Remarks on a conjecture of Lang (Q614480): Difference between revisions

Lang's conjecture predicts a lower bound for the canonical height of non-torsion points on elliptic curves over number fields. Silverman formulates a generalization for abelian varieties. Let \(g\geq 1\), \(k\) be a number field and \(A/k\) be an abelian variety of dimension \(g\). According to Lang-Silverman's conjecture there exists a constant \(c=c(k,g)>0\) such that for every point \(P\in A(k)\) satisfying \(\overline{\mathbb{Z}\cdot P}=A\) we have \[ \hat{h}(P)\geq c\,h_F(A/k)\,. \] Here \(\hat{h}(.)\) is the canonical height associated to any ample symmetric divisor from \(\text{Div}(A)\) and \(h_F(A/k)\) is the Faltings height of the abelian variety \(A\). The main theorem in this paper gives an asymptotic result on the height of Heegner points on the modular Jacobian \(J_{0}(N)\), with a proof building on an identity from [\textit{B. H. Gross} and \textit{D. B. Zagier}, Invent. Math. 84, 225--320 (1986; Zbl 0608.14019)] and an estimate from [\textit{P. Michel} and \textit{E. Ullmo}, Invent. Math. 131, No. 3, 645--674 (1998; Zbl 0991.11037)]. As a corollary the author shows that the constant in the Lang-Silverman's conjecture cannot be absolute but depend on the dimension \(g\) of the abelian variety \(A\). Furthermore, he recalls a classical fact that the constant \(c\) should depend on the field \(k\), and he constructs a non-trivial counterexample showing that the condition \(\overline{\mathbb{Z}\cdot P}=A\) is also necessary in the formulation of Lang-Silverman's conjecture.