Notions of rank and independence in countably categorical theories

Basic Information

• Paper ID: 2511.06113
• Title: Notions of rank and independence in countably categorical theories
• Author: Vera Koponen (Uppsala University, Sweden)
• Classification: math.LO (Mathematical Logic)
• Publication Date: November 8, 2025
• Paper Link: https://arxiv.org/abs/2511.06113

Abstract

This paper defines a hierarchical structure of ranks—$n$-ranks ($n$-rank), where $n < \omega$, for ω-categorical theories $T$ and their models $\mathcal{M}$. These ranks focus only on "imaginary elements below the $n$-th level," where the $n$-th level contains all elements of $M$ together with all imaginary elements that are equivalence classes of $\emptyset$-definable equivalence relations on $n$-tuples of $M$. Using $n$-ranks, the author defines the notion of $n$-independence ($n$-independence).

The main results include:

1. For all $n < \omega$, the $n$-independence relation restricted to $M_n$ satisfies all properties of Kim-Pillay independence relations, with symmetry as the only possible exception.
2. If the algebraic closure on $\mathcal{M}^{\mathrm{eq}}$ restricted to imaginary elements with $n$-rank 1 satisfies the exchange property, then $n$-independence is symmetric, thus becoming a complete independence relation.
3. If $n$-independence is symmetric for all $n < \omega$, then the theory $T$ is rosy.
4. Application: If $T$ has weak elimination of imaginaries and the algebraic closure on $M$ restricted to elements of 0-rank 1 satisfies the exchange property, then $T$ is superrosy with finite $U^\forking$-rank.

Research Background and Motivation

Core Problems to Address

The concepts of rank and independence in model theory have played a central role since Morley's pioneering work on uncountable categorical theories in the 1960s. These concepts are crucial for classifying complete first-order theories into different categories:

1. Development of Classification Theory: Shelah's stability theory divides theories into ω-stable, superstable, stable, or unstable. Kim and Pillay later generalized these concepts to the broader classes of simple and supersimple theories.
2. More General Independence Concepts: More general concepts such as thorn-independence were introduced, and the classes of rosy and superrosy theories were defined (Onshuus, Ealy, Adler, etc.).
3. Specificity of ω-Categorical Theories: ω-categorical theories (theories where all countable models are isomorphic) are important objects of study in model theory, including all finite homogeneous structures and Fraïssé limits.

Importance of the Problem

1. Completeness of Theory Classification: Rosy theories represent the largest known class of theories with independence relations satisfying certain natural properties. However, it remains unclear whether there exist non-rosy ω-categorical theories.
2. Boundary Line Problem: Finding the boundary between ω-categorical rosy theories and non-rosy theories is an important open problem.
3. Fine Structure Analysis: Understanding the fine structure of specific theories' models requires appropriate rank and independence tools.

Limitations of Existing Approaches

1. Restrictions of Global Properties: Existing independence concepts (such as forking independence, thorn independence) typically require considering all imaginary elements of the theory, which can be overly complex in certain cases.
2. Uncertainty of Symmetry: Many naturally defined independence relations satisfy all Kim-Pillay properties except symmetry, but symmetry is often difficult to verify.
3. Lack of Hierarchical Methods: There is no systematic approach to analyze the impact of imaginary elements on independence "layer by layer."

Research Motivation of This Paper

The core innovation of this paper lies in:

1. Hierarchical Rank Concept: Defining $n$-ranks that focus only on "imaginary elements below the $n$-th level," providing a tool for progressive analysis.
2. Exchange Property as Sufficient Condition for Symmetry: Isolating a parameterized property (Assumption 5.1) that is equivalent to the symmetry of $n$-independence.
3. Connection to Rosiness: Through hierarchical independence relations, establishing connections with rosy theories and providing new methods for determining whether ω-categorical theories are rosy.

Core Contributions

1. Hierarchical Ranks and Independence:
   • Defines $n$-ranks (Definition 3.2) and $n$-independence (Definition 4.1), where $n < \omega$
   • Proves that $n$-independence satisfies all properties of independence relations (except possibly symmetry) (Section 4)
2. Sufficient Conditions for Symmetry:
   • Proposes Assumption 5.1 (exchange property) and proves it is equivalent to the symmetry of $n$-independence (Theorem 5.15)
   • This provides a verifiable condition to determine whether independence is complete
3. Connection to Rosiness:
   • Proves that if Assumption 5.1 holds for all $n < \omega$, then the theory is rosy (Theorem 6.10)
   • This provides explicit obstacles for finding non-rosy ω-categorical theories
4. Concrete Applications:
   • For theories with weak elimination of imaginaries, provides sufficient conditions for superrosy with finite $U^\forking$-rank (Theorem 6.14)
   • Applies results to concrete examples such as atomless Boolean algebras
5. Technical Tools:
   • Introduces the concepts of $n$-coordination sequences ($n$-cs) and $n$-canonical coordination sequences ($n$-ccs) (Definitions 3.6, 5.3)
   • Establishes a pregeometry framework to analyze the exchange property (Lemma 5.2)

Detailed Methodology

Task Definition

The core task of this paper is:

• Input: An ω-categorical theory $T$ and its κ-saturated model $\mathcal{M}$ (κ an arbitrarily large infinite cardinal)
• Objectives:
  1. Define a series of rank functions $\text{rk}_n: \mathcal{P}_{\text{fin}}(M^{\text{eq}}) \times \mathcal{P}(M^{\text{eq}}) \to \omega \cup \{\infty\}$
  2. Define independence relations $\mathop{\smile\hskip -0.9em ^| \hskip 0.4em}_n$ based on these ranks
  3. Determine when these independence relations satisfy symmetry
  4. Use these results to determine the rosiness of the theory

Core Concept: Hierarchical Structure $M_n$

Definition (Definition 2.1):

• $M_0 = M$ ("real" elements)
• $M_{n+1} = M_n \cup X_{n+1}$, where $X_{n+1}$ is the set of all imaginary elements that are equivalence classes of $\emptyset$-definable equivalence relations on $M^{n+1}$
• $M_n$ is a substructure of $M^{\text{eq}}$ containing all imaginary elements "below the $n$-th level"

Key Properties:

• $M_0 \subseteq M_1 \subseteq M_2 \subseteq \cdots \subseteq M^{\text{eq}}$
• Each $M_n$ is interpretable in $M$, so $\text{Th}(M_n)$ is also ω-categorical (Fact 2.4)

Definition of $n$-Rank

Definition (Definition 3.2): For $A, B \subseteq M^{\text{eq}}$, the $n$-rank $\text{rk}_n(A/B)$ is recursively defined as:

1. $\text{rk}_n(A/B) \geq 0$ (always holds)
2. $\text{rk}_n(A/B) \geq \alpha + 1$ if and only if there exists $a \in \text{acl}_n(A) \setminus \text{acl}_n(B)$ such that $\text{rk}_n(A/\{a\} \cup B) \geq \alpha$
3. For limit ordinals $\alpha$, $\text{rk}_n(A/B) \geq \alpha$ if and only if for all $\beta < \alpha$, $\text{rk}_n(A/B) \geq \beta$

where $\text{acl}_n(A) = \text{acl}^{\text{eq}}(A) \cap M_n$ is the algebraic closure restricted to $M_n$.

Key Properties:

• For finite $A$, $\text{rk}_n(A/B)$ is always finite (Lemma 3.7)
• $\text{rk}_n(A/B) \leq \text{rk}_{n+1}(A/B)$ (Lemma 3.4)
• If the theory has soft elimination of imaginaries, then all $n$-ranks equal the 0-rank (Lemma 3.8)

$n$-Coordination Sequences ($n$-cs)

Definition (Definition 3.6): If $\text{rk}_n(A/B) = \alpha < \omega$, then a sequence $a_1, \ldots, a_\alpha \in \text{acl}_n(A)$ is an $n$-coordination sequence for $A/B$ if and only if for all $k = 1, \ldots, \alpha$: $a_k \notin \text{acl}_n(\{a_1, \ldots, a_{k-1}\} \cup B)$

Intuitive Understanding: An $n$-coordination sequence is a sequence that "witnesses" the rank, where each element is "new" relative to the preceding elements and $B$.

Key Lemma (Lemma 3.5):

• $\text{rk}_n(A/B) \geq \alpha$ if and only if there exists an $n$-cs of length $\alpha$
• If $a_1, \ldots, a_\alpha$ is an $n$-cs, then:
  • $\text{rk}_n(a_k/\{a_1, \ldots, a_{k-1}\} \cup B) = 1$ (the "incremental rank" of each element is 1)
  • $\text{acl}_n(A) \subseteq \text{acl}_n(\{a_1, \ldots, a_\alpha\} \cup B)$ (the sequence "spans" $A$)

$n$-Independence

Definition (Definition 4.1): A \mathop{\smile\hskip -0.9em ^| \hskip 0.4em}_n_C B if and only if for all finite $A' \subseteq A$: $\text{rk}_n(A'/BC) = \text{rk}_n(A'/C)$

Intuitive Understanding: $A$ is independent from $B$ relative to $C$ at level $n$ if adding $B$ does not change the rank of any finite subset of $A$ relative to $C$.

Proven Properties (Section 4):

1. Invariance (Invariance, Lemma 4.3): Preserved under elementary maps
2. Monotonicity (Monotonicity, Lemma 4.4): If A \mathop{\smile\hskip -0.9em ^| \hskip 0.4em}_n_B D and $B \subseteq C \subseteq D$, then A \mathop{\smile\hskip -0.9em ^| \hskip 0.4em}_n_B C and A \mathop{\smile\hskip -0.9em ^| \hskip 0.4em}_n_C D
3. Transitivity (Transitivity, Lemma 4.5)
4. Finite Character (Finite character, Lemma 4.6)
5. Locality (Locality, Lemma 4.7): For finite $A$, there exists finite $C \subseteq B$ such that A \mathop{\smile\hskip -0.9em ^| \hskip 0.4em}_n_C B
6. Extension (Extension, Lemma 4.8): Restricted to $M_n$

Missing Property: Symmetry—this is the central problem of the paper.

Exchange Property and Symmetry

Assumption 5.1 (Exchange Property): For $C \subseteq M^{\text{eq}}$, $2 \leq k < \omega$, $a_1, \ldots, a_k \in M_n$, if:

• For all $i = 1, \ldots, k$, $\text{rk}_n(a_i/C) = 1$
• $a_k \in \text{acl}^{\text{eq}}(\{a_1, \ldots, a_{k-1}\} \cup C) \setminus \text{acl}^{\text{eq}}(\{a_2, \ldots, a_{k-1}\} \cup C)$

then: $a_1 \in \text{acl}^{\text{eq}}(\{a_2, \ldots, a_k\} \cup C)$

Pregeometry Framework (Lemma 5.2): Define $X = \{d \in M_n : \text{rk}_n(d/C) = 1\}$, $\text{cl}(A) = \text{acl}_n(AC) \cap X$. Under Assumption 5.1, $(X, \text{cl})$ is a pregeometry.

$n$-Canonical Coordination Sequence ($n$-ccs, Definition 5.3): A sequence $a_1, \ldots, a_\alpha$ is an $n$-ccs for $A/B$ if there exist core indices $0 = k_0 < k_1 < \cdots < k_m = \alpha$ such that:

1. $\text{acl}_n(A) \subseteq \text{acl}_n(\{a_1, \ldots, a_\alpha\} \cup B)$
2. For all $j = 0, \ldots, m-1$, $\{a_{k_j+1}, \ldots, a_{k_{j+1}}\}$ is a basis of the set $\text{acl}_n(A) \cap \{d \in M_n : \text{rk}_n(d/\{a_1, \ldots, a_{k_j}\} \cup B) = 1\}$ (in the pregeometry sense)

Key Theorem (Theorem 5.15): If Assumption 5.1 holds for some $n$, then $\mathop{\smile\hskip -0.9em ^| \hskip 0.4em}_n$ restricted to $M_n$ is a complete independence relation (including symmetry).

Proof Strategy:

1. Use uniqueness of $n$-ccs (Lemma 5.4)
2. Prove via Proposition 5.13: If $\text{rk}_n(A/BC) < \text{rk}_n(A/C)$ (where $A, C \subseteq M^{\text{eq}}$, $B \subseteq M_n$ finite), then $\text{rk}_n(B/AC) < \text{rk}_n(B/C)$
3. Derive symmetry (Proposition 5.14)

Connection to Rosiness

Thorn-Independence (Definition 6.1): $\bar{a} \mathop{\smile\hskip -0.9em ^| \hskip 0.4em}^\forking_C \bar{b}$ if $\text{tp}(\bar{a}/C\bar{b})$ does not thorn-fork over $C$.

Key Lemma (Lemma 6.6): If Assumption 5.1 holds for $n$ and $\bar{a}, \bar{b}$ are finite sequences in $M_n$, then: $\bar{a} \mathop{\smile\hskip -0.9em ^| \hskip 0.4em}^\forking_{\!\!\!/\,\,\,\,\,} C \bar{b} \implies \bar{a} \mathop{\smile\hskip -0.9em ^| \hskip 0.4em}_n^{\!\!\!/\,\,\,\,\,} C \bar{b}$

Theorem (Theorem 6.10): If $T$ is ω-categorical and Assumption 5.1 holds for all $n < \omega$, then $T$ is rosy.

Proof Strategy:

1. Prove that thorn-independence has local character (Proposition 6.8)
2. Use the Ealy-Onshuus theorem (Theorem 6.9): A theory is rosy if and only if thorn-independence has local character

Technical Innovations

1. Hierarchical Treatment of Imaginary Elements

Innovation: Unlike traditional approaches that consider all imaginary elements $M^{\text{eq}}$ at once, this paper analyzes imaginary elements layer by layer through the hierarchical structure $M_0 \subseteq M_1 \subseteq \cdots$.

Advantages:

• Provides tools for progressive analysis
• For theories with soft elimination of imaginaries, the hierarchical structure "collapses" to the bottom level (Lemma 4.9)
• Allows fine control over the complexity of imaginary elements that need to be considered

2. Exchange Property as Sufficient Condition for Symmetry

Innovation: Assumption 5.1 transforms the symmetry problem into the exchange property of algebraic closure on specific element sets.

Technical Details:

• Only need to check elements with $\text{rk}_n = 1$
• Uses pregeometry theory (matroid theory)
• The exchange property is automatically satisfied in many natural examples (such as trivial algebraic closure)

Distinction from Baseline: Traditional approaches directly verify symmetry, while this paper provides an easier-to-verify equivalent condition.

3. Uniqueness of $n$-Canonical Coordination Sequences

Innovation: Lemma 5.4 proves the uniqueness of core indices in $n$-ccs, which is key to proving symmetry.

Technical Challenge: Requires proving equivalence of different bases under the pregeometry framework (Lemma 5.12).

4. Connecting Local and Global Properties

Innovation: Through Corollary 6.7, establishes the connection between local properties ($n$-independence) and global properties (thorn-independence): $\bar{a} \mathop{\smile\hskip -0.9em ^| \hskip 0.4em}^\forking_{\!\!\!/\,\,\,\,\,} C \bar{b} \implies \bar{a} \mathop{\smile\hskip -0.9em ^| \hskip 0.4em}_n^{\!\!\!/\,\,\,\,\,} C \bar{b} \text{ for all sufficiently large } n$

This allows deriving rosiness from hierarchical independence.

Experimental Setup

Note: This is a pure theoretical mathematics paper with no experimental component. All results are rigorous mathematical proofs.

Concrete Example Analysis

Example 3.3: Equivalence Relation Theory

Theory $T$: $E$ is an equivalence relation with infinitely many equivalence classes, each infinite.

Analysis:

• $\text{rk}_0(a) = 1$ (since $\text{acl}_0(A) = A$)
• $\text{rk}_1(a) = 2$ (since $[a]_E \in M_1$ is a new imaginary element)
• $\text{rk}_1(a/b) = 1$ if $[a]_E = [b]_E$
• Therefore a \mathop{\smile\hskip -0.9em ^| \hskip 0.4em}_0_\emptyset b but $a \mathop{\smile\hskip -0.9em ^| \hskip 0.4em}_1^{\!\!\!/\,\,\,\,\,} \emptyset b$

Significance: Different levels of independence can differ.

Atomless Boolean Algebras (End of Section 1)

Theory $T_B$: The complete theory of atomless Boolean algebras.

Known Results:

• Has quantifier elimination, so algebraic closure on real elements is trivial
• Has weak elimination of imaginaries (Newelski-Wencel)

New Result from This Paper (by Theorem 6.14): $T_B$ is superrosy with finite $U^\forking$-rank.

Significance: This extends Conant's results on free amalgamation Fraïssé limits.

Summary of Main Results

Theorem Hierarchy

1. Basic Properties (Section 4): $\mathop{\smile\hskip -0.9em ^| \hskip 0.4em}_n$ satisfies all independence relation properties except possibly symmetry
2. Symmetry Condition (Theorem 5.15): Assumption 5.1 (exchange property) $\iff$ $\mathop{\smile\hskip -0.9em ^| \hskip 0.4em}_n$ is symmetric
3. Rosiness Criterion (Theorem 6.10): If Assumption 5.1 holds for all $n < \omega$, then $T$ is rosy
4. Superrosiness Criterion (Theorem 6.14): If $T$ has soft elimination of imaginaries and Assumption 5.1 holds for $n=0$, then $T$ is superrosy with finite $U^\forking$-rank
5. Corollary (Theorem 6.13): If $T$ is ω-categorical and algebraic closure is trivial, then $T$ is rosy

Theoretical Significance

Contributions to Classification of ω-Categorical Theories:

• Provides new tools for determining whether ω-categorical theories are rosy
• If non-rosy ω-categorical theories exist, then there must exist some $n$ where Assumption 5.1 fails
• This provides explicit obstacles for finding non-rosy ω-categorical theories

Open Problem: The author notes "I am not aware of any example of an ω-categorical theory that is not rosy," so this paper's results may help:

1. Find non-rosy ω-categorical theories, or
2. Prove that all ω-categorical theories are rosy

Related Work

Ranks and Independence in Model Theory

1. Morley's Work (1965): Uncountable categorical theories, pioneering use of ranks
2. Shelah's Stability Theory (1990): ω-stable, superstable, stable theories
3. Kim-Pillay (1997): Simple theories and forking independence
4. Adler, Ealy, Onshuus (2006-2009): Thorn-independence and rosy theories

Specific Research on ω-Categorical Theories

1. Totally Categorical Theories (Ahlbrandt-Ziegler 1986, Hrushovski 1989, 1993)
2. ω-Categorical ω-Stable Theories (Cherlin-Harrington-Lachlan 1986)
3. Smoothly Approximable Structures (Cherlin-Hrushovski 2003)
4. Simple Finitely Homogeneous Structures (Baldwin-Freitag-Mutchnik 2024, Koponen 2018)
5. ω-Categorical NIP Theories (Simon 2022)

Elimination of Imaginaries

1. Hodges-Hodkinson-Macpherson (1990): Conditions for weak elimination of imaginaries
2. Newelski-Wencel (2001): Weak elimination of imaginaries for Boolean algebras
3. Conant (2017): Weak elimination of imaginaries for free amalgamation Fraïssé limits

Positioning of This Paper

This paper innovates in the following ways:

1. More General Framework: Does not assume simple or NIP, only ω-categoricity
2. Hierarchical Method: Systematically exploits the hierarchical structure of imaginary elements
3. Connecting Local and Global: Establishes connections between $n$-independence and thorn-independence
4. Broader Application Range: Covers a wider class of ω-categorical theories

Conclusions and Discussion

Main Conclusions

1. Methodological Contribution: Hierarchical ranks and independence provide new tools for analyzing ω-categorical theories
2. Theoretical Criterion: The exchange property (Assumption 5.1) is a necessary and sufficient condition for symmetry of $n$-independence
3. Classification Result: If the exchange property holds for all $n$, then the theory is rosy
4. Concrete Application: For theories with weak elimination of imaginaries, provides simple criteria for superrosiness

Limitations

1. Symmetry Problem: Symmetry of $n$-independence still requires additional assumptions (Assumption 5.1), not automatic
2. Computational Complexity: Verifying Assumption 5.1 may require checking infinitely many $n$
3. Scope of Applicability: Main results target ω-categorical theories; generalization to uncountable categorical theories is unclear
4. Concrete Examples: Paper contains few concrete examples, mainly theoretical framework

Future Directions

The author mentions in the Epilogue:

1. Finite Structure Sequences: Study finite structure sequences $(B_n : n < \omega)$ with "good" closure operators and their limits
2. Connection to Hrushovski Construction: Explore whether Theorem 6.10 or 6.14 apply to theories constructed via Hrushovski methods
3. Finding Non-Rosy Examples: This paper's results provide explicit obstacles for finding non-rosy ω-categorical theories

Open Problems:

• Are all ω-categorical theories rosy?
• If not, can this paper's methods find counterexamples?
• For theories not satisfying Assumption 5.1, can other useful independence concepts be defined?

In-Depth Evaluation

Strengths

1. Theoretical Innovation

• Original Concepts: The hierarchical method of $n$-ranks and $n$-independence is novel
• Technical Depth: Uses sophisticated tools from pregeometry theory, uniqueness of $n$-ccs, etc., with elegant proofs
• Unified Framework: Incorporates multiple known results (such as Conant's results on Fraïssé limits) into a unified framework

2. Mathematical Rigor

• Complete Proofs: All major results have detailed proofs
• Clear Logic: Logical chain from basic definitions to main theorems is complete
• Counterexample Awareness: Examples 3.3 and 4.2 illustrate that different levels of independence can differ

3. Problem Importance

• Central Problem: Whether all ω-categorical theories are rosy is an important open problem in model theory
• Methodological Value: Even if ultimately proving all ω-categorical theories are rosy, the hierarchical method has independent value
• Application Potential: Results apply to multiple concrete theory classes

4. Writing Quality

• Clear Structure: Progresses from preliminaries, $n$-ranks, $n$-independence, exchange property to rosiness, layer by layer
• Technical Details: Key lemmas (such as Lemmas 3.5, 5.4, 5.12) have detailed proofs
• Motivation: Introduction and Epilogue well explain research motivation and background

Weaknesses

1. Insufficient Concrete Examples

• Only two simple examples (Example 3.3 on equivalence relations, atomless Boolean algebras)
• Lacks detailed analysis of more complex theories (such as random graph, dense linear order)
• No examples where Assumption 5.1 fails

2. Computational Feasibility

• Verifying Assumption 5.1 requires checking all $n < \omega$, practically difficult
• No algorithms or heuristics provided for computing $n$-ranks in concrete theories
• The construction process of $n$-ccs (proof of Lemma 5.4) exists but may be computationally complex

3. Comparison with Existing Results

• For theories already known to be rosy (such as simple theories), no comparison of this paper's method efficiency with existing methods
• No discussion of relationship between this paper's rosiness criterion and other criteria (such as NIP + certain conditions)

4. Generalizability

• Main results limited to ω-categorical theories; applicability to more general theories unclear
• Necessity of Assumption 5.1 (whether it is the weakest condition) not discussed

Impact Assessment

Contribution to the Field

1. Theoretical Tools: Hierarchical ranks and independence may become standard tools for studying ω-categorical theories
2. Open Problems: Provides new attack angles for "whether all ω-categorical theories are rosy"
3. Unified Perspective: Incorporates multiple theory classes (theories with weak elimination of imaginaries, theories with trivial algebraic closure) into unified framework

Practical Value

1. Decision Methods: Theorems 6.10 and 6.14 provide practical criteria for determining rosiness/superrosiness
2. Concrete Applications: Provides new results for concrete theories such as atomless Boolean algebras
3. Methodology: Pregeometry framework and $n$-ccs methods may apply to other problems

Reproducibility

• Theoretical Results: All proofs are self-contained and can be independently verified
• Missing Implementation: No algorithms or computational implementations provided for computing $n$-ranks or verifying Assumption 5.1
• Example Verification: Given examples (Example 3.3) can be verified by hand

Applicable Scenarios

This paper's methods are particularly suitable for:

1. ω-Categorical Theories with Good Closure Properties:
   • Algebraic closure is trivial or nearly trivial
   • Has (weak) elimination of imaginaries
2. Finite Homogeneous Structures:
   • Fraïssé limits
   • Smoothly approximable structures
3. Cases Requiring Fine Rank Analysis:
   • When global rank is too coarse, $n$-ranks provide finer information
4. Studying Boundaries of Rosiness:
   • Finding non-rosy ω-categorical theories
   • Understanding the essence of rosiness

Unsuitable Scenarios:

• Uncountable categorical theories (methods depend on ω-categoricity)
• Theories with very complex algebraic closure ($n$-ranks may be difficult to compute)

Key References

1. Kim, B., Pillay, A. (1997). Simple theories. Annals of Pure and Applied Logic, 88, 149-164.
   • Defines standard properties of independence relations
2. Ealy, C., Onshuus, A. (2007). Characterizing rosy theories. The Journal of Symbolic Logic, 72, 919-940.
   • Proves rosy theories equivalent to thorn-independence having local character (this paper's Theorem 6.9)
3. Conant, G. (2017). An axiomatic approach to free amalgamation. The Journal of Symbolic Logic, 82, 648-671.
   • Proves free amalgamation Fraïssé limits have weak elimination of imaginaries and are superrosy
4. Shelah, S. (1990). Classification Theory, Revised Edition. North-Holland.
   • Classical work on stability theory
5. Cherlin, G., Harrington, L., Lachlan, A. H. (1986). ω-categorical ω-stable structures. Annals of Pure and Applied Mathematics, 28, 103-135.
   • Foundational work on ω-categorical ω-stable theories

Summary

This paper provides an innovative hierarchical framework for studying ω-categorical theories through $n$-ranks and $n$-independence, systematically analyzing the impact of imaginary elements. The core technical contribution is reducing the symmetry problem of $n$-independence to the exchange property of algebraic closure (Assumption 5.1), and using this result to establish connections with rosy theories. While the paper has some limitations in concrete examples and computational methods, its theoretical depth and contribution to important open problems make it significant work in model theory. In particular, this paper provides new tools and perspectives for answering the central question "whether all ω-categorical theories are rosy," and may have lasting impact on the field.