Dispute Resolution Without Disputing How the International [PDF]

Dispute Resolution Without Disputing: How the Interactional Organization of Mediation Hearings Minimizes Argument Author(s): Angela Garcia Reviewed work(s): Source: American Sociological Review, Vol. 56, No. 6 (Dec., 1991), pp. 818-835 Published by: American Sociological Association Stable URL: http://www.jstor.org/stable/2096258 . Accessed: 20/02/2012 11:10 Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

American Sociological Association is collaborating with JSTOR to digitize, preserve and extend access to American Sociological Review.

http://www.jstor.org

DISPUTE RESOLUTION WITHOUT DISPUTING: HOW THE INTERACTIONAL ORGANIZATION OF MEDIATION HEARINGS MINIMIZES ARGUMENT* ANGELAGARcIA Universityof Wisconsin,Eau Claire Mediation is an institutional interactional system in which disputing parties discuss and resolve differences with the help of a third party. Conflicts can be resolved with minimal confrontationor argumentin part because mediation de-emphasizes the adversarial nature of the situation and encourages cooperation. By analyzing the interactional organization of mediation hearings I show how mediation promotes agreement and minimizes argument. Mediation accomplishes these goals by an interactional organization that constrains how accusations and denials are positioned andformulated.Because this organizationprecludes the use of disputing techniques routinely used in ordinary conversation, disputes can be discussed and agreement reached withoutargument.

Closer [1956] (1964) provisionallydefined cess.' Some argumentsmay lead to resolutionof conflictas "struggleovervaluesandclaims to scarce status,power and resourcesin which the aimsof theopponentsareto neutralize,injure oreliminatetheirrivals"(p. 8). Inhis studyof the latentfunctionsof conflict outlinedby Simmel [1908] (1955),Coser(1964)pointedoutthatconflict has manypositiveeffects on groups,instituandsocietiesas a whole,e.g., tions,organizations, by aiding groupformation(p. 31), contributing "to the maintenanceof group boundariesand preventing] the withdrawalof membersfrom a group"(p. 8), creatingbalancebetween groups in a social system(p. 34), andhelpingto "maintain the total system by establishinga balance betweenits componentparts"(p. 35). Although conflict clearly has positive functions,thesecan oftenbe realizedonly if the conflict is ultimatelyresolved.Institutionalizeddispute resolutionprocedureslike trials, counseling, and mediationare specificallyorganizedto manageconflictsthatoccurwithinsocieties and groups. In ordinaryconversation,disputingand dispute resolutionmay be one and the same pro-

the conflict, e.g., by one side giving in (Vuchinich 1990)- othersmay not.People seek institutionalconflictresolutionprocedureswhentheir disputes are not resolved by informal means. Therefore,to be effective institutionalizedconflict resolutionproceduresshoulddifferin some way frominformaldisputing.The parameterson which these proceduresdiffer include the roles of the participants,authority/powerof specific roleholders,normativeexpectations,procedures followed, and interactionalorganization. I use mediationhearingsas a case studyof an institutionalconflict resolution procedureand comparethem to the interactionalorganization of disputing in ordinaryconversation.I argue that institutionalconflict resolutionprocedures (e.g., mediation)resolve conflict by eliminating specific conflictualprocesses from the interaction.In particular,the type of arguingthatoccurs in ordinaryconversationby participantsin informal disputesis eliminatedin the mediationprocess. Mediationaccomplishesthisby implementing an interactionalorganizationthatconstrains the presentationof accusationsanddenials,pro* Direct all correspondenceto Angela Garcia,De- vides opportunitiesfor disputantsto respondselectively to accusations,and providesfor mitipartmentof Sociology andAnthropology,University formulationof accusationsand denials. gated of Wisconsin,Eau Claire,WI 54702. I thankSteven Because this organizationprecludesthe use of Clayman,Douglas Maynard,Neal Slone, Don Zimdisputing techniquesused routinelyin ordinary merman,and the anonymousASRreviewersfor providing commentson earlierversionsof this paper.In addition,I'd like to thankthe mediatorsand media- Grantfromthe Universityof Wisconsin,Eau Claire. tion clients for allowingme to observeandvideotape ' Sacks,Schegloff,andJefferson(1974) distinguish their hearings.Part of the work for this paper was two types of talk:institutionaltalk and informaltalk done undera Time ReassignmentIncentiveProgram (or "ordinaryconversation"). 818

AmericanSociological Review, 1991, Vol. 56 (December:818-835)

DISPUTERESOLUTIONIN MEDIATION conversation,issues can be discussedandagreementreachedwithoutargument.Whenarguments do occur,mediationprovidesan interactionalorganizationfor terminatingthem that is absent fromordinaryconversation. DATA Disputantsin mediationhearingsnegotiate an agreementwith the help of a thirdpartyrather thanby handingdecision-makingauthorityover to a thirdpartyas in small claims court.Mediation emphasizes cooperationand compromise (Worleyand Schwebel 1985) and de-emphasizes theadversarial natureof disputing,whichtends to be exacerbatedin litigation (Girdner1985). Practitionersbelieve thatmediationreducesthe antagonismbetween disputants,gives them an opportunityto listento andunderstandeach other's positions,andpromotesreconciliation(Bottomley 1985,p. 162;Dingwall 1986,p. 10;Roberts 1988, p. 538; Folberg1983, p. 9). I evaluatea mediationprogramsponsoredby a Californiacounty.This programservesas an alternativeto small claims courtfor disputessuch as landlord-tenantdisputes, monetarydisputes involving small sums of money, and disputes betweenneighborsor family members. I videotapedall hearings(ninetotal)held during a six-monthperiodin 1987. In two cases the disputantshad additionalproblemsafterthe initialhearinganda secondmeetingwas held.Since the structureof these two follow-up meetings differedfrom the initialmediations,the followups arenot analyzed.Disputantsweretold about thestudybeforeeachhearingbeganandall agreed to be videotaped.Althoughthe recordingequipmentwasvisible,its intrusiveness wasminimized. The mediatorsin this ongoing programare volunteersfrom the communitywho receive five daysof trainingfromthedirectorof theprogram. The nine hearingsinvolved a total of 43 people (includingmediators,disputants,andwitnesses) andrangedfrom40 minutesto almostthreehours in length.Morethan20 hoursof audiotapewere transcribedusing a modifiedversionof GailJefferson's transcriptionsystem (see Atkinsonand Heritage1984, pp. ix-xi). Mediationprogramsvary greatly in theirorganization. In this program, each hearing is chairedby two mediatorswho open the hearings by making introductions,describingthe rules, andgettingthemediationagreementsigned.2The 2 The seventhhearingwas assignedonly one mediator.

819 mediatorsthen solicit extensive storiesfromthe disputants,and then lead a discussion period. When a solutionis reached,the mediatorswrite up the agreementandclose the hearing.The disputants(referredto as complainantand respondent)representthemselves.If a mediatedagreement is not reached, the mediatorsarbitratea decisionfor the disputants. THE ORGANIZATIONOF DISPUTINGIN ORDINARYCONVERSATION To understandhow mediationprovidesan interactionalstructurethatminimizesdisputing,it is necessaryto understandhow peopleconductdisputesin everydaysituations.3The disputingprocess dependson the speechexchangesystemof ordinaryconversation.Speechexchangesystems (Sacks et al. 1974) are unique interactionalorganizationsthatcanbe distinguishedon thebasis of theturn-taking systemandparticipation frameworkemployed. The turn-takingsystemas describedby Sacks et al. (1974), consistsof the rulesandprocedures participantsuse to exchangeturnstalking.In ordinaryconversation,turntransitionproperlyoccurs at the end of a "unittype" (e.g., sentence, clause,orphrase).Speakerscan selectanotherto speak(e.g., by askinga question).If the current speakerdoes not select a next speaker,any participantmay speak. If a next speakerdoes not self select, the currentspeaker may continue. Turnsat talkingand types of turns(e.g., questionsandanswers)arenot predetermined or controlledby conventions,structures,or individuals outsidethe interaction,but are negotiatedin the contextof the talkitself. For any given interaction,the participation frameworkdescribeswhatpatternsof participation and addressoccur(Goffman1981).Participants in ordinaryconversationnegotiate their participationstatus(e.g., ratifiedparticipantvs. bystander,addressedvs. nonaddressedrecipient) in the contextof the talk. Many of the arguingtechniquesused in disputesderivein partfromthe interactionalorganizationof ordinaryconversation.Researchon arguing in ordinaryconversationin informalsettings shows thatit involves adjacent,directlyaddressedexchanges of oppositionalturns.Techniqueschildrenuse to disputeincludeaggravated 3 The organizationof talk in institutionalsettings derives from the organizationof ordinaryconversation (Greatbatch1988, p. 402; Sacks et al. 1974, pp. 730-31, Heritage1987, p. 257).

AMERICANSOCIOLOGICALREVIEW

820

Excerpt1: 1 Stan: Speakersare identifiedat the left with the initials of their 2 Karen: institutionalroles: C is the complainant;R is the respon3 dent;M is the mediator. 4 Stan: If morethanone personholds a specific role, the partiesare 5 labelledA and B (e.g., MA is MediatorA). The transcripts have been simplifiedfor ease of reading.Pseudonymshave 6 Stan: been used for all propernouns. 7 Karen: Definition Symbol 8 Inhalationsand exhalations,respectively .hh hh 9 Stan: ta::lk Colons indicatea syllable is drawnout 19 Karen: thatDash indicatesa word was cut off abruptly 11 Table 1. Conventionsfor TranscribingConversations

lot YOU 0cost0 (1.4) (talk)

Underliningindicatesstress or emphasis Capitallettersindicateincreasedvolume Degree symbolsindicatedecreasedvolume Numbersin parenthesesindicate length of pauses (in seconds) Wordsin parenthesesaretentativetranscriptions. Empty parenthesesindicate nontranscribable talk. Punctuationindicatesintonation,not grammaticalstructure. Laughterparticles are transcribedas pronounced. Bracketsindicatesimultaneousspeech.

heh, hunh, henh A: [a copy] B: [I have] A: yeah= Equalsigns indicateone utteranceor word B: =in order is attachedto another. Arrows point out lines in excerpts that illustrateconcepts describedin text

)= I wantto talk to you( =I DI:DN'T:(.3) HAVE ANY THING,= =YOU HAD (RIGHT)TO DO WITH=IT! [(YOU ARE ALWAYS)] [YOU KNOWTHAT IS] BULL I DIDN'T [YOU ALLOWEDIT] [( see it )]=I DIDN'T EVEN DO THATCRAPI DIDN'T SEE THAT.

Stan begins this sequence in a normaltone of voice (line1).Karenescalatesthevolumeto a yell in her response(lines 2 and 3). Stan's next turn matchesthe volume of Karen'sresponse.They maintainthis high volume untilline 10 in which Karen escalates the volume to a high-pitched scream. This excerpt also illustratesthe exchange of accusationsand denials.In lines 2 and 3 Karen disclaimsresponsibilityfor the problemStanhas complainedabout:"Ididn'thaveanything."Stan countersthisclaim in lines 4 and5 withan oppositionalutterance:"Youhad rightto do with it." Karen overlaps this utteranceto deny Stan's accusation:"Youknow thatis bull I didn't...." Suchexchangesof oppositionalutterancesoccur frequentlyin ordinarydisputes(also see Coulter 1990, p. 184). These disputingtechniquesrequirethe flexible speechexchangesystemof ordinaryconversationwherea disputantcan place a responseto an utteranceadjacentto that utterance.For example, to producean aggravateddisagreement, thedisagreementmustbe placedadjacentto what is being disagreedwith. A disagreementthat is delayedor displacedis a mitigateddisagreement (Pomerantz1975, 1984). Such disputingtechniquesareusedin theturn-takingsystemof ordinaryconversationwhich does not limit when a given partycan takea turn. Second, the speech exchange system of ordinaryconversationdoes notrestrictwhocanspeak to whom.-Thus, disputants typicallyusetechniques like repetition,escalation,andinversionin utterances addressedto the otherdisputant.This may is thereleadtoescalationbecausetheco-disputant by selectedas next speaker,givinghimor herthe floortoproducea disputational response.4Insum,

ratherthanmitigateddisagreements(M. Goodwin 1983), repetitionof the previousspeaker's utterance(incorporatingpartialrepeatsinto the oppositionalutteranceis referredto as "format tying"by GoodwinandGoodwin[1987]), escalation of volume, acceleration,and denying or negatingthepreviousspeaker'sutterance(referred to as "inversion"by BrenneisandLein [1977, p. 56-57]). Researchon adultarguingshows thatit follows similarpatterns.Forexample,the "contrastivelymatchedcounter"used by adultsin argumentsequences(Coulter1990,p. 195)is analogous to the "formattying"found by Goodwin andGoodwin(1987) in children'sdisputes. Seriousdisputes,or"verbalfights,"showmore clearlytheparallelsbetweenchildren'sandadults' disputingtechniques.A tape recordingof a domestic disputebetweenthreeadultswas played as evidencein a mediationhearing.In Excerpt1, Stanis arguingwithhis ex-wife over who wrote I This observationis consistentwith data cited in a rudecommenton his lastsupportcheck.(Table 1 providesa summaryof the transcribingcon- M. Goodwin(1983), GoodwinandGoodwin(1987), andCoulter(1990). ventionsused in all excerpts.)

DISPUTERESOLUTIONIN MEDIATION

821

ferredresponseto anaccusation.But,for accusations, denialsarethe preferredresponsebecause the absenceor delayof a denialmaybe interpreted as an admissionof/evidence of guilt (AtkinPreferencefor Agreement son andDrew 1979,p. 112-13;Heritage1984,p. Besides the speechexchangesystem,preference 269). The preferencesystem, which minimizes organizationalso affects arguingand disputing and mitigates disagreements,is not being folin ordinaryconversation.One way thattalk can lowed. Accusationscanthereforeleadto arguing be said to have a preferenceorganizationis that or disputing. some types of utterances(e.g., invitations)imply/projecta specific type of response(e.g., ac- Maintainingand EscalatingDisputes ceptanceor refusalof the invitation)in the next turnspace(SchegloffandSacks 1973;Schegloff Once an argumenthas begun, its structureconsequenc- tains the seeds of its continuance.Researchon 1987;Sacks1987).Thesetwo-utterance es arereferredto as adjacencypairs.The initial theorganizationof children'sdisputesshowsthat utteranceis thefirstpairpart, andtheresponseis children often bypass opportunitiesto resolve calledthe secondpair part (SchegloffandSacks conflictandinstead"activelyworkto maintaina (M.Goodwin1983,p. 661; 1973, p. 295). The type of responsepreferred stateof contradiction" dependson the natureof the first pairpart.Pre- see also Corsaro1985, p. 212-19; Goodwinand ferredresponsesoccur more frequently,are ex- Goodwin1987,p. 206).5Children'sdisputes(M. plicitly stated,and are producedwithoutdelay, Goodwin 1982; Corsaroand Rizzo 1990, p. 28) disputes(Vuchinich1984, responsesareproducedwith and intergenerational while"dispreferred" delays, accounts,or othermitigatingtechniques p. 219) arecommonlyunresolved. Once begun, argumentsmay be difficult to (Pomerantz1984, p. 64). Pomerantz(1984) and Heritage(1984; 1988) stop because accusationsengenderreturnaccu(Coulter1990),or dearguedthatrecipientsfollow preferenceorgani- sations,counter-assertions zationbecauseof a normativeconstraintfavor- nials. Denials are preferredsecond pair parts ing social solidarity.A responsethatis not what (hence producedwithout the delay that might was projectedis formulatedas a dispreferredsec- allow accusersto repairor mitigateaccusations). ondpairpartto displayorientationto thefactthat Such denialslack accountsthatmight lead to a the preferredresponse should have been pro- resolutionof the disputants'divergentpositions; duced. The preferencesystem thus reflects the othermitigatingtechniquesthatmightlessen the impactof the denialarealso abmoral orderas people's responsesdemonstrate face-threatening sent. Accusationslead to departurefrom a prefsolidaritywith theirco-interactants. Formost utterancetypes,thereis a preference erence for agreement,and this departureis refor agreement(Pomerantz1975, 1978a, 1984; invokedby the preferredresponseof denial. Denialsarealso a directdisagreementwiththe Sacks 1987). Disagreementsare formulatedor placed to demonstratetheir dispreferredstatus priorspeaker,puttinghim or her in a positionof (Pomerantz1975; 1984). However, Pomerantz eitherbackingdown (whichis face-threatening) (1984, p. 81) noted that supportfor one's co- or defendinghim or herself.If the accusermaininteractantis demonstratedby disagreeingwith tains his or her stand and re-issues the accusanotby agreeingwiththem.Sim- tion,the disputecontinuesfor anotheradjacency self-deprecations, ilarly, the preferencefor agreeing with assess- pair because the second accusationalso invites ments shouldlead to the acceptanceof compli- departurefrom a preferencefor agreementand from ments,but this puts one in the positionof prais- mayprovokea seconddenial.Thedeparture ing oneself (Pomerantz1978a).Recipientscope a preferencefor agreementthataccusationsmake with these conflictingconstraintsby, for exam- relevantcan thus be maintainedover a series of ple, downgradingcomplimentsor shifting the turns(e.g., see Whalen,Zimmerman,andWhalen referentsof compliments (Pomerantz 1978a). 1988, p. 353-54). Complimentacceptancesare therebyproduced as dispreferredratherthanpreferredsecondpair I There may be cross-culturaldifferencesin chilparts. dren's tendenciesto aggravateor mitigate disputes. Accusationsalso operatecontraryto the "pref- Forexample,Boggs (1978) describedan argumentin erenceforagreement."Preferencefor agreement whichtwo Japanesechildrenused mitigatedandindiwould lead to an admissionof guilt as the pre- rect arguing. the potentialfor directlyaddressedadjacentexchangesmakesdisputingpossible.

822

AMERICANSOCIOLOGICALREVIEW

Likewise,once a disputehas begun,thereare (.3) interactionaldisincentivesfor ending it. BreakM: 'She's twelve?' ing the cycle of oppositionalutterancesmay in(.2) volve some degradationof self. Coulter(1990, C: =* She's twelve. (.5) And the:n,we had pp. 189-90) describedhow participantsretreat twi:ns:?,(.3) who have been li:ving at from argumentative sequences by "backing ho:mewith- with KA:Ren,(.2) uhm, since down"from a previouslystatedposition. Such the dissolution. retreatsfrom explicit disagreementsmay result (.3) in loss of face (see Emmison1987, 1988;Great'The name of your twins?' M: batchforthcoming;Katriel1985;Corsaro1985, (.1) p. 219; Vuchinich1990).Thus,once disputesare C: = uh, STAcey andAriel, (1.9) For the most begun,a motivationfor continuingthemmay be part,(.1) u::hin the la:st five years, we've builtinto the talkitself. (.2) had an O:N andoff: (.8) flexibility, In sum, the speech exchange system of ordi(.3) type of arra~ngement.... naryconversationprovidesan interactionalenvironmentin which disputingtechniquescan be Whiledisputantshavethefloorto tell theirstoused to continueor escalatedisagreement.What differencesin the speech exchange system of ries, they are free to self-select as next speaker, mediationhearingsmakedisputeresolutionmore andto speakeven whennot selectedby a mediator. However,they do not use the full range of likely thanescalation? turn-taking optionsthatwouldbe availabletothem in ordinaryconversation.Specifically,disputants THE SPEECHEXCHANGESYSTEM refrainfrom using the "currentspeakerselects OF MEDIATION next" option to select a co-disputantto speak. However,they may use the "currentspeakerseTurnTaking lects next"optionto ask a mediatora question. Disputantswho are not telling a story rarely The turn-takingsystem of mediationdiffers in severalrespectsfrom thatof ordinaryconversa- speak duringa co-disputant'sstory, and when tion. Mediationis partlya pre-allocatedsystem, they do theirutterancesdisplayorientationto the as aretrials(AtkinsonandDrew 1979),andnews turn-takingsystemin mediation.Forexample,a interviews (Clayman 1987; Greatbatch1988; disputantmay speakduringthe otherdisputant's HeritageandGreatbatch1991).A mediatoropens storyif it is in responseto a mediator'squestion. the hearing, explains the mediation process, Excerpt3 is partof the respondent'sstory in a makes introductions,and then elicits the com- disputebetweenneighborsin an apartmentcomplainant'sstory.The orderof story-tellingis thus plex. The respondentis explainingwhy she and set - the complainant'sstory precedesthe re- the othertenantswrote a letterrequestingevicspondent's.Disputantsdo not begin theirstories tion of the complainant("Mrs.Norton"). untilafterthemediatorsolicitsthem,therebydis- Excerpt3: playingtheir orientationto the mediator'sconR: Because we have A:LL had repeated?, trolof the progressof the hearing. (.6) g sets?, (.3) fromMrs. 'Norton.' (.8) Afterthe initialstoriesare completed,mediaA::nd,if therewouldbe someway to work torsmay solicit secondor even thirdstoriesfrom this ou::t!,I would li::keto do it:. (.2) the disputants.The disputantsmay not interrupt bu::t,(.3) 1 don't fee::l, andno one else eachotherduringtheirstories,butthe mediators feels because (1.9) A:LL the reasonsthats:h[e claimed] mayinterruptto askquestionsor refocusthetopic. The story-tellerwill typicallyuse the "current MB:=* [Haveyou] seen this? speakerselects self' turn-takingoption to con(.2) tinue his or her story after answeringa mediaC: = No, I haven't? tor'squestion.Forexample, When thereare two or more complainantsor Excerpt2: in a hearing,brief asides between respondents C: =I've been divo:rcedfive years. (.2) are allowed. In Excerpt4, the com"partners" Ouhmo(.2) 0so::0,.h (.5) 0u:hmo,(.1) We plainantarguesthatthe respondentneglectedto have threeCHILDren?,(1.0) The eldest visit his children.RespondentB urges her husSHARon,has been li:ving with me, (.1) bandto writedown his responseto this accusafor (.4) just abouttwo years.

DISPUTERESOLUTIONIN MEDIATION

823

tion, since they're not allowed to speak during scribedif they use "actionprojections"at turnthe complainant'sstory. beginningsto requestpermissionto speak. Action projectionsare utteranceslike, "CanI ask Excerpt4: you a question?"which, by projectingan action He WASN't (.1) he didn'teven SEE:: C: (e.g., a question),createa space for preliminary them for=the-(.2) you=know?,(.5) beforeperformingthe projectedacinformation BAREly saw themat a:ll, for a couple tion (Schegloff 1980). Disputantsphrasethese uh=years.(.5) RightA:fterthe divo:rce. actionprojectionsas questionsaddressedto the (.5) A::ND mediatorsandthey do not producethe projected (.4) actionuntila mediatorhas grantedpermission. ' Writeit down' RB: C:

(.9) A:::NDDU::H (2.5) you know!, (.2) I=I=don't=knowwhathe WA:NTS:=uh me::.. .

The non-storytellingdisputantoccasionally requests or provides clarifying informationor repairserrorsin the disputant'sstory.Mediators generallylet suchbriefintrusionspass,even when addresseddirectly to the co-disputant.6In Excerpt5 fromthe neighbor'sdispute,MediatorA hasjustreadaloudtherespondent'sevictionletter.

Excerpt6: She was: very very upsetaboutthat!=and RA: (.1) .h (.4) made it perfectlyclearthat she=didn'twant (.1) anything'that had' to do with Ben! (.3) 'after thatt?= C: =* =COULD=I-(.1) could=Ia:sk a question Oatthis point?0 (.1)

MB: C:

Excerpt5: 'Okay, and' thatwas jus:t: R: (.2) MB:

hm::h?,

MA:

And [thi:swa]s A:prilsecond? [on:e]

(.1)

R:

(.3) Yes. Thatwas the [firsttime this was turnedin] C: => [Wasthatorganized]by you? R:

R:

(.4) No. This was organized, by Tess ...

0Sure!0

(.2) Was: (.3) wa:s: (.2) he:r:(.9) u::hinte:nt, in 0you=know?,uhO(.3) Did it SOU:ND to you: thatshe was TE:LLingyou:?, . . .

InExcerpt6, froma familydispute,thecomplainantselectshimselfto speakaftertherespondent's actionprojection,"Could story.Thecomplainant's I ask a questionat this point?"is formulatedand intonatedas a question.The complainantthen pauses.AfterMediatorB'sresponse("Sure!")the complainantasks the projectedquestion. By gettingpermissionfroma mediatorto continuethe disputanttransformsa self-selectedutteranceintoa responseto a mediator'sutterance, which is consistentwith the turn-takingsystem of mediation.7Althoughdisputantsdo notalways I

While disputantsin these hearingsroutinelyuse The disputantsand the mediatorsdisplay an actionprojectionsas requestsforpermissionto speak, orientationto the turn-takingsystem of media- mediatorsuse action projectionsat turn-beginnings tion when they requestpermissionto speak or as preliminariesto preliminaries(Schegloff 1980): employ sanctions.Throughoutthe hearings,the MA: That's fair enough 'cause you might s- be disputantwho is not telling a story may select sittinghere anotherhourA:nyhow! self to speak for purposesother than those de6These findingsdiffersomewhatfromGreatbatch's forthcomingfindingsfor news interviews.The news interview is designed to show disagreementwhile mediationis designedto produceagreement.Thismay accountfor differencesbetweenmediationandnews interviewsin how clarificationrequestsare used and respondedto. In mediation,brief interjectionsthat clarifya point may be helpfulfor the conflict resolution process,andhence areallowed.Mediatorsintervene only when clarificationsor errorcorrectionsdevelop into arguing.

R:

(1.3) 'Sure.'

(.1) MB:=* hh ah Let me s:ay I- It dependsa lot on your layer, I- d- uh- know that-There are good ones and thereare bad ones.=

are Mediators'actionprojectionsatturn-beginnings generallyphrasedand intonatedas statements(e.g., "let me ask you a question")ratherthanas questions (e.g., "canI askyou a question?").Pauseswithinrather thanat the end of unit-types,or the absenceof pauses, indicatethat a response is neitherexpected nor pro-

824 requestpermissionbeforespeaking,suchrequests indicatethatthey (andthe mediators)areoriented to the expectationthatdisputantsnot self-select duringthe otherdisputant'sstory.8 Disputantsrarelyproduceutterancesduringthe otherdisputant'sstoriesthatdo not displayorientationto their limited right to speak. When disputantsuse the "nextspeakerselects self' option to respondto an accusationduringthe other disputant'sstory,mediatorsusuallyintervene.In Excerpt7, a father(RA)criticizesthe stepfather's (C) treatmentof theirchildren. Excerpt7: ... the CHILDrencoming ho::meand him (.4) taking them into the BA::throom,(.4) andlooking in their EYE:S!,becausetheir:pupils mightbe di=h=lated'cause they've had=too=many-(.1) too much sugarfrom milkshakesthatthey drinkin at my HOU:SE! (.2) C: => 'That's [nottrueat all'] [And=MY-M]Y KI:DS:(.2) my kids RA: have cry: (.1) cried over [that.] MA: =E [Excuse]me for interrnt for just= a=minute.=Iforgotto, (.1) mention, onie=of=theGROU:NDru:les!,(.2) and thati:s when- (.2) you're telling your story,(.7) you say nothing. RA:

AMERICANSOCIOLOGICALREVIEW Disputantsaddresstheirutterancesto the mediatorsratherthanto theirco-disputants.In Excerpt 8, a disputeovervehiclerepairs,the complainant providesinformationthatthe otherdisputantsalreadyknowbutthatthemediatormightnotknow, demonstratingthatthis utteranceis designedfor the mediators,not the respondent. Excerpt8: Okay, .h Dan?,If you'd like to go ahead MA: then=and(.2) andtell us your side of the

C:

(.6) Okay. (.3) U:::h,(.5) think=it=was approximately:u:h (.1) 'think it was in eighty six0 (1.9) the date was u::h(.1) FI::veuh seven eighty 'seven I believe an' I-Vtook the:motorho::me,to u:h Mark'sAuto. (1.0) .hh chuh! (.1) for:=a see (.1) replacefan belts, replace hose, (.2) inspectthe air upper=radiator conditioning:,unit.

The complainantspecifies the date the vehicle was broughtto the shop, the shop's name, and the purposeof the visit for the benefitof the mediators.The complainantgazes downathis notes initially,thenlifts his gaze to the mediatorsduring this utterance. Disputantscan also indicatethatthe utterance is addressedto the mediatorsby referringto codisputantsin the thirdperson.In Excerpt9, the MediatorA sanctionsthe complainantfor speak- respondentrefersto the complainantin the third person("DAN","he"). ing duringRespondentA's story. In sum,thedisputants'requestsforpermission Excerpt9: to speakand the mediators'sanctionsshow that Whe::n(.7) DAN was ca:lled,andwas RA: participantstreatthe mediationhearingas havtold thatthe waterpumphousing:,(.2) ing specificturn-takingnorms. was lea:king,. . . Whenhe came dow:n TheParticipationFramework

andhe pickedup the car:,rightbefore=the coach (.4) 'uh' before it was adequately roadtested.

While participantsin ordinaryconversationare Recipientresponsesprovidefurtherevidence notrestrictedas to whentheycanbe therecipient of an utteranceor select specific othersto speak that disputant'sutterancesare addressedto the (Sackset al. 1974), disputants'participationsta- mediators.Minimalresponses(e.g., "umhmh," tuses in mediationhearingsare predetermined. "yeah"),whenprovided,areproducedby mediators,not by co-disputants. vided for. Responses are generallynot forthcoming, Excerpt10: demonstratingthat the parties interpretmediators' At tha:tsta:g:e,(.2) it is truethat(.1) C: actionprojectionsat turn-beginningsas preliminaries I=had(.2) suggested volunteer=help=that to preliminariesratherthanas requestsfor permission could be- (.2) available,(.1) didn't-.(.1) to speak. didn'tworkout. 8 Participants in news interviewsalso use requests (.2) forpermissionto speakwhendepartingfromtheturnMB: =w 'um=hmh.0= takingsystem of thatinstitutionalsetting(Greatbatch =But- the=eh-(.6) she was NOT- (.4) C: 1988; Clayman and Maynard 1990; Heritage and SHE:nor=anybody=in=myfamily Greatbatch1991).

DISPUTE RESOLUTIONIN MEDIATION was=thepitmary, (.3) volunteerhelp on which the- budgetwas (.2) ma:de...

825 rosy ey:ed, (.9) first time landlord.... ((9 lines omitted))... I just winged it?, (.3) you=know?,I just- trustedthatthatwas:= MB:=E =.hh Whatrentwere you: paying?,(.4) priorto Paul's purchase.Whatwere you: (.1) whatwere you payingto the o:ther landlord?

The directionof the complainant'sgaze also indicates he is addressing the mediators - he switcheshis gaze fromone mediatorto the other anddoes not look at the respondent,who is seat(.9) ed next to him. R: Ei::thertwo fifty, or threehundred,. . . In addition,co-disputantsoftenengagein side activities like note-takingduringthe co-disputant'sutterances.Thus,mediatorsanddisputants Excerpt12, fromthe vehicle repairdispute,also displayorientationto the conventionthatdispu- illustratesmediatorcontrolof disputants'involvetants'utterancesareaddressedto the mediators. mentin the discussionandof shiftsin topic:MediatorB changesthe focus from the bill for the air-conditioning repairto whetherthevehiclehad DiscussionFormat been road-testedor not. In additionto the storytellingduringwhicha dis- Excerpt12: putanthas the floor for an extendedturnandthe WhatMA: How aboutthe air-conditioner?, co-disputantremainssilent,thesemediationhear(.1) money, was spentthere?= ings haveextendedperiodsof discussion,primaRA: =Thirtyseven fifty. sequencrily comprisedof question-and-answer (.5) es. The organizationof these discussionperiods MA: And that's simply charging? differsfromordinaryconversationin thatmedia(.4) torscontrolthe topic of discussionandwho parThat'scorrect.(.2) And thenyou haveRA: ticipates.The mediatorsuse directedquestions (.2) uh you shouldhave freon, to switchthe talkfromone disputantto the other. somewhere,(.1) the:re. Excerpt 11 is from the end of an extendedex(.5) change of questions and answers between the mediatorsand the respondentin a landlord-ten- MA: 'Okay,plus (.5) antdispute.In the arrowedutterance,MediatorB shouldbe somefreonchargedonit RA: The::re, asksthecomplainantwhetherhe, as landlord,was (.3) awareof the respondent'spoorearningcapacity. 'Thereis.' MB: After the complainant'sresponse, MediatorB (5.0) directsa questionto the respondent. timeyoupickedup MB: = .h Didyou=know=the Excerpt11: thevehicleit hadnotbeenroad-tested? (.6)Wereyoutoldthat? MB: Couldyou tell us, (.3) whatyour income wa:s, priorto. (.4) .h in otherwords, (.1) (.3) C: No, theguy-(.1) u:hTim(.1)u:hcalled Priorto thattime? R: me aboutI thinkit wasa DA::Ybefore twodaysbeforehe saiddu:h(.7) .h (.3) SETto go except,bu::t EV'RYTHING'S MB: Au:gust. we've-. . . R:

(.3) We::ll,I=don't-(.4) 1 can I'm=not=sure=ifI can reme:mber!,hh (.1) u::hm:(.5) it was very lo:w. (.5)

MB:

Very low. (.1)

R:

Very [low u:::hm] MB:=X [Wereyou awa:reof that]Paul

Duringdiscussionperiods,disputantscontinue to addresstheirutterances(includingaccusatory utterances)to the mediators.9Responsesto accusationsarealso addressedto mediators,thus resultingin mediatedtalk. For example, in Excerpt 13, from the landlord-tenantdispute, the complainantassertsthat the respondentis consistentlybehindin herrent.MediatorA shiftsthe

(.9)

C:

Ye:s::.WE:LL?,(.2) .h I didn't-you= know?,(.1) make any u::hinvestigation? (.2) 1 was a ve::ry,u:hm(1.6) um: (1.2)

addressed exchangesbetweendisputants 9-Directly - exchanges of threeormoreturnsare areinfrequent in suchexchanges. rare.Mediators oftenintervene

826 focusto therespondentwithoutchangingthetopic. Therespondentthenrespondsto the substance of thecomplainant'scomplaint(by admittingthat she cannot,in fact,pay the rent)in the contextof a responseto the mediator'squestion: Excerpt13: She-in facthasn'tma:dethepaymentsas C: here(.2) in:(.2) [Ma:y] I: (.5)requested MA: [And=she] certainlyhasn'tmadetheJuly payment. (.3) Correct. C: (.5)

AMERICANSOCIOLOGICALREVIEW The complainantfirstrefersto the respondentas "she,"then switches to directaddress("your"). The mediator'sminimalresponse("umhm")reestablishestheframeworkin whichthecomplainant's complaintsshouldbe addressedto the mediator- the remainderof the complaintis addressedto the mediator,not the respondent. In Excerpt15, fromthe same hearing,the disputantshave been arguingover who has more time with the children.

Excerpt15: You'veha:dthem,a majority! of thelast C: .hhItwasMA:ry'sbirthday Saturdays!, (.1) party,Fridaynightattheboardwalk. .hhAbout=a-(.2)you=know?, (.2) six MA: Okay. (.3) Now you a:nswered,(.3) unableto makethese that=you're eightweeksago, h I hadthosesame paymentsatthistime..hh(.4)To to youin thatthing. pla:ns!,I relinquished tha:t?,(.3)as a question,(.2) rephra:se .hhNextweekend,it wasthepark.(.1) .h Does=thatmea:nthatyoucan=notpa:y, Thenextweekendit were-a- a few andsixtyfive dollars?, elevenhundred weekendsago:,it was=likeyouhad=duh (.2)permonthrent. dentalappointment on a Saturday atone weekendwasthe o'clock!,.hThe=LA:st (.2) (.2)It'slik[e-it-] parade!(.5)you=know? R: 'Ye:sit does.'(.9)At thistime. M: = [ItS]OU:NDSlikeyourflexibility,(.1) .hhis working?, Departures (.9) uponmyti:me! Right!Butit's infringing C: Duringstorytellingand discussionperiods,disif you can makethe Andmaybe (.3) putantsusually addresstheir accusationsto the PLA:NS:you=know=for(.2)for When a disputantdepartsfrom the mediators.10 eveninginstead=of SATurday!, frameworkof mediationby addressparticipation .hh FRIDA:Y,(.4)Fridayevening=and ing theotherdisputantdirectly,themediatormay (.1) attemptto restoremediatoraddress."IOne techM: => 'hm=[hmh'] nique is for mediatorsto produce minimal reC: [an]d,andSA[turday?] sponses,eventhoughtheutterancewas addressed => M: [A compro]mise? to the co-disputant.For example,in Excerpt14, (1.8) from a divorcedcouple's dispute,the complainM: => ?um=hmh.? (.4) Stanhowlonghaveyou antcriticizestherespondent(his ex-wife) for gobeenmarried? ing on vacationsanddating: Excerpt14: C: =She'sgoneto Mexico,andto Arizona, too?(.2)so andL.A.withyourboyfriends don'tgiveme tha:t? (.3)

The complainantdepartsfrom the participation frameworkof mediationby addressinghis complaintsto the respondentinsteadof themediator. Themediatorattemptsto restoremediatoraddress by askingthecomplainanta question,usingminM: => 'um hm0= imal responses,andinitiatinga topic change. C: =Okay?(.1)Titis forTAThere..h u::hm, Adjacentaccusationsanddenialsarealso sub.hh(.2)1 donotagreewith,(.5)three to sanctionby the mediators.In Excerpt16, ject DA:ys?(.3) a week.... fromtheneighbors'dispute,therespondentsparks '?Only 8 percent(25 out of 321) of the accusations conflictby addressinga challengedirectlyto the in the nine hearingshadsecond-personattributionsof complainant.

blame. " See ClaymanandWhalen(1988/1989)andSchegloff (1988/1989) for analysesof departuresfromthe speech exchange system of anotherform of institutionaltalk - the news interview.

Excerpt16: 1 C: =TheSA:MEwomanthatorganized thatparked: 2 ?eh?=sothatI couldnot getout=ofmyco:-h-mplex.= 3

DISPUTERESOLUTIONIN MEDIATION 4 5 6 7 8

MB

9

R:

10 11

R: C:

MB:

-12

13 14 15 16 17 18 19 20 21 22 23 24 25 26 27

R: MB:

C: MB:

R: MB:

C: MB:

C: MA:

827

=um::?, (.2) You were parked= =one morning. (.2)

The two disputantsare now engaged in a fullfledged argument.They have abandonedthe speech exchange system of mediation for the speech exchange system of ordinaryconversation which enablesthem to engage in disputing - here,the exchangeof accusationsanddenials. She was parked, in In line 26 MediatorA breaksin andsanctionsthe h[er space and she] [yeah,chum!yeahbut=I=WAS]=JUST two for"crosstalk",thusattemptingto restorethe mediationexchangestructure. [SAYING I- I REME:MBER y]ou

[(parkednext) to you:] you SA:Ying,you=know?, [Therewas nowhereelse to park.] [thatyou felt you] were o:::ver, whe::lmed,with [peop]le. [You still] (.2) can not parkin [some one else's parkingplace.] [um so I thought] [It] ha:ppensa::1lthuhti::me [jea:n,]Byftdoesn't she [Yes] do it with otherpeople, an' that's [(not A [I DON'T WANT ANY CROSS TALK, thankyou.]

In lines I to 3 the complainantattemptsto lessen thecredibilityof a tenanton therespondent'sside of thedisputeby recountinganoccasionwhenthe tenantblockedherintoa parkingspace.Inlines6, 9, and 10 the respondentthen departsfrom the turn-taking systemof mediationandrespondswith a counter-assertion (Coulter1990). She gazes at the complainantduringthe initialportionof her utterance("youwereparked")butthenturnsher gaze backto the mediatorsas she repairsher utteranceto addressthemediatorsandto referto the complainantin thethirdperson("shewasparked, in herspace").MediatorB's attemptto cutoff the incipientdisputeby shiftingtheconversationback to a priortopicin lines 11, 12, and 14 ("butI was just saying I remember you saying, . .") appears to be unsuccessful (see Claymanard Whalen (1988/1989) for an analysisof unsuccessfulattemptsto restorethe speechexchangesystemof newsinterviews).Thecomplainantdoesnotcomply withMediatorB's bid for a changeof topic, andcontinuesthe exchangewiththe respondent. She defendsheraction("Therewas nowhereelse to park")in line 15. IgnoringMediatorB's continuingutterance,the respondentreplies to the complainantin lines 18 and 19 ("Youstillcannot parkinsomeoneelse's parkingplace").Thecomplainantthenrespondsdirectlyto thisutterancein lines 21, 22, 24, and 25, addressingthe respondent by name and turningher gaze towardher.

Summary Theinteractionalorganizationof mediationhearings differsfromthatof ordinaryconversationin several ways. In some respects,mediationis a pre-determinedspeech exchangesystem. Some turns(e.g., storyrequests,stories,openings and closingsof hearings)arepre-allocatedto specific individuals on the basis of institutionalroles. Access to turn-takingoptionsis also tiedto institutionalroles- disputantstypicallydo not selfselect duringthe otherdisputant'sstory,butmediators are free to self-select as next speaker throughoutthe hearing. Disputantself-selectionis not treatedas a departurefromtheturn-takingsystemif it is framed as a requestfor permissionto speak,an aside, a repair,or a requestfor clarification.Utterances respondingdirectlyto the substanceof the other disputant'scomplaintaretreatedas departures. In the participationframeworkof mediation, disputantsdirecttheirutterancesto themediators ratherthan to each other. Although all participantsare recipientsof the stories,the otherdisputantis nottheaddressedrecipientanddoes not have the rightto reply immediatelyto the story (C. Goodwin 1987). The participationstatusof participantsin mediationhearingsis therefore subjectto constraintsthatdo notexist in ordinary conversation. When disputantsdepartfrom the speech exchange system of mediationand use the turntaking rules and participationstatusesof ordinaryconversation,theirtalktakeson the character of an argument.Mediatorsattemptto restore the speechexchangesystemof mediationby using sanctions,changingtopics,redirectinga question, utteringminimalresponses,andothersimilartechniques. IMPLICATIONS In mediation,the adjacentanddirectlyaddressed oppositionalutterancesthatconstituteargument generallydo not occur.In effect, the speech ex-

828 changesystemof mediationlimits"disputing"to thoseoccasionswhendisputantsdepartfromthe turn-taking systemandparticipation statusescharacteristicof mediation.Fourcharacteristics of the mediationprocessenableparticipantsto manage accusationsanddenialswhile savingface (Goffman 1967) which enables them to avoid arguments.(1) Accusationsanddenialsarenot adjacency pairsin mediation- becausea denial is not immediatelyrelevant,an accusationdoes not engenderan oppositionalresponseas a next action. (2) In the participationframeworkof mediation, accusationsand denials are addressedto mediatorsinsteadof to co-disputants- denials occur as responsesto mediatorqueries,hence, they are not "disagreements."(3) The delay of denials provides disputantswith the option of selectivelyrespondingto denials,potentiallyreducing the numberof issues underdispute.(4) orderof Theinteractional organization/normative mediationprovidesfor mitigatedratherthanaggravatedaccusationsand denials. This interactional organizationhas several implicationsfor the disputeresolutionprocess. ThePlacementof Accusationsand Denials Inordinaryconversation,accusationsarefirstpair partsof adjacencypairs, and denials or admissions of guilt arepossibleresponses.Denialsare typically preferredresponses, and tend to be placedimmediatelyafterthe accusation,without delay, accounts,or othermitigatingtechniques. Accusationsthereforemake disagreementrelevant and thus provide an interactionalenvironmentfor escalationintodisputesandarguments. Accusationsand denialscan providean interactionalbasis for disputingtechniqueslike the exchangeof accusationsanddenials(M. Goodwin 1983) or other techniquesthat dependon adjacency (e.g., escalation,repetition,formattying, or contrastively-matched counters),and providing a motivationfor theiruse. Thespeechexchangesystemof mediationprecludes the use of many of these disputingtechniques by eliminatingadjacentexchanges between disputants.This is done by breakingup some coursesof actionthatcould otherwiseoccuras adjacencypairs.Disputantscannotrespond adjacentlyto accusationswithoutdepartingfrom the turn-takingsystem of mediation- respondents may not speakuntilafterthe complainant hasfinishedhis or her"story"anda mediatorhas solicitedthe respondent'sstory.In addition,mediatorsmay ask the complainantquestionsdur-

AMERICANSOCIOLOGICALREVIEW ing and after the story. Hence, a denial or an admissionof guilt is not the relevantnext action afteran accusation.12 Thecomplainantin theneighbors'disputeproduces manyaccusationsin the courseof her story, butthe respondent'sdenialsdo not occuruntil muchlaterin thehearing.13Forexample,early in the hearingthe complainantaccuses the respondent(Jane)of vandalizinghercar. Excerpt17: hh(.3) earlyin C: Mycarwasvandalizedd, themo::rning, (.1) I cameba:ckI was away(.2)forthenight(.) I cameback, andtherewaspaint:,(.6)on theinterior?, (.8) (.7) uhlatexpaint, (1.5)garbag:e. powdered milk.(.4) I'vegotpicturesof that=if-(.3)you'dliketo see it. (.4) MA:

'um=hmh.' (.2)

C:

0Whatthatlookedlike.0(2.1)A:::nd:the onlyconceivable person:(.3)that=it= could=do::(.4) thatwouldbe responsible notherselff,(.1) whethershedid=it=or wouldbeJane!,. . .

Therespondentshakesherhead"No"as hername is mentioned,butdoes not gaze at the complainant. Since the complainantis looking at the mediators,it is not clear whethershe sees this ges12 What is an adjacencypair first pair part may dependnot only on the type of utterance(e.g., greeting, invitation,accusation),and its construction,e.g., as a question(Schegloff andSacks 1973, p. 295), but also on the speech exchange system in which it is produced.Accusationsin ordinaryconversationhave differentimplicationsfor the next turnspace thando accusationsin mediationhearings.For an analogous argumentwith regardto news interviews,see Clayman and Maynard(1990). 13 The impact of nonverbalresponses to accusations is minimizedin these hearingsbecause the disputantsare seated on the same side of a long table, with the mediatorsopposite them. Disputantsmust thereforeturn their heads 90 degrees to make eye contactwiththeirco-disputant.Facialexpressionsare thereforeprimarilyvisible to the mediators,and may be producedfor theirbenefit.Becauseonly one video camera was used, there is no record of mediators' nonverbalresponsesto disputants'nonverbalactions. InExcerpt17, therespondenton two occasionsshakes herhead"no"afterthe complainantvoices an accusation. This gestureis probablynot visible to the complainantandis not respondedto verballyby the mediators.Respondentsmay also indicatetheirobjections to an accusationby note-takingactivity,thusshowing an intentto respondwhen they get the floor.

829

DISPUTERESOLUTIONIN MEDIATION

ture,but neithershe northe mediatorsacknowl- Thecomplainantaddressesthe mediatorsandreedge it. The respondentdoes not replyto this ac- fers to the respondentwith the pronouns"she" cusationuntilover half an hourlater. and"her."The complainantis thusable to avoid direct accusations.MediatorA respondsto the Excerpt18: complainant'ssolicitationof a minimalresponse thatshe interprets ("youknow?"),demonstrating At the time, (.1) we firststartedthis, (.2) R: there=was=uh(.2) numerous people at this utteranceas addressedto her. The responthatco:mplex,with ma:nyreasons(.1) to dent's failureto respondindicatesthatshe does (.2) egvery- not interpretthe utteranceas addressedto her. do vandalism,=CHILDREN?, body. (.1) .hhI hadno:thingto do with it? Thus,accusationsin mediationmay be less facethreateningthan those in ordinaryconversation Becauseaccusationsin mediationhearingsare (Pomerantz1978b).This is anotherreasonescanot firstpairparts,they do not make an opposi- lationinto argumentis less likely. Inmediation,adjacentexchangesbetweendistional responserelevantin the next turnspace. Therefore,accusationsin mediationtypicallydo putantscan be terminatedif they occur.If a disnotmakearguingrelevant,therebyremovingone putantdepartsfrom the turn-takingsystem by placinga denialadjacentto an accusationor adsourceof disputeescalation. dressing an accusationdirectly to a disputant, mediatorsinterveneto preventa full-fledgedarDenials, Not Disagreements gument.5 The speech exchange system of mediation Not only are denials in mediationdelayed, but becausethey arenot secondpairpartsto accusa- thereforeprecludesthe use of the disputingtechtions they do not constitutea disagreementwith niques found in ordinaryconversation.Mediatheimmediatelypriorutterance.Therespondent's tion allows for theproductionof accusationsand responseto theaccusationoccursas a responseto denialswhile preventingtheiruse as oppositiona mediatorqueryratherthanas a denialof an ac- al moves in argumentativesequences.The subcusation.Thus,denialsin a mediationcontextare stance of the disputecan be discussed without argument. less likely to provokedisputationalresponses. Because accusationsand complaintsare addressedto the mediatorsratherthanthe co-dis- SelectiveResponsesto Accusations putant,blamecannotbe addresseddirectlyto the co-disputantwithoutdepartingfrom the speech In ordinaryconversation,accusationsand deniexchangesystem of mediation.Therefore,attri- als are adjacency pairs and an omitted denial butionsof blameare typicallyformulatedin the 'S In this excerpt from the divorced couple's disthirdperson.'4For example,in Excerpt19 from pute, the mediator terminates a dispute. theneighbors'dispute,thecomplainantdescribes YEA:H, I- (.) you=know?=I-I still don't feel C: an incident in which the respondentverbally zaod=about=it,because like this is my fle:sh abusedher. and blood!, you=know?,=and(.2) Excerpt19: The=first:(.3) knowledge I had of her C: dislike,u::hm(.2) .h uhmaggravationwith me, one time she:: (.9) I was co:ming:(.2) fromthecar,withmy childwho=wasabout two at the timee, (.6) And her daughter came up to me and said u:hm(.8) 'get her outof thewa:yor somethingandshe said.' ( 1.1) .h you=know?,She=ust (.1) hu::rled a lot=of accusations. I don't=knowhow muchde:tailit's worth,going into. But= it=was=a=lot=of(. 1) you=know?,(.5) ra:ther,(.1) vi::le obscenities,that:,(1.0) I onlyuse if I'm (.2) furi-(.1) you=[know?,] MA: [0um=hmh]yeah.0 '4 Greatbatch(1988) and Clayman (1987) found similarpatternsof addressin news interviews.

M:

you=know?,(.3) uh- I['m yeah-] [But]you shoutedabortion,for nine months:! [with Sharon] [Listen,we] are no:t= =Hey,= talkingg (a[bout)] [WHO]had the abortion?(.1) Y[ou wannaget] SMA:RT? [Waita minu:te!]

C: M:

(.1) DIDN'T SH[E JUST] HA:VE?ONE? [Hey wait!]

R: M: C: M: C:

M:

(.2) HO:LDIT!, (.3) WE'RE NOT IN HERETO TALK ABOUT THAT. I DIDN'T TALK ABOUT MY: PROBLEMS,OR WHATEVER=WE'VEA:LL GOT A STORY. (.2) 0.hThat's no: one's business:!'

AMERICANSOCIOLOGICALREVIEW

830 would be noticeablyabsentand subjectto pursuit. The absenceof the preferredresponse(denial) implies the dispreferredresponse (admission of guilt).In the mediationcontext,however, disputantscanfail to respondto accusationswithout implyingguilt. After one disputanthas completedhis or her story, the mediatorgives the floor to the other disputant.Most disputantstailortheir storiesto respond to accusations made by the other disputant.However, the institutionalizeddelay of responsesto accusationsin mediationprovides an opportunityfor strategicmoves. A disputant may choose to bypass some accusations,focus on the moreimportantaccusations,or ignoreaccusationsshe or he cannotcrediblydeny. Thus, the speechexchangesystem in mediationfacilitates the resolutionof conflict by allowing for selectiveresponseto accusations. Forexample,in the neighbors'disputeseveral of the complainant'saccusationsare never respondedto by the respondent.The complainant accusedtherespondentof verballyassaultingher in the gardenof their apartmentcomplex. She also complainedof bangingnoises comingfrom the respondent'sapartment,and of grass damaged by the respondent'stenant's motorcycle. These accusationswere not referredto again in the hearingby any of the participants. If a respondentfails to respondto an accusation,the complainantmay reissuethe accusation aftertherespondenthascompletedhisorherstory. Forexample,in a disputeovervehiclerepairs,an accusationoriginallyproducedin the complainant's story (Excerpt20) is reissuedby the complainantduring the discussion period after the respondent'sstoryis completed(Excerpt21).

Excerpt21: We're A:lso forgettingthe WAterpump C: was not leakingwhen I broughtit in there! (.5) Somebodycould=uhWHACKedit with=aHAMmerfor all I:=know!

In these hearings,bypassedaccusationsaretypically allowed to lapse. Thus, strategicselection of whichaccusationsto respondto appearsto be an effective techniqueto redirectthe debateor minimizethe numberof accusations"on the table." Delay of denialsreducesopportunitiesfor argumentby minimizingsituationsin whichdisputingtechniquescan be used, andby providing for strategicmoves on the respondent'spart e.g., selectiveresponsesto accusations. TheFormulationof Accusationsand Denials

Mediationprovides a "protected"interactional environmentfor producingaccusationsand denials. However,disputantsin these hearingsdo not takeadvantageof thisprotectedenvironment to produceaggravatedaccusations,perhapsbecause the goal of the mediationprocessis to resolve conflict in a nonadversarialway. Mediatorsemphasizethegoalsof compromiseandnoncompetitivenessat the beginningof each hearing.'6Disputantsdisplayan orientationto these normsby makingaccusationsandcomplaintsless strongthanis possiblegiven the turn-takingsystemof mediation.Techniquesusedby disputants to mitigate accusations include elision of the agent, displacementof the agent relativeto the of theagent, complaint,collectiverepresentations andothertechniquesto downgradeaccusations.'7 Elision of agent. Disputantsin these hearings often constructaccusationswith the agent elidExcerpt20: ed. This makesthe attributionof blame implicit And THEN HE said- (.1) u:h (.1) he C: rather thanexplicit.This does not suggestambicalled me up aboutuh DA:Y after-it was guity in who is to blame - only thatdisputants (three=clock)supposeto pick it up. (.4) He said you got a leak in your waterpump are formulatingaccusationsless strongly.They (.8) housing now. (1.2) I=saidwell,=I= didn't=have a leak when I broughtit in here.(.6) so, I didn'tkno:w!,(.1) what-what =the- (.1) the=problemwas here. (1.1) u::h(.6) And it TURNEDOU:T that=it NEE:deda: a- the- WAter-(.3) the water pumphousingwas cracked.(.6) Now I don't know whetherthatwas: (.3) it was cracked(.2) whether-when=themeCHAnicwas working=on=it?, h or it was crackedbeforeI broughtit in. But it wasn't leakingwhen I brought'it in.' (.1) ThatI kno:w. (.4) 1 would have seen puddlesin my driveway.

16 In addition,mediationis voluntaryand presumably disputantsaretherebecausethey wantto resolve the conflict. 17 O'Barrand Conley (1985, pp. 685, 689) found that disputantsin small claims court also tended to avoidexplicitattributionsof blamein theirnarratives. Althoughthe interactionalcontexts of mediation and news interviewsare somewhatsimilar,the tendency of participantsto escalate or lessen conflict is verydifferent(see Greatbatchforthcoming).Thismay be becausein a news interviewthe goal is to "win"an argument,while in mediationthe goal is to "cometo an agreementor compromise."

DISPUTERESOLUTIONIN MEDIATION

831

are thus displayingan orientationto the nonadversarial,cooperativenormsof mediation. Elisionof theagentis mostoftenaccomplished by using the passive voice. Occasionally,it is achievedby othermethods,such as usingan active constructionbut not statingthe subject.

ordinaryconversationto referto "blameworthy" persons.Thepassivevoice enablesthecomplainant to make accusationswithoutspecifying the wrongdoer.Complainantscan thus accomplish the delicate task of making accusationswhile maintainingpolitenessandavoidingface-threatening utterances. Excerpt22: Displacementof agent.DisputantscanalsodisC: U::::H(.3) .h LOT=OF=the U::h(.5) play orientationto the normativeorderof mediathe=la-theLAborthatwasu::hm(.2) .hhh onthemotorhomewas:wasnot tion by displacingthe agentfromthe accusation conducted throughdelay, hesitation,or sentencestructure. donein=uh-in=uhprofessional(1.6) In Excerpt24, the complainantfromthe land'in=uh'(.8)wa:y!,(.2) 0I=mean0= basically.(.3)'it wasn't,tuh'completed,pro- lord-tenantdisputeis explainingwhy therespondent'scontinuingtenancyin his house would be it=wasn't.hhit=isMY=uh-in fessionally MY:estimation it=wasn't=completed atall! problematic: Excerpt24:

Inthisexcerpt,thecomplainantdoes not say who C: It's been difficultfor us this yearbecause performedthelaboron themotorhome.He could of the: (.6) .h coincidence!of my have specifiedthe agentwhile using the passive deliveringher mail every day and seeing voice (e.g., "thelaborthatwas done by the meu:h (.8) 0u:ho(.1) businesspracticesthat chanic")but he formulateshis complaintso that I=don't-(.2) thinkhelp pa=h=ythe the agent of the "wrongdoing"is impliedrather re=heh=heh=nt!. . . thanstated. In Excerpt23, the complainantin the stepparThe agent is not completelyelided (the coments' disputesummarizeshis complaintagainst plainantrefersto "hermail"),butshe is not specthe respondent: ified in the actualcomplaint:"businesspractices I don't thinkhelp pay the rent."Hesitation, that Excerpt23: pauses, and "turnholders"furtherseparatethe C: I'm=notsurehowto put=itI FEE:LthatI agentfromtheaccusation.(Thepresenceof laughandemotionally BEEN::physically: assaulted!, (1.0)A::ND:(.1)I'dliketochange ter may also serveto mitigatethe complaint.) In a divorcedcouple's dispute,the respondent tha:tto a moresensitive(.9) 'u::hm'(1.5) accusesherex-husbandof lyingabouthis income. moresensitivecommunica tion. The complainantframes his accusationin the passivevoice andelides the actor.He couldhave framedthesecomplaintsin theactivevoice while stilladdressinghis utteranceto themediators(e.g., "Stan assaulted me" or "Stan is insensitive"). Overall,38 percent(121 out of 321) of the accusationsor complaintsin thesehearingswereconstructedwith agentelided.'8 The use of implicitreferencesin accusations may minimizeconflict-provokingthreatsto face, and hence departuresfrom the turn-takingsystem of mediation.By not referringexplicitly to the blamedparty,the "blow"is softenedand a defensefromthe blamedpartyis not immediately relevant. ThesefindingsareconsistentwithPomerantz's (1978b)findingsthatthe passivevoice andother techniquesfor eliding the subjectwere used in

Excerpt25: R: He sent some pay checks:to: (. 1) u:hm, (.3) my attorney? (.1) M: Oum=hm?0 (.2) R: uhm, (.2) Which I don't thinkwere quite corRE=heh=CT!but,

The agent("he")is placedin the firstpartof this utterance,which ends with questionintonation. The respondentthen pauses, and the mediator providesa minimalresponse.Afteranotherpause, the respondentproducesa turnholder ("uhm") followedby anotherpause.Shethenproducesthe accusation: "Which I don't think were quite correct."In additionto displacement,therespondent uses the qualifier"quite"to minimize the accusationand the uncertaintymarker"think." IxElisionof the agentis only one techniquefor The reversalof the negative("Idon't thinkwere accusations. mitigating However,60 percent(191out correct"insteadof "wereincorrect")also serves of 321)of accusations weremitigatedin someway. to mitigatethe accusation(Huebler1983).In ad-

832

AMERICANSOCIOLOGICALREVIEW

dition, we again see the mitigating laughter particles("corRE=heh=CT!"). Collectiverepresentationof agent.Disputants may also displayorientationto the nonadversarial normsof mediationby referringto the agent of wrongdoingas a collectiveagent.One way to do thisis by incorporating self as a blamedparty. In Excerpt26 froma disputebetweentwo brothers, the respondentincludesan admissionof his own guiltin a complaintthathis brotherhasbeen harassinghim:

thatsome of the workwas done correctly("part of it was doneokay").The complainantthenreiterateshis complaint,butin a way thatminimizes the problem:"two or three things that weren't completedproperly."This is a hedgedor understated form of "were completed improperly" (Huebler1983). In Excerpt29, fromthe divorcedcouple'sdispute,the complainantcomplainsthatthe respondent (his ex-wife) ignorestheireldest daughter, who lives with him.

Excerpt26: Since the beginningof this: (.2) yea:r, RA: we've been (.2) harassingeach other...

Excerpt29: C: I'm not saying this to=HURT=you=uhuh- I thinkyou=shou-(.3) m:maybemake an Effort to- even ca:ll her! once a week!, In Excerpt27 from the stepparents'dispute, (.2) to=findout how she's doing=or:.hh RespondentB is tryingto persuadethecomplain(.1) or: (.2) you=know?,make a special antto stopinterferingwiththe raisingof herhusnight=aweek?, (.2) for her!,

band'schildren(thecomplainant'sstepchildren). Excerpt27: I just thinkthat .hh we Justhave to know RB: our pla:ce?,(.3) and that(.1) we have to SHU:T=upa little bi:t,and let the PA:Rents:,(.1) raise the children..h And we're THEREas the (.1) .h the- the (.1) you=know?,to=HE::LP!,to nurture,to LOVE them, .hh and urnto let them, deal with it.

Everyone atthehearingknowsit is thecomplainantwho has been interferingin the raisingof the stepchildren,butRespondentB's useof "we"(the stepparents) mitigateshercomplaintsanddisplays orientationto the normsof mediation. Downgradingaccusations.Accusationscanbe mitigated by downgrading them. Excerpt 28 shows the complainant'sinitial formulationof his complaintin the disputeover vehicle repairs. Excerpt28: MA:

C: C:

'Okay.'(.2)Improper

mechanical] work:? [(Yeah)] (.1) u::h(3.2) work Yeah=im::proper completedby the mechanic I=would=imagine,=PArtof it was done, (.2) u::h(.4) okay, (1.2) but there=were(.2) two=orthreethingsthatwere (.2) that=weren't(.1) completed(.1) properly.

This potentiallythreateningaccusationbegins with a "pre-delicate"(Schegloff 1980), and incorporateshesitations,uncertaintymarkers("I think") and tentative words ("shou-" and "maybe").The complainantleaves the actualaccusationunsaid("youneglect yourdaughter"or "youneverpay any attentionto yourdaughter"), andinsteadimpliesitby specifyinghow shecould correctthis problem("callher once a week"). SUMMARY AND CONCLUSIONS Disputing techniquesin ordinaryconversation dependon a locally managedturn-takingsystem and participationframeworkthat allow participantsto adjacentlyanddirectlyaddresseachother. Thus, denials can be placed immediatelyafter accusationsandformulatedwith explicitattributionsof blame,therebyprovidingtheinteractional contextfor escalationinto argument.'9 In contrast,the interactionalorganizationof mediationprecludesthe use of disputingtechniquesfoundin ordinaryconversation.The turntakingsystemof mediationinstitutionalizesnonadjacentaccusationsand denials. The delayed placementof denialsreducesargumentby mini-

19Some types of institutionaltalkprovidefor argument.O'Donnell(1990, p. 214) studieda union-managementmeeting in which the goal of avoidingconflictualandconfrontationaltalk was not achieved.In The complainantfirst specifies the party who contrast,turn-takingrules and patternsof addressin performedthe improperwork ("themechanic") trials preclude arguingbetween disputants(see Atbut then immediatelyqualifiesthis with an un- kinsonandDrew 1979). Small claims courtsalso apcertaintymarker:"I would imagine."Attached pearto precludedirectexchangesbetweendisputants to theuncertaintymarkeris anacknowledgement (O'BarrandConley 1985;Conley and O'Barr1990).

833

DISPUTERESOLUTIONIN MEDIATION mizing situationsin which disputingtechniques canbe used.In addition,theopportunityto selectivelyrespondto accusationscanreducethenumber of accusationsunderconsiderationand thus facilitatesolution. The speechexchangesystemof mediationalso providesfor accusationsto be addressedto mediatorsratherthan to the other disputant.This patternof addressresultsin third-personreferences to blamedparties.Direct challenges and attributionsof blame thus do not confrontthe "blamed"party directly, lessening their facethreateningimpacts. Finally, the normativeorderof mediationas cooperativeand nonadversarialencouragesdisputantsto formulatetheiraccusations(anddenials) even less stronglythanpossible within the speechexchangesystemof mediation.This mitigationprovidesa furtherdeterrentto escalation into a dispute. These fouraspectsof the organizationof meand diationdeterargument.Thesecharacteristics, how disputantsuse themto avoidthreatsto face, may be one reasonmediationoften successfully resolves conflict withoutconfrontationor argument. To discoverhow mediationworks as a techproniqueforconflictresolution,theinteractional cess of mediationhearingsmust be examined. processminimizes"disputing" Thisinteractional betweenparticipantswho aremeetingto resolve a dispute. Complainantsand respondentscontributeto thisoutcomeby how theypositionand formulatereferencesto othersin theircomplaints, accusations,anddenials.Thus,partof mediation is mitigatingaccusationsby indirectformulations and delayed placementof denials. The advantage of mediationover other types of dispute resolution(e.g., trials,counseling,arguing)may of mediatorsor the diffilie not in characteristics culty of the dispute,but in the interactionalorganizationof mediationitself. The objectiveof thisanalysisis to show how a specific type of conflict resolutionis organized and how thatorganizationenables it to accomplish its goals. While the interactionalorganization of mediationderives from organizational principlesof ordinaryconversation,itdiffersfrom ordinaryconversationin ways that specifically preventargumentfromoccurring.

and disputeresolutioninface-to-face interaction.In addition to studies of the interactionalorganization of mediationhearings,she is workingon an analysis of gender differencesin justificationsand accounts, and on theorganizationof argumentsin ordinaryconversation. REFERENCES

Atkinson,J. MaxwellandPaulDrew. 1979. Orderin Court:The Organizationof VerbalInteractionin Judicial Settings.London:Macmillan. Atkinson,J. Maxwell and John Heritage,eds. 1984. Structuresof Social Action: Studies in Conversation Analysis. Cambridge, England:Cambridge UniversityPress. Boggs, StephenT. 1978. "TheDevelopmentof Verbal Disputing in Part-HawaiianChildren."Language in Society7:325-44. Bottomley,Ann. 1985. "WhatIs Happeningto Family Law? A FeministCritiqueof Conciliation."Pp. 162-87 in Womenin Law, editedby J. Brophyand C. Smart.London:RoutledgeandKeganPaul. Brenneis,DonaldandLauraLein. 1977. "'You Fruithead': A Sociolinguistic Approachto Children's DisputeSettlement."Pp.49-65 in ChildDiscourse, edited by S. Ervin-Trippand C. Mitchell-Kernan. New York:AcademicPress. Clayman,Steven. 1987. "GeneratingNews: The Interactional Organizationof News Interviews." Ph.D. dissertation,Departmentof Sociology, University of California,SantaBarbara. Clayman,StevenandDouglasMaynard.1990."TurnTakingand Addressin the Achievementof an InstitutionalSetting: The Case of the News Interview." Paper presentedat Midwest Sociological Association,April,Chicago,Illinois. Clayman,StevenandJackWhalen.1988/1989."When the Medium Becomes the Message: The Case of the Rather-BushEncounter." Research on Language and Social Interaction22:241-72. Conley,JohnM. andWilliamM. O'Barr.1990."Rules versus Relationshipsin Small Claims Disputes." Pp. 178-96 in ConflictTalk:SociolinguisticInvestigationsof Argumentsin Conversations,editedby A. Grimshaw.Cambridge,England: Cambridge UniversityPress. Corsaro,WilliamA. 1985. Friendshipand Peer Culture in the Early Years.Norwood,NJ: Ablex. Corsaro,William A. and Thomas A. Rizzo. 1990. "Disputesin the PeerCultureof AmericanandItalian Nursery-SchoolChildren."Pp. 21-66 in Conflict Talk: SociolinguisticInvestigationsof Arguments in Conversation,edited by A. Grimshaw. Cambridge,England:CambridgeUniversityPress. Coser,Lewis A. [1964] (1956). TheFunctionsof Social Conflict.Glencoe, Illinois:Free Press. is AssistantProfessorof Sociology-at GARCIA ANGELA the Universityof Wisconsin- Eau Claire's Depart- Coulter,Jeff. 1990. "ElementaryPropertiesof ArgumentSequences."Pp. 181-204in InteractionCommentof SociologyandAnthropology.Her currentresearch in conversationanalysisfocuses on disputing petence. Washington,DC: InternationalInstitute

834

AMERICANSOCIOLOGICALREVIEW

for EthnomethodologyandConversationAnalysis Katriel,Tamar.1985."Brogez:RitualandStrategyin IsraeliChildren'sConflicts."Languagein Society and UniversityPressof America. 14:467-90. Dingwall,Robert.1986. "Some Observationson Divorce Mediationin Britainandthe United States." O'Barr,WilliamM. andJohnM. Conley. 1985. "Litigant SatisfactionversusLegal Adequacyin Small MediationQuarterly5:5-23. Claims CourtNarratives."Law & SocietyReview Emmison,Michael. 1987. "Victorsand Vanquished: 19:661-701. The Social Organizationof CeremonialCongratulationsandCommiserations."Languageand Com- O'Donnell, Katherine.1990. "Differenceand Dominance: How Labor and ManagementTalk Conmunication7:93-110. flict."Pp. 210-40 in ConflictTalk:Sociolinguistic . 1988. "On the Interactional Management InvestigationsofArgumentsin Conversation,editof Defeat."Sociology 22:233-51. ed by A. Grimshaw.Cambridge,England:CamFolberg,Jay. 1983. "A MediationOverview:History bridgeUniversityPress. andDimensionsof Practice."MediationQuarterly Pomerantz,Anita. 1975. "Second Assessments: A 1:3-13. Studyof Some Featuresof Agreements/DisagreeGirdner,Linda K. 1985. "Adjudicationand Mediaments."Ph.D. dissertation.Departmentof Socioltion: A Comparisonof CustodyDecision-Making ogy, Universityof California,Irvine. ProcessesInvolvingThirdParties."Journalof Di. 1978a. "Compliment Responses: Notes on vorce 8:33-47. the Co-operationof MultipleConstraints."Pp. 79Goffman, Erving. 1967. InteractionRitual: Essays 112 in Studies in the Organizationof Conversaon Face-to-FaceBehavior.GardenCity,NY: Doutional Interaction,edited by J. Schenkein. New bleday. York:AcademicPress. . 1981. Forms of Talk. Oxford, England: . 1978b. "Attributions of Responsibility: Basil Blackwell. Blamings." Sociology 12:115-21. Goodwin,Charles. 1987. "Forgetfulnessas an Inter. 1984. "Agreeing and Disagreeing with active Resource." Social Psychology Quarterly Assessments: Some Featuresof Preferred/Dispre50:115-31. ferredTurn Shapes."Pp. 57-101 in Structuresof Goodwin,MarjorieHarness.1982."Processesof DisSocial Action: Studies in ConversationAnalysis, pute ManagementAmong UrbanBlack Children." edited by J. M. Atkinson and J. Heritage.CamAmericanEthnologist9:76-96. bridge,England:CambridgeUniversityPress. . 1983. "Aggravated Correction and Disagreementin Children'sConversations."Journal Roberts,Marian.1988.Mediationin FamilyDisputes. Hants,England:WildwoodHouse. of Pragmatics7:657-77. Goodwin, MarjorieHarness and CharlesGoodwin. Sacks, Harvey. 1987. "Onthe Preferencesfor Agreement and Contiguity in Sequences in Conversa1987. "Children'sArguing."Pp. 200-48 in Lantion."Pp. 54-69 in Talkand Social Organization, guage, Gender and Sex in ComparativePerspeceditedby G. ButtonandJ. Lee. Clevedon,England: tive, edited by S. Philips, S. Steele, and C. Tanz. MultilingualMatters. Cambridge,England:CambridgeUniversityPress. Greatbatch,David. 1988. "ATurn-TakingSystemfor Sacks, Harvey,EmanuelSchegloff, and Gail Jefferson. 1974. "A SimplestSystematicsfor the OrganBritish News Interviews."Language in Society ization of Turn-Takingfor Conversation."Lan17:401-30. . Forthcoming."The Managementof Disguage 50:696-735. agreementBetween News Interviewees."In Talk Schegloff, Emanuel.1980. "Preliminariesto Preliminaries: 'CanI Ask You a Question?"'Sociological at Work,edited by P. Drew and J. Heritage.CamInquiy 50:104-52. bridge,England:CambridgeUniversityPress. . 1988/1989. "From Interview to ConfronHeritage,John. 1984. Gaifinkeland Ethnomethodoltation:Observationsof the Bush/RatherEncounogy. Cambridge,England:Polity PIress. ter."Researchon LanguageandSocial Interaction . 1987. "Ethnomethodology." Pp. 224-72 22:215-40. in Social TheoryToday,editedby A. Giddensand . 1987. "Analyzing Single Episodes of InJ. Turner.Cambridge,England:Polity Press. teraction:An Exercise in ConversationAnalysis." . 1988. "Explanations as Accounts: A ConSocial PsychologyQuarterly50:101-14. versationAnalyticPerspective."Pp. 127-44 in Analysing Everyday Explanation: A Casebook of Schegloff,EmanuelandHarveySacks. 1973. "Opening up Closings."Semiotica7:289-327. Methods,editedby C. Antaki. London:Sage. Heritage,Johnand David Greatbatch.1991. "Onthe Simmel,Georg. [1908] 1955. Conflict.Translatedby K. Wolff. New York:Free Press. InstitutionalCharacterof InstitutionalTalk: The Case of News Interviews."Pp. 93-137 in Talkand Vuchinich, Samuel. 1984. "Sequencingand Social Structurein Family Conflict."Social Psychology Social Structure,edited by D. Boden and D. H. Quarterly47:217-34. Zimmerman.Cambridge,England:Polity Press. . 1990. "The Sequential Organization of andHedges in Huebler,Alex. 1983. Understatements JohnBenjamins. Closing in VerbalFamily Conflict."Pp. 118-138 English.Amsterdam/Philadelphia:

DISPUTE RESOLUTIONIN MEDIATION

835

in Conflict Talk: SociolinguisticInvestigationsof Analysis."Social Problems35:335-62. Argumentsin Conversations,edited by A. Grim- Worley, Susan M. and Andrew L. Schwebel. 1985. "TheEffect of Cooperationon Egocentrismin Dishaw. Cambridge,England:CambridgeUniversity Press. vorce Mediation:A SimulationStudy."Pp. 151-65 in Divorce Mediation:Perspectiveson the Field, Whalen, Jack, Don H. Zimmerman, and Marilyn Whalen. 1988. "WhenWordsFail: A Single Case editedby C. Everett.New York:HaworthPress.