REVIEW 3 major objections 8 minor 1 cited by
There may be an $n$-entangled set but no $n+1$-entangled sets
T0 review · 3 major / 8 minor · reviewed 2026-08-05 · deepseek-v4-flash
Pith's one-line read For every n ≥ 2, there is a consistent model with an n-entangled set but no (n+1)-entangled set.
desk verdict Answers a real open question, but the main theorem leans on an unproven imported consistency proof; worth refereeing with conditions. 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 key mechanism is the forcing $\mathbb{P}(X,E)$, whose conditions are finite $\in$-chains of countable elementary submodels — 'good pairs' $(s, x)$ with $x(i) \in s(i+1) \setminus s(i)$ — together with the constraint that the types of any two distinct $(n+1)$-tuples $x, y$ are non-constant. The chain of submodels makes the forcing proper and supports two duplication lemmas (Lemmas 3.7 and 3.8) that split the work according to whether a small piece of E sits in one or many submodel layers; these lemmas preserve the $n$-entangledness of E while the generic filter adds an uncountable family of $(n+1)$-tuples from X with no constant type, so X is no longer $(n+1)$-entangled. The parametrized axiom $\mathrm{PFA}_{n\text{-ent}}(E)$ then supplies the
What would settle it
Read Theorem 97 of [13] (cited here as Theorem 2.6) and check whether its consistency proof for $\mathrm{PFA}_{n\text{-ent}}(E)$ uses large cardinals beyond ZFC; if it does, the unqualified consistency claim in Theorem 0.2 is false as stated. Alternatively, attempt to construct an $(n+1)$-entangled set inside a model of $\mathrm{PFA}_{n\text{-ent}}(E)$; any such set would refute the paper's main implication.
Extended reading notes
Core claim
The paper's central claim is that for each fixed integer $n \ge 2$, the statement 'there is an $n$-entangled set of reals, but there is no $(n+1)$-entangled set of reals' is consistent with ZFC plus the P-ideal dichotomy and Martin's axiom. The proof works through a parametrized forcing axiom, $\mathrm{PFA}_{n\text{-ent}}(E)$, which asserts that a fixed set E is $n$-entangled and that every proper forcing preserving E's $n$-entangledness has a filter meeting any family of $\aleph_1$ dense sets. This axiom, imported from Guzmán and Todorcevic, is consistent and implies both MA and PID. The genuinely new piece is a forcing $\mathbb{P}(X,E)$ that, for any uncountable X, is proper, preserves E's $n$-entangledness, and forces X to stop being $(n+1$
Load-bearing premise
The paper's main theorem rests on the imported consistency of the parametrized forcing axiom $\mathrm{PFA}_{n\text{-ent}}(E)$ from [13], a theorem stated with no proof here and no explicit large-cardinal hypothesis, together with the omitted proof that $\mathbb{P}(X,E)$ is proper; if either fails, the consistency claim does not go through.
Editorial extensions
If this is right
- The hierarchy of entangledness is not well-ordered by consistency strength: each level n can be the top level in a model of PID+MA.
- Question 3.5 of [6] has a positive answer for every n ≥ 2.
- Under FCA^Δ, the full spectrum of exact entangledness levels occurs simultaneously among pairwise homeomorphic subsets of ℝ.
- Neither the existence of a 2-capturing scheme nor a 2-Δ-capturing scheme implies the existence of an entangled set, and a 2-Δ-capturing scheme can coexist with MA.
- The boundary between finite and infinite entangledness is not merely a matter of taking all n; each finite level behaves like an independent phenomenon.
Reading between the lines
- The good-pairs forcing ℙ(X,E) is a template: any property of X that can be killed by an uncountable family of finite tuples avoiding a definable set of types may be destructible this way while preserving a hereditary property of E.
- If the imported consistency of PFA_{n-ent}(E) turns out to require large cardinals, the main theorem can likely be repaired by re-running the argument inside a model of PFA for the right class of forcings, but the unqualified ZFC-consistency claim would need revision.
- The FCA^Δ result suggests that the Cohen model, where FCA^Δ holds, contains a complete ladder of exact entangledness levels; checking this would give a CH-based witness without any forcing axioms.
- Problem 0.6 — whether every 2-entangled set can be made n-entangled — might be approachable by iterating ℙ(X,E) along the uncountable targets X, destroying their higher entangledness one by one.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. This paper answers Question 3.5 of Carroy-Levin-Notaro [6] in the positive: for every n >= 2, the statement 'there is an n-entangled set but no n+1-entangled set' is shown consistent together with PID and MA (Theorem 0.2). The proof passes through a parametrized proper forcing axiom PFA_{n-ent}(E), argued to be consistent in [13] (Theorem 2.6) and to imply MA and PID (Theorems 2.7-2.8). The new technical core is a forcing P(X,E) built from finite chains of 'good pairs' (Section 3), which is proper, preserves the n-entangledness of a fixed E, and makes an arbitrary uncountable X non-(n+1)-entangled. The second half of the paper gives construction-scheme results: FCA^Delta produces n-entangled sets that are not n+1-entangled for each n (Theorem 0.3), and 2-capturing schemes are shown insufficient for constructing entangled sets (Theorems 0.4 and 0.5).
Significance. If Theorem 0.2 is established, it is a substantial result: it answers the open question from [6] and shows the finite n-entangledness hierarchy can be terminated at any prescribed level in models of two strong forcing axioms. The forcing P(X,E) and the preservation argument are original and constitute the main technical contribution; the construction-scheme results are clean and useful, and Theorem 0.3 unifies prior CH-based constructions. The paper is proof-based; no machine-checked proofs or code are provided. The significance is tempered by the fact that the lower-bound consistency theorem is imported from an unpublished source and the preservation proof currently has a gap; both are fixable but are load-bearing.
major comments (3)
- [§2, Theorem 2.6] The consistency of PFA_{n-ent}(E) is the lower bound for the main theorem, but it is quoted ('Essentially Theorem 97, [13]') without proof, without an explicit large-cardinal hypothesis, and deferred to an upcoming joint paper. The abstract and Theorem 0.2 state 'is consistent' unqualified; if the [13] proof uses a supercompact cardinal, as is typical for PFA-type axioms, the consistency claim must be qualified. Please state the exact hypothesis, give the proof or a detailed reference with the argument, and adjust the abstract accordingly.
- [§3, proof of Theorem 3.10] In the recursive construction, the case 'k ∈ P and a_k = a' is handled 'using Lemma 3.7'. Lemma 3.7 requires a to be of the first kind over the chain F_{n_r+k}; the proof has not shown this, and it can fail (a may have several elements in one F(i+1)∖F(i) block). In that situation Lemma 3.7 is inapplicable; Lemma 3.8 does apply because a ∩ F_{n_r+k}(0)=∅, and since σ_k=τ when a_k=a, its conclusion type(a,b)=τ is exactly what the recursion needs. The case split on first/second kind is therefore necessary.
- [§3, Theorem 3.4] The genericity criterion for P(X,E) — hence its properness — is stated and its proof omitted ('we leave it to the reader'). Properness is required to apply PFA_{n-ent}(E), and the analogous preservation proof (Theorem 3.10) contains the gap described above, so this omission cannot be regarded as routine. A proof or at least a detailed sketch should be included. The paper should also explicitly describe the ω1-many dense sets that turn the generic-filter object Q_G of Corollary 3.5 into a witness of non-(n+1)-entangledness under PFA.
minor comments (8)
- [§2, PFA definition] The displayed definition of PFA_{n-ent}(E) ends with 'then m(P)' with no conclusion; presumably 'm(P)=ω1'. As printed the axiom is incomplete.
- [§3, Definition 3.6] The definition of a_i^F uses F(i+1) for i ≤ n; the convention that the top model F(|F|) means the universe (stated earlier for chains) should be repeated here to avoid ambiguity.
- [§3, proof of Theorem 3.10] The recursion bound 'Assume k ≤ n' should be over the number n_r^p of new good pairs, not the fixed integer n; as written it is confusing.
- [§5, after Prop. 5.5] The homeomorphism between E_F, E_e, and E_o is asserted solely from the equality of the Δ-values. Since the definitions of e_α and o_α allow sign flips (e.g., e_α(l) = -Ξ_α(l)), the order at the first difference can be reversed, so the claimed homeomorphism needs a proof.
- [§5, proof of Prop. 5.4, Case 2] The inequality 'n = |C_odd| < |C_{l-1}^j| - 1 = n - 1' is garbled; the contradiction should be n ≤ n - 1 (or |C_odd| ≤ |C_{l-1}^j| - 1). Please fix.
- [§6, proof of Theorem 6.1] 'It is easy to see that P is a ccc-forcing provided that P is an entangled (thus, increasing) set' should say 'provided that E is an entangled set' — P is the forcing notion, not the entangled set.
- [§2, proof of Theorem 2.7] The proof uses 'Claim 1 of Lemma 8.5 in [1]' in a generalized n-entangled form. Provide a reference or a proof of this generalization, since [1] states it for entangled sets only.
- [References] [13] is cited without a venue. If it is a preprint, please state its status; given that Theorem 2.6 is essential, a published reference or an appendix with the argument would be preferable.
Circularity Check
No significant circularity; the central derivation is a new forcing-preservation proof resting on an external consistency theorem imported from [13].
full rationale
The paper's central derivation (Theorem 0.2) is not circular by construction. It starts from Theorem 2.6, an imported consistency theorem for the parametrized forcing axiom PFA_{n-ent}(E) ('Essentially Theorem 97, [13]'), and then proves new content in Section 3: the forcing P(X,E) destroys the n+1-entangledness of every uncountable X while preserving the n-entangledness of a fixed E. The forcing is explicitly designed to add an uncountable family of (n+1)-tuples with pairwise non-constant type; this directly witnesses the failure of n+1-entangledness, so no quantity is fitted and then renamed as a prediction. The preservation proof (Theorem 3.10) is a combinatorial argument using Lemmas 3.7-3.9, not a restatement of the desired conclusion. The secondary results under FCA^Δ and 𝔪_S > ω1 are conditional statements proved from prior axioms; citations to the author's own [8], [10], [11] point to independent earlier theorems (e.g., FCA^Δ from CH) rather than to the target result. The main caveat is not circularity: Theorem 2.6 is imported with no proof in this paper and with no explicit large-cardinal hypothesis, and Theorem 3.4 is said to be left to the reader. These are completeness/correctness risks about the external support, not cases where an equation reduces to an input or a fitted parameter is called a prediction.
Assumptions & free parameters
free parameters (1)
- scheme type growth bound n_{k+1} ≥ 2m_k
assumptions (6)
- standard math ZFC
- domain assumption Consistency of PFA_{n-ent}(E) (Theorems 2.5 and 2.6)
- domain assumption Claim 1 of Lemma 8.5 of [1]
- domain assumption FCA^Δ is consistent (Theorem 4.11 from [8])
- domain assumption Existence of a fully Δ-capturing scheme with rapidly growing type
- standard math Standard toolbox: Δ-system lemma, elementary submodels, proper forcing, Neeman iterations
invented entities (2)
-
PFA_{n-ent}(E), parametrized forcing axiom
-
Forcing P(X,E) with good pairs
Cite this review
Pith. "Pith review of There may be an $n$-entangled set but no $n+1$-entangled sets." pith.science (2026). https://pith.science/paper/RBAOGSYY
@misc{pith2026250901029,
author = {Pith},
title = {Pith review of: There may be an $n$-entangled set but no $n+1$-entangled sets},
year = {2026},
howpublished = {\url{https://pith.science/paper/RBAOGSYY}},
note = {Machine review of arXiv:2509.01029}
}
abstract
In this paper we show that for every $2\leq n\in \mathbb{N}$, the statement "there is an $n$-entangled set, but there are no $n+1$-entangled sets" is consistent. We also prove some theorems which improve our understanding of entangled sets in relation to construction schemes: (1) The axiom FCA$^\Delta$ introduced in \cite{finitizationclubch} implies the existence of $n$-entangled sets which are not $n+1$-entangled. (2) $\mathfrak{m}_\mathcal{F}>\omega_1$ implies the non-existence of entangled sets. Thus, $2$-capturing schemes alone are not sufficient to build these kinds of linear orders. (3) The existence of a 2-$\Delta$-capturing scheme is consistent with MA.
Forward citations
Cited by 1 Pith paper
-
Open Colorings and Baumgartner's Axiom
It is consistent that MA_ℵ1 and OCA_T hold while Baumgartner's Axiom fails, witnessed by a non-reversible, non-increasing ℵ1-dense set of reals.
Reference graph
Works this paper leans on
-
[6]
Some questions on entangled linear orders, 2025
Raphaël Carroy, Maxwell Levine, and Lorenzo Notaro. Some questions on entangled linear orders, 2025
work page 2025
-
[13]
The p-ideal dichotomy, martin’s axiom and entangled sets
Osvaldo Guzmán and Stevo Todorcevic. The p-ideal dichotomy, martin’s axiom and entangled sets
-
[1]
Uri Abraham, Matatyahu Rubin, and Saharon Shelah. On the consistency of some partition theorems for continuouscolorings,andthestructureof ℵ1-denserealordertypes. Ann.PureAppl.Logic ,29(2):123–206, 1985
work page 1985
-
[2]
Martin’s axiom does not imply that every twoℵ1-dense sets of reals are isomorphic
Uri Avraham and Saharon Shelah. Martin’s axiom does not imply that every twoℵ1-dense sets of reals are isomorphic. Israel J. Math., 38(1-2):161–176, 1981
work page 1981
- [3]
-
[4]
James E. Baumgartner. Allℵ1-dense sets of reals can be isomorphic.Fund. Math., 79(2):101–106, 1973
work page 1973
-
[5]
Hausdorff gaps and towers in(𝜔)∕Fin
Piotr Borodulin-Nadzieja and David Chodounský. Hausdorff gaps and towers in(𝜔)∕Fin. Fund. Math., 229(3):197–229, 2015
work page 2015
-
[7]
On construction schemes: building the uncountable from finite pieces, 2024
Jorge Antonio Cruz Chapital. On construction schemes: building the uncountable from finite pieces, 2024
work page 2024
Show all 28 references
-
[8]
Multiple gaps and some finitizations of club and ch, 2025
Jorge Antonio Cruz Chapital. Multiple gaps and some finitizations of club and ch, 2025
2025
-
[9]
Why Y-c.c.Ann
David Chodounský and Jindřich Zapletal. Why Y-c.c.Ann. Pure Appl. Logic, 166(11):1123–1149, 2015
2015
-
[10]
Construction schemes: Transferring structures from𝜔to𝜔1.submitted
Jorge Antonio Cruz Chapital, Osvaldo Guzmán González, and Stevo Todorcevic. Construction schemes: Transferring structures from𝜔to𝜔1.submitted
-
[11]
Forcing axioms and con- struction schemes.In preparation
Jorge Antonio Cruz Chapital, Osvaldo Guzmán González, and Stevo Todorcevic. Forcing axioms and con- struction schemes.In preparation
-
[12]
PhD thesis, University of Toronto (Canada), 2025
Emily Erlebach.Proper forcing, the𝑃-ideal dichotomy and the S-space problem. PhD thesis, University of Toronto (Canada), 2025
2025
-
[14]
Large irredundant sets in operator algebras.Canad
Clayton Suguio Hida and Piotr Koszmider. Large irredundant sets in operator algebras.Canad. J. Math., 72(4):988–1023, 2020
2020
-
[15]
ActaMath.Hungar
D.KalajdzievskiandF.Lopez.Forcingandconstructionschemes. ActaMath.Hungar. ,159(1):27–41,2019
2019
-
[16]
Entangledness in Suslin lines and trees.Topology Appl., 275:107157, 19, 2020
John Krueger. Entangledness in Suslin lines and trees.Topology Appl., 275:107157, 19, 2020
2020
-
[17]
Banach spaces from a construction scheme.J
Fulgencio Lopez. Banach spaces from a construction scheme.J. Math. Anal. Appl., 446(1):426–435, 2017
2017
-
[18]
Constructionschemesandtheirapplications .PhDthesis,UniversityofToronto(Canada), 2018
FulgencioLopez. Constructionschemesandtheirapplications .PhDthesis,UniversityofToronto(Canada), 2018
2018
-
[19]
Proc.Amer.Math.Soc
FulgencioLopezandStevoTodorcevic.Treesandgapsfromaconstructionscheme. Proc.Amer.Math.Soc. , 145(2):871–879, 2017
2017
-
[20]
Math., 251(1):35–68, 2020
TadatoshiMiyamotoandTeruyukiYorioka.AfragmentofAsperó-Mota’sfinitelyproperforcingaxiomand entangled sets of reals.Fund. Math., 251(1):35–68, 2020
2020
-
[21]
Baumgartner’s isomorphism problem forℵ2-dense suborders of ℝ.Arch
Justin Tatch Moore and Stevo Todorcevic. Baumgartner’s isomorphism problem forℵ2-dense suborders of ℝ.Arch. Math. Logic, 56(7-8):1105–1114, 2017
2017
-
[22]
Forcing with sequences of models of two types.Notre Dame J
Itay Neeman. Forcing with sequences of models of two types.Notre Dame J. Form. Log., 55(2):265–298, 2014
2014
-
[23]
Combinatorial dichotomies and cardinal invariants.Math
Dilip Raghavan and Stevo Todorcevic. Combinatorial dichotomies and cardinal invariants.Math. Res. Lett., 21(2):379–401, 2014
2014
-
[24]
J.Stepr¯ansandW.S.Watson.Homeomorphismsofmanifoldswithprescribedbehaviouronlargedensesets. Bull. London Math. Soc., 19(4):305–310, 1987
1987
-
[25]
Weak baumgartner axioms and universal spaces, 2025
Corey Bacal Switzer. Weak baumgartner axioms and universal spaces, 2025
2025
-
[26]
A construction scheme for non-separable structures.Adv
Stevo Todorcevic. A construction scheme for non-separable structures.Adv. Math., 313:564–589, 2017. THERE MAY BE AN𝑛-ENTANGLED SET BUT NO𝑛+1-ENTANGLED SETS 23
2017
-
[27]
Remarks on chain conditions in products.Compositio Math., 55(3):295–302, 1985
Stevo Todorčević. Remarks on chain conditions in products.Compositio Math., 55(3):295–302, 1985
1985
-
[28]
American Mathematical Society, Providence, RI, 1989
Stevo Todorčević.Partition problems in topology, volume 84 ofContemporary Mathematics. American Mathematical Society, Providence, RI, 1989. hlj Jorge Antonio Cruz Chapital Department of Mathematics, University of Toronto, Canada cruz.chapital at utoronto.ca
1989
Reviewed August 5, 2026 · model on record in the stance chip above.
Discussion (0). Sign in to comment.