Documentation

Linglib.Phenomena.Complementation.Studies.Cristofaro2013

Phenomena.Complementation.Studies.Cristofaro2013 #

@cite{noonan-2007} @cite{dryer-2013-wals}

Cross-linguistic complementation analysis anchored on Sonia Cristofaro's five WALS chapters on complement/subordinate clauses (Ch 124--128):

Ch 124A: 'Want' complement subjects (283 languages).
Ch 125A: Purpose clauses -- balanced vs deranked (170 languages).
Ch 126A: 'When' clauses -- balanced vs deranked (174 languages).
Ch 127A: Reason clauses -- balanced vs deranked (169 languages).
Ch 128A: Utterance complement clauses -- balanced vs deranked (143 languages).

The 19-language ComplementationProfile sample is the testbed for cross-chapter generalizations.

Cristofaro's central insight #

Complement clauses can be classified along a single binary dimension: balanced (retains main-clause morphology) vs. deranked (uses reduced/non-finite forms). This maps onto Noonan's finer typology: balanced ≈ indicative/subjunctive, deranked ≈ infinitive/nominalized/ participle. Different complement types differ systematically in how likely they are to be deranked: utterance complements resist deranking strongly, purpose clauses favor it.

NOTE: cristofaro-2013 has no entry in references.bib yet; cited via inheritance from Cristofaro's Ch 124--128 contribution to @cite{dryer-haspelmath-2013}.

Contents #

§1. The 19-language ComplementationProfile sample.
§2. Cross-chapter sample-grounded generalizations.

Out of scope #

The substrate types (BalancedDeranked, WantCompStrategy, ComplementationProfile, WALS converters) and corpus-only generalizations (utterance-mostly-balanced, purpose-more-deranked-than-utterance) live in Linglib/Typology/Complementation.lean. The 20-language Dryer subordination sample lives in Studies/Dryer2013.lean. Noonan's CTP per-verb data is in Studies/Noonan2007.lean.

def Phenomena.Complementation.Studies.Cristofaro2013.compEnglish :

Typology.Complementation.ComplementationProfile

English: equi-deletion with 'want'; balanced/deranked across purpose/when/reason; balanced utterance.

Equations

One or more equations did not get rendered due to their size.

Instances For

def Phenomena.Complementation.Studies.Cristofaro2013.compJapanese :

Typology.Complementation.ComplementationProfile

Japanese: desiderative suffix -tai; no independent 'want' CTP.

Equations

One or more equations did not get rendered due to their size.

Instances For

def Phenomena.Complementation.Studies.Cristofaro2013.compTurkish :

Typology.Complementation.ComplementationProfile

Turkish: strongly deranked (nominalized) across clause types.

Equations

One or more equations did not get rendered due to their size.

Instances For

def Phenomena.Complementation.Studies.Cristofaro2013.compHindi :

Typology.Complementation.ComplementationProfile

Hindi: not in the F125A sample; otherwise mixed.

Equations

One or more equations did not get rendered due to their size.

Instances For

def Phenomena.Complementation.Studies.Cristofaro2013.compMandarin :

Typology.Complementation.ComplementationProfile

Mandarin: isolating; uniformly balanced complements.

Equations

One or more equations did not get rendered due to their size.

Instances For

def Phenomena.Complementation.Studies.Cristofaro2013.compKorean :

Typology.Complementation.ComplementationProfile

Korean: balanced despite being agglutinative SOV.

Equations

One or more equations did not get rendered due to their size.

Instances For

def Phenomena.Complementation.Studies.Cristofaro2013.compGerman :

Typology.Complementation.ComplementationProfile

German: not in F128A sample; mixed otherwise.

Equations

One or more equations did not get rendered due to their size.

Instances For

def Phenomena.Complementation.Studies.Cristofaro2013.compRussian :

Typology.Complementation.ComplementationProfile

Russian: mixed balanced/deranked across subordinate clause types.

Equations

One or more equations did not get rendered due to their size.

Instances For

def Phenomena.Complementation.Studies.Cristofaro2013.compPersian :

Typology.Complementation.ComplementationProfile

Persian: overt complement subjects with 'want'.

Equations

One or more equations did not get rendered due to their size.

Instances For

def Phenomena.Complementation.Studies.Cristofaro2013.compIrish :

Typology.Complementation.ComplementationProfile

Irish: not in F124A sample; mixed otherwise.

Equations

One or more equations did not get rendered due to their size.

Instances For

def Phenomena.Complementation.Studies.Cristofaro2013.compBasque :

Typology.Complementation.ComplementationProfile

Basque: deranked purpose clauses like Turkish.

Equations

One or more equations did not get rendered due to their size.

Instances For

def Phenomena.Complementation.Studies.Cristofaro2013.compYoruba :

Typology.Complementation.ComplementationProfile

Yoruba: uniformly balanced (serial verb language).

Equations

One or more equations did not get rendered due to their size.

Instances For

def Phenomena.Complementation.Studies.Cristofaro2013.compTagalog :

Typology.Complementation.ComplementationProfile

Tagalog: V-initial Austronesian; deranked purpose.

Equations

One or more equations did not get rendered due to their size.

Instances For

def Phenomena.Complementation.Studies.Cristofaro2013.compNavajo :

Typology.Complementation.ComplementationProfile

Navajo: only in Ch 124A; overt complement subjects.

Equations

One or more equations did not get rendered due to their size.

Instances For

def Phenomena.Complementation.Studies.Cristofaro2013.compSwahili :

Typology.Complementation.ComplementationProfile

Swahili: only in Ch 128A.

Equations

One or more equations did not get rendered due to their size.

Instances For

def Phenomena.Complementation.Studies.Cristofaro2013.compArabic :

Typology.Complementation.ComplementationProfile

Arabic (Gulf): balanced+deranked across purpose/when/reason/utterance.

Equations

One or more equations did not get rendered due to their size.

Instances For

def Phenomena.Complementation.Studies.Cristofaro2013.compFinnish :

Typology.Complementation.ComplementationProfile

Finnish: negative auxiliary language; balanced+deranked across types.

Equations

One or more equations did not get rendered due to their size.

Instances For

def Phenomena.Complementation.Studies.Cristofaro2013.compSpanish :

Typology.Complementation.ComplementationProfile

Spanish: deranked purpose (para + INF); subjunctive complements.

Equations

One or more equations did not get rendered due to their size.

Instances For

def Phenomena.Complementation.Studies.Cristofaro2013.compFrench :

Typology.Complementation.ComplementationProfile

French: deranked purpose (pour + INF); subjunctive complements.

Equations

One or more equations did not get rendered due to their size.

Instances For

def Phenomena.Complementation.Studies.Cristofaro2013.allCompProfiles :

List Typology.Complementation.ComplementationProfile

The 19-language sample.

Equations

One or more equations did not get rendered due to their size.

Instances For

theorem Phenomena.Complementation.Studies.Cristofaro2013.comp_profile_count :

allCompProfiles.length = 19

theorem Phenomena.Complementation.Studies.Cristofaro2013.deranked_purpose_implies_at_least_mixed_when :

have derankedPurp := List.filter (fun (p : Typology.Complementation.ComplementationProfile) => p.purposeClause == some Typology.Complementation.BalancedDeranked.deranked) allCompProfiles; (derankedPurp.all fun (p : Typology.Complementation.ComplementationProfile) => p.whenClause == some Typology.Complementation.BalancedDeranked.balancedDeranked || p.whenClause == some Typology.Complementation.BalancedDeranked.deranked || p.whenClause == some Typology.Complementation.BalancedDeranked.balanced) = true

In the sample, languages with deranked purpose clauses tend to also have at least balanced/deranked 'when' clauses (purpose deranking is not isolated).

theorem Phenomena.Complementation.Studies.Cristofaro2013.sample_utterance_lean_balanced :

have withUtt := List.filter (fun (x : Typology.Complementation.ComplementationProfile) => x.utteranceComp.isSome) allCompProfiles; have balanced := List.filter (fun (p : Typology.Complementation.ComplementationProfile) => p.utteranceComp == some Typology.Complementation.BalancedDeranked.balanced || p.utteranceComp == some Typology.Complementation.BalancedDeranked.balancedDeranked) withUtt; balanced.length > withUtt.length / 2

Sample utterance complements lean balanced (consistent with Cristofaro's cross-chapter observation).

theorem Phenomena.Complementation.Studies.Cristofaro2013.sample_purpose_more_deranked_than_balanced :

have withPurp := List.filter (fun (x : Typology.Complementation.ComplementationProfile) => x.purposeClause.isSome) allCompProfiles; have pureDeranked := List.filter (fun (x : Typology.Complementation.ComplementationProfile) => x.purposeClause == some Typology.Complementation.BalancedDeranked.deranked) withPurp; have pureBalanced := List.filter (fun (x : Typology.Complementation.ComplementationProfile) => x.purposeClause == some Typology.Complementation.BalancedDeranked.balanced) withPurp; pureDeranked.length ≥ pureBalanced.length

Sample purpose clauses show more pure-deranked entries than pure-balanced (matching the WALS-aggregate trend in Typology.Complementation.purpose_more_deranked_than_utterance).

theorem Phenomena.Complementation.Studies.Cristofaro2013.sample_want_mostly_implicit :

have withWant := List.filter (fun (x : Typology.Complementation.ComplementationProfile) => x.wantComp.isSome) allCompProfiles; have implicitSubj := List.filter (fun (x : Typology.Complementation.ComplementationProfile) => x.wantComp == some Typology.Complementation.WantCompStrategy.subjectImplicit) withWant; implicitSubj.length * 2 > withWant.length

The 'want' complement subject is left implicit in the majority of sample languages (Cristofaro Ch 124A's headline pattern in our sample).

def Phenomena.Complementation.Studies.Cristofaro2013.noonanToCristofaro :

Typology.Complementation.NoonanCompType → Typology.Complementation.BalancedDeranked

Project @cite{noonan-2007}'s 6-way NoonanCompType onto Cristofaro's coarser BalancedDeranked typology. Noonan's reduced types (infinitive/nominalized/participle, captured by NoonanCompType.isReduced) correspond to Cristofaro's .deranked; finite types (indicative/subjunctive) correspond to .balanced. Paratactic falls on the balanced side: paratactic complements use main-clause morphology and behave like balanced finite clauses for Cristofaro's morphological test.

Equations

Instances For

theorem Phenomena.Complementation.Studies.Cristofaro2013.deranked_iff_reduced (t : Typology.Complementation.NoonanCompType) :

noonanToCristofaro t = Typology.Complementation.BalancedDeranked.deranked ↔ t.isReduced = true

The two typologies' "reduced" diagnostics agree by construction. NoonanCompType.isReduced t = true iff noonanToCristofaro t = .deranked. Promotes the docstring-only correspondence to a kernel-checked theorem.

theorem Phenomena.Complementation.Studies.Cristofaro2013.balanced_iff_not_reduced (t : Typology.Complementation.NoonanCompType) :

noonanToCristofaro t = Typology.Complementation.BalancedDeranked.balanced ↔ t.isReduced = false

Symmetric form: balanced iff non-reduced.