Universals over Subject Assignments #
A SubjectAssignment is the abstract interface a syntactic theory
must provide to be evaluated against pro-drop / overt-PRO universals:
a function from a SubjectContext (locus of decision) to an
Exponent (null vs overt).
Stating the universals over this abstract function type — rather than
over a particular framework's inventory structure (DM vocabulary
items, HPSG pro-form lexical entries, LFG f-structure features) —
keeps the typological claims framework-neutral. Each framework
provides its own projection from its inventory type to a
SubjectAssignment; the universals here apply to all of them
uniformly.
Honest quantification #
hasOvertPRO is ∀ p ∈ thematicPersons, a (.controlled p) = .overt
(decidable by decide). allowsProDrop is ∃ p ∈ thematicPersons, a (.matrixFinite p) = .null. The Ostrove universal then has its
strongest content: if EVERY thematic person is overt in PRO position,
NO thematic person can be dropped in matrix position.
Per-person predicates allowsProDropAt/hasOvertPROAt expose the
finer grain that partial-pro-drop languages (Hebrew: 1/2 only;
Brazilian Portuguese: 3rd-person dependent; Finnish: 1/2 only) and
anti-agreement languages (overt only when in situ) require.
The Satisfies predicate is defined exactly once #
The two-Boolean ProDropProfile (in Basic.lean) is the projected
view; SubjectAssignment.Satisfies is defined as
(toProDropProfile a).Satisfies rather than re-stipulated. The bridge
is then a real reduction (loss of per-person information), not the
identity.
A language's null-vs-overt decision at every cell of the cross-classification. The abstract interface against which pro-drop / overt-PRO universals are stated.
Equations
Instances For
Per-person controlled-subject (PRO) overtness.
Equations
Instances For
Per-person pro-drop: the language drops a matrix finite subject
of person p.
Equations
Instances For
Per-person anti-agreement licensing: the language drops a matrix
finite subject of person p under Ā-extraction even when it
cannot drop the same subject in situ (@cite{baier-2018}).
Equations
- One or more equations did not get rendered due to their size.
Instances For
The language has overt PRO iff EVERY thematic person realizes controlled-subject position overtly. The honest quantified version of @cite{ostrove-2026}'s "overt PRO" parameter.
Equations
- a.hasOvertPRO = ∀ p ∈ Phenomena.NullSubject.thematicPersons, a.hasOvertPROAt p
Instances For
The language allows pro-drop iff SOME thematic person can be dropped in matrix finite position.
Equations
- a.allowsProDrop = ∃ p ∈ Phenomena.NullSubject.thematicPersons, a.allowsProDropAt p
Instances For
The language is partially pro-drop: it drops some thematic persons but not all. Hebrew (1/2 only), Brazilian Portuguese (3rd-person dependent), Finnish (1/2 only).
Equations
- a.isPartialProDrop = (a.allowsProDrop ∧ ¬∀ p ∈ Phenomena.NullSubject.thematicPersons, a.allowsProDropAt p)
Instances For
The language exhibits the anti-agreement effect: SOME thematic person triggers null subjects only under Ā-extraction.
Equations
Instances For
Bridge to the legacy two-Boolean profile in Basic.lean. The Bool
fields of ProDropProfile are typological observables (the data);
the abstract SubjectAssignment predicates are Prop, so we project
via decide.
Equations
- a.toProDropProfile = { allowsProDrop := decide a.allowsProDrop, hasOvertPRO := decide a.hasOvertPRO }