REVIEW 2 major objections 6 minor 3 cited by
A metric lifting theorem
T0 review · 2 major / 6 minor · reviewed 2026-08-12 · deepseek-v4-flash
Pith's one-line read The paper proves that under OCA and $\mathrm{MA}_{\aleph_1}(\sigma\text{-linked})$, every coordinate respecting function between reduced products of separable metric spaces with uniformly bounded diameter is trivial, extending the…
desk verdict A well-executed metric generalization of the discrete lifting theorem; the main proof is plausible, but the load-bearing Lemma A.2 is only sketched and should be fully proved before the paper is accepted. 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 engine is the ideal $J_{\mathrm{prod}}=\bigcap_{\varepsilon>0}J^\varepsilon_{\mathrm{prod}}$, where $A\in J^\varepsilon_{\mathrm{prod}}$ means that on $A$ the fixed lifting $\Phi$ can be approximated to within $\varepsilon$ by a product-form sequence of factor maps. The proof first shows, for finite factor spaces, that $J^\varepsilon_{\mathrm{prod}}$ meets every uncountable almost disjoint family, so $J_{\mathrm{prod}}$ is nonmeager; a uniformization lemma then patches approximate liftings on two nonmeager subfamilies of $J_{\mathrm{prod}}$ to produce a $3\varepsilon$-lifting on all of $\mathbb{N}$. The passage from finite to separable factors uses countable dense subsets, a cofinal $K_1$-homogeneous set obtained from OCA, and a metric version of the partial-selector dichotomy (Lemma A.2), a dichotomy saying that a family of partial coordinate choices either splits into countably many pieces with bounded disagreement or contains a tree-like family with prescribed disagreements.
What would settle it
Find, under OCA, a family $F$ of partial selectors for a sequence of separable metric spaces and some $\varepsilon>0$ for which Lemma A.2's first alternative fails ($F$ is not a countable union of pieces with $|\mathrm{Diff}_\varepsilon(g,h)|\le n$) and its second alternative fails (no tree-like almost disjoint family $A$ and injection $f\colon Z\to F$ with the stated disagreement property), contradicting the lemma and removing the basis for Proposition 2.10.
Extended reading notes
Core claim
The central claim is Theorem 2.3: under OCA and $\mathrm{MA}_{\aleph_1}(\sigma\text{-linked})$, any coordinate respecting function $\phi\colon \prod_n M_n/\mathrm{Fin}\to \prod_n N_n/\mathrm{Fin}$ between reduced products of separable metric spaces of uniformly bounded diameter admits a lifting of product form, i.e. there is an almost permutation $f$ of $\mathbb{N}$ and maps $h_n\colon M_{f(n)}\to N_n$ with $\phi([a])=[h_n(a_{f(n)})]$. The metric structure enters through the pseudometrics $d_S([a],[b])=\limsup_{n\in S} d_n(a_n,b_n)$ indexed by $S\in \mathcal{P}(\mathbb{N})/\mathrm{Fin}$, and coordinate respecting means equality modulo each $d_S$ is preserved. Since OCA forces the associated automorphism of $\mathcal{P}(\mathbb{N})/\mathrm{Fin}$ to come from an almost permutation, the proof reduces to the identity case and then shows the ideal $J_{\mathrm{prod}}$ of index sets carrying exact product-form liftings must be all of $\mathbb{N}$.
Load-bearing premise
The argument's load-bearing premise is that the metric version of the partial-selector dichotomy (Lemma A.2) follows from the discrete version by routine modifications, a step the paper only sketches and on which the uniformization that yields $\mathbb{N}\in J_{\mathrm{prod}}$ depends.
Editorial extensions
If this is right
- If Theorem 2.3 is correct, then under OCA and $\mathrm{MA}_{\aleph_1}(\sigma\text{-linked})$ there are no non-trivial coordinate respecting maps between metric reduced products; every such map is a permutation of indices plus factorwise maps.
- Consequently, isomorphisms and automorphisms between such reduced products are all of this rigid form, so coordinate respecting maps cannot flexibly deform one asymptotic metric structure into another.
- The nonmeagerness of $J_{\mathrm{prod}}$ and the uniformization argument give a template for lifting theorems in other quotient structures: build approximate product-form liftings on a large ideal, then patch them.
- Under CH, the paper notes the opposite behaviour: non-trivial coordinate respecting functions between metric reduced products exist, so the rigidity is genuinely a consequence of the forcing axioms.
Reading between the lines
- Editorial inference: the same two-stage proof, finite factor spaces first and then separable factors via countable dense subsets, may carry over to reduced products over ideals other than $\mathrm{Fin}$, provided an analogue of the partial-selector dichotomy holds for those ideals.
- Editorial inference: since the theorem assumes uniformly bounded diameter, a natural test is whether the conclusion survives without that bound; the pseudometrics $d_S$ no longer align with asymptotic equality on $S$ when diameters are unbounded, so a different notion of coordinate respecting would be needed.
- Editorial inference: the use of OCA to convert the automorphism of $\mathcal{P}(\mathbb{N})/\mathrm{Fin}$ into an almost permutation suggests the result may extend to other quotient Boolean algebras whose automorphism groups are trivial under forcing axioms.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper states and proves a metric generalization of the lifting theorem of [1]: under OCA and MA_ℵ1(σ-linked), every coordinate respecting function between reduced products of sequences of separable metric spaces with uniformly bounded diameter is trivial, i.e., induced by an almost permutation of N and coordinatewise maps between the factor spaces. The proof first treats finite metric factor spaces, introducing ideals J^ε_prod of index sets admitting approximate product-form liftings, proving their nonmeagerness via an OCA argument (§2.1), and then uniformising exact liftings with the help of a metric partial-selector dichotomy (Lemma A.2) and a 3ε-diagonalization lemma (Lemma 2.9). The general separable case is reduced to countable factor spaces via dense subsets and then to finite approximations indexed by NN, with a final OCA-based uniformisation (§2.2).
Significance. If sound, the theorem is a genuine and useful extension of [1] from discrete to metric structures, relevant to rigidity questions for reduced products (e.g., in C*-algebra applications). The paper has several strengths: the reduction of the coordinate respecting automorphism to the identity, the careful handling of approximate liftings with explicit ε-arithmetic, the effective use of nonmeagerness of the ideals, and a clean finite-to-countable-to-separable strategy. The manuscript is also honest about its debts to [1], which is cited as 'to appear'. The main theorem is falsifiable: under CH there are coordinate respecting nontrivial maps, so the forcing axioms are essential. However, two load-bearing steps are not fully established: the metric partial-selector dichotomy is only sketched (and invokes OCA#, a strengthening not assumed in the theorem), and the uniform-k step in the final uniformisation is asserted without proof. These must be supplied before the result can be regarded as proved.
major comments (2)
- [Appendix A, Lemma A.2] Lemma A.2 is the load-bearing step of Proposition 2.10: Claim 2.12 uses its alternative (1) to partition J_prod into sets G_n with |Diff_ε(h^A,h^B)| ≤ n, and without that partition the iterative application of Lemma 2.9 cannot produce N ∈ J^{9ε}_prod. The proof of the lemma, however, is only a sketch: it says that the proof of [1, Proposition 3.7] goes through 'mutatis mutandis' and defers the derivation of alternatives (1) and (2) to the discrete case, while the number n + (4^{n+1}-1)/3 and the open-ball colourings W^n_m are introduced without a verification that they yield the perfect tree-like family with the 'for every A ∈ A' clause. Moreover, the proof applies OCA#, which is described as a formal strengthening of OCA, whereas the lemma and Theorem 2.3 assume only OCA; the paper does not show that OCA# is available for these particular colourings. Please provide a complete proof of Lemma A.2 (or state it under the exact axiom needed and prove that OCA suffices for the colourings at hand); as written, the main theorem is conditional on an unverified adaptation.
- [Section 2.2, proof of Theorem 2.3 (after Claim 2.14)] The step 'Since ≤* = ⋃_k ≤k ... we can find k such that G_{n̄} is ≤k-cofinal' swaps the quantifiers in the definition of ≤*-cofinality: for each f there are g ∈ G_{n̄} and k(f) with f ≤_{k(f)} g does not imply the existence of a single k working for all f. The subsequent construction of uniform functions h_n for all n ≥ k depends on this uniformity (it needs, for each n ≥ k and each x ∈ M_n, some g ∈ G_{n̄} with x ∈ M_{n,g(n)}). If the uniformity claim is true, a proof must be given; otherwise the argument should be modified, for example by using, for each a, a master g ∈ G_{n̄} with f_a ≤* g and comparing h_n(a_n) with h_{n,g}(a_n) on a tail via K1-homogeneity, together with a coordinatewise choice lemma for the values of h_n. As it stands, the final step of Theorem 2.3 is not fully justified.
minor comments (6)
- [Section 2.2] In the dense-subset reduction, the sentence 'for x ∈ M_n \ D_n, h̃_n(x) = h_n(y) where y is any element of M_n which has distance ≤ 1/n from x' is not correct as written: y should be chosen in D_n, since h_n is only defined there, and the verification that the extended maps lift φ should be spelled out (using that [a] = [b] for the chosen dense approximants).
- [Claim 2.12] The choice of distinct x,y with Δ(x,y) ≥ n should be strengthened to Δ(x,y) > n (i.e., ≥ n+1), so that the element j of (Diff_ε(h^{f(x)},h^{f(y)}) ∩ B) \ Δ(x,y) satisfies j > n(x) and hence j ∉ Diff_{ε/2}(h^{f(x)},h^B) for both f(x) and f(y); with Δ(x,y) ≥ n, the case j = n leaves the displayed triangle inequality unjustified.
- [Proposition 2.6] The inference 'As ε was arbitrary, each J^ε_prod intersects nontrivially all uncountable almost disjoint families, hence so does Jprod' needs the short diagonal argument: if A were an uncountable almost disjoint family with A ∩ Jprod = ∅, then for each A ∈ A there is m with A ∉ J^{1/m}_prod, so some B_m = {A ∈ A : A ∉ J^{1/m}_prod} would be uncountable, contradicting that J^{1/m}_prod intersects every uncountable almost disjoint family.
- [Section 2.2, topology on NN] In the definition of the basic open neighbourhoods of NN, the condition '∀x ∈ M_{n,f(n)}' should read '∀x ∈ M_{n,f(n)} ∩ M_{n,g(n)}', since h_{n,g} is only defined on M_{n,g(n)}.
- [Claim 2.12, proof] In the displayed triangle inequality '∂_n(h^{f(x)}_j(w), h^{f(y)}_j(w)) ≤ ...', the metric index should be j, not n, since the inequality is evaluated at the fixed index j.
- [Claim 2.7] The 'standard Martin's Axiom argument' that shrinks the K0-homogeneous set H to one with a_n ∈ {s^0_n, s^1_n} for all (a,A) ∈ H and n ∈ A should be stated at least in outline, since this is the main place where MA_ℵ1(σ-linked) is used in §2.1 and the translation from [1, Proposition 5.2] is not immediate in the metric setting.
Circularity Check
No circularity: the metric lifting theorem is proved by a genuine OCA/MA argument; the self-cited discrete dichotomy in [1] is independent of the target result, though Lemma A.2 is only sketched.
full rationale
Walking the derivation chain: Theorem 2.3 is reduced to the identity-automorphism case via [1, Theorem 1], and then to showing N∈Jprod. Lemma 2.4 establishes Jprod=∩J^ε_prod directly from the triangle inequality, with no circular use of triviality. Proposition 2.6 proves Jprod nonmeager by a self-contained OCA colouring (Claim 2.7) and a density argument (Claim 2.8); the only imported ingredient is the MA-based thinning 'translates readily' from [1, Prop. 5.2], a separate result not derived from Theorem 2.3. Proposition 2.10 uses Claim 2.11, which follows from the definition of Jprod, and Claim 2.12, whose combinatorial engine is Lemma A.2. Lemma A.2 is stated as a metric generalization of [1, Prop. 3.7]; the proof is only sketched ('we only sketch the required modifications') and invokes OCA#, a strengthening of OCA. This is a genuine load-bearing omitted proof and a correctness risk if the mutatis-mutandis adaptation fails, but it is not circular: [1, Prop. 3.7] is prior independent work by overlapping authors, is parameter-free, and does not assume the present theorem. The separable case in §2.2 applies the finitary theorem to dense subsets and uses OCA plus bounding number; no fitted parameter is renamed as a prediction. At no point is the conclusion 'φ is trivial' used as an input, and no quantity is defined in terms of the target statement. Hence no circular step; score 0, with the Appendix A proof sketch flagged as a rigor/support issue rather than circularity.
Assumptions & free parameters
assumptions (7)
- standard math ZFC
- domain assumption OCA (Open Colouring Axiom)
- domain assumption MA_ℵ1(σ-linked)
- domain assumption OCA# (formal strengthening of OCA)
- domain assumption All automorphisms of P(N)/Fin are induced by almost permutations under OCA
- standard math Nonmeager subsets of P(N) have nonmeager sections A_n for all sufficiently large n
- domain assumption The bounding number b is greater than ℵ1 under OCA
Cite this review
Pith. "Pith review of A metric lifting theorem." pith.science (2026). https://pith.science/paper/JX2BZMFH
@misc{pith2026241111127,
author = {Pith},
title = {Pith review of: A metric lifting theorem},
year = {2026},
howpublished = {\url{https://pith.science/paper/JX2BZMFH}},
note = {Machine review of arXiv:2411.11127}
}
read the original abstract
In a recent article by Farah and the authors, a strong lifting theorem was proved for a class of coordinate-respecting maps between reduced products of discrete structures, hereby working under mild Forcing Axioms. We generalise this lifting theorem to the metric setting.
Forward citations
Cited by 3 Pith papers
-
Rigidity of Higson coronas
Under OCA and MA_ℵ1, homeomorphic Higson coronas of uniformly locally finite metric spaces force coarse equivalence, a statement independent of ZFC.
-
On automorphism groups of metric reduced products of symmetric groups
Assuming OCA and Martin's axiom for sigma-linked posets, all isomorphisms between metric reduced products of symmetric groups are trivial, and the outer automorphism group is described explicitly.
-
Biba's trick, with applications
Assuming OCA_T and MA(σ-linked), every homomorphism from P(N) into a quotient by a countably 80-determined ideal has a continuous lifting on a nonmeager ideal, improving PFA-based results and removing MA for countably...
Reference graph
Works this paper leans on
-
[1]
B. De Bondt, I. Farah, and A. Vignati, Trivial isomorphisms between reduced products, to appear in Israel Journal of Mathematics, arXiv:2307.06731
-
[2]
E. G. Effros and J. Rosenberg, C -algebras with approximately inner flip , Pacific J. Math. 77 (1978), no. 2, 417--443
work page 1978
-
[3]
I. Farah, Analytic quotients: theory of liftings for quotients over analytic ideals on the integers, Mem. Amer. Math. Soc. 148 (2000), no. 702, xvi+177
work page 2000
-
[4]
, Between reduced powers and ultrapowers, J. Eur. Math. Soc. (JEMS) 25 (2023), no. 11, 4369--4394
work page 2023
- [5]
-
[6]
Ghasemi, Reduced products of metric structures: a metric F eferman-- V aught theorem , J
S. Ghasemi, Reduced products of metric structures: a metric F eferman-- V aught theorem , J. Symbolic Logic 81 (2016), no. 3, 856--875
work page 2016
Reviewed August 12, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.