Twists, Codazzi Tensors, and the $6$-sphere

This paper investigates $\psi$-twisted almost Hermitian structures, introducing a class of automorphisms called $g$-Codazzi maps due to their relationship with Codazzi tensors, and applies these findings to prove a non-integrability result for the standard nearly Kähler structure on the 6-sphere.

The Gist

Imagine you have a perfectly shaped, six-dimensional ball (the 6-sphere). On the surface of this ball, there is a special set of rules for how to rotate things and measure distances. Mathematicians call this an "almost Hermitian structure." It's like a complex dance floor with specific steps (geometry) and a specific way to turn (complex structure).

For a long time, mathematicians have wondered: Can we twist these rules around to make the dance floor perfectly "integrable"? In math terms, "integrable" means the rules are consistent everywhere, allowing for a smooth, perfect geometry (like a Calabi-Yau manifold). The standard rules on the 6-sphere are "nearly" perfect but have a tiny glitch that prevents them from being truly integrable.

This paper, written by David N. Pham, explores a specific way to "twist" these rules and asks: Does this twist fix the glitch, or does it make it worse?

The "Twist" Mechanism

Think of the 6-sphere as a giant, flexible rubber sheet.

• The Standard View: Usually, to change the shape of the sheet, you might stretch it or deform it (like squishing a balloon). This is hard to analyze because the stretching changes everything at once.
• The Author's View: Instead of stretching, imagine you have a magical pair of glasses (an automorphism, or "twist") that you put on the sheet. This glasses doesn't stretch the fabric; it just changes how you see and measure the fabric. It re-labels every point and direction.
  • You look at a distance $d$, but through the glasses, it looks like $d'$.
  • You look at a rotation $J$, but through the glasses, it looks like $J'$.

The author asks: If we put on these glasses, does the new view ($J'$) become a perfect, integrable structure?

The "Codazzi" Glasses

The author realizes that not all glasses are created equal. Some glasses are messy and hard to calculate with. He focuses on a special, very neat type of glasses called g-Codazzi maps.

The Analogy:
Imagine you are walking on a hill.

• A normal twist is like walking while wearing a helmet that makes you stumble randomly.
• A g-Codazzi twist is like walking with a perfectly balanced, symmetrical helmet. If you tilt your head left, the world tilts left in a perfectly predictable way. Mathematically, this means the "curvature" of the twist behaves very nicely, similar to how a perfectly balanced scale works.

The author proves a crucial fact about these special glasses on a sphere: They must be symmetrical. You can't have a "lopsided" Codazzi twist on a sphere; it has to be perfectly balanced.

The Big Discovery: The 6-Sphere is Stubborn

The main goal of the paper is to test the 6-Sphere Conjecture. This conjecture suggests that no matter how you twist the standard structure on a 6-sphere, you can never make it perfectly integrable. It's like trying to turn a square peg into a round hole; the geometry just won't cooperate.

The author takes the "g-Codazzi" glasses (the symmetrical, well-behaved ones) and tries to twist the 6-sphere.

1. He calculates what happens to the geometry.
2. He uses a clever trick involving the "kernel" (the part of the structure that doesn't move) and the "image" (the part that does move).
3. He finds that if the twisted structure were to become integrable, it would force the glasses to do something impossible: it would force the twist to cancel itself out completely, turning the new structure back into the old, broken one.

The Result:
The author proves that for this specific, well-behaved class of twists (g-Codazzi maps), the 6-sphere remains stubborn. You cannot twist it into a perfect, integrable structure. The "glitch" is permanent, at least for this type of twist.

Why This Matters

Before this paper, there was a partial answer from other mathematicians (Bor and Hernández-Lamoneda) that said, "If the twist is really strong and balanced in a specific way, then it won't work." But their method had a blind spot; it couldn't handle twists that were "weird" or didn't meet their strict strength requirements.

The author's new method is like a universal key. It doesn't care how strong or weak the twist is, as long as it's a "Codazzi" twist. It proves that none of them work.

The Takeaway

Think of the 6-sphere as a locked box.

• Old Key: Only worked on very specific, strong locks.
• New Key (This Paper): Works on any lock that has a specific symmetrical shape (g-Codazzi).

The author has successfully proven that for this entire category of symmetrical locks, the box cannot be opened to reveal a "perfect" geometry. The 6-sphere resists being made perfect, even when you try to twist it with the most well-behaved tools available.

This is a major step forward. While it doesn't solve the problem for every possible twist (the "general" case), it solves it for a huge, important class of them, bringing us closer to understanding the deep, stubborn nature of the 6-sphere.

Technical Summary

Here is a detailed technical summary of the paper "Twists, Codazzi Tensors, and the 6-Sphere" by David N. Pham.

1. Problem Statement and Motivation

The paper investigates the deformation of almost Hermitian structures $(M, g, J, \omega)$ via automorphisms of the tangent bundle, $\psi \in \text{Aut}(TM)$. Unlike standard deformations (which vary the structure tensors continuously), this approach "twists" the existing structure by conjugating with an automorphism $\psi$.

The central problem is to understand how such twists affect:

1. Integrability: Does the twisted almost complex structure $J_\psi = \psi \circ J \circ \psi^{-1}$ become integrable (i.e., is the Nijenhuis tensor $N_{J_\psi}$ zero)?
2. Riemannian Geometry: How do the metric $g_\psi$ and its curvature properties change?

Motivation: The author provides a motivating example on $S^3 \times S^3$ (equipped with a strictly nearly Kähler structure) where a specific non-integrable $J$ is twisted by an automorphism $\psi$ into an integrable complex structure $J_\psi$, which also yields an SKT (Strong Kähler with Torsion) metric. This demonstrates that twists can fundamentally alter the geometric nature of a manifold.

This leads to Conjecture 1.2: Let $J$ be the standard almost complex structure on the 6-sphere $S^6$ (induced by the $G_2$ cross product). Then, for all $\psi \in \text{Aut}(TS^6)$, the twisted structure $J_\psi$ is non-integrable.

2. Methodology and Key Definitions

2.1. The Twist Construction

Given an almost Hermitian manifold $(M, g, J, \omega)$ and $\psi \in \text{Aut}(TM)$, the twisted structure is defined as:

• $g_\psi(X, Y) = g(\psi^{-1}X, \psi^{-1}Y)$
• $J_\psi = \psi \circ J \circ \psi^{-1}$
• $\omega_\psi(X, Y) = \omega(\psi^{-1}X, \psi^{-1}Y)$

2.2. The $g$-Codazzi Map

The core methodological innovation is the restriction of the study to a specific class of automorphisms called $g$-Codazzi maps.

• Definition: An automorphism $\psi$ is a $g$-Codazzi map if its inverse $\psi^{-1}$ satisfies the Codazzi condition with respect to the Levi-Civita connection $\nabla^g$:
  $$(\nabla^g_X \psi^{-1})Y = (\nabla^g_Y \psi^{-1})X$$
• Significance: This condition drastically simplifies the transformation law for the Levi-Civita connection of the twisted metric $g_\psi$. If $\psi$ is a $g$-Codazzi map, the new connection is simply:
  $$\nabla^{g_\psi}_X Y = \psi \nabla^g_X (\psi^{-1}Y)$$
  This linearity allows for explicit calculations of curvature and integrability conditions that are otherwise intractable for general automorphisms.

2.3. Relationship to Codazzi Tensors

The paper establishes a correspondence (Proposition 2.3) between non-degenerate symmetric Codazzi tensors $A$ and self-adjoint $g$-Codazzi maps $\psi$ via $A(X, Y) = g(\psi^{-1}X, Y)$. On the round sphere $S^n$, Codazzi tensors are well-understood (related to Hessians of functions), providing a rich supply of test cases.

3. Key Contributions and Results

3.1. Riemannian Geometry of Twisted Metrics (Section 2)

• Curvature Transformation: The author derives a formula relating the curvature operator $R_{g_\psi}$ of the twisted metric to the original curvature $R_g$. For a $g$-Codazzi map $\psi$:
  $$R_{g_\psi} = R_g \circ \psi^*$$
  where $\psi^*$ is the pullback on 2-forms.
• Flatness Preservation: A direct consequence is that $g$ is flat if and only if $g_\psi$ is flat.
• Kähler Preservation: If the original manifold is Kähler ($\nabla^g J = 0$), the twisted manifold remains Kähler.

3.2. Geometry of the Round Sphere (Section 3)

The paper focuses on the round sphere $(S^n, g)$ for $n \geq 3$.

• Self-Adjointness Theorem (Theorem 3.5): A crucial technical result is proven: Every $g$-Codazzi map on the round sphere $S^n$ ($n \geq 3$) must be self-adjoint with respect to the metric $g$.
  • Proof Strategy: The author uses the fact that the curvature operator on $S^n$ is the identity. By analyzing the induced action on 2-forms and utilizing linear algebra lemmas regarding self-adjointness of pullbacks, it is shown that $\psi$ cannot be skew-adjoint (which would imply the existence of non-trivial parallel 2-forms, impossible on $S^n$).
• Implication: This reduces the study of twists on $S^6$ to the class of self-adjoint automorphisms, significantly constraining the possible forms of $\psi$.

3.3. Integrability on the 6-Sphere (Section 4)

The main result addresses Conjecture 1.2 for the specific class of $g$-Codazzi maps.

• Integrability Condition (Proposition 4.6): For a nearly Kähler manifold and a self-adjoint $g$-Codazzi map $\psi$, $J_\psi$ is integrable if and only if the operator $K = J_\psi + \psi J$ commutes with $\nabla^g_Z J$ for all vector fields $Z$.
• Main Theorem (Theorem 4.7): Let $(S^6, g, J, \omega)$ be the standard nearly Kähler structure. If $\psi$ is any $g$-Codazzi map, then $J_\psi$ is non-integrable.
  • Proof Sketch:
    1. Assume $J_\psi$ is integrable.
    2. By Theorem 3.5, $\psi$ is self-adjoint.
    3. Using the non-degeneracy of the Nijenhuis tensor of the standard $J$ on $S^6$ (linked to the $G_2$ structure), the commutation condition forces the operator $K$ to be a scalar multiple of $J$ (i.e., $K = aJ$).
    4. This implies $\psi J = -J \psi$ (anti-commutation).
    5. However, if $\psi$ anti-commutes with $J$, then $J_\psi = \psi J \psi^{-1} = -J$. Since the standard $J$ on $S^6$ is known to be non-integrable, $-J$ is also non-integrable, leading to a contradiction.
    6. Therefore, the assumption that $J_\psi$ is integrable must be false.

3.4. Comparison with Existing Literature

The paper compares its result with a theorem by Bor and Hernández-Lamoneda [7], which provides sufficient conditions for non-integrability based on the eigenvalues of the curvature operator ($\lambda_{\max}/\lambda_{\min} < 7/5$).

• Limitation of [7]: The author constructs examples of $g$-Codazzi maps where the eigenvalue ratio condition fails, rendering the [7] result inapplicable.
• Advantage of Theorem 4.7: The new theorem proves non-integrability for all $g$-Codazzi maps (provided they are invertible), regardless of the curvature eigenvalue ratios, thus providing a more robust and complete solution for this class.

4. Significance and Conclusion

• Resolution of a Specific Case: The paper provides a definitive proof that the standard almost complex structure on $S^6$ cannot be made integrable by twisting with any $g$-Codazzi map. This is a significant step toward resolving the long-standing open problem of whether $S^6$ admits any integrable complex structure (Conjecture 1.2).
• Methodological Framework: The introduction of $g$-Codazzi maps as a tractable class of automorphisms offers a new tool for studying deformations of geometric structures. The derived formulas for connection and curvature transformations are general and applicable beyond the sphere.
• Geometric Insight: The work highlights the rigidity of the nearly Kähler structure on $S^6$. Even when the metric is twisted in a way that preserves the "Codazzi" property (a strong geometric constraint), the underlying non-integrability of the complex structure persists.
• Future Directions: The author notes that while this solves the problem for $g$-Codazzi maps, the general conjecture for arbitrary automorphisms remains open. The techniques developed here serve as a foundation for tackling more general classes of twists.

In summary, David N. Pham's paper establishes a rigorous framework for analyzing "twisted" almost Hermitian structures, proves that the standard $S^6$ structure is robust against a wide class of such twists (specifically $g$-Codazzi maps), and demonstrates that the non-integrability of $S^6$ is a deep geometric feature resistant to these specific types of transformations.