Acton and Christopher Potts Abstract Drawing on previous analyses of the social meaning of demonstratives and other function words we argue that the semantics of demonstratives facilitates affective uses that can be characterized as attempts by the ID: 75524
Download Pdf The PPT/PDF document "That straight talk Sarah Palin and the s..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Inthenextsection,wereviewLakoff'sclaimsinmoredetailandsynthesizethemwithmorerecentcommentariesonthesocialeffectsofdemonstratives(Chen,1990;BowdleandWard,1995;Wolter,2006;Liberman,2008,2010;PottsandSchwarz,2010;DavisandPotts,2010).Thisleadstothecorehypothesisthatdemonstrativescanbeusedasaresourceforfosteringasenseofcom-mongroundandsharedperspectivebetweeninterlocutors—aneffectwetracetothesemanticsandpragmaticsofdemonstratives,whichrequireaspecialdegreeofcoordinationandperspectivalalignmentbetweendiscourseparticipantsinordertorefersuccessfully.InSection3,wetestforthiseffectwithaquantitativeanalysisofdemonstrativeusageacrossover200,000user-suppliedtextsfromtheonlinesocialnetworkExperienceProject(EP).Atthetimeofourdatacollection,EPallowedcommunitymemberstoannotateeachother'stextswith(amongotherthings)markersofsharedexperienceandperspective.Thisprovidesuswiththemeanstoidentifyandstudythesocialeffectsofdemonstrativesonalargescale,acrossmulti-plelinguisticexchanges.Wendthat,aspredicted,demonstrativeusageonEPcorrelateswithresponsesexpressingasenseofcommongroundandsharedperspective.TheEPanalysissupportstheclaimthat,acrossavastnumberoflanguageusers,demonstrativescanbeusedtoenhancediscourseparticipants'senseofsharedperspectiveandcommonground.Weseektobalancethishigh-levelperspectivewithacloseexaminationofuseattheleveloftheindividual,situatedinthecontextofabroaderpersona,linguisticstyle,andsociallandscape.Tothatend,inSection4,weturnourfocustoSarahPalin.Palin,theoncelittle-knowngovernorofAlaska,rstreceivedwidemediaattentionwhenshewasnamedSenatorJohnMcCain'srunningmateinthe2008U.S.presidentialelectionand,alongwithher`informal,'`colloquial'speechstyle,hasbeenthesubjectofmuchpubliccommentaryanddisagreementeversince.BuildingonworkbyLiberman(2008,2010)andMcWhorter(2010),weshowthatPalinisaprolicanddistinctiveuserofdemonstratives,andwearguethatthisispartofhermoregenerallinguisticstrategyofportrayingherselfasbeing`onewiththe“everydayAmericanpeople”'(Pinker2008;seealsoLabov2008).Palinisanidealcase-studynotonlybecauseofherpublicprominenceandfrequentuseofdemonstratives,butalsobecausethepoliticalrealminwhichsheoperatesstandsincontrasttotheEPcommunity.Thelatterisagenerallysupportiveenvironment,aplaceforuserstoshareandconnect,whereappealstosharedperspectivearelikelytobewellreceived.Theworldofpolitics,however,isaplaceofdivisionandopposition,andreactionstoPalin'sspeechareimpassionedandpolarized(Section5).Listenerswhosharehersocialandpoliticalattitudestendtoregardherlinguisticstyleasrelatableandwarm,whereaslistenerswhodisagreewithhertendtondherspeechgratingandobjectionable.Thishighlightsthefactthat,whiledemonstrativessupportattemptstofosterasenseofsharedperspectiveandcommonground,theydonotguaranteesuccessinthissocialstrategy.Onthecontrary,aspeaker'sdemonstrativescanbesociallyrepellentiftheyclashwiththeaddressee'sconceptionofherrelationtothespeaker.Thatis,theoverallsocialandpragmaticeffectsarefacilitatedbytheunderlyingsemantics,buttheyareultimatelydeterminedjointlybythediscourseparticipantsduringinteraction(Clark,1996).2AFFECTIVEDEMONSTRATIVESTheideathatdemonstrativeusebearsonthesocialrelationsbetweeninterlocutorsgoesbackatleastasfarasLakoff(1974)(seealsoFillmore1975).Lakoffclaimsthat,inadditiontotheir2 thisbook?')andpronominaluses(`Haveyoureadthis?').Pronominalones,whichlackanNPexpressingdescriptivecontent,typicallydependmoreheavilyoncontext.Toseethesignicanceoftheperspectivalrequirementsofdemonstratives,wecancontrast(5)abovewithademonstrativeversion:(6)Couldyouturnaroundandgetmethatboxofcereal?Here,unlikethecaseof(5),thedescriptivecontentoftheNPalonemaynotbeenoughtoestablishreference,andElaineshouldnotassumethatthereisonlyoneboxofcerealbehindher.For,asstatedabove,demonstrativesdon'tpresupposethatthereisauniqueindividualinthecontextsatisfyingthedescriptivecontentoftheNP.Instead,theypresupposethatthereisaparticularindividualthatsatisesthedescriptivecontentoftheNPandthattheaddresseeisequippedtoidentifyitbyconsultingthespeaker'sperspective.IfJerryutters(6),theremaybemultiplesalientcerealboxesinthecontext,buthepresupposesthatElainecanidentifytheoneheintendsbasedontheirsharedunderstandingofthephysicalenvironment,thepriordiscoursecontext,Jerry'stastesandattitudes,andsoforth.Thisdependenceonspeakerperspectiveisaprimaryforceinexplainingthesocialeffectsofinterest.First,theveryactofconsideringaspeaker'sperspective,oftryingtoseethingsasshedoes,providesanopportunityfortheaddresseetodevelopasenseofempathyandmutualunderstandingwithher.Second,aspeaker'suseofademonstrativeindicatestotheaddresseeabeliefthatthetwohaveenoughsharedexperienceandperspectivethatitisreasonabletoexpecttheaddresseetobeabletoconsiderthespeaker'spointofview.Thus,wherefelicitous,theactofinvokingandrelyingonperspectivalalignmentinusingademonstrativecanengenderorreinforceasenseofsharedperspectivebetweeninterlocutors.AswewillseeinthecaseofSarahPalin,however,wheretheconditionofperspectivalalignmentisnotmet,orsimplydeemedunmeritedorpresumptuousbytheaddressee,theuseofademonstrativecanbejarring;notonlymightthereferentialactfail,butdisparitiesinperspectivearebroughttothefore.Basedonthetheoreticaldiscussionabove,alldemonstrativeuseshavethepotentialtooccasionthesocialeffectsofinterest.Witheachuseofademonstrative,thespeakerinstructsheraddresseetoconsiderherperspectiveandindicatescondencethattheircommongroundandperspectivearerichenoughfortheaddresseetodoso.Atthesametime,however,certaindemonstrativesdoseemtopackastrongershared-perspectivepunchthanothers.Weidentifytwoleadingcontributors.First,usingademonstrativetointroduceintothediscoursesomethingthatisnotphysicallypresent,asin(3)and(4b),mightbeexpectedtoaddtothedemonstrative'ssocialimpact.Suchusespresupposethattheaddresseecanaccessthespeaker'sperspectiveinarelativelybroaddo-main,onethatextendsbeyondtheconversationtheyareengagedinandtheirimmediatephysicalenvironment,andhencethatthespeakerandaddresseehaveanontrivialdegreeofcommongroundandperspectivalalignment.Second,wherethedemonstrativeismorphosyntacticallyunnecessaryforthepurposesofse-curingreference,itmay,bybasicGriceanprinciples(Grice,1975),comeoffasespeciallyaffec-tive.Perhapstheclearestcaseofthiskindistheuseofademonstrativeasthedeterminerofapropername,asin(2)above.InEnglish,propernamesneverrequireasyntacticallyindependentdeterminer,sousingoneisalmostalwaysamarkedchoice:itmakesthephraselongerwithoutcontributingdirectlytotheactofreference.Thus,thedemonstrativeinanexamplelike(2)isref-erentiallysuperuous.Thespeakerof(2)isineffectgoingoutofherwaytouseademonstrative.Butaddresseesgenerallydonotexpectapersontosaymorethanisnecessaryfornoreasonatall5 thecontentisattimesembarrassing,sad,shocking,funny,heart-warming,andmixturesthereof.Atthetimeofourdatacollection,EPcommunitymemberscouldrespondtoindividualconfessionsbypostingtextcommentsor,separately,bychoosingfromvedifferentpresetreactioncategories:`yourock';`teehee';`Iunderstand';`sorry,hugs';and`wow,justwow'.Table1providessomesampleconfessiontextswithassociatedreactiondistributionsandreadercommenttexts.Atthesite,eachofthereactioncategorieswasassociatedwithacartoonfaceconveyingsomethingaboutthesocialmeaningofthesecategories.`Yourock'showedasmileyfacewithsunglasses,indicatingapositive,cheerfulreaction.The`teehee'faceseemedtobesmilingevenmorebroadly.The`Iunderstand'emoticonwasalsosmiling,butwithaknowingwink.`Sorry,hugs'wasportrayedwithafrown.Andthe`wow,justwow'facelookedshocked,aghast.Themeaningsofthesereactioncategoriesarevague,sociallycomplex,andparticulartotheEPcontext.Togainadditionalinsightintowhattheymeanttothecommunitymemberswhochosethem,wecanexaminethenumerousreadercommentsontheconfessiontexts.Thesecommentsarenottiedtospecicreactioncategorychoices,butweexpectgeneralalignmentbetweenthecommenttextsandthedistributionofreactioncategorychoicesforagivenconfession.Forex-ample,therstconfessioninTable1elicitedprimarily`Iunderstand'reactions,anditscommentsprimarilyexpressunderstanding.Themiddleconfessionreceivedamajorityof`teehee'responses,anditscommentsmostlyexpressamusement.Thethirdconfessionelicited`yourock'responsesonly,andthecommentsarecheeringandsupportive.Foramorecomprehensiveperspectiveonthesocialmeaningofthereactioncategories,wecanstudythebroaderpatternsinhowthecommentsandreactioncategoriesalign.Todothis,webuildasimpleprobabilitymodel.Ifxisaphraseandcisoneofthereactioncategories,thenCount(x;c)isthenumberofcreactionsfortextscontainingx.Forexample,ifTable1isourentirecorpus,thenCount(kudos;`yourock')=22,becausekudosoccursonceintwocommentsassociatedwiththethirdconfession,whichreceived11`yourock'reactions.Fromthesecounts,wecalculateP(xjc)=Count(x;c)=Px0Count(x0;c),wherex0rangesoverallphrasesofthesametypeasx.Thesenumbersaretinyandhardtocomparewitheachother,sowemakethemmoreintuitivebynormalizingtoobtainP(cjx)=P(xjc)=Pc0P(xjc0),wherecrangesoverthevereactioncategories.Thisdetermines,foreachphrasex,aconditionalprobabilitydistributionoverthereactioncategories. `sorry,hugs'`yourock'`teehee'`Iunderstand'`wow,justwow' 1foryourlossyourock!haha!usedtofeelafavorand2sorryaboutyourawesome!!lmao!!amgoingthroughwouldyoufeel3sosorryforhappybirthday!mademelaughiunderstandcompletelytoyourwife4amsosorryhappyforyoumademydayyouleastexpectthatwaybut5sorryforyouriagree!lmfao!!exactsamewaybetweenthetwo6imsosorryi'mhappyfortobecarefuli'mthesameyoufeelif7sosorrytotooshorttohahahathetimeiswouldyoudo8getthroughthistocomeoutihaven'thadamthesameicanjust9andyourfamilythankyou!onthecouchfeelingthesameshameonyou10i'msorryforkeepuptheon!!samewayandsexwithher Table2:Topcommenttrigramsbycategory.8 (a) (b)Figure1:DemonstrativesintheEPdata.usesofthatwithcomplementizerandrelativizeruses(anon-issuefortheotherdemonstratives).Similarly,thoughtheeffectisrobustforphrasaltheseandthose,wedonotgetevidencefortheeffectfortheirpronominaluses.Weofferthefollowingtwoexplanations.First,pronominaltheseandthoseareinfrequentinthecorpus;therearefewerthan4,300tokensofeach,comparedwithmorethan50,000tokensofpronominalthisandthat.Moreinterestingly,wemaybewit-nessingsomethingofatrade-offintheaffectiveproleofthepluralpronominalforms,whichshowhighlyelevated`yourock'reactions,perhapsrelatedtotheexclamativityeffectsidentiedbyLakoff(1974)andstudiedbyPottsandSchwarz(2010)andDavisandPotts(2010).5Theseconsiderationsnotwithstanding,theoverallpictureisclear:demonstrativeuseinEPconfessions,onthewhole,isstronglyassociatedwith`Iunderstand'reactions.Figure1suggeststhatphrasalandpronominalusesareequallylikelytostimulate`Iunderstand'reactions.However,wethinkitdoesnotfollowfromthisfactthatthetwoareequallycapableofsupportinghighlyaffectiveinterpretations.AttheendofSection2,wearguedthattheaffectivestrengthofauseofademonstrativedependsinpartonthenatureoftheexpressionsthatcouldhavebeenusedinitsplace.Thissuggeststhat,togainafullerunderstandingoftheemotiveeffectsofdemonstratives,weshouldlookatthemwithrespecttotheircomparisonclasses:alldeterminersinthecaseofphrasaldemonstratives,andallpronounsinthecaseofpronominalones.Todothis,weconductedtwosamplingexperiments.Fortherst,summarizedinFigure2(a),wecalculatedallofthevaluesP(`Iunderstand'jdet)forallofthedeterminersdetrepresentedinourdata(excludingdemonstratives).CallthevectorofvaluesV.Then,wesampledfourvaluesfromVandtooktheirmean,repeatingthissampling10;000times.ThedistributionoftheresultingvaluesisdepictedasahistograminFigure2(a).Forreference,wehavealsoincludedthe97.5thpercentilemark.Ifwedenethenullhypothesisassayingthat,incomparisontootherdeterminers,phrasaldemonstrativeshavemerelyaverageassociationwith`Iunderstand',thenthisexperimentgivesgoodreasontorejectthathypothesis,sincethedemonstrativeestimate(markedwith`DEMONSTRATIVES')isabovethismark.Thesecondexperiment,summarizedinFigure2(b),followedthesameprocedure,butthecom-parisonvectorVcontainedthevaluesP(`Iunderstand'jpro)foreachpronominalformpro(ex- 5Thedataandcodeforalltheanalysesinthispaperareavailablefromtheauthors'websites.10 (a)Phrasaldemonstrativescomparedwithotherdeterminers. (b)Pronominaldemonstrativescomparedwithotherpronouns.Figure2:Samplingexperiments.12 (10)`...somebodywhowantstoprotectourconstitution...shouldspeakoutonbehalfofthisindividual,whosebusinessisbeingharmedbythosewhoareintolerantandarebigotedandarehypocriticalbecausetheydon'tagreewiththisman'spersonalopinionandthesentimentthatheshared...'7Inall,theaveragedemonstrative-phraselengthforPalinwas1:5wordsgreaterthantheaveragefortheothers(4:7and3:2words,respectively).Thesedifferencesinlengthbearontherelativeaffectivityofthetwogroups'demonstratives.Crucially,longerdemonstrativephrasesgenerallypackinmorecontentthanshorteronesasregardstheidentityandcharacteroftheintendedref-erent.Thishasanumberofrelevantconsequences.Forone,itmeansthatPalin'sdemonstrativedeterminersinthecorpusaremorelikelytobeintroducinganewreferentintothediscourse,be-causelongerreferringexpressions,beingmoreinformative,carryinformationthatwouldlikelybesuperuousinanaphoricusage,butusefulinrstmentionsofareferent(WardandBirner,2004).TheexamplesfromPalinin(11)illustrate.Therst,two-worddemonstrativephraseisanaphoric,madewithreferencetoaquestionaboutheruseoftheterm`bloodlibel.'Thesecondintroducesanewreferentintothediscourseandishardtoimagineasbeinganaphoric.Thedemonstrativephrasesin(12),fromMicheleBachmannandElisabethHasselbeck,respectively,portraythesamekindoflength/newnesscontrast:(11)a.`Ithinkthecritics,again,wereusinganythingthattheycouldgatheroutofthatstatement...'8b.`But,youknow,Ithinkoneofmyproblemsinthiswholeprocessis,Idon'tliveforthatgameofthepunditcy,oftheopiningandspeculatingonwhoisdoingwhat...'9(12)a.`Itisastateissueandit'safederalissue.It'simportantforyourviewerstoknowthatfederallawwilltrumpstatelawonthisissue.'10b.`Butunfortunately,whenyougetintothisgrandschemeofsortofmediatelevisionetcetera,theconservativewomanisnowjudged.'11AsnotedinSection2,demonstrativesthatintroducenewreferentsintothediscoursehaveespe-ciallyhighaffectivepotentialbecausetheypresupposecommongroundandsharedperspectivethatextendsbeyondthecontextoftheconversation.ThesecondconsequenceofPalin'slongerdemonstrativephrasesisthatherdemonstrativesaremorelikelytobemorphosyntacticallyunnecessaryand,hence,asnotedbefore,moreaffective:(13)`Well,anybodywhoisaprotection—somebodywhowantstoprotectourConstitution,allofourconstitutionalrights,includingthatfreedomofspeech,shouldspeakoutonbehalfofthisindividual...'12 7OntheRecordwithGretaVanSusteren,July31,2012.8Hannity,January17,2011.9Hannity,May18,2011.10MicheleBachmannonFoxNewsSunday,June26,2011.11ElisabethHasselbeckonHannity,January19,2011.12PalinonOntheRecordwithGretaVanSusteren,July31,2012.16 traitsidentiedinSection2ascorrespondingtohighsolidarity-relatedaffectivity.Inaddition,weprovidedevidencethatPalin'sdemonstrativesaremorelikelytocarryevaluativecontentandthuspresupposegreatersharedsentimentbetweenspeakerandaddressee.WenowconsiderreactionstoPalin'sspeechinlightoftheforegoingdiscussion.OuraimistoshowthatPalin'sdemonstrativeusagehelpsexplainthepolarizednatureofthesereactions,and,atthesametime,thatthenatureofthesereactionsenrichesourunderstandingofthesocialmeaningofdemonstratives.5REACTIONSTOSARAHPALIN'SSPEECHTheprevioussectionshowedthatPalin'sdemonstrativedeterminersareunusuallyfrequentandunusuallyaffective.Thisaspectofherspeechhasbeenthesubjectofprofessionalcommentary,mostnotablyinLanguageLogpostsbyLiberman(2008,2010)andinaneditorialbyMcWhorter(2010),butithasnot,toourknowledge,beenwidelydiscussedbynon-linguists,oratleastnotexplicitlyso.Wetracethisnottoalackofsaliency,butrathertothefactthatitischallengingfornon-linguiststoconsciouslyidentifydemonstrativeusageasanelementoflinguisticstyle.Forexample,Liberman'sandMcWhorter'spostsreceivednumerouscomments;oncethelinguistshadidentiedthephenomenonandprovidedsometerminology,peoplewereabletoprovideanddiscussrelevantexamplesfromPalin'sspeech.ClearevidenceofthesalienceofPalin'sdemonstrativescanbefoundinparodiesofherspeech.OnthesketchcomedytelevisionshowSaturdayNightLive,actressTinaFeyimpersonatedPalininthe2008vicepresidentialdebate,withactorJasonSudekisasheropponentJoeBiden.Purnelletal.(2009)observethatFeyemployedsomeofPalin'sphonologicalfeaturesinherimpersonation,includinghighratesofthecoronalrealizationof-ing,andfrontingof/O/,butshedidnotincorporateothers,suchasprelateralmerger.Theauthorstakethisaspotentialevidencefortheperceptualrelevanceoftheformerrelativetothelatter.Inasimilarvein,wenotethatFeyasPalinusedtwelvedemonstrativedeterminers.Someexamples:(15)a.`Youknow,JohnMcCainandI,we'reacoupleofmavericks.And,goshdarnit,we'regonnatakethatmaverickenergyrighttoWashington.'b.`Gwen,wedon'tknowifthisclimatechangehoozie-what's-itisman-madeorifit'sjustanaturalpartofthe“EndofDays.”'c.`Oh,andforthoseJoeSix-Packsoutthereplayingadrinkinggameathome—Maverick.'FeyasPalinspoke620wordsinthesketch,forademonstrative-determinerrateof12/620=0.019.Incontrast,Sudeikis,asBiden,usedonlytwodemonstrativedeterminers,bothofwhichhadcleardiscourseantecedents.Hisrateofdemonstrativedeterminerswasjust2/491=0.004.Similarly,GarryTrudeau,inhisNovember29,2008,editionofDoonesbury,playsupPalin'sdemonstrativeuse,withoneaffectivedemonstrativeforeachofthecartoon'sfourcells.ThetextofthePalincharacter'sutterancesfromthiscomicstripisgivenin(16);thecontextisacharacterwatchingherontelevision.18 thelanguagebothdependsuponcertainsocialconnectionsandmakesthemmoresalient,itcomestooccupyanimportantplaceinthesocialdynamic,asbothasignalofthesocialconnectionsandareinforcerorevenprogenitorofthem.6CONCLUSIONDemonstrativesmightseematrsttobeunlikelyvehiclesforestablishingandmaintainingsocialbonds,sincetheyappeartobeblandfunctionalelements.However,whenwelookcloselyattheircontext-dependentsemantics,wendrichpotentialforsocialmeaning.Inparticular,wehaveshownthattheuseofdemonstrativesbothpresumesand,whenwelcome,reinforcesasenseofsharedperspectivebetweeninterlocutors.Itisthisdynamic,weargue,thatexplainswhylinguistsandlexicographersalikehavecharacterizedcertainusesofdemonstrativesas`colloquial'(Lakoff,1974)or`informal'—justliketaboowords,certainphoneticfeatures,termsofaddress,etc.,affectiveusesofdemonstrativesrequireadegreeoffamiliarityandfellowshipbetweenspeakerandhearertobelicensed.WehavealsoshownthatSarahPalinisanextraordinaryuserofdemonstratives.Thoughaf-fectivedemonstrativesareawidelyusedstylisticresource,Palinusesthemwithespeciallyhighfrequency,inespeciallyaffectiveways.Inturn,Palin'sspeechtendstopresupposeagreatdealofsharedperspectiveandcommongroundwithheraudience.ThishelpsexplicatetheparticularwaysinwhichPalin'svoiceissodivisive:thosewhoarealignedwithherpoliticalandculturalattitudeshearhertoneasgenuineandwarm,whereasthosewhodisagreewithherviewshearherstyleasdisingenuousanduntoward.Asnotedfromtheoutset,demonstrativesarefarfrombeingtheonlylinguisticresourcesavail-ablefor,toborrowLakoff'sphrase,`achievingcamaraderie.'Thus,underouranalysis,weshouldexpecttondthattheuseofaffectivedemonstrativescorrelateswithotherresourcesofthiskind.Analysisofsuchcorrelationsisbeyondthescopeofthispaper,butweofferafewpreliminaryndingsasarststepinthatdirection.First,concerningPalin'suseofinformaleuphemismslikeheckanddarn,pointedoutbyPurnelletal.(2009)amongothers,wendinourcorpusofFoxNewsinterviewsthatthreeofPalin'sveusesofheckhaveatleastonedemonstrativephraseintheverysamesentence,asin(19),andoneoftheothertwousesoccurswithademonstrativephraseintheimmediatelyprecedingsentence.(19)`This“ObamaCare”mandate,thisdecisionthatwasmadetoday–heck,ObamaevenproposingitandrammingitthroughanddownourthroatsthroughPelosiandReid'sem-bracingofit–that'saharbingerofthingstocome.'22Similarly,oneofPalin'stwousesofdarnisaccompaniedbyademonstrativephraseinthesamesentence.Second,wehavepreliminaryevidencethatPalinisarelativelyfrequentuserofrst-personpluralindexicalslikewe,us,andour,whichTalbot(1992)arguescancontributetoasenseofclosenessandalignmentbetweenspeakerandhearerbyreferringtobothaspartofthesamecol-lective.23Forinstance,in(20),fromthe2008vicepresidentialdebate,Palinusesverst-person 22OntherecordwithGretaVanSusteren,June28,2012.23Suchindexicalsmaybepolarizinginmuchthesamewayasaffectivedemonstratives:thosewhofeelalignedwiththespeakermaybewarmedbysuchlinguisticgesturestosolidarity,whilethosewhoseethemselvesasbeingatoddswiththespeakermayndthempresumptuous.21 Baayen,R.Harald.,DouglasJ.DavidsonandDouglasM.Bates.2008.Mixed-effectsmodelingwithcrossedrandomeffectsforsubjectsanditems.JournalofMemoryandLanguage59(4).390–412.Bowdle,BrianF.andGregoryWard.1995.Genericdemonstratives.InProceedingsofthetwenty-rstannualmeetingoftheberkeleylinguisticssociety,32–43.BerkeleyLinguisticsSociety.Campbell,SherlockR.andJamesW.Pennebaker.2003.Thesecretlifeofpronouns:Flexibilityinwritingstyleandphysicalhealth.PsychologicalScience14(1).60–65.Campbell-Kibler,Kathryn.2007.Accent,(ing),andthesociallogicoflistenerperceptions.Amer-icanSpeech82(1).32–64.Chen,Rong.1990.Englishdemonstratives:Acaseofsemanticexpansion.LanguageSciences12(2).139–153.Cheshire,Jenny.2005.Syntacticvariationandbeyond:Genderandsocialclassvariationintheuseofdiscourse-newmarkers.JournalofSociolinguistics9(4).479–508.Clark,HerbertH.1996.Usinglanguage.Cambridge:CambridgeUniversityPress.Davies,Mark.2008.TheCorpusofContemporaryAmericanEnglish:450millionwords,1990-present.Availableonlineathttp://corpus.byu.edu/coca/.Davis,ChristopherandChristopherPotts.2010.Affectivedemonstrativesandthedivisionofpragmaticlabor.InMariaAloni,HaraldBastiaanse,TikitudeJagerandKatrinSchulz(eds.),Logic,language,andmeaning:17thAmsterdamColloquiumrevisedselectedpapers,42–52.Berlin:Springer.Dixon,JohnA.andDonH.Foster.1997.Genderandhedging:Fromsexdifferencestosituatedpractice.JournalofPsycholinguisticResearch26(1).89–107.Elbourne,Paul.2008.Demonstrativesasindividualconcepts.LinguisticsandPhilosophy31(4).409–466.Erman,Britt.2001.Pragmaticmarkersrevisitedwithafocusonyouknowinadultandadolescenttalk.JournalofPragmatics33(9).1337–1359.Fillmore,CharlesJ.1975.SantaCruzlecturesondeixis,1971.Bloomington,IN:IndianaUniver-sityLinguisticsClub.Grice,H.Paul.1975.Logicandconversation.InPeterColeandJerryMorgan(eds.),Syntaxandsemantics,vol.3:SpeechActs,43–58.NewYork:AcademicPress.Horn,LaurenceR.1984.Towardanewtaxonomyforpragmaticinference:Q-basedandR-basedimplicature.InDeborahSchiffrin(ed.),Meaning,form,anduseincontext:Linguisticapplica-tions,11–42.Washington:GeorgetownUniversityPress.Jaeger,T.Florian.2008.Categoricaldataanalysis:AwayfromANOVAs(transformationornot)andtowardslogitmixedmodels.JournalofMemoryandLanguage59(4).434–446.Joshi,AravindK.1982.Mutualbeliefsinquestionansweringsystems.InNeilS.Smith(ed.),Mutualknowledge,181–197.London:AcademicPress.King,Jeffrey.2001.Complexdemonstratives:Aquanticationalaccount.Cambridge:MITPress.Kitagawa,Chisato.1979.Anoteon`sono'and`ano'.InGeorgeBedell(ed.),Explorationsinlinguistics:PapersinhonorofKazukoInoue,232–243.Tokyo:Kurosio.Klein,DanandChristopherD.Manning.2003.Accurateunlexicalizedparsing.InProceedingsofthe41stannualmeetingoftheAssociationforComputationalLinguistics,vol.1,423–430.Stroudsburg,PA:ACL.Labov,William.2008.Palin'saccentexamined.AllThingsConsidered,NationalPublicRadio.Lakoff,Robin.1974.Remarkson`this'and`that'.InProceedingsoftheChicagoLinguistics23 Thatstraighttalk:SarahPalinandthesociolinguisticsofdemonstrativesEricK.ActonandChristopherPottsDrawingonpreviousanalysesofthesocialmeaningofdemonstrativesandotherfunctionwords,wearguethatthesemanticsofdemonstrativesfacilitatesaffectiveusesthatcanbecharacterizedasattemptsbythespeakertofosterasenseofsharedperspectiveandcommongroundwithotherdiscourseparticipants.Wepresentlarge-scalequantitativeevidencethatthisstrategyiswidelyusedandcommunicativelyeffective.WethenconductafocusedcasestudyofthedemonstrativeuseofU.S.politicianandpublicgureSarahPalin,situatedinthewidercontextofPalin'spersona,style,andplaceinthesociallandscape.AnanalysisoftelevisioninterviewdatashowsthatPalinisadistinctiveandprolicuserofaffectivedemonstratives.Palin'susagehighlightsthecontext-dependenceofdemonstratives'socialmeaningandleadstoadeeperunderstandingofherrhetoricalstrategiesandthepolarizedreactionstheyhavereceived.Keywords:Demonstratives,socialmeaning,perspective,style,socialmedia,SarahPalin1INTRODUCTIONFunctionwordsandphrases,thenutsandboltsoflanguage,seematrsttobeanunlikelysourceforrichsocialmeaning,butpreviousresearchhastracedawidevarietyofpragmatic,stylistic,andperspectivaleffectstotheseexpressionsandtheroletheyplayininteraction,inbuildingcommonground,andinachievingreferenceincontext(TalbotStubbeandHolmesDixonandFosterCampbellandPennebakerMooreandPodesvaLakoff's()analysisofEnglishdemonstratives(,and)isanearlyandinuentialexampleofthislineofinvestigation.Lakoffsuggeststhatdemonstrativescanbeusedasatoolfor`achievingcamaraderie'(p.347)and`establishingemotionalclosenessbetweenspeakerandaddressee'(p.351),andclaimsthattheseeffectsarisefromthebasicsemanticsanduseconditionsofdemonstratives.ThecentralgoalofthispaperistosubstantiateandreneLakoff'scharacterizationofthiscomponentofdemonstratives'socialmeaning,throughacombinationofsemanticandpragmaticanalysis,large-scalecorpusanalysis,andasituatedcasestudyofoneparticularlydistinctiveuserofdemonstratives:U.S.politicianandpublicgureSarahPalin. OurthankstoSamBowman,PennyEckert,KatherineHilton,BenotMonin,RobPodesva,EwartThomas,andaudiencesatStanford,NWAV40,andNASSLLI5.Ourworkhasbenetedtremendouslyfromthecommentsofanonymousrefereesandtheeditorsofthisjournal.ThisresearchissupportedinpartbyNSFIIS-1159679.