Resolvability and monotone normality (Q948871): Difference between revisions

A space is said to be \textsl{\(\kappa\)-resolvable} (respectively, \textsl{almost \(\kappa\)-resolvable}) if it contains \(\kappa\) pairwise disjoint dense subsets (respectively \(\kappa\) dense subsets such that the intersection of any two is nowhere dense). A space is maximally resolvable if it is \(\Delta(X)\)-resolvable, where \(\Delta(X)\) denotes the minimum cardinality of a non-empty open subset of \(X\). It is known that all metric and all linearly ordered spaces are maximally resolvable and the paper under review studies resolvability properties in the more general class of monotonically normal spaces. The main result of Section 1 is that every crowded monotonically normal space is \(\omega\)-resolvable, while in Section 2 it is shown that each crowded monotonically normal space is almost \(\mu\)-resolvable (where \(\mu= \min(\mathfrak c, \omega_2)\)) and hence such a space is almost \(\omega_1\)-resolvable and is almost \(\omega_2\)-resolvable under the negation of the continuum hypothesis. Possibly the two most important results of the final section are that from a supercompact cardinal it is consistent to have a monotonically normal space \(X\) with \(| X| =\Delta(X)=\aleph_\omega\) which is not maximally resolvable and that all monotonically normal spaces of size less than \(\aleph_\omega\) are maximally resolvable.