Pith. sign in

REVIEW 2 major objections 4 minor 26 references

Kaplansky's second test problem in operator algebras

T0 review · 2 major / 4 minor · reviewed 2026-08-01 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Doubling an operator cannot hide a similarity failure: two copies similar forces one copy similar in the settings proved here.

desk verdict A serious partial answer to Kaplansky's second test problem; the flagged gap in Proposition 3.21 does not survive contact with the paper. read the letter →

arxiv 2607.20593 v1 pith:LT4GNKHH submitted 2026-07-22 math.FA math.OA

Chunlan Jiang , Minghui Ma , Rui Shi , Yuanhang Zhang This is my paper · ORCID
classification math.FAmath.OA MSC 47A4547A65
keywords Kaplansky'ssecondtestproblemsimilaritytypeI_nvonNeumannalgebraproperty(J)stronglyirreducibledecompositionJacobsonradicalessentiallyfinite-dimensionalcommutantdirectsums
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Kaplansky's second test problem asks whether similarity of doubled elements is enough to conclude similarity of the original elements. This paper proves that the answer is yes for a large class: any operator with property (J) inside a type I_n von Neumann algebra — meaning its relative commutant contains a bounded maximal abelian family of idempotents — and, in the same algebras with n≤3, for every operator without exception. It proves the analogous cancellation for elements of unital Banach algebras whose relative commutant is finite-dimensional modulo the Jacobson radical. The strategy is to decompose operators into strongly irreducible summands, prove that such decompositions are unique up to similarity, and then show that a similarity between doubled operators must match the summands one by one. If the results stand, the long-open test problem is settled affirmatively across these settings and tied to a concrete reduction to matrix-valued functions on ℓ∞.

What carries the argument

Property (J): an operator T in a von Neumann algebra M has property (J) when its relative commutant {T}'∩M contains a bounded maximal abelian family of idempotents. In type I_n algebras this is equivalent to having a finite strongly irreducible decomposition, and the paper's Theorem 3.26 shows such decompositions are unique up to similarity — the key mechanism that turns a similarity between T1⊕T1 and T2⊕T2 into a matching of irreducible summands. The Banach-algebra version replaces idempotent families by the quotient Q(T,B)=(T,B)'/Rad((T,B)'); finite dimensionality of this quotient forces a finite Wedderburn block decomposition, enabling the same matching argument.

What would settle it

Exhibit, for some n≥4 and abelian von Neumann algebra A, an operator T in Mn(A) such that T⊕T is similar to S⊕S in M2(Mn(A)) but T is not similar to S; equivalently, find t≥1 with ρ_n(t)=∞, or produce an idempotent in the relative commutant of a direct sum of strongly irreducible upper-triangular operators whose difference from its diagonal part is not in the radical.

Watch

Extended reading notes

Core claim

On its own terms, the paper establishes two cancellation theorems. Theorem 3.27: if T1 has property (J) in a type I_n von Neumann algebra M and T1⊕T1 is similar to T2⊕T2 in M2(M), then T1 is similar to T2 in M. Theorem 5.13: in any unital Banach algebra B, if T1 has essentially finite-dimensional commutant and T1^(n) is similar to T2^(n) in Mn(B), then T1 is similar to T2 in B. The proof hinges on showing that every finite strongly irreducible decomposition of an operator with property (J) is unique up to similarity: any two bounded maximal abelian families of idempotents in the relative commutant are conjugate by an invertible element of that commutant. For 1≤n≤3, the paper shows property (

Load-bearing premise

In Proposition 3.21, the assertion that an idempotent in the commutant of a direct sum of strongly irreducible upper-triangular operators is similar to the diagonal formed from its diagonal entries uses the claim that the strictly upper-triangular part of that commutant lies in its radical; this inclusion is asserted but not proved, and the later theorems inherit it.

Editorial extensions

If this is right

  • In every type I_n von Neumann algebra, operators with property (J) pass Kaplansky's second test problem: T1⊕T1∼T2⊕T2 implies T1∼T2.
  • For type I_1, I_2, and I_3 algebras the test problem has an unconditional affirmative answer, since property (J) holds automatically.
  • The reduction theorem makes the full type I_n question equivalent to a numerical condition on the similarity function ρ_n, so it suffices to analyze Mn(ℓ∞).
  • In unital Banach algebras, elements with finite-dimensional commutant modulo the radical also pass the test; in particular this covers strictly cyclic quasinilpotent weighted shifts.
  • As the paper notes, these results connect to local unitary equivalence of quantum states, where a related cancellation question has been posed.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If the same diagonalization mechanism works without property (J) in higher dimensions, Question 1.1 would be resolved; the natural test case is n=4 in Mn(ℓ∞), where the similarity function ρ_4 could be studied numerically or constructively.
  • The proof's reliance on the unproved containment of the strictly upper-triangular commutant in the radical suggests a precise stress test: check that containment in a concrete 3+2 block example, since Proposition 3.21's conclusion depends on it.
  • The Banach-algebra theorem may extend to elements whose commutant is finite-dimensional modulo some other two-sided ideal, provided the quotient remains semisimple enough for a Wedderburn-type decomposition to apply.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

2 major / 4 minor

Summary. The paper studies Kaplansky's second test problem for similarity: if T⊕T is similar to S⊕S in M2(B), must T be similar to S in B? The authors prove an affirmative answer for operators with property (J) in type In von Neumann algebras (Theorem 3.27), and for elements with essentially finite-dimensional commutant in unital Banach algebras (Theorem 5.13). They also prove a reduction theorem (Theorem 4.3) allowing reduction to Mn(ℓ∞), remove the property (J) assumption for 1 ≤ n ≤ 3 (Proposition 4.7), and discuss applications to weighted shifts and to local unitary equivalence of quantum states.

Significance. If the proofs are completed, this is a substantial contribution to a long-standing problem in operator theory. The Banach-algebra part, culminating in Theorem 5.13, is elegant and appears to be correct; the radical/Wedderburn argument is transparent and gives a genuinely new general result. The reduction theorem (Theorem 4.3) is also a valuable structural tool. The von Neumann-algebra part rests on a long structural analysis (Proposition 3.21) whose proof is not fully written out, and on a case analysis for n = 3 that contains an unjustified algebraic step. These gaps are local and likely repairable, but they are load-bearing for the main type-In claims.

major comments (2)
  1. [Proposition 3.21, §3.4] The proof of this key lemma is incomplete. After Case I and Case II, the text says that Case III follows by a combination of the two, but it only proves the single configuration N=7, n1=3, n2=n3=2, and the concluding 'General Case' paragraph merely reduces to Case III by grouping traces. Since Proposition 3.21 drives Corollaries 3.22 and 3.23 and hence Theorem 3.27, a complete argument for arbitrary block decompositions is needed, not just an illustrative example.
  2. [Proposition 4.7, Case 2.1, §4.2] The proof uses a11^{-1} in the construction around the claim '|p22,11| ≥ 1/2 IA up to similarity', while the standing assumption in Case 2 is only R(a11)=IA. In a commutative von Neumann algebra, full range projection does not imply invertibility (for example, multiplication by x on L∞[0,1] has range projection I but no bounded inverse). The argument needs a spectral-localization step (or a different device) before inverting a11. As written, this invalidates the proof of the n=3 case, which is a stated main result.
minor comments (4)
  1. [Proposition 3.21, Case II, §3.4] The step 'Since P11 and Tj,j+1 are diagonal, Pjj = P11' is terse. It is correct, but only after using that each diagonal entry of Tj,j+1 has range projection IA, so multiplication by that entry is injective on A and forces the off-diagonal entries of Pjj to vanish. Please spell this out; otherwise the claim appears to depend on invertibility of Tj,j+1, which is not available.
  2. [Proposition 3.21, Cases I and II] The assertion that a strictly (block) upper-triangular element of the relative commutant lies in the radical is not proved. It is true: the set of such elements is a two-sided ideal and is nilpotent. A one-sentence justification would improve the exposition.
  3. [Proposition 4.7, Case 2.1] The expression 'x = a11^{-1} a22 = (IA - p22,11)^{-1}p12,11' contains an undefined symbol 'a22' and appears garbled. The intended formula should be stated cleanly.
  4. [Theorem 3.27] The letter Q is used both for the maximal abelian family and for the idempotent X^{-1}(P1⊗0)X, which is confusing. Use different notation for the family and the idempotent.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity found; main theorems rest on internal lemmas, and self-citations are not load-bearing.

full rationale

The derivation chain for the two central theorems is internally supported rather than circular. Theorem 3.27 is proved from the finite strongly irreducible decomposition theorem (3.17) and the uniqueness-up-to-similarity theorem (3.26); Theorem 3.26 is proved from Proposition 3.21, whose proof uses internal structural lemmas (2.4, 2.5, 2.6, 3.8, 3.10, 3.18) and classical external tools such as Rosenblum's equation and Deckard–Pearcy determinacy results. Theorem 5.13 similarly builds on Wedderburn's theorem and an internal computation (Proposition 5.11) of essentially relative commutants. No fitted parameter is renamed as a prediction, and no equation in the paper reduces to its own conclusion by construction. The self-citations that occur—[10], [11], [13], [19], [20]—appear in the introduction or final remarks as motivation, context, or applications (including the quantum-state remark), and are not load-bearing premises for the main theorems. The reviewer concern about Proposition 3.21 Case II, where the inference P_jj = P_11 from an intertwining equation involving T_{j,j+1} requires invertibility of T_{j,j+1}, is a potential proof gap rather than a circular step, because the argument does not assume the theorem it is trying to prove. Accordingly, no circularity is exhibited, and the score is 0.

Assumptions & free parameters 0 free parameters · 5 assumptions · 0 invented entities

The proofs are built on standard operator-algebra machinery. No free parameters are fitted. The introduced definitions (property (J), essentially finite-dimensional commutant, similarity number) are properties or functions rather than postulated entities with independent evidence.

assumptions (5)
  • standard math Rosenblum's theorem: BX - XA = Q is solvable when the spectra of A and B are disjoint
    Used in Corollary 2.7 to show intertwiners vanish.
  • standard math Wedderburn's theorem for finite-dimensional semisimple algebras over C
    Used in Lemma 5.5 and Theorem 5.13 to decompose Q(T,B).
  • domain assumption Deckard-Pearcy selection theorem for continuous matrix-valued functions on Stonian spaces
    Bridges pointwise similarity bounds to global similarity in Theorem 4.3.
  • domain assumption Shields' commutant theorem and maximal ideal space results for H∞_β
    Used in Proposition 5.16 to compute Q(T) for weighted shifts.
  • standard math Analytic functional calculus for idempotents in von Neumann algebras and Banach algebras
    Used to construct idempotents from spectral projections in Proposition 2.6 and Lemma 5.5.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Kaplansky's second test problem in operator algebras." pith.science (2026). https://pith.science/paper/LT4GNKHH

@misc{pith2026260720593,
  author       = {Pith},
  title        = {Pith review of: Kaplansky's second test problem in operator algebras},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/LT4GNKHH}},
  note         = {Machine review of arXiv:2607.20593}
}
abstract

Kaplansky's second test problem on similarity asks: if $T$ and $S$ are elements in a unital Banach algebra $\mathcal{B}$ and $T\oplus T$ is similar to $S\oplus S$ in $\mathbb{M}_2(\mathcal{B})$, is $T$ similar to $S$ in $\mathcal{B}$? We answer this problem affirmatively if $T$ is an operator with property $(J)$ in a type $\mathrm{I}_n$ von Neumann algebra $\mathcal{M}$, i.e., $\{T\}'\cap\mathcal{M}$ contains a bounded maximal abelian family of idempotents. Moreover, the condition of property $(J)$ can be removed for $1\leqslant n\leqslant 3$. A similar result is proved if $T$ is an element in a unital Banach algebra $\mathcal{B}$ with essentially finite-dimensional commutant, i.e., the relative commutant of $T$ in $\mathcal{B}$ is finite-dimensional modulo its Jacobson radical. Finally, we point out that one of our main results can be applied to the implementation of local unitary (LU) equivalence of quantum states.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

26 extracted references · 1 canonical work pages

  1. [1]

    G. Allan. Introduction to Banach spaces and algebras. Prepared for publication and with a preface by H. Garth Dales. Oxf. Grad. Texts Math., 20 Oxford Uni- versity Press, Oxford, 2011. viii+371 pp. MR2761146

    2011

  2. [2]

    E. Azoff. Borel measurability in linear algebra. Proc. Amer. Math. Soc. 42 (1974), 346–350. MR0327799 KAPLANSKY’S SECOND TEST PROBLEM IN OPERATOR ALGEBRAS 53

    1974

  3. [3]

    Bercovici

    H. Bercovici. Three test problems for quasisimilarity. Canad. J. Math. 39 (1987), no. 4, 880–892. MR0915020

    1987

  4. [4]

    Cowen, R

    M. Cowen, R. Douglas. Complex geometry and operator theory. Acta Math. 141 (1978), no. 3-4, 187–261. MR0501368

    1978

  5. [5]

    Davidson

    K. Davidson. C ∗-algebras by example. Fields Inst. Monogr., 6 American Mathe- matical Society, Providence, RI, 1996. xiv+309 pp. MR1402012

    1996

  6. [6]

    Davidson, D

    K. Davidson, D. Herrero. The Jordan form of a bitriangular operator. J. Funct. Anal. 94 (1990), no. 1, 27–73. MR1077544

    1990

  7. [7]

    Deckard, C

    D. Deckard, C. Pearcy. On continuous matrix-valued functions on a Stonian space. Pacific J. Math. 14 (1964), 857–869. MR0172130

    1964

  8. [8]

    Gilfeather

    F. Gilfeather. Strong reducibility of operators. Indiana Univ. Math. J. 22 (1972/73), 393–397. MR0303322

    1972

Show all 26 references
  1. [9]

    Heisenberg

    W. Heisenberg. Über quantentheoretische Umdeutung kinematischer und mecha- nischer Beziehungen. Z. Physik 33, 879–893 (1925). doi.org/10.1007/BF01328377

    1925 doi

  2. [10]

    C. Jiang. Similarity classification of Cowen-Douglas operators. Canad. J. Math. 56 (2004), no. 4, 742–775. MR2074045

    2004

  3. [11]

    Jiang, Y

    C. Jiang, Y. Cao. Advance in operator theory and Jiang’s seven questions. Com- mun. Math. Res. 39 (2023), no. 2, 287–295. MR4580037

    2023

  4. [12]

    Jiang, X

    C. Jiang, X. Guo, K. Ji, K-group and similarity classification of operators. J. Funct. Anal. 225 (2005), no. 1, 167–192. MR2149922

    2005

  5. [13]

    Jiang, R

    C. Jiang, R. Shi. Direct integrals of strongly irreducible operators. J. Ramanujan Math. Soc. 26 (2011), no. 2, 165–180. MR2816786

    2011

  6. [14]

    Kadison, J

    R. Kadison, J. Ringrose. Fundamentals of the theory of operator algebras, I, Elementary theory. Academic Press, New York, 1983. MR0719020

    1983

  7. [15]

    Kadison, J

    R. Kadison, J. Ringrose. Fundamentals of the theory of operator algebras, II, Advanced theory. Academic Press, Orlando, FL, 1986. MR0859186

    1986

  8. [16]

    Kadison, I

    R. Kadison, I. Singer. Three test problems in operator theory. Pacific J. Math. 7 (1957), 1101–1106. MR0092123

    1957

  9. [17]

    Kaplansky

    I. Kaplansky. Infinite abelian groups. University of Michigan Press, Ann Arbor, MI, 1954. v+91 pp. MR0065561

    1954

  10. [18]

    B. Kraus. Local unitary equivalence of multipartite pure states. Phys. Rev. Lett. 104 (2010), no. 2, 020504, 4 pp. MR2585388

    2010

  11. [19]

    M. Ma, R. Shi. Bargmann invariants and local unitary equivalence. arXiv:2607.16878 [quant-ph] doi.org/10.48550/arXiv.2607.16878

    doi arXiv

  12. [20]

    Marcoux, H

    L. Marcoux, H. Radjavi, S. Troscheit, Y. Zhang. Stability relations for Hilbert space operators and a problem of Kaplansky. Math. Z. 308 (2024), no. 3, Paper No. 39, 46 pp. MR4803020 54 CHUNLAN JIANG, MINGHUI MA, RUI SHI, AND YUANHANG ZHANG

    2024

  13. [21]

    Rosenblum, On the operator equation BX XA = Q

    M. Rosenblum, On the operator equation BX XA = Q. Duke Math. J. 23 (1956), 263–269. MR0079235

    1956

  14. [22]

    D. Sherman. Divisible operators in von Neumann algebras. Illinois J. Math. 54 (2010), no. 2, 567–600. MR2846474

    2010

  15. [23]

    A. Shields. Weighted shift operators and analytic function theory. Topics in Op- erator Theory, Math. Surveys, vol. 13, Amer. Math. Soc., Providence, R.I., 1974, 49–128. MR0361899

    1974

  16. [24]

    M. P. Thomas. Quasinilpotent strictly cyclic unilateral weighted shift operators on `p which are not unicellular. Proc. London Math. Soc. (3) 51 (1985), no. 1, 127–145. MR0788853

    1985

  17. [25]

    Zhang, B

    L. Zhang, B. Xie. A Survey of Bargmann Invariants: Geometric Foundations and Application. 2026, arXiv:2601.01858

    2026

  18. [26]

    Zhang, B

    L. Zhang, B. Xie, Y. Tao. Bargmann-invariant framework for local unitary equiv- alence and entanglement. Phys. Rev. A 112 (2025), no. 5, Paper No. 052426, 28 pp. MR5002102 Chunlan Jiang, School of Mathematics, Hebei Normal University, Shijiazhuang, 050016, China Email address ...

    2025

Pith tools

Reviewed August 1, 2026 · model on record in the stance chip above.