Pith. sign in

REVIEW 1 major objections 3 minor 41 references

On the Lipschitz dimension of Cheeger-Kleiner

T0 review · 1 major / 3 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 This paper proves that every non-abelian Carnot group has infinite Lipschitz dimension: no Lipschitz light map into any Euclidean space exists.

desk verdict Solid, new results on Lipschitz dimension; the only flaw I found is a fixable normalization gap in the snowflake proof. read the letter →

arxiv 1908.04421 v1 pith:P645MRYK submitted 2019-08-12 math.MG

Guy C. David This is my paper · ORCID
classification math.MG MSC 30L9954F4553C23
keywords LipschitzdimensionCarnotgroupslightmapsPansudifferentiationquasi-isometricembeddingsNagataweaktangentsSierpinskicarpets
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 studies the Lipschitz dimension introduced by Cheeger and Kleiner, a quantitative analogue of topological dimension defined through Lipschitz light maps into Euclidean spaces. It establishes that non-abelian Carnot groups, such as the Heisenberg group with its sub-Riemannian metric, have infinite Lipschitz dimension. It also computes this dimension for products of metric trees, Euclidean buildings, snowflakes of Euclidean spaces, Sierpinski carpets, and the Sierpinski gasket, and compares it with Nagata, Hausdorff, Assouad, and Cheeger analytic dimensions. A corollary is a short proof that non-abelian Carnot groups do not quasi-isometrically embed into any space of finite Lipschitz dimension.

What carries the argument

The central object is the Lipschitz light map: a $C$-Lipschitz map $f:X\to Y$ such that for every $r>0$ and every $W\subset Y$ with $\operatorname{diam}(W)\le r$, the $r$-components of $f^{-1}(W)$ have diameter at most $Cr$. Lipschitz dimension is the smallest $n$ for which such a map into $\mathbb{R}^n$ exists. The proving engine is the weak-tangent calculus: rescaling domain and target and passing to Gromov-Hausdorff limits. Theorem 4.19 shows that, for complete doubling spaces, a Lipschitz map is Lipschitz light if and only if every weak tangent is light. For Carnot groups, Pansu differentiation supplies a weak tangent $Df(x)$ that is a Lie group homomorphism commuting with dilations; in a non-abelian Carnot group such a homomorphism must collapse the connected commutator subgroup, contradicting lightness. Other sections use a Lang-Schlichenmaier criterion to compute dimensions of trees and buildings, a self-covering argument for fractals, and convergence arguments for snowflakes and analytic dimension.

What would settle it

Exhibit a Lipschitz light map from the first Heisenberg group with its sub-Riemannian metric to $\mathbb{R}^n$ for some finite $n$; Theorem 5.8 says no such map exists. The direct check is to take any candidate $f$ and test the defining bound: for all $r>0$ and all $W\subset\mathbb{R}^n$ of diameter at most $r$, every $r$-component of $f^{-1}(W)$ has diameter at most $Cr$.

Watch

Extended reading notes

Core claim

The central claim is Theorem 5.8: if $G$ is a non-abelian Carnot group, then $\dim_L(G)=\infty$. Equivalently, for every $n$, every Lipschitz map $f:G\to\mathbb{R}^n$ fails to be Lipschitz light, meaning there is a scale $r$ and a small set $W\subset\mathbb{R}^n$ whose preimage contains arbitrarily large $r$-components. The proof combines Pansu's differentiation theorem with the paper's weak-tangent characterization of Lipschitz light maps: at almost every point, $f$ has a weak tangent $Df(x)$ that is a Lie group homomorphism commuting with the dilations of $G$, and any such homomorphism collapses the connected commutator subgroup of $G$ to a point. Because a weak tangent of a Lipschitz light map must itself be Lipschitz light, hence light, this collapse is impossible. Thus non-abelian Carnot groups admit no Lipschitz light map into any Euclidean space, and in particular no bi-Lipschitz embedding into one.

Load-bearing premise

The proof assumes Pansu's differentiation theorem: a Lipschitz map from a Carnot group to Euclidean space has a tangent group homomorphism at almost every point, and if that theorem did not apply, the contradiction in Theorem 5.8 would not follow.

Editorial extensions

If this is right

  • Non-abelian Carnot groups admit no bi-Lipschitz embedding into any Euclidean space, since such an embedding would be Lipschitz light.
  • Non-abelian Carnot groups do not quasi-isometrically embed into any space of finite Lipschitz dimension; in particular, they do not embed into finite products of metric trees or finite-rank Euclidean buildings.
  • Every compact positive-measure subset of a non-abelian Carnot group also has infinite Lipschitz dimension.
  • Products of $n$ metric trees, rank-$n$ Euclidean buildings, and snowflakes of $\mathbb{R}^n$ each have Lipschitz dimension exactly $n$; Sierpinski carpets and the Sierpinski gasket have Lipschitz dimension $1$.
  • For any complete Lipschitz differentiability space, the Cheeger analytic dimension is no larger than the Lipschitz dimension.

Reading between the lines

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

  • The proof gives a template for showing that a space has infinite Lipschitz dimension: find a weak tangent that is a homomorphism with a connected kernel. The same template may apply to other nilpotent Lie groups with dilation structures, not only Carnot groups.
  • Because Carnot groups have finite Nagata, Hausdorff, and Assouad dimensions, their infinite Lipschitz dimension shows that Lipschitz dimension records a genuinely different, more rigid quantitative obstruction, one sensitive to commutator structure at infinitesimal scales.
  • The quasi-isometric non-embedding corollary suggests a coarse-geometric counterpart: any space of finite Lipschitz dimension has weak tangents of finite Lipschitz dimension, so coarse embeddings into such spaces can be ruled out by exhibiting a weak tangent with infinite Lipschitz dimension.
  • For the Heisenberg group, the result implies that any projection or coding map into $\mathbb{R}^n$ must collapse a connected set at some scale, so quantitative finite-dimensional parametrizations of sub-Riemannian spaces cannot be faithful.
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

1 major / 3 minor

Summary. The paper develops the theory of the Lipschitz dimension dim_L introduced by Cheeger and Kleiner, which is defined as the minimal n for which a metric space admits a Lipschitz light map into R^n. The main contributions are: a weak-tangent characterization of Lipschitz light maps (Theorem 4.19); behavior of dim_L under products, unions, and Gromov-Hausdorff limits; computations of dim_L for metric trees, Euclidean buildings, snowflakes of Euclidean spaces, Carnot groups, subsets of R, and self-covering fractals; relationships with Nagata, Hausdorff, and Assouad dimensions; an upper bound on Cheeger's analytic dimension by dim_L; and mapping-theoretic results showing non-invariance under quasisymmetric and snowflake maps. A central result is Theorem 5.8, asserting that every non-abelian Carnot group has infinite Lipschitz dimension, which yields a short proof of quasi-isometric non-embedding into spaces of finite Lipschitz dimension.

Significance. If the results stand, this is a valuable contribution to the metric geometry of quantitative dimension theories. The most striking result, Theorem 5.8, gives a clean and conceptually simple proof that non-abelian Carnot groups have infinite Lipschitz dimension, using Pansu's theorem and the weak-tangent machinery; this is an elegant explanation of a previously known quasi-isometric non-embedding phenomenon. The weak-tangent characterization in Theorem 4.19 is a useful tool and is applied fruitfully throughout. The paper is careful and largely self-contained, with standard external inputs (Assouad embedding, Pansu differentiation, Lang-Schlichenmaier constructions, Cheeger's differentiation theory) clearly identified. I found one substantive proof gap, in the normalization of rescaled maps in Theorem 5.4, but it is local and easily repaired; the central arguments for Theorem 5.8 and its corollaries are sound.

major comments (1)
  1. [§5.2, proof of Theorem 5.4] The assertion that the maps h_λ(t) = λ^ε h(t/λ) subconverge to a bi-Lipschitz embedding of X = (R, |·|^ε) into R^{n-1} is not justified as written. Unless h(0) = 0, these maps are not normalized and can diverge, since h_λ(0) = λ^ε h(0) tends to infinity. The gap is repaired by replacing h_λ with t ↦ λ^ε (h(t/λ) - h(0)); the translated maps remain uniformly bi-Lipschitz, send 0 to 0, and then subconverge by Arzelà-Ascoli on compact intervals to the desired embedding. This fix is local and does not affect Theorem 5.8, Corollary 5.9, or Corollary 5.10, but the proof as printed should be corrected.
minor comments (3)
  1. [§5.2, Theorem 5.6] The statement appears to be trivially true as written: since α ∈ (0,1), the condition k > (n-1)/α forces k ≥ n, and then any E ⊆ R^n has dim_L(E) ≤ n by the isometric inclusion E ↪ R^n, which is Lipschitz light because preimages of sets of diameter ≤ r are contained in those sets. The proof, however, treats π^{-1}(p) as an (n-k)-plane, which is only meaningful for k ≤ n. Please clarify the intended range of k or simplify the theorem.
  2. [§8.3, Corollary 8.10] In the hypothesis, the expression H^n(g(X)) should presumably be H^n(g(Z)), since g is defined on Z and not on all of X.
  3. [§5.5, Definition 5.15] The phrase 'rescaled translates of K, inside K' is immediately qualified by allowing the copies to contain points outside K; please rephrase to avoid the apparent contradiction.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the central Carnot-group result derives from the external Pansu differentiation theorem plus internally developed weak-tangent lemmas.

full rationale

The paper's main result, Theorem 5.8, is not circular: it assumes a Lipschitz light map f:G→R^n and uses Pansu's differentiation theorem (Theorem 5.7, an external result) to obtain a weak tangent Df(x) that is a dilation-commuting homomorphism. Because Df(x) is a weak tangent of f, Corollary 4.16 makes it Lipschitz light; because it commutes with dilations and is a homomorphism, it collapses the connected commutator subgroup of the non-abelian Carnot group. These two facts contradict each other. Pansu's theorem does not assume Lipschitz lightness or Lipschitz dimension, so the conclusion is not built into the input. The weak-tangent framework in Section 4 is developed internally from Gromov-Hausdorff convergence, and Theorem 5.4 uses Assouad's embedding theorem and a minimal-dimension contradiction, again with no fitted parameter renamed as a prediction. Theorem 7.6 invokes Proposition 7.5, which the paper explicitly attributes to independent prior work by Schioppa [40] and the author [14]; this is external support, not a self-citation chain carrying the argument. No step in the derivation reduces by construction to its own inputs, so the appropriate finding is no circularity.

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

The paper's results rest on established theorems in metric geometry and sub-Riemannian geometry. No parameters are fitted to data. The only notable author-overlap is Proposition 7.5 (from the author's [14]), which is an independent published theorem. The many 'Question' passages are clearly marked as open, not results.

assumptions (6)
  • standard math Pansu's differentiation theorem (Theorem 5.7): Lipschitz maps between Carnot groups have almost-everywhere tangent homomorphisms Df(x) that commute with dilations.
    Invoked in the proof of Theorem 5.8 to produce a weak tangent of f that is a group homomorphism collapsing the commutator subgroup.
  • standard math Assouad's embedding theorem: every doubling metric space admits a bi-Lipschitz embedding into some Euclidean space R^n.
    Used in Theorem 5.4 and Corollary 8.4 to ensure snowflakes bi-Lipschitz embed into R^n so that weak tangents of projections can be studied.
  • standard math Lang-Schlichenmaier Theorems 3.2 and 3.3, as stated in Theorem 5.2: there exist maps f_T: T to R and f_X: X to R^n with associated flows h_T, h_X satisfying conditions (i)-(iii) of Lemma 5.1.
    This is the only input for Corollary 5.3 (trees and buildings have Lipschitz dimension n). The present paper does not reproduce the construction.
  • standard math Proposition 7.5: for a complete Lipschitz differentiability space with chart (U, phi: X to R^k), almost every tangent of phi is a Lipschitz quotient map onto R^k.
    Central input for Theorem 7.6; cited to [40, Theorem 5.56] and to the author's own [14, Corollary 5.1]. If this fails, the proof that dim_C X <= dim_L X breaks.
  • standard math David-Semmes [13, Proposition 12.8]: a weak tangent of a Lipschitz map between an Ahlfors regular space and a doubling space can be chosen David-Semmes regular when the image has positive measure.
    Used in the proof of Corollary 8.10.
  • standard math [32, Proposition 3.1]: a positive measure compact subset of a Carnot group has a tangent isometric to the whole group.
    Used in Corollary 5.9 to pass from dim_L(G)=infinity to the same for positive measure compact subsets.

how reviews work

0 comments
Cite this review

Pith. "Pith review of On the Lipschitz dimension of Cheeger-Kleiner." pith.science (2026). https://pith.science/paper/P645MRYK

@misc{pith2026190804421,
  author       = {Pith},
  title        = {Pith review of: On the Lipschitz dimension of Cheeger-Kleiner},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/P645MRYK}},
  note         = {Machine review of arXiv:1908.04421}
}
read the original abstract

In a 2013 paper, Cheeger and Kleiner introduced a new type of dimension for metric spaces, the "Lipschitz dimension". We study the dimension-theoretic properties of Lipschitz dimension, including its behavior under Gromov-Hausdorff convergence, its (non-)invariance under various classes of mappings, and its relationship to the Nagata dimension and Cheeger's "analytic dimension". We compute the Lipschitz dimension of various natural spaces, including Carnot groups, snowflakes of Euclidean spaces, metric trees, and Sierpinski carpets. As corollaries, we obtain a short proof of a quasi-isometric non-embedding result for Carnot groups and a necessary condition for the existence of non-degenerate Lipschitz maps between certain spaces.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

41 extracted references · 40 canonical work pages

  1. [14]

    G. C. David. Tangents and rectifiability of Ahlfors regu lar Lipschitz differentiability spaces. Geom. Funct. Anal. , 25(2):553–579, 2015

    work page 2015

  2. [40]

    Schioppa

    A. Schioppa. Derivations and Alberti representations . Adv. Math. , 293:436–528, 2016

    work page 2016

  3. [1]

    Ambrosio and B

    L. Ambrosio and B. Kirchheim. Rectifiable sets in metric a nd Banach spaces. Math. Ann. , 318(3):527–555, 2000

    work page 2000

  4. [2]

    D. Bate. Structure of measures in Lipschitz differentiab ility spaces. J. Amer. Math. Soc. , 28(2):421–482, 2015

    work page 2015

  5. [3]

    Bate and S

    D. Bate and S. Li. Characterizations of rectifiable metri c measure spaces. Ann. Sci. ´Ec. Norm. Sup´ er. (4), 50(1):1–37, 2017

    work page 2017

  6. [4]

    Bate and G

    D. Bate and G. Speight. Differentiability, porosity and d oubling in metric measure spaces. Proc. Amer. Math. Soc. , 141(3):971–985, 2013

    work page 2013

  7. [5]

    Bates, W

    S. Bates, W. B. Johnson, J. Lindenstrauss, D. Preiss, and G. Schechtman. Affine approxi- mation of Lipschitz functions and nonlinear quotients. Geom. Funct. Anal. , 9(6):1092–1127, 1999

    work page 1999

  8. [6]

    Bourdon and H

    M. Bourdon and H. Pajot. Poincar´ e inequalities and quas iconformal structure on the bound- ary of some hyperbolic buildings. Proc. Amer. Math. Soc. , 127(8):2315–2324, 1999

    work page 1999

Show all 41 references
  1. [7]

    Capogna, D

    L. Capogna, D. Danielli, S. D. Pauls, and J. T. Tyson. An introduction to the Heisenberg group and the sub-Riemannian isoperimetric problem , volume 259 of Progress in Mathematics . Birkh¨ auser Verlag, Basel, 2007

    2007

  2. [8]

    J. Cheeger. Differentiability of Lipschitz functions on metric measure spaces. Geom. Funct. Anal., 9(3):428–517, 1999

    1999

  3. [9]

    Cheeger and B

    J. Cheeger and B. Kleiner. Realization of metric spaces a s inverse limits, and bilipschitz embedding in L1. Geom. Funct. Anal. , 23(1):96–133, 2013

    2013

  4. [10]

    Cheeger and B

    J. Cheeger and B. Kleiner. Inverse limit spaces satisfy ing a Poincar´ e inequality.Anal. Geom. Metr. Spaces, 3:15–39, 2015

    2015

  5. [11]

    Cheeger, B

    J. Cheeger, B. Kleiner, and A. Schioppa. Infinitesimal S tructure of Differentiability Spaces, and Metric Differentiation. Anal. Geom. Metr. Spaces , 4:Art. 5, 2016

    2016

  6. [12]

    J. P. Chen and R. G. Niemeyer. Periodic billiard orbits o f self-similar Sierpi´ nski carpets. J. Math. Anal. Appl. , 416(2):969–994, 2014

    2014

  7. [13]

    Fractured fractals and broken d reams

    G. David and S. Semmes. “Fractured fractals and broken d reams”, volume 7 of Oxford Lecture Series in Mathematics and its Applications . The Clarendon Press, Oxford University Press, New York, 1997

    1997

  8. [15]

    G. C. David. Bi-Lipschitz pieces between manifolds. Rev. Mat. Iberoam., 32(1):175–218, 2016

    2016

  9. [16]

    G. C. David and K. Kinneberg. Lipschitz and bi-Lipschit z maps from PI spaces to Carnot groups. to appear, Indiana Univ. Math. J. , Preprint, 2017. arXiv:1711.03533

    2017 arXiv

  10. [17]

    G. C. David and B. Kleiner. Rectifiability of planes and A lberti representations. Ann. Sc. Norm. Super. Pisa Cl. Sci. (5) , 19(2):723–756, 2019. arXiv:1611.05284

    2019 arXiv

  11. [18]

    De Philippis, A

    G. De Philippis, A. Marchese, and F. Rindler. On a conjec ture of Cheeger. In Measure theory in non-smooth spaces , Partial Differ. Equ. Meas. Theory, pages 145–155. De Gruyte r Open, W arsaw, 2017. 36 GUY C. DA VID

    2017

  12. [19]

    Durand-Cartagena and J

    E. Durand-Cartagena and J. T. Tyson. Rectifiable curves in Sierpi´ nski carpets.Indiana Univ. Math. J. , 60(1):285–309, 2011

    2011

  13. [20]

    P. Erd¨ os. The dimension of the rational points in Hilbe rt space. Ann. of Math. (2) , 41:734– 736, 1940

    1940

  14. [21]

    Eriksson-Bique

    S. Eriksson-Bique. Characterizing spaces satisfying Poincar´ e inequalities and applications to differentiability. Geom. Funct. Anal. , 29(1):119–189, 2019

    2019

  15. [22]

    Gigli and E

    N. Gigli and E. Pasqualetto. Behaviour of the reference measure on RCD spaces under charts. Preprint, 2016. arXiv:1607.05188

    2016 arXiv

  16. [23]

    Lectures on analysis on metric spaces

    J. Heinonen. “Lectures on analysis on metric spaces”. U niversitext. Springer-Verlag, New York, 2001

    2001

  17. [24]

    W. B. Johnson, J. Lindenstrauss, D. Preiss, and G. Schec htman. Uniform quotient mappings of the plane. Michigan Math. J. , 47(1):15–31, 2000

    2000

  18. [25]

    S. Keith. A differentiable structure for metric measure spaces. Adv. Math. , 183(2):271–315, 2004

    2004

  19. [26]

    Kell and A

    M. Kell and A. Mondino. On the volume measure of non-smoo th spaces with Ricci curvature bounded below. Ann. Sc. Norm. Super. Pisa Cl. Sci. (5) , 18(2):593–610, 2018

    2018

  20. [27]

    Kleiner and B

    B. Kleiner and B. Leeb. Rigidity of quasi-isometries fo r symmetric spaces and Euclidean buildings. Inst. Hautes ´Etudes Sci. Publ. Math. , (86):115–197 (1998), 1997

    1998

  21. [28]

    Kleiner and A

    B. Kleiner and A. Schioppa. PI spaces with analytic dime nsion 1 and arbitrary topological dimension. Indiana Univ. Math. J. , 66(2):495–546, 2017

    2017

  22. [29]

    T. Laakso. Ahlfors Q-regular spaces with arbitrary Q > 1 admitting weak Poincar´ e inequality. Geom. Funct. Anal. , 10(1):111–123, 2000

    2000

  23. [30]

    U. Lang, B. Pavlovi´ c, and V. Schroeder. Extensions of L ipschitz maps into Hadamard spaces. Geom. Funct. Anal. , 10(6):1527–1553, 2000

    2000

  24. [31]

    Lang and T

    U. Lang and T. Schlichenmaier. Nagata dimension, quasi symmetric embeddings, and Lips- chitz extensions. Int. Math. Res. Not. , (58):3625–3655, 2005

    2005

  25. [32]

    Le Donne

    E. Le Donne. Metric spaces with unique tangents. Ann. Acad. Sci. Fenn. Math. , 36(2):683– 694, 2011

    2011

  26. [33]

    Le Donne

    E. Le Donne. A primer on Carnot groups: homogenous group s, Carnot-Carath´ eodory spaces, and regularity of their isometries. Anal. Geom. Metr. Spaces , 5(1):116–137, 2017

    2017

  27. [34]

    Le Donne and T

    E. Le Donne and T. Rajala. Assouad dimension, Nagata dim ension, and uniformly close metric tangents. Indiana Univ. Math. J. , 64(1):21–54, 2015

    2015

  28. [35]

    J. R. Lee and A. Sidiropoulos. Near-optimal distortion bounds for embedding doubling spaces into L1 [extended abstract]. In STOC’11—Proceedings of the 43rd ACM Symposium on Theory of Computing , pages 765–772. ACM, New York, 2011

    2011

  29. [36]

    Montgomery

    R. Montgomery. A tour of subriemannian geometries, their geodesics and app lications, vol- ume 91 of Mathematical Surveys and Monographs . American Mathematical Society, Provi- dence, RI, 2002

    2002

  30. [37]

    Modern dimension theory , volume 2 of Sigma Series in Pure Mathematics

    J Nagata. Modern dimension theory , volume 2 of Sigma Series in Pure Mathematics . Hel- dermann Verlag, Berlin, revised edition, 1983

    1983

  31. [38]

    P. Pansu. M´ etriques de Carnot-Carath´ eodory et quasiisom´ etries des espaces sym´ etriques de rang un. Ann. Math., (2) , 129(1):1–60, 1989

    1989

  32. [39]

    S. D. Pauls. The large scale geometry of nilpotent Lie gr oups. Comm. Anal. Geom. , 9(5):951– 982, 2001

    2001

  33. [41]

    S. Semmes. On the nonexistence of bi-Lipschitz paramet erizations and geometric problems about A∞ -weights. Rev. Mat. Iberoamericana, 12(2):337–410, 1996. Department of Mathematical Sciences, Ball State University , Muncie, IN 47306 E-mail address : gcdavid@bsu.edu

    1996

Pith tools

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