Woodin's Ω Conjecture: completeness of Ω-logic under a proper class of Woodin cardinals

Woodin's Ω Conjecture asserts that, assuming ZFC (Zermelo–Fraenkel set theory with Choice) and a proper class of Woodin cardinals, the semantic consequence relation of Ω-logic coincides with its quasi-syntactic proof relation: for every sentence $\varphi$ of set theory, $\emptyset \models_\Omega \varphi$ if and only if $\emptyset \vdash_\Omega \varphi$. The soundness direction is a theorem of Woodin; the completeness direction has been open since Ω-logic was introduced around 1999. No proof or refutation of the completeness direction has been published; the only known partial route is Woodin's conditional Theorem 7.38, which derives the Ω Conjecture from V = Ultimate-L together with the Θ₀ Conjecture, both themselves open problems. The problem posed is a proof or refutation of the conjecture in its standard formulation.

Background

Cantor's Continuum Hypothesis (CH) asserts that every infinite set of real numbers is either countable or in bijection with the set $\mathbb{R}$ of all reals; equivalently $2^{\aleph_0}=\aleph_1$. Gödel and Cohen proved CH independent of ZFC (Zermelo–Fraenkel set theory with the Axiom of Choice), the standard axiom system for set theory, so CH can only be settled, if at all, by new axioms. The leading candidates are large cardinal axioms — assertions that certain very large infinite cardinals with strong reflection properties exist. A Woodin cardinal is a cardinal $\delta$ such that for every function $f:\delta\to\delta$ there is an elementary embedding $j:V\to M$ (a map from the universe of sets into a transitive class $M$ preserving the truth of all first-order formulas) whose critical point, the first ordinal moved by $j$, lies below $\delta$ and whose target is closed under $f$ arbitrarily far below $\delta$. Large cardinals settle the theory of definable sets of reals, but a classical theorem of Lévy and Solovay shows that no standard large cardinal axiom decides CH.

To organize exactly what large cardinal axioms can decide, in a way immune to the forcing maneuvers behind independence results, Woodin (1999) introduced the strong logic called Ω-logic. Forcing passes from the universe $V$ of sets to a Boolean-valued extension $V^{\mathbb{B}}$, where $\mathbb{B}$ is a complete Boolean algebra; $V^{\mathbb{B}}_\alpha$ denotes the $\alpha$-th rank-initial segment of such an extension. For a countable theory $T$ and a sentence $\varphi$ in the language of set theory,

$T \models_\Omega \varphi$

holds if for every complete Boolean algebra $\mathbb{B}$ and every ordinal $\alpha$, whenever $V^{\mathbb{B}}_\alpha \models T$ then also $V^{\mathbb{B}}_\alpha \models \varphi$; a sentence is Ω-valid if $\emptyset \models_\Omega \varphi$. Woodin proved that under a proper class of Woodin cardinals this semantic relation is forcing-invariant: set forcing can never change whether $T \models_\Omega \varphi$ holds. Matching it is a quasi-syntactic proof relation $T \vdash_\Omega \varphi$: the "proofs" are universally Baire sets of reals (sets of reals coded by a pair of projecting trees that continue to determine the set and its complement in every set-forcing extension, a robustness property of definable sets of reals supplied by large cardinal axioms), and the test structures are countable transitive models (countable well-founded models of the membership relation) closed under these proofs. This proof relation is likewise forcing-invariant, and soundness is a theorem: $T \vdash_\Omega \varphi$ implies $T \models_\Omega \varphi$. A theory $T$ is Ω-complete for a class $\Gamma$ of sentences if for each $\varphi\in\Gamma$ either $T \models_\Omega \varphi$ or $T \models_\Omega \neg\varphi$; for example, under a proper class of Woodin cardinals ZFC is Ω-complete for sentences about $L(\mathbb{R})$, the constructible universe over the reals, yet it is never Ω-complete at the level of CH.

Koellner's 2006 study of whether absolutely undecidable statements exist builds on this framework. It shows that, under a proper class of Woodin cardinals, the maximality axiom $(*_0)$ for the structure $\langle H(\omega_2),\in,I_{NS},A\rangle$ — where $H(\omega_2)$ is the collection of sets whose transitive closures have size below $\omega_2$, $I_{NS}$ is the non-stationary ideal on $\omega_1$ (the family of subsets of $\omega_1$ disjoint from some closed unbounded set), and maximality is required over $\Pi_2$ sentences (those of the form "for every set there is a set such that …" followed only by bounded quantifiers) — is, assuming it is Ω-satisfiable (equivalently, assuming ZFC $+ (*_0)$ is Ω-consistent, a hypothesis implied by the Strong Ω Conjecture), Ω-complete for the theory of $H(\omega_2)$, with $2^{\aleph_0}=\aleph_2$ holding in the resulting Ω-complete theory; whether $(*_0)$ is in fact Ω-satisfiable over $V$ is left open (cf. Theorem 3.12 of the Stanford Encyclopedia of Philosophy entry on the Continuum Hypothesis, Reference 9: assuming the Strong Ω Conjecture, there is an axiom $A$ such that ZFC $+ A$ is Ω-satisfiable and Ω-complete for the structure $H(\omega_2)$, any such $A$ Ω-implies $\neg\mathrm{CH}$, and in the maximal such theory $2^{\aleph_0}=\aleph_2$). The hinge of the whole construction, and the paper's designated central open problem, is the Ω Conjecture: the assertion that Ω-logic is complete, so that Ω-validity is always witnessed by an Ω-proof. Under the Ω Conjecture, Ω-logic becomes exactly the logic of the true large cardinal axioms for $\Pi_2$ statements, realizing Gödel's 1946 proposal of an absolute notion of mathematical provability, and any axiom that is Ω-consistent with ZFC and Ω-complete for the theory of $H(\omega_2)$ must Ω-imply $\neg\mathrm{CH}$.

Problem Statement

The Ω Conjecture (Woodin). Assume ZFC and that there is a proper class of Woodin cardinals. Then for every sentence $\varphi$ of the language of set theory,

$\emptyset \models_\Omega \varphi \quad\text{if and only if}\quad \emptyset \vdash_\Omega \varphi,$

where $\models_\Omega$ is the semantic Ω-consequence relation (so $\emptyset \models_\Omega \varphi$ means that $\varphi$ is true in $V^{\mathbb{B}}_\alpha$ for every complete Boolean algebra $\mathbb{B}$ and every ordinal $\alpha$) and $\vdash_\Omega$ is the quasi-syntactic Ω-proof relation (whose proofs are universally Baire sets of reals, verified in closed countable transitive models).

The soundness direction ($\emptyset \vdash_\Omega \varphi$ implies $\emptyset \models_\Omega \varphi$) is a theorem. The open problem is the completeness direction: prove that, under the stated large cardinal hypothesis, every Ω-valid sentence has an Ω-proof; or refute the conjecture by establishing, in ZFC together with standard large cardinal axioms, that there exists a sentence $\varphi$ with $\emptyset \models_\Omega \varphi$ but $\emptyset \not\vdash_\Omega \varphi$.

Scope notes. The truth of the conjecture is itself invariant under set forcing (given a proper class of Woodin cardinals), so its truth value cannot be flipped by passing to a forcing extension, and neither direction can be settled by set forcing alone. The problem is stated in the standard formulation of the conjecture (Woodin 1999; Koellner 2006; the Stanford Encyclopedia of Philosophy entry on the Continuum Hypothesis). The same source lists companion open questions — the Ω-satisfiability of the axiom $(*_0)$ over $V$, and whether the embedding axiom for $L(V_{\lambda+1})$ with supplementary principles (A) and (B) settles CH — which are not part of this problem's completion boundary; a further companion question, the uniqueness of Ω-complete theories, was resolved negatively in 2009.

References

1. Peter Koellner, "On the Question of Absolute Undecidability", Philosophia Mathematica 14(2):153–188 (2006), DOI: 10.1093/philmat/nkj009, https://doi.org/10.1093/philmat/nkj009 (revised and reprinted in Kurt Gödel: Essays for his Centennial, Lecture Notes in Logic 33, 2009).
2. Peter Koellner and W. Hugh Woodin, "Incompatible Ω-Complete Theories", The Journal of Symbolic Logic 74(4):1155–1170 (2009), DOI: 10.2178/jsl/1254748685, https://doi.org/10.2178/jsl/1254748685
3. W. Hugh Woodin, "The Axiom of Determinacy, Forcing Axioms, and the Nonstationary Ideal", De Gruyter, Series in Logic and its Applications 1 (1999), ISBN 9783110157086, DOI: 10.1515/9783110804737, https://doi.org/10.1515/9783110804737
4. W. Hugh Woodin, "In Search of Ultimate-L: The 19th Midrasha Mathematicae Lectures", The Bulletin of Symbolic Logic 23(1):1–109 (2017), DOI: 10.1017/bsl.2016.34, https://doi.org/10.1017/bsl.2016.34
5. David Asperó and Ralf Schindler, "Martin's Maximum++ implies Woodin's axiom (*)", Annals of Mathematics 193(3):793–835 (2021), DOI: 10.4007/annals.2021.193.3.3, arXiv:1906.10213 (arXiv preprint titled "${\sf MM}^{++}$ implies $(*)$"), https://doi.org/10.4007/annals.2021.193.3.3
6. Matteo Viale, "Strong forcing axioms and the continuum problem [after Asperó's and Schindler's proof that MM++ implies Woodin's Axiom (*)]", Séminaire Bourbaki 2021/2022, exposé 1207, Astérisque 446, pp. 383–416 (2024), DOI: 10.24033/ast.1216, arXiv:2305.07784 (arXiv preprint version substitutes "following" for "after"), https://doi.org/10.24033/ast.1216
7. Vincenzo Dimonte, "I0 and rank-into-rank axioms", Bollettino dell'Unione Matematica Italiana 11(3):315–361 (2018; published online 2017), DOI: 10.1007/s40574-017-0136-y, arXiv:1707.02613, https://doi.org/10.1007/s40574-017-0136-y
8. Sandra Müller and Grigor Sargsyan, "Gödel's program in set theory", Monatshefte für Mathematik 208(4):729–750 (2025), DOI: 10.1007/s00605-025-02086-x, arXiv:2412.07325, https://doi.org/10.1007/s00605-025-02086-x
9. Peter Koellner, "The Continuum Hypothesis", Stanford Encyclopedia of Philosophy (2013), https://plato.stanford.edu/entries/continuum-hypothesis/