Precompactness notions in Kaplansky--Hilbert modules and extensions with discrete spectrum

This paper is a continuation of our work on the functional-analytic core of the classical Furstenberg-Zimmer theory. We introduce and study (in the framework of lattice-ordered spaces) the notions of total order-boundedness and uniform total order-boundedness. Either one generalizes the concept of ordinary precompactness known from metric space theory. These new notions are then used to define and characterize “compact extensions” of general measure-preserving systems (with no restrictions on the underlying probability spaces nor on the acting groups). In particular, it is (re)proved that compact extensions and extensions with discrete spectrum are one and the same thing. Finally, we show that under natural hypotheses a subset of a Kaplansky-Banach module is totally order bounded if and only if it is cyclically compact (in the sense of Kusraev).

💡 Research Summary

The paper continues the authors’ program of extracting the functional‑analytic core of the classical Furstenberg‑Zimmer structure theorem. Working in the setting of lattice‑normed spaces over a commutative unital C∗‑algebra A, the authors introduce two new compactness‑type notions: total order‑boundedness and uniform total order‑boundedness. A subset M of a lattice‑normed space E is called totally order‑bounded if there exists a decreasing net (uα) in the positive cone A⁺ converging to zero such that for each α one can find a finite set Fα⊂E with inf_{y∈Fα}|x−y|≤uα for every x∈M. Uniform total order‑boundedness replaces the net (uα) by an ordinary ε>0, demanding a single finite approximating set for each ε. When A=ℂ these notions collapse to the usual precompactness in normed spaces; otherwise they provide a natural order‑theoretic analogue.

Lemma 2.4 establishes basic stability properties of (uniformly) totally order‑bounded sets under finite unions, sums, bilinear maps, order‑closures and order‑bounded linear operators. Example 2.3 shows that total order‑boundedness does not imply uniform total order‑boundedness, even in a Kaplansky‑Hilbert (KH) module. The central compactness result, Proposition 2.5, proves that for a Stone algebra A (i.e., a complete lattice‑ordered C∗‑algebra) and a finite‑rank KH‑module E, the three conditions “order‑bounded”, “totally order‑bounded” and “uniformly totally order‑bounded” are equivalent. This is a Heine–Borel type theorem for finite‑rank KH‑modules.

In Section 3 the authors turn to abstract measure‑preserving systems (X,𝔅,μ,T) and their extensions Y→X without any standardness assumptions on the probability spaces or the acting groups. Using the order‑norm on L²(X|Y) they define “conditionally precompact” subsets (the uniform total order‑bounded sets) and call an extension compact if the unit ball of L²(X|Y) is conditionally precompact. Theorem 3.6 and Corollary 3.7 show that such compact extensions are exactly those with discrete spectrum (i.e., the associated Koopman representation decomposes into a direct sum of finite‑dimensional invariant subspaces). This recovers the classical equivalence of compactness and discrete spectrum, now valid for arbitrary probability spaces and groups. The proof still relies on the ambient Hilbert space structure, and the authors note in Remark 3.8 that removing this dependence remains an open problem.

Section 4 connects the new precompactness notions to cyclic compactness introduced by Kusraev. Under the natural hypothesis that the lattice‑normed space is “mix‑complete” (order‑complete and every order‑closed set is complete), Proposition 4.5 proves that a subset is totally order‑bounded if and only if it is relatively cyclically compact. Thus, in a Kaplansky‑Banach module, total order‑boundedness coincides with the cyclic compactness used in conditional set theory.

Overall, the paper achieves three major contributions: (1) it provides a purely order‑theoretic generalization of precompactness suitable for lattice‑normed spaces; (2) it uses this to give a clean, abstract definition of compact extensions and proves their equivalence with extensions of discrete spectrum in full generality; (3) it links these notions to cyclic compactness, thereby unifying several strands of functional analysis, ergodic theory, and conditional set theory. By doing so, the authors demonstrate that much of the Furstenberg‑Zimmer machinery can be recast in a functional‑analytic language that does not depend on classical measure‑theoretic regularity assumptions, opening the way for further extensions to non‑standard settings and to other categories where lattice‑normed modules arise.