Misplaced Pages

This article has multiple issues. Please help improve it or discuss these issues on the talk page. (Learn how and when to remove these messages) This article may be too technical for most readers to understand. Please help improve it to make it understandable to non-experts, without removing the technical details. (January 2020) (Learn how and when to remove this message) This article needs additional citations for verification. Please help improve this article by adding citations to reliable sources. Unsourced material may be challenged and removed. Find sources: "Fraïssé limit" – news · newspapers · books · scholar · JSTOR (January 2020) (Learn how and when to remove this message) (Learn how and when to remove this message)

In mathematical logic, specifically in the discipline of model theory, the Fraïssé limit (also called the Fraïssé construction or Fraïssé amalgamation) is a method used to construct (infinite) mathematical structures from their (finite) substructures. It is a special example of the more general concept of a direct limit in a category. The technique was developed in the 1950s by its namesake, French logician Roland Fraïssé.

The main point of Fraïssé's construction is to show how one can approximate a (countable) structure by its finitely generated substructures. Given a class ${\displaystyle \mathbf {K} }$ of finite relational structures, if ${\displaystyle \mathbf {K} }$ satisfies certain properties (described below), then there exists a unique countable structure ${\displaystyle \operatorname {Flim} (\mathbf {K} )}$, called the Fraïssé limit of ${\displaystyle \mathbf {K} }$, which contains all the elements of ${\displaystyle \mathbf {K} }$ as substructures.

The general study of Fraïssé limits and related notions is sometimes called Fraïssé theory. This field has seen wide applications to other parts of mathematics, including topological dynamics, functional analysis, and Ramsey theory.

Finitely generated substructures and age

Fix a language ${\displaystyle {\mathcal {L}}}$. By an ${\displaystyle {\mathcal {L}}}$-structure, we mean a logical structure having signature ${\displaystyle {\mathcal {L}}}$.

Given an ${\displaystyle {\mathcal {L}}}$-structure ${\displaystyle {\mathcal {M}}}$ with domain ${\displaystyle M}$, and a subset ${\displaystyle A\subseteq M}$, we use ${\displaystyle \langle A\rangle ^{\mathcal {M}}}$ to denote the least substructure of ${\displaystyle {\mathcal {M}}}$ whose domain contains ${\displaystyle A}$ (i.e. the closure of ${\displaystyle A}$ under all the function and constant symbols in ${\displaystyle {\mathcal {L}}}$).

A substructure ${\displaystyle {\mathcal {N}}}$ of ${\displaystyle {\mathcal {M}}}$ is then said to be finitely generated if ${\displaystyle {\mathcal {N}}=\langle A\rangle ^{\mathcal {M}}}$ for some finite subset ${\displaystyle A\subseteq M}$. The age of ${\displaystyle {\mathcal {M}}}$, denoted ${\displaystyle \operatorname {Age} ({\mathcal {M}})}$, is the class of all finitely generated substructures of ${\displaystyle {\mathcal {M}}}$.

One can prove that any class ${\displaystyle \mathbf {K} }$ that is the age of some structure satisfies the following two conditions:

Hereditary property (HP)

If ${\displaystyle A\in \mathbf {K} }$ and ${\displaystyle B}$ is a finitely generated substructure of ${\displaystyle A}$, then ${\displaystyle B}$ is isomorphic to some structure in ${\displaystyle \mathbf {K} }$.

Joint embedding property (JEP)

If ${\displaystyle A,B\in \mathbf {K} }$, then there exists ${\displaystyle C\in \mathbf {K} }$ such that both ${\displaystyle A}$ and ${\displaystyle B}$ are embeddable in ${\displaystyle C}$.

Fraïssé's theorem

As above, we noted that for any ${\displaystyle {\mathcal {L}}}$-structure ${\displaystyle {\mathcal {M}}}$, ${\displaystyle \operatorname {Age} ({\mathcal {M}})}$ satisfies the HP and JEP. Fraïssé proved a sort-of-converse result: when ${\displaystyle \mathbf {K} }$ is any non-empty, countable set of finitely generated ${\displaystyle {\mathcal {L}}}$-structures that has the above two properties, then it is the age of some countable structure.

Furthermore, suppose that ${\displaystyle \mathbf {K} }$ happens to satisfy the following additional properties.

Amalgamation property (AP)

For any structures ${\displaystyle A,B,C\in \mathbf {K} }$, such that there exist embeddings ${\displaystyle f:A\to B}$, ${\displaystyle g:A\to C}$, there exists a structure ${\displaystyle D\in \mathbf {K} }$ and embeddings ${\displaystyle f':B\to D}$, ${\displaystyle g':C\to D}$ such that ${\displaystyle f'\circ f=g'\circ g}$ (i.e. they coincide on the image of A in both structures).

Essential countability (EC)

Up to isomorphism, there are countably many structures in ${\displaystyle \mathbf {K} }$.

In that case, we say that K is a Fraïssé class, and there is a unique (up to isomorphism), countable, homogeneous structure ${\displaystyle \operatorname {Flim} (\mathbf {K} )}$ whose age is exactly ${\displaystyle \mathbf {K} }$. This structure is called the Fraïssé limit of ${\displaystyle \mathbf {K} }$.

Here, homogeneous means that any isomorphism ${\displaystyle \pi :A\to B}$ between two finitely generated substructures ${\displaystyle A,B\in \mathbf {K} }$ can be extended to an automorphism of the whole structure.

Examples

The archetypal example is the class ${\displaystyle \mathbf {FCh} }$ of all finite linear orderings, for which the Fraïssé limit is a dense linear order without endpoints (i.e. no smallest nor largest element). By Cantor's isomorphism theorem, up to isomorphism, this is always equivalent to the structure ${\displaystyle \langle \mathbb {Q} ,<\rangle }$, i.e. the rational numbers with the usual ordering.

As a non-example, note that neither ${\displaystyle \langle \mathbb {N} ,<\rangle }$ nor ${\displaystyle \langle \mathbb {Z} ,<\rangle }$ are the Fraïssé limit of ${\displaystyle \mathbf {FCh} }$. This is because, although both of them are countable and have ${\displaystyle \mathbf {FCh} }$ as their age, neither one is homogeneous. To see this, consider the substructures ${\displaystyle {\big \langle }\{1,3\},<{\big \rangle }}$ and ${\displaystyle {\big \langle }\{5,6\},<{\big \rangle }}$, and the isomorphism ${\displaystyle 1\mapsto 5,\ 3\mapsto 6}$ between them. This cannot be extended to an automorphism of ${\displaystyle \langle \mathbb {N} ,<\rangle }$ or ${\displaystyle \langle \mathbb {Z} ,<\rangle }$, since there is no element to which we could map ${\displaystyle 2}$, while still preserving the order.

Another example is the class ${\displaystyle \mathbf {Gph} }$ of all finite graphs, whose Fraïssé limit is the Rado graph.

For any prime p, the Fraïssé limit of the class of finite fields of characteristic p is the algebraic closure ${\displaystyle {\overline {\mathbb {F} }}_{p}}$.

The Fraïssé limit of the class of finite abelian p-groups is ${\displaystyle \mathbb {Z} (p^{\infty })^{(\omega )}}$ (the direct sum of countably many copies of the Prüfer group). The Fraïssé limit of the class of all finite abelian groups is ${\displaystyle \bigoplus _{p{\text{ prime}}}\mathbb {Z} (p^{\infty })^{(\omega )}\simeq (\mathbb {Q} /\mathbb {Z} )^{(\omega )}}$.

The Fraïssé limit of the class of all finite groups is Hall's universal group.

The Fraïssé limit of the class of nontrivial finite Boolean algebras is the unique countable atomless Boolean algebra.

ω-categoricity and quantifier elimination

The class ${\displaystyle \mathbf {K} }$ under consideration is called uniformly locally finite if for every ${\displaystyle n}$, there is a uniform bound on the size of ${\displaystyle n}$-generated (substructures of) structures in ${\displaystyle \mathbf {K} }$. The Fraïssé limit of ${\displaystyle \mathbf {K} }$ is ω-categorical if and only if ${\displaystyle \mathbf {K} }$ is uniformly locally finite. If ${\displaystyle \mathbf {K} }$ is uniformly locally finite, then the Fraïssé limit of ${\displaystyle \mathbf {K} }$ has quantifier elimination.

If the language of ${\displaystyle \mathbf {K} }$ is finite, and consists only of relations and constants, then ${\displaystyle \mathbf {K} }$ is uniformly locally finite automatically.

For example, the class of finite dimensional vector spaces over a fixed field is always a Fraïssé class, but it is uniformly locally finite only if the field is finite. The class of finite Boolean algebras is uniformly locally finite, whereas the classes of finite fields of a given characteristic, or finite groups or abelian groups, are not, as 1-generated structures in these classes may have arbitrarily large finite size.

See also

• Structural Ramsey theory
• Hrushovski construction

References

1. ^ "The n-Category Café". golem.ph.utexas.edu. Retrieved 2020-01-08.
2. Hodges, Wilfrid. (1997). A shorter model theory. Cambridge University Press. ISBN 0-521-58713-1. OCLC 468298248.
3. Lupini, Martino (November 2018). "Fraïssé limits in functional analysis" (PDF). Advances in Mathematics. 338: 93–174. doi:10.1016/j.aim.2018.08.012. ISSN 0001-8708.
4. Schlicht, Philipp (January 7, 2018). "An introduction to model theory (lecture notes), Defn 2.2.1" (PDF). Mathematical Institute of the University of Bonn.
5. Notes on infinite permutation groups. Bhattacharjee, M. (Meenaxi), 1965–. Berlin: Springer. 1998. ISBN 3-540-64965-4. OCLC 39700621.{{cite book}}: CS1 maint: others (link)
6. ^ Hodges, Wilfrid (1993-03-11). Model Theory. Cambridge University Press. p. 350. doi:10.1017/cbo9780511551574. ISBN 978-0-521-30442-9.

v t e Mathematical logic
General	Axiom list Cardinality First-order logic Formal proof Formal semantics Foundations of mathematics Information theory Lemma Logical consequence Model Theorem Theory Type theory
Theorems (list)  and paradoxes	Gödel's completeness and incompleteness theorems Tarski's undefinability Banach–Tarski paradox Cantor's theorem, paradox and diagonal argument Compactness Halting problem Lindström's Löwenheim–Skolem Russell's paradox
Logics	Traditional Classical logic Logical truth Tautology Proposition Inference Logical equivalence Consistency Equiconsistency Argument Soundness Validity Syllogism Square of opposition Venn diagram Propositional Boolean algebra Boolean functions Logical connectives Propositional calculus Propositional formula Truth tables Many-valued logic 3 finite ∞ Predicate First-order list Second-order Monadic Higher-order Fixed-point Free Quantifiers Predicate Monadic predicate calculus
Set theory	Set hereditary Class (Ur-)Element Ordinal number Extensionality Forcing Relation equivalence partition Set operations: intersection union complement Cartesian product power set identities Types of sets Countable Uncountable Empty Inhabited Singleton Finite Infinite Transitive Ultrafilter Recursive Fuzzy Universal Universe constructible Grothendieck Von Neumann Maps and cardinality Function/Map domain codomain image In/Sur/Bi-jection Schröder–Bernstein theorem Isomorphism Gödel numbering Enumeration Large cardinal inaccessible Aleph number Operation binary Set theories Zermelo–Fraenkel axiom of choice continuum hypothesis General Kripke–Platek Morse–Kelley Naive New Foundations Tarski–Grothendieck Von Neumann–Bernays–Gödel Ackermann Constructive
Formal systems (list), language and syntax	Alphabet Arity Automata Axiom schema Expression ground Extension by definition conservative Relation Formation rule Grammar Formula atomic closed ground open Free/bound variable Language Metalanguage Logical connective ¬ ∨ ∧ → ↔ = Predicate functional variable propositional variable Proof Quantifier ∃ ! ∀ rank Sentence atomic spectrum Signature String Substitution Symbol function logical/constant non-logical variable Term Theory list Example axiomatic systems (list) of arithmetic: Peano second-order elementary function primitive recursive Robinson Skolem of the real numbers Tarski's axiomatization of Boolean algebras canonical minimal axioms of geometry: Euclidean: Elements Hilbert's Tarski's non-Euclidean Principia Mathematica
Proof theory	Formal proof Natural deduction Logical consequence Rule of inference Sequent calculus Theorem Systems axiomatic deductive Hilbert list Complete theory Independence (from ZFC) Proof of impossibility Ordinal analysis Reverse mathematics Self-verifying theories
Model theory	Interpretation function of models Model equivalence finite saturated spectrum submodel Non-standard model of arithmetic Diagram elementary Categorical theory Model complete theory Satisfiability Semantics of logic Strength Theories of truth semantic Tarski's Kripke's T-schema Transfer principle Truth predicate Truth value Type Ultraproduct Validity
Computability theory	Church encoding Church–Turing thesis Computably enumerable Computable function Computable set Decision problem decidable undecidable P NP P versus NP problem Kolmogorov complexity Lambda calculus Primitive recursive function Recursion Recursive set Turing machine Type theory
Related	Abstract logic Algebraic logic Automated theorem proving Category theory Concrete/Abstract category Category of sets History of logic History of mathematical logic timeline Logicism Mathematical object Philosophy of mathematics Supertask
Mathematics portal

• Category theory
• Mathematical logic
• Model theory