Take the 2-minute tour ×
MathOverflow is a question and answer site for professional mathematicians. It's 100% free, no registration required.

A p.o. $\mathcal{P}$ is $(<\kappa)$-closure, if for every decreasing sequence of $<\kappa$ conditions in the forcing $p_0 \geq p_1 \geq \cdots$, there is a condition that is below all of them. Prikry forcing $\mathcal{P}$ is the set of pairs $(s,A)$ where s is a finite subset of a fixed measurable cardinal $\kappa$, and $A$ is an element of a fixed filter $D$ on $\kappa$. A condition $(s,A)$ is stronger than $(t, B)$ if $t$ is an initial segment of $s$ and $A \cup (s-t) \subseteq B$. Why doesn't Prikry forcing have this property? Could someone help me out with this?

share|improve this question
    
Noah's answer is the most direct, hence best, but let me point out another way to see this. A $<\kappa$-closed forcing adds no new $<\kappa$-sequences of ordinals. But Prikry forcing adds (not surprisingly) a Prikry sequence, which is a new $\omega$-sequence of ordinals. (If you take the natural name for the Prikry sequence, apply the proof that it wouldn't be new if the forcing were $<\kappa$-closed, and make the simplest choices wherever an arbitrary choice is needed in the proof, you'd recover Noah's counterexample.) –  Andreas Blass Jun 15 '12 at 12:49
add comment

1 Answer

up vote 2 down vote accepted

Consider the conditions $(s_i, \kappa)$ where $s_i=\{0, . . . , i\}$ for $i\in\omega$. This is a sequence of $\omega$-many conditions, and clearly $(s_{i+1}, \kappa)\le (s_i, \kappa)$, but there is obviously no single condition below all the $(s_i, \kappa)$: were $(p, A)$ such a condition, then $p$ would have to be infinite, which contradicts the definition of condition.

Note that we could alter the forcing to consider conditions of the form $(p, A)$, with $A$ as above, and $p$ a subset of $\kappa$ of size $<\kappa$. This forcing would then be $<\kappa$-closed (depending, I guess, on the filter $D$), but this is a very different forcing.

share|improve this answer
    
Thank you for your answer. –  um Haitham Jun 15 '12 at 11:04
    
You can accept an answer by clicking on the check mark next to the answer. –  Noah S Jun 21 '12 at 2:15
add comment

Your Answer

 
discard

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.