ProblemswithGriceanmaxims??1•Themaximsallowderivingallthepossibleimplicaturesratherthantheonesintendedbythespeaker.Whenshouldthehearerstopderivingimplicatures?2•Themaximsdonothaveequalimportanceingeneratingimplicatures.QuantityandRelevanceareresponsibleformostimplicatures.3•Wemaynotneedthatmanymaxims.E.g.,Quantity2(‘Donotmakeyourcontributionmoreinformativethanrequired’)followsfromRelation(‘Berelevant’).4.thedistinctionbetween‘whatissaid’and‘whatisimplicated’(truth-conditional/non-truth-conditionalmeaning;semantics/pragmatics)isnotasclear-cutasassumedbyGrice.Post-Griceandevelopments1.neo-Griceans(e.g.Horn1984,1988;Levinson1995,2000):rearrangedGrice’smaxims,reducedtheirnumber.Implicaturesareexplainedastheresultoftheresolutionofconflictsbetweenmaxims2.RelevanceTheory(SperberandWilson1986;Carston2002):reducedGrice’smaximstoonecognitiveprinciple.ConversationalimplicatureIII.ScalarimplicaturesII.Horn'sQ-principleandR-principleIII.Levinson’sQ-,I-,andQ/M-principlesI.ScalarimplicaturestheimplicaturestobeinferredinaccordancewiththemaximofQuantityproposedbyGazdar(1979a)anddeveloped,byHorn(1972and1973).Inlanguagethereareanumberofwords,whicharebasedontheweaknessandstrengthoftheinformation,formascaleofvaluesexpressingquantity.e.g.all,most,many,some,fewInthisscale,anystrongerformentailstheweakerform,butanyweakerformimplicatesthestrongerformdoesnotobtain.Otherexamples??always,often,sometimesmust,should,maynecessarily,possiblycertainlythatp,probablythatp,possiblythatpthe,a(n)/someClausalimplicatures:theimplicaturesconcerningcomplexsentences.know,believeIbelievethereislifeonMars.IdonotknowthereislifeonMars.IknowthereislifeonMars.II.Horn'sQ-principleandR-principleQ-quantityR-relevanceHornsuggeststhattheCPcanbemodifiedsothatthe),proposedtwoprinciples.E.g.a)Thebabycriedandthemotherpickeditup.Themotherwasthemotherofthecryingboy.b)MartandDavidboughtapiano.MartandDavidboughtapianotogether.Theimplicaturesa)andb)cannotbeinferredbyapplyingtheCPbutbyastereotypicalrelation.(markedness)Q-principle:makeyourcontributionsufficient(Sayasmuchasyoucan)hearer’seconomyR-principle:makeyourcontributionnecessary(Saynomorethanyoumust)speaker’seconomyThesetwoprinciplesconflictwitheachother.TosolvethesetwoantitheseHornexplainedbyinvokingthenotionofmarkedness.Theuseofamarked(relativelycomplexanti/orprolix)expressionwhenacorrespondingunmarked(simpler,lesseffortful)alternateexpressionisavailabletendstobeinterpretedasconveyingamarkedmessage(onewhichtheunmarkedalternativewouldnotorcouldnothaveconveyed).E.g.a)Canyouopenthedoor?b)Areyouabletoopenthedoor?Q-principleandR-principle重要性有区别??RisfloutedforQIII.Levinson’sQ-,I-,andQ/M-principles1)Q-principleSpeaker'smaxim:Donotprovideastatementthatisinformationallyweakerthanyourknowledgeofthisworldallows,unlessprovidingastrongerstatementwouldcontravenetheI-principle.Speaker:asinformativeaspossible信息量充分Recipient'scorollary:Takeitthatthespeakermadethestrongeststatementconsistentwithwhatheknows.hearer:understandtothedegreethatthespeakersayswheneveraspeakerusesaweakerstatementwherehecouldhaveusedastrongerone,thiswillQ-implicatethatthestrongeronedoesnothold:E.g.Someofthestudentsshowed.b)Allofthestudentsshowed.SpeakerQ-implicatesthatb)doesnotholdandthereforetheimplicatureis??notallofthestudentsshowed.ThistypeofimplicaturecanbeexplainedbytheHorn-scalerule.2)I-principleSpeaker'smaxim(theMaximofMinimization)Sayaslittleasnecessary,i.e.producetheminimallinguisticinformationsufficienttoachieveyourcommunicationalends(bearingtheQ-principleinmind).Speaker:asefficientaspossible力求简洁Recipient'scorollary(theenrichmentrule)Amplifytheinformationalcontentofthespeaker'sutterance,byfindingthemostspecificinterpretationuptowhatyoujudgetobem-intendedpoint.Hearer:benefitasmuchaspossibleSpecifically:Avoidinterpretationsthatmultiplyentitiesreferredto(assumereferentialparsimony);specifically,prefercoreferentialreadingofreducedNps(pronounsorzeroes).a)Billhasacar.Thewindowdoesnotclose.ThewindowofBill'scardoesnotclose.Notsomeoneelse’sb)Johnenteredtheroomandsatonthechair.JohnenteredtheroomandJohnsatonthechair.Orderingofapplicationoftheprinciples:QMILevinson(2000)–restatedtheprinciplesasheuristics:QHeuristic‘Whatisn’tsaid,isn’t.’IHeuristic‘Whatissaidsimplyisstereotypicallyexemplified.’MHeuristic‘What’ssaidinanabnormalwayisn’tnormal.’Evaluation??1)attentionhasbeendrawntotheconversationalmaximsorprinciplesinrelationtolinguisticforms.2)provideuswiththenewmechanismofmakinginferences,thatis,makinginferencesbywhatiscalledstereotypicalrelation.3)Theneo-Griceantheoryshowsthatpeopletendtomaketheleastefforttoachievethemosteffectivepurposeinlinguisticcommunication.Thisisverymuchlikefollowingtheprincipleofleasteffortinhumanbehaviour.Toputitsimply,wetrytosayaslittleasnecessary,butinferasmuchaswecan.2.RelevanceTheoryIntroductiontotheauthorsDanSperberisaFrenchanthropologist,linguistandcognitivescientist,currentlyaResearchDirectorattheJeanNicodInstitute.WiththeBritishlinguista