Abelian varieties over large algebraic fields with infinite torsion (Q273115): Difference between revisions

\textit{W.-D. Geyer} and \textit{M. Jarden} [Isr. J. Math. 31, 257--297 (1978; Zbl 0406.14025)] proved that if \(K\) is a finitely generated field, then for almost all \(\sigma\) in the absolute Galois group of \(K\), and for all elliptic curves defined over \(\bar{K}(\sigma)\), the torsion subgroup of \(E(\bar{K}(\sigma))\) contains a point of order \(\ell\) for infinitely many primes \(\ell\). Over here, \(\bar{K}(\sigma)\) denotes the fixed field of \(\sigma\), and `almost all' means outside a set of Haar measure \(0\). Furthermore, they conjectured that these statements remain true if the elliptic curve \(E\) is replaced by any abelian variety. In [\textit{W.-D. Geyer} and \textit{M. Jarden}, Finite Fields Appl. 11, No. 1, 123--150 (2005; Zbl 1088.12001)], a weak form of this conjecture was proved, namely that if \(A\) is an abelian variety over a number field \(K\), then there exists a finite Galois extension \(L\) of \(K\), such that \(A(\bar{L}(\sigma))\) contains points of order \(\ell\) for infinitely many primes \(\ell\) for almost all \(\sigma\) in the absolute Galois group of \(L\). In the paper under review, this result is strengthened to show that for an abelian variety \(A\) over a number field \(K\), and for almost all \(\sigma\) in the absolute Galois group of \(K\), the torsion subgroup of \(A(\bar{K}(\sigma))\) contains points of order \(\ell\) for infinitely many primes \(\ell\).