The Flat Cover Conjecture for Monoid Acts

Basic Information

• Paper ID: 2507.04155
• Title: The Flat Cover Conjecture for Monoid Acts
• Author: Sean Cox (Virginia Commonwealth University)
• Classification: math.CT (Category Theory), math.AT (Algebraic Topology), math.LO (Mathematical Logic), math.RA (Rings and Algebras)
• Publication Date: November 20, 2025 (arXiv v3)
• Paper Link: https://arxiv.org/abs/2507.04155
• Support: NSF grant DMS-2154141

Abstract

This paper proves the Flat Cover Conjecture (FCC) for the category of (right) acts over any right-reversible monoid $S$, provided that flat $S$-acts are closed under stable Rees extensions. The paper demonstrates that the class $\mathcal{F}$-Mono (monomorphisms of $S$-acts with flat Rees quotients) is cofibrantly generated in such categories, answering a question of Bailey and Renshaw. However, the cofibrantly generated property of $\mathcal{SF}$-Mono (monomorphisms with strongly flat Rees quotients) is stronger, as it implies a bound on the cardinality of indecomposable strongly flat acts. Similarly, the cofibrantly generated property of $\mathcal{U}_{\mathcal{F}}$ (unital monomorphisms with flat complements) implies a bound on the cardinality of indecomposable flat acts. The core tool is a new theorem characterizing the cofibrantly generated property of monomorphism classes via "almost everywhere" effectiveness.

Research Background and Motivation

Problem Background

1. Classical Results and Conjectures: Bass's theorem states that every $R$-module has a projective cover if and only if the ring $R$ is perfect. Enochs conjectured in the 1980s that every $R$-module has a flat cover, which is the famous Flat Cover Conjecture (FCC), proved around 2000.
2. Challenges in Non-Additive Categories: The FCC has many analogous results in additive categories, but the situation is more difficult in non-additive categories. For the category Act-$S$ of acts over a monoid $S$, Bailey and Renshaw proved: if the cardinalities of indecomposable flat $S$-acts are bounded, then FCC holds. However, such bounds do not always exist.
3. Key Questions: Bailey and Renshaw posed two core problems:
   • Question 1.1: Is $\mathcal{F}S$-Mono cofibrantly generated in Act-$S$?
   • Question 1.2: For monoids with a left zero element, is $\mathcal{F}S$-Mono cofibrantly generated in the category Act$_0$-$S$ of central acts?

Research Motivation

1. Theoretical Importance: Establishing homological algebra theory in non-additive categories requires understanding the covering properties of flat objects (rather than merely projective objects).
2. Technical Obstacles: The proofs in module categories "appear to depend on the additive structure of the category," and direct generalization to act categories faces difficulties.
3. Conceptual Differences: In non-additive categories, "geometric flatness" (preserving monomorphisms under tensor) and "categorical flatness" (directed colimit of finitely presented projectives) are typically not equivalent, increasing the problem's complexity.

Core Contributions

1. Main Theorem 1.3: For any monoid $S$, the following are equivalent:
   • $S$ is right-reversible and $\mathcal{F}S$-Mono is closed under composition in Act-$S$
   • $\mathcal{F}S$-Mono is cofibrantly generated in Act-$S$
   
   For $S$ with a left zero element, corresponding results are given for Act$_0$-$S$.
2. Corollary 1.4: If $S$ is a right-reversible monoid and flat $S$-acts are closed under stable Rees extensions, then FCC holds in Act-$S$.
3. Theorem 1.5: For right LO monoids (for any $s,t\in S$, there exists $u$ such that $su=t$ or $tu=s$), the class of pure monomorphisms with flat Rees quotients is cofibrantly generated, and FCC holds.
4. Theorem 1.6: Characterizes when $\mathcal{SF}$-Mono is cofibrantly generated:
   • $S$ is left-collapsible
   • $\mathcal{SF}$-Mono is closed under composition
   • The cardinalities of indecomposable strongly flat acts are bounded
5. Theorem 1.7: For act classes $\mathcal{X}$ satisfying the separating coproduct property, $\mathcal{U}_{\mathcal{X}}$ is cofibrantly generated if and only if the cardinalities of indecomposable $\mathcal{X}$-members are bounded.
6. Core Technical Theorem 3.12: Provides a new characterization of cofibrantly generated property—a monomorphism class $\mathcal{M}$ is cofibrantly generated if and only if it is "almost everywhere effective." This is a "top-down" characterization that completely avoids dealing with transfinite constructions.

Methodology Details

Core Concepts

Cofibrantly Generated: A class $\mathcal{M}$ is cofibrantly generated if it is the closure of some set $\mathcal{M}_0$ under pushouts, transfinite compositions, and retracts.

Cellularly Generated: $\mathcal{M} = \text{cell}(\mathcal{M}_0)$, i.e., closure under pushouts and transfinite compositions (but not retracts).

Rees Quotient: For a sub-act $A\subseteq B$, the Rees congruence is defined by $b_0\rho_A b_1$ if and only if $b_0=b_1$ or both are in $A$. $B/A$ denotes the quotient act.

Technical Innovation: Almost Everywhere Effectiveness

Definition 3.10 (Core Innovation): Let $\mathcal{M}$ be a cellularly closed monomorphism class. For a regular cardinal $\kappa>|S|$, $\mathcal{M}$ is called $\kappa$-almost everywhere effective if there exists a parameter $p$ such that:

For any $N$ satisfying $S\cup\{S,p\}\subset N\prec_{\Sigma_n}(V,\in)$ with $N\cap\kappa$ transitive, for all $f\in N\cap\mathcal{M}$:

• $f\restriction N\in\mathcal{M}$ (restriction to $N$)
• $r_{f,N}\in\mathcal{M}$ (map from pushout to codomain)

Here the diagram $D_{f,N}$ is involved:

A ----f----> B
|            |
v            v
N∩A ---> N∩B
|            
v  ⌜ pushout
P^{f,N} --r_{f,N}--> (implicit connection to B)

Actual form:

A ----------f----------> B
|                        ↑
|                        |
v                    r_{f,N}
N∩A --> N∩B --> P^{f,N} = A∪(N∩B)
      f↾N      ⌜

Key Theorem 3.12

Theorem 3.12: For a monoid $S$ and a cellularly closed monomorphism class $\mathcal{M}$, the following are equivalent:

1. $\mathcal{M}$ is almost everywhere effective
2. $\mathcal{M}$ is cellularly generated

Proof Strategy:

(I)⟹(II) Direction (Almost Everywhere Effective ⟹ Cellularly Generated):

1. Let $\kappa$ witness the almost everywhere effectiveness of $\mathcal{M}$
2. For $f:A\to B\in\mathcal{M}$, prove by induction on the cardinality of $|B|$ that $f\in\text{cell}(\mathcal{M}_{<\kappa})$
3. For $|B|=\lambda\geq\kappa$, construct an elementary submodel chain $\langle N_\alpha:\alpha\leq\text{cf}(\lambda)\rangle$ such that:
   • $|N_\alpha|<\lambda$
   • $B\subset N_{\text{cf}(\lambda)}$
   • Each $N_\alpha$ has transitive intersection with $\kappa$
4. Using almost everywhere effectiveness, prove that $f$ decomposes into a transfinite composition where each step has codomain of size $<\lambda$
5. Inductive hypothesis ensures each step is in $\text{cell}(\mathcal{M}_{<\kappa})$

(II)⟹(I) Direction (Cellularly Generated ⟹ Almost Everywhere Effective):

1. Let $\mathcal{M}=\text{cell}(\mathcal{M}_{<\kappa})$
2. Parameter $p=(\mathcal{M}_{<\kappa},\kappa)$
3. For $f\in N\cap\mathcal{M}$, $f$ has a transfinite decomposition $\vec{f}$ where each step is a pushout of morphisms in $\mathcal{M}_{<\kappa}$
4. By elementarity, $\vec{f}\in N$
5. For $\alpha\in N\cap\mu$, the pushout square restricted to $N$ remains a pushout (Lemma 3.11)
6. Therefore $f\restriction N$ is a transfinite composition of pushouts of $\mathcal{M}_{<\kappa}$-members
7. Similar argument shows $r_{f,N}\in\text{cell}(\mathcal{M}_{<\kappa})$

Set-Theoretic Tools

Elementary Submodels: Using $\Sigma_n$ elementarity $N\prec_{\Sigma_n}(V,\in)$ avoids technical issues from Gödel's incompleteness theorem. Obtained via Levy-Montague reflection principle and Löwenheim-Skolem theorem.

Key Lemma 3.8: If $S\cup\{S\}\subset N\prec^*(V,\in)$ and $B,\rho\in N$ ($\rho$ a congruence on $B$), then:

• $N\cap\rho$ is a congruence on $N\cap B$
• $N\cap(B/\rho)\cong(N\cap B)/(N\cap\rho)$
• For Rees quotients: $\frac{B}{A\cup(N\cap B)}\cong\frac{B/A}{N\cap(B/A)}$

Main Theorem Proofs

Proof of Theorem 1.3

(1)⟹(2): Let $S$ be right-reversible and $\mathcal{F}$-Mono closed under composition.

1. By 5, Theorem 3.11, $\mathcal{F}$-Mono is cellularly closed
2. Prove $\mathcal{F}$-Mono is $\kappa$-almost everywhere effective (for any $\kappa>|S|+\aleph_0$)
3. For $f:A\hookrightarrow B\in\mathcal{F}$-Mono with $f\in N\prec^*(V,\in)$:
   • By Lemma 3.8, $\frac{N\cap B}{N\cap A}\cong N\cap(B/A)$
   • By Corollary 4.3, $N\cap(B/A)$ is flat (since $B/A$ is flat and $B/A\in N$)
   • Similarly, $\frac{B/A}{N\cap(B/A)}$ is flat (using right-reversibility)
   • Therefore both $f\restriction N$ and $r_{f,N}$ are in $\mathcal{F}$-Mono
4. By Theorem 3.12, $\mathcal{F}$-Mono is cellularly generated

(2)⟹(1): If $\mathcal{F}$-Mono is cofibrantly generated, then in particular it is closed under composition. Use the contrapositive of Lemma 4.2: if $B/A$ is flat and $A\neq\emptyset$, then $S$ is right-reversible.

Proof of Theorem 1.6 (Strongly Flat Case)

(1)⟹(2) (Cofibrantly Generated ⟹ Left-Collapsible + Cardinality Bound):

1. Let $\mathcal{SF}$-Mono be cofibrantly generated; by Theorem 3.12, there exists $\kappa$ such that it is $\kappa$-almost everywhere effective
2. Prove $S$ is left-collapsible: Take any $f:\emptyset\to B\in\mathcal{SF}$-Mono, so $B$ is strongly flat. Construct $N$ such that $B/(N\cap B)$ is strongly flat with $N\cap B\neq\emptyset$; by 27, Theorem 6.2(7), $S$ is left-collapsible.
3. Prove indecomposable strongly flat acts have size $<\kappa$:
   • Assume there exists a locally cyclic strongly flat act $F$ with $|F|>\kappa$
   • Construct $M,N$ with $|M|=\kappa\subset M$, $|N|<\kappa$, $M\in N$
   • Consider $\emptyset\hookrightarrow F$; by almost everywhere effectiveness:
     • $M\cap F\hookrightarrow F$ is in $\mathcal{SF}$-Mono
     • $(M\cap F)\cup(N\cap F)\hookrightarrow F$ is in $\mathcal{SF}$-Mono
     • Therefore $\frac{F}{(M\cap F)\cup(N\cap F)}$ is strongly flat
   • By Lemma 5.2, the inclusion $(M\cap F)\cup(N\cap F)\hookrightarrow F$ is pure
   • But $|M\cap F|=\kappa$, so there exist $x\in(M\cap F)\setminus(N\cap F)$ and $y\in(N\cap F)\setminus(M\cap F)$
   • By local cyclicity, there exist $z,s,t$ with $zs=x$, $zt=y$
   • Purity requires existence of $z_P\in(M\cap F)\cup(N\cap F)$ with the same property, contradiction

(2)⟹(1) (Left-Collapsible + Cardinality Bound ⟹ Cofibrantly Generated):

1. Let indecomposable strongly flat acts have size $<\kappa$
2. For $f:A\hookrightarrow B\in\mathcal{SF}$-Mono, let $F:=B/A$ be strongly flat
3. By elementarity, $F=\bigsqcup_{i\in I}F_i$ decomposes into indecomposable parts with $\langle F_i\rangle_{i\in I}\in N$
4. For $i\in N\cap I$: $F_i\in N$ and $|F_i|<\kappa$, so by Fact 3.7, $F_i\subset N$
5. For $i\in I\setminus N$: $N\cap F_i=\emptyset$
6. Therefore $N\cap F=\bigsqcup_{i\in N\cap I}F_i$ is strongly flat, and $F\setminus(N\cap F)=\bigsqcup_{i\in I\setminus N}F_i$ is strongly flat
7. By left-collapsibility, the singleton act $\Theta_S$ is strongly flat, so $\frac{F}{N\cap F}\cong\Theta_S\sqcup\bigsqcup_{i\in I\setminus N}F_i$ is strongly flat

Proof of Theorem 1.7 (Unital Monomorphism Case)

For act classes $\mathcal{X}$ satisfying $\bigsqcup_{i\in I}X_i\in\mathcal{X}\Leftrightarrow\forall i\ X_i\in\mathcal{X}$:

Cofibrantly Generated ⟹ Cardinality Bound: Similar to Theorem 1.6, using almost everywhere effectiveness and local connectedness to derive contradiction.

Cardinality Bound ⟹ Cofibrantly Generated:

1. For $f:A\hookrightarrow B=A\sqcup X\in\mathcal{U}_{\mathcal{X}}$ (with $X\in\mathcal{X}$)
2. Decompose $X=\bigsqcup_{i\in I}X_i$ into connected components
3. $N\cap X=\bigsqcup_{i\in N\cap I}X_i\in\mathcal{X}$
4. $X\setminus(N\cap X)=\bigsqcup_{i\in I\setminus N}X_i\in\mathcal{X}$
5. Therefore both $f\restriction N$ and $r_{f,N}$ are in $\mathcal{U}_{\mathcal{X}}$

Experimental Setup

This is a pure theoretical mathematics paper with no experiments or numerical computations. All results are rigorous mathematical proofs.

Experimental Results

Not applicable (pure theoretical work).

Related Work

Classical Background

1. Bass's Theorem: Characterization of projective cover existence
2. Enochs's FCC: Proof in module categories (Bican, El Bashir, Enochs, 2001)
3. Generalizations in Additive Categories: Work by Crivei-Prest-Torrecillas, Positselski-Rosický, etc.

Research on Act Categories

1. Bailey-Renshaw Series:
   • 3 Proves FCC holds when indecomposable flat acts have bounded cardinality
   • 4 Case of right-cancellative monoids
   • 5 Weak factorization systems and cofibrantly generated property of $\mathcal{F}S$-Mono
   • 2 Strong Flat Cover Conjecture (SFCC)
2. Kruml 18: Provides counterexample to SFCC (non-right-reversible monoid)
3. Renshaw's Foundational Work:
   • 24,25 Relationships between flatness, stability, and purity
   • 27 Characterization of P-unital monomorphisms and strong flatness

Category-Theoretic Tools

1. Rosický 28: Cofibrantly generated property of $\mathcal{F}$-Mono in module categories
2. Borceux-Rosický 7: Purity in locally presentable categories, concept of effective coproducts
3. Makkai-Rosický 22: Cellular category theory
4. Lieberman et al. 20: Cofibrantly generated pure monomorphisms (in Ab without Borceux-Rosický conditions but still cofibrantly generated)

Advantages of This Work

1. Unified Framework: Theorem 3.12 provides a unified method for checking cofibrantly generated property applicable to multiple situations
2. Avoids Transfinite Constructions: Almost everywhere effectiveness is a "top-down" characterization, easier to verify
3. Exact Characterizations: Theorems 1.6 and 1.7 give necessary and sufficient conditions for strongly flat and unital monomorphism cases
4. Technical Breakthrough: Successfully applies set-theoretic methods to non-additive categories

Conclusions and Discussion

Main Conclusions

1. FCC holds for right-reversible monoids (condition: flat acts closed under stable Rees extensions)
2. FCC holds unconditionally for right LO monoids
3. New characterization of cofibrantly generated: Almost everywhere effectiveness ⟺ Cellularly generated
4. Complete characterization for strongly flat case: Cofibrantly generated ⟺ Left-collapsible + Closed under composition + Cardinality bound
5. Characterization for unital monomorphisms: Cofibrantly generated ⟺ Indecomposable members have bounded cardinality

Limitations

1. Stable Rees Extension Assumption: Theorems 1.3 and Corollary 1.4 require the assumption that "flat acts are closed under stable Rees extensions." The author acknowledges uncertainty about how restrictive this condition is, but notes:
   • Holds when every quasi-flat act is flat
   • Holds when every stable monomorphism is pure
   • Actually the "stable" modifier is redundant (Lemma 4.2)
2. Conjecture 7.1: The author conjectures that for right-reversible monoids, FCC should hold unconditionally (without the stable Rees extension assumption)
3. Pure Monomorphism Case: Conjecture 7.2 proposes that $\mathcal{F}$-PureMono should always be cellularly generated (for right-reversible monoids), but this is only proved for right LO monoids
4. Complete Characterization of SFCC: Theorem 1.6 gives necessary and sufficient conditions for $\mathcal{SF}$-Mono to be cofibrantly generated, but this is only a sufficient condition for SFCC to hold; complete characterization of when SFCC holds remains open

Future Directions

1. Eliminate Stable Rees Extension Assumption: Prove or disprove Conjecture 7.1
2. Pure Monomorphisms in General Case: Prove or disprove Conjecture 7.2
3. Category-Theoretic Translation (Problem 7.3): Translate "almost everywhere effective" into pure category-theoretic language
4. Generalization to Other Categories: The author notes Theorem 3.12 can be generalized to arbitrary locally presentable categories
5. Complete Characterization of SFCC: Find necessary and sufficient conditions for SFCC to hold

In-Depth Evaluation

Strengths

1. Theoretical Innovation

• Theorem 3.12 is a major breakthrough: Provides a novel characterization of cofibrantly generated property, avoiding the complexity of directly handling transfinite constructions. This is a "top-down" approach, more elegant than traditional "bottom-up" constructions.
• Deep integration of set theory and category theory: Creatively uses elementary submodels and reflection principles to address category-theoretic problems, which is novel in homological algebra of non-additive categories.

2. Technical Rigor

• Careful set-theoretic treatment: Section 3.2 specifically addresses technical issues arising from Gödel's incompleteness theorem, using $\Sigma_n$ elementarity rather than full elementarity, demonstrating high technical maturity.
• Precision of Key Lemma 3.8: The characterization of Rees quotient behavior under elementary submodels is the foundation for all subsequent proofs, with rigorous proof.
• Lemma 3.11: Clear proof that pushouts are preserved under restriction to elementary submodels, with generalizable technique.

3. Systematic Nature of Results

• Answers Bailey-Renshaw questions: Theorem 1.3 directly answers Questions 1.1 and 1.2
• Unified framework: Theorems 3.12, 1.6, 1.7 demonstrate application of the same technique in different settings
• Necessary and sufficient conditions: Theorems 1.6 and 1.7 provide exact characterizations, not merely sufficient conditions

4. Clarity of Exposition

• Well-structured: From motivation, background, core techniques to applications, logic is clear
• Technical details transparent: Section 3.2 specifically discusses set-theoretic technical issues, avoiding confusion
• Abundant remarks and discussion: Such as Remarks 3.2, 5.3, 6.1, etc., help readers understand technical choices

5. Mathematical Depth

• Corollary 3.14: Proves that intersections of cofibrantly generated classes remain cofibrantly generated, demonstrating robustness of the theory
• Lemmas 5.1 and 5.2: Precisely identify essential differences between strongly flat and flat
• Theorem 1.5: Unconditional result for right LO monoids, demonstrating power of the theory

Weaknesses

1. Restrictiveness of Main Assumptions

• Stable Rees Extension Assumption: The author admits uncertainty about how restrictive this assumption is. While some satisfying cases are given, systematic characterization is lacking.
• Lack of Counterexamples: No concrete examples of monoids not satisfying stable Rees extension closure are provided, making necessity of the assumption unclear.

2. Relationship with Existing Theory

• Relationship with 21: The author mentions Theorem 3.1 of 21 is "close" but has continuity assumptions. More detailed comparison would help.
• Problem 7.3 Unresolved: Admits inability to fully translate "almost everywhere effective" into pure category-theoretic language, which somewhat limits the result's category-theoretic "purity."

3. Application Scope

• Focus on Existence: The paper primarily proves existence of covers, with less discussion of constructivity, uniqueness, and practical applications.
• Computational Aspects: No discussion of how to verify in practice the condition "flat acts closed under stable Rees extensions."

4. Technical Complexity

• Set-Theoretic Barrier: While Section 3.2 attempts to lower the barrier, concepts like $\Sigma_n$ elementarity and Levy-Montague reflection may challenge non-logicians.
• Proof Length: The proof of Theorem 3.12 (especially the (I)⟹(II) direction) involves complex transfinite induction and diagram chasing, potentially difficult to fully master.

5. Open Problems

• Conjectures 7.1 and 7.2: Core results still depend on unresolved conjectures, limiting theoretical completeness.
• Complete Characterization of SFCC: Theorem 1.6 only gives conditions for $\mathcal{SF}$-Mono to be cofibrantly generated, which is sufficient but not necessary for SFCC; complete characterization remains open.

Impact

1. Contribution to the Field

• Methodological Breakthrough: Theorem 3.12 provides new technical tools potentially applicable to other locally presentable categories (author mentions generalization in future papers)
• Answers Long-Standing Questions: Bailey-Renshaw's questions posed years ago receive (conditional) affirmative answers
• Unified Perspective: Brings flat, strongly flat, and unital monomorphism cases into a unified framework

2. Practical Value

• Theoretical Foundation: Provides basis for developing relative homological algebra in act categories
• Verifiable Criteria: Theorems 1.6 and 1.7 provide checkable necessary and sufficient conditions
• Lemma 3.3 Application: Connects cofibrantly generated property to cover existence, providing practical tools

3. Reproducibility

• Complete Proofs: All major results have detailed proofs
• Sufficient Citations: External results used are precisely cited
• Technical Transparency: Detailed discussion of set-theoretic techniques enhances verifiability

4. Follow-Up Research

• Existing Follow-Up: Author mentions paper 10 with collaborators using similar techniques
• Generalization Potential: Remark 3.13 mentions plans to generalize to general locally presentable categories
• Open Problems: Section 7 provides clear directions for future research

Applicable Scenarios

1. Theoretical Research:
   • Homological algebra in non-additive categories
   • Monoid act theory
   • Model categories and homotopy theory
2. Concrete Applications:
   • Act categories over right-reversible monoids (e.g., commutative monoids)
   • Act categories over right LO monoids
   • Central acts over monoids with left zero element
3. Technical Tools:
   • Situations requiring proof that certain morphism classes are cofibrantly generated
   • Study of cover and envelope existence
   • Construction of weak factorization systems
4. Inapplicable Scenarios:
   • Non-right-reversible monoids (FCC may fail)
   • Applications requiring constructive algorithms (existence proofs don't provide construction)
   • Pure category-theoretic research avoiding set theory

Comparison with Related Work

Selected References

1. 3 Bailey-Renshaw (2014): Covers of acts over monoids and pure epimorphisms - Establishes connection between cardinality bounds on indecomposable flat acts and FCC
2. 5 Bailey-Renshaw (2014): Weak factorization systems for S-acts - Poses core questions answered in this paper
3. 6 Bican-El Bashir-Enochs (2001): All modules have flat covers - Proves original FCC
4. 7 Borceux-Rosický (2007): Purity in algebra - Introduces effective coproduct concept; Theorem 3.12 is its improvement
5. 22 Makkai-Rosický (2014): Cellular categories - Foundation of cellular generation theory
6. 27 Renshaw (2002): Stability and flatness in acts over monoids - Establishes relationships between stability, purity, flatness; source of Lemma 4.2
7. 28 Rosický (2002): Flat covers and factorizations - Cofibrantly generated in module categories; starting point for this paper's generalization

Overall Assessment: This is a technically profound and important pure mathematics paper. Theorem 3.12 is a genuine innovation, creatively applying set-theoretic elementary submodel techniques to category-theoretic problems. While main results depend on the "stable Rees extension closure" assumption (whose restrictiveness remains unclear), unconditional results are given for right LO monoids. The paper answers Bailey-Renshaw's long-standing questions and establishes foundations for homological algebra in act categories. Technical treatment is rigorous, exposition is clear, and this is an important contribution to the field. Main weaknesses are the unclear necessity of certain assumptions and several unresolved core conjectures. The paper's methodological value may exceed its concrete results, as the "almost everywhere effective" framework has broad generalization potential.