On Artin's Primitive Root Conjecture for Function Fields over $\mathbb{F}_{q}$

Abstract

In 1927, E. Artin proposed a conjecture for the natural density of primes $p$ for which $g$ generates $(\mathbb{Z}/p\mathbb{Z})^\times$. By carefully observing numerical deviations from Artin's originally predicted asymptotic, Derrick and Emma Lehmer (1957) identified the need for an additional correction factor; leading to a modified conjecture which was eventually proved to be correct by Hooley (1967) under the assumption of the generalised Riemann hypothesis. An appropriate analogue of Artin's primitive root conjecture may moreover be formulated for an algebraic function field $K$ of $r$ variables over $\mathbb{F}_{q}$. Relying on a soon to be established theorem of Weil (1948), Bilharz (1937) provided a proof in the particular case that $K$ is a global function field (i.e. $r=1$), which is correct under the assumption that $g \in K$ is a $\textit{geometric}$ element. Under these same assumptions, Pappalardi and Shparlinski (1995) established a quantitative version of Bilharz's result. In this paper we build upon these works by both generalizing to function fields in $r$ variables over $\mathbb{F}_{q}$ and removing the assumption that $g \in K$ is geometric; thereby completing a proof of Artin's primitive root conjecture for function fields over $\mathbb{F}_{q}$. In doing so, we moreover identify an interesting correction factor which emerges when $g$ is not geometric. A crucial feature of our work is an exponential sum estimate over varieties that we derive from Weil's Theorem.