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


I started to read the chapter 31 in Jechs book about proper forcing. Unfortunately it is written in a rather sketchy way and I do have some issues in proving a lemma about two equivalent definitions of $(M,P)$-genericity. Jech gives the following definition:

Let $(P,<)$ be a fixed notion of forcing, $\lambda > 2^{|P|}$ and let $M \prec (H_{\lambda}, \in, <, ..)$, where $H_{\lambda}$ is the set of all sets whose transitive closure has size less than $\lambda$, and $' <'$ is a well-ordering of $H_{\lambda}$. Then a condition $q \in P$ is $(M,P)$-generic if for every maximal antichain $A \in M$, the set $A \cap M$ is predense below $q$.

He moves on to state a small lemma, saying that its proof is a routine verification (that leads us to the assertion that I am not an old hand):

Assume the same situation as in our definition above. Then the following are equivalent:

(i) $q$ is $(M,P)$-generic.

(ii) $ q \Vdash \dot{G} \cap M$ is a filter on $P$ generic over $M$

The (ii) $\Rightarrow$ (i) direction is quite easy but I couldn't find a proof for (i) $\Rightarrow$ (ii).

Thank you

share|cite|improve this question
up vote 4 down vote accepted

Let $G$ be a $(V,P)$-generic filter containing $q$. We want to show that $G \cap M$ is a filter on $P$ generic over $M$. $G \cap M$ will be countable so it typically won't be closed upwards in $P$, the relevant point is that it's closed upwards in $P \cap M$. So it's not hard to see that it's a filter on $P \cap M$. Next we need to see it's generic over $M$.

Let $A \in M$ be a maximal antichain in $P$. Let $p$ be the sole element of $A \cap G$. Let $r$ extend both $p$ and $q$, which is possible because they're both in $G$ and $G$'s a filter. By (i), $A \cap M$ is predense below $q$, and $r \leq q$, so there's some $p' \in A \cap M$ compatible with $r$. But since $A$ is a maximal antichain, there's precisely one element of $A$ compatible with $r$, and we already know $p \in A$ is compatible with $r$ (in fact it's above $r$), so $p' = p$. Therefore $p \in A \cap G \cap M$, meaning $G \cap M$ meets $A$. Since $A$ was arbitrary, $G \cap M$ meets every maximal antichain belonging to $M$, hence it's generic over $M$.

Just as genericity of a filter can be defined in terms of dense sets, open dense sets, predense sets, or maximal antichains, it can so be done for genericity of a condition. See Lemma 1.2 in my notes below; it gives three equivalent conditions for $(M,P)$-genericity. The first and third are just like your (i) and (ii) but with antichains replaced with dense sets. The second one is also in terms of dense sets but slightly different from your (i) and (ii), and it's another useful way of looking at $(M,P)$-genericity.

share|cite|improve this answer
Nice notes! Looking forward to meeting you someday (I don't remember meeting you before, have we?), it looks like we have similar interests. – Andrés E. Caicedo Oct 27 '10 at 20:57
Thank you very much! – Stefan Hoffelner Oct 27 '10 at 21:42
oktan, welcome! Andres, thanks! I don't believe we've met, but it seems we're related in a sense, John Steel is my advisor here at Berkeley. I'm interested in forcing axioms and infinitary combinatorics, so we certainly do have common interests, although I have much much more to learn. – Amit Kumar Gupta Oct 28 '10 at 6:38

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.