James E. Hanson
Abstract:We show that constructible models of arbitrary complete continuous first-order theories are unique up to isomorphism.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is: **Prove that in continuous logic, the constructible models of any complete continuous first - order theory are unique up to isomorphism**.
Specifically, the goal of the paper is to generalize the classical results in discrete logic to continuous logic. In discrete logic, Vaught's theorem shows that for a countable complete theory \(T\), the following conditions are equivalent:
- \(T\) has a prime model.
- \(T\) has an atomic model.
- For each \(n\), the atomic types in \(S_n(T)\) are dense.
Furthermore, in discrete logic, a countable model is a prime model if and only if it is an atomic model, and any two prime models are isomorphic, and each prime model is \(\aleph_0\)-homogeneous.
However, in the case of uncountable theories, the situation becomes complicated. Primal and atomic are no longer equivalent, and Shelah has provided several examples with non - unique prime models. Early positive results on uncountable theories mainly focused on \(\aleph_0\)-stable theories. Ressayre proved the uniqueness of a special type of prime model in an unpublished result, and this proof technique has been widely used in instability theory.
In continuous logic, Vaught's results are known to be generalizable, but the proof needs to be modified, and the prime model is only guaranteed to be "approximately \(\aleph_0\)-homogeneous", not fully \(\aleph_0\)-homogeneous. In addition, some classical results in continuous logic (such as Vaught's never - two theorem and Lachlan's theorem on the number of countable models of superstable theories) cannot be directly generalized, mainly because of the d - finiteness problem of types in continuous logic.
Therefore, the core problem of the paper is **to prove that in continuous logic, any two constructible models are isomorphic**, thus filling this gap in the field and expanding the existing theoretical framework.
### Main Conclusions
The main conclusion of the paper is:
\[
\text{Any two constructible models are unique up to isomorphism.}
\]
That is, for any complete continuous first - order theory \(T\), its constructible models are unique up to isomorphism. This conclusion is proved by introducing concepts such as augmented construction sequence and self - sufficient pair, and using backward induction and the properties of dense sets.
### Related Definitions and Lemmas
To prove this conclusion, the paper introduces several key definitions and lemmas, including but not limited to:
1. **Construction Sequence**: If for each \(i\), \(\text{tp}(b_i / A b_{<i})\) is atomic, then \((b_i)_{i < \alpha}\) is called a construction sequence over \(A\).
2. **Augmented Construction Sequence**: A construction sequence plus distance predicates, parameter sets, and language sets.
3. **Self - Sufficient Pair**: A construction - closed pair that satisfies specific conditions, such that its corresponding sub - structure is dense in the original model.
Through these definitions and lemmas, the paper finally proves that in continuous logic, any two constructible models are isomorphic, thus solving this problem.