3
$\begingroup$

Let $\kappa$ be a cardinal number, define $\textsf{M}(\kappa)$ and $\textsf{BW}(\kappa)$ as follows:

  • $\textsf{M}(\kappa)$ : For every sequence $(f_{n}:\kappa\to \mathbb{R})_{n\in\mathbb{N}}$ of real-valued function defined on $\kappa$ there is a subsequence $(f_{n}:\kappa\to \mathbb{R})_{n\in A}$ such that for each $\alpha\in\kappa$ the sequence $(f_{n}(\alpha))_{n\in A}$ is eventually monotonic.

  • $\textsf{BW}(\kappa)$ : For every sequence $(f_{n}:\kappa\to \mathbb{R})_{n\in\mathbb{N}}$ of pointwise bounded real-valued functions defined on $\kappa$ there is a subsequence $(f_{n}:\kappa\to \mathbb{R})_{n\in A}$ which is pointwise convergent.

Proposition: For each cardinal number $\kappa$, we have that $\textsf{M}(\kappa) \Longrightarrow \textsf{BW}(\kappa)$

Proof: Let $(f_{n}:\kappa\to \mathbb{R})_{n\in\mathbb{N}}$ be a sequence of pointwise bounded real-valued functions defined on $\kappa$ (i.e., for each $n\in\mathbb{N}$, the set $\{f_{n}(\alpha):\alpha\in\kappa\}$ is bounded on $\mathbb{R}$). As $\textsf{M}(\kappa)$ holds, there exists a subsequence $(f_{n}:\kappa\to \mathbb{R})_{n\in A}$ such that for each $\alpha\in\kappa$ the sequence $(f_{n}(\alpha))_{n\in A}$ is eventually monotonic. Then, for each $\alpha\in\kappa$, there is an infinite subset $A_{\alpha}$ of $A$ such that $(f_{n}:\kappa\to \mathbb{R})_{n\in A_{\alpha}}$ is monotonic, consider $A^{\prime}=\bigcap_{\alpha<\kappa}A_{\alpha}$.

Can I conclude that $A^{\prime}$ is infinite and that $(f_{n}(\alpha))_{n\in A^{\prime}}$ is pointwise convergent (in fact, only I need that $(f_{n}(\alpha))_{n\in A^{\prime}}$ is bounded)? In case my reasoning is wrong, can someone give me an idea of how to fix it?

Thanks a lot

$\endgroup$
5
  • 3
    $\begingroup$ It seems unlikely that $A’$ would be infinite. $\endgroup$ Commented Mar 19, 2020 at 11:15
  • 2
    $\begingroup$ But since you only need pointwise convergence, does the statement not follow, because, or any $\alpha\in\kappa$, the sequence $(f_n(\alpha))_{n\in A}$ is eventually monotonic and therefore convergent (since it is also bounded)? $\endgroup$ Commented Mar 19, 2020 at 13:05
  • 1
    $\begingroup$ I think your notion of "pointwise bounded" is wrong -- don't you want $\{f_n(\alpha) \,:\, n \in \mathbb N\}$ to be bounded for any given $\alpha$? If you fix $n$ and let $\alpha$ vary, as in your definition, then the constant functions $f_n(\alpha) = n$ satisfy your hypotheses, and BW$(\kappa)$ is false for every $\kappa > 0$. $\endgroup$
    – Will Brian
    Commented Mar 19, 2020 at 13:27
  • 2
    $\begingroup$ Also, unless I'm misunderstanding your definition, I think $\mathsf{BW}(\kappa)$ is true if and only if $\kappa < \mathfrak{s}$. This follows from Theorem 3.2 in Andreas Blass' article in the Handbook of Set Theory (link: math.lsa.umich.edu/~ablass/hbk.pdf). $\endgroup$
    – Will Brian
    Commented Mar 19, 2020 at 13:29
  • 2
    $\begingroup$ One more thing: I think $\mathsf{M}(\kappa)$ is true if and only if $\kappa < \mathrm{min}\{\mathfrak{b},\mathfrak{s}\}$. This follows from Theorem 3.5 in Andreas' article. $\endgroup$
    – Will Brian
    Commented Mar 19, 2020 at 13:31

0

You must log in to answer this question.