The Wiegold problem and free products of left-orderable groups

Basic Information

• Paper ID: 2510.26073
• Title: The Wiegold problem and free products of left-orderable groups
• Authors: Lvzhou Chen (Purdue University) and Yash Lodha (Purdue University)
• Classification: math.GR (Group Theory), math.GT (Geometric Topology)
• Publication Date: October 30, 2025 (arXiv preprint)
• Paper Link: https://arxiv.org/abs/2510.26073

Abstract

The normal rank or weight of a group greater than 1 means that no single element can normally generate the group. The Wiegold problem, posed in 1976, asks whether there exists a finitely generated perfect group with normal rank greater than 1. This paper proves that the free product of any nontrivial left-orderable groups has normal rank greater than 1, thereby resolving the Wiegold problem by taking free products of finitely generated perfect left-orderable groups. The authors obtain estimates on normal rank through topological arguments, proving a spectral gap property for an unsigned version of stable commutator length. The key element of the proof is an ingenious new construction of left-order families on the free product of two left-orderable groups.

Research Background and Motivation

1. Core Problem

A fundamental concept in group theory is the normal rank of a group, defined as the cardinality of the minimal set of elements that normally generate the group. For free products of groups $G = A \star B$, a long-standing problem concerns: under what conditions is the normal rank greater than 1?

2. Significance of the Problem

• Theoretical Significance: The Wiegold problem posed in 1976 (Problem FP14 in BMS02, Problem 5.52 in KM23) asks whether there exists a finitely generated perfect group (a group equal to its own commutator subgroup) with normal rank greater than 1. This problem has seen little progress since its formulation.
• Topological Applications: Normal rank is closely related to 3-manifold topology. The normal rank of $\pi_1(M)$ provides a lower bound for the Dehn surgery number of a 3-manifold $M$. The Lickorish-Wallace theorem states that any closed oriented connected 3-manifold can be obtained by Dehn surgery on some $n$-component link, and the minimal such $n$ is the Dehn surgery number.
• Related Conjectures:
  • Gordon's Conjecture: For free products $G = \star_{i \in I} A_i$ with $|I| \geq 3$, the normal rank is always greater than 1
  • Scott-Wiegold Conjecture: Special case when factors are finite cyclic groups (proved by Howie)
  • Connected Sum Conjecture: The Dehn surgery number of a connected sum of three non-$S^3$ 3-manifolds is greater than 1

3. Limitations of Existing Methods

• For free products $G = A \star B$, torsion elements may result in normal rank 1. For example, in $G = \mathbb{Z}/2 \star \mathbb{Z}/3$, $w = ab$ normally generates the entire group.
• Since the 1980s, known results require stronger conditions: each free factor is locally indicable (every nontrivial finitely generated subgroup surjects onto $\mathbb{Z}$). However, locally indicable groups cannot be perfect.
• It is conjectured that results hold under torsion-free conditions, but proof techniques are lacking.

4. Motivation of This Work

This paper resolves the problem under the left-orderable condition. A group $G$ is left-orderable if there exists a total order $<$ such that $f < g$ implies $hf < hg$. This condition:

• Is stronger than torsion-freeness
• Is weaker than local indicability
• Encompasses rich classes of groups: free groups, surface groups, braid groups, fundamental groups of hyperbolic 3-manifolds, etc.
• Has abundant examples of finitely generated perfect left-orderable groups

Core Contributions

1. Resolution of the Wiegold Problem: Proves the existence of finitely generated (even finitely presented) perfect groups with normal rank greater than 1 (Theorem A).
2. Normal Rank Estimates for Free Products: Proves that the free product $G = A \star B$ of any nontrivial left-orderable groups has normal rank greater than 1 (Theorem B). More precisely, for any $w \in G$ not conjugate into $A$, the natural map $A \hookrightarrow A \star B$ induces an injection $A \hookrightarrow (A \star B)/\langle\langle w \rangle\rangle$.
3. Spectral Gap Property: Establishes a spectral gap property for the unsigned version of stable commutator length (Theorem C). For boundary-incompressible $w$-admissible surfaces $S$ (without 2-sphere or disk components), $-\chi(S) \geq \deg(S)$.
4. Relative Stacking Construction: Develops the concept and existence theory of relative stacking (Theorem E), an ingenious method for constructing special left-orders on free products.
5. Dynamical Arrangements Method: Introduces the dynamical arrangements technique, systematically constructing group actions satisfying specific equations and inequations.

Detailed Methods

Task Definition

Objective: Prove that for the free product $G = A \star B$ of left-orderable groups $A, B$, if $w \in G$ is not conjugate into $A$, then the image of $A$ in the quotient group $(A \star B)/\langle\langle w \rangle\rangle$ is nontrivial.

Strategy: Assume there exists $a \in A \setminus \{id\}$ such that $a \in \langle\langle w \rangle\rangle$, i.e., there exists an equation: $a = (g_1 w^{n_1} g_1^{-1})(g_2 w^{n_2} g_2^{-1}) \cdots (g_k w^{n_k} g_k^{-1})$ where $k \in \mathbb{Z}^+$, $g_i \in G$, $n_i \neq 0$. Derive a contradiction through topological and dynamical systems methods.

Model Architecture

The proof consists of two main parts:

1. Topological Argument (Theorem D)

Definition of Admissible Surfaces: Given $w \in G$, a map $f: S \to X$ (where $X$ is a $K(G,1)$ space) is called a $w$-admissible surface if:

• Each boundary component either represents a conjugacy class of $w^n$ ($n \neq 0$) or a conjugacy class in some $A_i$
• At least one $w$-boundary component exists

Degree: A $w$-boundary component representing $w^n$ has degree $|n|$; the degree $\deg(S)$ of surface $S$ is the sum of degrees of all $w$-boundary components.

Boundary Incompressibility: A surface $(f,S)$ is boundary-compressible if there exists an embedded subsurface $P \subset S$ homeomorphic to a pair of pants whose two boundaries on the $w$-boundary represent $w^n$ and $w^{-m}$ respectively, with the third boundary representing $w^{m-n}$ in the interior of $S$.

Simple Normal Form: Any $w$-admissible surface can be reduced to simple normal form, decomposed into disk-pieces and annulus-pieces. Each piece $P$ has:

• Valence $d(P)$: number of segments of the polygonal boundary
• Sign change number $sc(P)$: number of consistent segments

Euler Characteristic Formula: For a surface $S$ in simple normal form: $-\chi(S) = \sum_P \left[\frac{d(P)}{2} - \chi(P)\right]$

2. Dynamical Systems Argument (Theorem E)

Definition of Relative Stacking: For a cyclically reduced word $w = a_1b_1 \cdots a_nb_n$, an action $\sigma: G \to \text{Homeo}^+(R)$ and a point $x \in \mathbb{R}$ form a relative stacking if the orbit of $x$ under $w$$\Omega(w,x) = \{x \cdot \sigma(a_1b_1 \cdots a_ib_i) \mid 1 \leq i \leq n\} \cup \{x \cdot \sigma(a_1b_1 \cdots a_i) \mid 1 \leq i \leq n\}$ is stable, i.e.:

1. Each element in $\Omega(w,x)$ appears exactly once
2. $x \cdot \sigma(w) = x$

Existence: For $w$ that is not a proper power, a relative stacking exists.

Technical Innovations

1. $\hat{\lambda}$-Orientation and Sign Changes

From the relative stacking, obtain a valuation $\lambda: J_w \to \mathbb{R}$ (where $J_w$ is the set of junction points on $S^1_w$), pulled back to the surface boundary to get $\hat{\lambda}: J_S \to \mathbb{R}$. This defines the $\hat{\lambda}$-orientation of each arc $\gamma \in F$.

Key Lemma 4.5: The total number of inconsistent segments is at least $2\deg(S)$, so the total number of consistent segments is at most $(|w|-2)\deg(S)$.

Key Lemma 4.6: For each piece $P$, the sign change number $sc(P)$ is even, and $sc(P) > 0$ for disk pieces.

Corollary 4.7: $sc(P) \geq 2\chi(P)$.

2. Proof of Spectral Gap Inequality

Combining the above lemmas: $-\chi(S) = \sum_P \left[\frac{d(P)}{2} - \chi(P)\right] \geq \frac{1}{2}\sum_P [d(P) - sc(P)]$$\geq \frac{1}{2}[|w|\deg(S) - (|w|-2)\deg(S)] = \deg(S)$

3. Dynamical Arrangements Construction

Catenation: Ordered tuples $(I_1, J_1, \ldots, I_n, J_n)$ where each interval has length 2 with endpoints in $\mathbb{Z}$, satisfying specific linking conditions.

Lemmas 5.9 and 5.10: For a reduced word $w = a_1b_1 \cdots a_kb_k$ and catenation intervals, one can construct actions such that specific points under $w$ have images in designated ranges.

System Resolution: Through blow-up construction, reduce solving systems of equations and inequations to constructing actions satisfying specific inclusion and disjointness conditions.

Lemma 5.3: If finite systems $\Lambda_1, \ldots, \Lambda_m$ are each solvable, then their combined system $\Xi(\Lambda_1, \ldots, \Lambda_m)$ is solvable.

4. Existence of Relative Stacking

Proposition 5.5: For $w$ that is not a proper power and any proper prefix $w_1$, there exist an action $\tau$ and a point $x$ such that $x \cdot \tau(w) = x$ but $x \cdot \tau(w_1) \neq x$.

The proof divides into two cases:

• Case 1: $w_1 = a_1b_1 \cdots a_k$
• Case 2: $w_1 = a_1b_1 \cdots a_kb_k$ (requires $b_k \neq b_n$, using the non-proper-power condition)

Through carefully designed dynamical arrangements, construct a closed interval $I$ such that $I \cdot \sigma(w) \subseteq I$ but $(I \cdot \sigma(w_1)) \cap I = \emptyset$; the intermediate value theorem yields the desired fixed point.

Experimental Setup

This is a pure theoretical mathematics paper with no experimental setup, datasets, or numerical experiments. All results are obtained through rigorous mathematical proofs.

Experimental Results

The paper contains no experimental results; all conclusions are theoretical theorems.

Related Work

1. Results on Locally Indicable Groups

• Brodskĭı (1984), Howie (1981), Short (1983): Independently proved Theorem B when each free factor is locally indicable. However, locally indicable groups cannot be perfect.

2. Analogous Results for HNN Extensions

• Howie (Kervaire-Laudenbach) Conjecture: For $G = A \star \mathbb{Z}$, if $p(w) \neq 0$ (where $p: G \to \mathbb{Z}$ is the projection), then $w$ does not normally generate $G$.
• Gerstenhaber-Rothaus (1962): Proof when $A$ is residually finite
• Klyachko (1993): Proof when $A$ is torsion-free
• Chen (2025): New proof method

3. Stable Commutator Length

• Duncan-Howie (1991): Genus problem for single-relation quotients
• Chen (2018), Chen-Heuer (2019), Heuer (2019): Spectral gaps for scl in free products and graphs of groups
• This paper studies the unsigned degree version, requiring stronger left-orderability conditions

4. 3-Manifold Topology

• Gordon (1983): Conjecture on Dehn surgery numbers
• Auckly (1997), Hom-Karakurt-Lidman (2016), Liu-Piccirillo (2024): Bounds on Dehn surgery numbers
• L-space Conjecture: Relationship between left-orderability and fundamental groups of 3-manifolds

5. Other Directions

• Osin-Thom (2013): Conjectured connection between normal rank and first $l^2$-Betti number
• Monod-Ozawa-Thom (2012): Wiegold problem version in irng

Conclusions and Discussion

Main Conclusions

1. Theorem A: There exist finitely generated (even finitely presented) perfect groups with normal rank greater than 1. This completely resolves the Wiegold problem posed in 1976.
2. Theorem B: The free product $G = A \star B$ of any nontrivial left-orderable groups has normal rank greater than 1.
3. Theorem C: Establishes a spectral gap property: for boundary-incompressible $w$-admissible surfaces $S$ (without sphere or disk components), $-\chi(S) \geq \deg(S)$.
4. Corollary 6.1: If $u \in G = A \star B$ is not conjugate into $A$ or $B$ and is not a proper power, then for any $k \geq 2$, the image of $u$ in $G/\langle\langle u^k \rangle\rangle$ has order $k$.

Limitations

1. Condition Restrictions: Results require free factors to be left-orderable. While this encompasses broad classes of groups, it does not cover all torsion-free groups.
2. Number of Free Factors: Theorem B handles only two factors. For three or more factors, Gordon's conjecture remains open.
3. Right-Orderability Question: Problem 6.2 asks whether the quotient group $G/\langle\langle w \rangle\rangle$ is torsion-free or right-orderable (when $w$ is not a proper power). This was resolved by Howie (1982) in the locally indicable case but remains unknown in the left-orderable case.
4. Constructivity: The existence proof for relative stacking is constructive, but the construction process is complex and difficult to implement explicitly.

Future Directions

1. Problem 1.1: Is the normal rank of the free product of $n$ nontrivial left-orderable groups at least $n$? An affirmative answer would imply that the Dehn surgery number of a connected sum of $n$ 3-manifolds is at least $n$.
2. Problem 6.2: Study torsion-freeness and right-orderability of the quotient group $G/\langle\langle w \rangle\rangle$ under left-orderability assumptions.
3. Generalization to Graphs of Groups: Extend relative stacking and dynamical arrangements methods to more general graphs of groups settings.
4. Connection with $l^2$-Betti Numbers: Explore the possibility of the Osin-Thom conjecture in the left-orderable groups case.
5. Computational Aspects: Develop algorithms to determine whether elements in free products of given left-orderable groups normally generate.

In-Depth Evaluation

Strengths

1. Major Breakthrough: Resolves a nearly 50-year-old open problem (Wiegold problem), representing significant progress in group theory.
2. Methodological Innovation:
   • The relative stacking concept cleverly transforms algebraic problems into dynamical systems problems
   • Dynamical arrangements technique provides new tools for systematically constructing group actions
   • Perfect combination of topological arguments (Euler characteristic estimates) with dynamical systems arguments (group action construction)
3. Technical Depth:
   • Introduction of $\hat{\lambda}$-orientation and analysis of sign change numbers demonstrate profound geometric insight
   • Blow-up construction and system resolution method (Lemma 5.3) have universal applicability
   • Careful analysis of two cases (prefix ending in $A$ or $B$) demonstrates rigor
4. Broad Impact:
   • Provides numerous new examples of perfect groups with normal rank greater than 1
   • Direct applications to 3-manifold topology (Dehn surgery numbers)
   • Relative stacking methods may apply to other group-theoretic problems
5. Writing Quality: Paper structure is clear, progressing from simple examples (Example 5.8) to general cases, facilitating understanding.

Weaknesses

1. Technical Complexity:
   • Existence proof for relative stacking (Section 5) contains numerous technical details, particularly in Case 2
   • Dynamical arrangements construction requires careful design of catenation intervals, lacking intuitive appeal
   • Some lemmas (e.g., Lemma 4.3) have correct but highly technical proofs
2. Condition Restrictions:
   • Left-orderability is an intermediate condition between torsion-freeness and local indicability; ideally only torsion-freeness would be needed
   • Only handles two-factor free products; multi-factor case remains open
3. Computability:
   • While the construction is theoretically sound, no discussion of how to actually compute or verify whether given groups satisfy the conditions
   • Explicit construction of relative stacking in concrete examples may be extremely complex
4. Generalizability:
   • Unclear whether methods can be extended beyond HNN extensions to more general graphs of groups
   • Relationship to signed version of stable commutator length could be further explored

Impact

1. Theoretical Significance:
   • Resolves long-standing open problem, will be cited in group theory textbooks
   • Relative stacking and dynamical arrangements methods may become standard tools for studying group actions
   • Provides new perspective on studying normal rank
2. Applied Value:
   • Direct contribution to Dehn surgery number research in 3-manifold topology
   • May provide insights for related problems like the L-space conjecture
   • Constructed examples of perfect left-orderable groups useful for other research
3. Reproducibility:
   • Proof is entirely theoretical and can be verified step-by-step
   • Main theorem statements are clear and easy to cite
   • Technical details, while complex, are complete

Applicable Scenarios

1. Group Theory Research:
   • Study normal rank of free products, HNN extensions, and graphs of groups
   • Construct perfect groups with specific properties
   • Study structural theory of left-orderable groups
2. Geometric Topology:
   • Estimate Dehn surgery numbers of 3-manifolds
   • Study L-spaces and fundamental groups with left-orderable structures
   • Analyze surface subgroups and one-relator groups
3. Dynamical Systems:
   • Construct group actions with specific dynamical properties
   • Study homeomorphism groups of the real line
   • Analyze relationships between order structures and group actions
4. Combinatorial Group Theory:
   • Study systems of equations and inequations in groups
   • Analyze variants of word and conjugacy problems
   • Construct counterexamples and pathological examples

References

The paper cites extensive literature, with key references including:

1. Wiegold Problem: BMS02 Baumslag-Myasnikov-Shpilrain, KM23 Kourovka Notebook
2. Locally Indicable Groups: Bro84 Brodskĭı, How81 Howie, Sho83 Short
3. Left-Orderable Groups: DNR14 Deroin-Navas-Rivas survey, GS87 Ghys-Sergiescu, HL19, HL25 Hyde-Lodha
4. Stable Commutator Length: Cal09 Calegari, Che18, Che20, Che25 Chen, DH91 Duncan-Howie
5. 3-Manifold Topology: Gor83 Gordon, LP24 Liu-Piccirillo
6. Stacking Methods: LW17 Louder-Wilton, Mil21 Millard

Summary: This is a high-quality theoretical paper resolving an important open problem, introducing innovative techniques of relative stacking and dynamical arrangements, cleverly combining topology, dynamical systems, and group theory. Despite technical complexity, the methods are universal and profound, with significant impact on both group theory and geometric topology. The paper's main contribution lies not only in resolving the Wiegold problem but also in developing new tools potentially applicable to broader problems.