Jet spaces of varieties over differential and difference fields (Q1879034): Difference between revisions

The paper deals with finite dimensional differential algebraic varieties and difference algebraic varieties, aiming at singling out under which conditions these varieties are {``algebraic''}, in the sense that they entirely arise from algebraic geometry. Language and notions from model theory intervene in this perspective (see the statements of the main theorems below); but, in spite of this, the declared purpose of the authors is just to avoid any crucial use of model theoretic tools in their proofs. Indeed their treatment has an {``elementary''} flavour and employ a generalized Gauss jet map together with the theory of linear differential (difference) equations. In more detail, in the differential case it is shown that, if \(U\) is a big differentially closed field \(U\) of characteristic 0, \(X\) is a finite dimensional definable subset of \(U\) defined over a countable algebraically closed differential subfield \(k\) of \(U\), \(a \in X\), \(k'\) is a countable algebraically closed differential subfield extending \(k\) and \(c\) denotes a canonical base of the type of \(a\) over \(k'\), then the type of \(c\) over \(k \langle a \rangle\) is internal to the field of constants of \(U\). Similarly, in the difference case, an existentially closed difference field \(U\) is considered (again in characteristic 0) and it is proved that, if \(X\) is a finite dimensional definable set, defined over an algebraically closed difference subfield \(k\) of \(U\), \(a \in X\), \(k'\) is an algebraically closed difference subfield of \(U\) containing \(k\) and \(c\) denotes, as before, the canonical parameter of the type of \(a\) over \(k'\), then the type of \(c\) over \(k \langle a \rangle\) is almost internal to the fixed field of \(U\). The positive characteristic case is also partly considered. Finally, the authors examine the impact of their methods and results on the Hrushovski model theoretic approaches to the Mordell-Lang conjecture for function fields in all characteristics and to the Manin-Mumford conjecture over number fields. Again, the hope is to eliminate in those proofs any crucial use of model theoretic ideas. The authors discuss this purpose mainly in the characteristic 0 case.