The cyclic quotient and explicit torus data of a Sylow normalizer in PSL₂
ProvedGlauberman.Dickson.h821_borel_quotient_datadickson-classificationgroup-theorypsl2
For a nontrivial Sylow p-subgroup P of H ≤ PSL₂(F), where F is a finite field of cardinality p^f, the normalizer N modulo its normal subgroup P is cyclic and its order divides the split torus order. Nonidentity quotient representatives act without nontrivial fixed points on P. After an inner conjugation of H, its normalizer lies in the standard unipotent-by-diagonal Borel subgroup. The theorem retains the explicit injective unipotent character, split-torus homomorphism and matrix/conjugation formulas from Huppert II.8.21.
Preamble
import Mathlib.Algebra.BigOperators.Ring.Nat import Mathlib.Algebra.CharP.CharAndCard import Mathlib.Algebra.Group.AddChar import Mathlib.Algebra.Polynomial.SpecificDegree import Mathlib.FieldTheory.Finite.GaloisField import Mathlib.GroupTheory.GroupAction.ConjAct import Mathlib.GroupTheory.SchurZassenhaus import Mathlib.GroupTheory.SpecificGroups.Dihedral import Mathlib.GroupTheory.Sylow import Mathlib.GroupTheory.Transfer import Mathlib.LinearAlgebra.Eigenspace.Basic import Mathlib.LinearAlgebra.Matrix.GeneralLinearGroup.Card import Mathlib.LinearAlgebra.Matrix.GeneralLinearGroup.FinTwo import Mathlib.LinearAlgebra.Matrix.GeneralLinearGroup.Projective import Mathlib.LinearAlgebra.Matrix.ProjectiveSpecialLinearGroup set_option autoImplicit false universe u v open scoped Pointwise Classical
Formal statement
theorem Glauberman.Dickson.h821_borel_quotient_data
{F : Type u} [Field F] [Finite F] {p f : ℕ} [Fact p.Prime]
(hFcard : Nat.card F = p ^ f) (H : Subgroup (Matrix.ProjectiveSpecialLinearGroup (Fin 2) F))
(P : Sylow p H) (hP_ne_bot : (P : Subgroup H) ≠ ⊥)
(N : Subgroup H) (hN : N = Subgroup.normalizer (P : Set H))
(PN : Subgroup N) [PN.Normal]
(hPN : PN = (P : Subgroup H).subgroupOf N) :
IsCyclic (N ⧸ PN) ∧
Nat.card (N ⧸ PN) ∣
(Nat.card F - 1) / Nat.gcd (Nat.card F - 1) 2 ∧
(∀ n : N, n ∉ PN → ∀ x : (P : Subgroup H), x ≠ 1 →
(n : H) * (x : H) * (n : H)⁻¹ ≠ (x : H)) ∧
∃ U T : Subgroup (Matrix.ProjectiveSpecialLinearGroup (Fin 2) F),
∃ conjH : H →* Matrix.ProjectiveSpecialLinearGroup (Fin 2) F,
Function.Injective conjH ∧
(∃ g : Matrix.ProjectiveSpecialLinearGroup (Fin 2) F, ∀ h : H,
conjH h = g * (h : Matrix.ProjectiveSpecialLinearGroup (Fin 2) F) * g⁻¹) ∧
IsMulCommutative U ∧
IsCyclic T ∧
Nat.card T =
(Nat.card F - 1) / Nat.gcd (Nat.card F - 1) 2 ∧
(∀ t : Matrix.ProjectiveSpecialLinearGroup (Fin 2) F, t ∈ T →
∀ x : Matrix.ProjectiveSpecialLinearGroup (Fin 2) F, x ∈ U →
t * x * t⁻¹ = x → t = 1 ∨ x = 1) ∧
(P : Subgroup H).map conjH ≤ U ∧
U ⊔ T ≤ Subgroup.normalizer
(U : Set (Matrix.ProjectiveSpecialLinearGroup (Fin 2) F)) ∧
(∀ x : Matrix.ProjectiveSpecialLinearGroup (Fin 2) F, x ∈ U ⊔ T → x ∉ U →
∃ u : Matrix.ProjectiveSpecialLinearGroup (Fin 2) F, u ∈ U ∧
x ∈ T.map (MulAut.conj u).toMonoidHom) ∧
(∀ n : N, conjH (n : H) ∈ U ⊔ T) ∧
(∀ n : N, conjH (n : H) ∈ U ↔ n ∈ PN) ∧
∃ unipotent : AddChar F (Matrix.ProjectiveSpecialLinearGroup (Fin 2) F),
∃ splitTorus : Fˣ →* Matrix.ProjectiveSpecialLinearGroup (Fin 2) F,
Function.Injective unipotent ∧
U = unipotent.toMonoidHom.range ∧
T = splitTorus.range ∧
(∀ a : Fˣ, ∀ x : F,
splitTorus a * unipotent x * (splitTorus a)⁻¹ =
unipotent ((a : F) ^ 2 * x)) ∧
(∀ x : F, unipotent x =
QuotientGroup.mk'
(Subgroup.center
(Matrix.SpecialLinearGroup (Fin 2) F))
(⟨!![1, x; 0, 1], by simp [Matrix.det_fin_two]⟩ :
Matrix.SpecialLinearGroup (Fin 2) F)) ∧
∀ a : Fˣ, splitTorus a =
QuotientGroup.mk'
(Subgroup.center
(Matrix.SpecialLinearGroup (Fin 2) F))
(⟨!![(a : F), 0; 0, (a⁻¹ : F)],
by simp [Matrix.det_fin_two]⟩ :
Matrix.SpecialLinearGroup (Fin 2) F) := by sorrySource
Original formalization: Qiuzhen-CFSG/CFSG, original source authors, Apache-2.0; commit 96b2a02085dc678f3e0a97b334c31ada599c55fd; https://github.com/Qiuzhen-CFSG/CFSG/blob/96b2a02085dc678f3e0a97b334c31ada599c55fd/Glauberman/DicksonSylowNormalizer.lean; declaration Glauberman.Dickson.h821_borel_quotient_data. arexychen contributes extraction, target-environment replay, theorem-DAG packaging and validation, not original authorship of Dickson classification.