4
$\begingroup$

I am dealing with the basics of Radin forcing but there are some formal details which I was not able to keep with. For example, given $u$ a measure sequence with length an uncountable regular cardinal $\lambda<\kappa(u)$ it is known that forcing with the Radin forcing $\mathbb{P}_u$ changes the cofinality of $\kappa(u)$ to $\lambda$. However I did not find any reference in which this fact was proved in detail so I was wondering that someone here could help me either with a good reference or with a detailed explanation.

Thanks in advance.

$\endgroup$
1
  • $\begingroup$ If I understand your notation correctly, the special case of Radin forcing that you're asking about is known as Magidor forcing. $\endgroup$ Commented Jun 15, 2022 at 14:07

1 Answer 1

4
$\begingroup$

By induction on $2 \leq \delta=$length of (u)$< \kappa,$ one can show that $otp(C)= \omega^{\delta-1}$, if $\delta < \omega$ and $otp(C)= \omega^{\delta}$ if $\omega \leq \delta < \kappa$ where $C$ is the Radin club, in particular if $\delta$ is regular uncountable, then $otp(C)=\delta$. Further one can manage to not add any new subsets to $\delta$. Now the result should be clear.

Here is the basic idea:

The case $l(u)=2$ is just the case of Prikry forcing, and we have $otp(C)=\omega.$

Suppose $\delta=\gamma+1$ is a successor ordinal and it holds for $\gamma$. Then note that forcing with $\mathbb{P}_u$ adds an $\omega$-sequence $(w_i: i < \omega)$ of measure sequences each of them of length $\gamma.$ Now use the factorization of the forcing and the induction hypothesis to get the result for $\delta$.

For limit $\delta < \kappa,$ consider the set $A=\{ w \in U_\infty: l(w) < \delta \}\in \bigcap_{0<\alpha < \delta}u(\alpha)$, and show, using the induction hypothesis, that below $(u, A),$ the club filter $C$ is forced to have order type $\omega^\delta=sup_{i<\delta}\omega^i.$


Radin forcing changes cofinalities below $\kappa:$

Radin forcing changes cofinalities below $\kappa$ too, provided that $length (u)>2$. This is clear, because of the following: Let $(u_i: i<\theta)$ be the measure sequence added by generic filter such that the sequence $(\kappa_i=\kappa(u_i): i < \theta)$ is increasing. Recall that $C=\{ \kappa_i: i \}$ is the Radin club. We can easily manage all $\kappa_i$'s be measurable in $V,$ the ground model: just note that the set $A'=\{w \in U_\infty \cap V_{\kappa(u)}: \kappa(w) $ is measurable$ \}\in u(\alpha), 0 < \alpha < length (u)$, and it suffices to force below $(u, A')$.

As $length (u)> 2,$ so $\theta=otp(C) \geq \omega^2$, in particular all $\kappa_i, i \leq \omega^2$ are singular in the extension with cofinality $\omega$.

In fact it is possible to say much more: First assume $length (u)< \kappa$, so $\theta=\omega^{lenght(u)}$. Then we can also manage $\kappa_0=min(C)> \theta,$ so clearly each limit point of the Radin club is singular in the extension.

Again more is possible to say, if $length (u)=\kappa^+,$ then $C$ has order type $\kappa,$ and by a theorem of Mitchell, $\kappa$ remains inaccessible in the extension, but if we force below $(u, A''),$ where $A''=\{ w \in U_\infty \cap V_{\kappa(u)}: length(w) < \kappa(w)^+ \}$, then all limit points of the Radin club will be singular in the extension, so that $\kappa$ becomes inaccessible not Mahlo in the extension.

$\endgroup$
3
  • $\begingroup$ thank you for your answer! Anyway, could you give the details of the proof of $otp(C)=\omega^\delta$? Thank you once again! $\endgroup$
    – Ergonvi
    Commented Dec 20, 2016 at 8:31
  • 1
    $\begingroup$ I have also read that Radin forcing changes cofinalities below $\kappa$. This seems natural to me, but I am not able to keep with the right argument to see it. Please, could you give me some indications? Thank you! $\endgroup$
    – Ergonvi
    Commented Dec 22, 2016 at 9:55
  • 1
    $\begingroup$ Yes, it is true. I added some details regarding this point. $\endgroup$ Commented Dec 24, 2016 at 4:31

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .