Contravariantly infinite resolving subcategories

This paper defines contravariantly infinite subcategories of finitely generated modules over a commutative Noetherian ring and establishes several criteria for such infiniteness specifically in the case where the ring is a local complete intersection.

The Gist

Imagine you are a mathematician exploring a vast, complex city called Ring City. This city is built on a specific set of rules (it's a "commutative Noetherian ring"). Inside this city, there are millions of different buildings, which we call modules.

Some buildings are very sturdy and well-constructed (these are the "Maximal Cohen-Macaulay" modules), while others are a bit wobbly or have finite lifespans (modules with "finite projective dimension").

The paper you are reading is about a special group of buildings called a subcategory. Think of a subcategory as a specific neighborhood or a club within the city. The author, Gen Tanigawa, is investigating a very specific property of these neighborhoods: Are they "Contravariantly Infinite"?

That sounds like a mouthful, so let's break it down with an analogy.

The "Right Approximation" Game

Imagine you are standing outside a neighborhood (let's call it Neighborhood X). You have a building (Module M) that doesn't belong to this neighborhood.

You want to get a glimpse of what's inside Neighborhood X, so you try to build a bridge from a building inside the neighborhood to your building outside. In math terms, you are looking for a Right X-Approximation. It's like trying to find the "best possible look-alike" of your building that actually lives inside the neighborhood.

• Contravariantly Finite: If every building outside the neighborhood can find a "best look-alike" inside the neighborhood, the neighborhood is "finite." It's very accessible.
• Contravariantly Infinite: If there is at least one building outside the neighborhood that cannot find any look-alike inside (no matter how hard you try), then the neighborhood is "infinite." It has a hard, impenetrable wall that blocks certain outsiders.

The Main Discovery (The "Big Reveal")

The paper focuses on a special type of city: a Local Complete Intersection. Think of this as a city built on very stable, predictable ground (like a perfect grid).

The author proves a fascinating rule about these cities. If you have a neighborhood (X) that is "resolving" (a fancy way of saying it's built up logically from the ground up, containing the basic free modules and closed under certain operations), then three things are actually the same thing:

1. The Neighborhood is "Infinite" (Impenetrable): There are outsiders who can't find a look-alike inside.
2. The Neighborhood has a "Wobbly" Building: Inside the neighborhood, there is at least one building that has a "finite lifespan" (finite projective dimension) but isn't just a basic free building.
3. The Neighborhood has a "Wobbly" Building (Version 2): Inside the neighborhood, there is at least one building that is not one of the super-sturdy "Maximal Cohen-Macaulay" ones.

The Analogy:
Imagine a club for "Super-Sturdy Buildings."

• If the club is finite (accessible), it means the club only contains Super-Sturdy Buildings. If you try to bring in a wobbly building, the club can easily find a sturdy look-alike for it.
• If the club is infinite (impenetrable), it means the club has secretly invited a "wobbly" building inside. Because this wobbly building is there, the club becomes so strange that some outsiders can no longer find a match for them.

The Catch:
This rule only works if the city has positive dimensions (it's not just a single point). If the city is tiny (0-dimensional, like an Artinian ring), the rule breaks. The author shows a counter-example where a neighborhood is "infinite" even though it only contains the most basic, sturdy buildings. It's like a tiny island where the rules of the mainland don't apply.

The "Radius" Concept

The paper also talks about the Radius of a neighborhood.

• Imagine you have one "seed" building.
• You can build new buildings by gluing existing ones together (extensions).
• The Radius is the minimum number of "gluing steps" you need to build every building in the neighborhood starting from that one seed.
• Finite Radius: You can build the whole neighborhood in a few steps.
• Infinite Radius: The neighborhood is too complex to be built from a single seed in a finite number of steps.

The paper proves that for these stable cities, a neighborhood has an Infinite Radius if and only if it contains a "wobbly" building (is not fully made of Super-Sturdy ones).

The "Coherence" Mystery (The Second Half)

In the second half, the author asks a deeper question: What happens if the city is a "Gorenstein" ring (a slightly more complex type of city)?

He introduces a concept called Coherence.

• Imagine you have a "Look-Alike Machine" (the functor).
• Sometimes, the machine produces a "finitely generated" output (it's a bit messy but manageable).
• Coherence means: If the machine produces a manageable output, it automatically produces a "finitely presented" output (it's perfectly organized).

The author shows that for these specific cities, if your neighborhood is made entirely of Super-Sturdy buildings, the neighborhood is Coherent. It's a guarantee that the math works out neatly.

Summary in Plain English

1. The Goal: The author wants to know when a group of mathematical objects (a subcategory) becomes "impenetrable" to outsiders (Contravariantly Infinite).
2. The Rule: In a well-behaved city (Complete Intersection), a group is impenetrable if and only if it contains a "weak link" (a module that isn't perfectly sturdy). If the group is made only of perfect, sturdy modules, it remains accessible to everyone.
3. The Exception: This rule fails if the city is too small (0-dimensional).
4. The Bonus: If the group is made of perfect modules, it also has a nice property called "Coherence," meaning its internal structure is very orderly.

The Takeaway:
This paper is like a detective story. The detective (the author) finds a clue: "If you see a wobbly building in this neighborhood, you know the neighborhood is closed off to the outside world." It connects the physical structure of the buildings (their stability) to the accessibility of the neighborhood (whether outsiders can find a match).

Technical Summary

Here is a detailed technical summary of the paper "Contravariantly Infinite Resolving Subcategories" by Gen Tanigawa.

1. Problem Statement and Motivation

The paper addresses the classification and structural properties of resolving subcategories within the category of finitely generated modules ($\text{mod } R$) over a commutative Noetherian local ring $R$.

• Background: The concept of contravariantly finite subcategories (where every module admits a right approximation by objects in the subcategory) was introduced by Auslander and Smalø to study almost split sequences and tilting theory. A landmark result by Takahashi established that over a complete Gorenstein local ring, the only contravariantly finite resolving subcategories are $\text{mod } R$, the category of maximal Cohen-Macaulay (MCM) modules ($\text{CM } R$), and the category of free modules ($\text{add } R$).
• The Gap: This paper investigates the "opposite" notion: contravariantly infinite subcategories. A full subcategory $\mathcal{X}$ is defined as contravariantly infinite if no module outside $\mathcal{X}$ admits a right $\mathcal{X}$-approximation.
• Core Question: Under what conditions is a resolving subcategory $\mathcal{X}$ contravariantly infinite? Specifically, the author seeks to characterize these subcategories over local complete intersection (CI) rings and explore whether these characterizations extend to general Gorenstein rings.

2. Methodology and Framework

The author employs a combination of homological algebra, approximation theory, and category theory. Key methodological tools include:

• Resolving Subcategories: Utilizing the definition by Auslander and Bridger (closed under direct summands, extensions, and kernels of epimorphisms, containing projectives).
• Approximation Theory: Analyzing the subcategory $\text{rap } \mathcal{X}$ (modules admitting right $\mathcal{X}$-approximations) and its relationship with $\mathcal{X}^\perp$ (modules $M$ where $\text{Ext}^i_R(X, M) = 0$ for all $X \in \mathcal{X}, i > 0$).
• Radius of Subcategories: Utilizing the notion of "radius" introduced by Dao and Takahashi, which measures the complexity of building modules in a subcategory via extensions.
• Functor Categories: Introducing the subcategory $\mathcal{C}$ of modules $M$ such that the restricted functor $\text{Hom}_R(-, M)|_{\mathcal{X}}$ is finitely presented in the functor category of $\mathcal{X}$. This bridges the gap between approximation properties and functorial finiteness.
• Specific Ring Classes: The analysis focuses heavily on Henselian local complete intersection rings and AB rings (rings satisfying the Auslander-Reiten condition), leveraging their specific homological properties (e.g., finite complexity, existence of canonical modules).

3. Key Contributions and Results

A. Characterization over Complete Intersection Rings (Main Theorem)

The primary result (Theorem 1.1 and Theorem 3.4) provides a complete characterization of contravariantly infinite resolving subcategories over a henselian local complete intersection ring $R$ with $\dim R > 0$.

For a resolving subcategory $\mathcal{X} \subseteq \text{mod } R$, the following are equivalent:

1. $\mathcal{X}$ is contravariantly infinite.
2. $\mathcal{X}$ contains a module of finite and positive projective dimension.
3. $\mathcal{X}$ contains a module that is not maximal Cohen-Macaulay (i.e., $\mathcal{X} \not\subseteq \text{CM } R$).
4. (If $R$ is quasi-excellent) $\mathcal{X}$ has infinite radius.

Significance: This establishes a sharp dichotomy: a resolving subcategory is contravariantly infinite if and only if it "escapes" the realm of MCM modules by containing modules with finite projective dimension. If it stays within MCM modules, it is contravariantly finite.

B. Necessity of Dimension Assumption

The author proves that the assumption $\dim R > 0$ is indispensable.

• Counterexample (Example 3.6): For an Artinian (dimension 0) complete intersection ring with codimension $c \geq 2$, there exists a resolving subcategory $\mathcal{E}$ (modules with complexity $< c$) that is contravariantly infinite, yet it contains no modules of finite positive projective dimension (as all non-free modules have infinite projective dimension in this context). This highlights a fundamental difference between positive-dimensional and zero-dimensional CI rings.

C. Extension to Gorenstein Rings and Functorial Properties

The paper investigates whether Theorem 1.1 holds for general Gorenstein rings (Question 4.1). While a complete answer is not found, the author introduces the concept of coherent resolving subcategories.

• Definition: A resolving subcategory $\mathcal{X}$ is coherent if the subcategory $\mathcal{C}$ (where $\text{Hom}(-, M)|_{\mathcal{X}}$ is finitely presented) coincides with $\text{rap } \mathcal{X}$.
• Result (Theorem 1.2): If $R$ is a henselian local complete intersection and $\mathcal{X}$ consists entirely of MCM modules, then for any finitely generated module $M$, if $\text{Hom}_R(-, M)|_{\mathcal{X}}$ is finitely generated, it is automatically finitely presented.
• Generalization: The paper proves that if $R$ is Gorenstein, satisfies the Auslander-Reiten condition (ARC), and $\mathcal{X} \subseteq \text{CM } R$ satisfies a specific condition (A), then $\mathcal{X}$ is coherent. This implies that for complete intersections, resolving subcategories contained in $\text{CM } R$ are always coherent.

4. Significance and Impact

• Classification: The paper significantly advances the classification of resolving subcategories. It clarifies that over complete intersections, the property of being "contravariantly infinite" is entirely determined by the presence of non-MCM modules (specifically those with finite projective dimension).
• Refinement of Approximation Theory: By introducing the subcategory $\mathcal{C}$ and the notion of coherence, the paper deepens the understanding of the relationship between right approximations and the finiteness properties of functors.
• Boundary Conditions: The explicit construction of counterexamples for Artinian rings delineates the precise boundaries of these homological phenomena, preventing over-generalization of results from positive-dimensional rings to zero-dimensional ones.
• Future Directions: The work sets the stage for further research into Gorenstein rings that are not complete intersections, specifically regarding the coherence of subcategories and the validity of the main theorem's equivalences in broader contexts.

In summary, Tanigawa provides a rigorous structural theorem for resolving subcategories over complete intersections, linking the geometric property of dimension and the homological property of projective dimension to the categorical property of contravariant infiniteness, while simultaneously laying the groundwork for extending these results to the broader Gorenstein setting.