Criteria for exact saturation and singular compactness (Q2041960)

The paper under review presents a detailed study on the interplay of theories, saturated models and singular cardinals. Let \(T\) be a theory; \(T\) has exact saturation at a singular cardinal \(\kappa\) if \(T\) has a \(\kappa\)-saturated model which is not \(\kappa^+\)-saturated. When \(T\) is not stable, then \(T\) has exact saturation at any regular \(|T|<\kappa\). In a previous paper [J. Math. Log. 17, No. 1, Article ID 1750001, 18 p. (2017; Zbl 1420.03066)], the first author et al. showed that a simple theory \(T\) has exact saturation at \(\mu\), where \(\mu\) is singular, \(|T|<\kappa=cf(\mu)\), \(\mu^+=2^\mu\) and \(\Box_\mu\) holds. They also proved that when \(\kappa\) is a singular cardinal, \(\kappa^+=2^\kappa \), a NIP theory \(T\) of size less than \(\kappa\) is distal if and only if it does not have exact saturation at \(\kappa\) (for a definition of distal see paragraph 4.1.2 of that paper). The theory of dense linear orders is a nice example of a theory without exact saturation at singular \(\kappa\). This suggests that order is a property to take care of. In order to find criteria for exact saturation in broader classes of theories, the authors introduce in the article under review shredding (Definition 2.1, p. 7), a notion that seems to refine forking. It captures the obstacle to ensuring that one can realize a formula such that a large interval of a given indiscernible sequence is additionally indiscernible over the realization. Then, they introduce the class of unshreddable theories; they are the theories with a bound on the number of times a type can be shred. NIP and simple theories are unshreddable. The main theorem of the paper establishes that it is possible to build exactly saturated models of unshreddable theories with the independence property for singular cardinals under some set-theoretic hypotheses as above and some extra considerations on the theory in question. They examine some examples from the point of view of the notions previously introduced. In particular, they consider the \(SOP_n\) hierarchy (theories with the \(n\)th strong order property) and the corresponding notion of \(NSOP_n\). Finally, they provide a sufficient condition for having singular compactness, which is the negation of exact saturation. The article is very well written, easy to follow for a reader with an adequate background.