Pith. sign in

REVIEW 6 minor 6 references

The ideal structure of C*-algebras of etale groupoids with isotropy groups of local polynomial growth

T0 review · 0 major / 6 minor · reviewed 2026-08-11 · deepseek-v4-flash

Pith's one-line read For amenable second countable étale groupoids with isotropy of local polynomial growth, an induced primitive ideal lies in the closure of a set of induced ideals exactly when an intersection of subgroup-induced ideals is contained in it…

desk verdict A genuine advance on the Mackey program for étale groupoids: full Jacobson topology for nonabelian isotropy of local polynomial growth, with a credible SL3(Z) computation. read the letter →

arxiv 2412.11805 v4 pith:ZTFTTAFQ submitted 2024-12-16 math.OA math.RT

classification math.OAmath.RT MSC 46L0546L5522D25
keywords groupoidC*-algebrasprimitiveidealspaceJacobsontopologyétalegroupoidsisotropygroupslocalpolynomialgrowthinducedrepresentationsweakcontainment
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

This paper aims to give a complete description of the primitive ideal space Prim C*(G) — the Jacobson topology on the set of kernels of irreducible representations — for a wide class of C*-algebras coming from étale groupoids. The class covers amenable second countable Hausdorff étale groupoids whose isotropy groups have local polynomial growth, and more generally belong to a class M defined by a maximal-ideal-intersection property. The central claim is that the topology on Prim C*(G) is fully encoded by the representation theory of the isotropy groups and their subgroups: an induced primitive ideal lies in the closure of a set of induced ideals exactly when a certain intersection of ideals induced from subgroups is contained in it. If true, this gives a concrete description of primitive spectra for all such groupoids, including crossed products by actions of groups of local polynomial growth, and yields explicit computations such as the primitive spectrum of SL3(Z) acting on SL3(R)/U3(R). The result extends the earlier abelian-isotropy case to noncommutative isotropy, provided the isotropy groups satisfy the maximal-ideal condition.

What carries the argument

The engine is a weak-containment criterion for induced representations (Theorem 3.3). Given an amenable isotropy group G^x_x and a family of representations induced from other isotropy fibers, the criterion translates weak containment of one induced representation in a direct sum of others into an asymptotic condition on approximate Hilbert–Schmidt intertwiners, and then, under a maximal-ideal-intersection hypothesis on the restrictions π|Γ_k to a cofinal family of subgroups Γ_k, into a statement about convergence in the Fell topology on Sub(G)prim — the space of pairs (S,I) where S is a subgroup of G and I is a primitive ideal of C*(S). The central identity is a weak form of Frobenius reciprocity connecting induction and restriction: π ≺ $Ind^{{G^x_x}}$_S π_ω together with continuity of induction along converging subgroups gives the sufficiency direction, while the necessity direction uses the maximal-ideal-intersection property to decompose arbitrary states into convex combinations of states on maximal-ideal quotients. This is how the topology on the unit space and the representation theory of isotropy groups combine to determine the Jacobson topology.

What would settle it

Exhibit a finitely generated amenable group Γ such that every finite subset lies in a subgroup whose primitive ideals are all maximal (so Γ belongs to class M), but C*(Γ) has a primitive ideal that is not an intersection of maximal ideals; taking the groupoid with one unit and isotropy Γ would then violate the implication (2)⇒(1) of Theorem A, since the intersection condition could hold while the induced ideal is not in the closure.

Watch

Extended reading notes

Core claim

Theorem A (Theorem 4.2) states that for an amenable second countable Hausdorff locally compact étale groupoid G whose isotropy groups lie in class M, the induction map Ind : Stab(G)prim → Prim C*(G) is surjective, and for any G-invariant subset Ω of Stab(G)prim and any (x,J) ∈ Stab(G)prim, Ind(x,J) lies in the closure of Ind Ω exactly when the intersection of $Ind^{{G^x_x}}$_S I over all pairs (S,I) in the closure of Ω in Sub(G)prim with S ⊂ G^x_x is contained in J. Equivalently, π_J is weakly contained in the direct sum of the representations $Ind^{{G^x_x}}$_S π_I over those pairs. For the smaller class of FC-hypercentral isotropy groups, which includes virtually nilpotent groups, Theorem B simplifies the condition: the closure of Ω must contain a point (S,I) with S ⊂ G^x_x and π_I weakly contained in the restriction of π_J to S. The paper also proves a version for transformation groupoids with locally finite stabilizers, and as a worked example computes the primitive spectrum of SL3(Z) ⋉ C0(SL3(R)/U3(R)) together with its topology, where the stabilizers are trivial, isomorphic to $Z^{2}$, or isomorphic to the discrete Heisenberg group.

Load-bearing premise

The whole description stands on the assumption that every finite piece of every isotropy group can be placed inside a subgroup whose irreducible-representation kernels are all maximal — no kernel strictly contains another — which is known for groups of local polynomial growth but is not yet characterized algebraically.

Editorial extensions

If this is right

  • For every action Γ ↷ X of a countable group of local polynomial growth, Prim(Γ ⋉ C0(X)) is now completely described; previously only abelian stabilizers were tractable in this generality.
  • For groupoids with FC-hypercentral isotropy, the topology takes a simple form: closure is detected by a single subgroup-representation pair (S,I) with π_I weakly contained in the restriction of π_J, a description that covers all type I étale groupoid C*-algebras.
  • For amenable actions with locally finite stabilizers, the criterion becomes checkable via finite-dimensional representations of finite subgroups of the stabilizers.
  • The worked example computes Prim(SL3(Z) ⋉ C0(SL3(R)/U3(R))) explicitly, both as a set and with its Jacobson topology, using the description of quasi-orbits from unipotent flow theory.
  • The results cover actions of countable groups that are relatively hyperbolic with respect to a family of local-polynomial-growth or FC-hypercentral subgroups, including free products and lattices in semisimple Lie groups of real rank one.

Reading between the lines

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

  • Testable extension: if the intersection formula in Theorem A is as robust as it appears, the same criterion should describe Prim C*(G) for any amenable second countable étale groupoid whose isotropy groups satisfy the maximal-ideal-intersection property, even when the class M has no algebraic characterization; searching for a counterexample among groups not in M would sharpen the boundary.
  • Conjectural extension: the paper's suspicion that no topology on Stab(G)prim makes the induction map a homeomorphism in general suggests that primitive spectra of groupoids are finer than quasi-orbit spaces of isotropy data; the S^fin_∞ ⋉ K^N example would be a natural place to test this.
  • One algorithmic consequence not stated in the paper: when the groupoid is given by an action of a finitely presented group, the explicit SL3(Z) computation indicates that the intersection condition can in principle be checked from finite data on subgroup closures, potentially leading to computational tools for primitive spectra.
  • The examples show class M contains exponential-growth groups such as the ax+b group, so a purely geometric characterization of M would connect the C*-algebraic maximal-ideal condition to group growth and relative hyperbolicity; this remains an open question raised by the paper.
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.

Referee Report

0 major / 6 minor

Summary. This paper gives a description of the Jacobson topology on Prim C*(G) for amenable second countable Hausdorff locally compact étale groupoids G whose isotropy groups lie in a class M (Definition 4.1) that contains all groups of local polynomial growth. The main theorem (Theorem A, restated as Theorem 4.2) asserts that for any G-invariant subset Ω of Stab(G)_prim and any (x,J) ∈ Stab(G)_prim, Ind(x,J) lies in the closure of Ind Ω if and only if the intersection of the induced ideals Ind^{G^x_x}_S I over all points (S,I) ∈ Sub(G)_prim in the closure of Ω with S ⊂ G^x_x is contained in J. The proof combines the surjectivity of the induction map (Ionescu–Williams, the generalized Effros–Hahn theorem) with a technical weak-containment criterion, Theorem 3.3, relating weak containment of induced representations of the groupoid to the asymptotic behaviour of states on the isotropy groups and their subgroups under the maximal-ideal-intersection hypothesis. Sharper criteria are obtained under additional hypotheses: Theorem 4.6 for FC-hypercentral isotropy groups and Theorem 4.9 for actions with locally finite stabilizers. The final section implements the theory for the crossed product SL3(Z) ⋉ C0(SL3(R)/U3(R)): using Ratner's theorem and results of Dani–Margulis, the paper computes the primitive spectrum as a set (Theorem 5.20) and describes the Jacobson topology (Theorem 5.21) in terms of the quasi-orbit space of SL2(Z) on R^2 × T^2 analyzed in Section 5.2.

Significance. Assuming the main results hold, this is a substantial advance over the abelian-isotropy treatment of [CN24b] and over the classical results of Glimm and Williams: it is, to my knowledge, the first general topological description of Prim for groupoid C*-algebras with discontinuous nonabelian isotropy. A particular strength is that the maximal-ideal-intersection hypothesis is transparent: the class M is defined precisely so that the hypotheses of Theorem 3.3 are automatically satisfied, and the paper is explicit that the truly external input is the surjectivity of Ind (Ionescu–Williams), a published theorem cited correctly. The state-extension argument in Theorem 3.3 (2)⇒(3), the self-contained weak Frobenius reciprocity toolkit of Section 2, and the careful quasi-orbit analysis in Section 5.2 are all coherent and detailed. The paper is also unusually honest about limitations: Question 4.13 states that the class M has no known algebraic characterization, and Question 4.14 openly doubts the existence of a topology on Stab(G)_prim realizing the description, an instructive contrast with the abelian case.

minor comments (6)
  1. [Theorem 4.2, proof, passage (2')⇒(2)] The step 'by Lemma 1.8 and density of {K:(y_n,K)∈Ω} in hull(ker π_{y_n})' compresses a diagonal/subsequence argument that converts a pointwise state approximation into actual convergence of points (S_n, I_n) → (S, I) with (y_n, I_n) ∈ Ω; since the same step recurs verbatim in Theorem 4.6 and is echoed in Theorem 4.22, expanding this passage by two or three sentences would noticeably improve verifiability.
  2. [Section 5.5, before Lemma 5.18] There is a typographical error in the definition of the representation for ord z = n: the text reads 'π(X) := a1/nu, , π(Y) := b1/nv' with a stray double comma.
  3. [Introduction and Abstract] The names 'TheoremA' and 'TheoremB' appear without a space in several places; this should be corrected in the journal version.
  4. [Theorem 5.20(2)] The notation z = (a,b) ∈ T^2 = bΓ_i is confusing because the coordinate b and the character-group symbol bΓ_i are conflated, and for i = 1 the coordinate ar b appears in Q_θ(a,\bar b); writing \widehat{Γ_i} and explicitly stating which coordinate corresponds to the Z-character would remove ambiguity.
  5. [Remark 4.25] The claimed correction of [BL20, Theorem 3.18] would be much easier for a reader to check if the authors pinpointed the precise countability assumption in the original proof that fails; as written, the reader must reconstruct the comparison between the two arguments unaided.
  6. [Lemma 5.6(2)] The authors state that the complete description of cluster points in Lemma 5.6(2) is 'not difficult' and omit it; since Lemmas 5.4 and 5.6 are the tools the reader needs to verify Theorem 5.21, a one-sentence description of the missing case would make the section more self-contained.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the main theorems are derived from stated hypotheses and external surjectivity of induction, not from their conclusions.

full rationale

The derivation chain is self-contained in the relevant sense. Theorem 3.3, the technical engine, assumes that for a cofinal family of subgroups Γ_k the ideal ker(π|Γ_k) is an intersection of maximal ideals; this is exactly the property used to define the classes M0 and M in Definition 4.1. Thus in Theorem 4.2 the assumption that each isotropy group lies in M is precisely the hypothesis required to apply Theorem 3.3, not a conclusion smuggled in through the back door. The paper explicitly acknowledges that class M has no known algebraic characterization (Question 4.13), which is a limitation of scope rather than circularity. The surjectivity of the induction map Ind : Stab(G)prim → Prim C*(G), which upgrades the weak-containment description to a description of all of Prim C*(G), is attributed to the external theorem of Ionescu and Williams [IW09b], not to the authors' own prior work. The dependence on [CN24b] is for the abelian-isotropy special case and for comparison; the nonabelian extension in Theorems 4.2 and 4.6 is proved from Theorem 3.3 and standard lemmas, and the authors remark in Section 5 that the closed-orbit part of the concrete example could be obtained already from [CN24b], explicitly treating that part as illustrative rather than load-bearing. No fitted parameter is renamed as a prediction, and no uniqueness theorem is imported from the authors' earlier papers to force the choice of the main result. The maximal-ideal-intersection condition is an honestly stated sufficient hypothesis, not a self-definitional circular step.

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

The paper introduces no free parameters or invented entities. It rests on several deep external theorems in operator algebras, geometric group theory, and dynamics, listed in the axioms. The authors' own prior work [CN24b] is used for background and for special cases, but the nonabelian results are derived rather than assumed.

assumptions (5)
  • standard math The induction map Ind: Stab(G)prim -> Prim C*(G) is surjective for amenable second countable Hausdorff locally compact etale groupoids ([IW09b]).
    Used in Theorems 4.2, 4.6, 4.9 and Section 5; without surjectivity the closure criterion would not exhaust Prim C*(G).
  • standard math For groups in class M0, every primitive ideal of C*(Gamma) is an intersection of maximal ideals (Poguntke, Echterhoff, Moore-Rosenberg for virtually nilpotent or FC-hypercentral groups).
    The defining property of M0 in Definition 4.1; it is the key hypothesis of Theorem 3.3 that makes nonabelian weak containment tractable.
  • standard math Finitely generated groups of polynomial growth are virtually nilpotent (Gromov; Wolf for the converse).
    Shows groups of local polynomial growth lie in class M, so Theorem A applies to them; used in Section 4.1.
  • standard math Ratner's orbit closure theorem and the Dani-Margulis classification of orbit closures for unipotent flows in SL3(R)/U3(R).
    Used in Lemma 5.9 to describe quasi-orbits of SL3(Z), which underlies the primitive-spectrum computation in Section 5.
  • standard math Thoma's classification of virtually abelian groups and the characterization of type I groupoid C*-algebras (Tho64, Goo73, Cla07, vW18).
    Used in Proposition 4.8 to show Theorem B covers every type I etale groupoid, completing the Mackey program in the etale case.

how reviews work

0 comments
Cite this review

Pith. "Pith review of The ideal structure of C*-algebras of etale groupoids with isotropy groups of local polynomial growth." pith.science (2026). https://pith.science/paper/ZTFTTAFQ

@misc{pith2026241211805,
  author       = {Pith},
  title        = {Pith review of: The ideal structure of C*-algebras of etale groupoids with isotropy groups of local polynomial growth},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ZTFTTAFQ}},
  note         = {Machine review of arXiv:2412.11805}
}
abstract

Given an amenable second countable Hausdorff locally compact \'etale groupoid $\mathcal G$ such that each isotropy group $\mathcal G^x_x$ has local polynomial growth, we give a description of $\operatorname{Prim} C^*(\mathcal G)$ as a topological space in terms of the topology on $\mathcal G$ and representation theory of the isotropy groups and their subgroups. The description simplifies when either the isotropy groups are FC-hypercentral or $\mathcal G$ is the transformation groupoid $\Gamma\ltimes X$ defined by an action $\Gamma\curvearrowright X$ with locally finite stabilizers. To illustrate the class of C$^*$-algebras for which our results can provide a complete description of the ideal structure, we compute the primitive spectrum of $\mathrm{SL}_3(\mathbb Z)\ltimes C_0(\mathrm{SL}_3(\mathbb R)/U_3(\mathbb R))$, where $U_3(\mathbb R)$ is the group of unipotent upper triangular matrices.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

6 extracted references · 4 canonical work pages

  1. [1967]

    Gaps in √n mod 1 and ergodic theory

    MR0227310 [EM04] N. D. Elkies and C. T. McMullen,Gaps in √n mod 1 and ergodic theory, Duke Math. J.123 (2004), no. 1, 95–139,DOI:10.1215/S0012-7094-04-12314-0. MR2060024 [EM05] N. D. Elkies and C. T. McMullen, Correction to: “Gaps in √n mod 1 and ergodic theory” [Duke Math. J. 123 (2004), no. 1, 95–139; MR2060024], Duke Math. J. 129 (2005), no. 2, 405–406...

  2. [1969]

    Ionescu and D

    [IW09a] M. Ionescu and D. P. Williams,Irreducible representations of groupoidC ∗-algebras, Proc. Amer. Math. Soc. 137 (2009), no. 4, 1323–1332,DOI:10.1090/S0002-9939-08-09782-7. MR2465655 [IW09b] M. Ionescu and D. P. Williams,The generalized Effros–Hahn conjecture for groupoids, Indiana Univ. Math. J. 58 (2009), no. 6, 2489–2508,DOI:10.1512/iumj.2009.58.3...

  3. [1980]

    Renault,The ideal structure of groupoid crossed productC ∗-algebras, J

    MR584266 [Ren91] J. Renault,The ideal structure of groupoid crossed productC ∗-algebras, J. Operator Theory25 (1991), no. 1, 3–36. With an appendix by Georges Skandalis. MR1191252 [Sau77] J.-L. Sauvageot,Idéaux primitifs de certains produits croisés, Math. Ann.231 (1977/78), no. 1, 61–76, DOI:10.1007/BF01360030 (French). MR0473355 [Sca20] E. Scarparo,A to...

  4. [2000]

    MR1799683 [AS94] R. J. Archbold and J. S. Spielberg,Topologically free actions and ideals in discreteC ∗-dynamical systems, Proc. Edinburgh Math. Soc. (2)37(1994), no. 1, 119–124,DOI:10.1017/S0013091500018733. MR1258035 [Bag68] L. Baggett,A description of the topology on the dual spaces of certain locally compact groups, Trans. Amer. Math. Soc.132 (1968),...

  5. [2008]

    Bruce and E

    MR2391387 [BS24] C. Bruce and E. Scarparo,A tracial characterization of Furstenberg’s×p, ×q conjecture, Canad. Math. Bull. 67 (2024), no. 1, 244–256,DOI:10.4153/s0008439523000693. MR4706815 [Bro73] I. D. Brown,Dual topology of a nilpotent Lie group, Ann. Sci. École Norm. Sup. (4)6 (1973), 407–411. MR0352326 [BM25] A.BussandD.Martinez, Essential groupoid a...

  6. [2020]

    Theory 19 (2016), no

    MR4321941 [SW16] A.SimsandD.P.Williams, The primitive ideals of some étale groupoid C∗-algebras,Algebr.Represent. Theory 19 (2016), no. 2, 255–276,DOI:10.1007/s10468-015-9573-4. [Tho64] E. Thoma,Über unitäre Darstellungen abzählbarer, diskreter Gruppen, Math. Ann.153 (1964), 111– 138, DOI:10.1007/BF01361180 (German). MR0160118 [Tuk94] P.Tukia,Convergence ...

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.