Universally meager sets in the Miller model and similar ones
Valentin Haberl, Piotr Szewczak, Lyubomyr Zdomskyy · 2025
Read Universally meager sets in the Miller model and similar ones by Valentin Haberl, Piotr Szewczak, Lyubomyr Zdomskyy (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-2512-15490.txt.
Introduction
Valentin Haberl, Piotr Szewczak, and Lyubomyr Zdomskyy bound the size of universally meager sets in certain set theory models in a 2025 paper. Universally meager sets of reals are small in a strong sense that survives many mappings. The question is how large such sets can be forced to remain in specific models of set theory.
Working in the Miller model and in a model built by Goldstern, Judah, and Shelah, the authors prove all universally meager sets have size at most the first uncountable cardinal. The same limits follow for Rothberger and Hurewicz spaces with no copy of the Cantor set. In the Miller model, strong measure zero sets obey the same bound.
Catalog record
| CREATOR | Valentin Haberl, Piotr Szewczak, Lyubomyr Zdomskyy |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 67 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2512-15490
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Universally meager sets in the Miller model and similar ones Thanks: The research of the first and the third authors was funded in whole by the Austrian Science Fund (FWF) [10.55776/I5930 and 10.55776/PAT5730424]. The research of the second author was funded by the National Science Center, Poland Weave-UNISONO call in the Weave programme Project: Set-theoretic aspects of topological selections 2021/03/Y/ST1/00122 Valentin Haberl, Piotr Szewczak, Lyubomyr Zdomskyy Abstract . We work in the realm of sets of reals. We prove that in the Miller model and in a model constructed by Goldstern–Judah–Shelah all universally meager sets have size at most ω 1 \omega_{1} . Some relations between combinatorial covering properties in these models allow to obtain the same limitations for sizes of Rothberger spaces and Hurewicz spaces with no homeomorphic copy of the Cantor set inside. It follows from our results that the existence of a strong measure zero set of size ω 2 \omega_{2} does not imply the existence of a Rothberger space of size ω 2 \omega_{2} . We also prove that in the Miller model all strong measure zero sets have size at most ω 1 \omega_{1} . 2020 Mathematics Subject Classification Primary: 03E35, 54D20, 54A35. Secondary: 03E75. 1. Introduction Definition 1.1 ( [ 40 ] ) . A subspace X\subseteq 2^{\omega} is universally meager , if every Borel isomorphic image of X is meager in {2^{\omega}} . The definition of universally meager subspaces was introduced by Zakrzewski [ 40 ] and shown by him to be equivalent [ 40 ] with an earlier studied by Grzegorek notion of absolutely of the first category sets [ 16 , 17 , 18 ] . Universally meager subspaces were also later studied by Zakrzewski who proved in [ 41 ] several characterizations thereof which are crucial for us (see Lemma 3.1 ). In some models of set theory there are limitations for sizes of universally meager subspaces. This is the case, e.g., in the Cohen model (a forcing extension of a ground model of CH using a finite support iteration (equivalently, product) of the Cohen forcing, of length \omega_{2} ), where all universally meager subspaces have size at most \omega_{1} , see the discussion after Problem 6.5 for more details. In our work we consider these limitations in the Miller model (a forcing extension of a ground model of GCH using a countable support iteration forcing of length \omega_{2} , introduced by Miller [ 29 ] ) and in the model constructed by Goldstern–Judah–Shelah [ 13 , the proof of Theorem 0.16] . Theorem 1.2 . In the Miller model and in the Goldstern–Judah–Shelah model, every universally meager subspace has size at most \omega_{1} . The proof of Theorem 1.2 is based on the fact that forcing notions involved satisfy the property (\dagger) introduced in [ 32 ] . It turns out that also the Cohen forcing shares the (\dagger) property. Our results mentioned above concern universally meager subspaces but the origin of our investigations is related to other properties of topological spaces and Theorem 1.2 helps to solve more problems. A set X\subseteq 2^{\omega} has strong measure zero ( SMZ , in short) if for every sequence {\langle\,\epsilon_{n}:n\in\omega\,\rangle} of positive reals there exists a sequence {\langle\,a_{n}:n\in\omega\,\rangle} of “centers” such that the family \{\,\mathrm{B}(a_{n},\epsilon_{n}):n\in\omega\,\} covers X , where \mathrm{B}(a,\epsilon):=\{\,x\in{2^{\omega}}:\rho(x,a)<\epsilon\,\} and \rho is any metric generating the standard topology on 2^{\omega} . This notion was introduced by Borel who conjectured [ 11 ] that only countable sets have this property. The Borel conjecture has been consistently refuted by Sierpiński [ 36 ] in 1928, and proved to be consistent in 1976 by Laver [ 25 ] , and since then appears in different contexts as well as serves as a blueprint for similar conjectures in set theory of the reals, see, e.g., [ 15 ] . Rothberger noticed [ 33 ] , that being a strong measure set might not be a topological property and he modified it to the following topological covering property. By space we mean a topological space homeomorphic with a subspace of the Cantor cube {2^{\omega}} . A space X is Rothberger if for any sequence {\langle\,\mathscr{U}_{n}:n\in\omega\,\rangle} of open covers of X , there are sets U_{n}\in\mathscr{U}_{n} for n\in\omega , such that the family \{\,U_{n}:n\in\omega\,\} covers X . It is easy to see that each Rothberger subset of {2^{\omega}} has SMZ. The first inspiration to this paper comes from our work [ 19 ] on concentrated and \gamma -spaces. A space X is concentrated on a countable set D\subseteq X , if the set X\setminus U is countable for any open set U\subseteq X containing D . A space X is concentrated if it is uncountable and concentrated on some of its countable subsets. Bartoszyński–Halbeisen conjectured [ 1 , Proposition 3.2] that there exists a concentrated space of size \omega_{2} in the Miller model. This conjecture has been refuted [ 19 ] , where we have also showed that any \gamma -space has size at most \omega_{1} in the Miller model (see [ 12 ] for the definition of \gamma -spaces). Since the class of all Rothberger spaces of {2^{\omega}} contains all concentrated and \gamma -spaces [ 12 ] , we have asked in our previous work [ 19 ] , the following question. Problem 1.3 ( [ 19 , Problem 4.2] ) . Does every Rothberger space have size at most \omega_{1} , in the Miller model? We get the following result which also solves Problem 1.3 . Theorem 1.4 . In the Miller model and in the Goldstern–Judah–Shelah model, every Rothberger space has size at most \omega_{1} . Now we explain how Theorem 1.2 delivers the above results. To this end we need an auxiliary notion of Hurewicz spaces [ 20 ] . We shall not work directly with Hurewicz spaces, but rather cite their known properties, hence we refer the reader to a work of Tsaban [ 38 ] for more information of such spaces. For definitions and properties of cardinal characteristics of the continuum used below, we refer the reader to a work of Blass [ 10 ] . There are four main steps to prove Theorem 1.4 . • The inequality \mathfrak{u}<\mathfrak{g} holds in the Miller model and in the Goldstern–Judah–Shelah model. • In models of \mathfrak{u}<\mathfrak{g} , every Rothberger space is Hurewicz and totally imperfect , i.e., does not contain a homeomorphic copy of {2^{\omega}} inside [ 42 ] ; • Every Hurewicz totally imperfect subspace of 2^{\omega} is universally meager [ 40 ] . • Apply Theorem 1.2 . The following diagram presents relations between considered here properties for subspaces of {2^{\omega}} (only one implication requests some additional assumptions beyond ZFC and it holds, e.g., when \mathfrak{u}<\mathfrak{g} ). concentrated {\lx@inpgf@ignorespaces\begin{matrix}\text{Hurewicz and}\\ \text{totally imperfect}\end{matrix}} {\lx@inpgf@ignorespaces\begin{matrix}\text{universally}\\ \text{meager}\end{matrix}} Rothberger {\lx@inpgf@ignorespaces\text{property }\gamma} strong measure zero \scriptstyle{\lx@inpgf@ignorespaces\mathfrak{u}<\mathfrak{g}} Some additional properties of the Goldstern–Judah–Shelah model allows to prove that, in this model, a space is Rothberger if and only if it is Hurewicz and totally imperfect (see Theorem 4.12 ). In the above model, we have \omega_{2}=\mathfrak{c}=\mathfrak{d}>\mathfrak{b} and any union of less than \mathfrak{c} many SMZ sets has SMZ (in particular each set of size less than \mathfrak{c} has SMZ). If \mathfrak{c}=\mathfrak{d} and each set of