On the Jordan-Chevalley decomposition problem for operator fields in small dimensions and Tempesta-Tondo conjecture

This paper establishes tensorial conditions for operator fields in dimensions three and four to admit local coordinates where they take a strictly upper triangular form, and proves the Tempesta-Tondo conjecture regarding higher-order Frölicher-Nijenhuis brackets.

The Gist

Imagine you are a detective trying to solve a mystery about a shape-shifting object called an Operator Field (let's call it "L").

In the world of mathematics, this object "L" acts like a machine that stretches, squashes, or rotates space at every single point. Sometimes, this machine is messy and chaotic. But often, mathematicians want to know: "Can we find a special map (a coordinate system) where this machine looks simple and neat?"

Specifically, they want to know if we can find a map where the machine looks like a strictly upper-triangular matrix.

• The Analogy: Imagine a pyramid of blocks. If you look at it from the side, the blocks are stacked neatly, with empty space below the diagonal. If the machine looks like this, it means the system is "integrable"—it follows a predictable, step-by-step logic that we can understand and solve.

The paper by Bolsinov, Konyaev, and Matveev is about figuring out when this neat, pyramid-like structure is possible, and how to prove it without having to actually build the map.

Here is the breakdown of their discovery in everyday terms:

1. The "Smoothness" Test (The Torsion)

To see if the machine can be simplified, mathematicians use a tool called Torsion. Think of torsion as a "friction test."

• If the machine is perfectly smooth and orderly, the friction (torsion) is zero.
• If the friction is non-zero, the machine is twisted and messy, and you can't easily simplify it.

For a long time, mathematicians knew a rule (the Haantjes Criterion): If the machine is "simple" (diagonal) and the friction is zero, you can find your neat map.

The Problem: What if the machine isn't diagonal, but it's a "nilpotent" block? (Imagine a machine that eventually grinds to a halt after a few steps, like a gear that slips and stops).

• In 2D and 3D, the old friction test (Haantjes torsion) worked perfectly. If the friction was zero, the machine could be simplified.
• In 4D and higher, the old test failed. The authors found a tricky machine that had zero friction but was still too messy to be simplified. It was a "false positive."

2. The New Detective Tool (Theorem 2)

Since the old test failed in 4D, the authors invented a new, more sensitive detector (Theorem 2).

• They created a new tensor (a complex mathematical object) called T.
• Think of T as a "super-friction" meter. It looks at the machine's behavior in a much more detailed way.
• The Discovery: In 4D, if this new meter T reads zero, then you guaranteed can find the neat, pyramid-like map. If it reads anything else, you can't.

They also showed you how to build this new meter for any size of machine, not just 4D. It's like giving future detectives a blueprint to build their own super-sensors.

3. The "Tempesta–Tondo" Mystery (The Conjecture)

The second half of the paper solves a riddle about two machines, L and M, working together.

• Imagine two gears turning in perfect sync (they "commute").
• If both gears are already in that neat, pyramid shape (strictly upper-triangular), a famous conjecture asked: "Does their combined 'super-friction' vanish?"
• The authors proved YES.
• The Metaphor: If you have two perfectly organized, step-by-step processes, and you mix them together in a specific way, the resulting chaos is zero. The system remains perfectly orderly. This confirms a guess made by other mathematicians (Tempesta and Tondo) and closes a chapter on how these complex systems behave.

Summary of the "Why it Matters"

Why do we care about these abstract machines?

• Real World: These "machines" describe how fluids flow, how waves move, and how energy separates in physics.
• The Goal: If we can prove a system is "triangular" (simple), we can solve the equations that describe it. If it's messy, we can't.
• The Contribution: This paper gives us the exact checklist (Theorems 1, 2, and 3) to know when a system is solvable in 3D and 4D, and it fixes the broken tools we were using before.

In a nutshell: The authors found a way to tell if a complex, twisting mathematical machine can be straightened out into a simple, predictable shape. They fixed a broken test for 4D machines and proved that two synchronized, simple machines stay simple when mixed together.

Technical Summary

1. Problem Statement

The paper addresses the Jordan–Chevalley decomposition problem for operator fields (tensor fields of type $(1,1)$) on smooth manifolds. Specifically, it investigates the conditions under which an operator field $L$, which is pointwise similar to a single nilpotent Jordan block (or a scalar multiple of the identity plus such a block), can be transformed into a strictly upper triangular form (or upper triangular with constant diagonal) via a local coordinate change.

The core question is: Does the vanishing of specific torsion tensors (Nijenhuis or Haantjes) provide a necessary and sufficient condition for the existence of such "good" coordinates?

• Context: The Haantjes criterion states that for a semi-simple operator with real spectrum, the vanishing of the Haantjes torsion $H_L$ is necessary and sufficient for diagonalizability.
• Gap: For nilpotent operators (similar to a Jordan block), the situation is more complex. While it is known that if $L$ is strictly upper triangular, then the $(n-1)$-th order Haantjes torsion $T^{(n-1)}_L$ vanishes, the converse was an open question.
• Counter-example: The authors note that for dimensions $n \geq 4$, the vanishing of $T^{(n-1)}_L$ is not sufficient to guarantee triangularizability (Example 1.1).

2. Methodology

The authors employ a combination of differential geometry, linear algebra, and computer-assisted symbolic computation.

• Linearization Approach: To analyze the local behavior near a point $p$, the authors linearize the operator field $L(x)$. They assume $L(x)$ is similar to a Jordan block $J_n(\lambda(x))$ and express $L(x)$ via a similarity transformation $A(x)$:
  $$L(x) = A^{-1}(x) J_n(\lambda(x)) A(x)$$
  They expand $A(x)$ and $\lambda(x)$ to first order around $p=0$.
• Integrability Conditions: The existence of a coordinate system where $L$ is upper triangular is equivalent to the integrability of a specific flag of distributions (kernels of powers of $L - \lambda I$). This integrability translates into a system of linear equations on the coefficients of the linearization of $A(x)$.
• Tensor Construction: The authors construct candidate tensor fields $T$ (polynomials in the components of $L$ and linear in their first derivatives) and evaluate them on the linearized operator.
• Algebraic Equivalence: They compare the system of equations derived from the vanishing of the candidate tensor against the system of equations derived from the integrability of the distributions. If these two systems are algebraically equivalent, the tensor provides the necessary and sufficient condition.
• Dimensional Analysis: The proof is explicitly carried out for dimensions $n=3$ and $n=4$, where the systems are small enough to be solved by hand or simple computer algebra.

3. Key Contributions and Results

A. Dimension 3: The Haantjes Torsion is Sufficient

Theorem 1: For a 3-dimensional operator field $L$ with a single eigenvalue of geometric multiplicity 1, the following are equivalent:

1. $L$ can be locally transformed into an upper triangular form.
2. The classical Haantjes torsion $H_L$ (level 2) vanishes.

• Significance: This resolves the problem positively for $n=3$, showing that the classical Haantjes torsion is the correct obstruction.

B. Dimension 4: A New Tensor Condition

Theorem 2: For a 4-dimensional operator field $L$ with a single eigenvalue of geometric multiplicity 1, the classical Haantjes torsion $H_L$ is not sufficient (as shown in Example 1.4).
Instead, the authors construct a new tensor field $T$ defined by:
$$T^i_{jk} = \tilde{L}^i_s H^s_{rk} \tilde{L}^r_j - \tilde{L}^i_s H^s_{jr} \tilde{L}^r_k + H^i_{sk} \tilde{L}^s_r \tilde{L}^r_j$$
where $\tilde{L} = L - \frac{1}{4}(\text{tr } L)I$.
Result: $L$ is locally upper triangularizable if and only if this tensor $T$ vanishes.

• Significance: This provides the first explicit tensorial condition for triangularizability in dimension 4, correcting the insufficiency of the standard Haantjes torsion.

C. The Tempesta–Tondo Conjecture

Theorem 3: The authors prove a conjecture regarding higher-order Frölicher-Nijenhuis brackets.

• Statement: Let $L$ and $M$ be two commuting operators that are both given by strictly upper triangular matrices in some local coordinates. Then, the generalized Haantjes bracket of level $(n-1)$, denoted $H^{(n-1)}_{L,M}$, vanishes identically.
• Proof Strategy: The proof relies on the algebraic properties of strictly upper triangular matrices. Since $L$ and $M$ lower the index of basis vectors, any term in the bracket expansion involving powers summing to $\geq n$ vanishes. The authors show that for the specific level $n-1$, all non-vanishing terms in the expansion are forced to zero due to the skew-symmetry and the specific index constraints of the basis vectors.

4. Technical Details and Algorithm

The paper outlines a general algorithm (Section 2.1) to find such tensors in higher dimensions:

1. Construct a basis of tensor fields of type $(1,2)$ algebraically from $L$ and its Nijenhuis torsion (e.g., $N, LN, NL, LNL, \dots$).
2. Form a linear combination with unknown coefficients.
3. Substitute the linearized operator form (from the Jordan block assumption) into this combination.
4. Solve the resulting linear system to find coefficients that make the tensor vanish whenever the integrability conditions are met.
5. Verify that the vanishing of this specific tensor is equivalent to the integrability conditions.

5. Significance and Implications

• Integrable Systems: The results are crucial for the theory of integrable systems, particularly in the context of hydrodynamic-type evolutionary PDEs and separation of variables. The existence of "good" coordinates (triangular or diagonal) is often a prerequisite for solving these systems.
• Generalization of Haantjes Criterion: The work extends the famous Haantjes criterion beyond semi-simple operators to the nilpotent case, clarifying exactly when the classical torsion fails and how to fix it with higher-order or modified tensors.
• Computational Geometry: The paper demonstrates how computer algebra can be used to solve geometric problems involving differential invariants, providing a template for analyzing operator fields in dimensions $n > 4$.
• Resolution of Open Problems: It settles the Tempesta–Tondo conjecture and provides a definitive answer to the triangularizability problem for $n=3$ and $n=4$.

In summary, the paper establishes that while the vanishing of the standard Haantjes torsion is sufficient for triangularizability in dimension 3, dimension 4 requires a more complex tensorial condition. Furthermore, it confirms that high-order brackets of commuting triangular operators vanish, solidifying the algebraic structure of Haantjes algebras.