🔢 mathematics

A Hilbert 90 Property for S-Class Groups and Applications to the Gross--Kuz'min Conjecture

This paper establishes a computable, finite-level criterion for the $\mathbf{cl}^S$ -Hilbert 90 property in cyclic extensions of number fields and demonstrates that its satisfaction in $\mathbb{Z}_p$ -extensions implies the finiteness of Kuz'min-Tate module coinvariants, thereby offering a new approach to the Gross--Kuz'min conjecture supported by numerical evidence and a random matrix heuristic.

Original authors: Julian Feuerpfeil

Published 2026-08-25

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Julian Feuerpfeil

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

In the vast landscape of number theory, mathematicians study whole numbers not just as isolated digits, but as members of complex families that interact through rules of division and multiplication. One of the most enduring puzzles in this field involves understanding how these numbers behave when we expand our view from a single set of integers to a larger, more intricate system known as a number field. Imagine taking a familiar set of numbers and stretching it into a new dimension where familiar rules of arithmetic take on new, sometimes surprising shapes. Within these expanded systems, there are hidden structures called class groups that measure how far the system is from having a perfect, orderly structure. When mathematicians look at how these structures change as they move from a smaller system to a larger one, they often encounter a phenomenon where information seems to vanish or transform in ways that are difficult to predict. For over a century, a famous result known as Hilbert's Theorem 90 has served as a reliable guide, acting like a compass that tells researchers exactly when certain patterns must hold true in these expanded systems. However, this compass does not always work when applied to the more complex class groups that arise in modern number theory, leaving a gap in our understanding of how these systems evolve.

A recent paper by Julian Feuerpfeil addresses this gap by investigating a specific question: under what conditions does a similar rule of predictability hold true for these more complex class groups? The author focuses on a scenario where one number field is built upon another in a cyclic, repeating pattern, much like a set of nested rings. The study asks whether the "kernel" of a specific arithmetic operation—the collection of elements that disappear when mapped from the larger system back to the smaller one—can always be explained by a simple, repeating transformation. If this property holds, it means the system behaves with a high degree of regularity and order. Feuerpfeil develops a practical test to determine if this regularity exists without needing to know the detailed inner workings of the larger system, which is often impossible to calculate directly. Instead, the test relies only on the known properties of the smaller, base system and how the two systems connect.

The paper establishes that if this regularity holds for the very first step of an infinite chain of number fields, it automatically holds for every subsequent step in that chain. This is a powerful finding because it allows mathematicians to make definitive statements about infinite towers of number fields by checking only a single, finite layer. The author connects this discovery to a major unsolved problem in the field known as the Gross–Kuz'min conjecture, which predicts that certain complex structures in these infinite towers remain finite and manageable. By proving that the regularity condition implies this finiteness, the paper provides a new, concrete way to verify the conjecture in many cases. The research shows that for a wide variety of number fields, this regularity condition is satisfied, suggesting that the conjecture is likely true for almost all such systems.

To support this theoretical framework, the author introduces a specific mathematical map that acts like a filter, checking whether the connections between different parts of the number system are strong enough to maintain order. The paper demonstrates that if this map works perfectly for a specific type of prime number, the entire system behaves predictably. Through extensive computer simulations, the author tested this map across thousands of different number fields and prime numbers. The results showed that the map worked correctly in the vast majority of cases, failing only in a tiny fraction of instances. This statistical evidence leads to a strong prediction: for any given totally real number field, there are likely only a finite number of prime numbers for which this regularity breaks down. In other words, the orderly behavior described by the theorem is the rule, not the exception.

The study also identifies specific situations where this regularity might fail, particularly in systems that possess a certain type of symmetry related to complex numbers. By analyzing these exceptions, the author refines the prediction, showing that while the rule holds for most systems, the structure of the number field itself can sometimes create obstacles. The paper concludes by offering a heuristic, or an educated guess based on probability, that explains why these failures are so rare. This approach treats the behavior of the number systems as if they were random, yet the simulations confirm that the underlying mathematical structure forces them to behave in a highly ordered way. The work does not solve the Gross–Kuz'min conjecture in its entirety, but it provides a robust, computable criterion that confirms the conjecture for a vast array of cases and offers a clear path for future verification.

Ultimately, this research transforms a deep, abstract question about the nature of infinite number systems into a concrete, testable condition. It moves the field from a state of uncertainty, where the behavior of these systems was largely unknown, to a state of high confidence, where the regularity of these systems can be predicted with precision. By linking the behavior of finite steps to infinite towers, the paper offers a new lens through which to view the fundamental architecture of number theory, suggesting that even in the most complex expansions of our number systems, a deep and persistent order prevails.

Technical Summary: A Hilbert 90 Property for S-Class Groups and Connections with the Gross–Kuz'min Conjecture

Problem Statement
The paper investigates a variant of Hilbert's Theorem 90 adapted to the context of $S$ -class groups of number fields. While Hilbert's original theorem characterizes elements of norm 1 in a cyclic extension $L/K$ as those of the form $\sigma(y)y^{-1}$ , this property does not generally hold for $S$ -class groups ( $cl_S$ ). The author defines the $cl_S$ -Hilbert 90 property: a cyclic extension $L/K$ possesses this property if the kernel of the arithmetic norm map $N: cl_S(L) \to cl_S(K)$ coincides with the image of the map $1-\sigma$ acting on $cl_S(L)$ .

The primary motivation is to establish criteria for this property that rely solely on the arithmetic of the base field $K$ and the ramification behavior of the extension, without requiring knowledge of the class group of the extension field $L$ . A secondary, significant goal is to apply these findings to Iwasawa theory, specifically to the Gross–Kuz'min Conjecture, which asserts the finiteness of the coinvariants of the Kuz'min–Tate module associated with the cyclotomic $\mathbb{Z}_p$ -extension.

Methodology
The paper employs a combination of cohomological algebra, class field theory, and probabilistic heuristics:

Cohomological Mackey Functors: The author formalizes the problem using the framework of cohomological Mackey functors (or $G$ -modulations). This categorical framework allows for the definition of section cohomology groups $c_1(G, X)$ and $k_0(G, X)$ . The $cl_S$ -Hilbert 90 property is equivalent to the vanishing of $c_1(G, cl_S)$ .
Local-Global Criteria: By utilizing the ambiguous class number formula and Chevalley–Herbrand theorem, the author derives a criterion (Theorem 3.3) for the $cl_S$ -Hilbert 90 property. This criterion depends on the surjectivity of a map from the $S$ -units of $K$ to a direct sum of local unit quotients. Crucially, this criterion is expressed entirely in terms of the base field $K$ and the ramification data of the extension.
Governing Fields: The construction of specific cyclic extensions satisfying the property relies on theorems by Gras (and Gras–Munnier) regarding the existence of extensions with prescribed ramification over governing fields (Kummer extensions involving $S$ -units).
Iwasawa Theory: The paper connects the finite-level $cl_{S_p}$ -Hilbert 90 property to the finiteness of coinvariants in the Iwasawa module $T_p(K_\infty/K)$ . Using the topological Nakayama Lemma, it is shown that if the property holds for the first layer of a $\mathbb{Z}_p$ -extension (under mild assumptions), it holds for all layers, implying the finiteness of the coinvariants (the Finite Coinvariants Property, FCP).
Random Matrix Heuristics: To understand the frequency with which the derived criteria are satisfied, the author models the relevant local map $\tau_p$ (related to Fermat quotients) as a random matrix over $\mathbb{F}_p$ . This leads to a heuristic prediction regarding the density of primes for which the property holds.

Key Contributions and Results

Criterion for $cl_S$ -Hilbert 90 (Theorem 3.3): The paper establishes a practical, computable criterion for the $cl_S$ -Hilbert 90 property. It states that for a cyclic extension $L/K$ , the property holds if certain natural maps between $S$ -class groups of the base field (involving conductors and ramification indices) are isomorphisms. This allows for the verification of the property without computing $cl_S(L)$ .
Construction of Examples: Using the criterion and Gras's theorems, the author constructs large families of extensions (including $\mathbb{Q}(\zeta_{p^n})/\mathbb{Q}$ ) satisfying the property.
Connection to Gross–Kuz'min (Theorem A & B):
- Theorem A: For a $\mathbb{Z}_p$ -extension $K_\infty/K$ where $m_0(K_\infty/K)=0$ (places above $p$ are either totally ramified or split), if the first layer $K_1/K$ has the $cl_{S_p}$ -Hilbert 90 property, then the extension has the Finite Coinvariants Property (FCP).
- Theorem B: A sufficient condition for the FCP is the surjectivity of a specific local map $\tau_p: E_p \to \bigoplus_{q \neq p} U_q/U_q^p$ . If this map is surjective for some place $p$ above $p$ , then every $\mathbb{Z}_p$ -extension in the family $\mathcal{F}(K, p)$ has the FCP. If additionally the $p$ -part of the class number $h_p(K)=1$ , the FCP holds for all $\mathbb{Z}_p$ -extensions with $m_0=0$ .
Explicit Interpretations: For quadratic fields, the surjectivity of $\tau_p$ is linked to the valuation of Fermat quotients, connecting the problem to the Wieferich prime problem. For real quadratic fields, the condition is related to $p$ -rationality and $p$ -adic regulators.
Heuristic and Conjecture (Conjecture 5.4): Based on random matrix theory and numerical experiments (performed using the OSCAR system), the author proposes that for any totally real number field $K$ , there are at most finitely many primes $p$ for which the map $\tau_p$ fails to be surjective. This implies that the Gross–Kuz'min Conjecture (and the FCP) holds for almost all primes in the totally real case.
Exceptional Cases: The paper identifies structural obstructions (Proposition 5.5) for CM fields where the map $\tau_p$ cannot be surjective due to the action of complex conjugation, refining the heuristic for these cases.

Significance
The paper provides a new, finite-level criterion for the Gross–Kuz'min Conjecture that is independent of the class group of the infinite extension. By reducing the problem to the surjectivity of a map defined on the base field (Theorem B), it offers a concrete method to verify the conjecture for specific fields and primes. The formulation of Conjecture 5.4 extends previous work by Jaulent, suggesting a strong uniformity in the behavior of $S$ -class groups across $\mathbb{Z}_p$ -towers for totally real fields. The work bridges the gap between classical class field theory (Hilbert 90, capitulation) and modern Iwasawa theory, providing a unified framework for understanding the growth of class groups in towers. The numerical evidence and heuristic arguments support the view that the Gross–Kuz'min Conjecture is likely true for all totally real fields, with only finitely many exceptional primes.

More like this