🔢 mathematics

The complex projective plane as a ball quotient

This paper classifies all ball quotient structures on the complex projective plane $\mathbb{P}^2$ with smooth pairwise normal-crossing branch divisors, proving that they are isomorphic to either the 1986 Deligne-Mostow example or a specific degree 9 cover of it.

Original authors: Cindy Tan

Published 2026-07-22

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Cindy Tan

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine you are an architect trying to build a perfect, infinite room out of a special kind of curved glass. In the world of mathematics, this "room" is called a ball, but it's not the round kind you play with; it's a high-dimensional space with its own unique rules for distance and shape, known as complex hyperbolic space. Now, imagine you want to fold this infinite room up into a finite, manageable shape, like tucking a giant map into a pocket. To do this, you have to glue parts of the room together. But here's the catch: you can't just glue it anywhere. If you glue it too tightly, the paper tears; if you glue it too loosely, it falls apart. The places where you glue the edges are called branch points, and the lines or curves where these glues happen are the branch divisors.

Mathematicians have long been fascinated by what happens when you fold these infinite rooms into shapes we already know, like the complex projective plane (think of it as a flat, infinite canvas that loops back on itself, similar to how the surface of the Earth is finite but has no edge). The big question is: Which patterns of glue lines allow you to fold this infinite room into that specific canvas without breaking the rules? In 1882, a genius named Poincaré solved this puzzle for a one-dimensional line (a circle). But for the two-dimensional canvas, the answer was a mystery. This paper, written by Cindy Tan, steps into that mystery to see if we can finally map out all the possible ways to fold this infinite room into a flat, two-dimensional world.

The Great Folding Puzzle

In this paper, Cindy Tan tackles the problem of classifying these "folding patterns" on a two-dimensional canvas. She isn't just looking for any pattern; she is looking for patterns made of smooth, straight lines that cross each other cleanly, like the grid lines on graph paper or the edges of a stained-glass window. She wants to know: if you take an infinite, curved room and fold it into a flat plane, what are the only possible arrangements of lines that make this mathematically possible?

The paper proves a very specific and tidy result: There are only two ways to do this.

The Complete Quadrilateral: Imagine drawing six lines on a piece of paper. If you arrange them so that they form a specific shape with four points where three lines meet (called triple points), and you assign specific "weights" (or glue strengths) to them—three lines get a weight of 3, and the other three get a weight of 2—you get a perfect fold. This is a famous shape discovered by Deligne and Mostow in 1986.
The Dual Hesse Arrangement: Now, imagine a more complex pattern with nine lines. If you arrange these nine lines so they create twelve points where three lines meet, and you give every single line a weight of 2, you get the second and final valid pattern.

The paper shows that if you try to use any other number of lines, or any other arrangement of crossings, the math simply doesn't work. The infinite room refuses to fold into the flat plane.

Ruling Out the "Almost" Solutions

One of the most important parts of this work is what it rules out. You might think that if you have a pattern that looks almost right, maybe it's just a matter of tweaking the numbers. But Tan proves that this isn't the case.

No Smooth Normal-Crossing Divisors: The paper explicitly argues against the idea that you can have a "smooth" pattern where every line crosses every other line at a single point without any triple meetings. It turns out, such a perfectly clean, simple crossing pattern is impossible for this specific folding problem.
No Other Line Counts: If you try to use 4 lines, or 5 lines, or 8 lines, the math breaks down. The paper calculates that the "energy" or "curvature" of the shape doesn't balance out unless you have exactly 6 lines (in the first case) or 9 lines (in the second case).
No Weird Weights: You can't just slap random numbers on the lines. For the 6-line pattern, the weights must be 3, 3, 3, 2, 2, 2. For the 9-line pattern, they must all be 2. If you change even one weight, the structure collapses.

How Sure Are We?

This isn't a guess or a simulation; it is a rigorous mathematical proof. The author uses a powerful toolkit of "numerical invariants"—think of these as mathematical rulers and scales that measure the shape of the folded room. If the room is folded correctly, these measurements must equal zero.

Tan calculated these measurements for every possible arrangement of lines that crosses cleanly. She found that for every single arrangement except the two mentioned above, the measurements were not zero. This means those arrangements are mathematically impossible. For the two special arrangements, she showed that the measurements do equal zero, confirming they are the only valid solutions.

The paper also clarifies that while there are other, more complicated ways to fold the room (using curves that aren't straight lines, or lines that cross in messy, non-perfect ways), those are outside the scope of this specific puzzle. Within the world of clean, straight, crossing lines, the answer is definitive: there are only two solutions, and we have found them both.

So, the next time you see a stained-glass window or a grid of streets, remember: most patterns are just pretty pictures. But if you were to fold the universe itself into a flat sheet, you'd be stuck with one of these two very specific, very special designs.

Technical Summary: The Complex Projective Plane as a Ball Quotient

Problem Statement
This paper addresses the classification of "ball quotient structures" on the complex projective plane $\mathbb{P}^2$ . Specifically, it investigates which divisors $D$ on $\mathbb{P}^2$ can arise as the branch divisor of a quotient map $B^2 \to \mathbb{P}^2$ , where $B^2$ is the complex 2-ball and the quotient is taken by a lattice $\Gamma < \text{PU}(1,2)$ . The study is restricted to the case where the divisor $D$ is an arrangement of smooth curves that are pairwise normal-crossing (i.e., any two components intersect transversely, though three or more may intersect at a point). The goal is to determine all such pairs $(\mathbb{P}^2, D)$ that form a ball quotient orbifold. This problem is presented as the natural dimension-2 analogue of Poincaré's 1882 classification of ball quotient structures on $\mathbb{P}^1$ .

Methodology
The paper employs a combination of numerical constraints derived from complex hyperbolic geometry and the theory of complex reflection groups. The core methodology involves:

Numerical Invariants: The author utilizes Hirzebruch proportionality and Enoki's results for curves to define numerical invariants for smooth compact orbifolds $(X, D)$ . Specifically, the paper defines an orbifold proportionality invariant $\text{Prop}(X, D)$ and a curve-specific invariant $\text{prop}_{(X,D)}(D_i)$ . For a smooth compact ball quotient orbifold, these invariants must vanish (Proposition 3.9). These invariants are expressed in terms of the canonical class $K_X$ , the divisor $D$ , and local contributions from singularities.
Singularity Classification: The analysis relies heavily on the fact that singularities of orbifold divisors on smooth surfaces must be "reflection group singularities." By the Chevalley–Shephard–Todd theorem, the local group at any singularity is a complex reflection group. The paper utilizes the Shephard–Todd classification of these groups in dimension 2 to enumerate all possible local singularity types (Table 1) and compute their specific numerical contributions ( $\lambda_{x,D}$ and $\alpha_{x,D}$ ) to the global invariants.
Uniformization Theorem: To verify that specific candidates are indeed ball quotients, the paper applies the Kobayashi–Nakamura–Sakai theorem (Theorem 3.12), an orbifold analogue of Yau's uniformization theorem. This theorem states that a smooth compact orbifold $(X, D)$ is a ball quotient if and only if $K_X + D$ is ample and the orbifold Chern numbers satisfy $c_1^2(X, D) = 3c_2(X, D)$ .
Case Analysis: The proof proceeds by reducing the general problem to specific cases based on the number of lines and their weights. The author first proves that any such divisor must be a line arrangement (Proposition 4.2). Subsequent propositions constrain the possible weights (Proposition 4.3) and enumerate the specific configurations of lines and singularities that satisfy the vanishing conditions of the invariants.

Key Results
The main result, Theorem 1.1, provides a complete classification of ball quotient divisors on $\mathbb{P}^2$ under the hypothesis that the components are smooth and pairwise normal-crossing. The theorem states that such a pair $(\mathbb{P}^2, D)$ is a ball quotient orbifold if and only if $D$ is projectively equivalent to one of the following two arrangements:

The Deligne–Mostow Example ( $D_Q$ ): A complete quadrilateral consisting of 6 lines with 4 triple points. Three lines meeting at a triple point have weight 3, while the remaining three lines have weight 2. This corresponds to the weights $(\frac{5}{6}, \frac{5}{12}, \frac{1}{4}, \frac{1}{4}, \frac{1}{4})$ in the Deligne–Mostow construction.
The Dual Hesse Arrangement ( $D_H$ ): An arrangement of 9 lines with 12 triple points, where every line has weight 2. This structure is realized as a degree 9 orbifold cover of the Deligne–Mostow example via the map $(X:Y:Z) \mapsto (X^3:Y^3:Z^3)$ .

The paper explicitly rules out all other configurations, including line arrangements with fewer than 5 lines, arrangements with weights other than 2 or 3 (with specific exceptions that fail the ampleness or vanishing conditions), and arrangements with different singularity configurations.

Significance and Claims
The paper claims to provide the first direct proof that the Deligne–Mostow example ( $D_Q$ ) and the dual Hesse arrangement ( $D_H$ ) are the only ball quotient divisors on $\mathbb{P}^2$ satisfying the smooth pairwise normal-crossing condition. While the fact that $D_Q$ and $D_H$ are ball quotients was previously known (via Deligne–Mostow's moduli space construction and Thomas Höfer's work), the non-existence of any other such structures in this class is a new result.

The author notes that the hypothesis of "pairwise normal-crossing" is strictly weaker than assuming the divisor is a normal-crossing divisor (which would forbid triple points). A consequence of the classification is that no smooth normal-crossing divisor on $\mathbb{P}^2$ can be a ball quotient divisor. The paper also acknowledges that ball quotient divisors with more complex singularities (e.g., those involving conics or non-pairwise normal-crossing arrangements) exist but fall outside the scope of this specific classification, requiring different methods to handle.

The work generalizes Poincaré's classification of $\mathbb{P}^1$ ball quotients to dimension 2 and establishes a rigid structural constraint on ball quotient orbifolds on the projective plane, contrasting with the more flexible constructions on blow-ups of $\mathbb{P}^2$ studied by Hirzebruch and others.

The Great Folding Puzzle

Ruling Out the "Almost" Solutions

How Sure Are We?

More like this