Working below a \(low_ 2\) recursively enumerable degree (Q584251)

The results obtained by the authors are connected with investigations of jump classes of r.e. Turing degrees. The authors wish to find a formula \(\theta\) (x) which is true for low degrees but false for high degree and so separates one jump class from another. They prove that Robinson's splitting theorem, which is valid for low degrees, is not valid in the class of high degrees. Since Robinson's theorem was extended to \(low_ 2\) degrees by \textit{L. Harrington} (unpublished result), they separate the \(low_ 2\) degrees from the high ones (this theorem is proved in a forthcoming paper of the authors [Splitting and density cannot be combined below any high r.e. degree (to appear)]). The authors also prove some other results in the reviewed paper. They find conditions when an embedding of a partial order \({\mathcal X}\) into a structure of degrees below a low degree \({\mathfrak a}\) can be extended to an embedding of an order \({\mathcal Y}\) such that \({\mathcal X}\subseteq {\mathcal Y}\). This gives a decision procedure for the fragment of the \(\forall \exists\)-theory of the degrees below a fixed \(low_ 2\) degree \({\mathfrak a}\). Another result of the paper is the proof of the splitting theorem for \(low_ 2\) degrees.