Motivic pieces of curves: L L -functions and periods

Harry Spencer · 2026

Read Motivic pieces of curves: L L -functions and periods by Harry Spencer (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-2601-21934.txt.

Introduction

Harry Spencer works where number theory meets the geometry of curves. Start with a curve defined over a number field, and suppose a finite group of symmetries acts on it. That setup lets the curve's L-function, a key arithmetic object, split into factors.

Each factor belongs to what Spencer calls a motivic piece of the curve, tied to an irreducible representation of the symmetry group. His 2026 preprint develops these pieces and connects their L-functions with periods, numbers that link geometry to integrals. Splitting a hard object into symmetry-labeled pieces is a classic strategy in modern number theory, and this work sharpens it for curves with group actions.

Catalog record

CREATORHarry Spencer
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE57 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2601-21934

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Motivic pieces of curves: L L -functions and periods Harry Spencer Abstract . Given a curve C C over a number field K K equipped with the action of a finite group G G by K K -automorphisms, one obtains a factorisation of L ⁡ ( C , s ) L(C,s) into a product of L L -functions of ‘motivic pieces of curves’ associated to irreducible G G -representations. We describe an algorithm for explicitly computing values of these L L -functions, demonstrating implementations in the cases of certain curves with actions by C 3 C_{3} , C 4 C_{4} and D 10 D_{10} . We explain how this algorithm can be used to factor L L -functions of curves with endomorphisms of Hecke type. Towards applications, we explicitly formulate and numerically verify a version of Deligne’s Period Conjecture for hitherto-uninvestigated L L -functions arising from motivic pieces of superelliptic curves. [columns=2, intoc] 1. Introduction Given a curve C over a number field K , one associates to it an L -function L(C,s)=\prod_{(0)\neq\mathfrak{p}\hskip 0.28436pt\mathrel{\reflectbox{$\mathrel{\vtop{\halign{#\cr\raisebox{-1.50694pt}{\reflectbox{\rotatebox{90.0}{$\nshortmid$}}}\cr$\triangleright$\cr}}\mkern-3.0mu}$}}\hskip 1.42271pt\mathcal{O}_{K}}L_{\mathfrak{p}}(C,N(\mathfrak{p})^{-s})^{-1}, where L_{\mathfrak{p}}(C,T)=\det(1-\Frob_{\mathfrak{p}}^{-1}\cdot T\mid(T_{\ell}\Jac(C)\otimes_{\mathbb{Z}_{\ell}}\overline{\mathbb{Q}}_{\ell})^{*,I_{\mathfrak{p}}}) for any \ell\nmid\mathfrak{p} and these determinants are computed in practice by counting points on the special fibre of a regular model of C at each prime of \mathcal{O}_{K} . A priori , this is defined only for \Re(s)>3/2 , but the Hasse–Weil Conjecture predicts that L(C,s) admits analytic continuation to all of \mathbb{C} and, famously, the Birch–Swinnerton-Dyer Conjecture (BSD) describes the order of vanishing and leading term of the Taylor expansion at s=1 in terms of arithmetic invariants. If, further, C admits an action by K -automorphisms of a finite group G , then T_{\ell}\Jac(C)\otimes_{\mathbb{Z}_{\ell}}\overline{\mathbb{Q}}_{\ell} decomposes as a direct sum of irreducible G -representations. There is a corresponding factorisation— a priori for \Re(s)>3/2 —on the level of L -functions: (1.1) L(C,s)=\prod_{\tau\text{ irred.}}L(C^{\tau},s)^{\dim\tau}, where L(C^{\tau},s) is the L -function associated to a ‘motivic piece’, as described in [ 8 , §2] (cf. Definition 2.1 ). These will be the central objects of our study. The main achievement of the present work is Algorithm 3.1 , an algorithm for computing the Dirichlet series coefficients of L(C^{\tau},s) . The primary novelty of this is the reduction of the computation of Euler factors to point counts on curves—analogously to the standard algorithm for L(C,s) . We implement this algorithm in several cases using the Magma [ 1 ] adaptation of T. Dokchitser’s L -function calculator [ 6 ] . There were several motivations for the development of this algorithm, including: explicitly recovering cases where this decomposition matches with known factorisations of L -functions, improving L -function computations efficiency for curves with extra automorphisms (or endomorphisms), and numerical testing of many conjectures. On this latter point: formally, the motivic piece C^{\tau} is a motive over K with coefficients in \mathbb{Q}(\tau) ; it is the motive associated to the first cohomology of C and the \tau -idempotent of \mathbb{C}[G] . For the purposes of this work, we eschew any categorical discussion of motives—instead adopting the viewpoint of Deligne (cf. [ 4 , (0.12)] ), from which we see motives as a collection of cohomologies H_{B},H_{\dR},(H_{\ell})_{\ell} with additional data, including comparison isomorphisms between them (see § 5.1 ). The arithmetic of L -functions arising from motives is widely believed to be governed by a vast conjectural edifice which generalises BSD—for example, we are in the precise setting of the equivariant Tamagawa Number Conjecture of Burns and Flach [ 3 , §4] , after Bloch and Kato, for the pair (h^{1}(C),\mathbb{C}[G]) . We do not pursue verification of this conjecture herein, instead working on the lower levels of this edifice. At this level, several standard conjectures are interpreted explicitly for the L -functions L(C^{\tau},s) in [ 8 ] (see [ loc. cit. , Conjectures 2.20 & 2.23, Theorem 2.24]). See § 2 for explanations of the terms in ( 1.3 ). Conjecture 1.2 . The L -functions L(C^{\tau},s) satisfy the following properties: Hasse–Weil: L(C^{\tau},s) admits meromorphic continuation to the complex plane and the completed L -function \Lambda(C^{\tau},s) satisfies the functional equation (1.3) \Lambda(C^{\tau},s)=w(C^{\tau})\cdot\Lambda(C^{\tau^{*}},2-s). Deligne: For \sigma\in\Gal(\mathbb{Q}(\tau)/\mathbb{Q}) , we have \ord_{s=1}L(C^{\tau},s)=\ord_{s=1}L(C^{\sigma(\tau)},s) . Weak BSD: \ord_{s=1}L(C^{\tau},s)=\langle\tau,\Jac(C)(K)\otimes_{\mathbb{Z}}\mathbb{C}\rangle . Parity: For \tau self-dual, w(C^{\tau})=(-1)^{\langle\tau,\Jac(C)(K)\otimes_{\mathbb{Z}}\mathbb{C}\rangle} . Further to the conjecture of his appearing above, Deligne makes a conjecture ( [ 4 , Conjecture 2.8] , cf. Conjecture 5.9 ) linking the ‘irrational parts’ of motivic L -values to Deligne periods (Definition 5.6 ). We refer to this as Deligne’s Period Conjecture, numerical verification in new settings of which will be our primary focus. To do so, we work with superelliptic curves y^{n}=f(x) , a special case affording an explicit, testable formulation (Conjecture 5.25 ). In particular, in § 5.3 we discuss how the results of [ 25 ] and the associated Magma implementation [ 24 ] can be used to numerically compute the Deligne periods of motivic pieces of superelliptic curves. In § 6 , we combine this implementation with Algorithm 3.1 to numerically verify Deligne’s Period Conjecture for the motivic pieces of several superelliptic curves. Theorem 1.4 . For the 14 genus 3 superelliptic curves with L(C,1)\neq 0 considered in § 6 , there exist a,b,c\in\mathbb{Z} with |a|,|b|\leq 16 and |c|\leq 160 such that (1.5) \left|\frac{L^{\text{approx}}(C^{\tau},1)}{\Omega^{\text{approx}}_{C^{\tau}}}-\frac{a+b\cdot\omega}{c}\right|<\frac{1}{c^{4}}, where \tau is a non- \mathbb{Q} -rational character of C_{3} or C_{4} , L^{\text{approx}}(C^{\tau},1) is our numerical value for L(C^{\tau},1) , \Omega^{\text{approx}}_{C^{\tau}} is that for the Deligne period as in § 5.3 —both computed to precision at least 10 —and \omega=i or \omega=\zeta_{3} depending on the exponent of C . Remark 1.6 . The statement of Theorem 1.4 references approximate values because, although our computations of Dirichlet series coefficients are provably correct, the algorithm of [ 6 ] is not provably accurate without knowing the Hasse–Weil Conjecture for L(C^{\tau},s) . In all of the examples herein, the functional equation is at least numerically verified to the precision of our calculations. Remark 1.7 . Given \varepsilon>0 , for almost all \alpha\in\mathbb{R} , including all algebraic \alpha , there are only finitely many coprime pairs (p,q) for which |\alpha-p/q|<1/q^{2+\varepsilon} (cf. [ 19 , Theorem 32] , [ 27 ] ). With this in mind, looking at the real part and the imaginary part (divided by \sqrt{3} if \omega=\zeta_{3} ) of ( 1.5 ) separately, we find vastly better approximations than one would expect for a non- \mathbb{Q}(\omega) -rational. Theorem 1.4 therefore provides compelling evidence for the rationality of L(C^{\tau},1)/\Omega_{C^{\tau}} and for the validity of Deligne’s Period Conjecture. Example 1.8 (= Example 6.1 ) . Consider the curve (1.9) C/\mathbb{Q}(\zeta_{3}):\hskip 5.69046pty^{3}=x^{4}+x^{3}+\zeta_{3}x^{2}+\zeta_{3}-2 and write \tau for a non-trivial character of the automorphism group C_{3}\cong\langle y\mapsto\zeta_{3}\cdot y\rangle . With 100,000 Dirichlet series coefficients, we compute L(C^{\tau},1)\approx 1.8460900297\ldots-i\cdot 1.5447030118\ldots and \Omega_{C^{\tau}}=-38.6473408677\ldots-i\cdot 105.727015294\ldots. Together, these computations yield (1.10) \frac{L(C^{\tau},1)}{\Omega_{C^{\tau}}}\approx-0.0185185185185\ldots+i\cdot 0.0106916716517\ldots\approx\frac{\zeta_{3}-1}{81}. Remark 1.11 . Naturally, it would be of interest in future work to go beyond the rationality predicted by Deligne’s Period Conjecture to say something about the rational L -values it predicts to exist, à la Birch and Swinnerton-Dyer. Between [ 7 ] and [ 2 ] , it is explained that L -values of Artin-twists of elliptic curves ought to depend on the Galois-module structure of the Tate–Shafarevich group. Utilising this philosophy, the recent work [ 22 ] in the same setting seeks to

Read the complete text