Powers of commutators in infinite groups

Daan Heus · 2025

Read Powers of commutators in infinite groups by Daan Heus (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2505-06130.txt.

Introduction

Daan Heus answers this 2025 algebra question about powers and commutators in groups. A commutator measures how far two elements are from commuting. In finite groups, raising elements to a power coprime to their orders preserves certain commutator relations, up to renaming. Heus asks for exactly which order values that preservation always holds.

The paper settles the question completely, for both the commutator version and a product version. Some value sets keep the property in every group, and others admit counterexamples. The proof mixes combinatorial group theory with facts about a classic group of two-by-two matrices over the real numbers. Results of this kind tidy the border between what powers preserve and what they break.

Catalog record

CREATORDaan Heus
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE67 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2505-06130

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Powers of commutators in infinite groups Daan Heus Abstract Given elements x , u , z x,u,z in a finite group G G such that z z is the commutator of x x and u u , and the orders of x x and z z divide respectively integers k , m ≥ 2 k,m\geq 2 , and given an integer r r that is coprime to k k and m m , there exists w ∈ G w\in G such that the commutator of x r x^{r} and w w is conjugate to z r z^{r} . If instead we are given elements x , y , z ∈ G x,y,z\in G such that x ​ y = z xy=z , whose respective orders divide integers k , l , m ≥ 2 k,l,m\geq 2 , and are given an integer r r that is coprime to k , l k,l and m m , then there exist x ′ x^{\prime} , y ′ y^{\prime} and z ′ z^{\prime} conjugate to respectively x r x^{r} , y r y^{r} and z r z^{r} such that x ′ ​ y ′ = z ′ x^{\prime}y^{\prime}=z^{\prime} . In this paper we completely answer the natural question for which values of k , l , m , r k,l,m,r every group has these properties. The proof uses combinatorial group theory and properties of the projective special linear group PSL 2 ​ ( ℝ ) \mathrm{PSL}_{2}(\mathbb{R}) . Keywords. Infinite groups, Commutators, Conjugacy classes, Combinatorial group theory, Projective special linear group. 2020 Mathematics Subject Classification. 20F12, 20E45. 1 Introduction A Honda group is a group in which every generator of every subgroup generated by a commutator is itself also a commutator. In algebraic terms we say that a group G is Honda if for every integer r and all x,u,z\in G one has [x,u]=z,\,\langle z\rangle=\langle z^{r}\rangle\Rightarrow\exists v,w\in G:[v,w]=z^{r}. Here we use the convention that [x,u]=xux^{-1}u^{-1} . In 1953, K. Honda [ 4 ] proved that every finite group is Honda (Proposition 2.5 below). A natural question to ask is if every group is Honda. However, in 1977 ( [ 8 ] , page 488, Result (C)) S.J. Pride gave a family of groups that he showed were not Honda. On the other hand, B. Martin proved in [ 6 ] that many linear algebraic groups are Honda. We write \sim for “is conjugate to”. With the same proof that K. Honda used to show that every finite group is Honda, one can also show that every finite group G has the following property: for every integer r and all x,u,z\in G one has [x,u]=z,\,\langle x\rangle=\langle x^{r}\rangle,\,\langle z\rangle=\langle z^{r}\rangle\Rightarrow\exists w\in G:[x^{r},w]\sim z^{r}. We call such a group a quasi-Honda group. It is easy to see that every quasi-Honda torsion group is Honda (Proposition 2.3 ). The groups given by S.J. Pride are quasi-Honda (Proposition 2.6 ), so in general quasi-Honda does not imply Honda. One may now wonder if every group is quasi-Honda. To prove that every finite group is Honda, K. Honda applied an argument that W. Burnside had used in 1911 to prove that every finite group G has the following property: for every integer r and all x,y,z\in G one has xy=z,\,\langle x\rangle=\langle x^{r}\rangle,\,\langle y\rangle=\langle y^{r}\rangle,\,\langle z\rangle=\langle z^{r}\rangle\Rightarrow\exists x^{\prime},y^{\prime},z^{\prime}\in G:x^{\prime}y^{\prime}=z^{\prime},\,x^{\prime}\sim x^{r},\,y^{\prime}\sim y^{r},\,z^{\prime}\sim z^{r} ( [ 1 ] , Chapter XV, Theorem VII). We will call a group with this property a quasi-Burnside group. Since it is easy to prove that every quasi-Burnside group is quasi-Honda (Proposition 2.4.(b) ), Honda’s theorem is a consequence of Burnside’s result. Just as in the case of the quasi-Honda property one may wonder if every group is quasi-Burnside. The answer to both of these questions is negative, but we were able to answer a more refined question, which we proceed to formulate. If, in the definitions of quasi-Honda and quasi-Burnside, one has r=-1 , then one can give explicit formulas for w and for x^{\prime},y^{\prime},z^{\prime} ; e.g. w=xu and x^{\prime}=x^{-1} , y^{\prime}=xy^{-1}x^{-1} , z^{\prime}=z^{-1} . However, for general r it is only meaningful to ask for such formulas if one guarantees that the conditions \langle x\rangle=\langle x^{r}\rangle , \langle z\rangle=\langle z^{r}\rangle , and (in the quasi-Burnside case) \langle y\rangle=\langle y^{r}\rangle are satisfied. If r\notin\{\pm 1\} , then these conditions are equivalent to the existence of positive integers k,l,m that are coprime to r for which one has x^{k}=1 , ( y^{l}=1 ,) and z^{m}=1 . This leads to the following definitions. Given a ring R , we write R^{*} for its unit group. Let k,l,m\geq 2 be integers and let r\in(\mathbb{Z}/\mathrm{lcm}(k,l,m)\mathbb{Z})^{*} . A group G is (k,l,m,r) -quasi-Burnside if for all x,y,z\in G one has xy=z,\,x^{k}=y^{l}=z^{m}=1\Rightarrow\exists x^{\prime},y^{\prime},z^{\prime}\in G:x^{\prime}y^{\prime}=z^{\prime},\,x^{\prime}\sim x^{r},\,y^{\prime}\sim y^{r},z^{\prime}\sim z^{r}. Now let r\in(\mathbb{Z}/\mathrm{lcm}(k,m)\mathbb{Z})^{*} . A group G is (k,m,r) -quasi-Honda if for all x,u,z\in G one has [x,u]=z,\,x^{k}=z^{m}=1\Rightarrow\exists w\in G:[x^{r},w]\sim z^{r}. Clearly, a group G is quasi-Burnside if and only if for all k,l,m and every r\in(\mathbb{Z}/\mathrm{lcm}(k,l,m)\mathbb{Z})^{*} the group G is (k,l,m,r) -quasi-Burnside. Analogously, a group G is quasi-Honda if and only if for all k,m and every r\in(\mathbb{Z}/\mathrm{lcm}(k,m)\mathbb{Z})^{*} the group G is (k,m,r) -quasi-Honda. In this paper we classify for which values of k,l,m,r respectively k,m,r every group is (k,l,m,r) -quasi-Burnside respectively (k,m,r) -quasi-Honda. Our results are summarized in the two theorems below. If \gcd(k,m)\leq 2 , and we are given r\in(\mathbb{Z}/\mathrm{lcm}(k,m)\mathbb{Z})^{*} , then we write r^{*} for the unique element in (\mathbb{Z}/\mathrm{lcm}(k,m)\mathbb{Z})^{*} with r^{*}\equiv r\pmod{k} and r^{*}\equiv-r\pmod{m} . This element r^{*} exists by Lemma 2.9 . Theorem 1.1 . Let k,m\geq 2 be integers and let r\in(\mathbb{Z}/\mathrm{lcm}(k,m)\mathbb{Z})^{*} . Then the following are equivalent. (1) Every group is (k,m,r) -quasi-Honda. (2) Every group is (k,k,m,r) -quasi-Burnside, or both \gcd(k,m)\leq 2 holds and every group is (k,k,m,r^{*}) -quasi-Burnside. (3) We have \frac{2}{k}+\frac{1}{m}\geq 1 , or r\in\{\pm 1\} , or both \gcd(k,m)\leq 2 and r^{*}\in\{\pm 1\} . The equivalence (1) \Leftrightarrow (2) of Theorem 1.1 is proved at the end of § 6 , and the equivalence (2) \Leftrightarrow (3) of Theorem 1.1 is a direct consequence of Theorem 1.2 , which we shall prove first. We write \mathrm{PSL}_{2}(\mathbb{R}) for the projective special linear group of degree 2 over the real numbers. Theorem 1.2 . Let k,l,m\geq 2 be integers and let r\in(\mathbb{Z}/\mathrm{lcm}(k,l,m)\mathbb{Z})^{*} . Then the following are equivalent. (1) Every group is (k,l,m,r) -quasi-Burnside. (2) The group \mathrm{PSL}_{2}(\mathbb{R}) is (k,l,m,r) -quasi-Burnside. (3) We have \frac{1}{k}+\frac{1}{l}+\frac{1}{m}\geq 1 or r\in\{\pm 1\} . The implication (1) \Rightarrow (2) of Theorem 1.2 is trivial, the implication (2) \Rightarrow (3) is proved in § 3 , just above Proposition 3.8 , and the implication (3) \Rightarrow (1) is proved at the end of § 2 . The proof of Theorem 1.2 uses facts about products of triples of conjugacy classes in \mathrm{PSL}_{2}(\mathbb{R}) , and the proof of Theorem 1.1 uses a special notion of a reduced word in the free product of a group and an infinite cyclic group. For the entirety of this paper we fix three integers k,l,m\geq 2 . In § 2 we prove some basic facts about (k,l,m,r) -quasi-Burnside and (k,m,r) -quasi-Honda groups. We show that every group being (k,l,m,r) -quasi-Burnside is equivalent to the existence of g,h\in B_{k,l,m}:=\langle a,c\mid a^{k}=(a^{-1}c)^{l}=c^{m}=1\rangle such that a^{r}\cdot g(a^{-1}c)^{r}g^{-1}=hc^{r}h^{-1} (Proposition 2.11.(a)

Read the complete text