0
$\begingroup$

Let $A$ be the free $\sigma$-algebra with $\omega_1$ free $\sigma$-generators, $X$ its Stone space, and $Ba(X)/M$ the algebra of Baire subsets of $X$ modulo meagre sets. Then $A$ is $\sigma$-isomorphic to $Ba(X)/M$ by the Loomis-Sikorski theorem.

Let $2^{\omega_1}$ be the Cantor cube of weight $\omega_1$ (i.e. the Stone space of the free Boolean algebra with $\omega_1$ free generators) and $Ba(2^{\omega_1})$ the $\sigma$-field of Baire subsets of $2^{\omega_1}$. Then it can be shown that $A$ is also $\sigma$-isomorphic to $Ba(2^{\omega_1})$.

Now, since $Ba(2^{\omega_1})$ is a $\sigma$-field of sets, it is $\sigma$-distributive. Hence, since $A$ is $\sigma$-isomorphic to $Ba(2^{\omega_1})$, $A$ is an example of an atomless $\sigma$-distributive $\sigma$-algebra.

My question: Since $A$ is also $\sigma$-isomorphic to $Ba(X)/M$, is $Ba(X)/M$ also $\sigma$-distributive? (I have a doubt.)

$\endgroup$
2
  • $\begingroup$ Baire subsets are the $\sigma$-algebra generated by the clopen subsets, correct? The first $B(X)$ should also be $\text{Ba}(X)$ ? $\endgroup$ Commented Aug 3, 2020 at 17:42
  • $\begingroup$ @HennoBrandsma Yes, Baire sets are elements of the sigma-algebra generated by the clopen sets. And also, yes, $B(X)$ should be read $Ba(X)$. I corrected. $\endgroup$
    – Beginner
    Commented Aug 3, 2020 at 17:52

1 Answer 1

0
$\begingroup$

Yes, it should be clear that two if two Boolean Algebras are isomorphic by an isomorphism that preserves all countable sups (i.e. $\sigma$-isomorphic), and one is $\sigma$-distributive, then the other one is too.

$\endgroup$
1
  • $\begingroup$ Thank you for confirming. $\endgroup$
    – Beginner
    Commented Aug 3, 2020 at 19:19

You must log in to answer this question.

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