betaFeedback

WorksMathematics

Constructions and reductions on the Hornich-Hlawka inequality and its generalizations

Abstract

We investigate five open problems concerning the Hornich-Hlawka inequality and its generalizations posed by Marinescu and Niculescu. We give constructive answers to three of them. First, we construct an equivalent norm on ℓ2\ell^2 with no three-dimensional Hornich-Hlawka subspace. Second, we present a Nakano space that is not Hornich-Hlawka. Third, we show that the αα-power generalization of the Hornich-Hlawka inequality fails to hold on nonzero normed spaces when α∈(1,2)∪(2,∞)α\in (1, 2) \cup (2, \infty). This implies that on every nonzero inner product space, the αα-power generalization of the Hornich-Hlawka inequality holds for all triples precisely when α∈(0,1]∪{2}α\in (0,1] \cup \{2\}. For the remaining two problems, we establish reductions. We prove that ℓp(N)\ell^p(N) is Hornich-Hlawka for all N≥3N \geq 3 and 2<p≤log⁡3/(log⁡3−log⁡2)2 < p \leq \log 3 /(\log 3 - \log 2) if and only if it is Hornich-Hlawka at N=3N = 3 and p=log⁡3/(log⁡3−log⁡2)p = \log 3 /(\log 3 - \log 2). Finally, for p(⋅):(0,1)→[1,2]p(\cdot): (0, 1) \to [1, 2] measurable and integers k≥0k \geq 0, we establish a Sobolev-Nakano equivalence: we prove that Wk,p(⋅)(0,1)W^{k,p(\cdot)}(0,1), equipped with the sum of the Luxemburg norms of its derivatives of orders 0,…,k0, \ldots, k, is Hornich-Hlawka if and only if Lp(⋅)(0,1)L^{p(\cdot)}(0,1) is Hornich-Hlawka.

Introduction

For every x,y,z∈Rnx,y,z \in\mathbb{R}^n, the following inequality holds, called the Hornich-Hlawka inequality:

∥x∥2+∥y∥2+∥z∥2+∥x+y+z∥2≥∥x+y∥2+∥y+z∥2+∥z+x∥2(1)\lVert x\rVert^2+\lVert y\rVert^2+\lVert z\rVert^2+\lVert x+y+z\rVert^2 \ge\lVert x+y\rVert^2+\lVert y+z\rVert^2+\lVert z+x\rVert^2 \tag*{(1)}

The inequality can be traced back to Hornich and Hlawka [10, 8], who stated and proved the inequality. Hlawka’s elementary proof involves multiplying LHS - RHS by the LHS and invoking the Fréchet identity

∥x∥2+∥y∥2+∥z∥2+∥x+y+z∥2=∥x+y∥2+∥y+z∥2+∥z+x∥2(2)\lVert x\rVert^2+\lVert y\rVert^2+\lVert z\rVert^2+\lVert x+y+z\rVert^2=\lVert x+y\rVert^2+\lVert y+z\rVert^2+\lVert z+x\rVert^2 \tag*{(2)}

which is equivalent to the parallelogram law. The Hornich-Hlawka inequality then reduces to the triangle inequality. This proof generalizes to all inner product spaces (and hence Hilbert spaces).

Let XX be a Banach space, and let x,y,z∈Xx,y,z \in X. We say that XX is a Hornich-Hlawka space if its norm satisfies the Hornich-Hlawka inequality

∥x∥+∥y∥+∥z∥+∥x+y+z∥≥∥x+y∥+∥y+z∥+∥z+x∥(3)\lVert x\rVert+\lVert y\rVert+\lVert z\rVert+\lVert x+y+z\rVert\ge\lVert x+y\rVert+\lVert y+z\rVert+\lVert z+x\rVert \tag*{(3)}

for all x,y,z∈Xx,y,z \in X. In this case, we also say that XX has the Hornich-Hlawka property, or simply that XX is Hornich-Hlawka. A Hornich-Hlawka space is also called a quadrilateral space in some literature [25, 24], as the Hornich-Hlawka inequality is equivalent to an inequality on quadrilaterals. Although every Hilbert space is Hornich-Hlawka, not every Banach space is Hornich-Hlawka.

Recently, Marinescu and Niculescu [17] conducted a broad survey of the Hornich-Hlawka inequality and its generalizations; as illustrated there, the Hornich-Hlawka inequality enjoys connections to several areas of mathematics, including operator theory [LP1968; AK2012], probability and statistics [13, 6], and convex geometry [Lev1949; VA1969; MPF1993]. The authors of [17] also posed five open problems in Banach space theory relating to the Hornich-Hlawka inequality and its generalizations that were described as seeming “of greatest interest.” We follow their conventions and number these problems as Problems 1, 2, 3, 4, and 5.

In this paper, we explore each of these problems in detail.

Main results

First, for Problem 1, we prove two reductions (Theorem 4.3 and Corollary 4.7) that imply the following:

Theorem 1.1. Let p0=log⁡3/(log⁡3−log⁡2)≈2.7095⋯p_0 = \log3/(\log3-\log2) \approx2.7095\cdots. Then the space ℓp(N)\ell^p(\mathbb{N}) is Hornich-Hlawka for all N≥3N \ge3 and 2<p≤p02 < p \le p_0 if and only if ℓp0(3)\ell^{p_0}(3) is Hornich-Hlawka.

Thus, if ℓp0(3)\ell^{p_0}(3) were proven to be Hornich-Hlawka, then we would obtain a complete classification of finite-dimensional ℓp\ell^p spaces that are Hornich-Hlawka (see Table 1). We also provide two convex-geometric reformulations of the Hornich-Hlawka inequality in Theorem 4.10.

1≤p≤21\le p\le22<p≤(log⁡3)/(log⁡3−log⁡2)2<p\le(\log3)/(\log3-\log2)(log⁡3)/(log⁡3−log⁡2)<p≤∞(\log3)/(\log3-\log2)<p\le\infty
N≤2N\le2Hornich-HlawkaHornich-HlawkaHornich-Hlawka
N≥3N\ge3Hornich-HlawkaUnknownnot Hornich-Hlawka

Table 1. Landscape of ℓp(N)\ell^p(N) spaces with regards to the Hornich-Hlawka property. The top row (Corollary 3.5) and leftmost column (Corollary 3.7) both follow from L1L^1-embeddability (Corollary 3.2); the bottom right cell is due to Corollary 3.10.

For Problem 2, we prove the existence of a renorming on ℓ2\ell^2 that admits an infinite-dimensional Banach space without a three-dimensional Hornich-Hlawka subspace:

Theorem 1.2. There exists an equivalent norm ∥⋅∥X\|\cdot\|_X on the real Hilbert space ℓ2\ell^2 such that no three-dimensional subspace of X=(ℓ2,∥⋅∥X)X = (\ell^2,\|\cdot\|_X) is Hornich-Hlawka.

For Problem 3, we exhibit an explicit function p(⋅):(0,1)→[1,2]p(\cdot) : (0,1) \to[1,2] such that the Nakano space Lp(⋅)(0,1)L^{p(\cdot)}(0,1) is not Hornich-Hlawka:

Theorem 1.3. Define the function

p(t)={1,0<t<35,5t−2,35≤t≤45,2,45<t<1p(t) = \begin{cases} 1, & 0 < t < \frac{3}{5}, \\ 5t - 2, & \frac{3}{5} \le t \le\frac{4}{5}, \\ 2, & \frac{4}{5} < t < 1 \end{cases}

Then Lp(⋅)(0,1)L^{p(\cdot)}(0,1) is not Hornich-Hlawka.

For Problem 4, we provide an explicit reduction of a Sobolev space Wk,p(⋅)(0,1)W^{k,p(\cdot)}(0,1) having the Hornich-Hlawka property to the associated Nakano space Lp(⋅)(0,1)L^{p(\cdot)}(0,1) when assuming a particular choice of norm for the Sobolev space.

Theorem 1.4. Let p(⋅):(0,1)→[1,2]p(\cdot) : (0,1) \to[1,2] be measurable and let k∈N∪{0}k \in\mathbb{N} \cup\{0\}. With the choice of the norm ((4)) for Wk,p(⋅)(0,1)W^{k,p(\cdot)}(0,1), the Sobolev space Wk,p(⋅)(0,1)W^{k,p(\cdot)}(0,1) is Hornich-Hlawka if and only if Lp(⋅)(0,1)L^{p(\cdot)}(0,1) is Hornich-Hlawka.

For Problem 5, we give two elementary examples that show that the α\alpha-power generalization of the Hornich-Hlawka inequality cannot hold for any nonzero normed space when α∈(1,2)∪(2,∞)\alpha\in(1,2) \cup(2,\infty). As the statements and proofs of both examples are short and elementary, we refer the reader to Theorems 7.2 and 7.3 in Section 7 for the details.

The remainder of the paper is organized as follows. § 2 provides the common notation that will be used throughout the paper. § 3 contains a review of preliminary results. Sections 4 through 7 discuss our original contributions to each numbered problem in order.

Notation and Terminology

When writing the variables pp and NN, we always assume that 1≤p≤∞1 \le p \le\infty and N∈NN \in\mathbb{N} unless their ranges are otherwise specified.

Following the notation of [17], we denote by ℓp(N)=(RN,∥⋅∥p)\ell^p(N) = (\mathbb{R}^N,\|\cdot\|_p) the space of NN-tuples equipped with the pp-norm, and we denote by ℓp\ell^p the infinite-dimensional sequence-space analogue. The ambient field is always assumed to be the field of real numbers R\mathbb{R}.

The triple (Ω,A,μ)(\Omega,\mathcal{A},\mu) will always denote a measure space. In the context of probability spaces, we will write P\mathbb{P} instead of μ\mu to emphasize that the measure function is a probability measure.

We write Lp(Ω,μ)L^p(\Omega,\mu) to denote the space of pp-th power Lebesgue integrable functions f:Ω→Rf:\Omega\to\mathbb{R} with respect to μ\mu, with the understanding that p=∞p=\infty means we take the essential supremum norm. Moreover, when XX is a Banach space, we write LXp(Ω,μ)L_X^p(\Omega,\mu) to denote the space of pp-th power Bochner integrable functions f:Ω→Xf:\Omega\to X with respect to μ\mu, again separating the p=∞p=\infty case to mean the essential supremum. In this setting, we call XX the codomain space of LXp(Ω,μ)L_X^p(\Omega,\mu). In particular, if X=RX=\mathbb{R}, then we recover the set of pp-th power Lebesgue integrable functions LRp(Ω,μ)=Lp(Ω,μ)L_{\mathbb{R}}^p(\Omega,\mu)=L^p(\Omega,\mu).

We write Lp(⋅)(0,1)L^{p(\cdot)}(0,1) to denote a Nakano space, where p(⋅):(0,1)→[1,2]p(\cdot):(0,1)\to[1,2] is variable exponent function that is measurable. Similarly, we write Wk,p(⋅)(0,1)W^{k,p(\cdot)}(0,1) to denote a Sobolev space, where kk is a nonnegative integer and p(⋅):(0,1)→[1,2]p(\cdot):(0,1)\to[1,2] is variable exponent function that is measurable. In particular, if k=0k=0, then these two spaces coincide.

Throughout this paper, we assume the standard Luxemburg norm

∥f∥p(⋅)=inf⁡{λ>0:∫01∣f(t)λ∣p(t) dt≤1}\|f\|_{p(\cdot)}=\inf\left\{\lambda>0:\int_0^1\left|\frac{f(t)}{\lambda}\right|^{p(t)}\,dt\le1\right\}

for the Nakano spaces Lp(⋅)(0,1)L^{p(\cdot)}(0,1), and the norm

∥f∥k,p(⋅)=∑j=0k∥f(j)∥p(⋅)(4)\|f\|_{k,p(\cdot)}=\sum_{j=0}^{k}\|f^{(j)}\|_{p(\cdot)} \tag*{(4)}

for the Sobolev spaces Wk,p(⋅)(0,1)W^{k,p(\cdot)}(0,1), where we interpret the derivatives as weak derivatives. We remark that although there are several topologically equivalent norms for Sobolev spaces, our specific choice will allow to give an elegant partial answer to Problem 4 as in Theorem 1.4.

Finally, let p0p_0 represent the constant

p0=log⁡3log⁡3−log⁡2≈2.7095⋯ .p_0=\frac{\log3}{\log3-\log2}\approx2.7095\cdots.

Preliminaries

The Hornich-Hlawka property of ℓp(N)\ell^p(N) spaces

First, we establish the current standings of which ℓp(N)\ell^p(N) spaces are known to be Hornich-Hlawka, known not to be Hornich-Hlawka, or unknown. Table 1 summarizes these standings.

We will survey the known cases (the green and red cells of Table 1). The organization and exposition will closely mirror those of [17].

The key idea behind the ℓp(N)\ell^p(N) spaces that are known to be Hornich-Hlawka is that they can be embedded into an L1L^1 space, which is itself Hornich-Hlawka. This is formally described in Proposition 3.1 and Corollary 3.2.

Proposition 3.1. [17 Lemma 1] If XX is a Hornich-Hlawka space, then every space LX1(Ω,μ)L_X^1(\Omega,\mu) is Hornich-Hlawka. In particular, every space L1(Ω,μ)L^1(\Omega,\mu) is Hornich-Hlawka.

Corollary 3.2. Every space that embeds linearly and isometrically into an L1L^1 space is Hornich-Hlawka.

Remark 3.3. The converse to Corollary 3.2 is false; Neyman [21 Remark 3.5] showed that there is no finite list of inequalities that characterizes embeddability in L1(0,1)L^1(0,1). So we cannot conclude that a space that is not linearly and isometrically embeddable into an L1L^1 space is not Hornich-Hlawka.

Corollary 3.2 together with the following proposition allows us to verify the Hornich-Hlawka property of some ℓp(N)\ell^p(N) spaces.

Proposition 3.4. [6 Theorem 1, 7 Theorem 2, Lin1964, Theorem 3] If XX is a two-dimensional Banach space, then XX is a Hornich-Hlawka space.

Corollary 3.5. If N≤2N \le2, then ℓp(N)\ell^p(N) is a Hornich-Hlawka space.

Proposition 3.6. [3 Theorems 1, 2, 18 Theorem 1.1] If 1≤r≤p≤21 \le r \le p \le2 and N∈NN \in\mathbb{N}, then the spaces ℓp(N)\ell^p(N), ℓp\ell^p, and Lp(0,1)L^p(0,1) embed linearly and isometrically into Lr(0,1)L^r(0,1).

The following corollary follows from Propositions 3.1 and 3.6.

Corollary 3.7. If 1≤p≤21 \le p \le2, then ℓp(N)\ell^p(N), ℓp\ell^p, and Lp(0,1)L^p(0,1) are all Hornich-Hlawka spaces.

We also have some elementary negative results. We may verify the following by a straightforward calculation.

Proposition 3.8. [Wit1973, Proposition 2] If p0<p≤∞p_0 < p \le\infty, then ℓp(3)\ell^p(3) is not a Hornich-Hlawka space. In particular,

x=(−1,1,1),y=(1,−1,1),z=(1,1,−1)x=(-1,1,1),\qquad y=(1,-1,1),\qquad z=(1,1,-1)

provides a counterexample to the Hornich-Hlawka inequality (3).

The following proposition allows us to extend the results of Proposition 3.8 to ℓp(N)\ell^p(N) spaces where N≥3N \ge3.

Proposition 3.9. Let N≥3N \ge3. If ℓp(3)\ell^p(3) is not Hornich-Hlawka, then ℓp(N)\ell^p(N) is not Hornich-Hlawka.

Proof. Assume x,y,z∈ℓp(3)x,y,z \in\ell^p(3) violate the Hornich-Hlawka inequality (3). Then the zero-padded tuple

References

  1. [1]K. M. R. Audenaert and F. Kittaneh. “Problems and Conjectures in Matrix and Operator Inequalities.” In: arXiv preprint arXiv:1201.5232v3 (2012), pp. 12–14.arxiv.org/abs/1201.5232
  2. [2]A. V. Balakrishnan. “Fractional Powers of Closed Operators and the Semigroups Generated by Them.” In: Pacific Journal of Mathematics 10.2 (1960), pp. 419–437.DOI
  3. [3]J. Bretagnolle, D. Dacunha-Castelle, and J.-L. Krivine. “Lois Stables et Espaces Lₚ.” In: Annales de l’Institut Henri Poincaré 2 (1966), pp. 231–259.DOI
  4. [4]L. E. Dor. “Potentials and isometric embeddings in L₁.” In: Israel Journal of Mathematics 24 (1976), pp. 260–268.
  5. [5]E. Esa. Dimension-independent Hlawka constants for Schatten norms. Palomar, PALOMAR-2026-09-25-000006 v1. 2026. URL: https://palomar-registry.org/entry?id=PALOMAR-2026-09-25-000006&version=1.
  6. [6]T. S. Ferguson. “A Representation of the Symmetric Bivariate Cauchy Distribution.” In: Annals of Mathematical Statistics 33.4 (1962), pp. 1256–1266.DOI
  7. [7]C. S. Herz. “A Class of Negative-Definite Functions.” In: Proceedings of the American Mathematical Society 14.4 (1963), pp. 670–676.DOI
  8. [8]E. Hlawka. Ungleichungen. Vorlesungen über Mathematik. MANZ Verlags- und Universitätsbuch, 1990, p. 86.
  9. [9]A. Honda, Y. Okazaki, and Y. Takahashi. “Generalizations of the Hlawka’s Inequality.” In: Bulletin of the Kyushu Institute of Technology Pure and Applied Mathematics 45 (1998), pp. 9–15.DOI
  10. [10]H. Hornich. “Eine Ungleichung für Vektorlängen.” In: Mathematische Zeitschrift 48 (1942), pp. 268–274.DOI
  11. [11]S. Janson. “Stable distributions.” In: arXiv preprint arXiv:1112.0220v3 (2022).
  12. [13]P. Levy. “Théorie de l’addition des variables aléatoires.” In: Bulletin de la S. M. F. 67 (1939), pp. 1–41.
  13. [16]A. Marchaud. “Sur les dérivées et sur les différences des fonctions de variables réelles.” In: Journal de mathématiques pures et appliquées. 9th ser. 6 (1927), pp. 337–425.
  14. [17]D. S. Marinescu and C. P. Niculescu. “A New Look at the Hornich-Hlawka Inequality.” In: Results in Mathematics 80.149 (2025), pp. 1–21.DOI
  15. [18]T. M. McNicholl. “Continuous logic and embeddings of Lebesgue spaces.” In: Archive for Mathematical Logic 60 (2021), pp. 105–119.arxiv.org/abs/1805.11561
  16. [20]D. Moon. Hlawka’s inequality in ℓₚ for 2 ≤ p ≤ log 3/log(3/2). 2026. URL: https://prove2.me/theorems/9e0c08c7-583b-41b0-b8a3-57fa9532b2c1.
  17. [21]A. Neyman. “Representation of Lₚ-norms and isometric embedding in Lₚ-spaces.” In: Israel Journal of Mathematics 48.2–3 (1984), pp. 129–138.
  18. [22]W. Rudin. Functional Analysis. 2nd ed. McGraw-Hill, 1991, pp. 92–103.
  19. [24]A. Simon and P. Volkmann. “On Two Geometric Inequalities.” In: Annales Mathematicae Silesianae 9 (1995), pp. 137–140.
  20. [25]D. M. Smiley and M. F. Smiley. “The Polygonal Inequalities.” In: The American Mathematical Monthly 71.7 (1964), pp. 755–760.DOI
  21. [26]P. R. Stinga. User’s guide to the fractional Laplacian and the method of semigroups. Ed. by A. Kochubei and Y. Luchko. De Gruyter Reference. De Gruyter Brill, 2019, pp. 235–266.arxiv.org/abs/1808.05159
  22. [28]H. Weyl. “Bemerkungen zum Begriff des Differentialquotienten gebrochener Ordnung.” In: Vierteljahrsschrift der Naturforschenden Gesellschaft in Zürich 62 (1917), pp. 296–302.

Paper details

vlnc:W_4fnpw8ex
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 9

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Mathematics

Gemini 2.5 Pro (bare)

View statement
Looking for related papers…
Contents