No universal group in a cardinal

This paper introduces the "olive property" as a new sufficient condition for the non-existence of universal groups in certain cardinals, demonstrating that the class of groups satisfies this property despite failing the previously known SOP$_4$ criterion.

The Gist

Here is an explanation of Saharon Shelah's paper, "No Universal Group in a Cardinal," translated into simple language with creative analogies.

The Big Picture: The Search for the "Ultimate Container"

Imagine you are a librarian trying to organize a library. You have a specific rule for your books: they must all be groups (in the mathematical sense, like collections of numbers or shapes that follow specific combination rules).

You want to find the Ultimate Container. This is a single, massive book (a mathematical model) that is so big and so flexible that every other book in your library can be copied into it without breaking any rules. If you have this one "Universal Book," you don't need to keep the rest of the library; you just keep this one giant one.

The Question: Does this "Universal Book" exist for the class of all mathematical groups?

The Answer: It depends on the size of the library (the "cardinal" number).

• If the library is "nice" and follows standard rules of size (like the Generalized Continuum Hypothesis, or GCH), then Yes, you can usually find a Universal Book.
• But if the library is "messy" (where the rules of size get weird and unpredictable), Shelah proves that No, you cannot find a single book that holds everything.

The Problem with "Messy" Libraries

In the past, mathematicians knew that if a class of objects was "too complicated" (like Linear Orders, which are just lists of things arranged from smallest to biggest), you couldn't find a Universal Book in certain messy sizes.

However, Groups are tricky. They are more complex than simple lists, but they aren't as obviously chaotic as some other structures. For a long time, mathematicians weren't sure if Groups were "complicated enough" to prevent a Universal Book from existing.

The New Tool: The "Olive Property"

Shelah introduces a new concept called the "Olive Property." Think of this as a special security camera or a trap.

• The Metaphor: Imagine you are trying to build a giant warehouse (the Universal Group) to hold millions of smaller crates (smaller groups).
• The Trap: The "Olive Property" is a specific pattern of interactions between the crates. If your class of objects (Groups) has this property, it means the crates interact in a way that creates a "knot" or a "contradiction" if you try to stuff them all into one warehouse.
• The Result: If the Olive Property is present, no matter how big you make your warehouse, there will always be a new crate that doesn't fit, or a rule that breaks.

Shelah proves that Groups have the Olive Property. This means Groups are "complicated enough" that, in certain weird sizes, a single Universal Group simply cannot exist.

How the Proof Works (The "Guessing Game")

To prove this, Shelah uses a technique involving guessing.

1. The Setup: Imagine you have a giant list of instructions (a sequence of 0s and 1s) that tells you how to build a group.
2. The Challenge: You try to build a "Universal Group" that can mimic every possible version of these instructions.
3. The Olive Trap: Shelah constructs a scenario where the instructions force the group to do two contradictory things at once.
   • Analogy: Imagine a group of people trying to agree on a password. The "Olive Property" is a rule that says, "If Person A agrees with Person B, and Person B agrees with Person C, then Person A must disagree with Person C."
   • If you try to put everyone in one room (the Universal Group), the rules force a logical explosion. You can't satisfy everyone simultaneously.
4. The "Club" Guessing: To make this work, Shelah uses a set-theoretic tool called "guessing clubs." Imagine a security guard walking around a building. If the building is "messy" (doesn't follow GCH), the guard can't predict where the doors are. Shelah shows that because the guard can't predict the doors, the "Universal Group" fails to lock everyone in.

Why This Matters

1. It Solves a Mystery: Before this paper, we didn't know if Groups were "simple" enough to have a Universal member in all sizes, or "complex" enough to fail. Shelah says: They are complex.
2. It's a New Standard: The "Olive Property" is a new way to measure complexity. It's weaker than previous methods (meaning it catches more cases) but strong enough to prove that Groups are "non-structural" in these specific sizes.
3. Locally Finite Groups: The paper also looks at a specific type of group called "Locally Finite Groups" (groups where every small piece is finite). It turns out these are even harder to contain, and the same "No Universal" rule applies to them in messy sizes.

The "Amenability" Twist

In the final section, Shelah makes a correction. He previously thought Groups were "amenable" (a fancy math word meaning they behave nicely and can be predicted). He now proves they are not amenable.

• The Metaphor: "Amenable" is like a well-behaved dog that always sits when told. "Non-amenable" is like a dog that runs in circles and does the unexpected.
• The Conclusion: Groups are the wild dogs of mathematics. In certain sizes, you cannot predict their behavior well enough to build a single model that contains them all.

Summary in One Sentence

Saharon Shelah proves that mathematical Groups are too chaotic and complex to be contained in a single "Universal" model when the size of the universe is "messy," using a new detection method called the Olive Property to show that the rules of Groups inevitably lead to contradictions if you try to force them all into one box.

Technical Summary

Here is a detailed technical summary of Saharon Shelah's paper "No Universal Group in a Cardinal" (SH1029).

1. Problem Statement and Background

The central problem addressed is the existence of universal models within specific classes of structures (specifically groups) in a given cardinal $\lambda$.

• Context: In model theory, a model $M$ of cardinality $\lambda$ is universal for a class $K$ if every other model $N \in K$ of cardinality $\lambda$ can be elementarily embedded into $M$.
• Known Results:
  • If cardinal arithmetic satisfies the Generalized Continuum Hypothesis (GCH) or is "close" to it (e.g., $\lambda = 2^{<\lambda}$), many classes (including those of complete first-order theories) possess universal models.
  • If a theory $T$ has the Strict Order Property (SOP) or the stronger SOP$_4$, it is known that universal models often fail to exist in regular cardinals where arithmetic is "far" from GCH (e.g., $\mu^+ < \lambda < 2^\mu$).
• The Gap: The class of groups was known to have SOP$_3$ but NSOP$_4$ (No SOP$_4$). It was an open question whether the class of groups (and locally finite groups) admits a universal model in cardinals where GCH fails. Previous conditions for non-existence (like SOP$_4$) did not apply to groups.

2. Methodology: The "Olive Property"

To bridge the gap between SOP$_4$ and the behavior of groups, Shelah introduces a new sufficient condition for the non-existence of universal models called the Olive Property.

Definition of the Olive Property

A class of models (or a theory $T$) has the olive property if there exist quantifier-free formulas $(\varphi_0, \varphi_1, \psi)$ and a model $C$ such that:

1. Encoding Functions: For any sequence of functions $\bar{f} = \langle f_\alpha : \alpha < \lambda \rangle$ where $f_\alpha: \alpha \to \{0,1\}$, one can find a model $M$ and a sequence of tuples $\bar{a}_\alpha$ such that the truth of $\varphi_0$ or $\varphi_1$ between $\bar{a}_\alpha$ and $\bar{a}_\beta$ encodes the value of $f_\beta(\alpha)$.
2. Encoding Products: The formula $\psi$ encodes a specific condition on triples $(\alpha, \beta, \gamma)$ related to the constancy of the function $f_\gamma$ on the interval $[\alpha, \beta]$.
3. Forbidden Configuration: There is a specific "forbidden" configuration of four elements (or tuples) that cannot exist in any model of the class. Specifically, a configuration satisfying $\varphi_0, \varphi_1$ in a specific pattern and $\psi$ simultaneously is impossible.

Key Technical Insight: The olive property is strictly weaker than SOP$_4$ but stronger than SOP$_3$. It allows for the construction of "invariants" (using club guessing sequences) that prevent a single model from embedding all possible variations of the class.

Set-Theoretic Conditions ($Qr_1$)

The proof relies on a set-theoretic condition denoted as $Qr_1(\chi_2, \chi_1, \lambda)$. This condition involves:

• A stationary set $S \subseteq \lambda$.
• A sequence of clubs $\bar{C} = \langle C_\delta : \delta \in S \rangle$.
• A sequence of functions $\bar{g}$ and sets $\bar{A}$ that allow for "guessing" the behavior of functions on these clubs.
• The condition essentially states that if $\lambda$ is "far" from GCH (e.g., $\mu^+ < \lambda < 2^\mu$), one can construct a large family of non-embeddable models using these guessing sequences.

3. Key Contributions and Results

A. The Olive Property for Groups

The primary result of the paper is Theorem 2.1, which proves:

The class of groups has the olive property.

• Construction: Shelah explicitly constructs the witnessing formulas $(\varphi_0, \varphi_1, \psi)$ using group words.
  • $\varphi_0$: Conjugation relation ($y^{-1}x_0y = x_2$).
  • $\varphi_1$: Commutativity/Conjugation relations.
  • $\psi$: A complex word equation involving a specific group word $\sigma^*$ (constructed in Claim 2.3) that ensures a contradiction if the "forbidden" configuration exists.
• Proof Strategy: The proof involves constructing a group $G_{\bar{f}}$ generated by free generators subject to relations dictated by a function sequence $\bar{f}$. Using HNN extensions and free amalgamation, Shelah shows that for any sequence of functions $\bar{f}$, a group exists that encodes $\bar{f}$ via the olive formulas, but no single group can encode all possible $\bar{f}$ simultaneously due to the forbidden configuration.

B. Non-Existence of Universal Groups

Applying the olive property to the set-theoretic framework yields Theorem 1.9:

If a class has the olive property and the set-theoretic condition $Qr_1(\chi_2, \chi_1, \lambda)$ holds (which is true when $\mu^+ < \lambda < 2^\mu$ for regular $\mu, \lambda$), then there is no universal model in cardinality $\lambda$.

Specific Corollaries for Groups:

• General Groups: There is no universal group of cardinality $\lambda$ if $\mu^+ < \lambda < 2^\mu$ (where $\mu$ is regular).
• Locally Finite Groups: The paper extends this to the pair of classes (Locally Finite Groups, Groups). Conclusion 2.17 states that if $\mu = \text{cf}(\mu)$ and $\mu^+ < \lambda < 2^\mu$, there is no universal locally finite group of cardinality $\lambda$.

C. Non-Amenability of the Class of Groups

The paper corrects a previous claim (in earlier versions of Shelah's work) that the class of groups is "amenable" (a property related to the existence of universal models in forcing extensions).

• Claim 2.18: The class of groups is not amenable.
• Reasoning: In forcing extensions where GCH fails (specifically where $\lambda = \mu^+ = 2^\mu$), the olive property combined with club guessing implies the non-existence of universal models. This contradicts the definition of amenability.

4. Significance

1. Resolution of Open Questions: The paper answers Question 0.4 from Shelah's previous work, establishing that the class of groups behaves like linear orders (which have SOP) regarding the non-existence of universal models in "bad" cardinals, despite being NSOP$_4$.
2. New Classification Tool: The Olive Property provides a new, strictly intermediate dividing line in classification theory between SOP$_3$ and SOP$_4$. It demonstrates that SOP$_3$ is not sufficient to guarantee universal models, and a new, more nuanced condition is required to characterize non-structure.
3. Group Theory Implications: It settles the question of universal groups in ZFC (under specific cardinal arithmetic assumptions), showing that the complexity of the class of groups is high enough to prevent universality in a wide range of cardinals, even without the full strength of SOP$_4$.
4. Methodological Advance: The use of HNN extensions to construct groups encoding arbitrary function sequences, coupled with the specific "forbidden" configuration of the olive property, offers a powerful new technique for proving non-structure results in algebraic model theory.

Summary Conclusion

Saharon Shelah proves that the class of groups (and locally finite groups) lacks universal models in cardinals $\lambda$ where the cardinal arithmetic is sufficiently far from GCH (specifically $\mu^+ < \lambda < 2^\mu$). This is achieved by introducing the Olive Property, a condition weaker than SOP$_4$ but strong enough to trigger non-structure theorems via club guessing. This result refines the classification of groups in model theory, showing they are "complicated" enough to fail universality in many contexts, and corrects previous assumptions about their amenability.