Abelian varieties and algebraic independence. I (Q1173115)

Let \(\wp\) be an elliptic function of Weierstrass with algebraic invariants \(g_2, g_3\), \(\beta\) an algebraic number of degree \(d>2\) over the field \(k\) of endomorphisms of \(\wp\), and \(\wp\) a complex number such that \(u, \beta u, \ldots, \beta^{d-1}u\) are not poles of \(\wp\). One conjecture is that the transcendence degree of the field generated by \(\wp(u),\wp(\beta u),\ldots, \wp(\beta^{d-1}u)\) over \(\mathrm{Q}\) is at least \(d-1\). \textit{D. W. Masser} and \textit{G. Wüstholz} gave a lower bound for this transcendence degree which is a logarithmic function of \(d\) [Invent. Math. 72, 407--464 (1983; Zbl 0516.10027)]. For the exponential analog (problem of Gel'fond-Schneider), the currently best known estimate is also a function of the order of magnitude of \(\log d\) [see \textit{Yu. V. Nesterenko}, Approximations diophantiennes et nombres transcendants, Prog. Math. 31, 199--220 (1983; Zbl 0513.10035)]. Here, for the first time, an estimate is achieved with a linear function of \(d\). The results were already announced by the author [C. R. Acad. Sci., Paris, Sér. I 294, 257--259 (1982; Zbl 0498.14009)]. One main tool is a very interesting new criterion for algebraic independence, involving commutative algebra like in Nesterenko's work. Another essential idea is to use the action of endomorphisms for a power of the variety in a rather subtle way. Recently the author sharpened his criterion (thereby solving his problem p.300), and as a consequence, refined the results of this paper. In part II [Variétés abéliennes et indépendance algébrique. II: Un analogue abélien du théorème de Lindemann-Weierstrass, Invent. Math. 72, 389--405 (1983; Zbl 0516.10026)], the author describes another application of his method.