CONVERSATIONANALYSIS04.12.2006SindyKermerMelanieMüller*1Whatisconversation?•awayofusinglanguagesocially,of“doingthingswithwords”•aninteractionoftwoormoreparticipants•numberofparticipantsandlengthofcontributiontotheconversationcanvary•open-ended,hasthepotentialtodevelopinanywayplannedoccasionsforspeaking,suchasmeetingsordebatesWHATISCONVERSATIONANALYSIS?CAisthestudyofrecorded,naturallyoccurringtalk-in-interaction.Itsobjectofstudyistheinteractionalorganizationofsocialactivities.CAaimsatdiscoveringhowparticipantsunderstandandrespondtooneanotherintheirturnsattalk,withacentralfocusonhowsequencesofactionsaregenerated.CAisonlymarginallyinterestedinlanguageassuch,butfirstandforemostinlanguageasapracticalsocialaccomplishment.BackgroundtoCA•HaroldGarfinkel,1960s,ethnomethodological/sociologicalapproach•organizationoftalk-in-interaction•empiricalapproachwhichavoidsprematuretheoryconstruction→methodsareinductive-searchforrecurringpatterns→gatheringdataandanalysisofdataofactualpiecesoflanguage,real-life-conversations→data-driventheorizingMindMapConversationalAnalysisPracticalExamplesConversationsbetweenfriendsRelationshipcounsellingsessionsLegalhearingsFieldsEthnomethodologyDiscursivePsychologyQualitativeResearchPeopleHarveySacksEmanuelSchegloffAudio/videorecordingsTranscriptsAnalysesarebothcontext-free=(-)ethnographiccontextcontext-renewing=(+)conversationalcontext•avoidsanalysesbasedonasingletext→asmanyinstancesaspossibleofsomeparticularphenomenonexaminedacrosstexts→discoverthesystematicpropertiesofthesequentialorganizationoftalkandthewaysinwhichutterancesaredesignedtomanagesuchsequencesBasicnotionsturn:basicunitofconversation→maycontainmanyillocutions,iseverythingaspeakercommunicatesduringaunitofconversation•turn-taking:basicformoforganizationforconversation→speaker-changeoccurs→mostly,onespeakertalksatatime→transitionfromoneturntothenextwithoutgaporoverlap→turnorderandsizenotfixedTurn-Taking→lengthandtopicofcontributionnotspecifiedinadvance→currentspeakermayselectanotherspeakerorpartiesmayself-selectinstartingtotalk→transitionfromoneturntothenextwithoutgaporoverlap→turnorderandsizenotfixed→repairmechanisms:dealwithturn-takingerrorsandviolationsSizeofturnsisvariablebutthereisatendencytowardminimalizationTransitionRelevancePlaces(TRP)•transition:arelayoftherighttospeaktothenextspeaker→mechanismsofselection(self-orother-)→TRPcanbeexploitedbythespeakerholdingthefloor…directly,forthepurposeofallocatingtherighttospeaktoanextspeakerofhis/herchoiceindirectly,bythrowingthefloorwideopentowhoever→speakermayjustignoretheTRPandcontinuepastATransitionRelevancePlace(TRP)isapointatwhichaturnchangecanoccur.Typicaltechniquesforselectingnextspeaker:Question(offerorrequest)+addresstermTaggedassertion+addressfeatureRepairinitiatorsWho?Youdidwhat?Pardon?PreviewingTRPs•Whyareweoftenabletopredicttheendofsomebody’sspeech?→AdjacencyPairs→changesofspeeddelivery→intonation→word-choicepatterns04.12.2006SindyKermerMelanieMüller*AdjacencyPairs•discoverythatbecameastartingpointforawholenewapproach(similarasspeechactstopragmatics)•twosubsequentutterancesconstitutingaconversationalexchange•distinctionbetween‘firstpairpart’and‘secondpairpart’AdjacencyPairs:Examples•Complaint/denialKen:Heyyuhtookmychairbythewayan’Idon’tthinkthatwasveryniceAl:Ididn’ttakeyerchair,it’smychair.•Compliment/rejectionA:I’mgladIhaveyouforafriend.B:That’sbecauseyoudon’thaveanyothers.PREFERENCEORGANIZATIONOFADJACENCYPAIRSAninferentialaspectofadjacencypairsstemsfromthefactthatcertainfirstpairpartsmakealternativeactionsrelevantinsecondposition.Insomeadjacencypairsthereisachoiceoftwolikelyresponses,ofwhichoneistermedpreferredresponse(becauseitoccursmorefrequently),andtheotherdispreferred(becauseitislesscommon).Preferredresponse=unmarkedresponseDispreferredresponse=markedresponsePreferredresponsetoaninvitationisanacceptanceDispreferredresponseisadeclinationAdjacencyPairsFirst–APreferredSecond–B1DispreferredSecond–B2Offer/InvitationAcceptanceRefusalRequestComplianceRefusalAssessmentAgreementDisagreementBlameDenialAdmissionQuestionAnExpectedAnswerUnexpected(orlackof)AnswerPREFERENCEORGANIZATION1.OfferA:Likealift?-acceptance(preferred)B:Yousavedmylife.-refusal(dispreferred)B:Thanks,butI’mwaitingformyfriend2.ComplimentA:That’saniceshirt.-acceptance(preferred)B:Thanks-rejection(dispreferred)B:Well,Ithinkitmakesmelookold-agreement(preferred)B:It’squitenice,isn’tit?-shiftB:Judyfounditforme.-returnB:Thanks,Ilikeyourstoo.3.BlameA:Youbroketheglass-denial(preferred)B:Ididn’tdoit.-admission(dispref)B:Sorry,Ididn’tseeit.Structuralcharacteristicsofadispreferredresponse;combinationsof:PausesorsilencesprecedingdispreferredturnPrefacemarkingtokensUhmWellOtherpertubations(lengthenings,cut-offsetc)INSERTIONSEQUENCES(Pre-sequences)Thesesequencesarecalledadjacencypairsbecause,ideally,thetwopartsshouldbeproducednexttoeachother.Thepointisthatsomeclassesofutterancesareconventionallypairedsuchthat,ontheproductionofafirstpairpart,thesecondbecomesrelevantandremainsso,evenifitisnotproducedinthenextturn.Thenextturninanadjacencypair‘sequence’isarelevantsecondpairpart.Butthatneednotbethenextturnintheseriesofturnsmakingupsomeparticularconversation.Example:(Levinson,1983)A:CanIhaveabottleofMich?Q1B:Areyouovertwenty-one?Ins1A:No.Ins.2B:No.A1InsertionSequencesSchegloffgivesthreereasonsforInsertionSequences.Thehearerdoesnotunder-standwhatthespeakersaid.Thehearerdoesnotwanttocommithimselfuntilhehasmoreinformation.Thehearerisdelayingananswer.(no)FirsttopicslotNopriorconstraintonwhatshouldbetalkedaboutUsuallythemaintopicofthecall01R:whydon’tweallhavelunch02C:okaysothatwouldbeinStJude’swouldit?03R:yes04(0.7)05C:okayso:::06R:oneo’clockinthebar07C:okay08R:okay?09C:okthanksverymuchindeedGeorge=10R:=allright11C:[seeyouthere12R:[seeyouthere13C:okay14R:okay[bye15C:[byeclosureofsecond(third,fourth)topicsareclosure-relevantarrangementsREPAIRSRepairisagenerictermusedinCAtocoverawiderangeofphenomena,from---seemingerrorsinturn-taking,suchasoverlappingtalk,-toanyoftheformsofwhatiscommonlycalled‘corrections’–thatis,substantivefaultsinthecontentsofwhatsomeonehassaid.THEORGANIZATIONOFREPAIRSRepairtypesTherepairsystemembodiesadistinctionbetween1)theinitiationofrepair(markingsomethingasasourceoftrouble),and2)theactualrepairitself.Thereisalsoadistinctionbetween1)repairinitiatedbyself(thespeakerwhoproducedthetroublesource),and2)repairinitiatedbyother.Consequently,therearefourvarietiesofrepair:SELF-INITIATEDSELF-REPAIRRepairisbothinitiatedandcarriedoutbythespeakerofthetroublesource.EXAMPLE1.I:Isitflu:you’vegot?2.→N:NoIdon’tthink-IrefusetohaveallthesethingsSELF-INITIATEDOTHER-REPAIRThespeakerofatroublesourcemaytryandgettherecipienttorepairthetrouble–forexampleifanameisprovingtroublesometoremember.EXAMPLE:Inthefollowingexamplethefirstspeaker’sreferencetohistroublerememberingsomeone’snameinitiatesthesecondspeaker’srepair.1B:HehadthisuhMistuhW-mwhatever,Ican’tthinkofhisfirstname,Wattson,theonethatwrote/thatpiece2A:/DanWatts.OTHER-INITIATEDOTHER-REPAIRTherecipientofatrouble-sourceturnbothinitiatesandcarriesouttherepair.Thisisclosesttowhatisconventionallyunderstoodby‘correction’.EXAMPLE:Inthefollowingexamplethereisanexplicitcorrectionwhichisthenacknowledgedandacceptedinthesubsequentturn:1Milly:andthentheysaidsomethingaboutKruschevhasleukemiasoIthoughtohit’sallabigputon.2.→Jean:Breshnev.THEPREFERENCEFORSELFREPAIRSThereareseveralwaysinwhichturnsaredesignedtofacilitateself-repair,ordisplaythespeaker’ssensitivitytotheappropriatenessofself-repair.ConsiderthefollowingextractfromacalltotheBritishAirwaysflightinformationserviceandtrytoanalyseit:1A:thetimeforyou,/h2C:/yes3A:isohonesevenfivenight(.)5A:/sevenfiveni:ne,/((Smileyvoice))6C:/sevenfivewhat.(.)/yesResearchinabiggerwayComplexityofturntakingsystemsystemofordinaryconversationacrossculturesSomedifferentturntakingconventionsBurunditurnallocationconventionsInstitutionaltalkClassroomsCourtsNewsconferencesDebatesResearchinanotherwayAlternativeexplanations(whichdon’tworkTurntakingbasedon:SignalsEyegazeFunctionalunitsratherthansequentialunitsSpeechactsQ.ConversationAnalysisvs.DiscourseAnalysis?Sacks,H./Schegloff,E.A./Jefferson,G.ASimplestSystematicsfortheOrganizationofTurn-TakingforConversationLanguage,Vol.50,No.4,Part1.(Dec.1974),pp.696-735tenHave,P(1999)DoingConversationAnalysis.London:SagePublicationsHutchby,IandWooffitt,R(1998)ConversationAnalysis.Cambridge:PolityPressHowtotranscribe?Sowhereoughtwestop?AusabletranscriptTheywereoriginallydevelopedbyGailJeffersonintheearlydaysofofCA,andhaveprovedsorobustandanalyticallyusefulthattheyhavebecomeindispensable.http://www.linguistics.ucsb.edu/projects/transcription/representing