1. In the proof of Pospisil's Theorem (theorem 7.6) that there are $2^{2^\kappa}$ uniform ultrafilters on $\kappa \geq \omega$, the author writes : Let $\mathcal{A}$ be an independent family of subsets of $\kappa$. For every function $f : \mathcal{A} \to \{0,1\}$, consider this family of subsets for $\kappa$ : \begin{align} G_f = \{X : |\kappa - X| < \kappa\} \cup \{X : f(X) = 1\} \cup \{\kappa - X : f(X) = 0\}. \end{align} The family $G_f$ has the f.i.p. so there exists an ultrafilter $D_f$ etc etc... Why do we have to include $\{X : |\kappa - X| < \kappa\}$ in $G_f$?
  2. In the proof of theorem 7.8 that CH implies the existence of a Ramsey ultrafilter, the author writes : Let $\mathcal{A}_\alpha$, $\alpha < \omega_1$ enumerate all partitions of $\omega$. We will construct an $\omega_1$ sequence of infinite subsets of $\omega$ as follows : Given $X_\alpha$, let $X_{\alpha+1} \subset X_\alpha$ be such that either $X_{a+1} \subset A$ for some $A \in \mathcal{A}_\alpha$, or that $|X_{\alpha+1} \cap A| \leq 1$ for all $A \in \mathcal{A}_\alpha$. If $\alpha$ limit, let $X_\alpha$ be such that $X_\alpha - X_\beta$ is finite for all $\beta < \alpha$. Jech notes that such $X_\alpha$ exists because $\alpha$ is countable. Why is that though?
  3. Lastly. At some point (page 79) he mentions that : If $h : B \to C$ is a one-to-one mapping (B, C are Boolean algebras) such that $u \leq v$ if and only if $h(u) \leq h(v)$, then $h$ is an isomorphism. I don't see why h is onto though. It is indeed a homomorphism, but isn't it trivial to find homomorphisms that respect $\leq$ that are not onto. Am I missing something?
  • 2
    $\begingroup$ You copied wrong, in $G_f$ the third part of the union is actually $$\{\kappa\setminus X: f(X)=0\}$$. $\endgroup$ – Asaf Karagila Sep 3 '12 at 12:56
  1. Since we want to generate a uniform ultrafilter (i.e., $|X| = \kappa$ for all $X \in U$), we must ensure that it contains all co-$<\kappa$ sets, which is exactly what this condition guarantees.
  2. As $\{ X_\beta \}_{\beta < \alpha}$ is decreasing, we need only ensure that $X_\alpha \setminus X_\beta$ is finite for a cofinal sequence of $\beta < \alpha$. As $\alpha$ is countable, it in particular has countable cofinality, so let $\{ \beta_n \}_{n \in \omega}$ be an increasing cofinal sequence in $\alpha$. For each $n < \omega$ pick some $i_n \in A_{\beta_n}$, and ensure that $i_{n+1} > i_n$. Then $X_\alpha = \{ i_n : n \in \omega \}$ is as desired. (This is related to the so-called pseudointersection number $\mathfrak{p}$.)
  3. Jech actually says that a one-to-one homomorphism of $B$ onto $C$ is called an isomorphism, so I guess you are missing the onto part of the definition.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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