(This question assumes familiarity with combinatorial cardinal characteristics of the continnum.)

Let $[\mathbb{N}]^\infty$ be the family of infinite subsets of $\mathbb{N}$, partially ordered by $\subseteq^*$, where $a\subseteq^* b$ means $a\setminus b$ is finite.

Let $\mathfrak{ht}$ (sometimes called $\operatorname{Depth}^+([\mathbb{N}]^\infty)$) be the minimal cardinal number $\kappa$ such that there is no $\subset^*$-decreasing $\kappa$-sequence in $[\mathbb{N}]^\infty$. Then $\mathfrak{t}<\mathfrak{ht}\le\mathfrak{c}^+$.

A classic result from Kunen's thesis asserts that, when adding $\kappa\ge\aleph_1$ Cohen reals to a model of CH, we obtain $\mathfrak{ht}=\aleph_2$.

Open-ended question. Can the hypothesis $\mathfrak{b}<\mathfrak{ht}$ be expressed using cardinal characteristics other than $\mathfrak{ht}$?

It is easy to see that "$\mathfrak{t}=\mathfrak{b}$ or $\mathfrak{b}<\mathfrak{d}$" implies "$\mathfrak{b}<\mathfrak{ht}$".

Question 1. Is it consistent that "$\aleph_1=\mathfrak{t}<\mathfrak{b}=\mathfrak{c}=\aleph_2<\mathfrak{ht}$"?

In the Laver model, we have $\aleph_1=\mathfrak{t}<\mathfrak{ht}=\mathfrak{b}=\mathfrak{c}=\aleph_2$.

Question 2. What is the value of $\mathfrak{ht}$ in the Hechler and Mathias models?

Update: Will Brian answers Question 1 in the positive below, in the Hechler model (thus also answering Question 2 for Hechler). I thought that $\mathfrak{b}<\mathfrak{ht}$ implies there is a nontrivial set of reals satisfying the selection principle $\operatorname{S}_1(\Gamma,\Gamma)$, by the the linked paper. But Brian's comments make it clear that I oversimplified the question for this purpose. I opened a new question that fits better the intended application.

  • 1
    $\begingroup$ A model for Q1 with arbitrarily large continuum can be obtained by adding $\aleph_1$ random reals to a model of MA plus not CH, $\endgroup$ – Ashutosh May 10 '16 at 15:48

In the Hechler model, $\aleph_1 = \mathfrak{t} < \mathfrak{b} = \mathfrak{c} = \aleph_2 < \mathfrak{ht}$.

(By "the Hechler model" I mean the result of a length-$\omega_2$ finite support iteration of the forcing to adjoin a dominating real.)

That $\aleph_1 = \mathfrak{t} < \mathfrak{b} = \mathfrak{c} = \aleph_2$ is discussed in Section 11.6 of Blass's handbook article. That $\mathfrak{ht} > \mathfrak{c}$ follows from Theorem 4.1 in

J. Baumgartner and P. Dordal, "Adjoining dominating functions," Journal of Symbolic Logic vol. 50 (1985), pp. 94 - 101, available here.

Interestingly, Baumgartner and Dordal show that, while there are $\subset^*$-decreasing sequences of length $\omega_2$ in this model, none of them are maximal -- all maximal sequences have cofinality $\omega_1$.

This answers question 1 and the first part of question 2. I don't know what $\mathfrak{ht}$ is in the Mathias model, and I also don't know the answer to your open-ended question (though I think it's an interesting one).

  • $\begingroup$ I'm missing something: Take a nonmaximal decreasing sequence of length $\mathfrak{c}$. Then there is a pseudointersection, right? So, on top of this pseudointersection you can build a tower of height $\mathfrak{t}$, and end up with a tower of height continuum. $\endgroup$ – Boaz Tsaban May 10 '16 at 17:04
  • $\begingroup$ @BoazTsaban: Yes, that construction would work. The resulting tower would have cardinality $\mathfrak{c}$, but its cofinality would still have to be $\omega_1$ (your maximal $\subset^*$-descreasing sequence might have order type $\omega_2+\omega_1$, for example). $\endgroup$ – Will Brian May 10 '16 at 17:23
  • $\begingroup$ @BoazTsaban: Perhaps it's confusing for me to use "$\mathfrak{c}$" when referring to an order type, not a cardinality. I'll edit accordingly. $\endgroup$ – Will Brian May 10 '16 at 17:26

Your Answer

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Not the answer you're looking for? Browse other questions tagged or ask your own question.