pith. sign in

arxiv: 2604.10771 · v1 · submitted 2026-04-12 · 🧮 math.FA

Finite-codimensional subspaces of Daugavet spaces: projection constants and minimal projections

Tomasz Kania , Grzegorz Lewicki This is my paper

Pith reviewed 2026-05-10 15:12 UTC · model grok-4.3

classification 🧮 math.FA MSC 46B2046E15
keywords Daugavet propertyprojection constantsfinite-codimensional subspacesminimal projectionsC[0,1]L1 spaceBanach space dualityhyperplanes
0
0 comments X

The pith

Finite-codimensional subspaces Y of Daugavet spaces satisfy λ(Y,X)=1+λ(W,X*) with W the span of the defining functionals, yielding subspaces of C[0,1] with any projection constant Λ≥2 where the infimum is never attained.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper proves that in any Banach space X with the Daugavet property, if Y is the intersection of the kernels of n functionals whose span is W in the dual, then the projection constant of Y equals one plus the projection constant of W. This immediately implies that every hyperplane has projection constant exactly 2, and admits a minimal projection if and only if its defining functional attains its norm. The authors then apply a transfer map that sends certain finite-dimensional subspaces of ℓ1 to piecewise-constant subspaces of L1 inside the dual of C[0,1]; the map preserves projection constants while ensuring no weak*-continuous minimal projections exist on the image. Passing to annihilators produces the desired finite-codimensional subspaces of C[0,1].

Core claim

In a space X with the Daugavet property, for Y = ∩ ker f_j and W = span{f1,...,fn} ⊂ X*, one has λ(Y,X) = 1 + λ(W,X*), and minimal projections onto Y are in one-to-one correspondence with weak*-continuous minimal projections onto W. Specializing to the real space C[0,1] and using a transfer principle from duplication-stable subspaces of ℓ1^N to piecewise-constant subspaces of L1[0,1] ⊂ C[0,1]*, the authors obtain, for every Λ ≥ 2, a finite-codimensional subspace Y of C[0,1] such that λ(Y,C[0,1]) = Λ and the infimum is not attained; when the codimension is an even integer n, every value in (2,1+β_n] is realized, where β_n is the expected absolute value of the sum of n Rademacher variables.

What carries the argument

The duality identity λ(Y,X)=1+λ(W,X*) together with the transfer principle that maps duplication-stable subspaces of ℓ1^N to piecewise-constant subspaces of L1[0,1] while preserving projection constants and destroying weak*-continuous minimal projections.

If this is right

  • Every hyperplane in a Daugavet space has projection constant exactly 2.
  • A hyperplane in a Daugavet space admits a minimal projection if and only if its defining functional attains its norm.
  • For every real number Λ at least 2 there exists a finite-codimensional subspace Y of the real space C[0,1] with λ(Y,C[0,1])=Λ whose infimum is not attained.
  • When the codimension is an even integer n, every value in the interval (2,1+β_n] is realized as such a projection constant, where β_n ∼ √(2n/π).

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The set of attainable projection constants for finite-codimensional subspaces of C[0,1] is the entire half-line [2,∞).
  • Non-attainment of the infimum defining the projection constant may be generic for subspaces of higher codimension in spaces with the Daugavet property.
  • The same duality-plus-transfer technique could be tested on other classical spaces possessing the Daugavet property, such as L1 or the disc algebra.

Load-bearing premise

The transfer principle from duplication-stable finite-dimensional subspaces of ℓ1^N to piecewise-constant subspaces of L1[0,1] preserves projection constants and ensures that no weak*-continuous minimal projections exist on the transferred subspaces.

What would settle it

An explicit computation, for small even n, of a minimal projection onto one of the transferred piecewise-constant subspaces of L1[0,1] would show that the infimum is attained after all.

read the original abstract

Over the real or complex field, we establish a duality formula for projection constants of finite-codimensional subspaces of Banach spaces with the Daugavet property. If \[ Y=\bigcap_{j=1}^n \ker f_j \subset X, \qquad W=\operatorname{span}\{f_1,\dots,f_n\} \subset X^*, \] then \[ \lambda(Y,X)=1+\lambda(W,X^*), \] and minimal projections onto $Y$ correspond exactly to weak$^*$-continuous minimal projections onto $W$. This yields, in particular, a complete description of the hyperplane case: every hyperplane has projection constant $2$, and $\ker f$ admits a minimal projection if and only if $f$ attains its norm. We then specialise to the real space $X=C[0,1]$. Our second ingredient is a transfer principle from duplication-stable finite-dimensional subspaces of $\ell_1^N$ to piecewise-constant subspaces of $L_1[0,1]\subset M[0,1]=C[0,1]^*$. For the regular symmetric spaces constructed by Chalmers and the second-named author and the second named author and Prophet, respectively, the transferred subspaces retain their projection constants but admit no weak$^*$-continuous minimal projections. Passing to annihilators yields finite-codimensional subspaces of the real space $C[0,1]$ for which the infimum defining the projection constant is not attained. As a consequence, for every $\Lambda\in[2,\infty)$ there exists a finite-codimensional subspace $Y$ of the real space $C[0,1]$ such that \[ \lambda(Y,C[0,1])=\Lambda, \] and the infimum defining $\lambda(Y,C[0,1])$ is not attained. For each even codimension $n$ we moreover realise every value in the interval $(2,1+\beta_n]$, where \[ \beta_n = \mathsf E_{{\mathsf P}_n}\Bigl|\sum_{j=1}^n \varepsilon_j\Bigr| = n2^{-n}\binom{n}{n/2} \sim \sqrt{\frac{2n}{\pi}}, \] $(\varepsilon_j)$ is a Rademacher family on $\Omega_n=\{-1,1\}^n$, and $\mathsf{P}_n$ is the uniform probability measure.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The paper establishes a duality formula λ(Y,X)=1+λ(W,X*) for finite-codimensional subspaces Y=∩ker f_j in Daugavet spaces X, with W=span{f_j}⊂X*, and shows that minimal projections onto Y correspond exactly to weak*-continuous minimal projections onto W. Specializing to real C[0,1], a transfer principle maps duplication-stable finite-dimensional subspaces of ℓ₁^N to piecewise-constant subspaces of L₁[0,1]⊂M[0,1] that retain the projection constant but admit no weak*-continuous minimal projections; passing to annihilators yields finite-codimensional Y⊂C[0,1] with prescribed λ(Y,C[0,1])=Λ for any Λ≥2 where the infimum is not attained. For even codimension n the paper realizes every value in (2,1+β_n] with β_n=E_{P_n}|∑ε_j|∼√(2n/π).

Significance. If the duality and transfer hold, the results give a complete description of hyperplane projection constants (always 2, attained iff the functional attains its norm) and show that every value ≥2 is attained as a projection constant of some finite-codimensional subspace of C[0,1], with explicit non-attainment of the infimum. This advances the theory of projection constants in classical spaces by combining the Daugavet property with known constructions of symmetric spaces.

major comments (2)
  1. [Transfer principle] The transfer principle (detailed after the duality section) is load-bearing for the existence claim: it must map duplication-stable Z⊂ℓ₁^N to W⊂L₁[0,1] such that λ(W,M[0,1]) exactly equals λ(Z,ℓ₁^N) while excluding all weak*-continuous minimal projections from M[0,1]. The manuscript sketch does not explicitly rule out the possibility that singular measures in M[0,1] produce a strictly smaller projection constant or a weak*-continuous attaining projection, which would invalidate the correspondence to non-attained λ(Y,C[0,1]).
  2. [Duality formula] In the duality formula, the exact correspondence between minimal projections onto Y and weak*-continuous minimal projections onto W is asserted but the argument that no non-weak*-continuous projection onto W can achieve the bound λ(W,X*) must be verified in detail; any gap here propagates directly to the non-attainment statements for C[0,1].
minor comments (2)
  1. [Consequence for even codimensions] The binomial expression for β_n is given but the Rademacher average could be cross-referenced to a standard lemma for immediate verification.
  2. [Transfer principle] Notation for the spaces (M[0,1] as dual of C[0,1]) is standard but a brief reminder of the weak* topology induced by C[0,1] would aid readability in the transfer section.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the careful reading and for identifying areas where the exposition of the transfer principle and duality formula requires additional detail. We will revise the manuscript accordingly to strengthen these sections while preserving the core results.

read point-by-point responses

  1. Referee: [Transfer principle] The transfer principle (detailed after the duality section) is load-bearing for the existence claim: it must map duplication-stable Z⊂ℓ₁^N to W⊂L₁[0,1] such that λ(W,M[0,1]) exactly equals λ(Z,ℓ₁^N) while excluding all weak*-continuous minimal projections from M[0,1]. The manuscript sketch does not explicitly rule out the possibility that singular measures in M[0,1] produce a strictly smaller projection constant or a weak*-continuous attaining projection, which would invalidate the correspondence to non-attained λ(Y,C[0,1]).

    Authors: We agree that the sketch of the transfer principle needs to be expanded for full rigor. In the revised version we will add explicit arguments showing that the projection constant is preserved: any projection onto the transferred piecewise-constant subspace W in M[0,1] restricts, via averaging over the duplication structure, to a projection onto the original Z in ℓ₁^N with norm at most as large, forcing equality of the infima. We will also verify directly that no weak*-continuous minimal projection onto W exists, by showing that any such projection would induce a minimal projection in the discrete symmetric-space setting, contradicting the known non-attainment properties of those spaces. Singular measures cannot produce a strictly smaller constant because the total-variation norm on M[0,1] and the absolute continuity of the target subspace together imply that any singular component strictly increases the operator norm; this will be stated as a short lemma. revision: yes

  2. Referee: [Duality formula] In the duality formula, the exact correspondence between minimal projections onto Y and weak*-continuous minimal projections onto W is asserted but the argument that no non-weak*-continuous projection onto W can achieve the bound λ(W,X*) must be verified in detail; any gap here propagates directly to the non-attainment statements for C[0,1].

    Authors: We thank the referee for this observation. The proof of the duality formula already establishes both the equality λ(Y,X)=1+λ(W,X*) and the bijection between minimal projections onto Y and weak*-continuous minimal projections onto W. To close the potential gap, the revised manuscript will include an additional paragraph showing that, in a Daugavet space, if a non-weak*-continuous projection onto W attained λ(W,X*), then the Daugavet property would allow construction of a weak*-continuous projection of the same norm (via weak*-approximation and the norm-attaining properties used in the original argument), contradicting the assumption that only non-weak* projections attain. This verification will be placed immediately after the statement of the correspondence, ensuring the non-attainment claims for the subspaces of C[0,1] follow directly. revision: yes

Circularity Check

0 steps flagged

No significant circularity; derivations are self-contained via duality and explicit transfer

full rationale

The paper derives the duality formula λ(Y,X)=1+λ(W,X*) directly from the Daugavet property and standard annihilator arguments without any self-referential definitions or fitted inputs. The transfer principle is an explicit construction mapping duplication-stable subspaces of ℓ₁^N to piecewise-constant subspaces of L₁[0,1], preserving projection constants by direct verification rather than by renaming or self-citation load-bearing. Prior works by the authors supply example spaces but are not invoked as uniqueness theorems or ansatzes that force the current results; the non-attainment of minimal projections follows from the weak*-topology distinction between M[0,1] and C[0,1]*, which is independently established. The existence of Y with λ(Y,C[0,1])=Λ for arbitrary Λ≥2 is therefore a genuine construction, not a reduction to inputs.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The paper relies on the Daugavet property as a domain assumption and on a transfer principle for symmetric spaces; no free parameters or new entities are introduced.

axioms (2)
  • domain assumption X has the Daugavet property.
    Invoked to establish the duality formula for finite-codimensional subspaces.
  • domain assumption The transfer principle from duplication-stable subspaces of ℓ1^N to piecewise-constant subspaces of L1[0,1] preserves projection constants.
    Used to obtain the C[0,1] examples with prescribed constants and non-attainment.

pith-pipeline@v0.9.0 · 5758 in / 1380 out tokens · 33630 ms · 2026-05-10T15:12:33.651485+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Transfinite Daugavet property

    math.FA 2026-04 unverdicted novelty 8.0

    Transfinite Daugavet properties are defined and characterized for C(K) spaces by a cardinal index r(K) generalizing the reaping number, with the perfect version equivalent to K having no Gδ-points.

  2. Transfinite Daugavet property

    math.FA 2026-04 unverdicted novelty 6.0

    The authors extend the Daugavet property to transfinite cardinals, characterize the transfinite Daugavet C(K) spaces via a new cardinal index r(K) generalizing the reaping number, and prove inheritance results for ide...

Reference graph

Works this paper leans on

28 extracted references · 28 canonical work pages · cited by 1 Pith paper

  1. [1]

    Albiac and N

    F. Albiac and N. J. Kalton,Topics in Banach Space Theory, 2nd ed., Graduate Texts in Mathematics, vol. 233, Springer, Cham, 2016

    work page 2016

  2. [2]

    B. L. Chalmers and G. Lewicki,Minimal projections onto some subspaces ofℓ(n) 1 , Funct. Approx. Comment. Math.26(1998), 85–92

    work page 1998

  3. [3]

    B. L. Chalmers and G. Lewicki,Symmetric subspaces ofℓ1 with large projection constants, Studia Math. 134(1999), no. 2, 119–133. FINITE-CODIMENSIONAL SUBSPACES OF DAUGA VET SPACES 15

    work page 1999

  4. [4]

    B. L. Chalmers and G. Lewicki,Two-dimensional real symmetric spaces with maximal projection constant, Ann. Polon. Math.73(2000), no. 2, 119–134

    work page 2000

  5. [5]

    B. L. Chalmers and G. Lewicki,Symmetric spaces with maximal projection constants, J. Funct. Anal.200 (2003), no. 1, 1–22

    work page 2003

  6. [6]

    B. L. Chalmers and F. T. Metcalf,The determination of minimal projections and extensions inL1, Trans. Amer. Math. Soc.329(1992), 289–305

    work page 1992

  7. [7]

    B. L. Chalmers and F. T. Metcalf,A characterization and equations for minimal projections and extensions, J. Operator Theory32(1994), 31–46

    work page 1994

  8. [8]

    E. W. Cheney and C. Franchetti,Minimal projections inL1-space, Duke Math. J.43(1976), 501–510

    work page 1976

  9. [9]

    E. W. Cheney, C. R. Hobby, P. D. Morris, F. Schurer, and D. E. Wulbert,On the minimal property of the Fourier projection, Trans. Amer. Math. Soc.143(1969), 249–258

    work page 1969

  10. [10]

    E. W. Cheney and P. D. Morris,On the existence and characterization of minimal projections, J. Reine Angew. Math.270(1974), 61–76

    work page 1974

  11. [11]

    Deręgowska, M

    B. Deręgowska, M. Fickus, S. Foucart, and B. Lewandowska,On the value of the fifth maximal projection constant, J. Funct. Anal.283(2022), no. 10, 109634

    work page 2022

  12. [12]

    Deręgowska and B

    B. Deręgowska and B. Lewandowska,A simple proof of the Grünbaum conjecture, J. Funct. Anal.285 (2023), no. 2, 109950

    work page 2023

  13. [13]

    Feller,An Introduction to Probability Theory and Its Applications

    W. Feller,An Introduction to Probability Theory and Its Applications. Vol. I, 3rd ed., Wiley, New York, 1968

    work page 1968

  14. [14]

    S. D. Fisher, P. D. Morris, and D. E. Wulbert,Unique minimality of Fourier projections, Trans. Amer. Math. Soc.265(1981), 235–246

    work page 1981

  15. [15]

    Foucart and L

    S. Foucart and L. Skrzypek,On maximal relative projection constants, J. Math. Anal. Appl.447(2017), no. 1, 309–328

    work page 2017

  16. [16]

    Grünbaum,Projection constants, Trans

    B. Grünbaum,Projection constants, Trans. Amer. Math. Soc.95(1960), 451–465

    work page 1960

  17. [17]

    V. M. Kadets, R. V. Shvidkoy, G. G. Sirotkin, and D. Werner,Banach spaces with the Daugavet property, Trans. Amer. Math. Soc.352(2000), no. 2, 855–873

    work page 2000

  18. [18]

    Kania and G

    T. Kania and G. Lewicki,(λ+)-injective Banach spaces, arXiv:2603.09710 [math.FA], 2026

    work page arXiv 2026

  19. [19]

    König,Spaces with large projection constants, Israel J

    H. König,Spaces with large projection constants, Israel J. Math.50(1985), 181–188

    work page 1985

  20. [20]

    König, D

    H. König, D. R. Lewis, and P.-K. Lin,Finite dimensional projection constants, Studia Math.75(1983), no. 3, 341–358

    work page 1983

  21. [21]

    König, C

    H. König, C. Schütt, and N. Tomczak-Jaegermann,Projection constants of symmetric spaces and variants of Khinchine’s inequality, J. Reine Angew. Math.511(1999), 1–22

    work page 1999

  22. [22]

    König and N

    H. König and N. Tomczak-Jaegermann,Bounds for projection constants and1-summing norms, Trans. Amer. Math. Soc.320(1990), 799–823

    work page 1990

  23. [23]

    Lewicki,Minimal extensions in tensor product spaces, J

    G. Lewicki,Minimal extensions in tensor product spaces, J. Approx. Theory97(1999), 366–383

    work page 1999

  24. [24]

    Lewicki and M

    G. Lewicki and M. Mastyło,Asymptotic behavior of factorization and projection constants, J. Geom. Anal. 27(2017), no. 1, 335–365

    work page 2017

  25. [25]

    Lewicki and M

    G. Lewicki and M. Prophet,The Chalmers–Metcalf operator and minimal extensions, J. Funct. Anal.280 (2021), no. 1, 108800

    work page 2021

  26. [26]

    Lewicki and L

    G. Lewicki and L. Skrzypek,Chalmers–Metcalf operator and uniqueness of minimal projections, J. Approx. Theory148(2007), 71–91

    work page 2007

  27. [27]

    Lewicki and L

    G. Lewicki and L. Skrzypek,Minimal projections onto hyperplanes inℓn p, J. Approx. Theory202(2016), 42–63

    work page 2016

  28. [28]

    Shekhtman and L

    B. Shekhtman and L. Skrzypek,Uniqueness of minimal projections onto two-dimensional subspaces, Studia Math.168(2005), 273–284. (T. Kania)Mathematical Institute, Czech Academy of Sciences, Žitná 25, 115 67 Praha 1, Czech Republic and Institute of Mathematics and Computer Science, Jagiellonian University, Łojasiewicza 6, 30-348 Kraków, Poland Email address:...

    work page 2005