Documentation

Q2Presentation.Boundary.MaximalPro2

The maximal pro-2 quotient of a profinite group (Stage C, work order: MaximalPro2) #

This file is a general, Q₂-independent module of profinite group theory. For an arbitrary profinite group P it constructs the maximal pro-2 quotient MaximalPro2Quotient P, the canonical surjection maxPro2Proj P, and proves the universal property characterising it. In the manuscript (reference/q2_manuscript.tex) this is the construction underlying Proposition 3.10 (prop:pro2, the maximal pro-2 quotient Π of the candidate Γ_A) and the marked local Demushkin comparison (the maximal pro-2 quotient of G_{Q₂} is ≅ ℤ₂ on the unramified line); the boundary isomorphisms consume maxPro2_universal as their load-bearing tool.

Mathlib status #

Mathlib v4.28.0 has no pro-p / maximal pro-p notion (a search for IsProP, MaximalPro, "pro-p" finds only unrelated CategoryTheory.ObjectProperty.Ind files). We therefore build it here, mirroring three proven mathlib/​project templates:

Construction #

The index is the full subcategory Pro2Index P of OpenNormalSubgroup P on those open normal U whose quotient P ⧸ U is a finite 2-group (Pro2Prop). It is cofiltered (instIsCofilteredPro2Index): closed under ⊓ because P ⧸ (U ⊓ V) embeds in P ⧸ U × P ⧸ V, a product of 2-groups. The maximal pro-2 quotient is the inverse limit, in ProfiniteGrp, of the restricted finite-quotient diagram (pro2Diagram = (Pro2Prop P).ι ⋙ ProfiniteGrp.diagram P):

Pro-2 groups #

A profinite group P is pro-2 if every finite continuous quotient is a 2-group. Finite continuous quotients are exactly the quotients P ⧸ U by open normal subgroups U, so this says P ⧸ U is a 2-group for every U : OpenNormalSubgroup P.

Equations
Instances For

    A 2-group from an intersection: the inf-closure ingredient #

    theorem Q2Presentation.isPGroup_two_prod {A : Type u_1} {B : Type u_2} [Group A] [Group B] (hA : IsPGroup 2 A) (hB : IsPGroup 2 B) :
    IsPGroup 2 (A × B)

    A product of two 2-groups is a 2-group (used to show the pro-2 index is closed under ⊓).

    noncomputable def Q2Presentation.quotInfProdHom {G : Type u_1} [Group G] (U V : Subgroup G) [U.Normal] [V.Normal] :
    G ⧸ U ⊓ V →* (G ⧸ U) × G ⧸ V

    The homomorphism P ⧸ (U ⊓ V) →* (P ⧸ U) × (P ⧸ V) induced by the two quotient maps; it is injective, exhibiting P ⧸ (U ⊓ V) as (isomorphic to) a subgroup of the 2-group P ⧸ U × P ⧸ V.

    Equations
    Instances For
      theorem Q2Presentation.isPGroup_two_quotient_inf {G : Type u_1} [Group G] {U V : Subgroup G} [U.Normal] [V.Normal] (hU : IsPGroup 2 (G ⧸ U)) (hV : IsPGroup 2 (G ⧸ V)) :
      IsPGroup 2 (G ⧸ U ⊓ V)

      The quotient by an intersection of two open normal subgroups with 2-group quotients is again a 2-group.

      The pro-2 index category #

      The property selecting the open normal subgroups U ⊴ P whose quotient P ⧸ U is a 2-group.

      Equations
      Instances For
        @[reducible, inline]

        The pro-2 index category: the full subcategory of OpenNormalSubgroup P on the open normal subgroups with 2-group quotient. The maximal pro-2 quotient is the limit over this category.

        Equations
        Instances For

          The whole group ⊤ as an open normal subgroup.

          Equations
          Instances For

            The whole group as a pro-2 index object (P ⧸ ⊤ is trivial, hence a 2-group); witnesses nonemptiness of the index.

            Equations
            Instances For

              The intersection of two pro-2 index objects, again a pro-2 index object (isPGroup_two_quotient_inf).

              Equations
              Instances For

                The pro-2 index category is cofiltered. Binary cones are intersections infIndex; the parallel-arrow half is automatic (the index is a preorder, hence thin); nonemptiness is topIndex.

                theorem Q2Presentation.pro2_exists_dominating {P : ProfiniteGrp.{u}} (J : Finset (Pro2Index P)) :
                ∃ (B : Pro2Index P), ∀ A ∈ J, Nonempty (B ⟶ A)

                Finite cofiltered domination. Any finite set of pro-2 indices is dominated by a single index admitting a morphism to each (mirrors GammaA.exists_dominating).

                The maximal pro-2 quotient as an inverse limit #

                The restricted finite-quotient diagram: each pro-2 index U maps to the finite 2-group P ⧸ U (viewed as a discrete profinite group).

                Equations
                Instances For

                  The maximal pro-2 quotient MaximalPro2Quotient P, the cofiltered inverse limit (in ProfiniteGrp) of the finite 2-group quotients of P.

                  Equations
                  Instances For

                    The coordinate projection MaximalPro2Quotient P ⟶ P ⧸ U at a pro-2 index U (evaluation of the compatible family at U; mirrors Boundary/Z2Limit.z2Projection).

                    Equations
                    Instances For

                      Two points of MaximalPro2Quotient P agree once all their coordinate projections agree.

                      The underlying monoid homomorphism of the canonical projection: p ↦ (U ↦ mk p) (mirrors ProfiniteGrp.toLimitFun).

                      Equations
                      Instances For

                        The canonical projection maxPro2Proj P : P ⟶ MaximalPro2Quotient P (mirrors ProfiniteGrp.toLimit).

                        Equations
                        Instances For

                          The canonical projection followed by the coordinate at U is the finite quotient projection P ⟶ P ⧸ U.

                          The maximal pro-2 quotient is pro-2 #

                          Coordinate domination. Every open normal subgroup W of MaximalPro2Quotient P contains the kernel of some coordinate projection coordProj U. (The coordinate kernels form a neighbourhood basis of 1; a single pro-2 index U dominates the finitely many coordinates pinned down by the open set W, using cofilteredness of the index.)

                          The maximal pro-2 quotient is pro-2: every finite continuous quotient of MaximalPro2Quotient P is a 2-group. Given an open normal W, the domination lemma provides a coordinate projection φ = coordProj U with ker φ ≤ W; then M ⧸ ker φ ≅ range φ ≤ P ⧸ U is a 2-group, and M ⧸ W is a quotient of it.

                          The canonical projection is surjective #

                          The image of maxPro2Proj P is dense (mirrors ProfiniteGrp.denseRange_toLimit): a nonempty open set fixes finitely many coordinates; a single dominating pro-2 index U₀ collapses them; and P surjects onto the coordinate P ⧸ U₀, so the lift of the chosen value agrees with the target point on those coordinates.

                          The canonical projection is surjective. Its range is closed (continuous image of a compact space) and dense (maxPro2Proj_denseRange), hence everything.

                          The universal property #

                          For a pro-2 target Q and f : P ⟶ Q, every open normal V ⊴ Q pulls back to a pro-2 index comapIndex f hQ V = f⁻¹V of P (its quotient embeds in the 2-group Q ⧸ V). The induced finite quotient maps P ⧸ f⁻¹V ⟶ Q ⧸ V form a cone over diagram Q with apex MaximalPro2Quotient P; lifting it through Q.isLimitCone yields the universal g. This mirrors ProfiniteGrp.ProfiniteCompletion.lift/lift_unique.

                          theorem Q2Presentation.isPGroup_two_quotient_comap {G : Type u_1} {H : Type u_2} [Group G] [Group H] (φ : G →* H) {V : Subgroup H} [V.Normal] (hV : IsPGroup 2 (H ⧸ V)) :

                          Pulling back an open normal subgroup V ⊴ Q along f to a pro-2 index of P: its quotient P ⧸ f⁻¹V injects into the 2-group Q ⧸ V.

                          noncomputable def Q2Presentation.comapIndex {P Q : ProfiniteGrp.{u}} (f : P ⟶ Q) (hQ : IsProP2 Q) (V : OpenNormalSubgroup ↑Q.toProfinite.toTop) :

                          The pro-2 index f⁻¹V of P attached to an open normal V ⊴ Q (with Q pro-2).

                          Equations
                          • One or more equations did not get rendered due to their size.
                          Instances For
                            noncomputable def Q2Presentation.comapIndexHom {P Q : ProfiniteGrp.{u}} (f : P ⟶ Q) (hQ : IsProP2 Q) {V W : OpenNormalSubgroup ↑Q.toProfinite.toTop} (h : V ≤ W) :
                            comapIndex f hQ V ⟶ comapIndex f hQ W

                            Monotonicity of comapIndex: a transition V ⟶ W of open normals of Q induces a transition f⁻¹V ⟶ f⁻¹W of pro-2 indices of P.

                            Equations
                            Instances For
                              noncomputable def Q2Presentation.quotientMapFG {P Q : ProfiniteGrp.{u}} (f : P ⟶ Q) (hQ : IsProP2 Q) (V : OpenNormalSubgroup ↑Q.toProfinite.toTop) :

                              The finite quotient map P ⧸ f⁻¹V ⟶ Q ⧸ V induced by f.

                              Equations
                              • One or more equations did not get rendered due to their size.
                              Instances For

                                The cone over diagram Q with apex MaximalPro2Quotient P assembling the induced finite quotient maps; its limit lift is the universal morphism.

                                Equations
                                • One or more equations did not get rendered due to their size.
                                Instances For
                                  noncomputable def Q2Presentation.maxPro2Lift {P Q : ProfiniteGrp.{u}} (f : P ⟶ Q) (hQ : IsProP2 Q) :

                                  The universal morphism MaximalPro2Quotient P ⟶ Q lifting f : P ⟶ Q (for Q pro-2).

                                  Equations
                                  Instances For

                                    The induced finite quotient projection P ⟶ P ⧸ f⁻¹V followed by the quotient map P ⧸ f⁻¹V ⟶ Q ⧸ V equals f followed by Q ⟶ Q ⧸ V.

                                    Uniqueness: any g with f = maxPro2Proj P ≫ g equals maxPro2Lift f hQ. Two such morphisms agree on the dense range of maxPro2Proj P.

                                    Universal property of the maximal pro-2 quotient. For any pro-2 group Q and continuous f : P ⟶ Q, there is a unique g : MaximalPro2Quotient P ⟶ Q with f = maxPro2Proj P ≫ g. This is the load-bearing tool used by the boundary isomorphisms.