Entropy and affine actions for surface groups

Fran\c{c}ois Labourie

REVIEW 3 major objections 5 minor 27 references

Entropy and affine actions for surface groups

T0 review · 3 major / 5 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read A surface group with Hitchin linear part cannot act properly on affine space, proved here by entropy methods.

desk verdict Independent proof of Danciger-Zhang by entropy methods: the route is mostly solid and the smoothness theorem is real, but Theorem 5.2, the linchpin, is still a sketch. read the letter →

arxiv 1908.00599 v3 pith:H25M6GME submitted 2019-08-01 math.DG math.DS

François Labourie This is my paper · ORCID

classification math.DGmath.DS MSC 37D4037D3522E40

keywords surfacegroupsHitchinrepresentationsaffineactionsproperMargulisinvarianttopologicalentropyAnosovlastrootflow

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 proves that a surface group — the fundamental group of a closed oriented surface of genus at least two — whose linear part is a Hitchin representation, one that can be deformed to a Fuchsian representation, can never act properly on affine space. The theorem was recently established by other means; the contribution here is an independent, substantially shorter proof that runs through ergodic theory and thermodynamic formalism. The decisive mechanism is that the last root flow, a reparametrisation of the geodesic flow whose periodic orbits have lengths $\log \lambda_{p-1}(\gamma)+\log \lambda_p(\gamma)$, has topological entropy exactly 1 near the Hitchin locus, and that this constancy forces the Labourie-Margulis diffusion (a measure-valued extension of the Margulis invariant) to vanish on the Bowen-Margulis measure. An existing criterion then converts that vanishing into non-properness of the affine action. If the proof is correct, it gives a dynamical template for studying proper affine actions and related obstructions.

What carries the argument

The engine of the proof is the last root flow: the reparametrisation $\psi^t$ of the geodesic flow whose closed orbit over $\gamma$ has length $\log \lambda_{p-1}(\rho(\gamma))+\log \lambda_p(\rho(\gamma))$, where $\lambda_i$ are the ordered eigenvalues of $\rho(\gamma)$. Its topological entropy is shown to be constantly 1 (Theorem 5.2). The second ingredient is the Labourie-Margulis diffusion, the functional $M(\mu)=\int Q(\varepsilon_p,\omega(X))\,d\mu$ on flow-invariant measures; Lemma 6.2 ties its value on a closed orbit to the derivative of the $p$-th eigenvalue, and Corollary 6.3 ties it to the derivative of the last root flow reparametrisation. The smoothness theorem (Theorem 4.1), stating that the isotropic limit curve $\Theta$ is $C^1$ with tangent space $\Lambda^2(E^*_{p-1}\oplus E^*_p)$, is what permits the entropy computation, and it is obtained from a proximality lemma for line bundles plus the transversality property (5).

What would settle it

Compute the topological entropy of the last root flow for an explicit non-Fuchsian Hitchin representation in $\mathrm{SO}(p,p-1)$ from the periodic-orbit lengths $\log\lambda_{p-1}(\gamma)+\log\lambda_p(\gamma)$; if the value differs from 1, Theorem 5.2 and the present proof of non-properness fail. Equally decisive would be a Borel Anosov $\mathrm{SO}(p,p)$ representation satisfying the transversality property (5) whose isotropic limit curve is not $C^1$.

Watch

Extended reading notes

Core claim

The paper's central claim is Theorem 1.1: if $\Gamma$ is a surface group and $\rho:\Gamma\to\mathrm{Aff}(\mathbb{R}^{2p-1})$ has linear part a Hitchin representation in $\mathrm{SO}(p,p-1)$, then the action is not proper. The proof first reduces to this signature using a spectral fact: an element of a properly acting affine group must have 1 as an eigenvalue of its linear part, so the Zariski closure cannot be Zariski dense in $\mathrm{SL}(2p-1)$ and the representation lands in $\mathrm{SO}(p,p-1)$. The affine deformation is encoded as a variation of the linear representation inside $\mathrm{SO}(p,p)$, and Lemma 6.2 identifies the derivative of the $p$-th eigenvalue along a closed orbit with half the diffusion of that orbit. Together with the constancy of the last root flow entropy (Theorem 5.2) and Abramov's lemma, this gives a zero integral of the reparametrisation derivative against the Bowen-Margulis measure, hence a zero diffusion $M(\mu)=0$. An existing criterion then asserts that any invariant measure annihilating the diffusion implies the affine action is not proper. The smoothness theorem for the isotropic limit curve (Theorem 4.1) is the geometric input that makes the entropy theorem go through.

Load-bearing premise

The load-bearing premise is that the last root flow has topological entropy exactly 1 for every Hitchin representation close to the Fuchsian locus; if that constancy fails, the diffusion need not vanish and the proof of non-properness collapses.

Editorial extensions

If this is right

Every affine action of a surface group whose linear part is Hitchin is non-proper; in particular no such group admits a properly discontinuous affine action.
The diffusion criterion is now effective for surface groups: exhibiting one invariant measure with zero diffusion is enough to obstruct properness, and this paper produces that measure dynamically rather than by construction.
For representations close to the Fuchsian locus in $\mathrm{SO}(p,p-1)$, the last root flow has entropy exactly 1, so the vanishing of the diffusion on the Bowen-Margulis measure is a robust, open phenomenon.
The smoothness theorem gives a new regularity statement for isotropic limit curves of Anosov representations satisfying transversality, a property that holds on an open neighbourhood of Fuchsian representations.

Reading between the lines

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

If entropy-one for the last root flow holds on the entire Hitchin component, as the paper suspects, the same proof would show non-properness directly for every Hitchin linear part without any closeness assumption.
The mechanism suggests a testable numerical criterion: approximate the topological entropy of the last root flow from a finite set of periodic orbits for a non-Fuchsian Hitchin representation; the prediction is exactly 1, and any deviation would show where the analytic-continuation step breaks.
The same combination — a constant-entropy root flow plus a diffusion vanishing on the equilibrium state — may obstruct proper affine actions for other Anosov representations in indefinite orthogonal groups, not just the Hitchin case.
The transversality property (5) may characterize Hitchin representations inside Borel Anosov representations of $\mathrm{SO}(p,p)$, which would extend the smoothness theorem to the whole component.

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 short version: this is not a new theorem. It is an independent proof, via thermodynamic formalism and entropy, of Danciger-Zhang's result that surface groups with Hitchin linear part never act properly on affine space. Labourie says this plainly in the abstract, and the value is in the route, not the destination.

What is new and good: the Smoothness Theorem 4.1, giving C1 isotropic limit curves under the transversality property, and the overall ergodic strategy. Lemma 4.3 on proximal bundles is clean and self-contained, and the final chain — Lemma 6.2 reading the Margulis invariant as eigenvalue variation, Corollary 6.3 converting the reparametrization derivative into the diffusion, Lemma 6.4 applying Abramov to turn entropy constancy into a zero integral — is elegant and checkable. The paper is honest about debts: the main theorem is attributed to [7] and the proof cites published, community-checked inputs. No circularity.

The soft spot is exactly where the stress-test note lands. Theorem 5.2, the constancy of the last-root-flow entropy, is the linchpin, and its proof is a sketch. The local step ("the same discussion as in Potrie-Sambarino using SRB measures") is a hand-wave: no SRB measure is constructed, no entropy identity is stated, and the role of C1 smoothness is asserted rather than argued. The global step invokes "the analyticity of the entropy obtained in [3]" without a theorem number or a word about connectedness of the relevant domain. These are missing verifications, not contradictions, and they probably close. But for a paper whose purpose is an independent proof, those two sentences are load-bearing and need to become real arguments or precise references.

Smaller issues: the reduction to SO(p,p−1) uses a Guichard theorem without a citation, and Theorem 5.2 is proved only near the Fuchsian locus, so the analytic continuation is carrying more weight than the prose admits.

Who benefits: researchers in Anosov representations and higher Teichmüller theory. Even those who know the Danciger-Zhang theorem will take the smoothness result and entropy method seriously.

Recommendation: accept for peer review. The known theorem is not a disqualifier — the method and auxiliary results are the contribution. Assign a referee and tell them to push hard on Theorem 5.2.

Referee Report

3 major / 5 minor

Summary. The paper gives a new proof of the Danciger–Zhang theorem: if a surface group acts on affine space with Hitchin linear part, then the action is not proper. The approach is ergodic-theoretic. The author embeds a Hitchin representation in SO(p,p-1) into SO(p,p), considers the associated last root flow, and proves (Theorem 5.2) that its topological entropy is constantly equal to 1 for representations close to Hitchin. The proof of Theorem 5.2 has two steps: a local statement near Fuchsian representations obtained via an SRB/equilibrium-state argument after proving that the isotropic limit curve is C1 (Theorem 4.1), and a global step invoking analyticity of the entropy from [3] to pass to a neighborhood of the Hitchin component. The paper then uses the Abramov lemma to show that the reparametrization derivative integrates to zero against the Bowen–Margulis measure of the last root flow, interprets this integral via Lemma 6.2 and Corollary 6.3 as the Labourie–Margulis diffusion, and finally applies the Ghosh–Treib criterion to conclude that the affine action is not proper.

Significance. If the proof is completed, this is a valuable contribution: it offers a fundamentally different route to an important recent theorem, and it provides two results of independent interest, the C1 smoothness of the isotropic limit curve (Theorem 4.1) and the entropy constancy of the last root flow (Theorem 5.2). The paper is clearly written and builds on a coherent chain of lemmas: Lemma 4.3 gives a quantitative proximality estimate, Theorem 4.1 converts it into smoothness, Corollary 6.3 relates the reparametrization derivative to the Margulis diffusion, and the Ghosh–Treib criterion converts the vanishing of the diffusion into non-properness. The proof does not assume its conclusion; it re-derives the theorem from published inputs. However, the central Theorem 5.2 is currently only sketched, and because the entire proof of Theorem 1.1 depends on the constancy of the entropy, the missing details are load-bearing.

major comments (3)

[Section 5, Theorem 5.2] The proof of Theorem 5.2 is not self-contained. The local step is dispatched with the sentence: 'The same discussion as in Potrie–Sambarino using SRB measures gives us the result in the neighbourhood of the Fuchsian representation by Corollary 3.6 and Theorem 4.1 since the isotropic limit curve is C1.' No SRB measure is constructed, no entropy identity is stated, and the precise role of the C1-smoothness of the isotropic limit curve is not explained. Since Theorem 5.2 is the only input that makes the entropy of the last root flow constant in Lemma 6.4, and hence the only source of the vanishing of the Margulis diffusion in the proof of Theorem 1.1, this missing verification is load-bearing. The authors should either provide a complete proof of the local entropy statement or state and prove a precise theorem with explicit hypotheses and references to the specific arguments in [24].
[Section 5, analytic continuation] The global step in the proof of Theorem 5.2 is not precise. The sentence 'the analyticity of the entropy obtained in [3] implies that the entropy is constant and equal to 1 on the neighbourhood of the Hitchin representations in SO(p,p−1)' does not cite a theorem number in [3], nor does it verify that the Hitchin component of SO(p,p−1) is contained in the domain of analyticity and that this domain is connected to the Fuchsian neighborhood. The theorem statement itself is ambiguous: 'For ρ close enough to a Hitchin representation in SO(p,p−1)' could mean a neighborhood of an arbitrary Hitchin representation or a neighborhood of the Fuchsian locus. The closing remark that the theorem may apply to the whole Hitchin component suggests that the current proof may not cover all cases needed for Theorem 1.1. This point must be clarified and proved.
[Section 5, notation] The proof of Theorem 5.2 refers to objects Ξq and Mq without defining them anywhere in the manuscript. This makes the proof unverifiable as written. Since this is the central theorem, the notation must be introduced, and the claimed fixed-point and contraction-rate statements must be made explicit.

minor comments (5)

[Section 6.2, Lemma 6.2] In the proof of Lemma 6.2, the reference 'proposition 3.6' appears to be wrong: the orthogonality ⟨qργ(εk)|ε2p−k⟩=0 for k<p follows from the defining condition of the subspace H in Proposition 6.1, not from the transversality/corollary numbered 3.6. Please correct the reference.
[Section 6.2, paragraph before Lemma 6.2] The text says 'Let εp be the section of norm 1 of the spacelike line bundle Ep.' In the SO(p,p−1) decomposition, Vp is the timelike trivial bundle, and the SO(p,p) lines Ep and Ep are lightlike. The notation is confusing; the intended object appears to be the unit vector in the timelike line Vp. Please clarify the terminology.
[Section 6.4, proof of Theorem 1.1] The family (ρs)s∈[0,1] is introduced at the beginning of the proof, but the arguments using Theorem 5.2 and Lemma 6.4 are local in s, and it is not explained why the family remains in the relevant Anosov/entropy-conserving regime for all s in [0,1]. The proof should specify that the family is considered for s in a sufficiently small interval around 0, or justify the global statement.
[Theorem 4.1, statement] The statement begins 'Let SO(p,p) be Anosov representation satisfying the Transversality Property (5)' and should read 'Let ρ be an Anosov representation into SO(p,p) satisfying...'. This is a small grammatical issue but worth correcting.
[Throughout] There are several typographical errors that should be corrected in a final version, e.g., 'Danciger of Zhang' in the abstract should be 'Danciger and Zhang', 'cann ot' should be 'cannot', 'neig hbourhood' in the proof of Theorem 5.2 should be 'neighborhood', and 'a ﬃne' should be 'affine'.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: Theorem 1.1 follows from independently published lemmas; the sketchy Entropy Theorem is a verification gap, not a circular step.

full rationale

The central claim Theorem 1.1 is not derived from itself. It is assembled from external or prior published inputs: the Ghosh-Treib criterion [10] converts M(mu)=0 into non-properness; Lemma 6.2 computes the derivative of the p-th eigenvalue (from [19]); Corollary 6.3 converts that derivative into the integral of the reparametrization derivative using the closed-orbit identity lambda_p=1; Lemma 6.4 is a valid consequence of Abramov's lemma once entropy is constant; and Theorem 5.2 supplies that constancy. Theorem 5.2 is the only point where the paper is concise: the proof is a sketch that says 'the same discussion as in Potrie-Sambarino using SRB measures' and then invokes 'the analyticity of the entropy obtained in [3]' to move from a Fuchsian neighbourhood to all nearby Hitchin representations. This is an omitted verification, and if the SRB or analytic-continuation step fails the proof of Theorem 1.1 collapses. But it is not circular: [24] is Potrie-Sambarino's published argument, [3] is a published analyticity theorem, and neither is a restatement of Theorem 1.1 or of the constancy of the last-root-flow entropy in the form being proven. No parameter is fit to the target conclusion, and the vanishing of the Margulis invariant is not put in by hand. Hence no specific reduction of the form Eq. X = Eq. Y by construction can be exhibited, and the appropriate finding is no significant circularity.

Assumptions & free parameters 0 free parameters · 7 assumptions · 0 invented entities

Pure mathematics proof with no fitted constants, no experimental data, and no new postulated objects beyond standard constructions (last root flow, limit curves) already in the cited literature. The axioms are the external theorems the proof leans on; the two most load-bearing are the Ghosh-Treib criterion and the analyticity of entropy from [3].

assumptions (7)

domain assumption Hitchin representations are Borel Anosov and their limit curves depend continuously, analytically for entropy, on the representation (Guichard-Wienhard [13], Bridgeman-Canary-Labourie-Sambarino [3]).
Invoked in Proposition 3.3, Corollary 3.6, and Section 6.2 to decompose the associated bundle into contracting line bundles and to differentiate along deformation families.
domain assumption Ghosh-Treib criterion: a flow-invariant measure mu with M(mu)=0 implies the affine action is not proper ([10, Theorem 7.1 and Definition 4.4]).
The bridging step in Section 6.4 that converts the entropy computation into the non-properness conclusion. If the criterion has hidden hypotheses unmet here, the final step fails.
standard math For representations close to Hitchin in SO(p,p-1), the last root flow exists and associates to a closed orbit gamma the length log(lambda_p) + log(lambda_{p-1}) ([4, Proposition 2.4]).
Gives the flow used throughout Sections 5 and 6; cited from the author's prior work with coauthors.
standard math The Abramov lemma for reparametrized flows: h(m^s) = h(m) / integral f^s dm ([26, Lemma 2.4]).
The basis of Lemma 6.4, which yields integral g d-mu = 0 from entropy constancy.
domain assumption The Zariski closure of a Hitchin representation contains the principal SL2(R) and, if not Zariski dense, lies in Sp(2p) or SO(p,p-1); a proper affine action forces 1 into the spectrum of the Zariski closure (Guichard, by name only; Abels-Margulis-Soifer background).
The reduction in Section 1.1 that restricts the theorem to SO(p,p-1), stated without a bibliographic reference for Guichard's result.
standard math Lusztig positivity: the explicit SL2(R)-action matrix alpha_{k,m} is nonzero for m >= k, used to prove the transversality property (Proposition 3.5), following [8] and [20].
The computation in the proof of Proposition 3.5 needs the nonzero coefficients of the unipotent action; the paper attributes this to Lusztig positivity as used in [8].
domain assumption The entropy of the last root flow is an analytic function of the representation on the relevant connected domain ([3]).
Used in the proof of Theorem 5.2 to extend equality with 1 from a neighborhood of the Fuchsian locus to nearby Hitchin representations. If the domain is not connected as assumed, the extension fails.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Entropy and affine actions for surface groups." pith.science (2026). https://pith.science/paper/H25M6GME

@misc{pith2026190800599,
  author       = {Pith},
  title        = {Pith review of: Entropy and affine actions for surface groups},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/H25M6GME}},
  note         = {Machine review of arXiv:1908.00599}
}

read the original abstract

We give an independent proof of a theorem of Danciger of Zhang: surface groups with Hitchin linear part cannot act properly on the affine space

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

27 extracted references · 27 canonical work pages

[7]

Je ﬀrey Danciger and Tengren Zhang, Aﬃne actions with Hitchin linear part , arXiv .org (2018)

work page 2018
[3]

4, 1089–1179

Martin J Bridgeman, Richard Canary , François Labourie, a nd Andres Sambarino, The pressure metric for Anosov representations , Geometric And Functional Analysis 25 (2015), no. 4, 1089–1179

work page 2015
[24]

3, 885–925

Rafael Potrie and Andrés Sambarino, Eigenvalues and Entropy of a Hitchin representation , Inventiones Mathematicae (2017), no. 3, 885–925

work page 2017
[1]

Herbert Abels, Gregory Margulis, and Gregory Soifer, The Auslander conjecture for dimen- sion less than 7 , arXiv .org (2012)

work page 2012
[2]

Louis Auslander, The structure of complete locally a ﬃne manifolds , Topology 3 (1964), no. suppl. 1, 131–139. ENTROPY AND AFFINE ACTIONS FOR SURFACE GROUPS 13

work page 1964
[4]

Dedicata 192, (2018)

Martin J Bridgeman, Richard Canary , François Labourie, and Andres Sambarino, Simple root ﬂows for Hitchin representations , Geom. Dedicata 192, (2018)

work page 2018
[5]

Proper actions of discrete groups of affine transformations

Je ﬀrey Danciger, Todd A. Drumm, William M. Goldman, and Ilia Smi lga, Proper actions of discrete groups of a ﬃne transformations, arXiv:2002.09520, to be published in Geometry , Groups and Dynamics

work page Pith review arXiv 2002
[6]

Je ﬀrey Danciger, François Guéritaud, and Fanny Kassel, Proper a ﬃne actions for right- angled Coxeter groups, arXiv .org (2018)

work page 2018

Show all 27 references

[8]

Vladimir V Fock and Alexander B Goncharov , Moduli spaces of local systems and higher Teichmüller theory, Publ. Math. Inst. Hautes Études Sci. (2006), no. 103, 1–211

2006
[9]

Sourav Ghosh, Avatars of Margulis invariants and proper actions , arXiv .org (2018)

2018
[10]

, arXiv .org (2017)

Sourav Ghosh and Nicolaus Treib, Aﬃne Anosov representations and Proper actions. , arXiv .org (2017)

2017
[11]

Mar gulis, Proper aﬃne actions and geodesic ﬂows for hyperbolic surfaces , Annals of Maths 170 (2009), no

William M Goldman, François Labourie, and Gregory A. Mar gulis, Proper aﬃne actions and geodesic ﬂows for hyperbolic surfaces , Annals of Maths 170 (2009), no. 3, 1051–1083

2009
[12]

Margulis, Flat Lorentz 3-manifolds and cocompact Fuchsian groups, vol

William M Goldman and Gregory A. Margulis, Flat Lorentz 3-manifolds and cocompact Fuchsian groups, vol. 262, 2000

2000
[13]

2, 357–438

Olivier Guichard and Anna Wienhard, Anosov representations: domains of discontinuity and applications, Inventiones Mathematicae 190 (2012), no. 2, 357–438

2012
[14]

3, 449–473

Nigel J Hitchin, Lie groups and Teichmüller space , Topology 31 (1992), no. 3, 449–473

1992
[15]

Jeremy Kahn, François Labourie, and Shahar Mozes, Surface groups in uniform lattices of some semi-simple groups, arXiv .org (2018)

2018
[16]

3, 1127–1190

Jeremy Kahn and Vladimir Markovic, Immersing almost geodesic surfaces in a closed hyper- bolic three manifold, Annals of Mathematics 175 (2012), no. 3, 1127–1190

2012
[17]

1, 15–31

François Labourie, Fuchsian aﬃne actions of surface groups , Journal of Di ﬀerential Geom- etry 59 (2001), no. 1, 15–31

2001
[18]

1, 51–114

, Anosov ﬂows, surface groups and curves in projective space , Inventiones Mathemat- icae 165 (2006), no. 1, 51–114

2006
[19]

Quatrième Sér ie 51 (2018), no

François Labourie and Richard Wentworth, Variations along the Fuchsian locus , Annales Scientiﬁques de l’Ecole Normale Supérieure. Quatrième Sér ie 51 (2018), no. 2, 487–547

2018
[20]

123, 1994

George Lusztig, T otal positivity in reductive groups, vol. 123, 1994

1994
[21]

Margulis, Free completely discontinuous groups of a ﬃne transformations, Dokl

Gregory A. Margulis, Free completely discontinuous groups of a ﬃne transformations, Dokl. Akad. Nauk SSSR 272 (1983), no. 4, 785–788

1983
[22]

, Complete a ﬃne locally ﬂat manifolds with a free fundamental group , Zap. Nauchn. Sem. Leningrad. Otdel. Mat. Inst. Steklov . (LOMI) 134 (1984), 190–205

1984
[23]

Geo ﬀrey Mess, Lorentz spacetimes of constant curvature , Geometriae Dedicata 126 (2007), 3–45

2007
[25]

Beatrice Pozzetti, Andrés Sambarino, and Anna Wienhard , Conformality for a robust class of non-conformal attractors, arXiv .org (2019)

2019
[26]

2, 443–488

Andrés Sambarino, Quantitative properties of convex representations , Commentarii Mathe- matici Helvetici 89 (2014), no. 2, 443–488

2014
[27]

Ilia Smilga, Proper aﬃne actions on semisimple Lie algebras , Ann. Inst. Fourier (Grenoble), 66, (2016),no. 2, 785–831

2016

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.

[1] [7]

Je ﬀrey Danciger and Tengren Zhang, Aﬃne actions with Hitchin linear part , arXiv .org (2018)

work page 2018

[2] [3]

4, 1089–1179

Martin J Bridgeman, Richard Canary , François Labourie, a nd Andres Sambarino, The pressure metric for Anosov representations , Geometric And Functional Analysis 25 (2015), no. 4, 1089–1179

work page 2015

[3] [24]

3, 885–925

Rafael Potrie and Andrés Sambarino, Eigenvalues and Entropy of a Hitchin representation , Inventiones Mathematicae (2017), no. 3, 885–925

work page 2017

[4] [1]

Herbert Abels, Gregory Margulis, and Gregory Soifer, The Auslander conjecture for dimen- sion less than 7 , arXiv .org (2012)

work page 2012

[5] [2]

Louis Auslander, The structure of complete locally a ﬃne manifolds , Topology 3 (1964), no. suppl. 1, 131–139. ENTROPY AND AFFINE ACTIONS FOR SURFACE GROUPS 13

work page 1964

[6] [4]

Dedicata 192, (2018)

Martin J Bridgeman, Richard Canary , François Labourie, and Andres Sambarino, Simple root ﬂows for Hitchin representations , Geom. Dedicata 192, (2018)

work page 2018

[7] [5]

Proper actions of discrete groups of affine transformations

Je ﬀrey Danciger, Todd A. Drumm, William M. Goldman, and Ilia Smi lga, Proper actions of discrete groups of a ﬃne transformations, arXiv:2002.09520, to be published in Geometry , Groups and Dynamics

work page Pith review arXiv 2002

[8] [6]

Je ﬀrey Danciger, François Guéritaud, and Fanny Kassel, Proper a ﬃne actions for right- angled Coxeter groups, arXiv .org (2018)

work page 2018