Embeddings of $L^p$-operator algebras

Eusebio Gardella; Jan Gundelach

REVIEW 1 major objections 4 minor 34 references

Embeddings of $L^p$-operator algebras

T0 review · 1 major / 4 minor · reviewed 2026-08-03 · deepseek-v4-flash

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

Pith's one-line read For p≠2, a unital contractive homomorphism between reduced L^p-algebras of Weyl twists is isometric and respects the diagonal exactly when it is induced by a groupoid morphism, and this rigidity rules out several embeddings that exist in th

desk verdict Strong and ambitious paper, but the proof of the key automatic-preservation step (Prop 3.33) has a real gap that the main theorem leans on. read the letter →

arxiv 2601.15204 v2 pith:OMT7FVV5 submitted 2026-01-21 math.FA math.OA

Eusebio Gardella , Jan Gundelach This is my paper · ORCID

classification math.FAmath.OA MSC 22A2246H0546L0547L10

keywords L^p-operatoralgebrasgroupoidC*-algebrasWeyltwistsspatialnormalizersactorsAF-embeddabilityCuntztopologicalfullgroups

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 tries to establish a rigidity theorem for L^p-operator algebras at p≠2: algebras built from twisted etale groupoids admit almost no 'soft' embeddings, because every reasonable embedding is forced to respect the groupoid structure underneath. The central claim is that a unital contractive homomorphism between reduced L^p-algebras of Weyl twists with compact unit spaces is isometric and respects the canonical projection onto the diagonal subalgebra if and only if it is induced by a diagram of groupoid morphisms—an 'actor'—so the embedding is really the shadow of a groupoid morphism. A sympathetic reader should care because this converts otherwise intractable embedding questions into concrete groupoid questions, and it leads to results that contradict the C*-world: irrational-rotation L^p-algebras do not embed into spatial AF L^p-algebras, and O_2^p ⊗_p O_2^p does not embed into O_2^p, so the classical O_2-embedding theorem of C*-algebra theory has no L^p analogue for p≠2.

What carries the argument

The two central objects are spatial normalizers and actors. A spatial normalizer is an element of the algebra that realizes a partial homeomorphism of the diagonal; it is built from an MP-partial isometry (the Banach-algebraic analogue of a partial isometry) in the double dual, and off p=2 the structure theorem for partial isometries on L^p-spaces identifies these with the spatial partial isometries. An actor between twists is a consistent lifting operation: a free action of one twisted groupoid on another that commutes with right multiplication, equivalently an inverse semigroup homomorphism of bisections plus an equivariant anchor map of unit spaces. The machinery does two things: Theorem

What would settle it

Find a p∈(1,∞), p≠2 and two Weyl twists (G,Σ),(H,Ω) with compact unit spaces for which there exists a unital isometric homomorphism φ:F_p^λ(G,Σ)→F_p^λ(H,Ω) with φ∘E_G=E_H∘φ that is not of the form ι_*∘π_* for any intermediate Weyl twist; Theorem 4.20 says no such map exists. A more local falsifier is a single unital contractive homomorphism in this setting that does not send one spatial normalizer to a spatial normalizer, which would refute the automatic-preservation theorem and the actor construction.

Watch

Extended reading notes

Core claim

The central discovery is Theorem 4.20. Let (G,Σ) and (H,Ω) be Weyl twists—twisted etale groupoids whose base is Hausdorff and effective—with compact unit spaces, and let p∈(1,∞), p≠2. A unital contractive homomorphism φ:F_p^λ(G,Σ)→F_p^λ(H,Ω) is isometric and intertwines the conditional expectations (equivalently, is injective on the diagonal C(X)) if and only if there is an intermediate Weyl twist (G·_h Y, Σ·_β Y) and twist homomorphisms π:Σ→Σ·_β Y (surjective, fiberwise bijective) and ι:Σ·_β Y→Ω (injective, unitwise bijective, open image) such that φ=ι_*∘π_*, i.e. φ(f) is the zero-extension of f∘π_β. In words: every embedding is a groupoid morphism in disguise. The proof first shows (Theore

Load-bearing premise

The load-bearing assumption is the automatic-preservation theorem for p≠2: a unital contractive homomorphism must send the diagonal subalgebra and its normalizing partial isometries into the corresponding objects of the target; if that preservation fails for some pair of algebras, the induced actor—and with it the entire groupoid-level description—collapses.

Editorial extensions

If this is right

Embeddability between reduced L^p-groupoid algebras for p≠2 becomes a groupoid problem: one looks for an actor diagram, and non-existence can often be certified by groupoid invariants alone.
A principal Weyl groupoid algebra embeds into a spatial AF L^p-algebra if and only if the underlying groupoid is AF; consequently the L^p irrational rotation algebra A_θ^p has no unital contractive embedding into a spatial AF L^p-algebra (Corollary 5.6).
An isometric embedding of L^p-groupoid algebras induces an embedding of the associated topological full groups (Theorem 6.11), opening a route from Banach-algebra rigidity to group rigidity.
For p≠2 there is no unital contractive map O_2^p ⊗_p O_2^p → O_2^p, so no L^p analogue of the classical O_2-embedding theorem exists; more generally, a unital contractive map between tensor products of L^p-Cuntz algebras forces m≤n (Theorem 7.8 and Corollary 7.9).

Reading between the lines

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

Editorial inference: the sharp break at p=2 suggests an actual phase transition in embeddability rather than a technical gap; one could test whether the class of embeddable L^p-groupoid algebras shrinks continuously as p moves away from 2, or whether the transition is abrupt.
Editorial inference: the m≤n theorem does not say whether m≤n is sufficient; a natural test is to try to construct an actor from a product of m shifts of finite type to a product of n shifts whenever m≤n, which would show the bound is sharp.
Editorial inference: because the proof uses compact unit spaces, Hausdorffness, and effectiveness, extending the normalizer-preservation argument to non-Hausdorff or non-effective twists would expose which hypothesis is truly load-bearing; the main theorem predicts rigidity should fail or require new tools there.

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.

Desk Editor's Note

read the letter

I’ll cut to the chase: this is a strong, ambitious paper, and the main theorem is a real advance—but there’s a proof gap in a load-bearing spot that needs fixing before I’d trust the applications.

The genuinely new content is Theorem 4.20: for p≠2, unital contractive homomorphisms between reduced L^p-groupoid algebras that preserve the canonical conditional expectation are exactly those induced by a free actor between Weyl twists. That extends the isometric-isomorphism rigidity of Hetland–Ortega to arbitrary embeddings, and it uses spatial normalizers and actor theory in a way that is genuinely different from the C*-setting. The corollaries are sharp: irrational rotation L^p-algebras don’t embed into spatial AF L^p-algebras, and there’s no unital contractive map from O_2^p ⊗_p O_2^p to O_2^p. The topological-full-group application to Brin–Thompson groups is a nice bonus, and the paper is well-written and well-organized.

The soft spot is Proposition 3.33. The proof defines Φ(u) as a weak-* limit of φ(ue_i) and then says that because Φ inherits unitality and contractivity, it preserves MP-partial isometries by Remark 3.13. That’s not an argument. The MP-conditions Φ(u)Φ(u†)Φ(u)=Φ(u), Φ(u†)Φ(u)Φ(u†)=Φ(u†), and the hermitianity of the products are not automatic under weak-* limits, and the proof never establishes that Φ is multiplicative on the relevant partial isometries. The earlier claims about nets in the proof show the limit is independent of the approximating net, but they don’t give multiplicativity. This matters because Theorem 3.34 uses Banach–Lamperti on Φ(u) to conclude it is a spatial partial isometry; without the MP-partial-isometry conclusion, that step fails, and Theorem 4.12 and the whole characterization collapse. It may well be fixable—one can imagine working with the algebra elements φ(uh) and canceling φ(h) on its support—but it’s not in the paper.

A separate, lesser concern: Corollary 7.9 leans on an unpublished result [22, Cor. 11.19] for Brin–Thompson group rigidity. That’s a dependency, not a flaw, but it makes the O_2^p statement conditional.

Specialists in L^p-operator algebras and groupoid rigidity should read this paper, and it deserves a serious referee. I’d send it back asking for a repaired Proposition 3.33, and I wouldn’t cite the main theorem in my own work until that’s settled.

Referee Report

1 major / 4 minor

Summary. The manuscript develops a rigidity theory for embeddings of L^p-operator algebras associated with étale groupoids. Its main result, Theorem 4.20, asserts that for p∈(1,∞)\{2\} and compact-unit Weyl twists, a unital contractive homomorphism between reduced twisted groupoid algebras that intertwines the canonical conditional expectations is necessarily induced by a groupoid-level actor diagram. The proof passes through a theory of spatial normalizers and core inclusions modeled on Renault's C*-algebraic reconstruction. The advertised applications are: (i) for p≠2, a reduced L^p-groupoid algebra of a principal Weyl groupoid embeds into a spatial AF L^p-algebra only if the groupoid is AF, giving non-embeddability of irrational rotation L^p-algebras; (ii) for p≠2, there is no unital contractive homomorphism from O_2^p ⊗_p O_2^p to O_2^p, so no L^p-analogue of Kirchberg's O_2-embedding theorem. The paper also shows that embeddings of L^p-groupoid algebras induce embeddings of topological full groups and identifies the relevant full groups with generalized Brin–Thompson groups.

Significance. If the main theorem is correct, the paper gives a striking and unexpected contrast with C*-algebra theory: in the L^p setting with p≠2, embeddings of reduced groupoid algebras are rigid enough to be described entirely by morphisms of the underlying groupoids. The applications to AF-embeddability and to Cuntz-algebra tensor products are substantial and falsifiable. The paper is well structured, carefully written, and contains a considerable amount of original technical machinery, including the spatial-normalizer action and the actor reinterpretation. However, the central technical step in Section 3 is not proved as written, and one auxiliary claim in Section 5 is false. These issues are load-bearing for the main results, so significant revision is required before the paper can be accepted.

major comments (1)

[§4, Proposition 4.19] In the proof of Proposition 4.19 the symbol 'supp′' appears without definition; it should be 'supp' or should be explicitly defined.

minor comments (4)

[§3, Theorem 3.34] The first line contains a typo: 'φ(core(A))⊆φ(core(B))' should read 'φ(core(A))⊆core(B)'.
[§3, Proposition 3.33] The notation 'e∈φ(C0(U))B∗' for the weak-* closure is not defined; use an overline or a verbal description.
[§1, Introduction] There is a typo in the introduction: 'Pismner-Voiculescu' should be 'Pimsner-Voiculescu'.
[References] Reference [33] has a stray '2.' at the end of the arXiv identifier; reference [22] should be marked as an unpublished preprint with its current status.

Circularity Check

0 steps flagged · score 2.0 of 10

No constructional circularity in the main actor characterization; the flagged Proposition 3.33 step is a proof gap, not a circular reduction. Minor non-load-bearing self-citation from [8].

full rationale

The main theorem (4.20) is derived, not assumed: φ is first converted, via Theorem 3.35 and Theorem 4.12, into an actor β between the Weyl twists; conditional-expectation intertwining gives freeness (Prop 4.15); and Prop 4.19 reconstructs φ as ι_*∘π_* by comparing spatial-normalizer supports and j-values. No equation of the paper defines the conclusion into the hypotheses. The only cited result from the authors' prior work is the core computation Prop 3.5 / Thm 3.2, citing [8] (Choi–Gardella–Thiel). That cited work is published, does not contain Theorem 4.20, and is used only to identify core(F_p^λ(G,Σ)) with C(G^(0)); it is therefore real evidence, not a circularity chain. The skeptical concern about Prop 3.33 is a genuine proof gap: the sentence "By definition, Φ is multiplicative ... Φ inherits unitality and contractivity from φ and therefore preserves MP-partial isometries by Remark 3.13" asserts, without deriving, the multiplicativity/MP-inverse identities for a partially defined weak-* limit. This is an omitted proof that threatens Theorem 3.34 and hence Theorem 4.20, but it is a correctness risk rather than a constructional circular step: no fitted parameter, no self-definition, and no equation makes the target equivalent to an input.

Assumptions & free parameters 0 free parameters · 6 assumptions · 0 invented entities

The paper is a theorem-proving paper with no fitted parameters and no newly postulated physical or algebraic entities; 'spatial normalizers' and 'actors' are defined objects built from existing data. The ledger records the external theorems that do the work, including one key rigidity result from an unpublished preprint.

assumptions (6)

standard math For p≠2, MP-partial isometries on L^p are exactly spatial partial isometries (Banach-Lamperti classification).
Invoked in Theorem 3.34 to convert abstract normalizers into spatial set automorphisms; taken from [3, Theorem 2.28].
standard math The core of a unital L^p-operator algebra is the largest C*-subalgebra and is commutative for p≠2.
Used in Propositions 3.5 and 6.5 to identify core(F_λ^p(G,Σ)) with C(G^(0)); from [8, Theorem 3.2 / Proposition 5.1].
standard math L^p-operator algebras are Arens regular for 1<p<∞, so their double duals are dual Banach algebras.
Used in Example 3.17 and Proposition 3.33 for weak-* limits; from [10, Theorem 1] and [9, Corollary 6.3].
domain assumption If m>n, any homomorphism [[G_{k1}×...×G_km]] → [[G_l1×...×G_ln]] has abelian image.
Used in the proof of Theorem 7.8 to force abelian image, contradicting the nonabelian Brin-Thompson group; taken from [22, Cor 11.19], an arXiv preprint not reproduced in this paper.
domain assumption Inclusions of open subgroupoids induce isometric maps on reduced L^p-groupoid algebras for all p.
Used in Proposition 4.17 to ensure ι_* is isometric; stated as a generalization of [5, Lemmas 3.2 and 3.4] to L^p.
domain assumption Simplicity of the relevant L^p-Cuntz tensor products and existence of SFT groupoid models for spatial tensor products.
Used in Section 7 to make unital contractive maps injective and to translate between algebras and products of SFTs; from [29, Theorem 5.14], [8, Theorem 7.6], and [16, Theorem 7.7].

how reviews work

0 comments

Cite this review

Pith. "Pith review of Embeddings of $L^p$-operator algebras." pith.science (2026). https://pith.science/paper/OMT7FVV5

@misc{pith2026260115204,
  author       = {Pith},
  title        = {Pith review of: Embeddings of $L^p$-operator algebras},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/OMT7FVV5}},
  note         = {Machine review of arXiv:2601.15204}
}

abstract

We study embeddings of $L^p$-operator algebras arising from (twis\-ted) \'etale groupoids, with particular emphasis on rigidity phenomena for $p\neq 2$. Our methods rely on a detailed analysis of core normalizers and their functorial behavior under algebra homomorphisms. Using the notion of actors between groupoids, we show that under natural hypotheses, embeddings between reduced $L^p$-groupoid algebras can be described entirely in terms of morphisms of the underlying groupoids. We further show that embeddings of $L^p$-groupoid algebras induce embeddings of the associated topological full groups. Our results provide new tools for studying embeddability questions in the $L^p$-setting, and are particularly helpful when ruling out the existence of embeddings. As applications, we obtain strong embeddability results both for spatial AF $L^p$-operator algebras and for tensor products of $L^p$-Cuntz algebras. For $p\not \in \{1,2\}$, a reduced $L^p$-groupoid algebra associated with a principal \'etale groupoid embeds into a spatial AF $L^p$-operator algebra if and only if the underlying groupoid is AF. In particular, and in contrast with classical results of Pimsner-Voiculescu, irrational $L^p$-noncommutative tori do not embed into spatial AF $L^p$-operator algebras for $p\neq 2$. Furthermore, if $p\neq 2$, there is no unital contractive homomorphism from $\mathcal{O}_2^p \otimes_p \mathcal{O}_2^p$ into $\mathcal{O}_2^p$, showing that there is no $L^p$-analog of Kirchberg's $\mathcal{O}_2$-embedding theorem.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

34 extracted references · 5 linked inside Pith

[1]

W. Arendt. Spectral properties of Lamperti operators.Indiana Univ. Math. J., 32(2):199– 215, 1983

1983
[2]

W. Arveson. On subalgebras of C*-algebras.Acta Math., 123:141–224, 1969. 39

1969
[3]

Bardadyn, B

K. Bardadyn, B. Kwa´ sniewski, and A. McKee. Banach algebras associated to twisted ´ etale groupoids: inverse semigroup disintegration and representations onL p-spaces.J. Funct. Anal., 289:111163, 2025

2025
[4]

Bardadyn, B

K. Bardadyn, B. Kwa´ sniewski, and A. McKee. Banach algebras associated to twisted ´ etale groupoids: simplicity and pure infiniteness.Trans. Amer. Math. Soc., to appear
[5]

Barlak and X

S. Barlak and X. Li. Cartan subalgebras and the UCT problem, II.Math. Ann., 378(1-2):255– 287, 2020

2020
[6]

M. Brin. Higher dimensional Thompson groups.Geom. Dedicata, 108:163–192, 2004

2004
[7]

Brownlowe, A

N. Brownlowe, A. Sørensen.L 2,Z ⊗L 2,Z does not embed intoL 2,Z.J. Algebra, 456:1–22, 2016

2016
[8]

Y. Choi, E. Gardella, and H. Thiel. Rigidity results forL p-operator algebras and applications. Adv. Math., 452:Paper No. 109747, 47, 2024

2024

Show all 34 references

[9]

Civin and B

P. Civin and B. Yood. The second conjugate space of a Banach algebra as an algebra.Pacific J. Math., 11:847–870, 1961

1961
[10]

M. Daws. Arens regularity of the algebra of operators on a Banach space.Bull. London Math. Soc., 36(4):493–503, 2004

2004
[11]

Deaconu, A

V. Deaconu, A. Kumjian, and B. Ramazan. Fell bundles associated to groupoid morphisms. Math. Scand., 102(2):305–319, 2008

2008
[12]

Defant and K

A. Defant and K. Floret.Tensor norms and operator ideals, volume 176 ofNorth-Holland Mathematics Studies. North-Holland Publishing Co., Amsterdam, 1993

1993
[13]

Dicks and C

W. Dicks and C. Mart ´ ınez-P´ erez. Isomorphisms of Brin-Higman-Thompson groups.Israel J. Math., 199(1):189–218, 2014

2014
[14]

R. Exel. Inverse semigroups and combinatorialC ∗-algebras.Bull. Braz. Math. Soc. (N.S.), 39(2):191–313, 2008

2008
[15]

Gardella

E. Gardella. A modern look to algebras of operators onL p-spaces.Expo. Math420–453, 2021

2021
[16]

Gardella and M

E. Gardella and M. Lupini. Representations of ´ etale groupoids onL p-spaces.Adv. Math., 318:233–278, 2017

2017
[17]

Gardella and H

E. Gardella and H. Thiel. Isomorphisms of algebras of convolution operators.Ann. Sci. ´Ec. Norm. Sup´ er. (4), 55(5):1433–1471, 2022

2022
[18]

Giordano, I

T. Giordano, I. Putnam, and C. Skau. Affable equivalence relations and orbit structure of Cantor dynamical systems.Ergodic Theory Dynam. Systems, 24(2):441–475, 2004

2004
[19]

Hetland and E

E. Hetland and E. Ortega. Rigidity of twisted groupoidL p-operator algebras.J. Funct. Anal., 285(6):Paper No. 110037, 45, 2023

2023
[20]

Kirchberg and N

E. Kirchberg and N. C. Phillips. Embedding of exact C*-algebras in the Cuntz algebraO 2. J. Reine Angew. Math., 525:17–53, 2000

2000
[21]

X. Li. Every classifiable simple C ∗-algebra has a Cartan subalgebra.Invent. Math., 219(2):653–699, 2020

2020
[22]

Matte Bon

N. Matte Bon. Rigidity properties of full groups of pseudogroups over the Cantor set. 2018. Preprint, arXiv:1801.10133

2018 arXiv
[23]

H. Matui. Homology and topological full groups of ´ etale groupoids on totally disconnected spaces.Proc. Lond. Math. Soc. (3), 104(1):27–56, 2012

2012
[24]

H. Matui. Topological full groups of one-sided shifts of finite type.J. Reine Angew. Math., 705:35–84, 2015

2015
[25]

H. Matui. ´Etale groupoids arising from products of shifts of finite type.Adv. Math., 303:502– 548, 2016

2016
[26]

Meyer and C

R. Meyer and C. Zhu. Groupoids in categories with pretopology.Theory Appl. Categ., 30:Pa- per No. 55, 1906–1998, 2015

1906
[27]

Nekrashevych

V. Nekrashevych. Cuntz-Pimsner algebras of group actions.J. Operator Theory, 52(2):223– 249, 2004

2004
[28]

N. C. Phillips. Analogs of Cuntz algebras onL p-spaces. 2012. Preprint, arXiv:1201.4196

2012 arXiv
[29]

N. C. Phillips. Simplicity of UHF and Cuntz algebras onL p spaces. 2013. Preprint, arXiv:1309.0115

2013 arXiv
[30]

J. Renault. Cartan subalgebras in C*-algebras.Irish Math. Soc. Bull., 61:29–63, 2008

2008
[31]

V. Runde. Amenability for dual Banach algebras.Studia Math., 148(1):47–66, 2001

2001
[32]

Schafhauser

C. Schafhauser. Subalgebras of simple AF-algebras.Ann. of Math. (2), 192(2):309–352, 2020

2020
[33]

A. Sims. Hausdorff ´ etale groupoids and their C*-algebras. 2017. Preprint, arXiv:1710.10897.2

2017 arXiv
[34]

J. Taylor. Functoriality for groupoid and Fell bundleC ∗-algebras. 2023. Preprint, arXiv:2310.03126v1. (Eusebio Gardella and Jan Gundelach)Department of Mathematical Sciences, Chalmers University of Technology and University of Gothenburg, Gothenburg 412 96, Sweden. Email addr...

2023 arXiv

Pith tools

Reviewed August 3, 2026 · model on record in the stance chip above.

[1] [1]

W. Arendt. Spectral properties of Lamperti operators.Indiana Univ. Math. J., 32(2):199– 215, 1983

1983

[2] [2]

W. Arveson. On subalgebras of C*-algebras.Acta Math., 123:141–224, 1969. 39

1969

[3] [3]

Bardadyn, B

K. Bardadyn, B. Kwa´ sniewski, and A. McKee. Banach algebras associated to twisted ´ etale groupoids: inverse semigroup disintegration and representations onL p-spaces.J. Funct. Anal., 289:111163, 2025

2025

[4] [4]

Bardadyn, B

K. Bardadyn, B. Kwa´ sniewski, and A. McKee. Banach algebras associated to twisted ´ etale groupoids: simplicity and pure infiniteness.Trans. Amer. Math. Soc., to appear

[5] [5]

Barlak and X

S. Barlak and X. Li. Cartan subalgebras and the UCT problem, II.Math. Ann., 378(1-2):255– 287, 2020

2020

[6] [6]

M. Brin. Higher dimensional Thompson groups.Geom. Dedicata, 108:163–192, 2004

2004

[7] [7]

Brownlowe, A

N. Brownlowe, A. Sørensen.L 2,Z ⊗L 2,Z does not embed intoL 2,Z.J. Algebra, 456:1–22, 2016

2016

[8] [8]

Y. Choi, E. Gardella, and H. Thiel. Rigidity results forL p-operator algebras and applications. Adv. Math., 452:Paper No. 109747, 47, 2024

2024