The existential theory of equicharacteristic Henselian valued fields (Q299143): Difference between revisions

Let \(\mathbb F\) be a finite field of characteristic \(p\). To understand the first-order theory of the field of power series \(\mathbb F((t))\) is still one of the outstanding open problems in the model theory of (valued) fields. Until this paper, the situation can be summarized by saying that not much is known except that the first-order theory does have the independence property [\textit{I. Kaplan} et al., Isr. J. Math. 185, 141--153 (2011; Zbl 1261.03120)], and that its existential part is conditionally decidable assuming resolution of singularities in characteristic \(p\), in the usual language of rings augmented by a constant for the indeterminate \(t\) [\textit{J. Denef} and \textit{H. Schoutens}, in: Valuation theory and its applications. Volume II. Proceedings of the international conference and workshop, University of Saskatchewan, Saskatoon, Canada, July 28--August 11, 1999. Providence, RI: American Mathematical Society (AMS). 43--60 (2003; Zbl 1046.12006)]. In this paper, relying mainly on the work of \textit{F.-V. Kuhlmann} [J. Reine Angew. Math. 719, 1--43 (2016; Zbl 1401.13011)], the authors prove transfer theorems for the existential theories of equicharacteristic Henselian valued fields. They work in a standard basic three-sorted language for valued fields. The main result is that two equicharacteristic Henselian valued fields have the same existential theory iff their residue fields have the same existential theory. In contrast with previous state of affairs, the situation is very different in Henselian valued fields of mixed characteristic and the authors give examples where the above theorem fails. They also show that the existential theory of an equicharacteristic Henselian valued field is decidable iff the existential theory of its residue field is. It follows that the existential theory of the valued field \(\mathbb F((t))\) is decidable (note that this is unconditionally). The authors compare their work and the presence of a constant symbol for \(t\) in the work of Denef-Schoutens in the following way: they show that the existential theory of \(\mathbb F((t))\) in the language of valued fields with the extra constant for \(t\) is equivalent to its universal-existential theory without the extra constant but where formulas have only one universal quantifier ranging only over the base valued field.