Documentation

Linglib.Phenomena.Complementation.Studies.Cacchioli2025

@cite{cacchioli-2025} — Empirical Data @cite{cacchioli-2025} #

Pure empirical data from @cite{cacchioli-2025} "The Syntax of Clausal Prefixes in Tigrinya." No theory imports — this file contains only observed patterns, grammaticality judgments, and co-occurrence restrictions.

Key observations #

  1. Four prefixes: zɨ-, kɨ-, kəmzi-, ʔay-...-n
  2. Complementary distribution: No two prefixes co-occur
  3. Verb class selection: The matrix verb determines which prefix appears
  4. Fixed linear order: Prefix always precedes the verbal complex
  5. Agreement asymmetry: kɨ- and ʔay-...-n take agreement; zɨ- and kəmzi- don't

The four clausal prefixes attested in Tigrinya.

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

      A grammaticality judgment for a prefix combination. true = grammatical, false = ungrammatical.

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

          No two prefixes can co-occur (complementary distribution).

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

            All prefix combinations are ungrammatical (complementary distribution).

            A selection datum: a matrix verb selects a particular prefix.

            • verb : String
            • verbGloss : String
            • verbClass : String
            • selectedPrefix : TigrinyaPrefix
            • example_ : String
            • grammatical : Bool
            Instances For
              Equations
              • One or more equations did not get rendered due to their size.
              Instances For
                Equations
                Instances For

                  Verb class selection data from @cite{cacchioli-2025}.

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

                    Agreement data: which prefixes take agreement suffixes.

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

                            Map CTP reality status to [±finite] selection.

                            Realis CTPs (utterance, knowledge, commentative,...) select [+finite] complements — indicative/realis clauses whose Fin head bears [+finite].

                            Irrealis CTPs (desiderative, manipulative, modal,...) select [-finite] complements — subjunctive/irrealis clauses whose Fin head bears [-finite].

                            Some classes are variable (perception can take both finite and non-finite complements), so they return none.

                            Equations
                            Instances For

                              Desiderative verbs select [-finite] (subjunctive/irrealis) complements.

                              Irrealis CTP classes always map to [-finite] when they have a value. The converse (realis → [+finite]) does not hold universally: phasal verbs are realis but take non-finite complements.

                              Cross-paper divergence theorem: phasal CTPs witness the failure of the realis ↔ finite correspondence. @cite{noonan-2007} §3.1.1 puts phasal complements in DTR, but the substrate's ctpRealityStatus classifies them as realis (event is asserted as actual). Cacchioli's feature-based analysis classifies the same complements as [-finite] (reduced). The contradiction makes visible a real Noonan- internal tension between the realis-as-asserted-fact criterion and the DTR → irrealis groupings of Table 2.3.

                              The witness: phasal is realis AND takes non-finite (irrealis-aligned) complements. Any unification forcing realis ↔ +finite would have to reclassify either Noonan's reality status of phasal or Cacchioli's finiteness assignment.

                              Knowledge verbs select [+finite] → predicts kəmzi- (factive).

                              Desiderative verbs select [-finite] → predicts kɨ- (subjunctive).

                              Commentative verbs select [+finite] → predicts kəmzi- (factive).

                              Manipulative verbs select [-finite] → predicts kɨ- (subjunctive).

                              The negative circumfix surfaces correctly for a sample verb.