Vol. 207, No. 2, 2002

Download This Article
with up-to-date links in citations
Download this article. For Screen
For Printing
Recent Issues
Vol. 243: 1  2
Vol. 242: 1  2
Vol. 241: 1  2
Vol. 240: 1  2
Vol. 239: 1  2
Vol. 238: 1  2
Vol. 237: 1  2
Vol. 236: 1  2
Online Archive
Volume:
Issue:
     
Volumes 1–176are stored at Project Euclid
The Journal
Cover Page
Editorial Board
How To
Submissions Guidelines
Submissions Page
Subscriptions
Elect. License Agreement
Test your IP address
Contacts
To Appear

E. Kowalski & P. Michel

Abstract

For many L-functions of arithmetic interest, the values on or close to the edge of the region of absolute convergence are of great importance, as shown for instance by the proof of the Prime Number Theorem (equivalent to non-vanishing of ζ(s) for mathrmRe(s) = 1). Other examples are the Dirichlet L-functions (e.g., because of the Dirichlet class-number formula) and the symmetric square L-functions of classical automorphic forms. For analytic purposes, in the absence of the Generalized Riemann Hypothesis, it is very useful to have an upper-bound, on average, for the number of zeros of the L-functions which are very close to 1. We prove a very general statement of this type for forms on GL(n)Q for any n 1, comparable to the log-free density theorems for Dirichlet characters first proved by Linnik.

Authors
E. Kowalski
Université Bordeaux I - A2X
351, cours de la Libération
33405 Talence Cedex
France
P. Michel
CC 051, Mathématiques
Université Montpellier II
Place Eugène Bataillon
34095 Montpellier, Cedex 5
France