A Pontryagin class obstruction for purely electric and purely magnetic Weyl curvature tensors

This paper establishes that the existence of purely electric or purely magnetic Weyl curvature tensors on compact $4k$-dimensional manifolds is obstructed by the vanishing of specific products of Pontryagin classes in the top-degree de Rham cohomology, thereby providing nontrivial topological constraints on such Lorentzian metrics and their associated Petrov subtypes.

The Gist

Imagine you are an architect trying to build a house (a manifold) that can hold a very specific type of furniture (a metric). In the world of physics, specifically General Relativity, this "house" is the fabric of spacetime, and the "furniture" is the curvature caused by gravity.

For decades, physicists have been looking for special kinds of "rooms" in this house where the gravity behaves in a very simple, predictable way. They call these Purely Electric (PE) or Purely Magnetic (PM) rooms.

• Purely Electric: Think of this like a room where gravity only pulls or stretches (like a tidal force), with no "twisting" or swirling.
• Purely Magnetic: Think of this as a room where gravity only swirls or twists, with no pulling or stretching.

Most of the time, gravity does a mix of both. But sometimes, in very specific theoretical scenarios, it might be purely one or the other.

The Big Question

The paper asks a simple but profound question: "Can we build any house that allows for these special rooms?"

In math terms: If a shape (manifold) is flexible enough to hold a spacetime metric at all, is it guaranteed to be able to hold a metric where gravity is purely electric or purely magnetic everywhere?

The Answer: The "Topological Lock"

The author, Thijs de Kok, says: Not necessarily.

He discovers a hidden "lock" on the shape of the universe that prevents these special rooms from existing in certain buildings. This lock is made of something called Pontryagin Classes.

To understand this, imagine the shape of your house has a "DNA" or a "fingerprint" made of numbers. These numbers (the Pontryagin classes) describe the global twists and turns of the shape itself, regardless of what furniture you put inside.

The Discovery:
De Kok proves that if a building has a specific "DNA" (non-zero Pontryagin classes), it is impossible to arrange the furniture inside so that the gravity is purely electric or purely magnetic everywhere. The shape of the building itself forbids it.

How the Magic Trick Works (The Analogy)

The author uses a clever mathematical trick involving mirrors and symmetry.

1. The Mirror Test: Imagine you have a piece of fabric (the curvature tensor). If you look at it in a mirror (an orientation-reversing isometry), does it look the same?
   • If it looks the same, it's Even (Purely Electric).
   • If it looks like a negative reflection, it's Odd (Purely Magnetic).
2. The Paradox: The author shows that if you have a fabric that is perfectly Even or perfectly Odd under this mirror test, and you try to calculate a specific "total twist" of the fabric (the Pontryagin form), the math forces that total twist to be zero.
3. The Contradiction: But, we know that some shapes (like a specific 4D donut shape combined with a sphere) have a "DNA" where this total twist is not zero.
4. The Conclusion: Therefore, those specific shapes cannot have a fabric that is perfectly Even or Odd. They cannot host Purely Electric or Purely Magnetic gravity.

Why Should You Care?

This isn't just abstract math; it helps physicists understand the limits of the universe.

1. Filtering Solutions: Physicists spend years finding exact solutions to Einstein's equations (equations that describe how gravity works). This paper gives them a quick "checklist." If a proposed solution has the wrong "DNA" (non-zero Pontryagin classes), they can throw it out immediately because it's mathematically impossible for that shape to exist in our universe.
2. New Types of Gravity: It tells us that the universe is more restricted than we thought. We can't just assume that any shape of spacetime can support these "pure" gravity states.
3. Time Slices: The paper also applies this to "time slices" (moments in time). It suggests that if you try to slice a universe into layers where every layer is perfectly round and smooth (umbilic hypersurfaces), the universe's shape might forbid it.

The Bottom Line

Think of the universe as a puzzle. You have a box of pieces (shapes) and a set of rules (gravity). This paper found a new rule: "If the puzzle piece has a certain twist in its shape, you simply cannot fit the 'Purely Electric' or 'Purely Magnetic' gravity piece into it."

It's a beautiful example of how the shape of space itself dictates what kind of physics can happen inside it.

Technical Summary

1. Problem Statement

The paper addresses a fundamental question in general relativity and differential geometry: Do all manifolds that admit Lorentzian (or pseudo-Riemannian) metrics also admit metrics with a purely electric (PE) or purely magnetic (PM) Weyl curvature tensor?

• Context: In 4-dimensional Lorentzian geometry, the Weyl curvature tensor can be split into "electric" ($E$) and "magnetic" ($B$) parts relative to a timelike unit vector field. A spacetime is Purely Electric (PE) if $B=0$ and Purely Magnetic (PM) if $E=0$.
• Motivation: PE spacetimes are common (e.g., static spacetimes, irrotational perfect fluids), whereas PM spacetimes are rare and often inconsistent with the Einstein equations (e.g., no PM vacuum solutions exist).
• Gap: While specific examples of PE/PM spacetimes exist, there is no general topological obstruction known for the existence of such metrics on arbitrary manifolds, particularly in higher dimensions or for general Petrov types.
• Goal: To determine if the topological invariants of a manifold (specifically Pontryagin classes) impose constraints on the existence of metrics with globally PE or PM Weyl curvature tensors.

2. Methodology

The author employs a combination of algebraic curvature tensor theory, representation theory, and characteristic class theory.

A. Algebraic Framework

• Algebraic Curvature Tensors: The study is generalized from Riemannian manifolds to algebraic curvature tensors on scalar product spaces $(V, g)$.
• PE/PM as Symmetry: Following Hervik et al., the paper characterizes PE and PM tensors algebraically. A tensor is PE (or PM) with respect to a unit vector $u$ if it is even (or odd) under the action of the reflection $\theta_u$ across the hyperplane orthogonal to $u$.
  • $\theta_u(u) = -u$ and $\theta_u|_{u^\perp} = \text{id}$.
  • $\theta_u$ is an orientation-reversing isometry ($\det(\theta_u) = -1$).
• Generalization: The definition is extended to any algebraic curvature tensor that is even or odd under an orientation-reversing isometry $\theta$ with $\det(\theta) = -1$, not just those arising from Lorentzian signatures.

B. Pontryagin Forms and Higher Curvature Operators

• Higher Curvature Operators: The paper utilizes Thorpe's higher curvature operators ($\hat{C}^{*k}$), which act on the exterior powers $\wedge^{2k}V$.
• Pontryagin Forms: The $k$-th Pontryagin form $\varpi_k(C)$ is defined for an algebraic curvature tensor $C$. Crucially, the paper establishes (Theorem 2.18) that $\varpi_k(C)$ can be expressed as a specific bilinear form involving the $k$-th curvature operator:
  $$\varpi_k(C) \propto F_{2k}(\hat{C}^{*k}, \hat{C}^{*k})$$
  where $F_{2k}$ is a specific $2k$-form constructed from the inner product on exterior powers.
• Key Property: The Pontryagin form of a curvature tensor $C$ is identical to that of its Weyl component $W_C$ (Theorem 2.19). This allows the obstruction to be derived solely from the Weyl tensor properties.

C. The Vanishing Argument

The core proof relies on the behavior of differential forms under orientation-reversing isometries:

1. If a tensor $C$ is $\theta$-even or $\theta$-odd, the induced map on the exterior algebra fixes the Pontryagin forms (Lemma 3.8).
2. Specifically, $\wedge^{4k}\theta^*(\varpi_\alpha(C)) = \varpi_\alpha(C)$.
3. However, if $\dim(V) = 4k$ and $\det(\theta) = -1$, the action of $\theta$ on the top-degree exterior power $\wedge^{4k}V^*$ is multiplication by $-1$.
4. Therefore, $\varpi_\alpha(C) = -\varpi_\alpha(C)$, implying $\varpi_\alpha(C) = 0$.

3. Key Contributions and Results

Theorem 1.1 (General Vanishing Theorem)

For a $4k$-dimensional pseudo-Riemannian manifold $(M, g)$ where the Weyl curvature tensor $W$ is PE or PM at every point:

• All products of Pontryagin forms $\varpi_\alpha(W) = \varpi_{\alpha_1} \wedge \dots \wedge \varpi_{\alpha_l}$ vanish identically on $M$ for any multi-index $\alpha$ such that $\sum \alpha_i = k$.
• Consequently, the corresponding product of Pontryagin classes in de Rham cohomology vanishes:
  $$p_\alpha(M) = p_{\alpha_1}(M) \smile \dots \smile p_{\alpha_l}(M) = 0 \in H^{4k}_{dR}(M).$$

Theorem 1.2 (Petrov Type Obstruction)

Applied to 4-dimensional Lorentzian manifolds ($k=1$):

• If the Weyl tensor has Petrov Type I with real or purely imaginary eigenvalues, the first Pontryagin class $p_1(M)$ must vanish.
• This extends previous results (Avez, Zund, Porter & Thompson) which covered Types O, D, II, N, and III. It specifically adds Type I with specific eigenvalue constraints to the list of types forcing $p_1(M)=0$.

Theorem 1.3 (Foliations by Umbilic Hypersurfaces)

• If a pseudo-Riemannian manifold is foliated by nondegenerate umbilic hypersurfaces, the Weyl tensor is even with respect to the reflection across the hypersurface normal.
• Therefore, if such a foliation exists, all products of Pontryagin classes of degree $4k$ must vanish.

Counter-Examples and Optimality

• Example 3.16: The author constructs a compact, orientable 4-manifold $M = T^4 \# T^4 \# \mathbb{CP}^2 \# \mathbb{CP}^2$.
  • $\chi(M) = 0$ (admits Lorentzian metrics).
  • $\sigma(M) = 2 \implies p_1(M) \neq 0$.
  • Conclusion: $M$ admits Lorentzian metrics, but none of them can have a globally PE or PM Weyl tensor (nor any of the restricted Petrov types).
• Optimality (Example 3.13): The theorem is shown to be sharp. For $k \ge 2$, one can construct manifolds where lower-degree Pontryagin products are non-zero, but the top-degree product vanishes due to the PE/PM condition.

4. Significance

1. Topological Obstructions: This work provides the first general cohomological obstructions to the existence of PE/PM metrics. It shifts the question from local geometric analysis to global topological constraints.
2. Unification of Petrov Types: It unifies various known results regarding the vanishing of Pontryagin classes for specific Petrov types under a single algebraic framework (symmetry under reflection).
3. Application to Exact Solutions: Since many exact solutions to the Einstein equations (e.g., static spacetimes, certain fluid models) possess PE Weyl tensors, these results can be used to rule out the existence of such solutions on manifolds with non-vanishing top Pontryagin classes.
4. Foliations and Cosmology: The result regarding umbilic hypersurfaces is significant for cosmology, where spacelike hypersurfaces (time slices) are often assumed to be umbilic. The paper shows that such a foliation structure is topologically forbidden on manifolds with non-trivial top Pontryagin classes.
5. Algebraic Insight: The proof demonstrates that the vanishing of Pontryagin forms is a direct consequence of the algebraic symmetries of the curvature tensor under orientation-reversing isometries, independent of the specific field equations (Einstein equations), making the result robust across different physical theories.

In summary, de Kok establishes that the existence of globally PE or PM Weyl curvature tensors is a strong topological constraint, forcing the vanishing of specific products of Pontryagin classes, thereby preventing such metrics from existing on a wide class of compact manifolds.