A prime geodesic theorem for SL₄

Selberg zeta functions and regularized determinants; applications to spectral theory, Dirichlet series, Eisenstein series, etc. (explicit formulas) (11M36) Class numbers, class groups, discriminants (11R29)

Abstract: A prime geodesic theorem is proven for singular geodesics in quotients of SL(4). This is a case where regularity assumptions of previous papers fail. As a consequence, the analysis becomes much more involved. For applications in number theory (class number asymptotics) it is, however, necessary to consider this case, too.

Review text

The main theorem of this complicated paper is elegant and simply stated: Theorem 4.1.1 (Prime Geodesic Theorem) Assume that \(\Gamma\) is regular or of SL-type. For \(x\in\infty\) we have \(\pi(x)\sim{2x\over \log x}\). More sharply, \(\pi(x)= 2\,\text{li}(x)+ O({x^{3/4}\over\log x})\) as \(x\to\infty\), where \(\text{li}(x)= \int^x_2{1\over\log t} \,dt\) is the integral logarithm. (Here \(\Gamma\) is a discrete co-compact subgroup of \(\text{SL}_4(\mathbb{R})\) and the definitions of regular and SL-type, two involved for this summary, are given on pages 173, 174, resp.) The introduction offers a glimpse of the new difficulties arising in the treatment of the groups \(\Gamma\) above: ``Our strategy is that we define a suitable zeta function and use the trace formula with special test-functions to derive analytic continuation as in [\textit{A. Juhl}, Cohomological theory of dynamical zeta functions, Progress in Mathematics 194. Basel: Birkhäuser (2001; Zbl 1099.37017)] or [(*) \textit{A. Deitmar}, Am. J. Math. 122, No. 5, 887--926 (2000; Zbl 0999.22015)] so that we can use the techniques of the proof of the prime number theorem. It is in the analytic continuation that we encounter the first obstacle. We say that an element \(g\in G= \text{SL}_4(\mathbb{R})\) is weakly neat [\textit{A. Borel}, Introduction aux groupes arithmétiques. Paris: Hermann (1969; Zbl 0186.33202)] if the adjoint \(\text{Ad}(g)\) has no non-trivial roots of unity as eigenvalues. A subgroup of \(G\) is weakly neat if every element is. The results of [(*)] use the assumptions that the group \(G\) has trivial centre and the group \(\Gamma\) is weakly neat. Note that, if \(G\) has trivial centre then \(\Gamma\) weakly neat implies \(\Gamma\) torsion-free, since any non-weakly neat torsion elements must be central. To generalise the results of [(*)] the case where \(\Gamma\) is not weakly neat requires two things. First, we need to extend the definition of the zeta function to the more general situation when \(\Gamma\) is not necessarily weakly neat. Second, the definition of the zeta function includes the first higher Euler characteristic of the spaces \(X_{\Gamma_\gamma}\) for \(y\in{\mathcal E}(\Gamma)\). In [(*)] these are only defined for \(\Gamma_\gamma\) torsion-free so we need to broaden the definition. The writing style of this paper renders it best suited to well-versed experts. There is little exposition of previous works, and important definitions are embedded in the text (albeit in italics). More casual readers will want ready access to the extensive Bibliography. The application to the class number issues announced in the abstract are reserved for a later paper.

Recommendations

Cites work

Cited in

(6)

This page was built for publication: A prime geodesic theorem for \(\mathrm{SL}_4\)

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2481347)