Alphabeta Math
TheoremStatement: Literature-sourcedProof: AI-adaptedPipeline-generatedprecheck passjudge pass (gpt-6.1-sol)audited 2026-10-08
How statement and proof provenance work

The first chip identifies the source of the statement or construction; the second identifies the source of its local proof or verification.

  • Literature-sourced: the exact statement appears in a cited source; only wording and notation differ.
  • AI-adapted: a semantically identical restatement of literature-sourced material, modulo indexing, notation, and boundary cases adopted by the library.
  • AI-generated: a genuinely novel statement formulated by AI, with no source for the claim itself.

These labels describe origin, not correctness: citations and verification chips remain separate evidence.

The double commutant theorem for concrete von Neumann algebras

Statement

Assume the Axiom of Choice. Let H be a complex Hilbert space and let M⊆B(H) be a unital ∗-subalgebra closed in the weak operator topology. Then M′′=M, and M is also closed in the strong operator topology. Consequently the concrete von Neumann algebras of Von Neumann algebras and commutants are exactly the unital ∗-subalgebras that are closed in either of these topologies. For an arbitrary set S⊆B(H) one has W∗(S)′′=W∗(S).

Facts & Assumptions

Given: AC, a complex Hilbert space H, the bounded-operator space B(H), and either a unital ∗-subalgebra M closed in WOT or a set S⊆B(H).

[F1]

A concrete von Neumann algebra is a unital ∗-subalgebra closed in WOT; for any set A, its commutant is WOT-closed, and if A is self-adjoint then A′ is a unital ∗-subalgebra. The generated algebra W∗(S) is the WOT closure of the unital ∗-algebra generated by S (Von Neumann algebras and commutants).

[F2]

SOT is initial for the maps T↦Tξ in norm, whereas WOT is initial for the scalar maps T↦φ(Tξ); hence SOT is finer than WOT (Strong and weak operator topologies).

[F3]

The finite Hilbert direct sum H⊕n has norm ∥(ξj)∥2=∑j=1n∥ξj∥2, and its coordinate inclusions and projections are bounded (Hilbert direct sums of unitary representations).

[F4]

Every closed linear subspace C of a Hilbert space has a unique orthogonal decomposition H⊕n=C⊕C⊥; the orthogonal projection PC is the map selecting the C-component (Orthogonal decomposition by a closed subspace, The Hilbert orthogonal projection onto a closed subspace).

[F5]

Every bounded operator has a unique Hilbert adjoint satisfying ⟨Rξ,η⟩=⟨ξ,R∗η⟩, and adjoints respect composition; for a diagonal operator on H⊕n, the same identity on each coordinate gives (R⊕n)∗=(R∗)⊕n (The Hilbert-space adjoint of a bounded operator, Hilbert-adjoint identities).

[F6]

The operator norm is the unit-ball supremum and satisfies ∥Rξ∥≤∥R∥∥ξ∥ (The operator norm as the least bound and as the unit-sphere or unit-ball supremum).

[F7]

AC implies Countable Choice, which is the premise of the orthogonal-decomposition and Hilbert-adjoint suppliers (The Axiom of Choice, AC implies DC implies countable choice, The Axiom of Countable Choice (ACω)).

Proof

technique · direct finite-block approximation

Given: AC, H, and the algebra or set in the Statement.

1.1F1F3F4F5F6F7construct

Let A⊆B(H) be any unital ∗-subalgebra and fix T∈A′′. Fix a finite tuple ξ1,…,ξn∈H; if n=0 there is nothing to prove, so assume n≥1. Put K=H⊕n and ξ=(ξ1,…,ξn), and define DR=R⊕n for R∈B(H). By [F3,F6], ∥DRζ∥2=∑j=1n∥Rζj∥2≤∥R∥2∥ζ∥2, so DR is bounded; [F5] gives DR∗=DR∗. The orbit {DRξ:R∈A} is a linear subspace because R↦DRξ is linear, so its closure C is a closed subspace. For each R∈A, DRDSξ=DRSξ for S∈A, so DRC⊆C by continuity of the bounded operator DR; because R∗∈A, the same argument gives DR∗C⊆C. If ζ∈C⊥ and c∈C, then ⟨DRζ,c⟩=⟨ζ,DR∗c⟩=0, so DRC⊥⊆C⊥. Let PC be the orthogonal projection from [F4]. Uniqueness of the orthogonal decomposition makes PC linear, and orthogonality gives ∥ζ∥2=∥PCζ∥2+∥ζ−PCζ∥2, so it is bounded. Since DR preserves both C and C⊥, it commutes with PC. For coordinate inclusions ιj:H→K and projections πi:K→H, put Pij=πiPCιj∈B(H). Comparing the (i,j) blocks of PCDR=DRPC gives PijR=RPij; equality of all finite blocks is equality of the operators for every R∈A, so Pij∈A′. The condition T∈A′′ gives TPij=PijT for every i,j, hence DTPC=PCDT. Since IH∈A, ξ=DIHξ∈C and PCξ=ξ; therefore DTξ=PCDTξ∈C. For any ε>0, the definition of C supplies R∈A with ∥DTξ−DRξ∥<ε, which yields ∥Tξj−Rξj∥<ε for every j. Thus one element of A approximates T simultaneously on any prescribed finite tuple.

2.1F1F2step 1.1

If M is a unital ∗-subalgebra closed in WOT and T∈M′′, step 1.1 puts T in the SOT closure of M, since its finite-tuple conclusion is exactly the SOT neighborhood test [F2]. WOT is coarser than SOT [F2], so a WOT-closed set is SOT-closed and T∈M. Conversely, M⊆M′′ by the commutant definition [F1]. Hence M′′=M, and M is SOT-closed.

2.2F1step 1.1

If A is a unital ∗-subalgebra closed in SOT, then A⊆A′′ and step 1.1 gives A′′⊆A‾ SOT=A. Thus A=A′′. Since A is self-adjoint, A′ is a unital ∗-subalgebra and is WOT-closed [F1]; its commutant A′′ is WOT-closed as well [F1]. Therefore A is WOT-closed, proving the reverse closure implication.

3.1F1F2step 1.1step 2.1∎

For any S⊆B(H), let A=Alg⁡∗(S∪{IH}) and W=A‾ WOT=W∗(S) [F1]. Step 1.1 and the SOT-to-WOT continuity in [F2] give A′′⊆W. On the other hand, A′′ is WOT-closed and contains A [F1], so the minimality of WOT closure gives W⊆A′′. Thus W=A′′; in particular W is a unital ∗-subalgebra closed in WOT, and step 2.1 applied to W gives W′′=W. Therefore W∗(S)′′=W∗(S).

Source qualifications

Blackadar's I.9.1.1 explicitly labels its proof an outline: it reduces finite-tuple approximation to a one-vector orbit and cites I.2.5.4 for the tensor-block computation. The argument above writes the finite direct-sum block computation out. Bekka--de la Harpe state the closure/bicommutant equivalences in Theorem A.K.1 and refer its proof to Dixmier--von Neumann, Chapter I, §3, no. 4; their cited theorem is not treated as a proof here.

Depends on

Used by

Dependency tree · two levels

41 results within two dependency steps of this one, each drawn at its shortest distance from it. An arrow runs from a result to what uses it, so the chart reads left to right and ends at this result, which carries a heavier outline. Every node is a link to that result. Click elsewhere on the chart to enlarge it.

Sources