On ergodicity of linear actions on $\mathbb{R}^n$ and factoriality of group von Neumann algebras

Chinmay Tamhankar; Soham Chakraborty

REVIEW 2 major objections 6 minor 12 references

On ergodicity of linear actions on $\mathbb{R}^n$ and factoriality of group von Neumann algebras

T0 review · 2 major / 6 minor · reviewed 2026-08-04 · deepseek-v4-flash

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

Pith's one-line read For n≥3, ergodicity of a linear action on R^n is not preserved under taking transposes of the acting group.

desk verdict Theorem A is a real counterexample and the paper is worth refereeing, but the write-up needs a cleanup pass: a broken reference, a dimension typo, and an underspecified citation in the key proof. read the letter →

arxiv 2510.23864 v3 pith:7OYGSCGD submitted 2025-10-27 math.OA math.DS

Soham Chakraborty , Chinmay Tamhankar This is my paper · ORCID

classification math.OAmath.DS MSC 37A1546L1022D25

keywords ergodicactionsgroupvonNeumannalgebrasfactorialitylinearonR^ndualactionsemidirectproductslocallycompactabeliangroupsessentialrangeofcocycles

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

The paper establishes that ergodicity of a measure-preserving linear action of a countable group on R^n does not imply ergodicity of the dual action (obtained by transposing and inverting the matrices), for every n≥3. It constructs a concrete subgroup Γ0 of SL(n,Z) whose linear action is ergodic but whose dual action is not. The paper then gives sufficient conditions – split action, faithful dually doubly ergodic connected part, and infinite orbits on the discrete quotient – under which the group von Neumann algebra L(N⋊Γ) is a factor for abelian Lie-type groups N. These conditions yield large families of examples, including actions of lattices and dense subgroups with torsion or orbit-infinite discrete parts.

What carries the argument

The main objects are the subgroup Γ0={A∈SL(n,Z):Ae1=e1}, the skew-product decomposition of its linear action over the ergodic action on R^{n-1} with a Z^{n-1}-valued cocycle, and the essential-range criterion for skew-product ergodicity. For the factoriality theorems, the key machinery is the isomorphism L(N⋊Γ)≅L^∞(N̂)⋊Γ, the diagonal (split) action decomposition N=N^0×N/N^0, and the use of double ergodicity (weak mixing) or mixing of one component to force ergodicity of the diagonal dual action.

What would settle it

Check whether the corollary of Schmidt's theorem quoted for the skew product requires the base action to be probability-preserving and the cocycle target to be compact. If it does, the essential-range computation E(c)=R alone is insufficient, and a direct search for a Γ0-invariant non-null-non-conull subset of R^n would settle ergodicity. Additionally, testing the Γ0-action on bounded measurable functions that depend on the first coordinate could reveal an invariant function.

Watch

Extended reading notes

Core claim

The central claim is an asymmetry: for n≥3, the map A↦(A^T)^{-1} on SL(n,R) is not inner, and the subgroup Γ0={A∈SL(n,Z):Ae1=e1} realizes the asymmetry. Its linear action on R^n is ergodic, proved by writing it as a skew product over the ergodic action of Γ0 on R^{n-1} and showing the associated cocycle has essential range all of R. Its dual action is non-ergodic because Γ0^T fixes the character e1, giving a non-scalar invariant function. The paper further proves that factoriality of L(N⋊Γ) for abelian Lie-type N without compact connected subgroups follows from a split action whose connected-component action is faithful and dually doubly ergodic and whose discrete-component action has infini

Load-bearing premise

The proof that Γ0 acts ergodically on R^n rests on an unstated hypothesis in the cited skew-product ergodicity criterion; if that criterion does not cover cocycles with noncompact target over infinite-measure base actions, then Theorem A's ergodicity claim is not established.

Editorial extensions

If this is right

Question 1.1 (ergodicity of an action iff ergodicity of its dual) has a negative answer in every dimension ≥3.
Any intrinsic characterization of ergodic countable subgroups of SL(n,R) cannot be invariant under the automorphism A↦(A^T)^{-1}.
For n≥3, lattices and countable dense subgroups Γ<SL(n,R), combined with discrete abelian D with infinite Γ-orbits, yield factorial L((R^n×D)⋊Γ).
When D is a torsion abelian group, every action Γ↷R^n×D splits automatically, so the same factoriality conclusion holds without the split hypothesis.
Factorial examples cover both injective and non-injective type II_∞ factors (e.g., SL(2,Z) vs lattices in SL(n,R), n≥3).

Reading between the lines

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

The non-equivalence of ergodicity under transpose suggests that no purely spectral or rigidity-based criterion for ergodicity of linear actions can be invariant under this automorphism, so the search for an intrinsic characterization may need orbit-equivalence or cocycle invariants.
The essential-range computation may generalize to other fixed-character subgroups, potentially giving more counterexamples in SL(n,R).
The split-action condition is probably removable in greater generality: the paper's own torsion case (Theorem 5.5) shows the twisted cocycle vanishes under a purely algebraic hypothesis; one might test whether finite-generation or cohomological vanishing of the discrete part achieves the same.
The type II_∞ examples here suggest a systematic route to factorial locally compact groups with prescribed modular behavior, using non-measure-preserving actions as the paper hints.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Desk Editor's Note

read the letter

The main result is genuine. For each n≥3, the subgroup Γ0 = {A∈SL(n,Z) : Ae1=e1} gives a linear action on R^n that is ergodic while its transpose action is not. The construction is simple and the essential-range computation is convincing. I think the proof goes through; the direct argument is available anyway since for a.e. v the translations {w·v} are dense in R, so an invariant function cannot depend on the first coordinate, and then SL(n-1,Z)-ergodicity of the base forces constancy. The paper's reliance on [Sch95, Cor. 5.4] without stating its hypotheses is a real gap in presentation—the stress-test concern about infinite-measure base and noncompact target is worth taking seriously, but it does not invalidate the result.

The factoriality criteria in Section 5 are useful. The split-action theorem (Theorem 5.2) is a clean combination of Glasner-Weiss weak mixing and the Schmidt-Walters mixing criterion, and the torsion-group variant (Theorem 5.5) gives a large class of examples without assuming the action splits. The authors are honest about the split-action restriction and correctly point to recent work by Morando and Miyamoto for nearby territory. The citation pattern is fine: the reference to [BCDK24] is to a published independent result, not to the present paper.

The soft spots are real but mostly cosmetic. Theorem 5.2's converse cites a nonexistent 'Theorem Thm: double ergodicity and cocycle superrigid implies full action ergodic'; that should be replaced by the obvious fact that for a diagonal product action, ergodicity of the product implies ergodicity of each factor. Proposition 5.11 has a dimension error—a subgroup of SL(2,Z) acts on R^2, not R^n, so the statement should be R^2. Corollary 5.6's proof is a one-liner and contains an R^m/R^n typo; the result is a direct consequence of Theorem 5.5, so the fix is straightforward. There are also a handful of small typos ('reamins', 'Z^d') that a careful revision should catch.

None of these issues undermines the central claims. The paper deserves a serious referee and should be accepted after a cleanup. I would bring it to a reading group if anyone cares about factorial locally compact group von Neumann algebras or ergodic linear actions.

Referee Report

2 major / 6 minor

Summary. The paper studies factoriality of the group von Neumann algebra L(N⋊Γ) when N is an abelian locally compact group of Lie type and Γ is countable discrete. By Pontryagin duality, L(N⋊Γ) is identified with the crossed product L^∞(\hat N)⋊Γ, so factoriality is analyzed through ergodicity and essential freeness of the dual action. The main results are: (Theorem A, proved in Proposition 4.5) for every n≥3, the group Γ0={A∈SL(n,Z): Ae1=e1} acts ergodically on R^n while its transpose Γ0^T acts non-ergodically, giving a negative answer to the paper's Question 1.1; and (Theorem B, realized in Theorems 5.2, 5.5, 5.9, 5.10) sufficient conditions for factoriality when N=R^m×D and the action splits, with corollaries covering actions of lattices and dense subgroups of SL(n,R), torsion discrete quotients, and Bernoulli-type discrete parts. The paper also contains criteria for actions on discrete abelian groups and examples involving SL(2,Z) subgroups.

Significance. If the main claims hold, Theorem A is a striking counterexample: ergodicity of a linear action on R^n is not invariant under the transpose automorphism for n≥3, despite being true for n=2. Theorem B provides a broad, explicit class of factorial group von Neumann algebras, including II_∞ examples and, via remark 5.12, non-injective factors from SL(n,R)-lattices. The paper is a pure theorem paper: there are no fitted parameters or numerical experiments, and the statements are constructive and checkable from standard results in ergodic theory and group duality. The overall strategy—using the essential range of a cocycle and applying Glasner–Weiss or Schmidt–Walters—is sound in principle. However, the current manuscript contains a load-bearing gap in the proof of Theorem A (the unstated hypotheses of the Schmidt essential-range theorem) and an undefined theorem reference in the proof of the necessity half of Theorem 5.2. These issues are local and fixable, but they must be addressed before the claims can be accepted.

major comments (2)

[§4.5 / §2.2] Proposition 4.5's proof of ergodicity of Γ0↷R^n is the decisive step for Theorem A. The proof computes that the base action α on R^{n-1} is ergodic and that the essential range of the cocycle c is R, then invokes [Sch95, Corollary 5.4] (and in §2.2 [Sch77, Corollary 5.4]) to conclude that the skew product is ergodic. However, the paper nowhere states the hypotheses of those corollaries. Here the base is R^{n-1} with infinite Lebesgue measure and the target is the noncompact group R. If Schmidt's result is stated only for probability-preserving bases or compact targets, the argument does not go through as written. Please state the theorem in full and verify its hypotheses, or supply a direct proof (for instance, using density of {w·v} for v in a co-null set and SL(n−1,Z)-ergodicity on R^{n-1}).
[§5.2, proof of Theorem 5.2] The converse part of Theorem 5.2, which supports the necessity claims in Theorem B, is justified by an undefined reference: 'by Theorem Thm: double ergodicity and cocycle superrigid implies full action ergodic'. No such theorem appears in the paper or in the reference list. The proof is incomplete as written. This should be replaced by a concrete argument or citation—for example, that factoriality of L∞(R^m×\hat D)⋊Γ implies ergodicity of the underlying dual action, together with Proposition 3.3 for the orbit condition.

minor comments (6)

[§4.5, Proposition 4.5] The sentence 'Γ0 acts trivially on the first copy of R' is false: the action sends (x,v) to (x + w^T v, Av), so the first coordinate is moved by the cocycle. What is meant is that R^n is realized as a skew product over the base R^{n-1}. Please rephrase.
[§4.5, Proposition 4.5] The notation 'g·F_w = F_w' is ambiguous: under the skew action the first coordinate changes, so the equality can only refer to the base action α_g. Clarify that α_g fixes F_w pointwise for the chosen g.
[§5, Proposition 5.11] The statement writes L((R^n × ⊕_Γ Λ)⋊Γ), but Γ is a finite-index subgroup of SL(2,Z), so the linear action is on R^2. Either replace R^n by R^2 or adjust the hypothesis to a lattice in SL(n,R) for the appropriate n.
[§5.2, Theorem 5.5 and Corollary 5.10] There are small wording/typos: Theorem 5.5 condition 2 says 'doubly dually ergodic' instead of 'dually doubly ergodic', and Corollary 5.10 says 'Φ|Γ is faithful' where it should say 'Φ|_{N^∘}' or 'η is faithful'. Please correct.
[§4.1, Lemma 4.1] In the statement of Lemma 4.1, 'SL(n,R^n)' should be 'SL(n,R)'.
[§2.2] The notation E(ω) is used for both the essential-value set in the one-point compactification and for its intersection with T; consider distinguishing these to avoid confusion.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: Theorem A and Theorem B derivations are self-contained reductions to external ergodic-theory results; the [BCDK24] self-citation is contextual and not load-bearing.

full rationale

The paper is a pure theorem paper: no parameters are fitted and no data are predicted, so the fitted-input pattern does not apply. Proposition 4.5 (Theorem A) proves ergodicity of Γ0↷R^n by decomposing the action as a skew product over α on R^{n-1}, establishing α is ergodic via SL(n−1,Z) (external Moore ergodicity), and proving E(c)=R directly by an explicit open-set argument; the final inference uses [Sch95, Cor. 5.4], an external standard theorem. The reviewer-identified issue that the hypotheses of that corollary (infinite-measure base, noncompact target) are not verified is a correctness/missing-support concern, not a circular reduction: no displayed equation identifies ergodicity with essential range by construction. Theorem B’s factoriality criteria are proved by reducing L(N⋊Γ) via Theorem 2.2 to crossed products and then applying [GW16, Theorem 1.1], [SW82, Theorem 2.3], and Proposition 3.3; the hypotheses (faithful, dually doubly ergodic/dually ergodic, infinite orbits/stabilizers) are properties of the dual action, not restatements of the factoriality conclusion. The self-citation [BCDK24] appears in the introduction and around Theorem 3.4, but the paper gives direct proofs of the needed equivalences (3⇔4 from [AP10]; 1⇔2 from Propositions 3.2–3.3; 2⇔3 in the text), so it is not load-bearing. Editorial defects (e.g., the unresolved cross-reference in Theorem 5.2's converse, and the stray 'instead?' before Corollary 4.8) are presentation gaps, not circularity. No circular step satisfies the reduction-by-construction standard, so the honest finding is 0.

Assumptions & free parameters 0 free parameters · 8 assumptions · 0 invented entities

No free parameters or invented entities. The paper's proof rests on standard duality and structure theorems, several published ergodic-theoretic black boxes, and an explicit splitting restriction.

assumptions (8)

standard math Pontryagin duality and structure theorem for second-countable LCA groups: Lie-type N ≅ R^m × T^n × D, and factoriality forces T^n trivial (Thm 2.5, Cor 2.7).
Reduces N to R^m×D throughout; standard background result.
standard math L(N⋊Γ) ≅ L∞(\hat N)⋊Γ for discrete Γ acting on abelian LCA N by automorphisms (Thm 2.2, [BH20]).
Bridges group von Neumann algebras and crossed products; central to all factoriality arguments.
standard math [GW16, Thm 1.1]: double ergodicity of a nonsingular action implies weak mixing.
Used in Thm 5.1 to combine α with pmp β.
standard math [SW82, Thm 2.3]: mixing pmp × properly ergodic nonsingular is ergodic.
Used in Thm 5.9.
standard math [Sch95, Cor 5.4]: skew product ergodic if base ergodic and cocycle has full essential range.
Used in Prop 4.5; hypotheses not stated in the paper.
standard math [Sch95, Thm 1.6]: dual action on \hat D is mixing iff every non-trivial stabilizer on D is finite.
Used in Thm 5.9 and Cor 5.10.
standard math [PV11, Lemma 5.6]: lattice actions Γ↷R^n are doubly ergodic for n≥3; dense subgroups inherit double ergodicity.
Provides the examples in Cor 5.3 and Cor 5.6.
domain assumption Action splits: Φ(N/N^0)=N/N^0 (or N/N^0 torsion), so the dual action is diagonal.
Assumed in Thms 5.2 and 5.9; the Introduction states nonsplit actions are not handled.

how reviews work

0 comments

Cite this review

Pith. "Pith review of On ergodicity of linear actions on $\mathbb{R}^n$ and factoriality of group von Neumann algebras." pith.science (2026). https://pith.science/paper/7OYGSCGD

@misc{pith2026251023864,
  author       = {Pith},
  title        = {Pith review of: On ergodicity of linear actions on $\mathbbR^n$ and factoriality of group von Neumann algebras},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/7OYGSCGD}},
  note         = {Machine review of arXiv:2510.23864}
}

read the original abstract

We give some natural conditions on actions of discrete countable groups on abelian locally compact groups of Lie type that imply factoriality of the group von Neumann algebras of their semidirect products. This allows us to give a fairly large class of examples of locally compact groups whose group von Neumann algebras are factors.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

12 extracted references · 1 linked inside Pith

[1]

Macmillan Co

Macmillan Lectures in Mathematics. Macmillan Co. of India, Ltd., Delhi, 1977, p

1977
[3]

The development of the theory ofp-groups

Publications of the Mathematical Society of Japan. Mathematical Society of Japan, Tokyo, 1956, pp. vii+99. [Hil02] Paul Hill. “The development of the theory ofp-groups”. In:Pro- ceedings of the Second Honolulu Conference on Abelian Groups and Modules (Honolulu, HI, 2001). Vol

1956
[7]

[Mor25] Basile Morando.The regular representation of Neretin groups is factorial

arXiv:2505.21253 [math.OA]. [Mor25] Basile Morando.The regular representation of Neretin groups is factorial

arXiv
[8]

On rings of operators. IV

arXiv:2506.24029 [math.OA]. [MN43] F. J. Murray and J. von Neumann. “On rings of operators. IV”. In:Ann. of Math. (2)44 (1943), pp. 716–808. [Neu40] J. v. Neumann. “On rings of operators. III”. In:Ann. of Math. (2)41 (1940), pp. 94–161. [PV11] Sorin Popa and Stefaan Vaes. “Cocycle and orbit superrigidity for lattices in SL(n,R) acting on homogeneous space...

arXiv 1943
[25]

Skew product action

De Gruyter Studies in Mathematics. De Gruyter, Berlin, 2023, pp. xli+1032. [Kim06] Hyun Jung Kim. “Skew product action”. In:Int. J. Contemp. Math. Sci.1.5-8 (2006), pp. 205–211. [Kul45] Leonid Kulikov. “On the theory of abelian groups of arbitrary cardinality”. In:Mat. Sb.16(58).2 (1945), pp. 129–162. 26 [Mel76] O. V. Melnikov. “Compactness conditions for...

2023
[29]

Gen` eve: Universit´ e de Gen` eve, 1981, pp

Monograph of the En- seignement Math´ ematique. Gen` eve: Universit´ e de Gen` eve, 1981, pp. 39–46. [BH20] Bachir Bekka and Pierre de la Harpe.Unitary representations of groups, duals, and characters. Vol

1981
[81]

Birkh¨ auser Verlag, Basel, 1984, pp

Monographs in Mathematics. Birkh¨ auser Verlag, Basel, 1984, pp. x+209. 28

1984
[250]

Factoriality of groupoid von Neumann algebras

Mathematical Surveys and Monographs. American Mathematical Society, Providence, RI, 2020, pp. xi+474. [BCDK24] Tey Berendschot, Soham Chakraborty, Milan Donvil, and Se- Jin Kim. “Factoriality of groupoid von Neumann algebras”. In: to appear in Journal of Operator Theory(2024). [Dan01] S. G. Dani. “On ergodicZ d-actions on Lie groups by automor- phisms”. I...

2020

Show all 12 references

[1371]

Factoriality of twisted locally compact group von Neumann algebras

27 [Vae20] Stefaan Vaes.Unusual crossed product constructions being fac- tors. 2020.url:https://mathoverflow.net/q/362329(vis- ited on 01/17/2024). [Vae25] Stefaan Vaes. “Factoriality of twisted locally compact group von Neumann algebras”. In:C. R. Math. Acad. Sci. Paris363 (2...

2020
[1995]

Mildly mixing actions of locally compact groups

[SW82] Klaus Schmidt and Peter Walters. “Mildly mixing actions of locally compact groups”. In:Proc. London Math. Soc. (3)45.3 (1982), pp. 506–518. [Sut78] Colin E. Sutherland. “Type analysis of the regular representa- tion of a nonunimodular group”. In:Pacific J. Math.79.1 (19...

1982
[2010]

Deux actions de SL(2,Z)

25 [Aub81] P.-L. Aubert. “Deux actions de SL(2,Z)”. In:Ergodic The- ory (Les Plans-sur-Bex, 1980). Vol

1980
[2025]

[Hal56] Paul R

arXiv:2504.08107 [math.OA]. [Hal56] Paul R. Halmos.Lectures on ergodic theory. Vol

Pith tools

Reviewed August 4, 2026 · model on record in the stance chip above.

[1] [1]

Macmillan Co

Macmillan Lectures in Mathematics. Macmillan Co. of India, Ltd., Delhi, 1977, p

1977

[2] [3]

The development of the theory ofp-groups

Publications of the Mathematical Society of Japan. Mathematical Society of Japan, Tokyo, 1956, pp. vii+99. [Hil02] Paul Hill. “The development of the theory ofp-groups”. In:Pro- ceedings of the Second Honolulu Conference on Abelian Groups and Modules (Honolulu, HI, 2001). Vol

1956

[3] [7]

[Mor25] Basile Morando.The regular representation of Neretin groups is factorial

arXiv:2505.21253 [math.OA]. [Mor25] Basile Morando.The regular representation of Neretin groups is factorial

arXiv

[4] [8]

On rings of operators. IV

arXiv:2506.24029 [math.OA]. [MN43] F. J. Murray and J. von Neumann. “On rings of operators. IV”. In:Ann. of Math. (2)44 (1943), pp. 716–808. [Neu40] J. v. Neumann. “On rings of operators. III”. In:Ann. of Math. (2)41 (1940), pp. 94–161. [PV11] Sorin Popa and Stefaan Vaes. “Cocycle and orbit superrigidity for lattices in SL(n,R) acting on homogeneous space...

arXiv 1943

[5] [25]

Skew product action

De Gruyter Studies in Mathematics. De Gruyter, Berlin, 2023, pp. xli+1032. [Kim06] Hyun Jung Kim. “Skew product action”. In:Int. J. Contemp. Math. Sci.1.5-8 (2006), pp. 205–211. [Kul45] Leonid Kulikov. “On the theory of abelian groups of arbitrary cardinality”. In:Mat. Sb.16(58).2 (1945), pp. 129–162. 26 [Mel76] O. V. Melnikov. “Compactness conditions for...

2023

[6] [29]

Gen` eve: Universit´ e de Gen` eve, 1981, pp

Monograph of the En- seignement Math´ ematique. Gen` eve: Universit´ e de Gen` eve, 1981, pp. 39–46. [BH20] Bachir Bekka and Pierre de la Harpe.Unitary representations of groups, duals, and characters. Vol

1981

[7] [81]

Birkh¨ auser Verlag, Basel, 1984, pp

Monographs in Mathematics. Birkh¨ auser Verlag, Basel, 1984, pp. x+209. 28

1984

[8] [250]

Factoriality of groupoid von Neumann algebras

Mathematical Surveys and Monographs. American Mathematical Society, Providence, RI, 2020, pp. xi+474. [BCDK24] Tey Berendschot, Soham Chakraborty, Milan Donvil, and Se- Jin Kim. “Factoriality of groupoid von Neumann algebras”. In: to appear in Journal of Operator Theory(2024). [Dan01] S. G. Dani. “On ergodicZ d-actions on Lie groups by automor- phisms”. I...

2020