHH1]
  refine six_three_HH1_le (LK := K.index) (KH := H.relIndex K) (HA := A.relIndex H)
    (HH1 := H₁.relIndex H) hKHpos hHH1le ?_
  -- the (6.2) real bound, recast into the `six_three_HH1_le` shape.
  rw [hcardKA, key1, key2, hcardHA] at h62
  set s : ℝ := Real.sqrt (A.relIndex H : ℝ) with hs
  calc (H.relIndex K : ℝ) * (A.relIndex H : ℝ) - 1
      = (A.relIndex H : ℝ) * (H.relIndex K : ℝ) - 1 := by ring
    _ ≤ 2 * ((H.relIndex K : ℝ) * (K.index : ℝ)) * s := by push_cast at h62 ⊢; linarith [h62]
    _ = 2 * (K.index : ℝ) * (H.relIndex K : ℝ) * s := by ring

/-- **Image in the centre of a quotient ⟺ commutator condition.**

For `N ◁ Γ` and `D ≤ Γ`, the image `D ⧸ N ↦ Γ ⧸ N` lands in the centre `Z(Γ ⧸ N)` iff every
`d ∈ D` commutes with every `g ∈ Γ` modulo `N` (`g d g⁻¹ d⁻¹ ∈ N`).  This is the elementwise
characterisation of "`D ⧸ N` central in `Γ ⧸ N`", letting the centrality hypothesis be transported
across the natural isomorphism `↥(H.subgroupOf K) ≃* ↥H` in `inducedMember_re_le_general`. -/
theorem map_mk'_le_center_iff {Γ : Type*} [Group Γ] {N D : Subgroup Γ} [N.Normal] :
    D.map (QuotientGroup.mk' N) ≤ Subgroup.center (Γ ⧸ N) ↔
      ∀ d ∈ D, ∀ g : Γ, g * d * g⁻¹ * d⁻¹ ∈ N := by
  rw [Subgroup.map_le_iff_le_comap]
  constructor
  · intro hle d hd g
    have hmem : QuotientGroup.mk' N d ∈ Subgroup.center (Γ ⧸ N) :=
      Subgroup.mem_comap.mp (hle hd)
    rw [Subgroup.mem_center_iff] at hmem
    have h2 := hmem (QuotientGroup.mk' N g)
    rw [← QuotientGroup.ker_mk' N, MonoidHom.mem_ker]
    simp only [map_mul, map_inv]
    rw [h2]; group
  · intro hcomm d hd
    rw [Subgroup.mem_comap, Subgroup.mem_center_iff]
    intro q
    obtain ⟨g, rfl⟩ := QuotientGroup.mk'_surjective N q
    have hthis : QuotientGroup.mk' N (g * d * g⁻¹ * d⁻¹) = 1 := by
      rw [← MonoidHom.mem_ker, QuotientGroup.ker_mk']; exact hcomm d hd g
    simp only [map_mul, map_inv] at hthis
    rw [mul_inv_eq_one] at hthis
    calc QuotientGroup.mk' N g * QuotientGroup.mk' N d
        = (QuotientGroup.mk' N g * QuotientGroup.mk' N d * (QuotientGroup.mk' N g)⁻¹)
            * QuotientGroup.mk' N g := by group
      _ = QuotientGroup.mk' N d * QuotientGroup.mk' N g := by rw [hthis]

open scoped Classical in
omit [Fintype G] [Invertible (Nat.card G : ℂ)] [Invertible (Nat.card ↥L : ℂ)] in
/-- **Peterfalvi (6.2) θ-bound for an induced member of a general (solvable) kernel `K`.**

For a section `B ≤ D ≤ C ≤ K` (Peterfalvi's `B ⊂ D ⊂ C ⊂ K`) and an irreducible character `θ` of `K`
trivial on `B`, with `D ⧸ B` central in `C ⧸ B`, the induced member `ψ = Ind_K^L θ` has degree
bounded by `ψ(1) ≤ |L:C|·√|C:D|`.  The §11/§13 consumer applies this two ways: `(C, D) = (H, A)`
(the central case feeding (6.3)) and `(C, D) = (HC, HC)` (Peterfalvi (11.4), `√1 = 1`).

This is the half of `(6.2)` the consumer needs in its **general** form: the kernel `K` (the derived
subgroup `M'`, solvable) is strictly larger than the section group `C`, so unlike the Sibley central
case (`psi_degree_le_of_source_central`, `K = H = C`) the Clifford a-half is required.  The proof
feeds the fully general `theta_degree_le_index_mul_sqrt_index` (a-half + b-half) the section
`CK = C.subgroupOf K`, `NB = (B.subgroupOf K).subgroupOf CK ≤ DK = (D.subgroupOf K).subgroupOf CK`:

* the kernel hypothesis `NB ⊆ Ker(Res_{CK} θ)` is `characterKernel_restrict_subgroupOf`;
* the index identities `|CK| = |K:C|` and `|DK| = |C:D|` are `rfl` / `relIndex_subgroupOf`;
* the centrality `DK ⧸ NB ⊆ Z(CK ⧸ NB)` is transported from the given `D ⧸ B ⊆ Z(C ⧸ B)` across the
  natural isomorphism `e : ↥CK ≃* ↥C` (`subgroupOfEquivOfLe`, which preserves the underlying `↥L`
  element) using the commutator characterisation `map_mk'_le_center_iff`;
* `ψ(1) = |L:K|·θ(1)` (`induce_apply_one`) and `|L:K|·|K:C| = |L:C|` (`relIndex_mul_index`). -/
theorem inducedMember_re_le_general
    {K C D B : Subgroup ↥L} [Invertible (Nat.card ↥K : ℂ)] [B.Normal]
    (hBD : B ≤ D) (hCK : C ≤ K)
    (θ : IrreducibleCharacter ↥K)
    (hθB : (↑(B.subgroupOf K) : Set ↥K) ⊆ OddOrder.Peterfalvi.S03.characterKernel
        (θ : ClassFunction ↥K ℂ))
    (hcentral : (D.subgroupOf C).map (QuotientGroup.mk' (B.subgroupOf C)) ≤
        Subgroup.center (↥C ⧸ B.subgroupOf C)) :
    (ClassFunction.induce K (θ : ClassFunction ↥K ℂ) 1).re ≤
      (C.index : ℝ) * Real.sqrt (Nat.card (↥C ⧸ D.subgroupOf C) : ℝ) := by
  have : Fintype ↥K := Fintype.ofFinite _
  have : (B.subgroupOf C).Normal := (‹B.Normal›).subgroupOf C
  set CK : Subgroup ↥K := C.subgroupOf K with hCK_def
  have : Fintype ↥CK := Fintype.ofFinite _
  have : Invertible (Nat.card ↥CK : ℂ) :=
    invertibleOfNonzero (by exact_mod_cast (Nat.card_pos (α := ↥CK)).ne')
  have : (B.subgroupOf K).Normal := (‹B.Normal›).subgroupOf K
  set NB : Subgroup ↥CK := (B.subgroupOf K).subgroupOf CK with hNB
  set DK : Subgroup ↥CK := (D.subgroupOf K).subgroupOf CK with hDK
  have : NB.Normal := (‹(B.subgroupOf K).Normal›).subgroupOf CK
  have hNBDK : NB ≤ DK := Subgroup.subgroupOf_mono CK (Subgroup.subgroupOf_mono K hBD)
  -- the kernel hypothesis: `θ` trivial on `B` ⟹ `Res_{C∩K} θ` trivial on `NB`.
  have hθNB : (↑NB : Set ↥CK) ⊆ OddOrder.Peterfalvi.S03.characterKernel
      (ClassFunction.restrict CK (θ : ClassFunction ↥K ℂ)) :=
    characterKernel_restrict_subgroupOf CK hθB
  -- centrality transport `D ⧸ B ⊆ Z(C ⧸ B)` ⟹ `DK ⧸ NB ⊆ Z(CK ⧸ NB)` across `e : ↥CK ≃* ↥C`.
  have hcentralCK : DK.map (QuotientGroup.mk' NB) ≤ Subgroup.center (↥CK ⧸ NB) := by
    rw [map_mk'_le_center_iff]
    rw [map_mk'_le_center_iff] at hcentral
    intro d hd c
    set e := Subgroup.subgroupOfEquivOfLe hCK with he_def
    have hcoe : ∀ x : ↥CK, ((e x : ↥C) : ↥L) = ((x : ↥K) : ↥L) := fun _ => rfl
    have hdD : e d ∈ D.subgroupOf C := by
      rw [Subgroup.mem_subgroupOf, hcoe]
      exact Subgroup.mem_subgroupOf.mp (Subgroup.mem_subgroupOf.mp hd)
    have key : (e c) * (e d) * (e c)⁻¹ * (e d)⁻¹ ∈ B.subgroupOf C := hcentral (e d) hdD (e c)
    have hcomm : (e c) * (e d) * (e c)⁻¹ * (e d)⁻¹ = e (c * d * c⁻¹ * d⁻¹) := by
      simp only [map_mul, map_inv]
    rw [hcomm, Subgroup.mem_subgroupOf, hcoe] at key
    rw [hNB, Subgroup.mem_subgroupOf, Subgroup.mem_subgroupOf]
    exact key
  -- the general (6.2) θ-bound: `θ(1) ≤ |K:C|·√|C:D|`.
  have hθbound :=
    theta_degree_le_index_mul_sqrt_index (K := ↥K) θ CK (N := NB) DK hNBDK hθNB hcentralCK
  -- index identities `|CK| = |K:C|` and `|DK| = |C:D|`.
  have hCKindex : CK.index = C.relIndex K := rfl
  have hDKindex : DK.index = Nat.card (↥C ⧸ D.subgroupOf C) := by
    have hd0 : DK.index = (D.subgroupOf K).relIndex CK := rfl
    rw [hd0, hCK_def, Subgroup.relIndex_subgroupOf hCK]; rfl
  -- `ψ(1) = |L:K|·θ(1)`.
  have hind : (ClassFunction.induce K (θ : ClassFunction ↥K ℂ) 1).re
      = (K.index : ℝ) * ((θ : ClassFunction ↥K ℂ) 1).re := by
    rw [ClassFunction.induce_apply_one, Complex.mul_re, Complex.natCast_re, Complex.natCast_im]
    ring
  rw [hind]
  calc (K.index : ℝ) * ((θ : ClassFunction ↥K ℂ) 1).re
      ≤ (K.index : ℝ) * ((CK.index : ℝ) * Real.sqrt (DK.index : ℝ)) :=
        mul_le_mul_of_nonneg_left hθbound (Nat.cast_nonneg _)
    _ = (C.index : ℝ) * Real.sqrt (Nat.card (↥C ⧸ D.subgroupOf C) : ℝ) := by
        rw [hDKindex, hCKindex, ← Subgroup.relIndex_mul_index hCK]; push_cast; ring

omit [Fintype G] [Invertible (Nat.card G : ℂ)] [Invertible (Nat.card ↥L : ℂ)] in
/-- **Peterfalvi (6.2), general Hypothesis (6.1) form** (`K` solvable, section `B ⊆ D ⊆ C ⊆ K`).

Given the **(5.6) norm-weighted coherence bound** `h56` — proved from Hypothesis (5.2)'s image
families by `S08.exists_source_index_le_two_psi_of_imageData`
(`S08_SixTwoThreeFromImageFamilies`, issue 0153), and in the concrete §11/§13 context from the
`muGrid`/`columnSum` columns (`S12.Hypothesis.exists_source_of_coherence_dichotomy`) — this
produces the (6.2) real-inequality `|K:A| − 1 ≤ 2|L:C|·√|C:D|`.  The §11/§13 consumer instantiates
it two ways: `(C, D) = (H, A)` feeds `six_three_index_bound_general` (hence (6.3) /
`S13.coherent_S_of_coherent_SH0C`, Peterfalvi (11.3)); `(C, D) = (HC, HC)` (`√1 = 1`) gives
`|M':H₁| − 1 ≤ 2|M:HC|`, Peterfalvi (11.4) `S13.coherent_quotient_bound`.

The oracle `h56` packages the break member as `ψ = Ind_K^L θ` (`θ ∈ Irr K`, `θ` trivial on `B`,
`θ ≠ 1`; every member of the induced family `S` has this shape) with the degree bound
`|K:A| − 1 ≤ 2ψ(1)`.  Everything *downstream* of `h56` is proved here generally: the θ-degree bound
`ψ(1) ≤ |L:C|·√|C:D|` (`inducedMember_re_le_general`) and the arithmetic, so the §11/§13 consumer's
remaining obligation is exactly `h56`. -/
theorem six_two_general
    {K A C D B : Subgroup ↥L} [Invertible (Nat.card ↥K : ℂ)] [B.Normal]
    (hBD : B ≤ D) (hCK : C ≤ K)
    (hcentral : (D.subgroupOf C).map (QuotientGroup.mk' (B.subgroupOf C)) ≤
        Subgroup.center (↥C ⧸ B.subgroupOf C))
    (h56 : ∃ θ : IrreducibleCharacter ↥K,
        (↑(B.subgroupOf K) : Set ↥K) ⊆ OddOrder.Peterfalvi.S03.characterKernel
            (θ : ClassFunction ↥K ℂ) ∧
        (Nat.card (↥K ⧸ A.subgroupOf K) : ℝ) - 1 ≤
            2 * (ClassFunction.induce K (θ : ClassFunction ↥K ℂ) 1).re) :
    (Nat.card (↥K ⧸ A.subgroupOf K) : ℝ) - 1 ≤
        2 * (C.index : ℝ) * Real.sqrt (Nat.card (↥C ⧸ D.subgroupOf C) : ℝ) := by
  obtain ⟨θ, hθB, hψbound⟩ := h56
  have hdeg := inducedMember_re_le_general hBD hCK θ hθB hcentral
  calc (Nat.card (↥K ⧸ A.subgroupOf K) : ℝ) - 1
      ≤ 2 * (ClassFunction.induce K (θ : ClassFunction ↥K ℂ) 1).re := hψbound
    _ ≤ 2 * ((C.index : ℝ) * Real.sqrt (Nat.card (↥C ⧸ D.subgroupOf C) : ℝ)) := by linarith [hdeg]
    _ = 2 * (C.index : ℝ) * Real.sqrt (Nat.card (↥C ⧸ D.subgroupOf C) : ℝ) := by ring

/-- **Peterfalvi (6.3), general (6.1) form, reduced to the (5.6) coherence oracle** — the
single-cite producer for the §11/§13 consumer.

Bundles the full standalone chain `six_three_descent ∘ six_three_index_bound_general ∘
six_two_general`: with `K` solvable normal, `H ≤ K` nilpotent (`K ≠ H` allowed), `M ≤ H₁ < H`, if
`S(H₁)` is coherent, `|H:H₁| > 4|L:K|² + 1`, **and** the per-section **(5.6) break-member oracle**
`h56` holds, then `S(M)` is coherent.

`h56` is the only remaining hypothesis with genuine character-theoretic content: for each section
`B ≤ A ≤ H₁` with `A ⧸ B` central in `H ⧸ B`, `S(A)` coherent and `S(B)` not, it asks for a break
member `ψ = Ind_K^L θ` (`θ ∈ Irr K` trivial on `B`) with `|K:A| − 1 ≤ 2ψ(1)`.  It is *proved* from
Hypothesis (5.2)'s image families in `S08_SixTwoThreeFromImageFamilies` (giving the oracle-free
`six_three_of_imageData`), and in the §11/§13 context from the `muGrid` columns
(`S12.Hypothesis.exists_source_of_coherence_dichotomy`).  Everything else — minimal-`A`/maximal-`B`
descent, nilpotency-forces-centrality, the θ-degree bound, and the `√`-arithmetic — is discharged
here.  This is the theorem `S13.coherent_S_of_coherent_SH0C` (Peterfalvi (11.3)) cites, applied
with `(L,K,M,H,H₁) = (M, M', 1, HC, H₀C)`. -/
theorem six_three_of_six_two_oracle
    {K H M H₁ : Subgroup ↥L} [Group.IsSolvable ↥K]
    [Invertible (Nat.card ↥K : ℂ)]
    [M.Normal] [H₁.Normal]
    [Group.IsNilpotent (↥H ⧸ M.subgroupOf H)] (hHnorm : H.Normal)
    (hMH₁ : M ≤ H₁) (hH₁H : H₁ < H) (hHK : H ≤ K)
    (τ : OddOrder.Peterfalvi.S07.IntegralCharacterMap ↥L G) (A0 : Set ↥L)
    (SOf : Subgroup ↥L → Set (ClassFunction ↥L ℂ))
    (h56 : ∀ (A B : Subgroup ↥L) [A.Normal] [B.Normal], B ≤ A → A ≤ H₁ →
        (A.subgroupOf H).map (QuotientGroup.mk' (B.subgroupOf H)) ≤
          Subgroup.center (↥H ⧸ B.subgroupOf H) →
        Nonempty (OddOrder.Peterfalvi.S07.IsCoherent τ (SOf A) A0) →
        ¬ Nonempty (OddOrder.Peterfalvi.S07.IsCoherent τ (SOf B) A0) →
        ∃ θ : IrreducibleCharacter ↥K,
          (↑(B.subgroupOf K) : Set ↥K) ⊆ OddOrder.Peterfalvi.S03.characterKernel
              (θ : ClassFunction ↥K ℂ) ∧
          (Nat.card (↥K ⧸ A.subgroupOf K) : ℝ) - 1 ≤
              2 * (ClassFunction.induce K (θ : ClassFunction ↥K ℂ) 1).re)
    (hcoh : Nonempty (OddOrder.Peterfalvi.S07.IsCoherent τ (SOf H₁) A0))
    (hbound : 4 * K.index ^ 2 + 1 < Nat.card (↥H ⧸ H₁.subgroupOf H)) :
    Nonempty (OddOrder.Peterfalvi.S07.IsCoherent τ (SOf M) A0) := by
  refine six_three_descent hHnorm hMH₁ hH₁H hHK τ A0 SOf ?_ hcoh hbound
  intro A B _ _ hBA hAH₁ _hAcomm hcentral hAcoh hBncoh
  exact six_three_index_bound_general hAH₁ (hAH₁.trans hH₁H.le) hHK
    (six_two_general (A := A) (C := H) (D := A) (B := B) hBA hHK hcentral
      (h56 A B hBA hAH₁ hcentral hAcoh hBncoh))

end OddOrder.Peterfalvi.S08
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/S08_Theorem62_63_Standalone.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