Does V have a model that looks like V, according to Morse-Kelly set theory?

99 Views Asked by At

Let's talk about V, the class of all sets. We can't talk about this in ZFC, of course, so we will use Morse-Kelly set theory instead, which has classes.

My question is, can Morse-Kelly set theory prove (or disprove!) that their exists a model (U, $\in_U$) of set theory (where $U$ is a set, and therefore $(U, \in_U) \in V$) such that (V, $\in$) and (U, $\in_U$) are elementary equivalent as models of set theory?

1

There are 1 best solutions below

10
On

Yes. MK can define the satisfaction relation for $(V,\in)$ (and more generally for class models), and so it can carry out the usual proof of downward Löwenheim-Skolem for $(V,\in)$ (using global choice to get Skolem functions). This produces a countable elementary submodel $(U,\in_U)\prec (V,\in)$.

Alternatively, if you just want a structure that is elementarily equivalent to $(V,\in)$, since MK can define satisfaction in $(V,\in)$ it can define the theory of $(V,\in)$ and knows that this theory is consistent. The first-order completeness theorem now shows the theory of $(V,\in)$ has a (set) model.