MathOverflow is a question and answer site for professional mathematicians. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Jensen's covering theorem states that if $0^\sharp$ doesn't exist, then every uncountable set of ordinals can be covered by a constructible set of the same cardinality.

Now consider the following (somewhat) dual statements:

  1. Every uncountable set of ordinals covers a constructible uncountable set of ordinals.
  2. Every uncountable set of ordinals covers a constructible set of ordinals of the same cardinality.

I have two questions:

  1. Does any of the above statements follow from the non-existence of $0^\sharp$?
  2. If the answer is "no", are they still known to be consistent (in order to avoid trivialities, we may assume that V=L doesn't hold)?
share|cite|improve this question
up vote 5 down vote accepted

The answer to the first question is no: You can add a club subset of a stationary subset of $\omega_1$ by forcing. The closure of any uncountable subset of the generic club is a club contained in the stationary set, so if we begin in $L$ with a stationary-costationary subset of $\omega_1$, and add a club through it, the extension contradicts both statements.

As for your second question, add a Cohen real to $L$. Suppose $A\subseteq{\rm ORD}$ and $|A|$ has uncountable cofinality in the extension. Since Cohen's forcing is countable, there is a condition $p$ that is in the generic and such that $A_p=\{\alpha\mid p$ forces $\alpha\in\dot A\}$ has the same size as $A$. Note that this is a constructible subset of $A$ in the ground model.

In fact, the same holds for uncountable sets of cofinality $\omega$: Suppose now that $A$ has countable cofinality, let $\alpha$ be its supremum, let $\gamma$ be least such that $A\cap\gamma$ is uncountable. For each $\beta$ between $\gamma$ and $\alpha$ there is a $p_\beta$ in the generic that decides a subset of $A\cap\beta$ that is in $L$ and has size $|A\cap\beta|$. There must be a $p$ that appears as $p_\beta$ unboundedly often, and $A_p$ is as wanted.

In summary: The extension of $L$ obtained by adding a Cohen real is a model of your two statements.

share|cite|improve this answer
Thanks, Andres. :) – Haim Jul 31 '11 at 14:57

Perhaps Magidor's covering lemma may be mentioned: if $0^\sharp$ does not exist and $A$ is a set of ordinals which is closed under the primitive recursive set functions, then $A$ is the union of countably many constructible sets. M. Magidor: Representing sets of ordinals..., Transactions of the AMS, 317(1990), 91-126.

share|cite|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

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