Katětov Order: Structure & Complexity

• Katětov order is a relation on ideals and filters on ω, defined via reduction functions that pull back small sets and compare combinatorial complexity.
• It underpins diverse applications in forcing, cardinal invariants, and game-theoretic frameworks, linking set theory with topology and computability.
• The framework demonstrates non-linearity by lacking minimal tall Borel ideals and embedding complex structures like Boolean algebras and algebraic hierarchies.

The Katětov order is a central notion in the study of ideals and filters on countable sets, organizing them according to the existence of reduction functions that transfer properties, and controlling a wide array of combinatorial, topological, and descriptive complexity phenomena. Deep connections exist between the Katětov order, forcing, cardinal invariants, effective topos theory, and computability. Modern research advances include a game-theoretic variant, intricate applications to Borel ideals, and isomorphisms with algebraic hierarchies such as the Lawvere-Tierney order in the Effective Topos.

1. Definition of the Katětov Order

Let $\mathcal{I}$ and $\mathcal{J}$ be ideals on $\omega$. The Katětov order, denoted $\leq_K$, is given by: $$\mathcal{I} \leq_K \mathcal{J} \iff \exists f:\omega\to\omega \; \forall A\in\mathcal{I} \; (\,f^{-1}[A]\in\mathcal{J}\,)$$ This states that all $\mathcal{I}$-small sets pull back along $f$ to $\mathcal{J}$-small sets. Dually for filters $\mathcal{U},\mathcal{V}$: $$\mathcal{U} \leq_K \mathcal{V} \iff \exists h:\omega\to\omega \; \forall A\subseteq\omega \; [A\in\mathcal{U} \Rightarrow h^{-1}[A]\in\mathcal{V}]$$ Two ideals are Katětov-equivalent if each is reducible to the other, written $\mathcal{J}$0. The Katětov–Blass order $\mathcal{J}$1 strengthens this by requiring finite-to-one reductions, but for most naturally encountered Borel ideals, $\mathcal{J}$2 and $\mathcal{J}$3 coincide (Das et al., 2020).

The Katětov order measures the relative "size" or combinatorial complexity of ideals: if $\mathcal{J}$4, then every combinatorial property holding for $\mathcal{J}$5-positive sets is, via pullback, inherited by $\mathcal{J}$6-positive sets (Grebík et al., 2017).

2. Central Structural Phenomena

Nonexistence of Minimal Tall Borel Ideals

A tall ideal intersects every infinite subset of $\mathcal{J}$7 in an infinite set. There is no minimal tall Borel ideal for $\mathcal{J}$8: for every tall Borel ideal $\mathcal{J}$9, there exists another tall Borel ideal $\omega$0 with $\omega$1. The set of all codes for tall $\omega$2 ideals is $\omega$3-complete, whereas for any fixed $\omega$4 the set of codes above $\omega$5 in the Katětov order is $\omega$6; their intersection cannot be both unless the projective hierarchy collapses, so no minimal ideal exists (Grebík et al., 2017).

Cardinal and Combinatorial Invariants

The Katětov order allows analysis of ideal- and filter-associated cardinal invariants (additivity, covering, non-covering, cofinality, their $\omega$7-variants), which play roles in the structure theory of the continuum and forcing. For example, for certain “critical ideals” $\omega$8 arising in topology:

• $\omega$9
• $\leq_K$0, $\leq_K$1 These values match the standard Cichon–Blass spectrum for Borel ideals (Kowalczuk, 28 Feb 2026).

Borel and Combinatorial Ideals

Between classical ideals such as $\leq_K$2 and $\leq_K$3, the Katětov order has a copy of the Boolean algebra $\leq_K$4, producing antichains of size continuum and increasing/decreasing chains of length $\leq_K$5 (Das et al., 2020). Various combinatorial ideals (Hindman, Ramsey, Summable, van der Waerden) are pairwise Katětov-incomparable, demonstrating that the poset is highly non-linear and has extensive complexity (Filipów et al., 2023).

3. Katětov-Theoretic Characterization of Forcing and Forcings from Ideals

The Katětov order governs precise thresholds at which certain combinatorial forcing phenomena occur:

• Cohen Real Addition: For any ideal $\leq_K$6, Laver forcing associated with the dual coideal $\leq_K$7 adds a Cohen real if and only if there is some $\leq_K$8 such that $\leq_K$9, where $$\mathcal{I} \leq_K \mathcal{J} \iff \exists f:\omega\to\omega \; \forall A\in\mathcal{I} \; (\,f^{-1}[A]\in\mathcal{J}\,)$$0 is the ideal of nowhere-dense subsets of $$\mathcal{I} \leq_K \mathcal{J} \iff \exists f:\omega\to\omega \; \forall A\in\mathcal{I} \; (\,f^{-1}[A]\in\mathcal{J}\,)$$1 (Rosa et al., 14 Jan 2026).
• Random Real Non-Addition: No such forcing adds random reals, as none add bounded eventually-different (BED) reals; the associated proof uses pure-decision and fusion in the forcing (Rosa et al., 14 Jan 2026).
• Half-Cohen Reals: For the ideal $$\mathcal{I} \leq_K \mathcal{J} \iff \exists f:\omega\to\omega \; \forall A\in\mathcal{I} \; (\,f^{-1}[A]\in\mathcal{J}\,)$$2 associated with infinitely-often-equal reals, one has $$\mathcal{I} \leq_K \mathcal{J} \iff \exists f:\omega\to\omega \; \forall A\in\mathcal{I} \; (\,f^{-1}[A]\in\mathcal{J}\,)$$3 a forcing associated to $$\mathcal{I} \leq_K \mathcal{J} \iff \exists f:\omega\to\omega \; \forall A\in\mathcal{I} \; (\,f^{-1}[A]\in\mathcal{J}\,)$$4 or $$\mathcal{I} \leq_K \mathcal{J} \iff \exists f:\omega\to\omega \; \forall A\in\mathcal{I} \; (\,f^{-1}[A]\in\mathcal{J}\,)$$5 adds a half-Cohen real (Rosa et al., 14 Jan 2026).
• Structural Separation (Laver Property): The Katětov order distinguishes between addition of Cohen reals and satisfaction of the Laver property even for ultrafilters. There exist ultrafilters $$\mathcal{I} \leq_K \mathcal{J} \iff \exists f:\omega\to\omega \; \forall A\in\mathcal{I} \; (\,f^{-1}[A]\in\mathcal{J}\,)$$6 such that Laver forcing $$\mathcal{I} \leq_K \mathcal{J} \iff \exists f:\omega\to\omega \; \forall A\in\mathcal{I} \; (\,f^{-1}[A]\in\mathcal{J}\,)$$7 adds no Cohen real (i.e., $$\mathcal{I} \leq_K \mathcal{J} \iff \exists f:\omega\to\omega \; \forall A\in\mathcal{I} \; (\,f^{-1}[A]\in\mathcal{J}\,)$$8) but nevertheless fails the Laver property. This is controlled via Katětov-comparison with canonical “slalom” ideals $$\mathcal{I} \leq_K \mathcal{J} \iff \exists f:\omega\to\omega \; \forall A\in\mathcal{I} \; (\,f^{-1}[A]\in\mathcal{J}\,)$$9 (Rosa et al., 14 Jan 2026).

4. Game-Theoretic and Extended Katětov Orders

Recent advances define a gamified or game-theoretic Katětov order $\mathcal{I}$0 for filters:

• Two-player Game Formulation: Maker and Breaker interact through a finite-query process involving the choice of sets from filters and responding via strategies (which may be continuous or computable) (Kihara et al., 8 Feb 2026).
• Gamified vs. Classical Order: $\mathcal{I}$1 is strictly coarser than classical $\mathcal{I}$2: it collapses all maximal almost-disjoint (MAD) family ideals to a single equivalence class, while retaining infinite strictly increasing chains among structured ideals like finite-support Fubini products (Kihara et al., 8 Feb 2026, Kihara et al., 20 May 2026).
• Non-linearity and Complexity: The gamified order embeds $\mathcal{I}$3, yielding continuum many pairwise incomparable classes. While it identifies larger classes, within these, rich and complex behavior persists, especially related to Ramsey-theoretic separations (Kihara et al., 20 May 2026).
• Isomorphism with Lawvere–Tierney Order: For computable strategies, the computable gamified Katětov order is isomorphic to the Lawvere-Tierney order $\mathcal{I}$4 on basic topologies of the effective topos, revealing a deep connection between combinatorial and logical complexity (Kihara et al., 8 Feb 2026, Kihara et al., 13 May 2026).

5. Katětov Order and Topological/Categorical Structures

The Katětov order underlies the description and comparison of subtoposes in the Effective Topos, particularly via the Lawvere–Tierney order. The computable gamified Katětov order precisely models $\mathcal{I}$5 on subtoposes: every LT-topology is a join of basic ones, and the comparison reduces to Katětov-type reducibility on the induced upper sets. Furthermore, to every filter $\mathcal{I}$6 can be associated a spectrum of Turing degrees, $\mathcal{I}$7, which always forms a proper initial segment of the Turing degrees; in the $\mathcal{I}$8 case, exactly the hyperarithmetical degrees are obtained (Kihara et al., 8 Feb 2026).

This reveals that not only does the Katětov order classify combinatorial complexity for sets and ideals, it also controls logical and computable complexity in topological/categorical structures (Kihara et al., 13 May 2026).

6. Broader Landscape and Open Directions

The global Katětov hierarchy is non-linear and lacks both minimal tall Borel ideals and maximal elements among standard combinatorial Borel families. Embedding results (e.g., $\mathcal{I}$9) show the order is as wild as any partially ordered set of the continuum’s size (Das et al., 2020, Kihara et al., 20 May 2026). Classes of ideals remain pairwise incomparable (Hindman, Ramsey, Summable, van der Waerden) even in the gamified variant (Filipów et al., 2023, Kihara et al., 20 May 2026).

Key open questions include:

• The full characterization of dividing lines among particular classes of Borel ideals.
• Structural invariants distinguishing finer subclasses within $f$0 ideals.
• The role and reach of the gamified Katětov order in identifying logical complexity across set theory, computability, and categorical frameworks.

The Katětov order remains an organizing principle for logical, topological, and combinatorial complexity, now understood as influencing and unifying diverse mechanisms once placed in the appropriate topological and categorical frameworks (Kihara et al., 13 May 2026).