Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

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

Jech defines a completion of a Boolean algebra $B$ to be a complete Boolean algebra $C$ such that $B$ is a dense subalgebra of $C$.

I am trying to prove that given two completions $C$ and $D$ of $B$, then the mapping $\pi: C \rightarrow D$ given by $\pi(c) = \sum^D \{ u \in B \ \vert \ u \le c \} $ is an isomorphism. I understand that $c \not= 0 \implies \pi(c) \not= 0$, and that given any $d \in D$, I can write it as $d = \sum^D \{ u \in B \ \vert \ u \le d \} $, but I am unsure how to proceed. Any help would be appreciated.

share|cite|improve this question
up vote 2 down vote accepted

I am afraid that it doesn't make sense that $\pi (c)\leq c$ because both elements live in different lattices $C$ and $D$, so they are not comparable. So it's better to denote differently both orders as $\leq_C$ and $\leq_D$.

So let $\pi (c)= \sum^D${$u\in B:u\leq_C c$}, and let $\rho: D\to C$ be such that $\rho(d)=\sum^C${$u\in B:u\leq_D d$}.

For any $c\in C$ and $u\in B$ we have $u\leq_C c\Rightarrow u\leq_D\pi(c)$, by the property of the supremum. Conversely, let $u\in B$ be such that $u\leq_D\pi(c)$. Then $u=u\wedge\pi(c)=u\wedge\sum^D${$b\in B:b\leq_C c$}$=\sum^D${$u\wedge b:b\leq_C c$}, since $D$ is a complete Boolean algebra, and so it has the general distributivity. Now, if $b\leq_C c$ then $u\wedge b\leq_Cu\wedge c\leq_C c$. Therefore $u\leq_C c$. We have shown that $u\leq_C c\Leftrightarrow u\leq_D\pi(c)$.

Therefore $c=\sum^C${$u\in B:u\leq_C c$}$=\sum^C${$u\in B:u\leq_D \pi(c)$} $=\rho(\pi(c))$.

Similarly $\pi(\rho(d))=d$ for each $d\in D$. So $\rho=\pi^{-1}$, and $\pi$ is an isomorphism.

share|cite|improve this answer

Edit: Ignore this answer. See Loronegro's.

Apparently I haven't done enough on Math.SE before, so I have to post this as an answer instead of a comment.

Hopefully this is helpful: first notice that $\pi c \le c$ for every $c\in C$ (the supremum of a bunch of things that are $\le c$ is also $\le c$). From this, you can deduce that, for every $u\in B$, $u\le c$ iff $u\le \pi c$. Therefore the sets $\{u\in B : u\le c\}$ and $\{u\in B : u\le \pi c\}$ are identical, so $\pi^{-1}\pi = 1$. By starting the argument with $\pi^{-1}c \le c$ instead of $\pi c \le c$, you can also conclude that $\pi\pi^{-1} = 1$.

You should feel cheated now, as I have avoided any mention of density. I guess proving that $\sum^C \{u\in B: u\le c\} = c$ requires the density of $B$.

share|cite|improve this answer
Also, hi Paul!! – Zach N Feb 5 '12 at 15:38
Thanks Zach. I basically needed $\pi(c) \le c$ which I stupidly did not see! See you on Wednesday, hopefully I can get to chapter 9 haha :p – Paul Slevin Feb 5 '12 at 15:49
You didn't see it because it is false! AHHH!!! See Loronegro's answer. – Zach N Feb 7 '12 at 22:19

Note that $\pi $ has an inverse, defined in the same way, $\pi^{-1}(d)=\sum^C\{u\in B: u\leq d\} $ and the maps are order-preserving.

share|cite|improve this answer
But how do I use density to show those 2 facts? – Paul Slevin Feb 5 '12 at 10:56
Well, I think I can see why each is order-preserving, but I am having some difficulty showing that $\pi \pi^{-1} = 1$. – Paul Slevin Feb 5 '12 at 13:04

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.