Symmetric power functoriality for holomorphic modular forms

From MaRDI portal
Publication:2067686



Abstract: Let f be a cuspidal Hecke eigenform of level 1. We prove the automorphy of the symmetric power lifting mathrmSymnf for every ngeq1. We establish the same result for a more general class of cuspidal Hecke eigenforms, including all those associated to semistable elliptic curves over mathbbQ.


Langlands' principle of functoriality, linking automorphic representations and Galois representations, predicts a theory of liftings of automorphic representations. In the function field case, for \(\mathrm{GL}(n)\), this follows from L. Lafforgue's proof of the principle. An example is the symmetric square lifting, \(\mathrm{Sym}^n\), from \(\mathrm{GL}(2)\) to \(\mathrm{GL}(n+1)\). In the number field case it would imply Ramanujan's conjecture. The initial case of \(n=2\) was done on using the theory of \(L\)-functions by \textit{S. Gelbart} and \textit{H. Jacquet} [Ann. Sci. Éc. Norm. Supér. (4) 11, No. 4, 471--542 (1978; Zbl 0406.10022)], and independently on using the trace formula by the reviewer in a series of papers; see Part 1 of his book [Automorphic representations of low rank groups. Hackensack, NJ: World Scientific (2006; Zbl 1191.11017)]. The later is in fact from \(\mathrm{SL}(2)\) to \(\mathrm{PGL}(3)\). It implies multiplicity one theorem for\( \mathrm{SL}(2,\mathbb A\)), \(\mathbb A\) denotes the ring of adèles of the number field \(F\). More recently \textit{L. Clozel} and \textit{J. A. Thorne} [Compos. Math. 150, No. 5, 729--748 (2014; Zbl 1304.11040); Ann. Math. (2) 181, No. 1, 303--359 (2015; Zbl 1339.11060); Duke Math. J. 166, No. 2, 325--402 (2017; Zbl 1372.11054)] proved the automorphy of \(\mathrm{Sym}^m(\pi)\) for \(m\le 8\) when ``the number field \(F\) is totally real and the automorphic representation \(\pi\) is regular algebraic, in the sense of [\textit{L. Clozel}, in: Automorphic forms, Shimura varieties, and L-functions. Vol. I, Proc. Conf., Ann Arbor/MI (USA) 1988, Perspect. Math. 10, 77--159 (1990; Zbl 0705.11029)]. This includes the most classical case of automorphic representations arising from holomorphic modular forms of weight \(k\geq 2\). The strategy in this interesting paper is to use properties of the Coleman-Mazur eigencurve. It concern the field of rational numbers \(F=\mathbb Q\), thus \(\mathbb A\) is \(\mathbb A_{\mathbb Q}\), \(n\ge 2\), and a ``regular algebraic cuspidal automorphic representation \(\pi\) of \(\mathrm{GL}(2,\mathbb A)\) of level 1 (i.e., which is everywhere unramified). So the non-Archimedean components \(\pi_v\) of \(\pi\) are unramified. The initial result, Theorem D, asserts that if ``the \(n\)th symmetric power lifting exists for one such \(\pi\), it exists for any other such \(\pi\). Theorem E asserts it exists for some such \(\pi\). Theorem A is the obvious join: ``If \(\pi\) is a regular algebraic cuspidal automorphic representation of \(\mathrm{GL}(2,\mathbb A_{\mathbb Q})\) of level 1, then for each integer \(n\geq 2\) the symmetric power lifting \(\mathrm{Sym}^{n-1}\pi\) exists as a regular algebraic cuspidal automorphic representation of \(\mathrm{GL}(n,\mathbb A_{\mathbb Q}\)). Theorem B extends to ramified cases: ``Let \(\pi\) be a regular algebraic cuspidal automorphic representation of \(\mathrm{GL}(2,\mathbb A_{\mathbb Q})\) of conductor \(N\geq 1\), such that there is no quadratic automorphic character \(\chi\) with \(\pi\otimes\chi\simeq\pi\). Suppose that for each prime \(p|N\), the local component \(\pi_p\) is not cuspidal. Then for each \(n\geq 2\), the symmetric power lifting Sym\(^{n-1}\pi\) exists as a regular algebraic cuspidal automorphic representation of \(\mathrm{GL}(n,\mathbb A_{\mathbb Q})\). The reduction of B to A uses ``an argument of `killing ramification' as in the proof of Serre's conjecture. The automorphic representations of Theorem B include ``all those associated to holomorphic newforms of level \(\Gamma_0(N)\) for some squarefree integer \(N\ge1\); in particular those associated to semistable elliptic curves over \(\mathbb Q\). Corollary C then states: ``Let \(E\) be a semistable elliptic curve over \(\mathbb Q\). Then for each \(n\geq 2\) the completed symmetric power \(L\)-function \(\Lambda(\mathrm {Sym}^n E,s)\) admits an analytic continuation to the entire complex plane.



Cites work


Cited in
(only showing first 100 items - show all)








This page was built for publication: Symmetric power functoriality for holomorphic modular forms

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2067686)