We answer an old question of Michael Hrušák by constructing a +-Ramsey MAD family without the need of any additional axioms beyond ZFC. We also prove that every Miller-indestructible MAD family is +-Ramsey, this improves a result of Michael Hrušák. * Definition 1 1. By A ⊥ we denote the set of all X ⊆ ω such that A∪{X} is almost disjoint.
IfIn [6] it is proved that there is a MAD family that is not +-Ramsey. On the other hand, +-Ramsey MAD families can be contructed under b = c, cov(M) = c, a < cov(M) or ♦ (b) (see [6] and [8]). Michael Hrušák asked the following, Problem 2 (Hrušák [6]) Is there a +-Ramsey MAD family in ZFC?In this note we will provide a positive answer to this question. In [20] (see also [7] and [15]) Saharon Shelah developed a novel and powerfull method to construct MAD families. He used it to prove that there is a completely separable MAD family if s ≤ a or a < s and a certain PCF-hypothesis holds. Our technique for constructing a +-Ramsey MAD is based on the technique of Shelah (however, in this case we were able to avoid the PCF-hypothesis). It is worth mentioning that the method of Shelah has been further developed in [19] and [15] where it is proved that weakly tight MAD families exist under s ≤ b. Our notation is mostly standard, the definition and basic properties of the cardinal invariants of the continuum used in this note can be found in [2].
PreliminariesA MAD family A is completely separable if for every X ∈ I (A) + there is A ∈ A such that A ⊆ X. This type of MAD families was introduced by Hechler in [5]. A year later, Shelah and Erdös asked the following question:Problem 3 (Erdös-Shelah) Is there a completely separable MAD family?It is easy to construct models where the previous question has a positive answer. It was shown by Balcar and Simon (see [1]) that such families exist assuming one of the following axioms: a = c, b = d, d ≤ a and s = ω 1 . In [20] (see also [7] and [15]) Shelah developed a novel and powerful method to construct completely separable MAD families. He used it to prove that there are such families if either s ≤ a or a < s and a certain (so called) PCF hypothesis