No covering with nowhere dense P -sets in the Cohen model

Alan Dow, Osvaldo Guzmán · 2025

Read No covering with nowhere dense P -sets in the Cohen model by Alan Dow, Osvaldo Guzmán (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2507-22476.txt.

Introduction

Dow and Guzmán present a 2025 set theory paper on covering the remainder of the natural numbers with thin P-sets. The setting is the Cohen model, built by adding Cohen reals to a universe where the continuum hypothesis holds. The theorem says that when fewer than a certain large count of such reals are added, no family of nowhere dense P-sets covers that remainder space. A matching form of the result guarantees an ultrafilter on the natural numbers that holds no tower inside it.

Catalog record

CREATORAlan Dow, Osvaldo Guzmán
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE49 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2507-22476

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

No covering with nowhere dense P -sets in the Cohen model Alan Dow, Osvaldo Guzmán Abstract We prove that if less than ℵ ω \aleph_{\omega} -many Cohen reals are added to a model of CH , then ω ∗ \omega^{\ast} can not be covered by nowhere dense P -sets (equivalently, there is an ultrafilter on ω \omega that does not contain a tower). 1 Introduction Recall that a subset B of a topological space 1 1 1 All spaces under discussion are Hausdorff. X is called a P-set if the intersection of countably many neighborhoods of B is still a neighborhood of it. Similarly, a point a\in X is a P -point if the singleton \left\{a\right\} is a P -set. While there are infinite spaces in which every point is a P -point, this is not the case for compact spaces; in other words, every infinite compact space contains a non P -point (see [ 10 ] ). This result lead Kunen, van Mill and Mills (see [ 19 ] ) to wonder the following: Can an infinite compact space be covered with nowhere dense P- sets? A nowhere dense set is a set whose closure has empty interior, such sets are often considered “small” in the topological sense. The motivation behind the question is that, although an infinite compact space must have non P -points, it might still be possible that every point lies in a “small” P -set. A particular case of the main theorem of [ 19 ] is the following: Theorem 1 (Kunen, van Mill, Mills [ 19 ] ) If X is a compact space of \pi -weight \omega_{1}, then it has a point that does not belong to any nowhere dense P- set. We now turn our attention to \omega^{\ast}=\beta\omega\setminus\omega, the remainder of the Čech-Stone compactification of the countable discrete space \omega (for an excellent introduction to this topic, the reader may consult [ 20 ] ). By a straightforward application of Stone duality (see [ 11 ] ), the following statements are equivalent: 1. \omega^{\ast} can not be covered by nowhere dense P -sets. 2. There is an (non-principal) ultrafilter on \omega that does not contain a nowhere dense P -filter. 2 2 2 These notions will be reviewed in the following section. 3. There is an (non-principal) ultrafilter on \omega that intersects every tall P -ideal. By Theorem 1 , it follows that the Continuum Hypothesis ( CH) implies that \omega^{\ast} can not be covered by nowhere dense P -sets. Surprisingly, it is also consistent that \omega^{\ast} can be covered by such sets. This remarkable result was first proved by Balcar, Frankiewicz and Mills in [ 1 ] . Since then, this topic has continued to attract attention and more results have been obtained. We now list some of them: Theorem 2 (Balcar, Frankiewicz, Mills [ 1 ] ) It is consistent that \omega^{\ast} can be covered by nowhere dense P -sets. Theorem 3 (D. [ 6 ] ) It is consistent with Martin’s Axiom (MA) and \mathfrak{c}=\omega_{2} that \omega^{\ast} can not be covered by nowhere dense P -sets. Theorem 4 (D. [ 7 ] ) If \omega_{2} Cohen reals are added to a model of CH + \square_{\omega_{1}}, then \omega^{\ast} can not be covered by nowhere dense P -sets. Theorem 5 (D. [ 7 ] ) In the Laver model, every ultrafilter contains a tower (hence, \omega^{\ast} can be covered with nowhere dense P -sets). Theorem 6 (Zhu [ 21 ] , Brendle, Farkas, Verner [ 3 ] ) The principle of Near Coherence of Filters ( NCF ) implies that \omega^{\ast} can be covered by nowhere dense P -sets. Theorem 7 (D., van Mill [ 9 ] ) No compact space can be covered by nowhere dense ccc P -sets. The purpose of this paper is to improve upon Theorem 4 . We show that if fewer than \aleph_{\omega} Cohen reals are added to a model of CH, then there is an ultrafilter that intersects every tall P -ideal (the principle \square_{\omega_{1}} is no longer needed). Our approach builds on the argument presented in [ 7 ] . However, by carrying out a more refined analysis of the interaction between the countable elementary submodels used in that proof, we were able to eliminate the need of \square_{\omega_{1}} and extend the result for more than \omega_{2} Cohen reals. Although our proof is heavily inspired by [ 7 ] , no familiarity with that paper is needed to follow the arguments we present here. This work shares some conceptual overlap with our earlier paper [ 8 ] . While we will reference several results from that work, the present paper remains self contained, readers only need to accept the quoted results and require no prior familiarity with [ 8 ] . The paper is organized as follows. After a review of notation and some preliminaries, in Section 5 we present a proof that CH implies that \omega^{\ast} can not be covered by nowhere dense P -sets. This result is not new and follows from Theorem 1 above. However, we included this result since the proof serves as motivation for some of the ideas that will be used in the main part of the paper. In Section 6 , we develop the main technical results that we will need in the next section. We obtain several combinatorial results regarding countable elementary submodels that may have more applications in the future. In Section 7 , we use the ideas and results previously developed to prove that after adding less \aleph_{\omega} Cohen or random reals to a model of CH , there is an ultrafilter that intersects every increasing tower. We use this result in Section 8 to conclude that if fewer than \aleph_{\omega} -many Cohen reals are added to a model of CH , then \omega^{\ast} can not be covered by nowhere dense P -sets. We finish the paper with some open questions. 2 Notation If X is any set, we denote by \mathcal{P}\left(X\right) its power set (the collection of all its subsets). We say that \mathcal{I}\subseteq\mathcal{P}\left(X\right) is an ideal on X if \emptyset\in\mathcal{I} , X\notin\mathcal{I} and it is closed under finite intersections and taking subsets. On the other hand, \mathcal{F\subseteq} \wp\left(X\right) is a filter on X if X\in\mathcal{F} , \emptyset\notin\mathcal{F}, and it is closed under finite unions and taking supersets. An ultrafilter is a maximal filter that does not contain any finite set (in this work, all our ultrafilters are assumed to be non-principal). Let \mathcal{B} be a collection of subsets of \omega, denote \mathcal{B}^{\ast}=\left\{\omega\setminus B\mid B\in\mathcal{B}\right\}. It is easy to see that if \mathcal{F} is a filter then \mathcal{F}^{\ast} is an ideal (which we call the dual ideal of \mathcal{F} ) and if \mathcal{I} an ideal, then \mathcal{I}^{\ast} is a filter ( the dual filter of \mathcal{I} ). For A,B\subseteq\omega, define A\subseteq^{\ast}B ( A is an almost subset of B ) if A\setminus B is finite. For \mathcal{P\subseteq} \left[\omega\right]^{\omega} and A,B\subseteq\omega, we say that A\ is a pseudointersection of \mathcal{P} if it is almost contained in all elements of \mathcal{P}, while B is a pseudounion of \mathcal{P} if it almost contains every element of \mathcal{P}. We say \mathcal{C}\subseteq\left[\omega\right]^{\omega} is centered if the

Read the complete text