On the emergence of finite structures from attributive (non-finite) constructions : Evidence from to ihikeru in classical Japanes

1.Introducti

on

NominalistHypothesisisthatnon-finitenominalizationconstructionsfrequentlydevelopinto stand-alonefiniteclauses(seeStarosta,Pawley&Reid1982,Kaufman2009,Yap,Grunow-Harsta& Wrona2011).InthelongrecordedhistoryofJapaneselanguage,spanningfrom the8th_centurytothe

present,-ruhasbeenoneofthenominalizationconstructions.Therewasacleardistinctionbetween attributiveandconclusiveformsin'say'constructionssuchastoihikeru(attributive)andtoihikeri (conclusive)contrast.Theseformsarecomposedofacomplimentizerto,converbalconnectiveform of 'say'ihi,andperfectiveauxiliaryverb;keruisattributiveform andkeriisconclusiveform ofthe auxiliaryverb.(1)and(2)areexamplesoftoihikeriandtoihikerurespectively.

attri

buti

ve(

non-fi

ni

te)constructi

ons:

Evi

dencefrom

toihikeruinclassicalJapanese

Mi

zuhoTAMAJI

Abstract

Itisknownthatnon-finitenominalizationconstructionsfrequentlydevelopintostand-alone finitestructures.IntheclassicalJapanese,thetoihikeru/toihikeriformsshowsaclearattri bu-tive/conclusivecontrastinthe10th_{century.However,theattributiveform beguntodevelopinto}

conclusiveusesshowingsignsofblurredattributive/conclusivedistinction,theconclusiveformto ihikeridisappearedinthe17th_{century,andultimatelytothedemiseofthekakarimusubisystem}

asawhole.ThispaperexamineshowattributiveconstructionsinJapanesearereanal yzedascon-clusiveconstructionsthroughanalysisoftheprocessthatattributivetoihikerudevelopedinto conclusiveuses.Ouranalysisrevealsthreedevelopmentalpathwaysthatattributiveform devel -opedintoconclusivestructure,namelykakarimusubisystem,copularcleftconstructionsand converbaluse.Insubordinationplayedacrucialroleonthedevelopmentoffinitestructureand eachpathwayhasdifferentinsubordinationstrategies.

Italsorevealsthatthedemiseofkakarimusubifocussystem andcopularuseduetocleft constructionscontributedtothereanalysisofattributivetoihikeruform asconclusiveuses. Crucially,from atypologicalperspective,thisextensionprovidesadditionalinsighti ntostrate-giesbywhichrelativizationandnominalizationconstructionsdevelopintofiniteclauses.

Keywords:attributive/conclusivedistinction,kakarimusubisystem,copularcl eftconstruc-tions,conclusiveuse

(1) "Koko yai doko" to ihikereba, Here FOC where COMP say.COND.PRF "Tosa no Tomari" to ihikeri.

PN GEN PN COMP say.PRF

'If(I)asked"Whereisitnow?",(he)said"(Itis)TosanoTomari."(TosaNikki,p.38,935) (2) Mukashi A be no Nakamaro to hikeru hito wa,

Ancient.time PN GEN PN COMP say.ATTR.PRF person TOP Morokoshi ni watarite kaeri ki kerutokini,

PN LOC across.CONV return PST A TTR.PRFtime

'Onceuponatime,whenapersoncalledAbenoNakamarowenttoChinaandreturned(toJapan), .' (TosaNikki,p.37,935)

WhileclassicalJapanesehadaclearattributive/conclusivecontrast,modernJapanesedoesnot havesuchcontrastand-ruisusedforbothattributiveandconclusiveforms.Itisthoughtthata blurring theattributive/conclusivecontrastduetothedemiseofkakarimusubifocussystem broughtdisappearanceofconclusiveform (seeOno1993).

KakarimusubiisafocusconstructioninclassicalJapanese,whichinvolveduseoffocusparticles suchaszo,mamuandyaandthepresenceoftheseparticlesrequiredattributiveform inthesentence finalposition.(3)istheexampleofkakarumusubi.

(3) hashi wo yattsu watseru niyorite bridgeACC eight stretch.acrossATTR.CAUS because namu yatsuhashi to ihikeru.

FOC eight.bridges COMP say.ATTR.PRF

'Because(we)stretcheightbridgesacross(theriver),thatiswhywecall(theplace)Yatsuhashi (i.e.EightBridges).' (IseMonogatari,p.116,10th_c.)

Inthisexample,'Yatsuhashitoihikeru(thatissaidtobeyatsuhashi)'isthefocusedelement.The clausebeforethefocusparticlenamu'hashioyattsuwataseruniyorite(because(we)stretcheight bridgesacross(theriver))'istherelativeclause,namely,nominalizedconstructionthatmodifiesthe focusedelement.

Thedemiseofkakarimusubisystem isconsideredasatriggerofthedisappearanceofconclusive form showingsignsofblurredattributive/conclusivedistinction.However,itisdifficulttoexplain whyattributiveform developedintofinitestructurebythedemiseofkakarimusubisystem.This studyexamineshow attributiveconstructionsinJapanesearereanalysedasconclusi veconstruc-tions－inotherwords,how noun-modifyingconstructionsinitiallyusedinthereferentialdomain arereinterpretedwithbroaderscopeasstand-alonefinitestructuresthatrepresententireevents, situationsorpropositions.Italsoexaminesthisreanalysisofattributiveconstructionsasconclusive constructionsledtoablurringoftheattributive/conclusivedistinctionthattriggeredthedemiseof

kakarimusubi.Theorganizationofthisstudyisasfollows:section2explainsdataandmethodology employedinthisstudy,section3analysestheprocessthatattributivetoihikerudevelopedi ntocon-clusivestructure,section4isconclusion.

2.Dataandmethodol

ogy

Diachronicdataforouranalysiscomefrom theTaikeiHonbunDatabase,theelectronicversionof NihonKotenBungakuTaikei['TheCollectionofJapaneseClassicalLiteraryTexts'].Thedatabase consistsof466,574wordsfrom 556textscomprisingpoems,historicaldocuments,ficti onalnarra-tives,essays,kyoogenorcomedydramascripts,andnovelsfrom the8th_to19th_{century(beforethe}

Meijiperiod).265tokensoftoihkeruconstructionswereextractedandcategorizedaccordingtotheir functionsovertime.

3.Anal

ysi

softheprocessthatattri

buti

ve

toihikerudevelopedintofi-ni

testructureandthemechani

smsi

nvol

vedi

nreanal

ysi

s

InTaikeiHonbunDatabase,bothtoihikeruandtoihikeriwereattestedasearlyas10th_century.

Whiletoihikeruhadbeenattestedfrom 10th_to19th_{century,toihikerihadbeenattestedfrom 10}th_to

17th_{century.Thisimpliesthatdevelopmentofatrributivetoihikeruintoconclusiveuseresultedin}

disappearanceofconclusivetoihikeri.

Theexample(4)below isthefirsttokenofattributivetoihikeruattestedinTaikeiHonbun Database,whichisintroducedasexample(2)intheprevioussection.Toihikeruhereisaccompanied byaheadnoun(hito:person),whichmakestoihikeruaheadedrelativeclauseconstruction.Since theheadnoun(hito)isageneralnoun,toihikeruisreanalysedasanew nominalizer.Thisenables -runominalizertobereanalysedasanattributivemarker(functionallyequivalenttotheEnglish relativizerthat).

(4) Mukashi Abe no Nakamaro to ihikeru

Ancient.time PN GEN PN COMPsay.ATTR.PRF hito wa, Morokoshi ni watarite kaeri ki person TOP PN LOC across.CONV return PST keru tokini,

ATTR.PRF time

'Onceuponatime,whenapersoncalledAbenoNakamarowenttoChinaandreturned(to Japan), .' (TosaNikki,p.37,935)

Theexample(5)isattributivetoihikeruwithoutaccompaniedbyaheadnoun,butiti saccompa-niedbyatopicmarkerwa.'Sokowoyatsuhashitoihikeruwa'means'thatwhichpeoplecalledthere yatsuhashiis'.Thisindicatesthattoihikeruhereisareferentialuseintopicpositionandaheadless relativeclausestructure.

(5) soko o yatsuhashi to ihikeru wa, tthereACC PN COMP say.ATTR.PRF TOP mizu yuku kawa no kumote nareba, hashi water go river GEN spider.hand be.COND bridge no kumote nareba, hashi wo yattsu wataseru

GEN spider.hand be.COND bridge A CC eight stretch.across.CAUS niyorite namu yatsuhashi to ihikeru.

because FOC PN COMP say.ATTR.PRF

'Because(we)stretch eightbridgesacross (theriver),thatiswhy wecall(theplace) Yatsuhashi(i.e.EightBridges).'(IseMonogatari,p.116,10th_c.)

Thereanalysisof-runominalizerasanattributivemarkerandtheemergenceofheadlessrelative structureplayedcrucialrolesontheemergenceofstand-aloneconclusiveconstructionoftoihikeru. (6)istheexampleofstand-aloneconclusiveconstructionoftoihikeruattestedin19th_century.

'Aruhitoiu(Assomeonesaid)'isintroducingquotationand'toifu(saidthat)'ismarkingendofthe quotation,thereforetoihikerulocatesinthesentencefinalpositionofthequotation.

(6) Aruhito iu Ooko wa jooge to wakarete, someone say PN TOP up.down COMP separate.CONV Asakusa Tennoochoo no atari yori Senju no

PN PN GEN around from PN GEN

hashi giwa made wo subete Sendukagoo toihikeru to ifu. bridge edge up.to A CC all PN EVID COMP say

SomeonesaidthatOokowasdividedintouptownanddowntownand(theplace)fromAsakusa TennoochoouptotheedgeofSenjuhashibridgeiscalledSendukagoo.

(EdoMeishoZue,1836,p.86)

However,suchstand-aloneconclusiveconstructiondidnotevolvedirectlyfrom attributi vecon-structionoftoihikeru.Ourdiachronicanalysisrevealsthreedevelopmentalpathwaysfrom attri bu-tiveconstructiontothestand-aloneconclusiveconstruction,kakarimusubisystem,copularcleft constructions(Yap,Grunow-Harsta& Wrona2011)andconverbaluse.Furthermore,syntactic reanalysisisalsonecessaryfortheattributiveconstructioninthesentence-medialpositi ontode-velopintoconclusiveconstructioninthesentence-finalposition.Themechanism involvedinsuch syntacticreanalysisisknownasinsubordination(Evans2007).Therefore,wewilldiscussthreedevel -opmentalpathwaysandinsubordinationstrategiesofeachpathwayonthedevelopmentofattri bu-tivetoihikeruintoconclusiveconstruction.

3.1.Pathwayofkakarimusubisystem

Aswehavealreadyintroduced,(3)isthefirstexampleofkakarimusibiattestedinthe10th_century.

zoislocatedbetweencomplementizertoandihikeru.Example(8)isanexampleattestedinthe17th

century.Thefocusparticlezoisinthesentencefinalpositionfollowingtoihikeru.Si ncefocuspar-ticlescanoccurinvarioussyntacticposition,zocanalsoappearinthesentencefinalposition.This indicatesthatzodevelopedintoasentence-finalparticle.

(3) hashi wo yattsu watseru niyorite bridgeACC eight stretch.acrossATTR.CAUS because namu yatsuhashi to ihikeru.

FOC eight.bridges COMP say.ATTR.PRF

'Because(we)stretcheightbridgesacross(theriver),thatiswhywecall(theplace)Yatsuhashi (i.e.EightBridges).' (IseMonogatari,p.116,10th_c.)

(7) Hana ohokinite, akakarikereba, 'Oohana no Kuroodoegoo' nose big.CONV red.COND.PRF PN GEN PN

atozamani wa koto nagashi tote, later TOP quite long because 'Hanakuroodo' to zo ihikeru.

PN COMP FOC say.ATTR.PRF

'(Hehad)abigandrednose,thereforepeoplecalledhim 'OohananoKuroodoegoo'.Sinceitis quitelong,peoplejustcalledhim 'hanakuroodo'lateron.'

(UjishuuiMonogatari,13th_c.,p.315)

(8) "Nanji nani no yuhe o motsute ka You what GEN reason ACC have.CONV Q furo ni wa hito hitori to ihikeru

bath LOC TOP person one.person COMP say.ATTR.PRF zo" to tohi tamaheba,

SFP COMP ask HON.COND

'Ifheasked,"Whydidyousaythatweshouldtakeabathonebyone?' (IsohoMonogatari,17th_c.,p.368)

Theseexamplesshowspositionalshiftoffocusparticles.Theseparticlesshiftedtowardsright (sentencefinalpositioninSOV languagesuchasJapanese),whichisregardedasrightperiphery. Suchpositionalchangealsoindicateschangeofthescopeoverthefocusedel ements.Ifthefocuspar-ticlesmovestowardsthesentencefinalposition,nominalizedconstructioninreferentialdomainis reanalyzedwithbroaderscopeasnon-referentialfinitestructurethatrepresententirepropositions. Suchshiftisconsideredasinsubordination,becausefocusconstructioninvolvingnominalizedclause isreanalyzedasanewmainclause.Thismakesfocusparticlesinthesentencefinalpositionoptional, whichconsequentlyledtotheemergenceofstand-alonetoihikeruasseeninexample(8).Thus,these suggeststhatinsubordination strategy inkakaru musubisystem isrightperiphery offocus

particlesandchangethefocusparticlesintosentencefinalparticles.Theseexamplesalsoshowsthat rightdislocationofthefocusparticlesblurredattributive/conclusivecontrastandconsequentlyled tothedemiseofkakarimucubisystem.

3.2.Pathwayofcopularcleftconstruction

Copularcleftconstructionisacleftconstructionsupportedwithcopular.Example(9)isthefirst exampleofsuchconstructionattestedinthe10th_{century.Itisoriginallyconsideredtobeasimple}

sentencelike'KonohitononawaOchikubonokimi(Thisperson'snameisPrincessOchikubo)'. However,itturnedouttobeacleftconstructioninordertoemphasizethepart'Ochikubonokimi towa(theonewhoissaidtobePrincessOchikubo)'.Nowthat'Ochikubonokimitowa'i sasubor-dinatedclause,anditisfollowedbyamain-clause.Correspondingtononimalizationofthesubordi -natedclause,main-clausewasalsonominalizedas'konohitononaoihikerunarikeri(shouldbesaid thatthisperson'sname)'.Then,suchattributiveconstructionofVPtoihikerunarikeri(shouldbe saidthatVP)isreanalysedasconclusiveconstruction(shouldbethat(NP)VP)inthesentencefinal positioninthemain-clause.

(9) Ochikubo no kimi to wa kono hito no PN GEN princess COMP TOP this person GEN

na o ihikeru nari keri.

name ACC ay.ATTR.PERF COP COP

'ItissaidtobethispersonwhoiscalledPrincessOchikubo.' (Ochikubo Monogatari,10th_c.,p.233)

Example(10)isanexampleattestedinthe17th_{century.Thestructureofthissentenceis'Isoho}

mooshikeruwaVPtoihikerunari(whatIsohosaidisVP)',thereforethisisanexampleofacleft construction.

(10) Isoho mooshikeru wa, 'Ima yori nochi wa PN say.HMB.ATTR.PRF TOP now from later TOP

onyurushi nashi totemo, gofudai no tokoro HON.permission NEG CONCESS hereditary.Daimyoo GEN place o ba yurusare moosu beshi' to ihikeru nari. ACC FOC permit.PASS say.HMB should COMP say.ATTR.PRF COP

(Literally:'WhatIsohosaidisthatitshouldbeallrightforustovisithereditaryDaimyoo's placewithoutpermissionfrom nowon.')

'Isohosaid,"itshouldbeallrightforustovisithereditaryDaimyoo'splacewithoutpermi s-sionfrom nowon.(IsohoMonogatari,17th_c.,p.372)

Comparedtoexample(9),thenumberofcopularsupportingtoihikeruinthesentencefinalposi -tiondecreasedinexample(10).Thisimpliestheelisionofcopularinthecleftconstruction,whichfi -nallyledtostand-alonesentencefinaltoihikeru.Itisalsoconsideredthattheelisionofcopularis

insubordinationstrategytoreanalyzeattributivetoihikeruasconclusivestructureincaseofthe cleftconstruction.

3.3Pathwayofconverbaluse

Inthe13th_{century,attributivetoihikerubegantobeusedinconverbalpositionasseeninfoll}

ow-ing examples.Attributivetoihikeruisaccompaniedby accusativemarkerwoin example(11), whereasitisnotaccompaniedbyaccusativemarkerinexample(12).Thesetoihikerufunctionsas clauselinkerwithorwithoutparticles,hencetheseareconsideredasconverbaluse.

(11) Tazunetohikeru naka ni, "Tama wo ya kafu" ask inquire.ATTR.PRF among LOC jewel A CC FOC buy

to ihikeru o, kikiru hito mo nakari

COMP say.ATTR.PRF ACC hear.ATTR person FOC NEG.be

keru ni,

PRF:ATTR CONCESS

'Oneofthosepeoplesaid,"Whydon'tyoubuysomejewels?",butnobodypaidattentionto him.'(UjishuuiMonogatari,pp.395,13th_c.)

(12) Narikata o meshite, "Fue esasemu PN A CC call.CONV flute get.CAUS.VOL

to ihikeru, hoi nari to yorokobite,

COMP say.ATTR.PRF real.intention COP COMP to.be.pleased

'(He)calledNarikataandsaid"SayingthatIwouldletyougetafluteismyrealintention", andhebecamehappy.(Jikkinshoo,1215,p.116)

Theclausescontainingtheseconverbalusesissubordinatedclause,andthereareusuallymain clausesfollowingthem.Ifthemainclausesareelidedduetopragmaticreasons,theformersubordi -natedclausesarereanalyzedasnewmain-clauses.Thistriggersreanalysisofconverbaluseofthese attributiveformsassentencefinalstructurewithpragmaticfunctions.Thisconsequentlyenabledto ihikeruestablishitselfasstand-alonefinitestructure.

4.Concl

usi

on

Thus,wehavebeendiscussinghowattributivetoihikerudevelopedintoconclusivestructureusing diachroniccorpus.Thesurveyrevealsthattherearethreepathwaysforattributi veformstoun-dergobeforeitdevelopedintotheconclusivestructure,kakarimusubisystem,copularcl eftconstruc-tionandconverbaluses.Thisstudyalsorevealsthatindubordinationofthecomplementclauseinto anew main clausecontributedtoreanalysisofattributivetoihikeru asconclusivestructure. Differenttypesofelisionfacilitatetheinsubordinationofthecomplementclause:elisi onofthesen-tencefinalparticlesforkakarimusubisystem,elisionofcopularforcopularcleftconstructionsand

elisionofmain-clauseforconverbaluse.Itisconsideredthatsuchvarietyofdevelopmentalpathways andinsubordinationstrategiesistypologicalfeatureofnominalizationinJapaneselanguage.

Itwasconsideredthatdemiseofkakarimusubisystem contributedtothereanalysisofattributive form asconclusivestructure(Ono1993).Thisstudy,however,revealsthatthedemiseofkakari musubisystem wasaffectedbyachainofeventsinwhichattributivetoihikerualsocametobeused asconclusivestructures.From atypologicalperspective,itisworthnotingthattheextendeduseof attributive forms as conclusive ones provides additionalinsight into strategies by which relativizationandnominalizationconstructionsdevelopintofinitestructures(seeDeLancy2011).

References

DeLancey,Scott.2011.Finitestructuresfrom nominalizationconstructionsinTibeto-Burman.In NominalizationinAsianLanguages:DiachronicandTypologicalPerspectives,FoongHaYap, KarenGrunow-Hrsta& JanickWrona(eds),pp.343-359.Amsterdam:JohnBenjamins. Evans.N.2007.'Insubordinationanditsuses',inIrinaNikolaeva(ed.),Finiteness:Theoreticaland

EmpiricalFoundations,OxfordUniversityPress,NewYork,pp.366-431.

Kaufman,D.2009.Austronesian typology and the nominalist hypothesis.InAustronesian HistoricalLinguisticsandCultureHistory:A FestschriftforRobertBlust,AlexanderAdelaar & A ndrewPawley(eds).Canberra:PacificLinguistics.

Ono,S.1993.Kakarimusubinokenkyuu(A studyofKakarimusubi).Tokyo:IwanamiShoten. Starosta,S.,A.K.Pawley,andL.A.Reid.1982.TheevolutionoffocusinAustronesian.InPapers

from theThirdInternationalConferenceonAustronesianLinguistics,Volume2:Trackingthe Travelers,AmranHalim,LoisCarrington& S.A.Worm (eds),pp.145-170.Canberra:Pacific Linguistics.

Yap.F.H.,K.Grunow-Harsta & J.Wrona.(eds),2011.Nominalization in A sian Languages: DiachronicandTypologicalPerspectives.JohnBenjaminsPublishingCompany.

GlossingbasedonLeipzigConvention

ACC accusative HMB humbleform

ATTR attributive HON honorificform CAUSE causative LOC locative COMP complementizer NEG negative CONCESS concessive PASS passive COND conditional PN person'sname

CONV converb PRF perfective

COP copular PST past

EVID evidentiality SFP sentencefinalparticle FOC focusparticle TOP topicmarker