Chuj Verb Building: Empirical Data and Bridge Theorems #

@cite{coon-2019}

Theory-neutral empirical data from @cite{coon-2019} "Building verbs in Chuj: Consequences for the nature of roots." Journal of Linguistics 55(1): 35–81.

Chuj is a Q'anjob'alan (Mayan) language spoken in Guatemala and Mexico. The data here encodes the paper's primary empirical observations about root classes, voice morphology, and argument structure, without committing to the theoretical analysis.

Data encoded #

Root classes (§§2–3): four morphosyntactic classes of roots (√TV, √ITV, √POS, √NOM), identified by their surface distribution.
Voice suffixes (Table 58/78): Ø, -ch, -j, -w with their morphological and distributional properties.
Paradigm grammaticality (§§2–5): which root×voice combinations are grammatical.
-aj distribution (§5): existential closure suffix tracks implicit arguments.
Agent diagnostics (§4.1–4.2): agent-oriented adverbs and by-phrases distinguish -ch from -j.
Example verbs with glosses, organized by root class.

Bridge theorems #

Chuj fragment bridge #

Connects the Chuj fragment (Fragments/Chuj/VerbBuilding.lean) to the empirical data.

Root class ↔ Root arity: The phenomena's CRootClass maps to the fragment's Root values. √TV = selectsTheme, others = noTheme.
Voice suffix ↔ VoiceHead: Each suffix maps to the fragment's VoiceHead, with matching properties (theta assignment, D feature, phase head status).
Paradigm predictions: The fragment's isGrammatical matches the data's paradigm attestation for all root×voice combinations.
-aj predictions: The fragment's hasImplicitExternal and triggersAj match the data's -aj distribution.
Agent diagnostics: The fragment's assignsTheta matches the data's agent adverb and by-phrase diagnostics.
Division of labor: The data's formsBareTransitive aligns with the fragment's arity distinction: only roots with selectsTheme form bare transitives.

Root typology bridge #

Connects the theory-side predictions of Theories/Morphology/RootTypology.lean (@cite{beavers-etal-2021} formalization) to the empirical data in Phenomena/Causatives/Studies/BeaversEtAl2021.lean.

Classification isomorphism: The theory's RootType and the phenomena's CoSRootClass are provably isomorphic — they describe the same partition.
Diagnostic alignment: The phenomena's semantic diagnostics (changeDenialTest, restitutiveAgainTest) agree exactly with the theory's Boolean correlates (entailsChange, allowsRestitutiveAgain).
Prediction ↔ attestation: The theory predicts PC roots HAVE simple statives and result roots LACK them; the empirical data confirms this (PC: 7/8 sample roots ≥ 50%; result: all 10 sample roots ≤ 10%).
Markedness prediction: The theory predicts PC verbs are marked and result verbs are unmarked; the statistical comparison confirms PC median (56.01%) exceeds result median (15.20%).
Fragment grounding: The Chuj fragment's Root values instantiate the theory's predictions — e.g., rootTV_res.entailsChange = true matches the theory's RootType.entailsChange.result = true.