Diophantine subsets of function fields of curves (Q1005866)

From MaRDI portal
scientific article
Language Label Description Also known as
English
Diophantine subsets of function fields of curves
scientific article

    Statements

    Diophantine subsets of function fields of curves (English)
    0 references
    0 references
    16 March 2009
    0 references
    In number theory, a Diophantine set in \({\mathbb Z}^k\) is a subset \(D \subset {\mathbb Z}^k\) for which there exists a polynomial \(F(x;y) = F(x_1,\dots,x_k; y_1,\dots,y_n)\) with integer coefficients such that \(F(x;y) = 0\) has a solution with integer \(x_i\) and natural \(y_j\) iff \(x = (x_1,\dots,x_k) \in D\). By the theorem of Matiyasevich, the Diophantine sets in \({\mathbb Z}^n\) are the same as the recursively enumerable sets in \({\mathbb Z}^k\). In this paper are studied basic questions regarding certain generalized Diophantine sets. By the definition on page 209, a subset \(D \subset R\) is Diophantine if there are polynomials \(F_i(x;y_1,\dots,y_n) \in R[x;y_1,\dots,y_n]\) such that the system \(F_i(x;y_1,\dots,y_n) = 0, \forall i\) has a solution \((y_1,\dots,y_n) \in R^n\) iff \(x \in D\). The main question about such generalized Diophantine sets is to find a similar characterization as in case regarded above. However this question seems to be hard in any particular nontrivial case; for example it is not known whether \({\mathbb Z}\) is a Diophantine subset of the rationals \(R = {\mathbb Q}\), see the introduction. In this paper is studied in more detail the case when \(R = k(t)\) where \(k\) is an uncountable field of characteristic \(0\). Theorem 4, the main technical result of the paper, shows that a Diophantine subset \(D \subset k(t)\) is, roughly speaking, either very small or very large. This yields two impressive conclusions: Corollary 1. \(k[t]\) is not a Diophantine subset of \(k(t)\); Corollary 2. If \(K_2 \supset K_1 \supset k(t)\) are finite field extensions, then \(K_1\) is a Diophantine subset of \(K_2\) iff \(K_1 = K_2\). Especially, the 2nd Corollary answers in this case to a question of Bogomolov: When a subfield \(K_1 \subset K_2\) is Diophantine in \(K_2\) ? The proof of Theorem 4 relies on the theory of rational curves on varieties, to which the author has the main contribution, see [\textit{J. Kollár}, Rational curves on algebraic varieties. Berlin: Springer Verlag (1995; Zbl 0877.14012)].
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    Diophantine set
    0 references
    rationally connected variety
    0 references
    0 references
    0 references