scientific article; zbMATH DE number 1388189

Publication date 16 January 2000

chaotic behavior classification by countable models complexity of orbits continuous actions of Polish groups definable cardinality definable equivalence relations effective cardinality of quotient spaces equivalence classes orbit equivalence relations Polish spaces turbulence turbulent action

Mathematics Subject Classification ID

Research exposition (monographs, survey articles) pertaining to mathematical logic and foundations (03-02) Descriptive set theory (03E15) Structure of general topological groups (22A05) Descriptive set theory (topological aspects of Borel, analytic, projective, etc. sets) (54H05)

Review text

This monograph is an important contribution to the ongoing study (see e.g.\ the book by \textit{H. Becker} and \textit{A. S. Kechris} [The descriptive set theory of Polish group actions (London Mathematical Society Lecture Note Series. 232) (1996)] and the surveys by \textit{G. Hjorth} and \textit{A. S. Kechris} [Bull. Symb. Log. 3, 329-346 (1997; Zbl 0889.03038)] and by \textit{A. S. Kechris} [Bull. Symb. Log. 5, 161-174 (1999; Zbl 0933.03057)]) of definable equivalence relations on Polish spaces. This study allows precise comparisons of the complexity of classification problems arising in many areas of mathematics and leads to the understanding of the effective cardinality of quotient spaces (e.g., \(\mathbb{R}\) has effective cardinality smaller than \(\mathbb{R} / \mathbb{Q}\), since there exists a definable injection of \(\mathbb{R}\) into \(\mathbb{R} / \mathbb{Q}\), but there is no definable injection of \(\mathbb{R} / \mathbb{Q}\) into \(\mathbb{R}\)). Many natural classification problems in mathematics are solved by assigning to each equivalence class either reals (in this case the equivalence relation is said to be smooth) or more complex objects, such as countable sets of reals, countable ordinals, countable sets of countable ordinals (this is e.g.\ the case for the Ulm classification of countable torsion abelian groups), countable groups, etc. All these invariants can be viewed as countable models (in the sense of model theory) for an appropriately chosen countable language, considered up to isomorphism. Understanding when an equivalence relation admits classification by countable models up to isomorphism is therefore a central problem, and the book under review is a remarkable step in this direction. The monograph focuses on orbit equivalence relations induced by continuous actions of Polish groups: these equivalence relations do not exhaust all Borel or analytic ones, but constitute a wide subclass containing most definable equivalence relations appearing in practice. Even though manuscripts have been circulating for a few years, here for the first time the definition of a turbulent action and its fundamental properties appear in print: turbulence can be viewed as a very strong chaotic behavior of the action, and on p. 41 the author attempts to give an intuitive feeling for the concept. The idea of turbulence constitutes a major breakthrough in the area since it provides a necessary and sufficient condition for an orbit equivalence relation to be unclassifiable by countable models. The book's major theorem is stated in the preface: if \(G\) is a Polish group acting continuously on a Polish space \(X\) then either \textit{a)} the orbit equivalence relation admits a ``reasonable (see below) classification by countable models, or \textit{b)} a turbulent \(G\)-space \(Y\) is continuously \(G\)-embedded into \(X\). We now outline the content of the book. In the preface many examples of definable equivalence relations are sketched and the notion of classifiability by countable model is informally introduced; the preface includes also a discussion of the notion of definable cardinality and the definition of turbulence. Chapter 1 helps the reader gain an overview of the wide range of applications for the new technique: in fact following the introduction of the notion of turbulence several natural equivalence relations have been shown to be reducible to turbulent orbit equivalences by Hjorth himself and others. Chapter 2 sets the stage by establishing notation and terminology and by reviewing many classical results: most proofs are provided (or given as exercises), but the reader with some background in descriptive set theory will skim this chapter and refer back to it when needed. In chapter 3 turbulence enters the picture and its sufficiency for non-classification by countable models is proved. Chapters 4 and 5 are devoted to applications of the notion of turbulence: the former deals with homeomorphisms of the unit interval up to conjugacy (which admit classification by countable models) and homeomorphism of the unit square (which do not), the latter shows that irreducible infinite-dimensional representations of the free group on countably many generators do not admit classification by countable models. Chapter 6 introduces a generalized Scott analysis of the complexity of orbits. While the classical Scott analysis (reworked in the first section of the chapter) allows an understanding of \(S_\infty\)-actions, its generalization allows an understanding of actions of \(GE\) groups: for these actions Borel classification by countable models is proved to be equivalent to the failure of \textit{b)} above, and hence the major theorem holds in this case with the weakest possible definition of reasonable reduction. Chapter 7 is devoted to the study of \(GE\) groups: in particular it is shown that this is a wide class, since it includes all abelian, invariantly metrizable, nilpotent, and countable products of locally compact groups. Chapter 8 is a bit off the main track of the book, and deals with ``pathological groups, such as non-Polishable Borel groups. Chapter 9 concludes the proof of the main theorem (here reasonable classification means classification by an absolutely \({\Delta}^1_2\) function) and studies cardinalities in \(L(\mathbb R)\). Chapter 10 concludes the book with a list of open problems and conjectures. This book is essential reading for anyone interested in the subject of definable equivalence relations, but is also important for mathematicians working in areas lying outside the usual boundaries of set theory. Indeed the author has strived to make the book as self-contained as possible and, with the admitted exception of chapter 9, no background in modern set theory is assumed. Many exercises of varying difficulty scattered throughout the text help the reader gain familiarity with the various concepts. The attempt of keeping the book self-contained is even excessive at times (e.g.\ on p. 7 the author spends several lines in proving that the union of two \(G_\delta\)'s is \(G_\delta\), and on p. 22 he feels the responsibility of explaining the notation \(\bigcup A\)), and this is a minor annoyance. There are a few typos and some minor inconsistencies (e.g.\ the result of exercise 2.19 is used in a proof 5 pages earlier) that do not prevent this book from being a milestone in our understanding of the structure of definable equivalence relations and effective cardinalities.

Recommendations

Cited in

(only showing first 100 items - show all)

This page was built for publication:

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4933738)