Skip

Operator algebras · 3 manuscript(s)

Strong Kadison–Kastler stability and its spatial boundaries

Verified by Lean
Open on GitHub

1 · The essentials

The central question is whether two sufficiently similar collections of operators must be related by a very small change of coordinates. Operators act on a Hilbert space, the mathematical setting shared by the collections in this article. For collections called unital von Neumann algebras, the main result says yes: sufficiently close collections are related by a unitary change of coordinates as close to doing nothing as desired. The required closeness depends only on how small that change must be, not on the particular algebras or Hilbert space. The main result has a Lean formalization. But the companion manuscripts show that the conditions matter: approximation in only one direction need not allow a small change of coordinates. They also show that arbitrarily close collections of the broader kind called norm-separable C*-algebras need not be related by any unitary change of coordinates at all.

A picture to help (an imperfect image, not to be taken literally)

As an imperfect image, imagine two collections of objects that nearly match, and ask whether one tiny rigid movement can align the entire collections. The main result guarantees a suitable analogue of that movement under its particular mathematical conditions; the companion results warn that one-way matching or a different kind of collection can defeat the conclusion.

Why it matters

The main manuscript describes its result as resolving the strong Kadison–Kastler conjecture, with a tolerance uniform across algebras, representations, and Hilbert spaces. The companion manuscripts distinguish closeness from the existence of a small implementing unitary, and even from the existence of any implementing unitary. Together, they identify both a stability statement and limits on extending it.

2 · Going further

The main problem concerns unital von Neumann algebras and acting on the same Hilbert space. Their Kadison–Kastler distance measures the operator-norm distance between their unit balls, requiring approximation in both directions. Universal strong stability says that, for every prescribed , there is a tolerance such that yields a unitary with and . Crucially, depends only on , uniformly over the algebras, representations, and Hilbert spaces.

The main abstract describes this as resolving the strong Kadison–Kastler conjecture. Its central assertion is the passage from two-sided approximation of unit balls to exact conjugacy by a unitary close to the identity. The supplied abstracts do not describe the argument that establishes this passage, so no proof mechanism can be extracted from them. The main result has a Lean formalization, although the supplied material gives no more detailed account of its scope.

The second manuscript examines one-sided near inclusions instead. It constructs pairs of unital von Neumann algebras with common identity on separable complex Hilbert spaces whose one-sided gaps tend to zero. Spatial embeddings of the source into the target exist, but every implementing unitary remains a fixed positive distance from the identity. Thus the obstruction shown in the manuscript is specifically to small implementing unitaries, not to spatial embeddings themselves.

The third manuscript changes the class of algebras. For every , it constructs unital, norm-separable C*-algebras with the same identity on a common separable complex Hilbert space, at Kadison–Kastler distance less than , that are not conjugate by any unitary. The two algebras even have the same von Neumann closure. This is presented as a negative answer to the separable C*-algebraic spatial form of the conjecture, rather than a contradiction of the von Neumann algebra result.

Useful notions

  • Operators on a Hilbert space
  • Operator norm and unit balls
  • Unital operator algebras
  • Unitary conjugacy
  • One-sided versus two-sided approximation

What was known before / what is new

Before

The source identifies the strong Kadison–Kastler conjecture, the unrestricted small-spatial-embedding problem, and the separable C*-algebraic spatial form of the conjecture as the questions being addressed. It does not provide earlier partial results, a chronology, or prior methods.

New

The main result supplies a universal tolerance guaranteeing conjugacy by a unitary arbitrarily close to the identity for sufficiently close unital von Neumann algebras. The second manuscript shows that one-sided near inclusion does not give an unrestricted guarantee of small spatial embeddings. The third shows that arbitrarily close unital, norm-separable C*-algebras can fail to be unitarily conjugate, even when their von Neumann closures agree.

Limits and assumptions

The positive result concerns sufficiently close unital von Neumann algebras on the same Hilbert space, with closeness measured in both directions between unit balls. The one-sided counterexamples concern small implementing unitaries despite the existence of spatial embeddings; the C*-algebra counterexamples rule out any ambient unitary conjugacy. The main result has a Lean formalization, but the supplied material does not specify its detailed coverage or state that either companion result is formalized. No argument details, explicit tolerance, or further remaining open questions are supplied.

Mini glossary

Kadison–Kastler distance
—
A measure of how closely the unit balls of two operator algebras approximate one another in operator norm, in both directions.
Unitary conjugacy
—
A relation in which a unitary change of coordinates takes one algebra exactly onto the other, expressed as .
One-sided near inclusion
—
An approximation condition from a source algebra toward a target algebra, without requiring approximation in the reverse direction.
Spatial embedding
—
An embedding of one algebra into another implemented by a unitary change of coordinates.
Universal tolerance
—
A closeness threshold depending only on the desired distance of the implementing unitary from the identity, not on the algebras, representations, or Hilbert spaces.

Keywords : Kadison–Kastler stability · Unitary conjugacy · Near inclusions

AI-generated explanation based on the manuscript. Unofficial. Refer to the manuscript for the authoritative version.

3 · Technical details and sources

Claim made in the manuscript (source text)

Proves that sufficiently close unital von Neumann algebras on the same Hilbert space are conjugate by a unitary arbitrarily close to the identity, with a universal tolerance in the operator-norm distance between unit balls. Counterexamples show that near-identity conjugacy fails for one-sided near inclusions, and that arbitrarily close norm-separable C∗-algebras need not be ambiently unitarily conjugate.

Verified by Lean: the Lean software checked the part of the reasoning described below, sometimes only part of the manuscript. This does not mean independent mathematicians have validated the result.

Scope of the Lean formalization (source text)

…

lean/docs/289.md

Manuscripts in this family

★

Universal strong Kadison–Kastler stability

23/09/2026
To be verified

We prove that sufficiently close unital von Neumann algebras are conjugate by a unitary arbitrarily close to the identity. The tolerance depends only on the prescribed distance of that unitary from the identity, uniformly over all algebras, representations, and Hilbert spaces. This resolves the strong Kadison–Kastler conjecture.

2.

Near Inclusions of von Neumann Algebras Without Small Spatial Embeddings

05/10/2026
To be verified

We give a negative answer to the unrestricted small-spatial-embedding problem for one-sided near inclusions of von Neumann algebras. On separable complex Hilbert spaces, we construct pairs of unital von Neumann algebras with common identity whose one-sided gaps tend to zero. Spatial embeddings of the source into the target exist, but every implementing unitary stays a fixed positive distance from the identity. The obstruction therefore concerns small implementing unitaries, rather than the existence of spatial embeddings.

3.

Close Separable C*-Algebras Without Spatial Conjugacy

05/10/2026
To be verified

We disprove the separable C∗-algebraic spatial form of the Kadison–Kastler conjecture. For every ε > 0, we construct unital, norm-separable C∗-algebras on a common separable complex Hilbert space, with the same identity and Kadison–Kastler distance less than ε, that are not conjugate by any unitary. The two algebras have the same von Neumann closure.