TheGrammarof Interactives
BERNDHEINE
GreatClarendonStreet,Oxford,OX26DP, UnitedKingdom
OxfordUniversityPressisadepartmentoftheUniversityofOxford. ItfurtherstheUniversity’sobjectiveofexcellenceinresearch,scholarship, andeducationbypublishingworldwide.Oxfordisaregisteredtrademarkof OxfordUniversityPressintheUKandincertainothercountries ©BerndHeine2023
Themoralrightsoftheauthorhavebeenasserted Impression:1
Allrightsreserved.Nopartofthispublicationmaybereproduced,storedin aretrievalsystem,ortransmitted,inanyformorbyanymeans,withoutthe priorpermissioninwritingofOxfordUniversityPress,orasexpresslypermitted bylaw,bylicenceorundertermsagreedwiththeappropriatereprographics rightsorganization.Enquiriesconcerningreproductionoutsidethescopeofthe aboveshouldbesenttotheRightsDepartment,OxfordUniversityPress,atthe addressabove
Youmustnotcirculatethisworkinanyotherform andyoumustimposethissameconditiononanyacquirer
PublishedintheUnitedStatesofAmericabyOxfordUniversityPress 198MadisonAvenue,NewYork,NY10016,UnitedStatesofAmerica
BritishLibraryCataloguinginPublicationData Dataavailable
LibraryofCongressControlNumber:2022940625
ISBN978–0–19–287149–7
DOI:10.1093/oso/9780192871497.001.0001
Printedandboundby CPIGroup(UK)Ltd,Croydon,CR04YY
LinkstothirdpartywebsitesareprovidedbyOxfordingoodfaithand forinformationonly.Oxforddisclaimsanyresponsibilityforthematerials containedinanythirdpartywebsitereferencedinthiswork.
2.2.4Morethanoneargumentstructure
3.1.4Discussion
3.1.5Theattentionsignal*hey
3.2Directives
3.2.1Introduction
3.2.2Workingdefinition
3.2.3Grammar
3.2.4Discussion
3.2.5Canonicalimperatives
3.2.6Specialdomainsofusage
3.3Discoursemarkers
3.3.1Introduction
3.3.2Workingdefinition
3.3.3Grammar
3.3.4Fillers
3.3.5AnoteonclicksasEnglishdiscoursemarkers
3.3.6Discoursemarkersinlanguagecontact
3.3.7Grammaticalization
3.3.8Discussion
3.4Evaluatives
3.4.1Introduction
3.4.2Workingdefinition
3.4.3Grammar
3.4.4Discussion
3.5Ideophones
3.5.1Introduction
3.5.2Workingdefinition
3.5.3Grammar
3.5.4Ideophonesvs.interjections
3.5.5Grammaticalization
3.5.6Anoteonsoundsymbolism
3.5.7Areideophonesinteractives?
3.5.8Ideophonesasapotentiallyopen-endedclass
3.6Interjections
3.6.1Introduction
3.6.2Workingdefinition
3.6.3Grammar
3.6.4Sub-types
3.6.5Discussion
3.7Responseelicitors
3.7.1Introduction
3.7.2Workingdefinition
3.7.3Grammar
3.7.4Functionalspace
3.7.5Other-initiatedrepairmarkers
3.8Responsesignals
3.8.1
3.9.1
3.9.2
3.9.3
3.9.4
3.9.5
3.9.6
3.11.4
3.11.5
3.11.6
4.3
4.2.1
4.2.2
4.2.3
4.2.4
4.2.5
4.4Grammaticalizationofideophones
4.4.2Acommonpathway:fromideophonetolexical category
5.Relatedapproaches
5.1Notionsofinteractivegrammar
5.1.1Grammartraditions
5.1.2Inserts
5.1.3Expressives
5.1.4Discussion
5.2Dual-processframeworks
5.2.1Anoverview
5.2.2Microgrammarvs.macrogrammar
5.2.3Sentencegrammarvs.theticalgrammar
5.2.4Propositionalstructurevs.interactionalstructure
5.3Neurolinguisticcorrelates
5.4Dualisminsocialpsychology
5.5Discussion
5.6Conclusions
6.Thestatusofinteractives
6.1Whereareinteractiveslocatedingrammaranddiscourse?
6.1.1Theyarepartoflanguagestructure
6.1.2Theyarelocatedattheperipheryoflanguage structure
6.1.3Theyareintermediatebetweenlanguageand somethingelse
6.1.4Theyarenotpartoflanguage
6.1.5Theyformadomainofgrammarseparatefrom sentencegrammar
6.1.6Conclusion
6.2Areinteractiveswords?
6.2.1Onterminology
6.2.2Whatisaword?
6.2.3Discussion
6.3Dointeractivesformasystem?
6.3.1Isthereasetofentitiesmakingupthesystem?
6.3.2Dotheentitiesinteractwithoneanother?
6.3.3Ifthereisasystem,whichpurposedoesitserve?
6.3.4Howisthesystemsetofffromothersystemsor phenomena?
6.4Areinteractiveslinguisticmarginalia?
6.5Interactivesinlanguagecontact
6.6Interactivesandlanguageacquisition
7.1Distinguishingfeatures
7.1.1Conceptualspace
7.1.2Meaning
7.1.3Mentalrepresentation
7.1.4Form
7.1.5Speedofprocessing
7.1.6Sequenceofactivation
7.1.7Thelocusofsemantic-pragmaticanchoring
7.1.8Theroleofcontext
7.1.9Constraintsonusage
7.1.10Useofgesture
7.2.1Usageofinteractivegrammarispredominant
7.2.2Bothgrammarsareequallyimportant
7.2.3Useofsentencegrammarispredominant
7.3Howdothetwogrammarscontributetodiscourse
7.3.1Generalobservations
Listoffiguresandtables
Figures
4.1.Anetworkofgrammaticalizationchainslinkingtypesofinteractives
5.1.ClassificationoftheC-unitsandinsertsofBiberetal.(1999)
Tables
1.1.Themaintypesofinteractives
1.2.ExamplesofEnglishinteractivesintheleftperipheryslotofanutterance 19
1.3.TonaldistinctionsintheGermaninteractives ach, hm,and oje 24
1.4.Englishinteractivesandnegation
1.5.UnusualphonologicalfeaturesinEnglishtypesofinteractives
1.6.ThemaintypesofinteractiveswithEnglishexamples
2.1.TwokindsofmeaningofinteractivesaccordingtoKaplan(1999)
2.2.Methods,functions,anddisplaysusedincommunication 69
2.3.Threebasicmodesofrepresentinginformationfordiscourseprocessing
2.4.TypesofEnglishinteractives
2.5.ApproximateparaphrasesoftheinteractivesinTable2.4
2.6.PrototypicalargumentstructureoftheinteractivesinTable2.5
3.1.Mainfunctionsofthetentypesofinteractives
3.2.Correspondencesbetweeninserts(Biberetal.1999)andinteractives 110
3.3.Correspondingformsoftheattentionsignal*hey indifferentlanguages 117
3.4.FunctionsexpressedbydirectivesintwoAustralianlanguages 124
3.5.(Logudorese)Sardiniandirectives 125
3.6.Suppletiveimperativestemsin!Xun(E1dialect,Kx’afamily) 126
3.7.PartialsuppletisminSwahiliimperativeforms 126
3.8.AspectualfeaturesofJapaneseadverbialideophones(‘mimetics’) 151
3.9.Themaindiscourseconstructionsforpresentingideophones 153
3.10.IdeophoneconstructionsinSiwu 157
3.11.Correlationsinsoundsymbolismobservedinideophones
3.12.Iconicsound–meaningdistinctionsinWestAfricanlanguages 166
3.13.Englishformsthathavebeenclassifiedasprimaryandsecondary interjections
3.14.TypesofEnglishinteractivesshowingaprimaryvs.secondaryform distinction
3.15.Structuralschemaimposedbyother-initiatedrepair(OIR)markersin conversations
3.16.Asampleof‘other-initiatedrepairmarkers’
3.17.DegreesofpolitenessinKoreanresponsesignalsbasedonspeechlevels
3.18.TonologicalvariantsoftheGermanresponsesignal hm
3.19.TypesofEnglishvocatives
3.20.ReductionofsomeItalianfirstnames
3.21.ReductionofsomeCatalanfirstnames
3.22.ReductionofsomeZulunouns(Bantu,Niger-Congo)
3.23.Extensionsoftheterm ı´iyá ‘mother’inDatooga(SouthernNilotic, Nilo-Saharan)
3.24.AnimaldirectivesfordeicticmotioninZargulla(Omotic,Afroasiatic)
3.25.FunctionsofanimaldirectivesformulesinAytHadiddu(Berber,Afroasiatic)
3.26.AnimaldirectivesinManambu(Ndufamily)
3.27.ClassesofanimaldirectivesinArusaMaasai(EasternNilotic,Nilo-Saharan)
3.28.FromanimaldirectivetoverbinMuna(Malayo-Polynesian,Austronesian)
3.29.NurserytermsderivedfromanimaldirectivesinAytHadiddu(Berber, Afroasiatic)
3.30.Distinguishinggrammaticalfeaturesofinteractives
4.1.CooptationofsomeideophonesfromverbsinSwahili(Bantu,Niger-Congo)
4.2.CooptationofideophonesfromverbsinMwera(Bantu,Niger-Congo)
4.3.Theparametersofgrammaticalization
4.4.Hypothesisonthedevelopmentofsomeinteractivestowardstextanchoring
4.5.Fromvocativetodiscoursemarker
4.6.Unidirectionalityinargumentdevelopment
4.7.Englishexamplesofcamouflaging
4.8.Germanexamplesofcamouflaging
4.9.Theinitialandfinalstagestobeexpectedinthegrammaticalizationof ideophones
4.10.Fromideophonetoverb,selectedexamples
4.11.FromideophonetonouninSouthernSotho(Bantu,Niger-Congo)
5.1.‘Interjections’distinguishedbyBloomfield([1933]1962)
5.2.‘Interjections’distinguishedbyLeechetal.(1982)
5.3.‘Discoursemarkers’distinguishedbyZwicky(1985)
5.4.‘Pragmaticmarkers’distinguishedbyFraser(1990;1999)qualifyingas interactives
5.5.‘Interjections’distinguishedbyNorrick(2009)
5.6.Classesof‘interjections’distinguishedbyAmekaandWilkins(2006)
5.7.Korean‘discourseadverbials’distinguishedbySohn(1999)
5.8.Wampis‘interjections’distinguishedbyPeña(2015)
5.9.Kisi‘interjections’distinguishedbyChilds(1995)
5.10.Zuluʽinterjectives’distinguishedbyDoke([1927]1988)
5.11.Thompsonʽexpletives’distinguishedbyThompsonandThompson(1992)
5.12.EnglishinsertsdistinguishedbyBiberetal.(1999)
5.13.Englishexamplesofformulaictheticals
5.14.Sentencegrammarvs.interactivegrammarinaspeechsampleofan English-speakingaphasicpatient
5.15.Somedistinguishingfeaturesassociatedwithprocessesofreasoningand judgment
5.16.Dual-processframeworksrelatingmoststronglytointeractivegrammar
6.1.ExamplesofEnglishinteractivesintheleftperipheryslotofanutterance
6.2.ExamplesofEnglishprimaryinteractives
7.1.Featuresprototypicallydistinguishingsentencegrammarfrominteractive grammar
7.2.ExampleofanopeningsectioninEnglishtelephoneconversations
Interactives
Oneshortinterjectionmaybemorepowerful,moretothepoint,more eloquentthanalongspeech.Infact,interjections,togetherwithgestures,themovementsofthemusclesofthemouth,andtheeye,would bequitesufficientforallpurposeswhichlanguageanswerswiththe majorityofmankind.
(MaxMu¨ller1861:410)
1.1 Introduction
Inmanymonographsoflinguisticsaviewisexpressedorimpliedthatisdescribed succinctlyby Givón(1984) inthefollowingway:
Mostlanguagesdisplaythismixed-bagcategorywithexpressionssuchas‘yes’,‘no’, ‘hey’,‘oh’,‘hi’,‘wow’,‘ouch’,etc.ortheirfunctionalequivalents.Itisnotaunified categoryfunctionally,morphologicallyorsyntacticallyanditishighlylanguage specific.
(Givón1984:84)¹
Thegoalofthepresentstudyistotakeissuewiththisview.² Itisarguedthatthere isinfactsuchacategoryandratherthanbeinga‘mixedbag’,itcanbedescribed acrosslanguagesasagrammaticalcategoryofitsown.Thiscategoryincludes butisnotrestrictedtotheitemsmentionedbyGivón,whicharereferredtohere summarilyasinteractiveformsor,inshort,as interactives. ConsiderthefollowingconversationthattookplaceinTjwao,alanguage spokeninwesternZimbabwe.
(1)Tjwao(Khoe(‘CentralKhoisan’); Andrason,Fehn,andPhiri2020:14,(17))
A: Yii!
B: A-a!
A: Yee!
B: Ehe!
¹ AmoredifferentiatedviewisfoundintherevisededitionofGivón’sintroductiontosyntax(Givón 2001:102).
² Throughoutthisbook,interactivesareprintedinbold.
TheGrammarofInteractives.BerndHeine,OxfordUniversityPress. ©BerndHeine(2023).DOI:10.1093/oso/9780192871497.003.0001
Theconversationconsistsentirelyofinteractives,thatis,ofthekindofexpressions mentionedbyGivón,called‘interjections’bytheauthorsof(1).Theconversation beginswithpersonAmeetingpersonB.AcallsB,usingtheattentionsignal yii. Hearingthis,Bexpresseshissurprisebymeansoftheemotiveinterjection a-a. ³ Subsequently,Aproducestheinterjection yee tovoicehisexcitementandhappinessaboutmeetingB.SpeakerBexperiencesthesamefeelingandexpressesthis bymeansoftheresponsesignal ehe!. ⁴
Examplesofthekindillustratedin(1)arenotisolatedcases.MovingfromAfrica toAustralia,casessuchastheonein(2)canbefound,onceagainconsisting entirelyofinteractives.Theexcerptofaconversationin(2)istakenfromtheMayalilanguageofWesternArnhemLand.Itcontainstwodirectives(boi!, njudj!)and twoinstancesoftheinterjection njonj-njonj!.
(2)Mayali(Gunwinyguan,Arnhem; Evans1992:227,(3))
A: Boi!
Hey,comehere!
[Oldwomantoyouth:]‘Hey,comehere!’[Oldwomanholdsupbaby]
B: Njonj-njonj!
Njudj! Njonj-njonj!
Whatalittlesweetheart!Blowyournose!Whatalittlesweetheart! […]
‘Isn’tshealittlesweetheart!Blowyournose!Isn’tshecute!’
Suchexamplesarenotrestrictedto‘interjections’,theymayaswellinvolveother kindsofinteractiveforms.InthefollowingexampletakenfromBaka,thelanguage ofatraditionalhunter-gathererpeopleofSoutheastCameroon,thereisa‘mininarrative’consistingofastringofsixideophones,‘whichevokesinthehearerthe illusionofadirectparticipation’(Kilian-Hatz2001:157).
(3)Baka(Ubangi,Niger-Congo; Kilian-Hatz2001:157)⁵ wòàwòàwòàpɔˋɔˋɔˋ kung wóoò hunters.discusschimp.interrupts.eatingspear.strikes.chimpfallsdown pao tung.
breaks.a.branchfalls.hard.on.ground
³ InaccordancewithadistinctionproposedbyCaffiandJanney(1994:328–9),interjectionsare treatedhereasbelongingtoemotiveratherthanemotionalcommunication—pendingfurtherresearch. Accordingtotheseauthors,emotionalcommunicationisa‘typeofspontaneous,unintentionalleakageorburstingoutofemotion’,whileemotivecommunicationis‘inherentlystrategic,persuasive, interactionalandother-directedbyitsverynature’(see Stange2016:29forfurtherdiscussion).
⁴ Throughoutthisbook,theterms‘speaker’and‘hearer’areusedtorefertointerlocutorsinconversationalexchanges.Thetermsareoccasionallyextendedtowritersandreaders,respectively,butour focusisonspokenratherthanwrittendiscourse.Notethatinmorethan90%ofthelanguagesofthe worldthereisonlyspokendiscourse,whilethereisessentiallyno‘natural’languageintheworldthat disposesonlyofwrittendiscourse(ignoringspecialcasesliketheclassicallanguagesLatinorSanskrit).
⁵ Theauthordoesnotprovideatranslationofthistextpiecebutitseemsthatthemeaningis recoverablefromtheglosses,apparentlydescribingachimpanzeehunt.
Butinteractivesarenotrestrictedtousageamonghumans—theyarealsofound cross-linguisticallyintheinteractionwithanimals.⁶Taketheutterancein(4)from theArusaMaasailanguageofnorthernTanzania:Amancallsforhisdonkey’s attention(kuk),instructsittomoveforward(mape),andexpresseshissurprise andannoyance(ʃie)ashedidnotexpecttheanimaltoslowdownandstopmoving.Theutteranceiscomposedentirelyofinteractives,where kuk and mape are directivesdedicatedspecificallytointeractwithanimalswhileʃieisaninterjection, signalingtheemotivestateofthespeaker.
(4)ArusaMaasai(EasternNilotic,Nilo-Saharan; AndrasonandKarani2021b: 31)⁷
Kukmape, ʃie! DIRDIRINT ‘Kuk [hey,saidtoadonkey] mape [let’sgo],ah!’
Movingonfromsomewhat‘exotic’languagesandformsofcommunication,considernowthefollowingconversationalexchangeinabetter-knownlanguage—a ritualexchangethatmostofusarefamiliarwithinsomeformorother.Likethe precedingexamples,(5)consistsentirelyofinteractives,namelythegreetingforms hello and hi andtheinterjection oh. ⁸
(5)English(Levinson1983:311)
C:(Ccausestelephonetoring)
R: Hello
C: Hi
R: Ohhi::
Theexamplein(5)presentstheopeningsectionofatelephoneconversationasit cancommonlybeobservedbetweenspeakersofEnglish.Beingpartoftheoverallorganizationofatelephonecall,itmaysomehowconstituteaself-contained interaction,consistingofanexchangeofgreetingtokensbetweentwopersons,C andR.Buttheinformationconveyedappearstobemorecomplex,assuggestedby Levinson(1983:311–12):Whilehelloisagreetingtoken,speakerRdoesnotseem touseitassuchinthissituation;rather,helloissimplyaresponsetoC’sphonecall andadisplayforrecognitionofR’sidentity.C’ssubsequentuseofhithenfunctions infactasagreetingtoken,andthistokenisreturnedbyRwiththegreetingtoken ohhi::.Thelatter,however,isnotrestrictedtogreetingbutinadditionsignalsthat RhasrecognizedC.
⁶ SeeSection3.11foradetaileddiscussionofanimaldirectives.
⁷ Thegloss‘DIR’standsforanimaldirectiveforms.Theauthors AndrasonandKarani(2021b:3) use‘CAC’instead,whichstandsfor‘conativeanimalcall’.
⁸ Boththeexamplein(5)anditsdiscussionarebasedontheanalysisby Levinson(1983:311–12). WewillreturntotheexampleinSection7.2.1.
Theexamplespresentedsofarconsistentirelyofcombinationsorsequencesof interactivesbut,morecommonly,interactivesarefoundco-occurringwithother piecesoftext,asinthefollowingexchange,basedonanactualexperienceofthe author:⁹
(6)English(Wharton2003:41,(4),(5))
Patient:Becarefulwiththatneedle!
Dentist: Oops.
Patient: Ouch!
Dentist: Hell!I’msorry.
Patient: Shit! Getthebloodythingoutofmycheek!
Whatalltheseexamplessuggestis,first,thatinteractiveshave,ormayhaverich socialmeaningcontent.Second,socialexchangesliketheonesin(1)to(6)can consistlargelyorentirelyofinteractives.Third,interactivescanandfrequently doconstituteutterancesorconversationalturnsoftheirown—ifIsayinEnglish Oops!, Ouch!, Hell!,or Bang! thenthesearecommonlyacceptedasself-contained utterances. ¹⁰ Fourth,interactivesexpressconventionalizedmeaningcontentsthat arestoredandretrievedlikeotherformsofalanguage(Section 1.4).Andfinally, thefactthatinteractivescanoccurasdistinctutterances,andthattheycanbe combinedandarrangedtofunctionasself-containedpiecesofdiscourse,would seemtosuggestthattheysomehowhavefeaturesofanindependentdomainof linguisticcommunication.
Suchobservationshavenotreallyoverawedstudentsoflinguistics(Section 6.1).Waybackinthenineteenthcentury,well-knownlinguist Mu¨ller(1861:346, 352)observedthatinterjections‘areplaythings,notthetoolsoflanguage’and that‘languagebeginswhereinterjectionsend’. Benfey(1869:295)concludedin hishistoryoflinguisticsthatinterjectionis‘thenegationoflanguage’,andfor Sapir(1921:6–7)interjectionswereatbesta‘decorativeedgingtotheample, complexfabric’oflanguageand‘thenearestofalllanguagesoundstoinstinctive utterance’.
Thesituationhasnotchangeddramaticallyintheensuinghistoryoflinguistics; sufficeittomentiontheinfluentialstudyby Goffman(1978:809–10),wherethe authormaintainsthat‘responsecries’,thatis,theinterjectionsstudiedbyhim,are avarietyof‘non-words’andassuch,‘can’tquitebecalledpartofalanguage.’
Inthecourseofthelastdecades,linguisticshasbeenenrichedbyaplethoraof referencegrammarsoflanguages,manyofwhicharespokenintheremotestcornersoftheworld.Thesegrammars,mostlybasedonyearsoffieldwork,describe
⁹ TimWharton(p.c.ofMarch5,2022).
¹⁰ Anʽutterance’istakenheretostandforapieceofself-containedtextproducedbyaparticular individualonaparticularoccasionforaparticularpurpose.Ourconcernhereismainly,thoughnot exclusively,withspokenutterances.Suchapiecefrequentlyis,butneednotbeasentence.
thegrammaticalstructureofthelanguageconcernedingreatdetailandprovide thelinguistwithgoldminesofinformation.Butsomewhatsurprisingly,asubstantialpartofthesegrammarscontainsnomentionofinteractives,andmanyother grammarsdealwithinteractivesatbestinafewcasualnotesorasakindofshort appendage.
Tomentionanotherkindofexample:TheKoreanlanguageisespeciallyrichin interactives,disposingofaninventoryofover5000ideophones(Section3.5.1)— thenumberinfactcomesclosetothatofnounsorverbs,andKoreanideophones havearemarkableimpactonthesemanticandgrammaticalstructureofthelanguage.Butthereadermaybesurprisedtofindnodescriptionofideophonesinthe influentialreferencegrammarofKoreanby Sohn(1999).Ideophonesarementionedinasectiononsoundsymbolism(Sohn1999:96–102)but—againperhaps surprisingly—manyoftheideophoneshavenoconceivablerelationshiptosound symbolism.¹¹
Forsomeonewhowantstolearnalanguage,thesegrammarsmaytherefore besomewhatdisappointing.Afterhavingbattledone’swaythroughthe400to 800pagesofthesegrammars,havinginternalizedthesentencesandotherstructuresofthelanguage,onemaystillhavethefeelingofnotbeing‘communicatively competent’,knowinghowtogreet,bepolite,exhort,warn,beseech,surprise,call, persuade,cooperate,disapprove,challenge,orentertainothers—inshort,todothe kindofthingsonewouldfeelobligedorliketodowhenbeingconfrontedwiththe communityspeakingtherelevantlanguage.
Suchdisregardforlanguageasatoolofsocialinteractionisbynomeans restrictedtothetraditionofgrammarwritingjustalludedto;itisquitecommoninallkindsoflinguisticpublicationsandschoolsoflinguistics.Suffice ittomentionaparadigmexample:Oneofthemostcomprehensivereference grammarsofEnglish, TheCambridgeGrammaroftheEnglishLanguage (HuddlestonandPullum2002)discussesmostaspectsofEnglishgrammaringreat detail,butinteractivesareblatantlyabsent.Whileatleastinterjectionsareclassifiedasoneofthenine‘lexicalcategories’ofEnglish,onlyadozenlinesare devotedtotheminaworkcomprising1842pages.Theattitudeofoneof thetwoauthorsofthisreferencegrammarisnicelyreflectedinthefollowing remark:
Interjectionsaresounimportanttothefabricofthelanguagethattheyarealmost completelyignoredingrammars.Thereisalmostnothingtosay.Theyhaveno syntacticpropertiesatall,youpoponeinwhenthespiritmovesyou.Andtheir basicmeaningissimplyexpressiveofatransitorystate.
(GeoffreyK.Pullum,January2005post.Quotedfrom Ameka2020:57)
¹¹ InhisearliergrammarofKoreanthough,Sohn(1994)hadawholechapterdevotedtoideophones andinterjections(Chapter 4).
Thisremarkisreminiscentofsomeviewsfoundinnineteenth-century linguistics—forexample,when Benfey(1869:295)concludedinhishistoryof linguisticsthat‘interjectionsareemployedwhenoneeithercannotorwillnot speak.’
Overall,achangeinattitudecanbeobservedinthehistoryofgrammarwriting.Priortothemidtwentiethcentury,authorsworkingonthelanguagesof Africa,Asia,Australia,ortheAmericaswerelikelytoatleastmentionsomeofthe socialfunctionsofthelanguagesstudiedbythem.Butwiththeriseoflinguistic structuralisminallitsmanyversions,aninterestinlanguageasatoolofsocial interactiondeclinedandgrammaticaldescriptionbecamelargelyrestrictedto apprehendingthestructuralpropertiestobeobservedinthecategoriesofsentence grammar.
Nevertheless,inthe1990s,interactiveswerethesubjectofarangeofseminal studies,revealingalltherichnessofthesemanticspacethatinteractivespresent (e.g., Ameka1992a, 1992b; Evans1992; Wierzbicka1992; Wilkins1992).These studieshadsomeinfluenceonensuingdiscussionsonmeaning(seeSection2.1). However,thestudieshadnomajorimpactonthetraditionofgrammarwriting oroflinguisticanalysisingeneral,aswitnessed,forexample,intheremarkby GeoffreyPullumjustquoted.
Underlyingmuchofmainstreamlinguisticsuptothepresentthereappearsto beawidespreadassumptiontotheeffectthatthereisnogrammartothelanguage ofsocialinteractionor,evenifthereis,itisnotworthyofscholarlyattention. Onthisview,interactivesarenotclearlypartoflanguage,conceivedsomehow as‘stowaways’tobelocatedsomewherebetweenlanguageandsomethingelse: Theyhavebeenclaimedtobeperipheraltothelanguagesystem,beinglocated attheboundarybetweenverbalandnon-verbalbehavior,orbeingparalinguistic andasystematicelements(seeSection6.1).Accordingly,theyarecalled‘response cries’,‘non-words’,‘semi-words’,‘quasi-words’,or‘partlynatural,partlyconventionalsemi-words’(see Ameka1992a:112; 2006:744–5; Nu¨bling2004; Ameka andWilkins2006:5; StangeandNu¨bling2014:1986; Wharton2016:21),or as Ferguson(1981:21)putsit,‘thelittlesnippetsofritualusedineveryday encounters’.
Ideophoneshavebeencalled‘astep-childofmodernlinguisticscience’(Voeltz and Kilian-Hatz2001:2)andhavebeenthesubjectof‘arecurrentnarrativeof marginalisation’(Dingemanse2018:2).Inasimilarway, O’ConnellandKowal (2008:133)characterizeinterjectionsas‘aphenomenonthathasbeenhistorically amarginal,thoroughlyneglectedlinguisticcategory’.
Insum,interactives,likeinterjections,ideophones,socialformulae,discourse markers,etc.,havebeeneitherneglectedorignoredinmostpreviouslinguistic work—bethatdescriptiveorprescriptivework.Understandably,thisattitudeis deeplydeploredbystudentsofinteractives.
Thegoalofthepresentstudyistoarguethatthistraditionalattitudeisinneedof reconsideration.Ratherthanbeingperipheral,oramarginalphenomenon,interactivesaredescribedasacategoryonitsown,contrastingwithwhat Haiman (2018) calls‘prosaic’or‘propositional’grammarandwhatisreferredtohere as‘sentencegrammar’.Whereassentencegrammarhasbeendescribedashavingdenotational,informational,conceptual,descriptive,truth-conditional,or objectivefunctions,interactivestendtobeportrayedasservingconnotative,or expressivefunctions.Perhapsmostcommonlytheyarereferredtoasexpressives, thatis,asverbalmeanshavinganexpressiveratherthanareferentialfunction (Foolen2016:473).Thereasonforcallingthemhere‘interactives’ratherthan ‘expressives’isthatthelattertermhasfrequentlybeenappliedalsotoarangeof phenomenafarbeyondthosethatarecoveredbythedefinitionofinteractivesto beprovidedinthenextsection.¹²
AfterdefininginteractivesinSection 1.2,thepropertiescontainedinthe definitionarediscussedinSection 1.3.Section 1.3 ismeanttodemonstratethat thegrammarofinteractivesdiffersfromthatofsentencegrammarinaprincipled way,whileSection 1.4 showsthatthetwoneverthelessshareacommonbase.An inventoryofthetentypesofinteractivesdistinguishedisprovidedinSection 1.5 withEnglishexamples,andSection1.6thenconcludesthechapterwithanoutline ofthecontentofthebook.
1.2 Definition
Whatdovocativeformslike Mom! or Sir! possiblyhaveincommonwithinterjectionslike oops or yuck,ordiscoursemarkerslike indeed or um?Oneimportant answerisprovidedinthissection:Theyallconformtothesamedefinition.
Interactiveformsor,inshort,interactivesprovideinsightsintohowspeakers conceivethemselvesintheworldofsocialcommunication.Theyareprefabricatedroutineformsandincludebutarenotrestrictedtowhat Ferguson(1981) describesassocialformulasand Coulmas(1981:2–3)asconversationalroutines, thatis,‘highlyconventionalizedprepackagedexpressionswhoseoccurrenceis tiedtomoreorlessstandardizedcommunicationsituations’.Theyare‘interactive’inthesensethattheyaregroundedinsocialinteractionandweproposethe prototypicaldefinitionin(1)forthem.
(1)Aninteractiveisaninvariabledeicticformthatisinsomewaysetofffrom thesurroundingtextsemantically,syntacticallyandprosodicallyandcan neitherbenegatednorquestioned.
¹² Foranoverviewofresearchonexpressivessee Foolen(1997, 2012, 2016).
Itgoeswithoutsayingthatlikeotherlinguisticelements,interactivesdonotoccur inisolation,sealedofffromtherestofthetextofwhichtheyareapart.Rather, theymayshowfeaturesofassimilationtotheirtextualenvironment,suchastraces ofprosodicandotherkindsofintegration.
Thedefinitionin(1)makesnomentionofthefunctionsofinteractives,forthe followingreason:AswewillseeinTable 1.1,interactivesincludeanumberof differenttypes,andeachtypehasitsownfunctionalfocus.ThereaderisthereforereferredtoChapter 3 forinformationonthefunctionsofthevarioustypesof interactives(seeTable3.1).
Thepropertiesmentionedin(1)arediscussedandlookedatinmoredetailin Section 1.3.1.Inaddition,anumberofotherpropertieshavebeenreportedand arelookedatinSection 1.3.2.Theterm‘interactive’isusedhereinamoregeneral senseinthatitincludesboththe‘expressives’and‘interactives’distinguishedinthe frameworkofFunctionalDiscourseGrammar(HengeveldandMackenzie2008: 77),¹³itlargelycorrespondstotheterm‘insert’ofBiberetal.(1999;seeSection5.1). Itincludesbutisnotrestrictedtothe‘interactionalstructure’of Wiltschko(2021; seeSection5.2.4).However,itdoesnotincludewhat Ogi(2017) calls‘interactive markers’inJapanesesincethelatterlacksalientfeaturesofinteractivesinthesense ofthetermusedhere,suchasdistinctsyntacticandprosodicstatus.
Followingotherauthorsdealingwithspecificcategoriesofinteractives,suchas Evans(1992:22), Gehweiler(2008), AndrasonandDlali(2020:164)oninterjectionsor Childs(1994) onideophones,wepropose(1)tobea prototypical definition.Thus,ratherthanasadiscretecategory,interactivesareassumedhere toinstantiateaprototypeoflinguisticforms.Indoingso,wearerelyingonthe notionofprototypetheoryasdevelopedinthetraditionof Rosch(1973),ignoringmorespecificproblemsandtheoreticalissuesraised,forexample,inwork onfamilyresemblance(cf. Wittgenstein1953)oronradialcategorization(Lakoff 1987).Prototypescanbecharacterizedinthefollowingway(e.g., Taylor1989): (a)Theyarecategoriesthatarenotdefinedintermsofasetofnecessaryand sufficientattributes.(b)Categorymembershipisgraded:Somemembersarebetterinstancesofthecategorythanothers.(c)Themostcentralmembersfunction ascognitivereferencepointsofthecategory,beingmorerepresentativeofthe categorythanperipheralmembers.
Intheframeworkproposedhere,‘members’areinstantiatedbyindividualinteractives.Interactivesshowingthewholesetofpropertiesarethemostprototypical
¹³ Whatdistinguishes‘expressives’and‘interactives’intheframeworkofFunctionalDiscourse Grammar(HengeveldandMackenzie2008)isthatunliketheformer,thelatterare‘clearlydirected totheAddressee’.AsarguedbyLachlanMackenzie(p.c.ofFebruary17,2022),expressives‘arelike involuntarycriesofpain,joy,etc.,i.e.theyarenotinherentlycommunicative’.
Wewillreturntothedistinctionbetweenpresenceandabsenceofaddresseeor,‘hearer’aswesay here,inSection2.2.
or‘centralmembers’ofthecategory,andtheyarethemainconcernofthe paragraphstofollow.
Thetermʽdeicticform,’or‘indexical’,referstothefactthatinteractives are immediatelyanchoredtothesituationofdiscourse. ¹⁴ Thetermrelatestothenotion ʽdiscourse-deixis’ofWeinreich([1966]1989:69),whichissaidtoprovidethecognitiveframeusedbyinterlocutorstodesignandinterpretspokenorwrittentexts; itconcerns‘thepresuppositionsaboutdiscoursecontext,or“discourseplacedness conditions”’as Evans(1992:228)putsitwithreferencetothenatureofhisorganizinginterjections.Beingsemantically,syntactically,andprosodicallyseparated, interactivescanbe‘stand-alones.’Theycanandquitecommonlydoformwhathas beenreferredtoas‘situationboundutterances’(FelixAmeka,p.c.ofJuly25,2021).
Accordingtothedefinitionin(1),alinguisticformthatdoesnothavethe wholesetofpropertiesisnotaprototypicalmemberofthecategory,andthemain concernofthisstudyiswithprototypical,orcoremembersofthecategory—in otherwords,withformsthatexhibitallthepropertiesin(1).Aswewillseein Section1.3.1,however,somepropertiesarenotentirelystable.Deviationsfromthe prototypearemostofalloftwokinds.Ontheonehand,aninteractivemayhave asetofvariants,someofwhichareinvariablewhileothersarenot.Ontheother hand,bybeinganchoredtothesituationofdiscourseratherthantothestructure ofaclause,interactivesarehighlycontext-sensitiveandtendtoexhibitanumber ofcontext-relatedusageswherenotallusagesconformtothepropertiesin(1).
Wewillnotgenerallydiscardlessprototypicalinstancesasfarastheyconcern problemswithonlyoneortwooftheproperties.If,however,centralissuesonthe natureofinteractivesareatstakewewillinsistthatallpropertiesbehonoredin thediscussionstofollow.
Interactives,astheyaredefinedin(1),includearangeofdifferentelements,and theseelementscanbeclassifiedintotentypes,namelytheoneslistedinTable 1.1 (seeSection 1.5 formoreinformation).Itisthesetypesthatwillbetheconcernof thechapterstofollow,andinChapter 3 wewilldealwitheachofthemingreater detail.
AstheclassificationproposedinTable 1.1 shows,theterm‘interjection’isused hereinanarrowersensethaninmanyearlierclassifications,wheretypessuchas attentionsignals,directives,responseelicitors,and/orresponsesignalswerefrequentlysubsumedundertheheading‘interjection’.Thereasonforthisdiscrepancy willbediscussedinSection3.6.
¹⁴ Theterms‘deixis’and‘indexicality’canbetracedbackeachtodifferenthistoriesofresearch traditionsandtendtobeassociatedwithslightlydifferentreferences,theformerbeingprevalentin linguisticsandthelatterinphilosophyandpragmatics(seealsoSection2.1.3.1).Inthepresentstudy, ‘deicticform’and‘indexical’arebothtreatedasreferringtolinguisticelementswhosemeaningisimmediatelyanchoredtothesituationofdiscourseand,hence,mustbeinterpretedwithreferencetothe ‘context’inwhichtherelevantlinguisticdiscoursetakesplace.
