Compact covering and game determinacy (Q1910715)

From MaRDI portal
scientific article
Language Label Description Also known as
English
Compact covering and game determinacy
scientific article

    Statements

    Compact covering and game determinacy (English)
    0 references
    0 references
    0 references
    28 October 1996
    0 references
    Let \(X\) and \(Y\) be separable metric spaces and let \(f\) be a continuous function from \(X\) onto \(Y\). Then \(f\) is said to be perfect if the inverse images of compact subsets of \(Y\) are compact subsets of \(X\). If \(f\) is perfect, the following two statements are true: CC Each compact subset of \(Y\) is the image of a compact subset of \(X\). IP For some subspace of \(X,f\) restricted to that subspace is a perfect continuous function onto \(Y\). It can happen that \(f\) is not perfect, and yet CC or IP is true. An \(f\) for which CC is true is said to be a compact covering map, while an \(f\) for which IP is true is said to be an inductively perfect map. Each inductively perfect map is a compact covering map. The authors show that compact covering maps need not be inductively perfect. The examples showing this are by many standards pathological. The main subject of the paper is whether non-pathological examples can be found. Non-pathological is taken to mean that the relevant spaces lie in the projective hierarchy of some compactification. Two of the several nice results of this paper suggest that classical mathematics cannot answer this question. In particular, the authors show that Gödel's Axiom of Constructibility implies that there are co-analytic \(X\) and \(Y\) with a compact covering map which is not inductively perfect. But if the Axiom of Analytic Determinacy is assumed, then every compact covering map between two co-analytic spaces is also inductively perfect. So, the Axiom of Analytic Determinacy implies that if \(X\) and \(Y\) are both Borel, or even if only \(X\) is assumed to be Borel, then every compact covering map from \(X\) to \(Y\) is also inductively perfect. It is not good enough to only assume that \(Y\) is Borel; the authors show that a \(G_\delta Y\) can be found for which some compact covering map from \(X\) to \(Y\) is not inductively perfect. They also solve a problem of Michael, and disprove a claimed theorem in published literature.
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    infinite game
    0 references
    winning strategy
    0 references
    axiom of constructibility
    0 references
    axiom of analytic determinacy
    0 references
    metric spaces
    0 references
    compact covering map
    0 references
    inductively perfect map
    0 references
    projective hierarchy
    0 references
    co-analytic spaces
    0 references
    0 references