🔢 mathematics

The top cohomology of principal congruence subgroups of special linear groups over Euclidean number rings

This paper generalizes the Lee–Szczarba question by proving that for principal congruence subgroups of $\text{SL}_n(R)$ over Euclidean number rings, the natural map from the top cohomology to the reduced homology of the associated Tits building quotient is always surjective, and it establishes sufficient conditions on the prime $p$ for this map to be an isomorphism.

Original authors: Urshita Pal

Published 2026-05-07

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Urshita Pal

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 count the number of "holes" in a very complex, multi-dimensional shape. In mathematics, this is called cohomology. The paper you are asking about is a deep dive into counting these holes for a specific type of shape that arises from a group of numbers called Special Linear Groups over Euclidean number rings.

To make this understandable, let's break it down using a few analogies.

1. The Players: The Grid and the Filter

Think of the Special Linear Group ( $SL_n$ ) as a giant, infinite grid of points in space. These points represent ways to arrange numbers in a grid (a matrix) so that they fit together perfectly (the determinant is 1).

Now, imagine you put a filter over this grid. This filter is a "congruence subgroup." It only lets through the points that look a certain way when you divide them by a specific number (let's call this number $p$ ).

The Analogy: Imagine a sieve. The big grid is flour. The sieve (the congruence subgroup) only lets through the tiny grains that match a specific pattern. The paper studies the shape formed by just these filtered grains.

2. The Goal: Finding the "Top" Hole

Mathematicians know that these shapes have holes at various levels. There's a limit to how high up you can find a hole. This limit is called the top cohomology.

The Analogy: Imagine a skyscraper made of Lego blocks. You know there are empty spaces (holes) inside the building. The "top cohomology" is the highest floor where you can still find an empty room. The paper asks: What does the highest empty room look like, and how many of them are there?

3. The Map: Connecting Two Worlds

The paper focuses on a specific question posed by mathematicians Lee and Szczarba. They wondered if there is a perfect map (an isomorphism) between two different ways of describing these highest holes:

The Algebraic Way: Counting holes directly in the filtered grid (the congruence subgroup).
The Geometric Way: Looking at a giant, abstract structure called the Tits Building. Think of the Tits Building as a massive, multi-dimensional "city" made of flags and towers. When you fold this city up according to the rules of your filter, you get a smaller, quotient city.

The Big Question: Is the number of holes in the filtered grid exactly the same as the number of holes in this folded-up city?

4. The Discovery: A One-Way Street and a Two-Way Street

The author, Urshita Pal, proves two main things:

The One-Way Street (Surjectivity): The author proves that you can always map the holes from the grid to the city without losing any information. Every hole in the city has a corresponding hole in the grid. It's like saying, "If you find a room in the city, you can definitely find a matching room in the grid."
The Two-Way Street (Isomorphism): The author also figures out exactly when the map works both ways (meaning the counts are identical). It turns out this happens under specific conditions regarding the "units" (special numbers that can be multiplied to get 1) in the number system.
- The Metaphor: Imagine the grid and the city are two different languages. The author found that you can always translate from Grid-English to City-English. However, you can only translate back perfectly (making them identical) if the language has certain "vocabulary rules" (specifically, how the units behave when you add them together).

5. The Toolkit: Building Blocks and Connectivity

To prove this, the author had to build new mathematical tools.

Simplicial Complexes: These are shapes built out of triangles, tetrahedrons, and their higher-dimensional cousins. The author built specific "complexes" (like a scaffolding) to hold the shape together.
Connectivity: The author proved that these scaffolding structures are "highly connected."
- The Analogy: Imagine a net. If the net is "highly connected," it means you can't pull it apart easily; it's very sturdy. The author proved that for certain number systems (like the Gaussian integers or Eisenstein integers), this net is so sturdy that it holds the shape together perfectly, allowing the map between the grid and the city to be a perfect match.

6. The Results: When Does It Work?

The paper provides a checklist. If your number system and your filter number ( $p$ ) satisfy certain conditions (like the list of units behaving nicely), then the "Top Hole" in the grid is exactly the same as the "Top Hole" in the folded city.

The author gives specific examples where this works, such as:

Using the number system of Gaussian Integers (numbers like $a + bi$) with the prime number 3.
Using Eisenstein Integers with specific primes like $4\omega + 1$ .

Summary

In simple terms, this paper solves a puzzle about counting the highest-level "empty spaces" in a complex mathematical structure.

It proves you can always translate the count from one mathematical object to another.
It gives a precise recipe for when that translation is a perfect, one-to-one match.
It does this by building sturdy, high-dimensional "nets" (simplicial complexes) and proving they don't fall apart under specific conditions.

This allows mathematicians to calculate the size of these top holes for many new types of number systems, not just the standard whole numbers.

Technical Summary: The Top Cohomology of Principal Congruence Subgroups of Special Linear Groups over Euclidean Number Rings

Problem Statement
The paper investigates the top-degree cohomology of principal congruence subgroups $\Gamma_n(\mathfrak{p})$ of the special linear group $SL_n(R)$ , where $R$ is a Euclidean number ring and $\mathfrak{p}$ is a prime ideal. By Borel–Serre duality, the cohomology of these arithmetic groups in the top degree $\nu$ is isomorphic to the coinvariants of the Steinberg module $St_n(K)$ (where $K$ is the fraction field of $R$ ) under the action of $\Gamma_n(\mathfrak{p})$ .

The central problem, originally posed by Lee and Szczarba for $R=\mathbb{Z}$ , is to determine when the natural map from the Steinberg coinvariants to the reduced homology of the quotient of the Tits building is an isomorphism:
$H^\nu(\Gamma_n(\mathfrak{p}); \mathbb{Q}) \twoheadrightarrow \tilde{H}_{n-2}(T_n(K)/\Gamma_n(\mathfrak{p}); \mathbb{Q})$
While Miller, Patzt, and Putman (MPP) resolved this for $R=\mathbb{Z}$ and primes $p \leq 5$ , the question remained open for other Euclidean domains (such as Gaussian and Eisenstein integers) and specific primes. The paper seeks to generalize the MPP results to arbitrary Euclidean number rings and identify sufficient conditions on the prime $\mathfrak{p}$ for the map to be an isomorphism.

Methodology
The proof strategy relies on two main components: a high-connectivity analysis of specific simplicial complexes and a spectral sequence argument relating these complexes to the Steinberg module.

Simplicial Complexes and Connectivity:
The authors define and analyze complexes of "determinant- $U$ partial $U$ -bases" ( $BDU_n$ ) and their "augmented" versions ( $BDAU_n$ ), where $U$ is the image of the units of $R$ in the residue field $F = R/\mathfrak{p}$ .
- They generalize the complexes $BDA^\pm_n(\mathbb{F}_p)$ used by MPP.
- Key Technical Innovation: Unlike MPP, who relied on the Bykovskii presentation of the Steinberg module (specifically for $\mathbb{Z}$ ) to prove high connectivity, this paper provides a self-contained proof of high connectivity for $BDAU_n(F)$ .
- The authors prove that if $F$ $F$ and $U$ $U$ satisfy specific algebraic conditions (detailed below), the complex $BDAU_n(F)$ $B D A U_{n} (F)$ is $(n-1)$ $(n - 1)$ -connected. This involves:
  - Constructing a retraction from a larger complex of bases to the determinant-complex.
  - Demonstrating that "initial D-triangle maps" and "initial D-suspend maps" (generators of the homotopy groups) are null-homotopic under the specified conditions.
  - Verifying the base case ( $n=2$ ) by showing $BDAU_2(F)$ is 1-connected (simply connected) for specific pairs of rings and primes.
Spectral Sequence and Steinberg Module:
The second component utilizes a map-of-posets spectral sequence to relate the homology of the simplicial complexes to the Steinberg module.
- The authors establish a surjection from the chain complex of $BDU_n(F)$ to the Steinberg module $St_n(K)$ .
- They prove that if $BDAU_n(F)$ is $(n-1)$ -connected, the map from the Steinberg coinvariants to the homology of the quotient Tits building is an isomorphism.
- Crucially, this approach avoids assuming the Bykovskii presentation for $St_n(K)$ , allowing the results to apply to rings like $\mathbb{Z}[\sqrt{-2}]$ where such presentations were not previously established.

Key Contributions and Results

Generalized Isomorphism Theorem (Theorem 1.5):
The paper proves that for a Euclidean domain $R$ and a prime $\mathfrak{p}$ , the map
$(St_n(K))_{\Gamma_n(\mathfrak{p})} \to \tilde{H}_{n-2}(T_n(K)/\Gamma_n(\mathfrak{p}))$
is always surjective. Furthermore, it is an isomorphism if the residue field $F = R/\mathfrak{p}$ and the image of the units $U \subset F^\times$ satisfy five specific conditions:
1. $F^\times/U \cong C_2$ (cyclic group of order 2).
2. A specific additive generation property for elements in $F^\times \setminus U$ using units in $U$ .
3. $F$ is additively generated by $U$ .
4. A condition regarding the element 2 or specific linear combinations of units.
5. The complex $BDAU_2(F)$ is 1-connected.
Specific Examples (Theorem 1.4):
The authors verify these conditions for several pairs of Euclidean rings and primes, establishing the isomorphism for:
- $(\mathbb{Z}[\sqrt{-2}], 1+\sqrt{-2})$
- $(\mathbb{Z}[i], 3)$ , $(\mathbb{Z}[i], 2i+1)$
- $(\mathbb{Z}[\omega], 4\omega+1)$ , $(\mathbb{Z}[\omega], 4\omega+3)$ , $(\mathbb{Z}[\omega], 3\omega+1)$ , $(\mathbb{Z}[\omega], 3\omega+2)$
  where $\omega$ is a primitive cube root of unity.
Recursive Rank Formula:
Leveraging the isomorphism, the paper provides a recursive formula (Proposition 3.24) to compute the rank of the top cohomology group $\tilde{H}_{n-2}(T_n(K)/\Gamma_n(\mathfrak{p}))$ , generalizing previous results for $\mathbb{Z}$ .

Significance and Claims
The paper claims to provide a "generalized version" of the Lee–Szczarba question's resolution. Its primary significance lies in:

Removing Dependence on Bykovskii Presentation: By avoiding the Bykovskii presentation of the Steinberg module (which was known only for $\mathbb{Z}$ , $\mathbb{Z}[i]$ , and $\mathbb{Z}[\omega]$ ), the authors extend the isomorphism results to other Euclidean rings, such as $\mathbb{Z}[\sqrt{-2}]$ .
Case-by-Case Verification: The paper acknowledges that condition (5) (1-connectivity of $BDAU_2(F)$ ) must be checked manually for each specific ring and prime. It demonstrates this feasibility for Gaussian and Eisenstein integers but notes that a general algorithm for all Euclidean rings is not provided.
Surjectivity: The paper establishes that the map is always surjective for $n \geq 2$ , regardless of the specific conditions on $F$ and $U$ , which is a robust result in itself.

The authors remain modest regarding the scope, noting that while they provide a sufficient set of conditions for the isomorphism, they do not claim these conditions are necessary, nor do they currently know if there are other pairs of rings and primes satisfying these conditions beyond the examples listed.