3
$\begingroup$

So, I don't know much about countable models of set theory, other than that they exist. To me, their existence is a very weird thing (and a reason to move away from first-order formulations). Here is one thing that really weirds me out:

We have definitions of infinity, some set is infinite if it can be put into 1-1 correspondence with the natural numbers or a set is infinite if it can be put into 1-1 correspondence with a proper subset of itself (Dedekind infinite, if I recall).

How do these definitions work when we're in a countable model of ZFC? In particular, given that we know that the powerset of a countably infinite set will be uncountably infinite, how are we to understand this given that our model is countable? What does "uncountable" mean in such a context?

$\endgroup$
4
$\begingroup$

Recall that a structure $\mathcal M\models\exists x\varphi(x)$ if and only if there exists some $m\in|\mathcal M|$ such that $\mathcal M\models\varphi(m)$.

Countable is a definable property. $\varphi(a)$ says that $a$ is countable if and only if there exists an injective function whose domain is $a$ and its range includes only "finite ordinals" (transitive sets which are well-ordered by $\in$, and satisfy one of many characterizations of finiteness).

So $\mathcal M\models\varphi(a)$ if and only if there is $f\in|\mathcal M|$ such that $\mathcal M\models f\colon a\to\omega^\mathcal M\text{ is injective}$. If $\mathcal M$ does not know about such $f$, then $\mathcal M$ would think that $a$ is uncountable.


The key point is that a model of $\sf ZF$ never knows about sets which lie outside that model. Much like how $\Bbb Q$ does not know about $\pi,e,i$ or $\sqrt 2$ since none of them is a rational number.

$\endgroup$
2
$\begingroup$

Even though $A$ and $B$ are both countable sets in the model, it may be that there is no bijection from $A$ to $B$ which is also in the model.

So "internally" the denizens of that model could say "We have an injection from $A$ to $B$, but no surjection, so $A$ has smaller cardinality", even while external to the model we can see that there is a bijection.

$\endgroup$
1
$\begingroup$

In a countable model, every set of course has an enumeration, but in the case of the powerset of an infinite set no such enumeration is a member of the model! (Sometimes one says there is no "internal" enumeration.)

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.