@cite{goldberg-jackendoff-2004}: The English Resultative as a Family of Constructions #
@cite{goldberg-jackendoff-2004}
Paper data and per-datum verifications for @cite{goldberg-jackendoff-2004}.
The construction-theoretic primitives (ResultativeSubconstruction,
SubeventDesc, the fusion machinery, the universal-quantifier theorems)
live in Theories/Syntax/ConstructionGrammar/Resultatives.lean, which
this file imports.
Key claims #
- The English resultative is not one construction but a family of four subconstructions organized along two dimensions: causative/noncausative × property/path RP
- Every resultative has a dual subevent structure: a verbal subevent (from the verb) and a constructional subevent (CAUSE + BECOME/GO from the construction)
- The verbal and constructional subevents are linked by typed semantic relations: MEANS, RESULT, INSTANCE, or CO-OCCURRENCE
- Full Argument Realization (FAR): all obligatory arguments of both verb and construction must be syntactically realized; shared arguments fuse
- Semantic Coherence: verb role rV and construction role rC may fuse only if rV is construable as an instance of rC
- Aspectual constraint: resultatives are telic iff the RP denotes a bounded path/property
- Temporal constraint: the constructional subevent cannot temporally precede the verbal subevent
File contents #
This file holds:
- The paper's eight resultative entries (
hammerFlat,kickIntoField, …) and theallEntrieslist — concrete data points the paper discusses. - Per-datum verification theorems demonstrating the paper's data is consistent with the construction theory.
- A separate empirical-data layer (
ResultativeType,ResultativeDatum,allExamples,aspectualContrasts) holding theory-neutral grammaticality judgments and aspectual contrasts drawn from §§2–8 of the paper.
Empirical data: resultative entries #
Equations
- One or more equations did not get rendered due to their size.
Instances For
Equations
- One or more equations did not get rendered due to their size.
Instances For
Equations
- One or more equations did not get rendered due to their size.
Instances For
Equations
- One or more equations did not get rendered due to their size.
Instances For
Equations
- One or more equations did not get rendered due to their size.
Instances For
Equations
- One or more equations did not get rendered due to their size.
Instances For
Equations
- One or more equations did not get rendered due to their size.
Instances For
Equations
- One or more equations did not get rendered due to their size.
Instances For
All resultative entries.
Equations
- One or more equations did not get rendered due to their size.
Instances For
Per-datum verification theorems #
All four core subconstructions use MEANS (§3, summary 97a–d). RESULT is reserved for sound-emission and disappearance subconstructions.
Causative entries have CAUSE in their derived constructional subevent.
Noncausative entries lack CAUSE in their derived constructional subevent.
All derived constructional subevents have BECOME.
Noncausative (intransitive) entries have no object selection.
All causative entries specify an object selection mode.
All entries with bounded RP are telic.
Theorems migrated from Theories.Semantics.Causation.Resultatives #
These theorems quantify over allEntries (paper-specific data) and
therefore belong with the paper, not in the Theory layer.
All causative entries in the data have CAUSE.
MEANS-relation causative entries all have CAUSE.
Activity verbs in the data with bounded RPs become accomplishments.
All resultative entries have BECOME.
Per-entry verb class participation #
The construction-verb interaction across the four canonical Levin classes that @cite{goldberg-jackendoff-2004} discuss: manner verbs (hit = hammer/kick), change-of-state verbs (otherCoS = freeze), motion verbs (mannerOfMotion = roll/swing), and removing verbs (wipe).
All entries acquire CoS from the construction, regardless of verb class.
All entries participate in the resultative alternation (none are instrument-spec).
Causative entries all acquire the causative alternation.
Noncausative entries do NOT acquire the causative alternation (unless the verb already has causation — freeze/otherCoS does).
Hammer (hit-class): no CoS or causation alone → both added by causative construction.
Freeze (otherCoS): already has CoS + causation → construction doesn't change profile.
Roll (manner-of-motion): gains CoS from construction; no causation (noncausative).
Laugh (performance): pure manner verb — construction adds CoS + causation.
Empirical data: grammaticality judgments #
Theory-neutral grammaticality judgments and aspectual contrasts drawn from §§2–8 of the paper. These provide the shared data layer that other studies (Dendikken, Tay, Levin) connect to their own analyses.
What type of resultative is exemplified.
Extends the paper's 2×2 matrix (§2) with fake reflexives (§5) and anticausative property resultatives (@cite{levin-2026}).
- causativeProperty : ResultativeType
- causativePath : ResultativeType
- noncausativeProperty : ResultativeType
- noncausativePath : ResultativeType
- fakeReflexive : ResultativeType
- anticausativeProperty : ResultativeType
Anticausative: verb doesn't alternate alone; construction licenses it (@cite{levin-2026}). Distinct from
noncausativeProperty(e.g., freeze solid) where the verb independently shows the causative alternation.
Instances For
Equations
- One or more equations did not get rendered due to their size.
Equations
- One or more equations did not get rendered due to their size.
Instances For
Equations
- Phenomena.Constructions.Resultatives.Studies.GoldbergJackendoff2004.instDecidableEqResultativeType x✝ y✝ = if h : x✝.ctorIdx = y✝.ctorIdx then isTrue ⋯ else isFalse ⋯
A single resultative example with judgment data.
- exId : String
Example identifier
- sentence : String
The sentence
- judgment : Features.Acceptability
Acceptability judgment
- resType : ResultativeType
Which resultative subtype
- phenomenon : String
What phenomenon this illustrates
Instances For
Equations
- One or more equations did not get rendered due to their size.
Equations
- One or more equations did not get rendered due to their size.
Instances For
Equations
- One or more equations did not get rendered due to their size.
- Phenomena.Constructions.Resultatives.Studies.GoldbergJackendoff2004.instBEqResultativeDatum.beq x✝¹ x✝ = false
Instances For
Equations
- One or more equations did not get rendered due to their size.
Causative property resultatives (§2, ex. 5a, 7a) #
Equations
- One or more equations did not get rendered due to their size.
Instances For
Equations
- One or more equations did not get rendered due to their size.
Instances For
Equations
- One or more equations did not get rendered due to their size.
Instances For
Causative path resultatives (§2, ex. 5b) #
Equations
- One or more equations did not get rendered due to their size.
Instances For
Equations
- One or more equations did not get rendered due to their size.
Instances For
Noncausative property resultatives (§2, ex. 6a) #
Equations
- One or more equations did not get rendered due to their size.
Instances For
Equations
- One or more equations did not get rendered due to their size.
Instances For
Noncausative path resultatives (§2, ex. 6b) #
Equations
- One or more equations did not get rendered due to their size.
Instances For
Equations
- One or more equations did not get rendered due to their size.
Instances For
Fake reflexive resultatives (§5, ex. 9) #
Equations
- One or more equations did not get rendered due to their size.
Instances For
Equations
- One or more equations did not get rendered due to their size.
Instances For
Equations
- One or more equations did not get rendered due to their size.
Instances For
Aspectual contrasts (§4, Principle 27) #
An aspectual contrast pair.
- sentence : String
Sentence with temporal adverbial
- judgment : Features.Acceptability
Acceptability
- adverbialType : String
Which adverbial type
- description : String
Description
Instances For
Equations
- One or more equations did not get rendered due to their size.
Equations
- One or more equations did not get rendered due to their size.
Instances For
Equations
- One or more equations did not get rendered due to their size.
Equations
- One or more equations did not get rendered due to their size.
- Phenomena.Constructions.Resultatives.Studies.GoldbergJackendoff2004.instBEqAspectualContrast.beq x✝¹ x✝ = false
Instances For
Equations
- One or more equations did not get rendered due to their size.
Instances For
Equations
- One or more equations did not get rendered due to their size.
Instances For
Equations
- One or more equations did not get rendered due to their size.
Instances For
Equations
- One or more equations did not get rendered due to their size.
Instances For
Unacceptable resultatives (§6, semantic coherence violations) #
Equations
- One or more equations did not get rendered due to their size.
Instances For
Equations
- One or more equations did not get rendered due to their size.
Instances For
Aggregate data #
Equations
- One or more equations did not get rendered due to their size.
Instances For
Equations
- One or more equations did not get rendered due to their size.
Instances For
Empirical verification #
All four resultative types are attested in the data.
Both grammatical and ungrammatical examples are represented.
The aspectual contrast data includes both in- and for-adverbials.
Telic resultatives accept in-adverbials and reject for-adverbials.
Atelic bare activities accept for-adverbials.