Lectures on automorphic \(L\)-functions (Q1889647): Difference between revisions

This book consists of a collection of three papers with a common theme. Starting in the 1960s R.~P.~Langlands formulated a set of ideas concering the general theory of automorphic forms which is now known, along with some further ones, under his name. The central concept in this complex was that of ``functoriality''. This asserts that the set of automorphic representations has a rich internal structure. This theory is rather more analytic or representation-theoretic than arithmetic in nature, although one of its aspects, base-change, is inherently arithmetic. Another complex of ideas, now, but less accurately, associated with the name of Langlands, is that the ``arithmetic'' cohomology of an algebraic variety defined over a number field (or a function field of arithmetic type) should split into atomic components (motives) and these should themselves be associated with automorphic forms. These questions are much more arithmetic in nature and, in as far as they can be answered at the present time, depend on far-reaching extensions of the classical theory of complex multiplication (Shimura varieties). There have been two approaches taken to study functoriality, the method of \(L\)-functions and the Selberg Trace Formula. Between them a certain atmosphere of competition has prevailed. Both approaches have proved to be technically very demanding. One can ask, before going further, what the immediate consequences of functoriality would be if it were proved. Leaving aside questions of base-change, that is, the most arithmetic aspect, the major applications would be to the proof of the generalized Ramanujan Conjecture and its archimedean analogue, the (generalized) Selberg Conjecture and certain results (conjectures of Sato-Tate type) on the distribution of eigenvalues of Hecke operators. Is that all? Not quite. The major hope, and it is only a hope, is that when functoriality has been properly understood the arithmetic questions can be investigated in some degree of generality, and that these will lead to a better understanding of the arithmetic of varieties. This is thinking in the long term, like planting trees in a garden that will mature in 100 years. The book under review consists of three papers. The first, by J.~ W.~ Cogdell, is entitled ``Lectures on \(L\)-functions, Converse Theorems and Functoriality for GL\((n)\)''. The theory of converse theorems, developed in the first place by I.~I.~Piatetski-Shapiro, is the basic tool in using \(L\)-functions to study functoriality. These theorems allow one to pass from \(L\)-functions to automorphic representations of general linear groups. Here Cogdell gives a survey of the present state of the art. Of course one also has to have techniques for establishing the analytic properties of the relevant \(L\)-functions. One of these methods, also due to Piatetski-Shapiro, is the method of Eulerian subgroups, where the aim is to construct generalizations of the Rankin--Selberg integral which also have an Euler product structure. Cogdell sketches this method and indicates how it can be applied to study certain ``lifts'' from classical groups. These applications can be used to study the Ramanujan and Selberg Conjectures. There is a separate method of determining the analytic properties of \(L\)-series; this was first used by Langlands and has been developed by F.~Shahidi, more recently in collaboration with H.~Kim. This involves using as much information as possible from the theory of Eisenstein series; the analytic properties follow from those of the Eisenstein series. In the second paper, in this book, ``Automorphic \(L\)-functions'', Kim describes in detail how this method works and applies it combined with converse theorems to proving functoriality for the ``symmetric cube'' and ``symmetric fourth power'' of automorphic representations of GL(2). The final paper by M.~Ram~Murty, ``Applications of Symmetric Power \(L\)-functions'' describes how the results of the other two papers can be used to investigate the questions concerning classical and Maaß\ modular forms. The first part of this paper is devoted to studying the nature of the Fourier coefficients of such forms, for example, of the degree to which various values are assumed within the range allowed by the Ramanujan conjecture. Since the results on symmetric powers can be regarded as components of a proof of a Sato-Tate conjecture, partial results lead to partial but nevertheless interesting consequences. One of these is an elegant observation made by P. Sarnak that if a reasonable approximation to the Sato-Tate Conjecture holds then the coefficients of a Maaß\ form cannot be integral; from this he deduced that an automorphic representation of GL(2) over a number field whose \(L\)-function has integral coefficients is actually an Artin \(L\)-function of a Galois representation with a solvable image. Other topics dicussed are of approximations to the Selberg and Ramanujan Conjectures for GL\((n)\), for which the best estimates at the present time are due to H.~Kim and P.~Sarnak and applications to the theory of Artin \(L\)-functions. This book provides a stimulating survey of an important part of modern number theory which has experienced a number of spectacular advances in recent years.