Three results on holonomic D-modules

This paper advances the theory of holonomic D-modules by establishing the invariance of Euler characteristics under localization and regular singular twists, proving local generic vanishing theorems via twisting with differential forms, and providing a direct construction of the Laplace transform for Stokes-filtered sheaves that completes the correspondence with holonomic D-modules.

The Gist

Imagine you are an architect trying to understand the shape and stability of a complex, multi-dimensional building. This building isn't made of bricks, but of mathematical equations that describe how things change and flow. In the world of advanced mathematics, these "buildings" are called D-modules.

Claude Sabbah's paper is like a guidebook for three specific tools that help architects (mathematicians) understand these buildings, especially when the structures get messy, irregular, or have "holes" in them.

Here is a breakdown of the three main results in the paper, translated into everyday language with some creative analogies.

---

Part 1: The "Euler Characteristic" (Counting the Rooms)

The Concept:
Imagine you have a building (a mathematical object) and you want to know its "Euler characteristic." In simple terms, this is a single number that summarizes the building's shape (like counting rooms minus hallways plus corners).

The Problem:
Sometimes, you need to look at the building from different angles. Maybe you zoom in on a specific wall (localization), or maybe you look at the building after adding a new, slightly different type of paint (tensoring with a connection). The worry is: Does the shape summary (the number) change just because we zoomed in or changed the paint?

The Solution:
Sabbah proves that the number stays the same.

• The Analogy: Imagine you have a sculpture made of clay. If you take a magnifying glass to a specific crack in the clay, or if you dip the whole sculpture in a clear, regular varnish, the "total amount of clay" (the Euler characteristic) doesn't magically change.
• Why it matters: This gives mathematicians confidence. They can break a complex problem into smaller, local pieces or tweak the equations slightly, knowing that the fundamental "shape" of the answer remains stable. It's like knowing that no matter how you slice a cake, the total volume of the cake is constant.

---

Part 2: The "Vanishing Theorem" (The Magic Disappearing Act)

The Concept:
Sometimes, when you push a mathematical object through a doorway (a process called "pushforward"), it gets messy or loses information. It's like trying to push a tangled ball of yarn through a small hole; it might get stuck or break.

The Problem:
Mathematicians want to know: When does the yarn pass through perfectly, without getting tangled or losing a strand? Usually, this only happens under very specific, rare conditions.

The Solution:
Sabbah shows that if you twist the yarn with a specific kind of "magic string" (a closed differential form) before pushing it through, the yarn passes through perfectly every time, as long as you pick the right twist.

• The Analogy: Imagine trying to push a stiff, awkwardly shaped box through a narrow tunnel. It gets stuck. But, if you first wrap the box in a specific, flexible rubber band (the "twist"), the box suddenly becomes flexible enough to slide through the tunnel smoothly.
• Why it matters: This is a "generic vanishing theorem." It tells us that for almost any random twist we choose, the messy math problem suddenly becomes clean and solvable. It turns a rare miracle into a common occurrence.

---

Part 3: The "Laplace Transform" (The Translator Machine)

The Concept:
The Laplace transform is a famous mathematical tool that switches between two different languages.

• Language A: Describes things in terms of time (how things change moment by moment).
• Language B: Describes things in terms of frequency (the "vibrations" or patterns inside the change).

The Problem:
In the world of "irregular" math (where things behave wildly and unpredictably), there are two ways to describe these wild behaviors:

1. The D-Module way: Using complex differential equations.
2. The Stokes way: Using geometric shapes and "filters" that sort data based on how fast it grows or shrinks (like sorting sand by grain size).

For a long time, mathematicians knew these two languages were related, but the translation manual was incomplete. They knew how to go from Equations to Shapes, but the reverse was fuzzy.

The Solution:
Sabbah completes the manual. He builds a precise, two-way translator between the "Equation World" and the "Shape World" for these wild, irregular systems.

• The Analogy: Imagine you have a chaotic storm.
  • Method A (Equations): You write down the wind speed, pressure, and temperature at every second.
  • Method B (Shapes): You draw a map of the storm's "Stokes layers"—invisible layers where the wind behaves differently.
  • Sabbah's work is like building a perfect Rosetta Stone that lets you translate the wind speed data directly into the map layers, and vice versa, without losing any information.
• Why it matters: This connects two huge areas of math (Analysis and Topology). It allows mathematicians to use the tools of geometry to solve problems in calculus, and vice versa. It's the final piece of a puzzle that connects how we calculate things to how we visualize them.

---

Summary: What is the Big Picture?

This paper is about stability, flexibility, and translation in the high-stakes world of advanced mathematics.

1. Stability: No matter how you zoom in or tweak the rules, the core "shape" of the problem remains the same.
2. Flexibility: If you apply the right "twist" to a messy problem, it suddenly becomes solvable.
3. Translation: We now have a complete dictionary to translate between the language of equations and the language of geometric shapes, even for the most chaotic, irregular systems.

Sabbah is essentially saying: "Don't be afraid of the messy, irregular parts of math. We have the tools to measure them, fix them, and translate them into something we can understand."

Technical Summary

Here is a detailed technical summary of the paper "Three Results on Holonomic D-Modules" by Claude Sabbah.

1. Overview and Problem Statement

The paper addresses three distinct but interconnected problems within the theory of holonomic $\mathcal{D}$-modules, specifically focusing on those with irregular singularities. The work leverages recent foundational results by K. Kedlaya and T. Mochizuki regarding the existence of Stokes structures in dimensions $\ge 2$.

The primary goals are:

1. To establish the invariance of Euler characteristics for de Rham complexes under localization and tensoring with rank-one connections.
2. To prove "local generic vanishing theorems" for algebraic holonomic $\mathcal{D}$-modules, extending results from Gabriel Ribeiro's thesis.
3. To provide a direct topological construction of the Laplace transform for Stokes-filtered constructible sheaves and verify its compatibility with the algebraic Laplace transform via the Riemann-Hilbert correspondence, complementing previous work by Yu and Zhang.

---

2. Methodology

Sabbah employs a blend of transcendental (analytic) methods and algebraic geometry, heavily relying on:

• Stokes Structures: Utilizing the decomposition of meromorphic flat bundles into formal elementary modules (Kedlaya-Mochizuki theorem) and the associated Stokes filtrations.
• Irregularity Complexes: Using Mebkhout's notion of the irregularity complex to compute local Euler characteristics.
• Real Blow-ups: Analyzing the behavior of sheaves and connections on the real blow-up of manifolds along divisors to handle asymptotic behavior and Stokes phenomena.
• Topological Laplace Transforms: Constructing functors between categories of constructible sheaves and Stokes-filtered local systems using pushforwards and pullbacks on product spaces involving real blow-ups.
• Newton Polyhedra: In the proof of generic vanishing, the author uses Newton polyhedra to analyze the poles of differential forms and ensure the "good wild" condition is preserved under proper modifications.

---

3. Key Contributions and Results

Part I: Euler Characteristic of the de Rham Complex

• Problem: Determine how the global and local Euler-Poincaré characteristics of the de Rham complex behave under localization ($M(\ast D)$), co-localization ($M(!D)$), and tensoring with a rank-one meromorphic connection $N$ with regular singularities.
• Main Result (Theorems 1.1 & 1.3):
  • The global Euler characteristic $\chi_{dR}(X, M)$ is invariant under these operations.
  • Locally, for any point $x$, the local Euler characteristic $\chi_{dR, x}(M)$ is the same for $M(!D)$, $M(\ast D)$, and $M \otimes N$.
• Methodology:
  • The proof reduces the problem to the irregularity number ($\text{irr}_x$).
  • Using the Kedlaya-Mochizuki normal form, the author reduces the problem to "good" meromorphic flat bundles.
  • By omitting the Stokes filtration (justified by a theorem in [Sab17]), the computation reduces to a topological calculation on the real blow-up, showing that the irregularity numbers of $M$, its dual $M^\vee$, and $M \otimes N$ coincide.
• Significance: This provides a robust analytic proof for a conjecture in the algebraic setting (via GAGA), confirming that twisting by regular singular connections does not alter the topological invariants of the de Rham complex.

Part II: Local Generic Vanishing Theorems

• Problem: Prove that for an algebraic holonomic $\mathcal{D}$-module $M_U$ on a smooth variety $U$, the natural morphism from the proper pushforward to the total pushforward ($Dj_! M \to Dj_* M$) becomes an isomorphism after twisting by a generic closed algebraic differential form.
• Main Result (Theorem 4.6):
  • Let $\eta_1, \dots, \eta_r$ be closed one-forms on $U$. If they satisfy specific conditions (either logarithmic with non-resonant residues or "good wild" with high-order poles), there exists a Zariski-dense open set $V \subset \mathbb{C}^r$ such that for any $a \in V$, the morphism $Dj_! (M_U \otimes N_{\eta, a}) \to Dj_* (M_U \otimes N_{\eta, a})$ is an isomorphism.
• Methodology:
  • The proof uses induction on the dimension of the support of the module.
  • It relies on Proposition 5.1, a criterion for local vanishing based on the eigenvalues of the transverse monodromy of the formal regular part.
  • The author constructs a resolution of singularities and analyzes the pullback of the forms to ensure the "good wild" or "non-resonant" conditions are preserved.
  • In the "good wild" case, Newton polyhedra are used to show that for generic parameters, the connection retains a pole of order $\ge 2$ along the divisor, satisfying the vanishing criterion.
• Significance: This generalizes previous results (e.g., for $G_m$) to higher dimensions and arbitrary tori, providing a powerful tool for studying the cohomology of twisted $\mathcal{D}$-modules.

Part III: Laplace Transform of Stokes-Filtered Sheaves

• Problem: Construct the inverse topological Laplace transform for Stokes-filtered constructible sheaves and prove its compatibility with the algebraic Laplace transform of $\mathcal{D}$-modules, completing the picture started in [Sab13] and [YZ24].
• Main Result (Theorem 7.3):
  • Establishes a commutative diagram linking:
    1. Regular holonomic $\mathcal{D}$-modules on $\mathbb{A}^1_t$ ($Mod^0_{holreg}$).
    2. Holonomic $\mathcal{D}$-modules on $\mathbb{A}^1_\tau$ with exponential type singularities ($Mod_{hol}^{exp}$).
    3. Perverse sheaves on $\mathbb{C}_t$ with vanishing cohomology ($Perv_0$).
    4. Stokes-filtered local systems on $S^1_\tau$ ($Sto$).
  • Specifically, defines a topological functor $F^{\infty, top}_-$ that is the quasi-inverse to the topological Laplace transform $F^{\infty, top}_+$, and proves it corresponds to the algebraic inverse Laplace transform via the Riemann-Hilbert-Birkhoff-Deligne-Malgrange correspondence.
• Methodology:
  • Construction: The author defines the transform using the pushforward of a subsheaf $L_{<0}$ (rapid decay sections) on the product space $S^1_\tau \times \mathbb{C}_t$.
  • Blow-up Technique: To handle the "non-goodness" of the exponential factors at infinity, the author performs a complex blow-up of $\mathbb{P}^1_\tau \times \mathbb{C}_t$ at points $(\infty, c)$. This creates a normal crossing divisor where the Stokes structure becomes "very good."
  • Hukuhara Levelt-Turrittin: Uses the Hukuhara theorem in dimension 2 to relate the moderate de Rham complex on the blown-up space to the Stokes-filtered local system.
  • Comparison: Directly compares the topological construction with the algebraic one, avoiding the indirect methods used in [YZ24] by fully utilizing Stokes structures in dimension 2.
• Significance: This provides a complete and direct topological realization of the Laplace transform for irregular connections, bridging the gap between algebraic $\mathcal{D}$-module theory and topological sheaf theory in the presence of irregular singularities.

---

4. Significance and Impact

• Unification of Methods: The paper demonstrates how modern techniques for irregular singularities (Stokes structures, good formal decompositions) can be applied to classical problems like Euler characteristics and vanishing theorems.
• Resolution of Conjectures: It provides analytic proofs for algebraic conjectures regarding the invariance of Euler characteristics under twisting, which were previously only known in specific cases or via indirect arguments.
• Topological-Algebraic Bridge: Part III is particularly significant for the Riemann-Hilbert correspondence. It explicitly constructs the topological inverse of the Laplace transform and proves its compatibility with the algebraic side, offering a more transparent alternative to the "co-Stokes" approach of Yu and Zhang.
• Generalization: The results extend known theorems from dimension 1 (curves) to higher dimensions, validating the utility of the Stokes filtration framework in higher-dimensional complex geometry.

In summary, Sabbah's paper is a rigorous technical contribution that solidifies the connection between the analytic theory of irregular connections and the topological theory of constructible sheaves, providing new tools for computing invariants and transforming objects across these categories.