The $\pi$-property of a Banach space along a filter

Jaros{\l}aw Swaczyna; Tomasz Kania

REVIEW 2 major objections 4 minor 16 references

For analytic filters F, the class of separable Banach spaces with the F-π-property is a Σ13 set; countably generated filters lower the complexity to Σ12.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

The class of separable Banach spaces with the π-property along an analytic filter is claimed to be Σ^1_3, and Σ^1_2 for countably generated filters, but the proof as written has a critical error.

T0 review reviewed 2026-08-05 challenge →

load-bearing objection The F-π-property and the reduction to analytic filters are interesting, but a quantifier error in the proof of Theorem A empties X3 for every proper subspace, so the main complexity bound is not established. the 2 major comments →

arxiv 2508.21242 v1 pith:4KFCFLJ4 submitted 2025-08-28 math.FA math.LO

The π-property of a Banach space along a filter

Tomasz Kania , Jaros{\l}aw Swaczyna This is my paper

classification math.FA math.LO MSC 46B2046B1503E15

keywords Filter convergenceπ-propertyBanach spacesprojective hierarchyanalytic filtersPolish spaces of separable Banach spacesfinite-dimensional decompositionEffros–Borel structure

verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

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 generalises the classical π-property of separable Banach spaces—having a sequence of finite-rank projections converge pointwise to the identity—by replacing ordinary convergence with convergence along a filter F on N. Its main theorem computes the descriptive complexity of the resulting class: for any analytic filter F, the collection SB_{F-π} of separable spaces with the F-π-property is a Σ13 set (and Σ1_{s+2} if F is Σ1_s); for countably generated filters it is Σ12. The proof works by encoding the property as the existence of growth constants and finite-support matrices that satisfy three norm inequalities, then reading off the projective level of the set of all such encodings. The paper also shows that no space needs more than an analytic filter to certify its F-π-property, and that the union over all filters is Σ13. The interest is that filter convergence raises the descriptive complexity relative to the classical π-property, which bears on the open question whether such spaces necessarily admit finite-dimensional decompositions.

Core claim

Working in the standard Borel space SB of closed subspaces of C(Δ), the paper introduces the F-π-property: a separable Banach space X has it if there is a sequence of finite-rank projections P_n on X such that P_n x converges to x along F for every x. The central result is that for analytic F, SB_{F-π} is Σ13 in SB (with the bound shifting to Σ1_{s+2} when F is itself Σ1_s), and for countably generated F it is Σ12. The proof hinges on Lemma 3.1, which characterises the property in terms of a sequence of constants λ_n and matrices σ_n recording how a dense linearly independent sequence is mapped; two of the conditions state that the induced finite-rank operators are uniformly bounded and almo

What carries the argument

The carrying mechanism is a three-part coding of the F-π-property. A refinement of a known approximation lemma continuously converts an almost-idempotent finite-rank operator into a genuine finite-rank projection, with a scalar Θ_n(λ) controlling the allowed idempotency error. Lemma 3.1 then replaces an F-π-basis by the data (λ_n, σ_n) used to define three norm inequalities: boundedness, almost-idempotence, and pointwise approximation along the filter. The first two are Borel conditions; the third is where the filter enters, through an existential quantifier over sets A in F. The projective class of the set X3 built from this third condition determines the theorem's bound.

Load-bearing premise

The proof assumes that a vector in the ambient space that satisfies the approximation inequality for some r must already lie in the subspace X; if an ambient vector outside X can pass the inequality, the set X3 that encodes condition (6) becomes empty for proper subspaces and the complexity computation collapses.

What would settle it

Examine the definition of W_r in Section 3 on a proper closed subspace X of C(Δ). Take an ambient unit vector x not in X and a dense linearly independent sequence (x_i) in X; if the inequality ∥x - f_i(X)∥ ≤ 1/(r·12λ_n) holds for some i and the operator matrix almost fixes f_i(X), then x satisfies the defining conditions of W_r despite lying outside X. In that case the sentence 'if (X,(λ_n),σ,x,A) ∈ W_r then x∈X' is false, and X3 cannot equal the set of spaces satisfying condition (6).

Watch this falsifier. Get emailed when new claim-graph text bears on it.

If this is right

For every analytic filter F, SB_{F-π} is a third-level projective set, so the F-π-property is generally not Borel; filter convergence is a genuine source of descriptive complexity.
For countably generated filters, including the Fréchet filter that gives the ordinary π-property, the bound improves to Σ12; this is coarser than the known bound for the ordinary case, and the paper suggests the jump is unavoidable.
Any space with the F-π-property has an analytic sub-filter that works with the same projections, so non-analytic filters add no new spaces to any single class.
The union of all classes SB_{F-π} over all filters is itself Σ13 in SB.
The complexity gap between the ordinary π-property and its filter versions points toward a negative answer to the paper's Question 1.2, that the F-π-property might not imply existence of a finite-dimensional decomposition.

Where Pith is reading between the lines

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

The λ_n-σ_n coding is likely transferable to filter versions of other approximation properties, such as the bounded approximation property or the metric π-property, with similar projective bounds determined by the filter's complexity class.
The collapse from Σ13 to Σ12 for countably generated filters identifies the existential quantifier over A∈F as the source of extra complexity; proving optimality would amount to finding a filter for which SB_{F-π} is Σ13-complete, which the paper leaves open.
A direct test of the paper's suggested answer to Question 1.2 would be to compute the descriptive complexity of the class of separable spaces with a finite-dimensional decomposition; if that class is Borel or low in the projective hierarchy, the gap would strongly support the claim that the F-π-property does not imply an FDD.
The coding scheme depends on having a dense linearly independent sequence; the paper explicitly leaves open whether such sequences can be chosen continuously, and a positive answer would make the complexity bounds uniform across all admissible topologies.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Desk Editor's Note

The F-π-property and the reduction to analytic filters are interesting, but a quantifier error in the proof of Theorem A empties X3 for every proper subspace, so the main complexity bound is not established.

read the letter

Quick take on arXiv:2508.21242. The core idea is good: define F-π-property, prove a reduction to analytic filters (Theorem B), and attempt projective complexity bounds. Theorem B is a neat observation, and Lemma 3.1 gives a useful finitary characterization. The paper also sensibly builds on Ghawadrah's work.

The soft spot is in the proof of Theorem A, around the definition of X3. The set W_r is defined inside X × B_{C(Δ)} × {0,1}^N, and the note claiming that membership in W_r forces x∈X is false for a fixed r. An x outside X can be close enough to some f_i(X) to satisfy the first inequality. Taking the intersection over r does force x∈X for any (X,x) that survives in V, but that means X3 requires B_{C(Δ)}⊆X, i.e., X=C(Δ). So X3 is empty for every proper subspace, and the equality SB_{F-π}=Π_{SB}[X1∩X2∩X3] cannot hold. The same construction appears in Theorem C, so that proof inherits the defect.

The fix is likely straightforward: quantify over B_X instead of B_{C(Δ)} in the definition of W_r (or otherwise restrict x to the subspace). The resulting Borel/projective complexity should be the same, but as written the central claims are unsupported.

Credit where due: the definitions are natural, Theorem B is a real contribution, and the authors are clearly drawing on a solid literature. The self-criticism of Ghawadrah's proof in Section 5 is a bit of a side comment, but not damaging.

This paper deserves a serious referee — the mistake is subtle and the topic is worth engaging — but I would not accept it in its present form. It needs a major revision to correct the quantifier domain. If fixed, Theorems A and C would be interesting results.

For the reading group: yes, it's a good example of a promising idea derailed by a small domain error. I wouldn't cite the main theorems until the fix is in.

Referee Report

2 major / 4 minor

Summary. The paper introduces a filter-dependent version of the classical π-property for separable Banach spaces and studies the descriptive complexity of the resulting class SB_{F-π} inside the standard space SB of closed subspaces of C(Δ). The main results are Theorem A (SB_{F-π} is Σ^1_3 for analytic F, and Σ^1_2 when F is countably generated), Theorem B (any F-π-basis is also an F'-π-basis for some analytic subfilter F'), and Theorem C (the union over all filters is Σ^1_3). The proofs combine a characterization of the F-π-property by approximate finite-rank operators (Lemma 3.1), the perturbation lemmas of Section 2, and a projective coding of the witnessing data in Section 3.

Significance. If the main theorems were valid, they would give a uniform descriptive classification of a natural class of Banach spaces that generalizes Ghawadrah's Borel-complexity result for the usual π-property. The paper also contains useful self-contained perturbation lemmas and a clean analytic-subfilter reduction. However, the central coding of the unit-ball quantifier in Lemma 3.1 is defective. The set X3 that is supposed to encode condition (6) is defined using the ambient unit ball B_{C(Δ)} rather than the unit ball of the varying subspace X. As written, X3 excludes every proper subspace, so the equality SB_{F-π}=Π_SB[X1∩X2∩X3] is false and Theorem A's proof collapses. The same defect invalidates Theorem C. The results may be repairable, but the present manuscript does not establish them.

major comments (2)

[§3, definition of X3] The equality SB_{F-π}=Π_SB[X1∩X2∩X3] is false as written. The set V is defined by projecting V_r to X×B_C(Δ), so V concerns ambient vectors x∈B_C(Δ). The note 'if (X,λ,σ,x,A)∈W_r, then x∈X' is false for a single r: if x∉X has dist(x,X)<1/(r·12λ_n), the inequality ∥x−f_i(X)∥≤1/(r·12λ_n) can hold for some i. The full intersection V does force x∈X, since for large r the radius is smaller than dist(x,X). But that is precisely the defect: for every proper closed subspace X, any y∈B_C(Δ)\X belongs to (X×B_C(Δ))\V, so the projection Π_X[(X×B_C(Δ))\V] contains (X,λ,σ). Hence X3 is empty for every proper X. Lemma 3.1 condition (6) quantifies only over x∈B_X, and many proper X satisfy it. Thus the central encoding of Theorem A is invalid.
[§4, definition of Y3 in Theorem C] The same ambient-unit-ball defect appears in the proof of Theorem C. The set U is obtained by projecting U_r to Y×B_C(Δ), and Y3 = Y\Π_Y[(Y×B_C(Δ))\U]. Since U’s projection also forces the ambient vector x to lie in X, every proper subspace X yields points (Y,λ,σ,x) with x∉X in the projected complement, so Y3 excludes all proper X. Consequently the claimed equality SB_{Filter-π}=Π_SB[Y1∩Y2∩Y3∩(X×AF)] is unsupported.

minor comments (4)

[§2, proof of Corollary 2.2] The sentence 'It is clear that the vectors f_n(X) remain linearly independent whenever dim X = c' should presumably read 'whenever dim X = ∞' or 'whenever X is infinite-dimensional'; c is ambiguous here.
[§3, notation around V_r] In the definition of V_r, the notation X×B_C(Δ)×F mixes the space X of triples (X,λ,σ) with the first-coordinate space X. Consider writing the product explicitly as (SB×N^N×S)×B_C(Δ)×F to avoid confusion.
[§1 and §4, set inclusion notation] The paper uses A⊂F and F′⊂F where strict containment is not intended. Use ⊆ for readability.
[§3, displayed sentence after W_r] The sentence 'Note that if (X,λ,σ,x,A)∈W_r, then x∈X' is not only unproved but false; it should be removed or replaced by a correct statement about the full intersection V.

Circularity Check

0 steps flagged

No significant circularity: the projective bounds follow from a self-contained equivalence (Lemma 3.1) and standard coding; prior self-citations are not load-bearing.

full rationale

The paper's central claim (Theorem A) reduces the F-π-property to the concrete analytic condition (4)-(6) in Lemma 3.1. That lemma is proved directly: the forward direction uses Corollary 2.6 and the reverse uses Corollary 2.8 (whose proof is given in full via Lemma 2.7). The encoding into X1, X2, X3 mirrors this equivalence, so the descriptive-set-theoretic conclusion is derived from the definition rather than assumed. Citations to the authors' prior work [7], [14] appear only in the introduction as background on filter bases and do not carry the proof of Theorem A, B, or C; Theorem C is proved by coding analytic filters directly, and Theorem B supplies its own construction of F′. Lemma 2.7 is introduced as a refinement of [4, Prop. 3.6] but the paper supplies the full argument, so it is independent evidence rather than a self-citation. No fitted parameters are renamed as predictions and no uniqueness/forced-choice claim rests on the authors' earlier results. The only in-scope anomaly worth flagging is a local mathematical assertion in §3—the note 'if (X,λ,σ,x,A)∈W_r, then x∈X' is not true for a fixed r unless the radius forces x into X; the later intersection over r is what forces x∈X. This concerns correctness of the coding, not circularity of the derivation. Hence the derivation chain is self-contained and the circularity score is 0.

Axiom & Free-Parameter Ledger

0 free parameters · 5 axioms · 0 invented entities

The paper introduces the F-π-property as a definition, not as an invented entity with independent physical or mathematical content. There are no fitted parameters, no new constants, and no new objects beyond the standard spaces and operators. All axioms are standard descriptive set theory and Banach space facts, plus the domain restriction that filters contain the Fréchet filter.

axioms (5)

standard math Standard descriptive set-theoretic framework for the space SB of closed subspaces of C(Δ) with the Effros-Borel σ-algebra and admissible Polish topologies.
Invoked throughout Section 2 and in the proofs of Theorems A and C to make sense of projective complexity in SB.
standard math Existence of continuous maps g_i : SB → C(Δ) with X = {g_i(X)} for every X∈SB (from [11, Theorem 4.1]).
Used in Corollary 2.2 to construct the Borel sequence f_i(X) of dense, linearly independent vectors.
standard math The set of finite-dimensional spaces in SB is Borel (from [11, Corollary 4.2]).
Needed in Corollary 2.2 to handle finite-dimensional and infinite-dimensional spaces separately.
domain assumption All filters considered contain the Fréchet filter.
Stated in the introduction (page 2); ensures every element of the filter is infinite and that filter convergence is a genuine generalization of usual convergence.
domain assumption The underlying filter F is an analytic (or Σ^1_s) subset of the Cantor set.
Used in Theorem A to estimate the complexity of the condition A∈F in the sets Vr and Ur.

reviewed 2026-08-05 · how reviews work

0 comments

Cite this review

Pith. "Pith review of The $\pi$-property of a Banach space along a filter." pith.science (2026). https://pith.science/paper/4KFCFLJ4

@misc{pith2026250821242,
  author       = {Pith},
  title        = {Pith review of: The $\pi$-property of a Banach space along a filter},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/4KFCFLJ4}},
  note         = {Machine review of arXiv:2508.21242}
}

Share X Bluesky LinkedIn Reddit HN

read the original abstract

We examine the analyticity of the class of separable Banach spaces possessing the $\pi$-property, defined in terms of convergence along a filter. Our results establish that this class is $\Sigma^1_3$ whenever the underlying filter is analytic (as a subset of the Cantor set $\Delta$). Furthermore, we demonstrate that if the filter is countably generated, the class of such spaces is $\Sigma^1_2$ with respect to any admissible Polish topology on the family of closed subspaces of $C(\Delta)$.

discussion (0)

Reference graph

Works this paper leans on

16 extracted references · 16 canonical work pages

[1]

Albiac, N

F. Albiac, N. J. Kalton,Topics in Banach Space Theory . Graduate Texts in Mathematics, 233. New York: Springer-Verlag, 2006

work page 2006
[2]

Avilés, V

A. Avilés, V. Kadets, A. Pérez, S. Solecki, Baire theorem for ideals of sets,J. Math. Anal. Appl. 445 (2) (2017) 1221–1231

work page 2017
[3]

P. G. Casazza. Approximation properties. In: Handbook of the geometry of Banach spaces, Vol. I. North-Holland, Amsterdam, 2001, pp. 271–316

work page 2001
[4]

P. G. Casazza, N. J. Kalton, Notes on approximation properties in separable Banach spaces.Lect. Notes Lond. Math. Soc. 158 (1991), 49–65

work page 1991
[5]

M. Cúth, M. Doležal, M. Doucha, and O. Kurka, Polish spaces of Banach spaces,Forum of Mathemat- ics, Sigma 10 (2022), e26

work page 2022
[6]

M. Cúth, M. Doležal, M. Doucha, and O. Kurka, Polish spaces of Banach spaces. Complexity of isometry and isomorphism classes,J. Inst. Math. Jussieu 23(4) (2024), 1919–1957

work page 2024
[7]

de Rancourt, T

N. de Rancourt, T. Kania and J. Swaczyna, Continuity of coordinate functionals of filter bases in Banach spaces. J. Funct. Anal.. 284 (2023), no. 9, 109869

work page 2023
[8]

Ghawadrah, The descriptive complexity of approximation properties in an admissible topology

G. Ghawadrah, The descriptive complexity of approximation properties in an admissible topology. Fund. Math. 249 (2020), no. 3, 303–309

work page 2020
[9]

Ghawadrah, The descriptive complexity of the family of Banach spaces with the bounded approxi- mation property.Houston J

G. Ghawadrah, The descriptive complexity of the family of Banach spaces with the bounded approxi- mation property.Houston J. Math. 43 (2017), no. 2, 395–401

work page 2017
[10]

Ghawadrah, The descriptive complexity of the family of Banach spaces with theπ-property

G. Ghawadrah, The descriptive complexity of the family of Banach spaces with theπ-property. Arab. J. Math. 4 (2015), no. 1, 35–39

work page 2015
[11]

Godefroy and J

G. Godefroy and J. Saint-Raymond, Descriptive complexity of some isomorphism classes of Banach spaces, J. Funct. Anal., 275 (2018), 1008–1022

work page 2018
[12]

Hájek and R

P. Hájek and R. Medina, Compact retractions and Schauder decompositions in Banach spaces,Trans. Amer. Math. Soc. 376 (2023), 1343–1372

work page 2023
[13]

Johnson, H.P

W.B. Johnson, H.P. Rosenthal, and M. Zippin, On bases, finite dimensional decompositions and weaker structures in Banach spaces.Israel J. Math. 9 (1971), 488–506

work page 1971
[14]

Kania, J

T. Kania, J. Swaczyna, Large cardinals and continuity of coordinate functionals of filter bases in Banach spaces. Bull. Lond. Math. Soc . 53 (2021), no. 1, 231–239

work page 2021
[15]

A. S. Kechris,Classical descriptive set theory , vol. 156 of Graduate Texts in Mathematics, Springer- Verlag, New York, 1995

work page 1995
[16]

Kochanek,F-bases with brackets and with individual brackets in Banach spaces,Studia Math

T. Kochanek,F-bases with brackets and with individual brackets in Banach spaces,Studia Math. 211 (2012), 259–268. (T. Kania) Mathematical Institute, Czech Academy of Sciences, Žitná 25, 115 67 Praha 1, Czech Republic and Institute of Mathematics and Computer Science, Jagiellonian Uni- versity, Łojasiewicza 6, 30-348 Kraków, Poland Email address: kania@mat...

work page 2012

This paper was first reviewed by deepseek-v4-flash on August 5, 2026.

[1] [1]

Albiac, N

F. Albiac, N. J. Kalton,Topics in Banach Space Theory . Graduate Texts in Mathematics, 233. New York: Springer-Verlag, 2006

work page 2006

[2] [2]

Avilés, V

A. Avilés, V. Kadets, A. Pérez, S. Solecki, Baire theorem for ideals of sets,J. Math. Anal. Appl. 445 (2) (2017) 1221–1231

work page 2017

[3] [3]

P. G. Casazza. Approximation properties. In: Handbook of the geometry of Banach spaces, Vol. I. North-Holland, Amsterdam, 2001, pp. 271–316

work page 2001

[4] [4]

P. G. Casazza, N. J. Kalton, Notes on approximation properties in separable Banach spaces.Lect. Notes Lond. Math. Soc. 158 (1991), 49–65

work page 1991

[5] [5]

M. Cúth, M. Doležal, M. Doucha, and O. Kurka, Polish spaces of Banach spaces,Forum of Mathemat- ics, Sigma 10 (2022), e26

work page 2022

[6] [6]

M. Cúth, M. Doležal, M. Doucha, and O. Kurka, Polish spaces of Banach spaces. Complexity of isometry and isomorphism classes,J. Inst. Math. Jussieu 23(4) (2024), 1919–1957

work page 2024

[7] [7]

de Rancourt, T

N. de Rancourt, T. Kania and J. Swaczyna, Continuity of coordinate functionals of filter bases in Banach spaces. J. Funct. Anal.. 284 (2023), no. 9, 109869

work page 2023

[8] [8]

Ghawadrah, The descriptive complexity of approximation properties in an admissible topology

G. Ghawadrah, The descriptive complexity of approximation properties in an admissible topology. Fund. Math. 249 (2020), no. 3, 303–309

work page 2020

[9] [9]

Ghawadrah, The descriptive complexity of the family of Banach spaces with the bounded approxi- mation property.Houston J

G. Ghawadrah, The descriptive complexity of the family of Banach spaces with the bounded approxi- mation property.Houston J. Math. 43 (2017), no. 2, 395–401

work page 2017

[10] [10]

Ghawadrah, The descriptive complexity of the family of Banach spaces with theπ-property

G. Ghawadrah, The descriptive complexity of the family of Banach spaces with theπ-property. Arab. J. Math. 4 (2015), no. 1, 35–39

work page 2015

[11] [11]

Godefroy and J

G. Godefroy and J. Saint-Raymond, Descriptive complexity of some isomorphism classes of Banach spaces, J. Funct. Anal., 275 (2018), 1008–1022

work page 2018

[12] [12]

Hájek and R

P. Hájek and R. Medina, Compact retractions and Schauder decompositions in Banach spaces,Trans. Amer. Math. Soc. 376 (2023), 1343–1372

work page 2023

[13] [13]

Johnson, H.P

W.B. Johnson, H.P. Rosenthal, and M. Zippin, On bases, finite dimensional decompositions and weaker structures in Banach spaces.Israel J. Math. 9 (1971), 488–506

work page 1971

[14] [14]

Kania, J

T. Kania, J. Swaczyna, Large cardinals and continuity of coordinate functionals of filter bases in Banach spaces. Bull. Lond. Math. Soc . 53 (2021), no. 1, 231–239

work page 2021

[15] [15]

A. S. Kechris,Classical descriptive set theory , vol. 156 of Graduate Texts in Mathematics, Springer- Verlag, New York, 1995

work page 1995

[16] [16]

Kochanek,F-bases with brackets and with individual brackets in Banach spaces,Studia Math

T. Kochanek,F-bases with brackets and with individual brackets in Banach spaces,Studia Math. 211 (2012), 259–268. (T. Kania) Mathematical Institute, Czech Academy of Sciences, Žitná 25, 115 67 Praha 1, Czech Republic and Institute of Mathematics and Computer Science, Jagiellonian Uni- versity, Łojasiewicza 6, 30-348 Kraków, Poland Email address: kania@mat...

work page 2012