I've been going through Hrbacek and Jech's Introduction to Set theory, 3rd edition, on my own. There's an exercise which states:

Let $\kappa$ be a limit cardinal, and $\lambda<\kappa$ be a regular infinite cardinal. Show that there exists an increasing sequence $\langle\alpha_\nu\;|\; \nu<cf(\kappa)\rangle$ of cardinals such that $\lim_{\nu\to cf(\kappa)} \alpha_\nu=\kappa$ and $cf(\alpha_\nu)=\lambda$ for all $\nu$.

I think I proved the opposite, $\kappa=\aleph_\omega$ is a counter example, since all the cardinals below it are successor cardinals, besides $\aleph_0$, and successor cardinals are regular.

Is my counterexample correct?

In detail: Consider $\kappa=\aleph_\omega$, a limit cardinal. It has cofinality $\omega$. Consider an increasing sequence of cardinals, $\langle\alpha_n \;|\; n<\omega\rangle$ with limit $\kappa$. $\aleph_0$ is regular, and every cardinal between $\aleph_0$ and $\aleph_\omega$ is a successor cardinal, and therefore regular. Hence for all $n$, $cf(\alpha_n)=\alpha_n$, which means there cannot be a single $\lambda$ equal to the cofinality of every element, the sequence is increasing. $\lambda=\aleph_1<\kappa$ is a concrete example of a regular infinite cardinal for which this statement fails.


  • $\begingroup$ You get a vote for finding a logical error in a textbook, even if it may only be a typo. $\endgroup$ Aug 22, 2017 at 18:14

1 Answer 1


You are correct. The statement should assume that $\lambda<cf(\kappa)$, instead of that $\lambda<\kappa$. This fixes your counterexample, since for $\kappa=\aleph_\omega$ there is no infinite regular cardinal $\lambda<cf(\kappa)$.

  • $\begingroup$ This might be a typo in the book...............+1 $\endgroup$ Aug 22, 2017 at 18:13
  • $\begingroup$ I thought it might be a typo, but I wasn't sure what the typo would be. Thanks. $\endgroup$ Aug 22, 2017 at 18:50

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.