Small models of every infinite size: the countable-symbol core (issue #11 unit 7a) #
Marker, Theorem 11.2, countable-symbol core: if φ has arbitrarily large models then for
every infinite κ it has a model of size EXACTLY κ realizing only countably many complete
L_{ω₁ω}-types (exists_small_model_of_hasArbLargeModels_countable_symbols).
The assembly keeps the identity of the local EM carrier throughout (it does NOT route through
the existential tailTemplateRealizable_of_localEMContext_cross, which forgets it): a source
model of size ≥ ℶ_{ω₁} feeds the schema term source (SchemaLocalEMSource.lean, the
mixin-satisfying intermediate that itself realizes φ and carries the pairwise-distinct
schemaSeq); over the highly order-transitive skeleton J of size κ
(HighlyTransitiveExistence.lean) the schema context's carrier then has: satisfaction of φ
(the stage-0 bridge realizes_stage0_sentence_of_skolemUniversal), size exactly
max ℵ₀ κ = κ (mk_carrier_eq, injectivity from schemaSeq_pairwise_ne), and countably many
realized types (lomega1omegaSmall on the localColim reduct, descended to the seed language
by Lomega1omegaSmall.of_expansion).
The countable-symbol small-model theorem (Marker, Theorem 11.2 core): a sentence with
arbitrarily large models has, at every infinite κ, a model of size exactly κ realizing only
countably many complete L_{ω₁ω}-types.