3
$\begingroup$

This is a continuation of my previous question. Recall that a subset $A \subseteq {}^\omega\omega$ is analytic if it is the continuous image of the Baire space. I would like to know if there exist two models $N \subseteq M$ such that:

  1. $M \models \mathsf{ZFC}$ (or just $M \models \mathsf{ZF} + \text{There exists a non-principal ultrafilter}$).

  2. $N \models \mathsf{ZF} + \text{Every subset of reals is analytic}$.

  3. $({}^\omega\omega)^M = ({}^\omega\omega)^N$.

From the comments of previous posts, it's worth noting that:

  1. Under $\mathsf{ZF} + \mathsf{DC}$, one can construct the universal analytic set and prove that it is not analytic. Thus, $N \not\models \mathsf{DC}$.

  2. A (somewhat trivial) example of a model in which every subset of reals is analytic is the Feferman-Levy model, where the reals is a countable union of countable sets.

$\endgroup$
10
  • 3
    $\begingroup$ Maybe I'm having a silly moment but I don't think "Every set is analytic" is consistent with $\mathsf{ZF}$ in the first place. We don't need choice to get a surjection $h$ from Baire space to the set of continuous maps on Baire space; now consider the set $\{x: x\not\in h(x)\}$. (Re: Feferman-Levy, note that in that model $\mathbb{R}$ isn't a countable union of "uniformly countable" sets, and without uniform countability I don't see why in FL every set should be analytic. But maybe I'm missing something.) $\endgroup$ Commented Jan 3, 2022 at 2:36
  • $\begingroup$ @NoahSchweber what is "uniformly countable"? $\endgroup$ Commented Jan 3, 2022 at 3:14
  • $\begingroup$ Without choice, for a family of sets $\mathscr{A}=(A_i)_{i\in I}$ the statements "Each $A_i\in\mathscr{A}$ is countable" and "There is a function assigning each $A_i$ to a surjection $\omega\rightarrow A_i$" are not automatically equivalent; the latter is what I mean by uniform countability. In the Feferman-Levy model $\mathbb{R}$ is a countable union of countable sets, but no model of $\mathsf{ZF}$ can have $\mathbb{R}$ be a countable union of uniformly countable sets. This old MSE answer of mine may be relevant. $\endgroup$ Commented Jan 3, 2022 at 3:16
  • $\begingroup$ I see, but I'm not sure why the following alternative approach fails: I believe its a theorem of $\mathsf{ZF}$ that every analytic set can be seen as the Suslin operation applied to a Suslin scheme of closed sets (see my linked question, but my Suslin operation is different). I also believe that one can see from the operation that this implies countable union of countable set is analytic (as singletons are closed). Therefore, every subset of $\mathbb{R}$ is analytic. $\endgroup$ Commented Jan 3, 2022 at 3:26
  • $\begingroup$ "I also believe that one can see from the operation that this implies countable union of countable set is analytic" That looks dubious to me (in $\mathsf{ZF}$ alone anyways); I'd try to prove it in detail. $\endgroup$ Commented Jan 3, 2022 at 3:28

1 Answer 1

8
$\begingroup$

In fact, the principle "Every set is analytic" is not consistent with $\mathsf{ZF}$ in the first place. We don't need choice to get a surjection $h$ from Baire space to the set of continuous maps on Baire space. But once we have such an $h$, the "diagonalizing" set $\{x: x\not\in h(x)\}$ can't be analytic.


Let me show how to get such an $h$ in $\mathsf{ZF}$ alone. Working in $\mathsf{ZF}$, let $C$ be the set of continuous maps $\omega^\omega\rightarrow\omega^\omega$. To each $\gamma\in C$ we assign the set $$S(\gamma)=\{(\sigma,\tau)\in\omega^{<\omega}\times\omega^{<\omega}: \forall f\succ \sigma(\gamma(f)\succ\tau)\}.$$ By your favorite coding mechanism we can identify $S(\gamma)$ with some $\hat{S}(\gamma)\in\omega^\omega$. But $\hat{S}(\gamma)=\hat{S}(\gamma')$ implies $\gamma=\gamma'$ for continuous $\gamma,\gamma'$, so in fact $\hat{S}$ gives an injection from $C$ to $\omega^\omega$.

Now turning a surjection into an injection without choice is hard, but the converse is trivial: for $x\in\omega^\omega$, let $h(r)=\hat{S}^{-1}(r)$ if $r\in ran(\hat{S})$, and let $h(r)$ be the always-zero map otherwise.

$\endgroup$

You must log in to answer this question.

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