REVIEW 5 major objections 5 minor 34 references
Bourgain-Morrey-Lorentz spaces and operators on them
T0 review · 5 major / 5 minor · reviewed 2026-08-07 · deepseek-v4-flash
Pith's one-line read New Morrey-type spaces have a dyadic block space as their predual, and this duality drives the paper's operator bounds.
desk verdict Genuinely new predual for Bourgain-Morrey-Lorentz spaces, but the dyadic-norm equivalence is a real gap that blocks the operator applications as written. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The central object is the Bourgain-Morrey-Lorentz norm $$\|f\|_{$M^{{t,r}}$_{p,q}} = \left(\sum_{Q\in\mathcal{D}} |Q|^{r/t-r/p}\|f\|_{$L^{{p,q}}$(Q)}^r\right)^{1/r},$$ which combines Morrey's scale-weighting with Lorentz's fine second parameter. The dual block space is built from $(p',q',t')$-blocks, so the duality is carried by the integral pairing and by two local ingredients: Hölder's inequality in Lorentz spaces (Lemma 2.4) and the fact that restricting a functional to a cube gives an $L^{p,q}$ function. A separate load-bearing mechanism is the equivalence between the standard dyadic norm and norms computed on any one of the $3^n$ shifted dyadic grids (Lemma 2.16); this is what lets the Hardy-Littlewood maximal operator be controlled by its dyadic pieces.
What would settle it
Compute the ratio of the Bourgain-Morrey-Lorentz norms on two different shifted dyadic grids for a sparse sum of characteristic functions of cubes chosen so that each is covered by a cube from only one grid; an unbounded ratio as the number of scales grows would falsify Lemma 2.16 and break the maximal-operator application. For the duality claim itself, exhibit a continuous linear functional on the block space that is not given by integration against any locally integrable function, or a nonzero $f\in M^{t,r}_{p,q}$ whose integral against every block vanishes; either would refute Theorem 3.5.
Extended reading notes
Core claim
The paper's main discovery is the duality $M^{t,r}_{p,q}(\mathbb{R}^n) \cong (H^{t',r'}_{p',q'}(\mathbb{R}^n))^*$ for $1<q<\infty$ and either $1<p<t<r<\infty$ or $1<p\le t<r=\infty$, with the pairing $\int_{\mathbb{R}^n} f g\,dx$ and the norm equivalences (3.3)-(3.4). Here a $(p',q',t')$-block is a function supported on a cube $Q$ with $\|b\|_{L^{p',q'}} \le |Q|^{1/p'-1/t'}$, and $H^{t',r'}_{p',q'}$ consists of sums of such blocks with $\ell^{r'}$ coefficients. The proof embeds $M$ into the dual by Hölder's inequality in Lorentz spaces, proves injectivity through local $L^{p,q}$ duality, and establishes surjectivity by showing that a functional on blocks restricts to each cube as an $L^{p,q}$ function and then assembling these local functions with an $\ell^{r'}$ estimate. As a by-product, the block space has the Fatou property and dense smooth compactly supported functions.
Load-bearing premise
The load-bearing premise is that measuring a function cube-by-cube on one fixed shifted dyadic grid gives the same total size as measuring on the standard dyadic grid; if that equivalence fails, the Hardy-Littlewood maximal bound on the new spaces collapses, independently of the duality theorem.
Editorial extensions
If this is right
- Every continuous linear functional on the block space $H^{t',r'}_{p',q'}$ is integration against a unique function in $M^{t,r}_{p,q}$, so the two spaces form a dual pair with the norm formulas (3.3)-(3.4).
- The Hardy-Littlewood maximal operator, the sharp maximal operator, Calderón-Zygmund operators, fractional integral operators, and commutators with BMO functions are bounded on $M^{t,r}_{p,q}$ and on the block spaces under the stated parameter conditions.
- A weak Hardy factorization holds: each function in $H^1(\mathbb{R}^n)$ can be written as an $\ell^1$ sum of terms $g_k T^*(h_k)-h_k T(g_k)$, with a norm equivalence involving the block and Bourgain-Morrey-Lorentz norms of the factors.
- A locally integrable function $b$ lies in BMO exactly when the commutator $[b,T]$ is bounded on $M^{t,r}_{p,q}$ for a homogeneous Calderón-Zygmund operator $T$, and it lies in CMO exactly when the commutator is compact.
- The spaces are nontrivial only in the parameter ranges $p<t<r<\infty$ or $p\le t<r=\infty$, and in those ranges they are not isomorphic to Lorentz spaces.
Reading between the lines
- The duality theorem itself does not rest on Lemma 2.16; if that lemma's proof fails, the maximal-operator and sharp-maximal results in Section 4 would need repair, but the predual identification could still stand.
- The block-space predual should give a natural atomic decomposition for $H^{t',r'}_{p',q'}$, making these spaces amenable to interpolation in the parameter $q$; the paper does not pursue this.
- The Hardy factorization and the BMO/CMO characterizations suggest that the same machinery could extend to weighted or vector-valued variants of Bourgain-Morrey-Lorentz spaces, where the dyadic-grid equivalence would need a weighted analogue.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper introduces Bourgain-Morrey-Lorentz spaces M^{t,r}_{p,q}(R^n), a Lorentz-parameter extension of Bourgain-Morrey spaces, and studies their fundamental properties, duality, and the boundedness of classical harmonic analysis operators on them. The main positive result is Theorem 3.5, which identifies the predual of M^{t,r}_{p,q}(R^n) with a block space H^{t',r'}_{p',q'}(R^n). Building on this duality, the paper claims boundedness of the Hardy-Littlewood maximal operator, sharp maximal operator, Calderón-Zygmund operators, fractional integrals, and commutators on these spaces, plus a weak Hardy factorization and a BMO characterization via commutator boundedness, and a compactness criterion for commutators.
Significance. If the main results are correct, the paper extends the known duality theory for Bourgain-Morrey and Morrey-Lorentz spaces and offers a unified framework for operator boundedness in this class. The predual theorem (Theorem 3.5) is the clear centerpiece and appears to be supported by a plausible argument. The paper also explicitly identifies prior sources ([15], [10], [25]) and avoids circular reasoning or parameter fitting. However, several load-bearing proofs, especially Lemma 2.16, Theorem 4.12/4.14, Theorem 4.25, and Theorem 5.6, contain gaps that affect the advertised applications. The core duality result may survive, but the operator-theoretic claims across Sections 4-6 require substantial repair.
major comments (5)
- [Theorem 4.12 and Corollary 4.14] The proof of Lemma 2.16 is defective. Lemma 2.15 produces, for each cube Q, a cube R in the union of all 3^n shifted dyadic grids, not necessarily in the fixed grid D. The display then sums over R∈D as though the assignment Q↦R stayed inside D. No argument establishes R∈D, and no bounded-multiplicity estimate for the map Q↦R is given. The reverse inequality is equally unjustified: from R⊂13Q one cannot sum over R∈D. This lemma is load-bearing because Theorem 4.1 uses it to transfer dyadic-maximal bounds from each shifted grid back to the standard M^{t,r}_{p,q} norm. A correct proof needs a covering by cubes of the same grid with controlled multiplicity (e.g., for each Q∈D find R∈D with Q⊂R and |R|≤C|Q|, or use the bounded overlap of the 3^n grids and adjust the ℓ^r summation). Until this is supplied, the maximal-operator boundedness and all later operator results that depend on it are not established.
- [Theorem 4.25] Theorem 4.12 and Corollary 4.14 state the impossible condition 1/t2 = 1/t2 − α/n. The intended condition is presumably 1/t2 = 1/t1 − α/n, as used in the proof. As written, the hypotheses are empty. Additionally, the proof of Theorem 4.12 applies Theorem 4.1 to the scaled parameters p2(1−s), q2(1−s), t2(1−s), r2(1−s) with s = t1α/n. Theorem 4.1 requires 1≤q≤∞, but q2(1−s) can be < 1 under the stated assumptions, so the invoked maximal bound may be out of range. The theorem should be restated with correct exponent relations and, if the fractional-integral result is intended for all allowed parameters, a proof that covers the quasi-norm case q<1.
- [Theorem 5.6] The proof of Theorem 4.25 asserts ∥M_η(f)∥_{M^{t,r}_{p,q}} ≲ ∥f∥_{M^{t,r}_{p,q}} for η∈(1,min(p,q)) without proof. Theorem 4.1 establishes boundedness of M only for η=1. The cited self-improvement result [28] applies to quasi-Banach lattices with the Fatou property, but the Fatou property is proved only for the block space H (Theorem 3.8), not for M^{t,r}_{p,q}. Unless the Fatou property for M^{t,r}_{p,q} is established and the self-improvement argument is invoked, or a direct proof of the M_η bound is given, the commutator boundedness theorem lacks the necessary justification at this step.
- [Theorem 6.2, Case 1] Theorem 5.6 claims the two-sided norm equivalence ∥f∥_{H^1} ≈ inf Σ |λ_{k,j}| ∥g_{k,j}∥_{H'} ∥h_{k,j}∥_M. The proof constructs a representation with the sum bounded by C∥f∥_{H^1}, which gives the upper bound for the infimum. The reverse inequality, inf ≥ c∥f∥_{H^1}, requires showing that every term gT*(h)−hT(g) lies in H^1 with norm controlled by ∥g∥_{H'}∥h∥_M. This membership is asserted without proof; the proof of Lemma 5.5 only bounds the H^1 distance between an atom and such a term, not the H^1 norm of the term itself. Since the claimed equivalence is used in the statement of the factorization theorem and in Theorem 5.7, the missing lower bound must be provided.
- [Theorem 4.15 proof] In the proof of Theorem 6.2, Case 1, the argument obtains convergence of [b,T](φ_k) to G in M^{t,r}_{p,q} and also convergence to 0 in M^{t2,r}_{p,q} with t2 > t, claiming a contradiction with ∥G∥_{M^{t,r}_{p,q}} ≳ 1. However, no embedding between M^{t,r}_{p,q} and M^{t2,r}_{p,q} is stated or proved; convergence in the second space does not automatically imply the limit in the first is zero. The contradiction requires an additional argument (e.g., an embedding or a localization estimate) that is not present.
minor comments (5)
- [Section 4.2] In the proof of Theorem 4.15, the notation H^{q',r'}_{p',q'} appears (in the estimate for I_k) where H^{t',r'}_{p',q'} is clearly intended.
- [Section 5.5] In the proof of Theorem 4.12, the power rule for the norm is written with compressed notation; defining s = 1 − t1α/n explicitly would make the chain of equalities much easier to follow.
- [Section 4.5] In the proof of Theorem 5.7, the interchange of the infinite sum over j,k with the limit L→∞ in ⟨b_L, f⟩ is not justified in the text. Since the terms g_{k,j}T*(h_{k,j})−h_{k,j}T(g_{k,j}) have compact support, a dominated-convergence argument is possible, but it should be written out.
- [References] The definition of the finite overlapping property (Definition 4.5) requires that every admissible decomposition of f be finite overlapping, which is a very strong condition. Theorem 4.7 claims the constructed decomposition of M(b) satisfies this property, but no proof is given. If the property is not actually needed for the subsequent arguments, it should be simplified or removed; if it is needed, a verification must be provided.
- [Throughout] There are minor typographical issues in the references, e.g., the title of [15] is garbled as 'J. Funct. Anal., 284 (1) (2023) 52' and should be completed; also the notation 'Mont' appears in the reference to [15] in Section 1.
Circularity Check
No significant circularity: the main duality theorem and operator bounds are derived from Lorentz-space duality, block decompositions, and external prior results; the only notable defect is a non-circular proof gap in Lemma 2.16.
full rationale
Cirularity analysis: I find no circular step. The main duality result, Theorem 3.5, is proved from Lorentz-space duality, Hölder's inequality, block decompositions, and weak-* compactness; the reverse inclusion (H)^* ⊂ M is a direct dual estimate with arbitrary finitely supported ℓ^{r'} coefficients, not an assumption of the conclusion. Theorem 3.12 is a corollary of Theorem 3.5 plus the Fatou property of block spaces. The operator bounds (Theorems 4.1, 4.20, 4.25) rest on standard external results: Lemma 2.15 from [15], Lemma 2.6 from [13], Theorem 1.1 from [28], the sharp-maximal pointwise estimate from [29]/[11], and duality with Theorem 3.5. The only self-references are [1] in the introduction, which is not load-bearing. The notable weakness is Lemma 2.16: the proof finds for each cube Q a cube R in the union of the 3^n shifted grids, but then sums over R ∈ D without proving R ∈ D or giving a bounded-multiplicity summation over the fixed grid; this is a genuine proof gap in the dyadic-reduction route to Theorem 4.1, but it is a correctness defect, not a circularity, because Lemma 2.16 is not assumed and its conclusion does not reduce to the theorem it supports. Thus the circularity score is 0.
Assumptions & free parameters
assumptions (5)
- standard math Duality (L^{p,q}(R^n))^* = L^{p',q'}(R^n) for 1<p,q<∞
- standard math Boundedness of the Hardy-Littlewood maximal operator on L^{p,q}(R^n) for 1<p<∞ and 1≤q≤∞
- standard math Self-improvement of the maximal operator on quasi-Banach lattices with the Fatou property (Theorem 1.1 of [28])
- standard math Compactness criterion for translation-invariant lattices (Theorem 2.3 of [4])
- ad hoc to paper The homogeneity condition on the Calderón-Zygmund operator in Definition 5.3
invented entities (2)
-
Bourgain-Morrey-Lorentz space M^{t,r}_{p,q}(R^n)
-
Block space H^{t',r'}_{p',q'}(R^n)
Cite this review
Pith. "Pith review of Bourgain-Morrey-Lorentz spaces and operators on them." pith.science (2026). https://pith.science/paper/B4PF6VPF
@misc{pith2026250519130,
author = {Pith},
title = {Pith review of: Bourgain-Morrey-Lorentz spaces and operators on them},
year = {2026},
howpublished = {\url{https://pith.science/paper/B4PF6VPF}},
note = {Machine review of arXiv:2505.19130}
}
abstract
We introduce Bourgain-Morrey-Lorentz spaces and give a description of the predual of Bourgain-Morrey-Lorentz spaces via the block spaces. As an application of duality, we obtain the boundedness of Hardy-Littlewood maximal operator, sharp maximal operator, Calder\'on-Zygmund operator, fractional integral operator, commutator on Bourgain-Morrey-Lorentz spaces. Moreover, we obtain a weak Hardy factorization terms of Calder\'on-Zygmund operator in Bourgain-Morrey-Lorentz spaces. Using this result, we obtain a characterization of functions in $\BMO$ (the functions of ``bounded mean oscillation'') via the boundedness of commutators generated by them and a homogeneous Calder\'on-Zygmund operator. In the last, we show that the commutator generated by a function $b$ and a homogeneous Calder\'on-Zygmund operator is a compact operator on Bourgain-Morrey-Lorentz spaces if and only if $b$ is the limit of compactly supported smooth functions in $\BMO$.
Reference graph
Works this paper leans on
- [15]
-
[10]
N. A. Dao, S. G, Krantz, On the predual of a Morrey-Lorentz space and its applications to the linear Calderón-Zygmund operators, Front. Math. (Beijing) 19 (3) (2024) 385–418
work page 2024
- [25]
-
[28]
A. Shalukhina, Self-improving boundedness of the maximal operator on quasi- banach lattices over spaces of homogeneous type, J. Math. Anal. Appl. 548 (2) (2025) 22
work page 2025
-
[1]
T. Bai, J. Xu, Weighted Bourgain-Morrey-Besov-Triebel-Lizorkin spaces asso- ciated with operators, Math. Nachr. 298 (3) (2025) 886–924
work page 2025
- [2]
-
[3]
Bourgain, On the restriction and multiplier problems inR3
J. Bourgain, On the restriction and multiplier problems inR3. In Geometric aspects of functional analysis. Proceedings of the Israel seminar (GAFA) 1989- 90, Berlin etc.: Springer-Verlag, (1991) 179–191
work page 1991
-
[4]
Brudnyi, Compactness criteria for spaces of measurable functions, St
Y. Brudnyi, Compactness criteria for spaces of measurable functions, St. Pe- tersbg. Math. J. 26 (1) (2015) 49–68. 61
work page 2015
Show all 34 references
-
[5]
M. J. Carro, H. Li, J. Soria, Q. Sun, Calderón-Zygmund operators and commu- tators on weighted Lorentz spaces, J. Geom. Anal. 31 (9) (2021) 8979–8990
2021
-
[6]
R. E. Castillo, H. C. Chaparro, Classical and multidimensional Lorentz spaces, De Gruyter, Berlin, 2021
2021
-
[7]
R. E. Castillo, H. Rafeiro, An introductory course in Lebesgue spaces, CMS Books Math./Ouvrages Math. SMC. Cham Springer, Cham, 2016
2016
-
[8]
P. G. Ciarlet, Linear and nonlinear functional analysis with applications, Society for Industrial and Applied Mathematics (SIAM)., Philadelphia, 2013
2013
-
[9]
F. Dai, L. Grafakos, Z. Pan, D. Yang, W. Yuan, Y. Zhang, The Bourgain-Brezis- Mironescu formula on ball Banach function spaces, Math. Ann. 388 (2) (2024) 1691–1768
2024
-
[11]
Di Fazio, M
G. Di Fazio, M. A. Ragusa, Commutators and Morrey spaces, Boll. Unione Mat. Ital., VII. Ser., A 5 (3) (1991) 323–332
1991
-
[12]
C. L. Fefferman, E. M. Stein,Hp spaces of several variables, Acta Math. 129 (1072) 137–193
-
[13]
Grafakos, Classical Fourier analysis, volume 249 of Grad
L. Grafakos, Classical Fourier analysis, volume 249 of Grad. Texts Math. Springer, New York, 2014
2014
-
[14]
Grafakos, Modern Fourier analysis, volume 250 of Grad
L. Grafakos, Modern Fourier analysis, volume 250 of Grad. Texts Math. Springer, New York, 2014
2014
-
[16]
P. Hu, Y. Li, D. Yang, Bourgain-Morrey spaces meet structure of Triebel- Lizorkin spaces, Math. Z. 304 (1) (2023) 49
2023
-
[17]
Komori, T
Y. Komori, T. Takahiro, Factorization of functions inH1(Rn)and generalized Morrey spaces, Math. Nachr. 279 (5-6) (2006) 619–624. 62
2006
-
[18]
A. K. Lerner, F. Nazarov, Intuitive dyadic calculus: the basics, Expo. Math. 37 (3) (2019) 225–265
2019
-
[19]
Lorist, Z
E. Lorist, Z. Nieraeth, Banach function spaces done right, Indag. Math., New Ser. 35 (2) (2024) 247–268
2024
-
[20]
Masaki, Two minimization problems on non-scattering solutions to mass- subcritical nonlinear Schrödinger equation, Preprint, arXiv:1605.09234, 2016
S. Masaki, Two minimization problems on non-scattering solutions to mass- subcritical nonlinear Schrödinger equation, Preprint, arXiv:1605.09234, 2016
2016 arXiv
-
[21]
Merle, L
F. Merle, L. Vega, Compactness at blow-up time forL2 solutions of the critical nonlinear Schrödinger equation in 2d, Int. Math. Res. Not. 1998 (8) (1998) 399– 425
1998
-
[22]
C. B. j. Morrey, Multiple integral problems in the calculus of variations and related topics. Univ. California Publ. Math., n. Ser. 1, (1943) 1-130
1943
-
[23]
Moyua, A
A. Moyua, A. Vargas, L. Vega, Restriction theorems and maximal operators related to oscillatory integrals inR3, Duke Math. J. 96 (3) (1999) 547–574
1999
-
[24]
M. A. Ragusa, Embeddings for Morrey-Lorentz spaces, J. Optim. Theory Appl. 154 (2) (2012) 491–499
2012
-
[26]
Sawano, S
Y. Sawano, S. R. El-Shabrawy, Weak Morrey spaces with applications, Math. Nachr. 291 (1) (2018) 178–186
2018
-
[27]
Sawano, H
Y. Sawano, H. Tanaka, The Fatou property of block spaces, J. Math. Sci., Tokyo 22 (3) (2015) 663–683
2015
-
[29]
Torchinsky, Real-variable methods in harmonic analysis, volume 123 of Pure Appl
A. Torchinsky, Real-variable methods in harmonic analysis, volume 123 of Pure Appl. Math., Academic Press, Academic Press, New York, 1986
1986
-
[30]
Uchiyama, On the compactness of operators of Hankel type
A. Uchiyama, On the compactness of operators of Hankel type. Tôhoku Math. J. 30 (2) (1978) 163–171. 63
1978
-
[31]
Zhang, D
Y. Zhang, D. Yang, Y. Zhao, Grand Besov-Bourgain-Morrey spaces and their applications to boundedness of operators, Anal. Math. Phys. 14 (4) (2024) 58
2024
-
[32]
Y. Zhao, Y. Sawano, J. Tao, D. Yang, W. Yuan, Bourgain-Morrey spaces mixed with structure of Besov spaces, Proc. Steklov Inst. Math. 323 (2023) 244–295
2023
-
[33]
C. Zhu, D. Yang, W. Yuan, Bourgain-Brezis-Mironescu-type characterization of inhomogeneous ball Banach Sobolev spaces on extension domains, J. Geom. Anal. 34 (10) (2024) 70
2024
-
[34]
C. Zhu, D. Yang, W. Yuan, Brezis-Seeger-Van Schaftingen-Yung-type character- ization of homogeneous ball Banach Sobolev spaces and its applications, Com- mun. Contemp. Math. 26 (8) (2024) 48
2024
Reviewed August 7, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.