\(\text{Pos}_n(\mathbb R)\) and Eisenstein series (Q2386503)

This book treats a family of Dirichlet series in several variables associated with positive definite quadratic forms; they were introduced by \textit{A. Selberg} in 1959 [see Number theory, trace formulas and discrete groups, Symp. in Honor of Atle Selberg, Oslo/Norway 1987, 467--484 (1989; Zbl 0675.10030), pp. 473--474]. Although they are regarded today primarily as Eisenstein series associated with general linear groups Selberg emphasized that they were generalizations of Epstein's zeta function and appears to have expected that they would also have direct arithmetical applications in the study of quadratic forms. Selberg only sketched the main results and has not since returned to this topic. The subject was taken up by \textit{R. P. Langlands} [On the functional equations satisfied by Eisenstein series. Lecture Notes in Mathematics. 544. (Berlin: Springer) (1976; Zbl 0332.10018), Appendix 1 which apparently dates from 1962], by H. Maaß\ [Sém. Delange-Pisot-Poitou 12 (1970/71), Théorie Nombres, No. 22, 18 p. (1972; Zbl 0244.10028), published in 1971] and by \textit{A. Terras} [Nagoya Math. J. 42, 173--188 (1971; Zbl 0212.07702), and Nagoya Math. J. 44, 89--95 (1971; Zbl 0212.07801), also published in 1971]. This method consists of several steps. To define the functions in question of all one needs the basic reduction theory of quadratic forms and the theory of eigenfunctions of the algebra of invariant operators, a theory mainly due to S. Bochner. These are applied to certain theta series which are sums over a lattice. Selberg gave a novel interpretation of a transformation used by Riemann in his paper on prime numbers to show that one could remove terms where the determinant, a polynomial function on this lattice, is zero by the application of a differential operator. This means that one could apply a Mellin transform to represent the Dirichlet series in question. This representation does not quite suffice to give the full analytic continuation with the functional equations. Selberg noted that it could be supplemented by an application of Bochner's Tube Theorem to achieve these ends. (Langlands, who carries out this argument from first principles, refers to this as `` involving a form of Hartog's [sic] lemma''. He also indicates (loc.cit., p. iv) that Godement also found the same arguments independently of Selberg, but his work does not seem to have been published.) The purpose of the book under review is to give a crisp version of this theory. It is developed \textit{ad ovo} -- in other words all of the necessary reduction theory and harmonic analysis is developed simply and efficiently. Although some of the details are, by their nature, intricate, this book reduces these to a minimum and provides an accessible account to this important and beautiful area of mathematics. Although the standard methods for treating Eisenstein series today replace theta series by the functional-analytic methods (such as that due to Roelcke and Selberg, as in Langlands' book quoted above), Langlands' admonition, that this method ``should not be forgotten'' (loc.cit., p. iv) remains just as valid now as it was thirty years ago. Indeed it has been experiencing a renaissance recently and promises to be fruitful in related areas. One final word - the reviewer received this book for review almost simultaneously with the sad news of the death of Serge Lang (12th September, 2005). The loss of this unique figure in the mathematics of the last half-century has made the preparation of this review a poignant task.