Algebraic varieties over PAC fields (Q2480585): Difference between revisions

The main object of the paper under review is a pseudo algebraically closed (PAC) field, i.e. a field \(k\) such that every geometrically integral \(k\)-variety has a \(k\)-point. The author proves two theorems on such fields thus solving two open problems from the monograph of \textit{M.~Fried} and \textit{M.~Jarden} [Field Arithmetic. 2nd revised and enlarged ed. Berlin: Springer (2005; Zbl 1055.12003)]. Theorem 1 gives a new characterization of PAC fields: \(k\) is PAC if and only if every absolutely irreducible homogeneous polynomial \(f(x,y,z)\in k[x,y,z]\) has a nontrivial zero in \(k^3\). The proof reduces to showing that if \(k\) is not PAC then there is a geometrically integral plane projective \(k\)-curve without rational points. This is proved in several steps. First, finite fields are excluded. Then, assuming the existence of a pointless geometrically integral \(k\)-variety \(X\), it is proved that there exists a pointless nonsingular projective \(k\)-curve. Then, using general projections, one gets a plane curve with at most one \(k\)-point, and, finally, this point is removed using a double cover. Theorem 2 asserts that if \(X\) is a geometrically integral \(k\)-variety over a PAC field \(k\), then \(X(k)\) is dense in \(X(\bar k)\) in the \(v\)-adic topology, where \(\bar k\) is an algebraic closure of \(k\) and \(v\) is a nontrivial valuation of \(\bar k\). The proof is based on the reduction to the case where \(X\) is a plane curve and using a trick of Prestel (invented for treating the case of the projective line).