Rethinking the work of Langlands on Eisenstein series
Devadatta G. Hegde · 2026
Read Rethinking the work of Langlands on Eisenstein series by Devadatta G. Hegde (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-22475.txt.
Introduction
Hegde rethinks Langlands' construction of the discrete automorphic spectrum. Hegde shows it regularises cuspidal Eisenstein series at distinguished points, tracking both zeros and poles. Unlike single-variable functions, zero and pole sets can meet. Those points behind the spectrum are often of this kind. Zeros, hidden in rank one, star in higher rank, on equal footing with poles. Redoing Langlands' G2 work with zeros tracked cuts it to plain maths. A program rethinks it from scratch, giving zeros and poles equal standing.
Catalog record
| CREATOR | Devadatta G. Hegde |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 72461 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.22475
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Rethinking the work of Langlands on Eisenstein series Devadatta G. Hegde Abstract . Chapter 7 7 of Langlands’ monograph On the functional equations satisfied by Eisenstein series employs a sophisticated residue scheme to construct a portion of the discrete automorphic spectrum. We show, by examples, applications, and heuristics, that this construction is a straightforward regularization of cuspidal Eisenstein series at distinguished points, and that the regularization must track both the zeros and the poles of these Eisenstein series. Unlike the one-variable case, the zero set and pole set of a several-complex-variable meromorphic function can intersect at a point. The distinguished points supporting the discrete spectrum are typically of this kind. The zeros of cuspidal Eisenstein series — largely invisible in the rank-one case — begin to play a starring role in higher rank situations, on equal footing with the poles. We redo Langlands’ famous G 2 G_{2} calculation and show that once zeros are tracked, the calculation reduces to elementary algebra. Drawing on rank-two examples, we introduce a program to re-think Langlands’ construction from first principles, giving zeros and poles of cuspidal Eisenstein series equal standing from the very beginning. The program has the advantage of making the underlying phenomenon transparent, though carrying it out in full generality will require substantial further work. Organization of the notes These notes are organized so that each section can be read on its own, with the SL(2) , SL(3) , and G_{2} examples (sections 2 , 3 , 4 ) forming the technical core. Section 1 describes two theorems on the poles of certain Eisenstein series. The first, due to the author [ 6 ] , concerns the poles of degenerate Eisenstein series. The second concerns the poles of the most general GL(n) Eisenstein series relevant for spectral decomposition; it can be derived by the methods of the author’s thesis, and is a reformulation, in more natural terms, of an obscure theorem of Moeglin–Waldspurger [ 16 ] . Together these results suggest a new theme: that the poles of Eisenstein series admit simple descriptions in terms of Arthur parameters. Looking for the origin and generalization of this pattern is what opened the door to a re-examination of Langlands’ construction of the residual spectrum. Section 2 rethinks the basic case of SL(2,\mathbb{Z})\backslash SL(2,\mathbb{R})/SO(2,\mathbb{R}) . The goal is to introduce the contour-deformation argument in its simplest setting and to clarify the relationship between Laurent coefficients of Eisenstein series at poles and spectral expansions. Carrying out the deformation under the a priori assumption that the poles need not be simple, one finds that simplicity is forced: without it, the deformation does not yield a spectral expansion. This gives a fresh perspective on the standard result that maximal-parabolic cuspidal Eisenstein series of reductive groups have simple poles in the right half-plane, and identifies a relationship — between Laurent coefficients and spectral expansions — whose lessons apply more broadly. Section 3 reworks the SL(3) calculation from Langlands’ Boulder paper [ 9 ] . The goal is to carry out higher-rank contour deformation in the simplest case, illustrate the role of zeros of Eisenstein series, and clarify a certain "turbid lemma" of Langlands. We treat this example completely, including the one-dimensional spectrum, so that the next section can focus on the discrete spectrum of G_{2} alone. The justification of the contour deformation requires a certain cancellation; a critical look at this cancellation, in § 3.11 , suggests that the locations of the poles of cuspidal Eisenstein series are sharply restricted. Section 4 reworks the G_{2} calculation from Appendix III of [ 11 ] . Langlands offered the example to help the reader navigate Chapter 7 of his monograph, and observed in his commentary that the complications arising in it could not be eliminated "unless a procedure radically different from that of the text be found." We carry out the calculation by tracking both the zeros and the poles of the corresponding Borel Eisenstein series. With this addition, the calculation proceeds by elementary algebra. The two-dimensional residual spectrum identified by Langlands is recovered in the form one expects from the SL(3) analysis of the previous section: as regularizations of the Borel Eisenstein series at two distinguished points. Section 5 outlines a program for rethinking the "jungle of Paragraph §7" with equal attention paid to the zeros and poles of cuspidal Eisenstein series from the outset. 1. New results on poles of Eisenstein series 1.1. Poles of degenerate Eisenstein series Notation Let • G be a connected split semisimple algebraic group over a number field F . • B=U\rtimes T be a fixed Levi decomposition of a fixed Borel subgroup of G . • P=N\rtimes M , with T\subset M , be the standard Levi decomposition of a maximal parabolic subgroup P\supset B of G . • [G]=G(F)\backslash G(\mathbb{A}) be the automorphic space. • \delta_{P}:P(\mathbb{A})\to(0,\infty) be the modulus character. • \varpi_{P} be the fundamental weight associated with P and |\varpi_{P}|:P(\mathbb{A})\to(0,\infty) be the corresponding character. • \mathbf{K}=K_{\infty}\prod_{v<\infty}K_{v}\subset G(\mathbb{A}) : a maximal compact subgroup. Here, each K_{v} at a finite place v is hyperspecial and the Iwasawa decomposition G(\mathbb{A})=P(\mathbb{A})\mathbf{K} holds. Degenerate Eisenstein series Extend \delta_{P},|\varpi_{P}|:G(\mathbb{A})\to(0,\infty) as \mathbf{K} -spherical functions. The degenerate Eisenstein series E_{P}(s;g)=\sum_{\gamma\in P(F)\backslash G(F)}\left(\delta_{P}^{\frac{1}{2}}\cdot|\varpi_{P}|^{s}\right)(\gamma g) is the simplest example of an Eisenstein series. This series converges absolutely for \Re(s) sufficiently large and uniformly on compact subsets of [G] . The map s\mapsto E_{P}(s):=E_{P}(s;\bullet)\in C^{\infty}\left([G]\right) admits meromorphic continuation to \mathbb{C} . Poles The poles of E_{P}(s) in the region \Re(s)\geq 0 are described via the principal homomorphism \kappa:SL(2,\mathbb{C})\to\widehat{M} and the adjoint action of \widehat{M} on \widehat{\mathfrak{n}}=\text{Lie}(\widehat{N}) , where \widehat{\bullet} denotes the Langlands dual. Decompose (Langlands-Shahidi) \widehat{\mathfrak{n}}=r_{1}\oplus\cdots\oplus r_{m} into irreducible representations r_{j} of \widehat{M} indexed in the standard order. Restricting via \kappa , r_{j}=\bigoplus_{\ell\geq 1}V_{\ell}^{\oplus m_{j}(\ell)},\quad m_{j}(\ell)\in\mathbb{Z}_{\geq 0} where V_{\ell} is the \ell -dimensional irreducible representation of SL(2,\mathbb{C}) . The following theorem is proved in the author’s doctoral thesis [ 6 ] . Theorem 1 . Define \mathsf{Q}(s)=\prod_{j=1}^{m}\prod_{\ell\geq 1}\left(\frac{js-\frac{\ell+1}{2}}{js+\frac{\ell-1}{2}}\right)^{m_{j}(\ell)} Then \mathsf{Q}(s)\cdot E_{P}(s) is holomorphic and nonzero for \Re(s)\geq 0 (hence the zeros of \mathsf{Q} give the poles of E_{P}(s) , including multiplicities). 1.2. Poles of GL(n) Eisenstein series Let (n_{1},\dots,n_{k}) be a partition of n=n_{1}+\cdots+n_{k} , with n_{i}=a_{i}d_{i} for a_{i},d_{i}\in\mathbb{Z}_{\geq 1} . To show that the above result is not isolated, we describe the poles of all GL(n) Eisenstein series that play a role in the spectral decomposition of GL(n) automorphic forms. • G=GL(n) and P=N\rtimes M be the standard Levi decomposition of the parabolic corresponding to (n_{1},\dots,n_{k}) . Then \widehat{M}=GL(n_{1},\mathbb{C})\times\cdots\times GL(n_{k},\mathbb{C}) • For a,d\in\mathbb{Z}_{\geq 1} and \tau a cuspidal representation of GL(d,\mathbb{A}) , let \mathcal{E}(\tau,a) be the Speh representation for GL(ad,\mathbb{A}) with Arthur parameter A_{\mathcal{E}(\tau,a)}=A_{\tau}\boxtimes V_{a}:\mathscr{L}_{F}\times SL(2,\mathbb{C})\to GL(ad,\mathbb{C}) where \mathscr{L}_{F} is the (conjectural) Langlands group and V_{a} is the a -dimensional irreducible representation of SL(2,\mathbb{C}) . Here A_{\tau}:\mathscr{L}_{F}\to GL(d,\mathbb{C}) is the (conjectural) Langlands parameter attached to \tau . • Let \pi\simeq\boxtimes_{i=1}^{k}\mathcal{E}(\tau_{i},a_{i}) where \tau_{i} is a cuspidal representation of GL(d_{i},\mathbb{A}) . Its Arthur parameter is A_{\pi}=\bigoplus_{i=1}^{k}A_{\mathcal{E}(\tau_{i},a_{i})}:\mathscr{L}_{F}\times SL(2,\mathbb{C})\to\widehat{M} The adjoint action gives \widehat{\mathfrak{n}}=\bigoplus_{(i,j):1\leq i<j\leq k}\widehat{\mathfrak{n}_{ij}}, where \widehat{\mathfrak{n}_{ij}}\simeq\left(A_{\tau_{i}}\boxtimes V_{a_{i}}\right)\otimes\left(A_{\tau_{j}}\boxtimes V_{a_{j}}\right)^{\vee} \simeq\left(A_{\tau_{i}}\otimes A_{\tau_{j}}^{\vee}\right)\boxtimes\left(V_{a_{i}}\otimes V_{a_{j}}\right). By Clebsch-Gordan, V_{a_{i}}\otimes V_{a_{j}}\simeq\bigoplus_{\ell\geq 1}V_{\ell}^{\oplus m_{ij}(\ell)} where m_{ij}(\ell)\in\mathbb{Z}_{\geq 0} . Let \alpha_{ij}^{\vee} denote the coroot associated to the pair (i,j) . For \Lambda=(s_{1},\dots,s_{k})\in\check{\mathfrak{a}}_{P,\mathbb{C}} , set \langle\Lambda,\alpha_{ij}^{\vee}\rangle:=s_{i}-s_{j} and define the closed positive tube