Virtual domination of 3-manifolds II

Hongbin Sun

Abstract

In the study of 3-manifold topology, the concept of "domination" provides a natural partial ordering of topological complexity. This paper establishes that any closed oriented 3-manifold $M$ with positive simplicial volume acts as a universal source of complexity: it virtually 1-dominates any other closed oriented 3-manifold $N$. This result generalizes previous findings restricted to hyperbolic domains and identifies positive simplicial volume as the definitive threshold for virtual 1-domination in three dimensions.

Introduction

A fundamental pursuit in topology is the classification and comparison of manifolds. In the 1970s, Mikhail Gromov introduced the notion of domination. A manifold $M$ is said to dominate $N$ if there exists a continuous map $f: M \to N$ with a non-zero degree. Conceptually, this suggests that $M$ is at least as "topologically complicated" as $N$.

A more robust version of this relationship is virtual 1-domination. We say $M$ virtually 1-dominates $N$ if there exists a finite-sheeted covering space $M'$ of $M$ that admits a degree-1 map $f: M' \to N$. A degree-1 map is particularly significant because it implies that $M'$ can be continuously deformed or "pinched" to perfectly represent the topology of $N$. The paper "Virtual Domination of 3-Manifolds II" by Hongbin Sun provides a definitive characterization of which 3-manifolds possess this universal dominating power.

The Criterion: Positive Simplicial Volume

The central result of the paper is that the existence of a degree-1 map from a finite cover of $M$ to any $N$ is determined by the simplicial volume $\|M\|$. Simplicial volume is a topological invariant that measures the complexity of a manifold in terms of the minimum number of singular simplices needed to represent its fundamental class.

In the context of 3-manifolds, the Geometrization Conjecture implies that $\|M\| > 0$ if and only if the manifold contains at least one hyperbolic component in its Jaco-Shalen-Johannson (JSJ) decomposition. Sun proves that this condition is sufficient for $M$ to virtually 1-dominate any closed oriented 3-manifold $N$. This establishes positive simplicial volume as the definitive criterion for "universal complexity" in 3-manifold topology.

Methodological Framework

The proof is a tour de force of geometric topology, integrating classical decomposition theorems with sophisticated modern techniques in geometric group theory.

1. Reduction to Hyperbolic Targets

Building on work by Boileau and Wang, the author first simplifies the problem. It is shown that if a manifold $M$ can virtually 1-dominate all hyperbolic 3-manifolds, it can, by extension, dominate all closed 3-manifolds. This reduces the "target" space $N$ to the hyperbolic case.

2. Leveraging the JSJ Decomposition

For a manifold $M$ with $\|M\| > 0$, the JSJ decomposition ensures the existence of at least one hyperbolic piece. This piece is typically a "cusped" hyperbolic manifold—a manifold with a complete hyperbolic metric whose ends are tori. The strategy focuses on using the inherent geometric richness of this hyperbolic piece to construct the required map.

3. The "Good Pants" Construction and Cusped Geometry

The core of the construction utilizes the "Good Pants" machinery developed by Kahn and Markovic. This technique involves building nearly-geodesic surfaces within hyperbolic manifolds by assembling "pairs of pants" (spheres with three boundary components). To handle the specific challenges of the cusped JSJ pieces, Sun employs the "Hamster Wheel" and "Umbrella" constructions developed by Kahn and Wright. These specialized geometric structures allow for the management of the "leaks" in geometry that occur near the cusps, ensuring the constructed surfaces behave predictably.

4. Homological Control and the Connection Principle

The final stage involves a "Connection Principle" with homological control. The author demonstrates that the constructed 2-complexes can be lifted to a finite cover $M'$ in such a way that the map to the target $N$ can be extended across the entire manifold while maintaining a degree of 1. This ensures that the homology of the manifold can be made sufficiently "rich" to support the map to $N$.

Key Insights and Implications

The paper establishes that hyperbolic geometry is the primary driver of topological complexity in three dimensions. Even if a 3-manifold is largely composed of "simple" components (such as Seifert fiber spaces), a single hyperbolic JSJ component provides sufficient "topological DNA" to reconstruct the architecture of any other 3-manifold within a finite cover.

Furthermore, the result implies that 3-manifolds with positive simplicial volume satisfy the Universal Realization of Cycles (URC) property. This means that for any homology class in any topological space, there exists a finite cover of $M$ that maps onto that class. Additionally, Sun shows that for any finite abelian group $A$, $M$ has a finite cover $M'$ such that $A$ is a direct summand of the first homology group $H_1(M'; \mathbb{Z})$.

Conclusion and Future Directions

Sun’s work completes the hierarchy for virtual 1-domination in three dimensions, showing that positive simplicial volume is the exact threshold for this property. However, several questions remain:

  • Higher Dimensions: For manifolds of dimension $n \ge 4$, a complete answer currently seems out of reach. The classification of 3-manifolds via Geometrization provides a roadmap that does not yet exist for higher-dimensional spaces.
  • Effective Bounds: While the existence of a finite cover $M'$ is proven, the paper does not establish a bound on the degree of the cover. Determining the number of "sheets" required to dominate a specific target $N$ remains an open challenge.
  • Boundary Conditions: This study focuses on closed manifolds. Extensions to manifolds with non-empty tori boundary are addressed in subsequent research by the author.

Summary updated

Abstract

For any closed oriented 3-manifold $M$ with positive simplicial volume and any closed oriented 3-manifold $N$, we prove that there exists a finite cover $M'$ of $M$ that admits a degree-1 map $f:M'\to M$, i.e. M virtually 1-dominates N. This result generalizes previous virtual domination results with closed hyperbolic domain to more general domains.
BibTeX
@article{sun2023virtual,
  author = {Hongbin Sun},
  title = {{Virtual domination of 3-manifolds II}},
  journal = {Journal of the London Mathematical Society},
  year = {2023},
  doi = {10.1112/jlms.12770},
}
Figure