4
$\begingroup$

Working in $L$, suppose $L \models \kappa$ is a cardinal and $(\mathbb{P}, <) \in L_\kappa$. Let $\varphi(x)$ be a $\Sigma_1^1$ formula. Let $\tau \in L_\kappa$ be a $\mathbb{P}$-name for an element of ${}^\omega\omega$. Is "$1_\mathbb{P} \Vdash \varphi(\tau)$" absolute between $L_\kappa$ and $L$?

It seems that if $L_\kappa$ satisfies enough of set theory to prove the Mostowski Absoluteness and some basic facts about the forcing relation, then this would be true. For instance if $\kappa$ was inaccessible or even regular uncountable. For any arbitrary cardinal $\kappa$, does $L_\kappa$ satisfy enough set theory for the Mostowski absoluteness.

A related question is what are the known fragments of ZF that all $L_\kappa$ (or the fine structures $J_\kappa$) where $\kappa$ is a cardinal satisfy?

Thanks for any information that can be provided.

$\endgroup$
0

1 Answer 1

6
$\begingroup$

The answer is yes, and much more. If $\kappa$ is any uncountable cardinal and $\mathbb{P}$ is a notion of forcing in $L_\kappa$, then for any projective statement $\varphi$, the assertion $1_{\mathbb{P}}\Vdash\varphi$ is absolute between $L_\kappa$ and $L$. To see this, note first that $G\subset\mathbb{P}$ is $L_\kappa$-generic if and only if it is $L$-generic, since all subsets of $\mathbb{P}$ in $L$ are already in $L_\kappa$. Next, observe that all the nice $\mathbb{P}$-names for reals that are in $L$ are already in $L_\kappa$, since a nice name for a real is essentially a function from $\omega$ to the antichains of $\mathbb{P}$, and all such functions that are in $L$ are already in $L_\kappa$, since subsets of set appear in $L$ before the next cardinal stage. It follows that if $\tau$ is any $\mathbb{P}$-name for a real with $\tau\in L$, then we may find a nice name $\sigma\in L_\kappa$ such that $1\Vdash \tau=\sigma$. Consequently, $$\mathbb{R}^{L[G]}=\mathbb{R}^{L_\kappa[G]},$$ and so any projective statement forced over $L$ is equally forced over $L_\kappa$ by the same condition.

$\endgroup$
0

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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