Does one always need some choice to show the existence of nonprincipal ultrafilters?

151 Views Asked by At

The last sentence of the section "Types and existence of ultrafilters" of Wikipedia's article on ultrafilter (current revision) says:

In ZF without the axiom of choice, it is possible that every ultrafilter is principal.{see p.316, [Halbeisen, L.J.] "Combinatorial Set Theory", Springer 2012}

This statement is incorrect isn't it? Shouldn't "ultrafilter" here be replaced by "ultrafilter on $\mathscr{P}(\omega)$ (the power set of the set of natural numbers)"?

I think that for some Boolean algebras, one can show the existence of a nonprincipal ultrafilter without assuming the axiom of choice (or the ultrafilter lemma or the Boolean prime ideal theorem). Let $B$ be the set of all finite and cofinite subsets of the set of natural numbers, and let $U$ be the set of all cofinite subsets of the set of natural numbers. Then $U$ is a nonprincipal ultrafilter on the Boolean algebra $B$, isn't it?

I would also like to ask the following more general question: for what kind of Boolean algebra, can one show the existence of a nonprincipal ultrafilter without assuming some choice?

EDIT: I am not asking whether there is an infinite Boolean algebra with a unique ultrafilter (unless "nonprincipal ultrafilter that can be found without choice" and "unique filter" should mean the same thing, which I fail to see). I am first asking whether my example is a counterexample to the statement from Wikipedia. Assuming that I am right about this first point, I am then asking in what cases you can find nonprincipal ultrafilters without using choice. I actually don't know what $\mathcal{P}(\omega)/\text{fin}$ denotes in the suggested question. In any case, the answer given there is only an example and cannot be an answer to my question. I would appreciate if you could explain how my question is a duplicate or reopen my question so that people could provide answers that I would be able to understand.