🔢 mathematics

The Bogomolov Property through Galois Representations

This expository paper surveys techniques and results regarding the Bogomolov property for algebraic extensions associated with Galois representations, reinterpreting classical findings and establishing new criteria—particularly for modular representations with large local images—by leveraging Sen's theorem on totally ramified $p$ -adic Lie extensions.

Original authors: Lea Terracini

Published 2026-06-26

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: Lea Terracini

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

The Big Picture: Finding a "Safety Net" for Numbers

Imagine the world of numbers as a vast, infinite ocean. In this ocean, there are special islands called roots of unity (like the number 1, -1, or complex numbers that circle back to 1). These islands are very special because they have a "height" of zero. In mathematics, "height" is a way of measuring how complicated a number is.

The Problem:
Mathematicians have long wondered: If you take any number in this ocean that isn't one of those special islands, is there a "safety net" underneath it? In other words, is there a minimum level of complexity that every non-special number must have? You can't get infinitely close to zero complexity without actually being one of the special islands.

This idea is called the Bogomolov Property. If a collection of numbers (a "field") has this property, it means there is a universal floor for how simple a non-special number can be.

The Old Way vs. The New Way

The Old Way (Habegger's Discovery):
A mathematician named Habegger proved that if you take an elliptic curve (a specific type of geometric shape defined by an equation) and collect all the "torsion points" (points that, if you add them to themselves enough times, land back at the starting point), the field created by these points has this safety net.

The New Approach (Terracini's Paper):
Terracini and her colleagues say: "Let's look at this through a different lens." Instead of just looking at the shapes (elliptic curves), let's look at the Galois Representations.

The Analogy: Think of a Galois representation as a security camera system watching the number ocean.
The camera records how numbers move and interact.
The "field" is the specific area of the ocean that the camera is watching.
The paper asks: "If we set up a specific type of camera (a specific representation), does the area it watches have a safety net (the Bogomolov property)?"

This allows them to check many different types of number fields, not just those coming from elliptic curves.

The Two Main Tools Used in the Paper

To prove that a field has this safety net, the authors use two main strategies, which they apply to different "parts" of the number system.

1. The "Local" Strategy (The $p$ -adic Part)

Imagine you are looking at a specific neighborhood in the number ocean (a prime number $p$ ).

The Old Trick: Previously, they needed a very rare condition (like a number being "strongly supersingular," which is like finding a needle in a haystack) to prove the safety net existed.
The New Trick (Sen's Theorem): The authors use a powerful tool called Sen's Theorem.
- Analogy: Imagine the neighborhood has a strict set of rules about how people can move (a "Lie group" structure). Sen's Theorem says that if the neighborhood follows these strict rules, the "traffic" (ramification) is so organized that you can't sneak in a number that is too simple.
- This allows them to prove the safety net exists for many more fields, even if they don't have that rare "needle in a haystack" condition.

2. The "Global" Strategy (The Prime-to- $p$ Part)

This looks at the rest of the ocean, away from that specific neighborhood.

They use a tool called the ADZ Lemma.
Analogy: This is like checking if the local traffic rules in one neighborhood are so restrictive that they force the entire city to have a safety net. If the "local" camera sees a specific pattern (like a central symmetry), the whole field is safe.
They apply this to Modular Forms (complex mathematical functions that act like blueprints for number systems). They show that if a blueprint has certain "supercuspidal" features (a fancy way of saying it's very "irreducible" or cannot be broken down), the resulting field has a safety net.

What Did They Actually Prove?

The paper doesn't just talk about theory; it provides concrete results:

For Modular Forms: They proved that for certain types of modular forms (specifically those that are "supercuspidal" at a prime $p$ ), the fields generated by them have the Bogomolov property. This extends Habegger's result from elliptic curves to a much wider class of mathematical objects.
For "Big" Images: They showed that if a Galois representation (the security camera) has a "large image" (meaning it sees a lot of movement and complexity in the local neighborhood), then the field it defines has the safety net.
For Families of Curves: They showed that if you have a whole family of elliptic curves (or similar objects) that vary in a smooth, predictable way, then most of the fields in that family have the safety net.

The "Missing Piece" (The Central Element)

There is one specific condition in their proof called the Central Element Condition.

Analogy: Imagine the security camera system needs a "master key" (a central element) to lock the door and keep the safety net in place.
The authors found that if this master key exists, the proof works perfectly.
They acknowledge that there are cases where this master key doesn't exist (like a specific field involving roots of 2). In those cases, the safety net fails (heights can get arbitrarily close to zero).
They mention a conjecture by Rémond which suggests that even without the master key, the safety net might still exist if we look at the problem slightly differently (focusing on " $p$ -divisible" numbers). They are working on a new version of their proof to tackle this, but it's not fully finished yet.

Summary

In simple terms, this paper is a guidebook for finding safety nets in the ocean of numbers.

It takes a known result about elliptic curves and generalizes it to a much wider world of Modular Forms and Galois Representations.
It uses a clever new method (Sen's Theorem) to bypass the need for rare, hard-to-find conditions.
It proves that as long as the "local traffic" in the number system is complex and well-organized, there is a guaranteed minimum level of complexity for any number in that system.

The paper is a "survey," meaning it collects existing tools and results, explains how they fit together, and points the way toward solving the remaining mysteries (like the case without the "master key").

Technical Summary: The Bogomolov Property through Galois Representations

Problem Statement
The paper addresses the Bogomolov property (B) for algebraic extensions of $\mathbb{Q}$ . An extension $\mathbb{L}$ satisfies property (B) if there exists a uniform positive lower bound for the absolute logarithmic Weil height $h(\alpha)$ for all $\alpha \in \mathbb{L}^\times$ excluding roots of unity. While Northcott's theorem guarantees this for number fields, the property is non-trivial for infinite extensions. The central problem is to determine which infinite Galois extensions possess property (B), particularly those arising from arithmetic geometry (e.g., torsion points of elliptic curves) and modular forms.

The paper focuses on a specific perspective: associating a Galois representation $\rho: G_{\mathbb{K}} \to \Omega$ with the algebraic extension $\mathbb{K}(\rho)$ fixed by its kernel, and investigating whether $\rho$ "has property (B)" (i.e., if $\mathbb{K}(\rho)$ has property (B)). This framework aims to unify known results (such as Habegger's theorem on elliptic curves) and generalize them to broader classes of representations, including those of non-geometric origin.

Methodology
The paper employs a strategy that decomposes the analysis of Galois representations into local and global components, specifically separating the $p$ -part and the prime-to- $p$ part of adelic representations.

The ADZ Criterion: The authors utilize the Amoroso-David-Zannier (ADZ) lemma, which establishes property (B) for extensions where the local degrees of the fixed field of the center of the Galois group are uniformly bounded. In the language of representations, if the image of a decomposition group in the projective linear group is finite, property (B) holds.
Local $p$ -adic Analysis (Sen's Theorem): A core methodological innovation is the application of Sen's theorem on totally ramified $p$ -adic Lie extensions. This theorem relates the Lie filtration of a $p$ -adic Lie group to its higher ramification filtration. By establishing that the Galois group of the extension is a $p$ -adic Lie group with specific structural properties, the authors derive local metric inequalities ( $|\sigma(\alpha) - \alpha|_p < C$ ) that propagate to global height bounds via the product formula.
Separation of Components:
- Prime-to- $p$ part: Handled via the ADZ lemma, often requiring conditions on the image of Frobenius (e.g., $a_p = 0$ for modular forms) to ensure centrality.
- $p$ -part: Handled via the study of $p$ -adic Lie groups. The authors replace the strong supersingularity condition ( $a_p=0$ ) with the assumption that the local Galois image is a $p$ -adic Lie group of positive dimension.

Key Contributions and Results

Generalization of Habegger's Theorem: The paper re-interprets Habegger's result (that $\mathbb{Q}(E_{tors})$ has property (B) for non-CM elliptic curves) as a consequence of the Galois representation $\hat{\rho}_E$ having a large image.
Modular Forms and Conjectures:
- The paper discusses Conjecture 3.1, which posits that Galois representations attached to modular forms have property (B).
- It reviews Theorem 3.3 (Amoroso-Terracini), which proves property (B) for modular forms under strong conditions: $p \nmid N$ , $p \geq 2k-1$ , $a_p=0$ (strong supersingularity), and a specific large image condition for the mod- $p$ representation.
- The paper highlights the difficulty of extending this to general modular forms, noting that the condition $a_p=0$ is exceptional for weights $k > 2$ .
New Criteria via $p$ -adic Lie Groups:
- Theorem 4.3: Provides a sufficient condition for a Galois extension $\mathbb{L}/\mathbb{K}$ $L / K$ to have property (B) if it is a $p$ $p$ -adic Lie extension satisfying three local conditions at a place $v|p$ $v ∣ p$ :
  1. (LPTR) Locally Potentially Totally Ramified.
  2. (NC) Normal Closure condition on the Lie filtration.
  3. (CE) Existence of a central element acting non-trivially on roots of unity.
- This theorem allows the proof of property (B) for representations where $a_p \neq 0$ , provided the local image is a $p$ -adic Lie group.
Applications to Non-Geometric Representations:
- Theorem 5.1: Establishes property (B) for any continuous $p$ -adic representation $\rho: G_{\mathbb{K}} \to GL_d(\mathbb{Z}_p)$ where the image of the inertia subgroup at a place over $p$ is open in $GL_d(\mathbb{Z}_p)$ .
- Theorem 5.4: Applies these criteria to supersingular modular forms with $a_p \neq 0$ , showing property (B) holds if the local representation is strongly absolutely irreducible.
- Theorem 5.5: Extends property (B) results to elliptic curves over arbitrary number fields (not just $\mathbb{Q}$ ) under conditions of absolute irreducibility and surjectivity of the residual representation, bypassing the need for Elkies' theorem on supersingular primes.
Families of Representations: The paper demonstrates that these criteria behave well in $p$ -adic analytic families (e.g., Hida families, eigencurves), deducing a generic Bogomolov property for representations in an open subset of the parameter space.

Significance and Scope
The paper serves as an expository survey of techniques developed in joint work with Amoroso, Conti, and Piras. Its primary significance lies in:

Unification: It provides a unified framework interpreting classical results (like those for abelian extensions and elliptic curves) through the lens of Galois representations.
Relaxation of Conditions: It moves beyond the restrictive "strong supersingularity" ( $a_p=0$ ) condition required in earlier proofs, replacing it with structural conditions on the $p$ -adic Lie group image (Theorem 4.3).
Expansion of Scope: It extends the validity of property (B) to representations not necessarily of geometric origin, provided they exhibit "large local image" (open inertia).

The authors modestly note that while they have generalized the criteria for the $p$ -part using Sen's theorem, the prime-to- $p$ part for general modular forms (without $a_p=0$ ) remains an open challenge, and Conjecture 3.1 is not fully resolved. The paper also identifies the Rémond $p$ -adic conjecture (regarding $p$ -divisible subgroups) as a natural direction for future application of these techniques, though it acknowledges that the general case of Rémond's original conjecture lies beyond current methods.