🔢 mathematics

Geometric families of multiple elliptic Gamma functions and arithmetic applications, I

This paper introduces geometric families of multiple elliptic Gamma functions for higher-rank lattices to establish their transformation properties under $\mathrm{SL}_n(\mathbb{Z})$ and utilizes the resulting Bernoulli rational functions to construct cocycles that compute partial zeta values at $s=0$ , thereby advancing the solution to Hilbert's 12th problem for number fields with exactly one complex place.

Original authors: Pierre L. L. Morain

Published 2026-01-27

📖 4 min read🧠 Deep dive

CC BY 4.0

Original authors: Pierre L. L. Morain

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine you are trying to solve a massive, ancient puzzle about the hidden structure of numbers. Mathematicians have long been fascinated by a specific challenge called Hilbert's 12th Problem. In simple terms, this problem asks: "Can we build a universal 'key' that unlocks the secrets of how numbers relate to each other in complex systems?"

This paper, written by Pierre L. L. Morain, is the first step in a new series of research aimed at creating a better set of tools to solve this puzzle. The author introduces a new framework using a special family of mathematical objects called Multiple Elliptic Gamma Functions.

Here is a breakdown of the paper's main ideas using everyday analogies:

1. The Tools: "Mathematical Transformers"

Think of the Multiple Elliptic Gamma Functions as highly sophisticated, shape-shifting tools.

The Old Way: Previously, mathematicians had a tool called the "Theta function" (like a basic wrench) that worked well for simple 2-dimensional number problems. They also had a slightly more complex tool called the "Elliptic Gamma function" (a power drill) for 3-dimensional problems.
The New Upgrade: Morain has taken these tools and upgraded them into a whole family of tools that can handle problems in any number of dimensions (not just 2 or 3). He calls these "Geometric Families."
The Analogy: Imagine you have a set of Lego bricks. The old tools were just a few specific bricks. Morain has invented a new rule for how to snap these bricks together so they can build towers of any height, not just short ones.

2. The "Glue": Bernoulli Rational Functions

When you use these new tools, they don't just fit perfectly; they leave behind a specific "residue" or "glue" when they change shape.

In the paper, this glue is called Bernoulli Rational Functions.
The Analogy: Imagine you are folding a piece of paper (the mathematical function). When you fold it, the crease leaves a specific pattern. Morain discovered that these creases (the Bernoulli functions) follow a very strict, predictable rule.
The Discovery: He proved that if you arrange these tools in a specific way, the "glue" they leave behind cancels out perfectly, leaving a clean slate (mathematically, the sum equals zero). This is a crucial property because it means the tools are consistent and reliable.

3. The Application: Counting "Number Islands"

The ultimate goal of this research is to calculate something called Partial Zeta Values at a specific point ( $s=0$ ).

The Analogy: Imagine a vast ocean of numbers. Some areas are calm, and some are stormy. The "Zeta function" is a map that tells you the "depth" or "weight" of these areas.
The Problem: For a long time, calculating the depth of these areas for complex, multi-dimensional number systems was incredibly difficult.
The Solution: Morain shows that you can use his new "Bernoulli glue" to calculate these depths. He proves that these complex values are actually just simple combinations of the "glue" patterns he discovered.
The Result: He demonstrates this with specific examples (like cubic fields, which are 3-dimensional number systems). He shows that by using his new geometric tools, you can compute these values with high precision, confirming that they are indeed rational numbers (simple fractions).

4. The "Cocycle" Connection

The paper uses a fancy word, Cocycle, to describe how these tools interact.

The Analogy: Think of a "cocycle" as a set of instructions for a dance. If you have three dancers (mathematical objects) and they move in a circle, the "cocycle" ensures that if they all follow the rules, they end up back where they started without getting lost.
Morain shows that his new Bernoulli functions form a perfect "dance routine" for specific groups of numbers (called unit groups in totally real number fields). Because they dance perfectly, we can use their movements to measure the "size" of the number systems they inhabit.

Summary

In short, this paper does not solve Hilbert's 12th Problem entirely yet. Instead, it builds the scaffolding needed to do so.

It creates a new, scalable family of mathematical functions (the geometric families).
It proves these functions have a hidden, perfect symmetry (the cocycle property).
It demonstrates that this symmetry can be used as a calculator to find specific, hard-to-reach values in number theory (the partial zeta values).

The author is essentially saying: "I have built a new, more powerful engine. In this first paper, I show you how the engine is constructed and prove that it runs smoothly. In future papers, I will use this engine to drive us to the final destination of solving the number theory puzzle."

Technical Summary: Geometric Families of Multiple Elliptic Gamma Functions and Arithmetic Applications, I

Problem Statement
This paper addresses the arithmetic applications of multiple elliptic Gamma functions, specifically aiming to construct a framework for solving Hilbert's 12th problem for general number fields with exactly one complex place. While the case for rank 3 lattices ( $n=3$ ) has been explored by Bergeron, Charollois, and Garc´ıa [BCG23] using elliptic Gamma functions, this work seeks to generalize these constructions to higher ranks ( $n \ge 3$ ). The core challenge involves defining geometric families of these functions on lattices of arbitrary rank and establishing their transformation properties under the action of the special linear group $SL_n(\mathbb{Z})$ . A secondary objective is to utilize the associated "modularity defect"—a collection of Bernoulli rational functions—to compute partial zeta values at $s=0$ for totally real number fields.

Methodology
The paper employs a blend of analytic number theory, algebraic geometry, and the theory of polyhedral cones.

Geometric Construction of $G_{r}$ Functions:
The authors upgrade the construction of geometric families of elliptic Gamma functions (originally defined for rank 3 lattices by Felder, Henriques, Rossi, and Zhu [FHRZ08]) to higher degrees. They define geometric families $G_{n-2, a_1, \dots, a_{n-1}}$ attached to families of $n-1$ linearly independent primitive linear forms on a rank $n$ lattice $L$ . These functions are built using Nishizawa's multiple elliptic Gamma functions $G_r$ as building blocks, adapted to the geometric setting of $SL_n(\mathbb{Z})$ acting on the lattice.
Cocycle Relations and Cone Theory:
The transformation properties of these geometric functions are linked to a collection of Bernoulli rational functions, denoted $B_{n, a_1, \dots, a_n}$ . The authors prove that these functions satisfy specific modular and equivariance properties. To establish the additive cocycle relations for $B_{n, \dots}$ , the paper utilizes the theory of indicator functions of closed polyhedral cones. By defining a Solomon-Hu pairing between the space of cone indicator functions and rational functions, the authors relate the geometric Bernoulli functions to coefficient extractions of generating series associated with cones.
Inductive Proofs on Cone Configurations:
A significant portion of the technical work involves proving a cocycle relation for indicator functions of closed cones (Theorem 2). This is achieved through a double induction on the dimension of the vector space and the number of negative coefficients in linear relations among the defining linear forms. The authors carefully distinguish between "good" configurations (where the cocycle relation vanishes) and "bad position" configurations (BP), where the relation may not hold or requires specific treatment.
Application to Zeta Functions:
The paper applies these results to Shintani's method for computing partial zeta functions. By decomposing fundamental domains for the action of totally positive units into signed unions of rational polyhedral cones, the authors express the values of partial zeta functions at $s=0$ as linear combinations of the geometric Bernoulli functions.

Key Contributions and Results

Theorem 1 (Modular and Equivariance Properties): The paper establishes that the geometric families $G_{n-2, a_1, \dots, a_{n-1}}$ satisfy a modular property involving a product of functions equal to the exponential of a Bernoulli rational function $B_{n, a_1, \dots, a_n}$ . Furthermore, both the $G$ functions and the $B$ functions are equivariant under the action of $SL_n(\mathbb{Z})$ . This generalizes the known relations for the $\theta$ function ( $n=2$ ) and the elliptic Gamma function ( $n=3$ ).
Theorem 2 (Cocycle Relation for Cones): The authors prove a technical result regarding the indicator functions of closed polyhedral cones. They show that for linear forms in general position (excluding "bad position"), a specific alternating sum of cone indicator functions is congruent to zero modulo the subspace generated by cones containing lines. This result is the technical heart of the paper and is of independent interest in the theory of polyhedral cones.
Corollary 1 (Cocycle Relation for Bernoulli Functions): Derived from Theorem 2, this corollary states that the alternating sum of the Bernoulli rational functions $B_{n, a_0, \dots, \hat{a}_j, \dots, a_n}$ vanishes for almost all configurations of the linear forms. This confirms that the collection of functions $\phi_{n,a}$ forms a collection of $(n-1)$ -cocycles for specific subgroups of $SL_n(\mathbb{Z})$ associated with unit groups.
Theorem 3 (Partial Zeta Values): The paper provides an explicit formula expressing the partial zeta values $\zeta_{\mathfrak{f}}(\mathfrak{b}, 0)$ for a totally real number field of degree $n$ as a trace of algebraic numbers. Specifically, these values are computed as a sum of evaluations of the geometric Bernoulli functions $B_{n, a_1, \dots, a_n}$ at $s=0$ , weighted by signs derived from a signed fundamental domain decomposition.
Explicit Computations: The authors demonstrate the efficacy of their method with explicit computations for real cubic fields ( $n=3$ ), recovering known results from the literature (e.g., [CGS00]) and verifying them using Pari/GP.

Significance and Claims
The paper claims to provide a comprehensive framework for the arithmetic applications of multiple elliptic Gamma functions in higher dimensions. Its primary significance lies in:

Generalization: It successfully extends the geometric construction of elliptic Gamma functions and their associated modular properties from rank 3 to arbitrary rank $n$ , offering a unified perspective on the $\theta$ function, the elliptic Gamma function, and their higher-rank analogues.
Cohomological Insight: It clarifies the cohomological nature of these functions, identifying them as cocycles (or related to cocycles) for $SL_n(\mathbb{Z})$ and linking them to the computation of arithmetic invariants.
Computational Tool: It offers a concrete method for computing partial zeta values at $s=0$ in totally real fields using the geometric Bernoulli functions, bridging the gap between the analytic theory of multiple Gamma functions and the arithmetic of number fields.

The authors note that this is the first in a series of papers. Future work in the series will focus on smoothing operations to simplify the modular relations, the evaluation of these cocycles to construct conjectural elliptic units for number fields with exactly one complex place (addressing Hilbert's 12th problem), and further examples for $n=4$ . The current paper modestly frames its results as a foundational step, providing the necessary geometric and cohomological machinery for these subsequent arithmetic applications.