Documentation

InfinitaryLogic.Admissible.Barwise.HenkinClosure

The Henkin closure of a set of formulas #

The producer side of the source-fragment adapter (issue #19): a closure operator that turns any set of formulas into a fragment with a Fragment.HenkinBasis, so that the v4.5.0 endpoint Fragment.exists_countable_model_of_aconsistent_withConstants applies without supplying the basis by hand.

henkinBasisSeed L   -- falsum, the equality template, every relation template
henkinClosure S     := negationClosure (S ∪ henkinBasisSeed L)

Fragment.negationClosure (generic, Lomega1omega/NegationClosure.lean) supplies component and negation closure; the seed supplies the atoms. Countability needs the relation-symbol sigma countable in addition to S, because the seed contains one template per symbol. The exact HF regression henkinClosure (hfFragment L).toSet = hfFragment L fixes the operator on the finitary fragment.

The construction is purely syntactic externally. Admissibility enters only when internalizing that construction and showing its codes remain inside the admissible language. No fragment structure gains a field.

The seed #

The Henkin basis seed: falsum, the equality template at arity two, and the relation template of every symbol at its arity.

Equations
  • One or more equations did not get rendered due to their size.
Instances For

    A fragment with a Henkin basis contains the seed.

    The closure #

    theorem FirstOrder.Language.Fragment.henkinClosure_le {L : Language} {S : Set ((n : ℕ) × L.BoundedFormulaω Empty n)} {A : L.Fragment} (hSA : S ⊆ A.toSet) (hB : A.HenkinBasis) :

    The Henkin closure is below every fragment with a basis containing S.

    A fragment with a basis is its own Henkin closure.

    Countability: the seed contributes one template per relation symbol.

    The HF regression, exact: the finitary fragment is its own Henkin closure.

    The endpoint with the basis discharged #

    The constants-expanded universe of a Henkin closure is Henkin-closed.

    Countable model existence over the Henkin closure of a countable set. The v4.5.0 source-fragment endpoint with the basis discharged by the closure. The consistency hypothesis is over the constants-expanded universe of the full closure, a larger permission set than the minimal interface HenkinClosedMin needs: the closure conveniently produces full closure, and the weaker interface permits weaker evidence, but nothing here transports consistency from a smaller universe to this one.