Search arXiv⌕ Search

arXiv · 2610.36497

Failure of the congruence Hasse principle for commutators in orientable surface groups

Abstract

Ghosh, Meiri and Sarnak showed that the commutator equation over $\mathrm{SL}_2(\mathbb{Z})$ satisfies a profinite local to global principle, and that for a prime $\ell\equiv\pm1$ mod $5$ it fails for $\mathrm{SL}_2(\mathbb{Z}[1/\ell])$ with a reciprocity obstruction. We extend the failure to surface groups. For the genus-$2$ surface group $Γ_2$ we construct an explicit representation $ρ:Γ_2\to\mathrm{SL}_2(\mathcal{O}_{K,S})$, where $K=\mathbb{Q}(\sqrt{5})$ and $\mathcal{O}_{K,S}=\mathbb{Z}[\frac{1+\sqrt{5}}{2},\frac{1}{10}]$, whose quotients modulo the maximal ideals are the full groups $\mathrm{SL}_2(\mathbb{F}_q)$ except in residue characteristic $3$, where the quotient has order $24$. For every $k\ge2$ the word $[a_1,b_1]^k$ is not a commutator in $Γ_2$, while its image is a commutator in every one of these quotients. The same holds for every surface group $Γ_g$ with $g\ge2$.

Explore related subjects

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Jianchun Wu. 2026-10-27. Failure of the congruence Hasse principle for commutators in orientable surface groups. https://arxiv.org/abs/2610.36497

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

The Ramanujan-Petersson and Selberg conjectures for Maass forms

We prove the Ramanujan-Petersson conjecture for Maass forms of the group $SL(2,Z)$, with the help of automorphic distribution theory and pseudodifferential analysis. The first notion is an alternative to classical automorphic function theory, in which the plane takes the place usually ascribed to the hyperbolic half-plane. The Selberg conjecture for Hecke's group $Γ_0(M)$ follows as well.

math.GR↗

Actions of acylindrically hyperbolic groups on $\ell^1$

We construct explicit affine uniformly Lipschitz actions on $\ell^1$ for certain groups with hyperbolic features. For acylindrically hyperbolic groups, our actions have unbounded orbits, while for residually finite hyperbolic groups and for mapping class groups, the actions have proper orbits, with the induced $\ell^1$-metric quasi-isometric (respectively, almost quasi-isometric, up to a logarithmic factor) to the word metric. This allows us to construct proper continuous affine uniformly Lipschitz actions on $L^1$, with undistorted orbits, for all simple groups of real rank one.

math.GR↗

Terminal Geometry of Rank-Two Nielsen Reduction on Simplicial Trees

Let $X$ be a locally finite simplicial tree, and let $Γ=\langleγ_1,γ_2\rangle\leq\mathrm{Aut}(X)$ be a discrete subgroup generated by hyperbolic automorphisms acting without inversions, whose axes have finite edge intersection. We locate explicitly, using the Euclidean algorithm, the first stopping state of rank-two Nielsen reduction and determine the exact characteristic-set geometry of its terminal Nielsen pair. Write $m_1\geq m_2>0$ for the normalized translation lengths, $l$ for the axis-overlap length, and $d=\gcd(m_1,m_2)$. If $l=0$, the initial pair satisfies the Schottky criterion. If $l>0$, Euclidean quotients and remainders partition $0<l\leq m_1+m_2-d$ into intervals locating the first stopping state and its ordered Nielsen generators. Interior points yield terminal Schottky pairs. At nonfinal right endpoints, an auxiliary translated-axis overlap $l_0$ distinguishes Schottky terminal pairs from pairs containing a nontrivial elliptic generator. At the final endpoint and for larger overlaps, a nontrivial elliptic terminal generator occurs. We compute the terminal translation lengths and the relative geometry of axes or fixed subtrees in every case. Each terminal pair is Nielsen-equivalent to the initial pair and generates $Γ$. If both terminal generators are hyperbolic, they form a Schottky pair; otherwise exactly one is nontrivial elliptic. By local finiteness and discreteness, this element has finite order, so $Γ$ contains nontrivial torsion and is not free of rank two.

math.GR↗