🔢 mathematics

A cohomological translation of the Kaplansky radical for profinite groups

This paper introduces a cohomological analogue of the Kaplansky radical for arbitrary profinite groups, formulates and proves a group-theoretic version of the H-conjecture for broad classes of fields and pro- $p$ groups, and demonstrates that this property is stable under various natural constructions while providing new examples beyond those arising from arithmetic.

Original authors: Simone Blumer, Julian Feuerpfeil, Lucas Correa Lopes, Claudio Quadrelli

Published 2026-07-01

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: Simone Blumer, Julian Feuerpfeil, Lucas Correa Lopes, Claudio Quadrelli

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: A New Way to Look at Numbers and Shapes

Imagine you are trying to understand a complex machine (like a field of numbers) by looking at its blueprint. For a long time, mathematicians studied a specific part of this machine called the Kaplansky radical. Think of the radical as a "special filter" that catches only the most important, stubborn numbers in a field.

In the 1980s, two mathematicians, Kijima and Nishi, made a guess (a conjecture) about how this filter behaves when you stretch the machine into a larger version (a field extension). They thought the filter would behave in a very neat, predictable way, similar to a famous rule in math called "Hilbert's Theorem 90."

However, about 30 years ago, this guess was proven wrong for some weird, complicated fields. But, it turned out to be true for many "nice" fields, like the rational numbers or local fields.

The Problem: The old definition of this "filter" only worked for numbers (fields). It couldn't be used on abstract shapes or groups.

The Solution: This paper introduces a new, universal version of the filter. Instead of looking at numbers, the authors look at groups (mathematical structures that describe symmetry) and use a tool called cohomology (which is like a way of measuring the "holes" or "twists" in a shape). They call this new filter the $\mathbb{F}_p$ -cup radical.

The Core Concept: The "Orthogonal" Filter

To understand the new filter, imagine a dance floor where every dancer represents a piece of information (a cohomology class).

The Cup Product: This is a rule that says, "If two dancers pair up, do they create a spark?" If they pair up and create a spark (a non-zero result), they are "connected."
The Radical (The Filter): The authors define the radical as the group of dancers who never create a spark with anyone else on the floor. They are the "invisible" dancers. If a dancer is in the radical, they are orthogonal (at a right angle) to everyone else.

The paper asks: If we shrink the dance floor (by looking at a smaller subgroup), do the "invisible" dancers of the small floor map perfectly onto the "invisible" dancers of the big floor?

If the answer is "yes," the group has the p-Kijima–Nishi property. This is the group-theoretic version of the old "H-conjecture."

What the Authors Discovered

The paper is divided into two main adventures: one with Fields (numbers) and one with Groups (shapes).

1. The Field Adventure (Numbers)

The authors checked if famous types of number fields have this "invisible dancer" property.

The Good News: They proved that for many important types of fields, the answer is YES. This includes:
- Local and Global Fields: Think of these as the "standard" number systems (like the rational numbers or p-adic numbers).
- Rational Function Fields: Fields made by adding variables (like $x$ ) to a base field.
- Elementary Type Fields: Fields whose symmetry groups are built from simple Lego blocks (Demushkin groups and free groups).
The Bad News: They also proved that if you build a field that is "too weird" or "too transcendental" (extremely complex), you can break the property. You can construct a field where the "invisible dancers" don't map correctly. This confirms that the property isn't universal, but it holds for the "well-behaved" worlds mathematicians usually care about.

2. The Group Adventure (Shapes)

Since the new definition works for any group, not just those coming from numbers, the authors explored what happens with pure mathematical shapes.

Building Blocks: They showed that if you build a complex group by gluing together simple groups (like free groups or Demushkin groups) using specific rules (free products, HNN extensions), the "invisible dancer" property is preserved.
New Examples: They found new types of groups that satisfy the property, even though these groups cannot exist as the symmetry groups of any number field.
- Analogy: Imagine finding a new type of crystal structure that is perfectly stable in a lab, but you know it could never form naturally in the earth's crust. The paper finds these "lab-only" groups and proves they still follow the same rules as the natural ones.
Graph Groups: They looked at groups defined by graphs (dots and lines). They found that for a wide variety of these "Right-Angled Artin" groups, the property holds.

The "Dictionary" Connection

One of the most important parts of the paper is the "dictionary" they built.

They proved that for fields containing specific roots of unity (like the square root of -1), their new group-theoretic radical is exactly the same as the old number-theoretic Kaplansky radical.
This means: If you want to know if a number field satisfies the old "H-conjecture," you don't need to do complex number crunching. You just need to look at the shape of its symmetry group. If the group has the "p-Kijima–Nishi property," the field satisfies the conjecture.

Summary of Results

Translation: They successfully translated a hard problem about numbers into a problem about shapes (groups).
Verification: They proved that for "nice" fields (local, global, rational, etc.), the property holds.
Counter-examples: They proved that for "wild" fields, the property fails.
Stability: They showed that the property is robust. If you glue "nice" groups together in specific ways, the result is still "nice."
New Worlds: They discovered many new groups that satisfy the property, expanding the universe of objects where this mathematical rule works, even if those objects aren't number fields.

The Takeaway

The paper is like a master key. It takes a specific lock (the Kaplansky radical for numbers) and creates a universal key (the cohomological radical for groups). This new key opens many more doors, allowing mathematicians to see that the rules governing "nice" numbers are actually part of a much larger, deeper pattern that governs the shape of symmetry itself.

Technical Summary: A Cohomological Translation of the Kaplansky Radical for Profinite Groups

Problem Statement
The Kaplansky radical of a field $\mathbb{K}$ (of characteristic $\neq 2$ ) is the set of nonzero elements represented by every binary quadratic form $X^2 - aY^2$ . In the 1980s, D. Kijima and M. Nishi conjectured that for quadratic extensions $\mathbb{K}(\sqrt{a})/\mathbb{K}$ , the Kaplansky radicals are related by the norm map in a manner analogous to Hilbert's Theorem 90. Specifically, they conjectured that $N_{\mathbb{K}(\sqrt{a})/\mathbb{K}}^{-1}(R_K(\mathbb{K})) = R_K(\mathbb{K}(\sqrt{a})) \cdot \mathbb{K}^{\times 2}$ . This "H-conjecture" was disproven by K.J. Becher and D.B. Leep, who constructed counterexamples. However, the conjecture holds for several important classes of fields.

R.P. Dario and A.J. Engler later generalized the Kaplansky radical to arbitrary primes $p$ . The central problem addressed in this paper is to translate the Kaplansky radical and the H-conjecture into a cohomological framework applicable to arbitrary profinite groups and primes $p$ . The authors aim to define a group-theoretic analogue of the radical, investigate the stability of the associated "Kijima–Nishi property" under various group constructions, and determine which fields satisfy this property based on the structure of their absolute Galois groups.

Methodology
The authors introduce the $\mathbb{F}_p$ -cup radical, denoted $R_p(G)$ , for a profinite group $G$ . It is defined as the orthogonal complement of the first cohomology group $H^1(G, \mathbb{F}_p)$ with respect to the cup product pairing:
$R_p(G) = \{ \alpha \in H^1(G, \mathbb{F}_p) \mid \alpha \smile \beta = 0 \text{ for all } \beta \in H^1(G, \mathbb{F}_p) \}.$
They define the $p$ -Kijima–Nishi property for a profinite group $G$ as the surjectivity of the corestriction map $c_{U,G}: R_p(U) \to R_p(G)$ for every normal open subgroup $U$ of index $p$ .

The methodology relies heavily on Galois cohomology and the theory of pro- $p$ groups. Key technical tools include:

Cohomological Dictionary: Establishing an isomorphism between the classical $p$ -radical of a field $\mathbb{K}$ (containing a primitive $p$ -th root of unity) and the group-theoretic radical of its absolute Galois group $G_\mathbb{K}$ , specifically $R_p(\mathbb{K})/\mathbb{K}^{\times p} \cong R_p(G_\mathbb{K})$ .
Minimal Presentations: Utilizing minimal presentations of pro- $p$ groups to compute cup products via the Zassenhaus filtration (Proposition 1.3).
Structural Decompositions: Analyzing the behavior of the radical under free pro- $p$ products, amalgamated products, HNN-extensions, and fundamental groups of graphs of pro- $p$ groups using Mayer–Vietoris sequences and Hochschild–Serre spectral sequences.
Arithmetic Reductions: Applying results from local and global class field theory, Kummer theory, and the theory of Severi–Brauer varieties to construct specific fields and analyze their Galois groups.

Key Contributions and Results

Cohomological Definition and Equivalence: The paper establishes that for fields containing a primitive $p$ -th root of unity, the field satisfies the H-conjecture if and only if its absolute Galois group possesses the $p$ -Kijima–Nishi property. This reduces the arithmetic conjecture to a group-theoretic property depending solely on the maximal pro- $p$ quotient of the Galois group.
Counterexamples (Theorem A): The authors prove that for any field $\mathbb{K}$ of characteristic $\neq p$ , there exists an extension $\mathbb{K}'/\mathbb{K}$ such that $G_{\mathbb{K}'}$ does not have the $p$ -Kijima–Nishi property. This construction relies on creating highly transcendental extensions to split central simple algebras, mirroring the counterexamples of Becher and Leep.
Positive Results for Arithmetic Fields (Theorem B): The $p$ $p$ -Kijima–Nishi property is proven to hold for the absolute Galois groups of:
- Local and global fields.
- Complete discretely valued fields with residue fields containing a primitive $p$ -th root of unity.
- Purely transcendental extensions of fields containing a primitive $p$ -th root of unity.
- Fields whose maximal pro- $p$ Galois group is of elementary type (constructed from $\mathbb{Z}_p$ and Demuškin groups via free products and specific semidirect products).
- Formally real Pythagorean fields of finite type (for $p=2$ ).
Stability under Group Constructions (Theorem C & D): The paper investigates the stability of the property under universal group constructions.
- It is shown that the property is preserved under free pro- $p$ products (Proposition 3.7).
- New classes of groups satisfying the property are identified, including generalized right-angled Artin pro- $p$ groups and $\Delta$ -right-angled Artin pro-$2$ groups (Theorem D).
- The authors provide conditions under which amalgamated products and HNN-extensions preserve the property (Propositions 4.2 and 4.5), noting that surjectivity of restriction maps is a necessary condition.
Generalization without Roots of Unity (Section 6): The authors propose a generalization of the radical for fields not containing a primitive $p$ -th root of unity by introducing $p$ -orientations and Tate twists. They show that for local and global fields in this broader setting, the generalized radical remains trivial.

Significance
The paper claims that its primary significance lies in providing a "group-cohomological alter ego" for the Kaplansky radical, thereby translating a deep arithmetic problem into the language of profinite group theory. This translation allows for the application of powerful group-theoretic tools (such as Bass–Serre theory and the theory of elementary type groups) to problems in field theory and quadratic forms.

Specifically, the work demonstrates that the validity of the H-conjecture for a field depends only on the structure of its maximal pro- $p$ Galois group, aligning with the spirit of Anabelian Geometry. By proving the property for broad classes of groups—including those that may not arise as Galois groups of fields (such as certain generalized right-angled Artin pro- $p$ groups)—the authors expand the scope of the conjecture beyond arithmetic contexts, offering new examples and structural insights into the behavior of cohomological radicals. The results also clarify the limitations of the conjecture, showing that it fails for general extensions but holds robustly for "small" or "well-structured" fields and groups.