On the dual positive cones and the algebraicity of a compact Kähler manifold

This paper establishes that a compact Kähler manifold whose dual Kähler cone contains a rational interior point has a projective Albanese variety, thereby resolving the Oguiso–Peternell problem for Ricci-flat manifolds and addressing related algebraicity questions for threefolds.

The Gist

Imagine you are a detective trying to solve a mystery about the hidden structure of complex geometric shapes. This paper, written by mathematician Hsueh-Yung Lin, is all about figuring out when a specific type of shape—called a Compact Kähler Manifold—is actually a "polite" shape (algebraic/projective) versus a "wild" shape (non-algebraic).

Here is the story of the paper, broken down with simple analogies.

The Main Characters: The Shapes

1. The Kähler Manifold (The Shape): Think of this as a complex, multi-dimensional doughnut or a twisted piece of fabric. It has a very specific, smooth geometry.
2. The Projective Shape (The Polite Guest): Some of these shapes are "algebraic." This means they can be drawn perfectly using polynomial equations (like $x^2 + y^2 = 1$). In math-speak, they are "projective." These shapes are well-behaved; they have a rigid structure that makes them easy to study.
3. The Non-Algebraic Shape (The Wild Guest): Other shapes look smooth but cannot be described by simple equations. They are "wild." They might twist in ways that don't fit into the standard grid of algebra.

The Mystery: The "Dual Kodaira" Clue

For a long time, mathematicians knew a rule called the Kodaira Embedding Theorem. It said: "If you find a specific type of 'positive' energy (a rational class) inside the shape's core, the shape is polite (projective)."

But the author and his colleagues were asking a reverse question (The Oguiso–Peternell Problem):
"What if we look at the dual side? What if we find a positive clue in the 'shadow' or the 'mirror image' of the shape's core? Does that also guarantee the shape is polite?"

The paper investigates this reverse clue. The authors call this the "Dual Kodaira Condition."

The Investigation: How They Solved It

The author uses a few clever strategies to prove that if this "dual clue" exists, the shape is almost certainly polite.

1. The "Shadow" Test (Dual Cones)

Imagine the shape has a "Kähler Cone." Think of this as a flashlight beam shining from the center of the shape.

• The Original Rule: If the flashlight beam hits a rational point (a grid point), the shape is polite.
• The New Rule: The author looks at the shadow cast by that flashlight (the Dual Cone). If the shadow contains a rational point in its very center, does that mean the shape is polite?
• The Finding: Yes! If the shadow has a rational point in its heart, the shape is polite.

2. The "Albanese" Mirror

Every shape has a "mirror image" called its Albanese Torus. Think of this as a reflection of the shape in a pool of water.

• The Breakthrough: The author proves that if the "dual clue" exists, this reflection (the Albanese Torus) is always a polite, projective shape.
• Why it matters: If the reflection is polite, the original shape is usually polite too, or at least very close to it.

3. The "Ricci-Flat" Special Case

Some shapes are "Ricci-flat," meaning they have a special kind of perfect balance (like a perfectly flat sheet of paper, but in higher dimensions).

• The Result: For these perfectly balanced shapes, the author proves that if the "dual clue" exists, the shape is 100% polite (projective). There is no wiggle room; it must be algebraic.

The Three-Dimensional Puzzle

The paper gets really interesting when looking at 3D shapes (Threefolds).

• The Problem: In 3D, there are some weird, hypothetical shapes called "Simple Non-Kummer" threefolds. Mathematicians aren't even sure if these things actually exist (they might be ghosts).
• The Solution: The author says, "If we ignore these ghost shapes, then every 3D shape with the 'dual clue' is polite."
• The "Connecting Family" Metaphor: To prove this, the author looks at how curves (lines) move inside the shape. He asks: "Can we connect any two random points in the shape by walking along a chain of curves?"
  • If the answer is yes, the shape is polite.
  • The paper shows that the "dual clue" forces these connecting chains to exist, effectively taming the wild shape.

The Big Picture Conclusion

The paper answers a decades-old question: Does a specific type of positive energy in the "dual" space guarantee a complex shape is algebraic?

The Answer: Yes.

• If you find this specific "dual" positive class, the shape's reflection (Albanese) is projective.
• If the shape is "Ricci-flat" (perfectly balanced), the shape itself is projective.
• If the shape is 3-dimensional (and not a weird ghost), it is projective.

Why Should You Care?

This is like finding a universal key. For a long time, mathematicians had to check many different conditions to see if a shape was "polite" (algebraic). This paper says, "Hey, if you just check this one specific 'dual' condition, you can be sure the shape is well-behaved." It simplifies the map of the mathematical universe, helping us understand which complex shapes can be described by simple equations and which ones are truly wild.

In short: The author proved that if a complex geometric shape has a specific "positive shadow," it cannot be wild; it must be a polite, algebraic shape that fits neatly into the grid of mathematics.

Technical Summary

Here is a detailed technical summary of the paper "On the dual positive cones and the algebraicity of compact Kähler manifolds" by Hsueh-Yung Lin.

1. Problem Statement and Context

The paper addresses the Oguiso–Peternell problem, which seeks a "dual" statement to the classical Kodaira Embedding Theorem.

• Kodaira Embedding Theorem: A compact Kähler manifold $X$ is projective if its Kähler cone $\mathcal{K}(X)$ contains a rational cohomology class.
• The Dual Problem: Let $\mathcal{K}(X)^\vee$ be the dual Kähler cone in $H^{n-1,n-1}(X, \mathbb{R})$ (the space of curve classes). If the interior of $\mathcal{K}(X)^\vee$ contains a rational class (a "dual Kodaira condition"), how algebraic is $X$? Specifically, is $X$ projective, or at least of high algebraic dimension?

The author investigates two specific conditions:

1. Condition (K): The interior of the dual Kähler cone $\text{Int}(\mathcal{K}(X)^\vee)$ contains a rational class.
2. Condition (P): The interior of the dual pseudoeffective cone $\text{Int}(\text{Psef}(X)^\vee)$ contains a rational class.

The central questions are:

• Does Condition (K) imply that the Albanese variety of $X$ is projective?
• Does Condition (K) or (P) imply that a compact Kähler manifold (specifically in dimension 3) is projective?

2. Methodology

The paper employs a combination of Hodge theory, birational geometry, and the study of fibrations. Key methodological components include:

• Bimeromorphic Invariance: The author establishes that the existence of rational classes in the interior of dual cones is invariant under bimeromorphic modifications (blow-ups) and dominant meromorphic maps. This allows the reduction of the problem to specific bimeromorphic models.
• Classification of Low Algebraic Dimension: Utilizing Fujiki's classification of compact Kähler threefolds with algebraic dimension $a(X) \leq 1$, the author reduces the general problem to analyzing specific geometric structures:
  • Fibrations over surfaces.
  • Fibrations in abelian varieties over curves.
  • Quotients of torus fibrations.
  • Quotients of 3-tori.
• Deligne Cohomology and Jacobian Fibrations: For torus and abelian fibrations, the author uses Deligne cohomology and the theory of Jacobian fibrations to relate the existence of rational classes in the dual cone to the existence of multi-sections.
• Fourier Transform on Tori: For complex tori, the proof utilizes the Fourier-Mukai transform and the Poincaré formula to relate curve classes to line bundles, proving that the existence of a "positive" rational curve class forces the torus to be projective.
• Analysis of 1-Cycles: The paper investigates Question 1.10, asking whether Hodge classes vanishing outside a subvariety $Y$ are images of Hodge classes from a resolution of $Y$. This is crucial for handling elliptic fibrations.

3. Key Contributions and Results

A. The Albanese Torus (General Dimension)

Theorem 1.4: If a compact Kähler manifold $X$ satisfies Condition (K) (dual Kähler cone contains a rational class), then its Albanese torus $\text{Alb}(X)$ is projective.

• Significance: This provides a lower bound for the algebraic dimension $a(X)$. If $X$ has maximal Albanese dimension, $X$ itself is projective.
• Proof Strategy: The proof reduces to the case of complex tori (Proposition 5.1). It shows that if a torus $T$ admits a rational class in the interior of its dual Kähler cone, the torus must be projective. This is extended to general manifolds via the Albanese map.

B. Ricci-Flat Manifolds

Theorem 1.6: If a compact Kähler manifold $X$ has $c_1(X) = 0$ (Ricci-flat) and satisfies Condition (K), then $X$ is projective.

• Significance: This answers the Oguiso–Peternell problem for a broad class of manifolds including Hyper-Kähler manifolds and Calabi-Yau manifolds. The proof uses the Beauville-Bogomolov decomposition theorem and the specific properties of the nef cone on Hyper-Kähler manifolds.

C. Compact Kähler Threefolds

The paper makes significant progress on the 3-dimensional case, excluding "simple non-Kummer" threefolds (which are conjectured not to exist).

• Theorem 1.8: If $X$ is a smooth compact Kähler threefold (not simple non-Kummer) satisfying Condition (K), then its algebraic dimension $a(X) \geq 2$.
  • This improves previous results by Oguiso and Peternell, which required the rational class to be a specific "curve class."
• Theorem 1.7: If $X$ is a smooth compact Kähler threefold (not simple non-Kummer) satisfying Condition (P) (dual pseudoeffective cone), then $X$ is projective.
  • Proof Strategy: The author rules out $a(X)=0$ and $a(X)=1$ using the classification in Proposition 2.13. For $a(X)=2$, $X$ is bimeromorphic to an elliptic fibration over a surface. The author proves that Condition (P) forces the base surface to be projective and the fibration to be algebraically connected, implying $X$ is projective.

D. Connection to 1-Cycles (Conditional Result)

Corollary 1.11: Assuming Question 1.10 has a positive answer (regarding the lifting of Hodge classes from subvarieties), then any compact Kähler threefold satisfying Condition (K) or (P) (excluding simple non-Kummer) is projective.

• This bridges the gap for the remaining cases in dimension 3, contingent on a specific Hodge-theoretic conjecture about 1-cycles.

4. Significance and Impact

1. Resolution of Open Problems: The paper provides a definitive answer to the Oguiso–Peternell problem for Ricci-flat manifolds and for Kähler threefolds of algebraic dimension $\geq 2$ (under mild assumptions).
2. Dual Kodaira Condition: It establishes that the dual Kodaira condition is a powerful indicator of algebraicity, specifically forcing the projectivity of the Albanese variety and, in many cases, the manifold itself.
3. Methodological Advances: The work introduces a robust framework for analyzing dual cones under dominant maps and blow-ups, and effectively utilizes the interplay between the geometry of fibrations (torus, abelian, elliptic) and Hodge theory.
4. Future Directions: The paper highlights the critical role of Question 1.10 (lifting Hodge classes). A positive resolution to this question would completely solve the Oguiso–Peternell problem for all compact Kähler threefolds.

In summary, Lin's work significantly advances the understanding of the relationship between the positivity of curve classes (dual cones) and the algebraic nature of Kähler manifolds, proving that the presence of a rational class in the interior of the dual Kähler cone is a strong constraint that often forces projectivity.