Dualizing complexes for algebraic stacks

This paper investigates dualizing complexes on algebraic stacks, specifically establishing their existence for tame Deligne–Mumford stacks of equicharacteristic under very general conditions.

The Gist

Here is an explanation of the paper "Dualizing Complexes for Algebraic Stacks" by Pat Lank, translated into everyday language with creative analogies.

The Big Picture: Mapping the Unmappable

Imagine you are an explorer trying to map a new, strange continent. In the world of mathematics, this continent is called Algebraic Stacks.

• Schemes (the old maps) are like flat, smooth sheets of paper. We know how to draw on them perfectly.
• Algebraic Spaces are like slightly crumpled paper; they are a bit more complex, but we can still handle them.
• Algebraic Stacks are like a continent made of shifting sand dunes, where some spots are "fuzzy" or have hidden symmetries (like a kaleidoscope). They are the most complex objects in this field, often used to describe families of shapes (moduli theory) or to solve deep geometry puzzles.

For a long time, mathematicians had a powerful tool called Grothendieck Duality. Think of this tool as a "Universal Translator" or a "Perfect Mirror." If you have a shape on one side of the mirror, the mirror tells you exactly what its "dual" (its reflection) looks like on the other side. This tool works beautifully for flat sheets (Schemes) and slightly crumpled paper (Spaces).

The Problem: When mathematicians tried to use this "Universal Translator" on the shifting sand dunes (Algebraic Stacks), it kept breaking. They couldn't be sure if a "dual" object even existed for these complex shapes. Without this tool, it's very hard to study the "singularities" (the weird, sharp, or broken points) of these shapes.

The Goal: Building a New Mirror

Pat Lank's paper is about successfully building a working "Universal Translator" specifically for these complex sand dunes (Algebraic Stacks).

The paper proves that for a very large and important class of these stacks (called Tame Deligne–Mumford stacks), we can construct these dual objects. This is a huge deal because it means we can now apply the same powerful mathematical techniques to these complex shapes that we've used for simpler ones for decades.

The Toolkit: How They Did It

To build this mirror, Lank didn't just invent a new tool from scratch. Instead, he combined two existing, powerful construction techniques like a master carpenter.

1. The "Compactification" Trick (Nagata Compactification)

Imagine you are trying to study a wild, infinite forest. It's too big to map all at once.

• The Trick: You build a fence around a specific part of the forest, creating a "compact" (finite, manageable) version of it.
• In Math: This is called Nagata Compactification. It allows mathematicians to take a messy, open-ended geometric shape and "close it up" into a nice, finite box.
• The Paper's Move: Lank uses a new, advanced version of this trick (developed by David Rydh) specifically for these "tame" stacks. This lets him turn a difficult, infinite problem into a finite one he can solve.

2. The "Smooth Patch" Strategy (Lisse-Étale Site)

Imagine you are trying to understand a bumpy, rocky mountain. It's too jagged to analyze directly.

• The Trick: You send out smooth, flat drones (smooth maps) to hover over different parts of the mountain. From the drone's perspective, the ground looks smooth and flat. You study the smooth ground, and then you stitch those smooth views together to understand the whole mountain.
• In Math: This is working on the lisse-étale site. Instead of looking at the stack directly, you look at it through "smooth lenses."
• The Paper's Move: Lank defines a "dualizing complex" not as a single rigid object, but as something that looks like a perfect dual object whenever you view it through these smooth lenses. If it works locally (on the smooth patches), it works globally (on the whole stack).

The "Tame" Condition: Why It Matters

The paper focuses on "Tame" stacks.

• The Analogy: Imagine a dance floor.
  • In a "Wild" stack, the dancers (symmetries) might spin so fast or in such chaotic ways that they tear the floor apart. This is "wild" behavior.
  • In a "Tame" stack, the dancers spin, but they do so in a controlled, predictable rhythm. They don't break the floor.
• Why it helps: Because the "dancers" are well-behaved (tame), the math doesn't explode. This allows the "Compactification" and "Smooth Patch" tricks to work together without falling apart.

The Main Result (The "Aha!" Moment)

The paper proves a specific rule (Theorem 1.1):
If you have a "dualizing complex" (a perfect mirror) for a base shape $X$, and you have a "tame" map $f$ leading to a new shape $Y$, then you can automatically generate the perfect mirror for $Y$ just by applying the map $f$ to the mirror of $X$.

In everyday terms:
If you have a perfect blueprint for a house, and you build a new house that is a "tame" variation of the first one (maybe it has a different roof but the same foundation), you don't need to draw a new blueprint from scratch. You can just use your old blueprint and apply a few simple adjustments to get the new one.

Why Should We Care? (The "So What?")

1. Solving Geometry Puzzles: Mathematicians are currently trying to solve the "Minimal Model Program," which is like trying to find the simplest, most efficient version of every possible shape. This requires understanding "singularities" (the broken parts). Now that we have these dualizing complexes, we can finally analyze these broken parts on complex stacks.
2. No "Properness" Limits: Previous attempts required the shapes to be "proper" (like a closed, finite box). This paper removes that restriction. We can now study open, infinite, or weirdly shaped stacks, which opens the door to many new discoveries in number theory and geometry.
3. Unification: It unifies the theory. Now, the rules for Schemes, Spaces, and Stacks are much more consistent. It's like finally realizing that the laws of physics for apples, oranges, and watermelons are all just different versions of the same gravity.

Summary

Pat Lank has successfully built a bridge between the simple, well-understood world of geometric shapes and the chaotic, complex world of algebraic stacks. By using a "fence" to contain the chaos (Compactification) and "smooth drones" to survey the terrain (Smooth Maps), he proved that the powerful "Universal Translator" (Dualizing Complexes) works for a vast new territory. This gives mathematicians a new set of keys to unlock some of the hardest problems in modern geometry.

Technical Summary

Here is a detailed technical summary of the paper "Dualizing Complexes for Algebraic Stacks" by Pat Lank.

1. Problem Statement

The paper addresses the existence and construction of dualizing complexes on algebraic stacks, a central object in Grothendieck duality theory.

• Context: While dualizing complexes are well-understood for schemes and algebraic spaces (where they are used to study singularities and derived categories), their existence on general algebraic stacks has been limited.
• The Gap: Previous attempts to define dualizing complexes on stacks often relied on the functor $f^\times$ (the right adjoint to the derived pushforward $Rf_*$) applied to structural morphisms. However, this approach fails in general (e.g., for separated étale morphisms between non-Gorenstein affine schemes) because $f^\times$ does not always coincide with the upper shriek functor $f^!$.
• Goal: To establish a robust existence theory for dualizing complexes on tame Deligne–Mumford (DM) stacks in equicharacteristic, utilizing the correct functorial framework ($f^!$) and overcoming the lack of properness constraints.

2. Methodology

The author employs a combination of modern functorial formalism, compactification techniques, and descent arguments.

• Site and Definitions:
  • The work is conducted on the lisse-étale site of the algebraic stack.
  • A dualizing complex $K$ on a stack $X$ is defined smooth-locally: $K$ is dualizing if $Ls^*K$ is a dualizing complex for every smooth morphism $s: U \to X$ from an algebraic space. This aligns with the intuition that duality should behave well under smooth base change.
• Functorial Framework:
  • The paper utilizes the formalism of concentrated morphisms (quasi-compact, quasi-separated, finite cohomological dimension) introduced by Hall and Rydh.
  • It relies heavily on the work of Amnon Neeman ([Nee23]), who established a functorial formalism for the upper shriek functor $f^!$ and its relationship with the right adjoint $f^\times$.
  • Key to the proof is the distinction between $f^\times$ and $f^!$. The paper shows that under specific conditions (universally quasi-proper morphisms), $f^\times \cong f^!$.
• Nagata Compactification:
  • The core technical engine is the Nagata compactification theorem. The author reduces the problem of existence on general separated morphisms to the case of universally quasi-proper (essentially proper) morphisms.
  • A morphism $f: Y \to X$ is factored as $Y \xrightarrow{j} Y' \xrightarrow{p} X$, where $j$ is a dominant flat monomorphism and $p$ is universally quasi-proper.
  • The existence of such compactifications for tame Deligne–Mumford stacks is cited from forthcoming work by David Rydh ([Ryd26]).
• Descent and Reduction:
  • The proofs utilize étale descent. By choosing a smooth/étale cover $s: U \to X$ from a scheme, the problem is reduced to the setting of algebraic spaces or schemes where results are known (e.g., [LM22], [Sta26]).
  • The paper defines a specific 2-subcategory $\mathcal{S}_e$ of Noetherian algebraic stacks closed under fiber products and étale morphisms, ensuring that the necessary compactifications and base change formulas hold within this category.

3. Key Contributions and Results

A. Definition and Characterization

• Definition 3.4: Formalizes the definition of a dualizing complex on a locally Noetherian algebraic stack as one that is dualizing smooth-locally.
• Proposition 3.8: Proves that a complex $K$ is dualizing on $X$ if and only if its pullback along some smooth surjective morphism from an algebraic space is dualizing. This reduces the stack problem to the algebraic space problem.

B. Main Existence Theorems

• Theorem 1.1 (Main Result): Let $k$ be a field. For a separated morphism $f: Y \to X$ of finite presentation between tame $k$-Deligne–Mumford stacks (where $X$ is Noetherian with separated diagonal), if $K$ is a dualizing complex on $X$, then $f^!K$ is a dualizing complex on $Y$.
  • Significance: This uses the upper shriek functor $f^!$, not $f^\times$, ensuring correctness in non-proper settings.
• Corollary 1.2: Dualizing complexes exist for any tame Deligne–Mumford $k$-stack that is of finite presentation and separated.
  • Scope: This covers all separated DM stacks of finite presentation over a field of characteristic zero (since all such stacks are tame).
  • Constraint: No properness assumption is required on the stack itself.
• Corollary 1.3 (Relative Proper Case): If $f: Y \to X$ is a tame proper morphism of Noetherian algebraic stacks, and $K$ is dualizing on $X$, then $f^\times K$ is a dualizing complex on $Y$.
  • Mechanism: In the proper case, $f^\times \cong f^!$, allowing the use of the right adjoint to the pushforward. This relies on a base change result for $f^\times$ from [Nee23].

C. Technical Lemmas

• Lemma 4.3: Establishes that the category of tame Noetherian DM $k$-stacks with separated diagonal satisfies the axioms of the 2-subcategory $\mathcal{S}_e$ required for Neeman's theorems (specifically regarding the existence of Nagata compactifications).
• Lemma 4.5 & Proposition 4.7: Prove the preservation of dualizing complexes under representable morphisms and Deligne–Mumford morphisms by reducing to the algebraic space case via étale covers.

4. Significance and Impact

• Foundational for Birational Geometry: The results provide the necessary duality tools for the Minimal Model Program (MMP) on algebraic stacks and spaces, particularly over $\mathbb{Q}$ (characteristic zero). The existence of dualizing complexes is crucial for studying singularities (e.g., canonical, log canonical) in these broader contexts.
• Resolution of Functorial Ambiguity: The paper clarifies the relationship between $f^\times$ and $f^!$ in the context of stacks. It demonstrates that while $f^\times$ is useful for proper morphisms, $f^!$ is the correct tool for general finite presentation morphisms, and proves that $f^!$ preserves the dualizing property in the tame DM setting.
• Generality: The results apply to a very broad class of stacks (tame DM stacks) without requiring the stacks to be proper, filling a significant gap in the literature where existence was previously limited or undefined.
• Future Applications: The author notes that as the theory of Nagata compactifications for stacks matures, these results are expected to extend to even broader classes of algebraic stacks (e.g., Artin stacks).

Summary

Pat Lank's paper successfully establishes the existence of dualizing complexes for a wide class of algebraic stacks (tame Deligne–Mumford stacks) by defining them via smooth-local properties and proving their stability under the upper shriek functor $f^!$. By leveraging Nagata compactification and Neeman's functorial formalism, the author bridges the gap between the well-understood duality theory of schemes and the more complex geometry of stacks, providing essential tools for modern birational geometry and singularity theory.