How expressive can $\mathcal{L}_{\kappa,\kappa}$ be?

(Note that the logic systems described in this question only refer to logic systems restricted to the language $\in$)

1. Can $\mathcal{L}_{\kappa,\kappa}$ express $n$-th order finitary logic? It is clear that it can express first-order logic ($\Pi_{<\omega}^0$), but it seems unlikely that $\mathcal{L}_{\kappa,\kappa}$ can express even $\Pi_{<\omega}^1$.

By the answer to this question: Gödel's Constructible Universe in Infinitary Logics (A Possible Approach to HOD Problem), it is known that $L^{\Pi_{<\omega}^1}=HOD$. Assuming $\mathcal{L}_{\kappa,\kappa}$ expresses $\Pi_{<\omega}^1$, it is then not difficult to show that $HOD\subseteq L^{\mathcal{L}_{\kappa,\kappa}}\subseteq L(V_\kappa)$. Thus, unless $L(V_\kappa)=V$ (which is true iff $L=V$), $V\neq HOD$.

Therefore, the existence of such a $\kappa$ is inconsistent with $V\neq L\land V=HOD$. This is very doubtful.

2. Can $\mathcal{L}_{\kappa,\kappa}$ express $\Pi_n^1$ for $n>0$? For $n=0$, it is once again easy to see that, yes, this is true ($\Pi_0^1=\Pi_0^0$) for every $\kappa$. Other than this, I have no information for this question.

• Are you sure you've written what you meant to? Note that since you've put the "there is a formula" after all the "for all"s, everything you've said collapses, and your definition amounts to: "For every $S\subseteq U^{<\omega}$there is an $\mathcal{L}_{\kappa\kappa}$-formula $\varphi$ such that for every finite sequence $\nu$ we have $\varphi(\nu)\iff \nu\in S$." Maybe you want to put some sort of requirement on what $T$ can be? – Noah Schweber Nov 2 '17 at 3:29
• Yes, obviously. Sorry, my mistake. – Keith Millar Nov 2 '17 at 3:42
• It doesn't look to me like you've fixed this ... – Noah Schweber Nov 2 '17 at 23:05
• @NoahSchweber I was a little busy, sorry. Here I have fixed it. – Keith Millar Nov 3 '17 at 2:28

Suppose that $\kappa$ is a Mahlo cardinal. This is $\Pi^1_1$ expressible in $V_\kappa$, since it amounts to the assertion that every closed unbounded subset of $\kappa$ contains a regular cardinal. But it cannot be expressible by an assertion of $\mathcal{L}_{\kappa,\kappa}$, since there is a forcing extension $V[G]$ in which $V_\kappa$ is preserved, but which destroys the Mahloness of $\kappa$. Since $V_\kappa$ is preserved, the truth of any $\mathcal{L}_{\kappa,\kappa}$ assertion is invariant by the forcing, but the Mahloness of $\kappa$ is not.

One can make other similar kinds of examples. For example, "$\kappa$ is regular" is expressible in $V_\kappa$ by a $\Pi^1_1$ assertion, but it cannot be provably equivalent to any $\mathcal{L}_{\kappa,\kappa}$ assertion, since for example with a measurable cardinal, this can be changed by forcing without changing $V_\kappa$.

Here is another way to argue, without forcing. Let $\kappa$ be an inaccessible cardinal. The assertion that $\kappa$ is regular is $\Pi^1_1$ expressible in $V_\kappa$. If we fix a $\mathcal{L}_{\kappa,\kappa}$ truth predicate for $V_\kappa$, we can find many $\gamma<\kappa$ with $V_\gamma\prec_{\mathcal{L}_{\gamma,\gamma}}V_\kappa$. In particular, there will be such $\gamma$ of cofinality $\omega$ above the size of any particular $\varphi$ in $\mathcal{L}_{\kappa,\kappa}$. So $V_\gamma$ and $V_\kappa$ will agree on the truth of $\varphi$, but not on the assertion that the ordinals are regular.

I believe that one can likely press this further to make an example in ZFC, without any large cardinals.

• For a ZFC example, what about the following: let $\lambda$ be a limit cardinal of cofinality $>\kappa^+$, and look at a generic extension $V[G]$ in which $cof(\lambda)=\kappa^+$ but $V[G]$ has no new $\kappa$-sequences of elements of $\lambda$. Then since "$cof(\alpha)=\beta$" is $\Pi^1_1$ over $V_\alpha$ with a parameter for $\beta$, and $\mathcal{L}_{\kappa\kappa}$-truth in a structure depends only on the length-$\kappa$ sequences of elements, we have a contradiction. Does this work? – Noah Schweber Nov 3 '17 at 16:31
• In the non-forcing example, why can we find $\gamma$ with $V_\gamma \preceq_{\mathcal{L}_{\kappa \kappa}} V_\kappa$? For every $\alpha < \kappa$ there is a formula asserting $\alpha$ exists. We really want to fix a small fragment in which the formula $\phi$ lies. – Douglas Ulrich Nov 3 '17 at 16:39
• @DouglasUlrich Yes, thannks. I had meant $V_\gamma\prec_{\mathcal{L}_{\gamma,\gamma}} V_\kappa$. For $\gamma$ large enough, this includes any $\varphi$ in $\mathcal{L}_{\kappa,\kappa}$. I have now editted to correct this. – Joel David Hamkins Nov 3 '17 at 16:44
• @NoahSchweber That seems to work. – Joel David Hamkins Nov 3 '17 at 16:48