[Rud25b]: Embedded Intonation and Quotative Complements #

Rudin, Deniz (2025/2026). "Embedded Intonation and Quotative Complements to Verbs of Speech." Linguistic Inquiry, early access. doi:10.1162/ling.a.554.

Empirical Generalizations #

The paper's central observation: verbs of speech systematically split on whether they accept rising-declarative ("quotative") complements:

Verb	"p"	"p?"	"p" loud	"p" whisp	that p	that wh / Q
say	✓	✓	✓	✓	✓	✗
assert	✓	✗	✓	✓	✓	✗
yell	✓	✓	✓	✗	✓	✗
whisper	✓	✓	✗	✓	✓	✗
ask	✗	✓	✗	✗	✗	✓

Architecture: One Definition, Not Three #

Following mathlib practice, this file has no parallel formalizations.

Felicitous M V c is the single, model-parameterized definition of felicity: a complement is felicitous with a verb predicate iff there exists a witness (event + performance/proposition) with the right ontological properties.
IsRudinModel M is a class with 30 fields, one per cell. This is the statement of [Rud25b]'s empirical claim — any model is tested against it.
rudinModel is the concrete SpeechVerbs ℕ Bool (FBPerformance Bool) (fbOntology Bool) instantiation — Farkas-Bruce-grounded, with verb predicates defined as the postulate RHS so the meaning postulates hold by rfl.
instance : IsRudinModel rudinModel discharges the 30 cells from the postulates + FB ontology axioms.

There is no separate empirical : Verb → Complement → Felicity function and no separate predicted decision function. The empirical matrix and its derivation are the same proposition.

Farkas-Bruce Performance Ontology Bridge #

[Rud25b] [FB10]

Provides a PerformanceOntology instance whose Commits and RaisesIssue are derived from Farkas-Bruce discourse-state updates, rather than stipulated as primitive properties.

The Bridge #

A performance in F&B terms is a discourse-state update determined by its sentence form (declarative/interrogative), its propositional content, and its prosodic profile (rising or not, loud/whispered/ neutral). The FBPerformance record bundles exactly the data needed to compute its discourse effect:

a non-rising declarative asserts its content (adds to dcS, pushes issue)
an interrogative polarQuestions its content (pushes issue, no dcS commit)
a rising declarative pushes its content as an issue (no dcS commit) — the intermediate prosodic case that drives [Rud25b]'s empirical engine

Commits and RaisesIssue are then F&B-grounded predicates: a performance Commits iff its update adds its content to dcS; it RaisesIssue iff its update grows the table. Verb-class meaning postulates in SpeechVerbs see the same Commits / RaisesIssue that the F&B bridge theorems (in Discourse/Commitment/Table.lean) reason about — the connection is true by construction, not provable as an equivalence.

Why this matters #

Without the bridge, Commits is an axiomatic property of performances in PerformanceOntology — we'd have to say that rising declaratives don't commit. With the bridge, the F&B update semantics makes them not commit (the update doesn't touch dcS), and the Demonstration postulates inherit that fact directly.

Anti-correspondences #

A FBPerformance whose lingmat field is false and rising is false represents a non-linguistic performance (e.g., karate gestures). We choose LingMat to disjoin lingmat = true ∨ rising = true so that every rising-declarative performance is automatically linguistic material — a structural rather than axiomatic fact.

The Volume enumeration (neutral, loud, whispered) makes loud_not_whispered true by construction: a single field cannot simultaneously be both values.