16 Omitting Types

Let M be an L-structure. Which types must be realised?

Definition 16.1. We say p∈SnM(A) is isolated if it is an isolated point with respect to topology on SnM(A) (i.e. {p} is open).

Example. For a∈A⊆M, tpM(a∕A) is isolated by x=a ({tpM(a∕M)}=[x=a]).

Proposition 16.2. Assuming that:

  • p∈SnM(A).

Then the following are equivalent:
  • (i)
    p is isolated
  • (ii)
    {p}=[φ(z)] for some LA-formula φ(x¯). In this case we say φ(x¯) isolates p.
  • (iii)
    There is an LA-formula φ(x¯)∈p such that for any ψ(x¯)∈p,
    ThA(M)⊨∀⁡x(φ(x¯)→ψ(x¯)).

PIC

Proof. (i) ⟺ (ii): Obvious.

(ii) ⟹ (iii): Assume φ(x¯) isolates p. Fix an LA-formula ψ(x¯). We want to show M⊨∀⁡x¯(φ(x¯)→ψ(x¯)). So suppose M⊨φ(a¯). Then tpM(a¯∕A)∈[φ(x¯)]={p}.

So tpM(a¯∕A)=p, hence M⊨ψ(a¯).

(iii) ⟹ (ii): By assumption, for every LA-formula we have ψ(x¯)∈p, [φ(x¯)]⊆[ψ(x¯)]. Thus if qsin⁡[φ(x¯)], a∈[ψ(x¯)]. So ψ(x¯)∈q, so p⊆q, so p=q. □

Proposition 16.3. Assuming that:

  • T is complete and consistent

  • p∈Sn(T) is isolated

Then p is realised in every M⊨T.

Proof. Fix p∈Sn(T), isolated by φ(x¯). Fix M⊨T.

By Proposition 13.4, there is some N≽M realising p.

So N⊨∃⁡x¯φ(x¯), so M⊨∃⁡x¯φ(x¯).

Fix a¯∈Mn such that M⊨φ(a¯). Then a¯⊨p as for any φ(x¯)∈p we have

M⊨∀⁡x¯(φ(x¯)→ψ(x¯)).

So M⊨φ(a¯). □

Theorem 16.4 (Omitting Types Theorem). Assuming that:

  • L is countable

  • p∈Sn(T) is non-isolated

Then there is a countable M⊨T such that p is not realised in M (M omits p).

Proof. Let L∗=L∪C, with C a countably infinite set of new constants.

An L∗-theory has the witness property if for any L∗-formula φ(x¯) there is a constant c∈C such that T∗⊨∃⁡xφ(x)→ψ(c).

Fact: Suppose T∗ is a complete, satisfiable L∗-theory with the witness property.

Define ∼ on C such that c∼d if and only if T∗⊨c=d. Let M=C∕∼, and define an L∗-structure on M such that:

Then M is a well-defined L∗-structure and M⊨T∗.

Note we have M⊨φ([c1],…,[cn]) if and only if T∗⊨φ(c1,…,cn). We call M the Henkin model of T∗.

Fix p∈Sn(T) non-isolated.

Aim: build a complete, satisfiable L∗-theory T∗⊇T, with the witness property, .

Such that for all c1,…,cn∈C there is some φ(x¯)∈p such that T∗⊨¬⁡φ(c1,…,cn). Then the Henkin model of T∗ omits p.

Enumerate all the L∗-sentences φ0,φ1,… and all c(n={c¯1,c¯2,…}. We build a satisfiable L∗-theory T∪{𝜃1,𝜃2,…} such that

Let 𝜃0 be ∀⁡v(v=v), and suppose we have 𝜃0,…,𝜃m.

Case 1: m+1=3i+1 for some i.

If T∪{𝜃m,φi} is satisfiable then 𝜃m+1=𝜃m∧φi. Otherwise 𝜃m+1=𝜃m∧¬⁡φ.

So T∪{𝜃m+1} is satisfiable by construction.

Case 2: m+1=3i+2 for some i.

Suppose φi is ∃⁡v,ψ(v) for some ψ an L∗-formula, and ⊨𝜃i→φi (otherwise, let 𝜃m+1=𝜃m).

Choose a c∈C not used in 𝜃m. Let 𝜃m+1 be 𝜃m∧ψ(i).

Exercise: check T∪{𝜃m+1} is satisfiable.

Case 3: m+1=3i+3 for some i.

Let c¯i=(c1,…,cn). Without loss of generality assume x1,…,xn not used in 𝜃m. We build an L-formula as follows:

Then φ(x¯) doesn’t isolated p.

By Proposition 16.2, there is some ψ(x¯)∈p with

T⁄⊨∀⁡x(φ(x¯)→ψ(x¯)).

Let 𝜃m+1 be 𝜃m∧¬⁡ψ(c1,…,cn). Check 𝜃m+1 is satisfiable.

TODO □

Definition 16.5 (Atomic, prime). Fix M⊨T.

  • We say M is atomic if every n-type over ∅ realised in M is isolated.

  • We say M is prime if for any N⊨T there is an elementary embedding M→N.

Example. Let K⊨ACF⁡0. Then ℚ¯=ℚalg⊆K, and ℚ¯<κ by quantifier elimination. So ℚ¯ is the prime model of ACF⁡0.

Assume L is countable.

Fact: M is prime if and only if M is countable and atomic.

Theorem 16.6. Assuming that:

  • L countable

Then the following are equivalent:
  • (i)
    T has a prime model.
  • (ii)
    T has an atomic model.
  • (iii)
    For all n≥1, the isolated types are dense.

Theorem 16.7.

  • (a)
    Suppose |Sn(T)|<2ℵ0 for all n. Then T has a prime model and a countable saturated model.
  • (b)
    If T has a countable saturated model, then it has a prime model.

Example. What if |Sn(T)|=2ℵ0?

Th(ℤ,+,0) has no countable saturated model, no prime model.

Th(ℤ,+,0,1) has a prime model, but no countable saturated model.

Definition 16.8. For κ≥ℵ0, let I(T,κ) be the number of models of T of size κ (modulo isomorphism).

What size can I(T,κ) be?

Theorem 16.9 (Ryll-Nardsewski / Engeler / Svenonius 59). Assuming that:

  • L countable

  • T is a complete L-theory with infinite models

Then the following are equivalent:
  • (i) T is ℵ0-categorical.
  • (ii) For all n≥1, every type in Sn(T) is isolated.
  • (iii) For all n≥1, Sn(T) is finite.
  • (iv) For all n≥1, the number of L-formulas with x1,…,xn free variables is finite, modulo T.

Corollary 16.10. Assuming that:

  • G an infinite group

  • Th(G) is ℵ0-categorical (in Lgroups)

Then G has finite exponent (there exists n∈ℕ such that ∀⁡g∈G, gn=1).

Fact: Any abelian group with finite exponent has an ℵ0-categorical complete theory.