By Mostow Rigidity, a closed hyperbolic 3-manifold has a unique, inflexible geometric structure. Its volume, its length spectrum, and its fundamental group are completely locked in place. Yet, if we shift our perspective from the manifold itself to its infinite family of finite-sheeted covering spaces, this geometric stiffness gives way to an astonishing degree of topological freedom.
This phenomenon is known as virtual universality. It reveals that the family of finite covers of a single hyperbolic 3-manifold is rich enough to mimic or "dominate" the topology of completely arbitrary 3-manifolds.
The Concept of Virtual Domination
In geometric topology, a manifold $M$ is said to dominate another manifold $N$ if there exists a continuous map from $M$ to $N$ with a non-zero degree. Intuitively, this means $M$ is at least as topologically complex as $N$; it can be wrapped around $N$ a finite number of times without collapsing.
A manifold virtually dominates $N$ if it possesses a finite-sheeted covering space $\widetilde{M}$ that dominates $N$. In Virtual domination of 3–manifolds, Hongbin Sun proved a startling universality result: every closed oriented hyperbolic 3-manifold virtually dominates every closed oriented 3-manifold.
[ Rigid Hyperbolic Manifold M ]
│
▼ (Pass to a large finite cover)
[ Flexible Covering Space M̃ ]
│
▼ (Degree-1 Map via Skeleton Mimicry)
[ Arbitrary Target Manifold N ]
This was later sharpened by Sun and Yi Liu in Virtual 1-domination of 3-manifolds, demonstrating that the map can achieve exactly degree 1. A degree-1 map implies that the fundamental group of the target $N$ is a quotient of the fundamental group of the cover $\widetilde{M}$, and its homology maps surjectively onto $N$. This means that a single hyperbolic manifold contains the structural blueprint for all possible 3-dimensional shapes within its family of finite covers.
The Mechanism: Topological Skeleton Mimicry
How can a geometrically rigid space map onto an arbitrary target? The underlying mechanism relies on replicating the "skeleton" of the target manifold inside a large finite cover of the host hyperbolic manifold.
Every 3-manifold can be decomposed into simpler pieces, such as handlebodies or link complements. For instance, the global topology of these spaces is often characterized by how internal boundaries or links behave, a theme explored by Bellettini et al. in A table of n-component handlebody links of genus n+1 up to six crossings. To construct a degree-1 map from a cover $\widetilde{M}$ to a target $N$, one must find a complex inside $\widetilde{M}$ that perfectly replicates the handlebody skeleton of $N$.
Because the fundamental group of a hyperbolic manifold is incredibly rich, it contains a vast web of virtual subgroups. By carefully selecting loops and surfaces that match the intersection patterns of the target $N$'s skeleton, researchers can isolate a matching subcomplex. By passing to a sufficiently large finite cover, the self-intersections of this subcomplex can be "unrolled" and eliminated, leaving a cleanly embedded template. The rest of the cover $\widetilde{M}$ is then collapsed away, leaving a well-defined, degree-1 map onto $N$.
Subgroup Separability and the Virtual Space
The ability to cleanly unroll self-intersections in a finite cover depends on a group-theoretic property known as subgroup separability or LERF (Locally Extended Residually Finite). If a fundamental group is LERF, any finitely generated subgroup (such as the group representing an essential surface or a skeleton component) can be isolated in a finite-sheeted cover.
This property is deeply tied to several major milestones in 3-manifold topology:
- The Virtually Haken Conjecture: Marc Lackenby in Heegaard splittings, the virtually Haken conjecture and Property (τ) and Covering spaces of 3-orbifolds demonstrated how the complexity of a space (measured via Heegaard gradient and linear homology growth) scales predictably under finite covers, providing a pathway to finding essential surfaces.
- The Virtually Fibred Conjecture: This property guarantees that complex geometric structures can unfold into highly organized configurations. For example, Xiao Guo showed in Virtually fibred Montesinos links of type SL2 how specific geometric link complements unfold into surface bundles over a circle within finite covers.
- Spectral Geometry: The way these covers unroll also dictates their geometric invariants. Long et al. showed in LERF and the Lubotzky-Sarnak conjecture that Cheeger constants can vanish in sequences of covers, while Marc Lackenby established in Spectral geometry, link complements and surgery diagrams that combinatorial data from surgery diagrams directly bounds these spectral geometric invariants.
However, this virtual flexibility has strict dimensional boundaries. In Non-LERFness of arithmetic hyperbolic manifold groups and mixed 3-manifold groups, Hongbin Sun demonstrated that while subgroup separability works reliably in three dimensions, it fails for general arithmetic hyperbolic manifolds in higher dimensions. This highlights 3-manifolds as a unique sweet spot where algebraic separability and topological flexibility perfectly align.
Hyperbolic Dominance in Bounded and Mixed Manifolds
The power of hyperbolic geometry to generate virtual complexity is not restricted to closed manifolds; it extends to manifolds with boundaries and mixed geometric structures.
In Virtual domination of 3‐manifolds II, Hongbin Sun extended these results to cusped hyperbolic 3-manifolds (such as those formed by the complements of alternating links studied by Adams et al. in Augmented cellular alternating links in thickened surfaces are hyperbolic). The author established virtual 8-domination, proving that cusped hyperbolic manifolds can map onto any compact 3-manifold with torus boundaries.
Crucially, a manifold does not even need to be fully hyperbolic to inherit this universal complexity. According to the JSJ decomposition, a 3-manifold can be cut along tori into a mix of hyperbolic and Seifert-fibered (locally symmetric) pieces. In Virtual domination of 3-manifolds, III, the author proved that if a mixed 3-manifold contains at least one hyperbolic component that intersects the boundary, that single component acts as an engine of complexity. Its infinite family of covers can absorb and dominate the more rigid Seifert-fibered pieces, allowing the entire mixed manifold to virtually dominate other spaces.
This algebraic richness is so profound that 3-manifold groups even dictate constraints when stepping up a dimension, as seen in work by Wang et al. in Minimal Euler characteristics of 4-manifolds with 3-manifold groups. Ultimately, the theory of virtual domination shows that the infinite landscape of finite covering spaces transforms hyperbolic 3-manifolds from rigid geometric objects into universal templates for low-dimensional topology.
Key Foundational Work
Heegaard splittings, the virtually Haken conjecture and Property (τ) (2006) — contributed 2 concept elements to this direction.
Covering spaces of 3-orbifolds (2005) — contributed 2 concept elements to this direction.