Self-mapping Degrees of 3-Manifolds

Hongbin Sun, Shicheng Wang, Jianchun Wu, Hao ZhengOsaka 2008 · DOI · arXiv

TL;DR

This paper provides a complete classification of the "self-mapping degrees" for all closed, oriented 3-manifolds within Thurston’s geometrization framework. By leveraging the unique geometric structures of 3D space, the authors provide explicit sets of integers that represent possible degrees for self-maps across all eight Thurston geometries.

What Problem Does This Paper Address?

In topology, a fundamental question is: given a space $M$, what are the possible "degrees" of a continuous map from $M$ to itself? The degree of a map essentially counts how many times the domain "wraps around" the target. For a circle ($S^1$), the degrees are simply all integers ($\mathbb{Z}$). For surfaces, the answer is also well-understood. However, as the dimension $n$ increases, the problem becomes notoriously difficult because we lack a complete classification of manifolds for $n > 3$.

The authors focus on the 3-dimensional case. Thanks to Thurston’s Geometrization Conjecture (which was famously proven by Grigori Perelman shortly before this paper was finalized), we know that every prime 3-manifold can be decomposed into pieces that each support one of eight distinct geometric structures. This paper fills the remaining gaps in the literature to provide a comprehensive list of self-mapping degrees, denoted as $D(M)$, for all such manifolds.

Methods

The authors employ a "divide and conquer" strategy based on the geometric properties of the manifolds. They categorize 3-manifolds into five classes:

  1. Class 1: Manifolds with $S^3$ (spherical) or $S^2 \times E^1$ geometry.
  2. Class 2: Connected sums of Class 1 manifolds (non-prime manifolds).
  3. Class 3: Torus bundles and semi-bundles (supporting $E^3$, Sol, or Nil geometries).
  4. Class 4: Nil 3-manifolds that are not torus bundles.
  5. Class 5: Manifolds supporting $H^2 \times E^1$ geometry.

The methodology involves a mix of algebraic topology and differential geometry. For many cases, the authors use the fact that a non-zero degree map between these manifolds must induce an injective homomorphism between their fundamental groups. They specifically utilize Seifert fiber space theory, orbifold covering data, and the properties of fundamental groups to derive explicit formulas for the sets of degrees.

Datasets

Not applicable. This is a theoretical mathematics paper; it does not use empirical datasets or benchmarks.

Key Insights and Ideas

  • The "Degree > 1" Threshold: A pivotal insight (Theorem 1.0) is that a 3-manifold admits a self-map of degree $> 1$ if and only if it belongs to specific geometric families (like being covered by a torus bundle or $F \times S^1$). For manifolds outside these families, the possible degrees are restricted to a very small set, typically $\{0, 1\}$ or $\{0, 1, -1\}$.
  • Geometric Rigidity: The paper demonstrates how the underlying geometry of a manifold strictly constrains its topological maps. For instance, in Nil and Sol geometries, the possible degrees are often restricted to squares ($l^2$) or specific arithmetic progressions.
  • Connected Sum Complexity: Calculating degrees for connected sums (Class 2) is particularly tricky. The authors show that the degree set of a connected sum is essentially the intersection of the degree sets of its individual components, provided they are not lens spaces.
  • Orbifold Coverings: For Class 4 and 5 (Seifert manifolds), the authors translate the problem of 3-manifold maps into the language of 2-dimensional orbifold coverings, which are much easier to compute.

Key Results

The paper culminates in a series of theorems (1.1 through 1.5) that provide the exact set $D(M)$ for each class:

  • Spherical Manifolds: For $S^3$ geometry, the degrees are often of the form $\{k^2 \pmod{|G|}\}$, where $|G|$ is the order of the fundamental group.
  • Nil Manifolds: For certain Nil manifolds, the degrees are exactly the set of squares $\{l^2 \mid l \equiv 1 \pmod k\}$ for specific values of $k$ (e.g., $k=3, 4, 6$).
  • Hyperbolic-related Manifolds: For $H^2 \times E^1$ manifolds, the degree set is determined by a complex intersection of sets derived from the manifold's Seifert invariants.
  • Orientation Reversal: The authors provide specific conditions under which a manifold admits a degree $-1$ map (meaning it is "amphicheiral" or admits an orientation-reversing self-map).

Limitations and Future Work

The primary limitation is that the results are strictly bounded by "Thurston’s picture." While the Geometrization Conjecture covers all closed oriented 3-manifolds, the paper does not extensively address manifolds with boundaries or non-orientable manifolds.

Additionally, while the paper provides the sets of degrees, it doesn't always describe the homotopy classes of the maps achieving those degrees. A natural next step in this field is the "Mapping Degree Problem": given two different manifolds $M$ and $N$, what is the set of degrees of maps from $M$ to $N$? This paper provides the foundation for that broader inquiry by solving the self-mapping case.

Summary updated

Abstract

For each closed oriented 3-manifold $M$ in Thurston's picture, the set of degrees of self-maps on $M$ is given.
BibTeX
@inproceedings{sun2008self,
  author = {Hongbin Sun and Shicheng Wang and Jianchun Wu and Hao Zheng},
  title = {{Self-mapping Degrees of 3-Manifolds}},
  booktitle = {Osaka},
  year = {2008},
  doi = {10.48550/arxiv.0810.1801},
}
Figure