Lippi-Green-Accent, Standard Language and

7/27/2019 Lippi-Green-Accent, Standard Language and ...

1/36

Language in Society 23, 163-198. Printed in the United States of America

Accent, standard anguage deology, anddiscriminatorypretext n the courtsROSINA LIPPI-GREEN

Department of Germanic Languages and LiteraturesUniversity of MichiganAnn Arbor, Michigan 48109-1275

ABSTRACTTitleVIIof the U.S. CivilRightsAct clearly orbidsanemployero dis-criminateagainstpersonsof color for reasonsof personalor customerpreference.Similarly,a qualified ob applicantmay not be rejectedonthe basisof linguisticraits inked o nationalorigin.Incontrasto racialdiscrimination, owever,an employerhas considerableatitude n mat-ters of language,providedn partby a judicialsystemwhichrecognizesin theory the linkbetween anguageand social identity,but in practiceis often confoundedby blindadherenceo a standardanguagedeology.Thenatureandrepercussionsf thistypeof linguisticdiscriminationrehereexplored. Language ndlaw,accent,discrimination,tandardan-guageideology,critical anguagestudies)*

"The stranger within my gate,He may be true or kind,But he does not talk my talk -I cannot feel his mind.I see the face and the eye and the mouth,But not the soul behind.The men of my own stock,They may do ill or well,But they tell the lies I am wonted to,They are used to the lies I tell;And we do not need interpretersWhen we go to buy and sell."

Rudyard Kipling

In 1965,at theageof 29, SulochanaMandhareeft her homein Maharash-tra, India, and came to the United States. At that point in her life, Ms.? 1994 Cambridge University Press 0047-4045/94 $5.00 + .00

163


2/36

ROSINA LIPPI-GREEN

Mandhare a nativespeakerof Marathi had beenstudyingEnglishforalmost 20 years.Ms. Mandhare s soft-spoken; he speaksan Englishwhichis character-ized by full vowels in unstressed yllables,distinctive ntonationpatterns,aspirated ricatives, nda lack of distinction etweennitial v/ and /w/. Sheis an intelligentand articulatewoman,and she tells her storyin a clearandcompletelycomprehensibleanguage.After some timein the U.S., Ms. Mandhare elates, she decided o con-tinuehereducation.She had arrivedwithundergraduateegreesn bothlib-eral artsandeducation;but she returnedo school, and in 1972completeda master'sdegree n educationat New Orleans'sLoyola University. n 1979she wascertifiedas a school librarianaftercompletinga programat Nich-ols StateUniversity.After workingfor one yearas an elementary choollibrarian,Ms. Mandhare pplied or and wasgivena job as a librarian t aschool servingkindergartenhroughsecondgrade n the Lafargue,Louisi-ana, school district,for the 1980-81 schoolyear.Ms. Mandhare peaks of that yearas a happyand successfulone. Herresponsibilities ere o oversee hesmall ibrary, eadstories o thechildren,and introduce hemto using heresources;heenjoyed hiswork.Therefore,whenin April 1981 she was told that her contractwould not be renewedbecauseof her"heavyaccent,speechpatterns,andgrammar roblems" inspiteof her excellent killsas a librarianMandhare 985:240-41) she wasstunnedand angry.'She investigatedheroptions;and becauseshe under-stoodthat the U.S. CivilRightsActprohibitsdiscriminationynationalori-gin in the workplace,she filed suit. This civil action was decidedin Ms.Mandhare'savor, but the decisionwasreversedby the U.S. Court of Ap-pealsin favorof the schoolboard.2Ms. Mandhare'sase,and others ikehers,are mportant ecausehey pro-videreal-lifeexamplesof many phenomenawhich havelong been of inter-est to linguists.There is a body of work on the processes nvolvedwhenlistenersevaluatespeakers Lambertet al. 1960, Carranza& Ryan 1975,Rickford1985),on social stereotypingbased on language Lambert1967,Giles&Ryan 1982),on thepsychologicalrocessesnvolved n speechaccom-modation(Giles 1984,Giles& Coupland1991),on the cognitiveprocesseswhichstructure ollaborationn discourseClark& Wilkes-Gibbs 986),andon language-focusediscriminationLabov1969,Giles1971,Kalin&Rayko1978, Milroy& Milroy 1985, Rickford& Traugott 1985).Morerecently,therehasdevelopeda bodyof workon the relationship mong institution-alizedpower constructs, deology, and language (Thompson 1984, Kress1985, Fairclough1989).Butin spiteof such extensive nquiry,many areasremainunexplored.Onesuch areais the rangeof ways in which accent isdefined,and how it is put to use.Accent is used by phoneticians o discuss pitch or stress, or by orthog-164 Language in Society 23(2)


3/36

ACCENT AND DISCRIMINATORY PRETEXT IN THE COURTS

raphers o refer to specificdiacritics.Moregenerally,however,the term isused as a looselydefinedreference o setsof distinctive ifferences ver geo-graphicor socialspace, mostusuallyphonologicalandintonationfeatures.In the case of secondlanguage earning,accent may refer to the carryoverof nativelanguagephonologyand intonation nto a targetlanguage.One of the first, and sometimesmost difficult, lessons for a linguist intrainingis the abandonmentof subjectiveevaluations.In the pursuit ofknowledgeaboutthe structure ndfunctionof language,heavily nfluencedby scientificmethod,beliefhasno place; t can serveonly to obscure he pro-cess of discovery.Linguistsproceed on the assumption hat all naturallyoccurringanguages,whether r nottheyhave a literateradition,are equallyfunctionalandhave the samepotential o developfurther unctionsas nec-essary; herehas been no evidence n the many yearsof inquiry o disprovethis basicthesis.Linguistsurtherdifferentiateanguage romspeech,speechfromcommunication,andfluency fromcommunicative ompetence. Likeaccent, luency is a general ermwithout technicaldefinition.)The crucialconceptof communicative ompetence s definedas the abilityto use andinterpret anguage n a stylisticallyandculturallyappropriatemanner.Thismovesfarbeyond heset of phonological ndintonation eatureswhichbun-dledtogethermaybe markedas accent.

Thegeneralpublic,however,does not make suchdistinctions.For mostpeople,accent s a dustbincategory: t includesall the technicalmeanings,and a moregeneralandsubjectiveone:accent s how theotherspeaks.It isthe firstdiagnostic or identificationof geographicor socialoutsiders.Fora native of the north side of Chicago- a cab driver,elementaryschoolteacher,or district udge- all the following"havean accent":peoplefromsouthern ndiana,Georgia,Brooklyn,England,or SouthAfrica;the nativespeakerof African AmericanEnglishVernacularwho lives down the streetor west of theLoop;the co-worker romJamaica; nd themansellingpaperson the cornerwhose Guatemalanphonologyand intonation shinethroughintohisEnglish.No distinction s madebetweenpidginor creole,sociallyorgeographically asedvariation,nativeor nonnativeanguage:heyare alljustaccents,whichmaybe describedas adenoidal,barbarous,broad, cute, dis-tinct, educated,flat, foreign, funny, guttural,harsh, heavy, lilting, nasal,posh, provincial,quaint,rough, rustic,sing-song,strong,and uneducated(McArthur1992:10).The subjectivenatureof thesequalifiers s clear.Muchof linguisticvariations structured roundsocialidentity.Linguistsknowthis, butnonlinguistsknowit too, and act on it: accentbecomesbothmannerand means for exclusion. The fact is, however,that whenpeoplerejectan accent,theyalso rejectthe identityof the personspeaking:his orherrace,ethnicheritage,nationalorigin,regionalaffiliation,or economicclass.3Thusthe conceptof accent,so all-encompassingn the mind of thepublic,is a powerfulone whichneedsto be investigated.Language in Society 23(2) 165


4/36

ROSINA LIPPI-GREEN

Inthe remainder f this article,my goal is to illustratehenatureand someof therepercussionsf accentdiscrimination.nthe process,I hopeto dem-onstrate hat accent particularly henassociatedwithracial,ethnic,or cul-turalminorities is most likelyto posea barriero effectivecommunicationwhen two elementsare lacking.The first element s a basiclevel of commu-nicativecompetenceon the partof the speaker, ndependent f Li phonol-ogy and intonation.Thesecondelement,even more mportantbut farmoredifficult to assess, is the listener'sgoodwill. Without that goodwill, thespeaker's ommandof thelanguage, .e. his or herdegreeof communicativecompetence,s irrelevant. rejudicedisteners annothearwhata personhasto say, becauseaccent,as a mirrorof social identityand a litmus test forexclusion,is moreimportant.After a moregeneraldiscussionof backgroundssues, theexamination faccentdiscrimination, eferred o heremore specificallyas language-traitfocused (LTF)discrimination,s limitedbecauseof spaceconsiderationsothe workplaceand the courts. Moregenerally his is the beginningof anexplorationof whyso manyof us continueto use linguistic raits to ratio-nalize andjustifydiscrimination f all kinds- and to toleratesuchdiscrim-ination, even when it is directed owardourselves.STANDARD LANGUAGE IDEOLOGYIn mattersof languagehistory,structure, unction,andstandardization,heaverageindividual s, for the most part, simultaneouslyuninformedandhighly opinionated.Whenasked directlyaboutlanguageuse, most peoplewilldrawa verysolid basicdistinction f "standard"proper, orrect)Englishvs. everything lse. If askedfor a more exactingdefinition,mostwillnot beable to provide t, or willcouchit in termsof salientfeaturesof nonmain-stream anguagevarieties:"ProperEnglish s havingyoursubjectsand verbsagree"; "Why can't they see that the word is spelled a-s-k, not a-x?";"[kwifi] - that sounds so ignorant."LTFdiscriminationtemsprimarilyrom theacceptance f a standard an-guage deology a termcoinedby Milroy&Milroy1985).The definitionusedhere s: a biastowardanabstracted,dealized,homogeneouspoken anguagewhich s imposedfrom above, and whichtakes as its modelthe written an-guage.The most salientfeature s thegoalof suppression f variationof allkinds.4What is the source of the standard anguage (SL) ideology?How is it"imposed rom above"?Who is responsible or its propagation?SL ideology s partof a greaterpowerconstruct,a set of socialpracticeson whichpeople dependwithoutcloseanalysisof underlying ssumptions.In a thought-provokingiscussionof therelationship etween anguageandsocialpower,Fairclough 1989:33)pointsout thatthis institutionalizationf166 Language in Society 23(2)


5/36


behaviorswhichoriginatewiththe dominantbloc (an allianceof thosewhosee theirinterestsas tied to capitaland capitalism) unctionsto keep sepa-rate the poweredand the disempowered:Ideologicalpower,the powerto projectone'spracticesas universaland"commonsense,"is a significantcomplement o economicand politicalpower, and of particular ignificancehere becauseit is exercised n dis-course . . . There are . . . in gross terms two ways in which those whohave powercan exerciset and keep t:throughcoercingothers o go alongwiththem, with the ultimatesanctionsof physicalviolence or death;orthroughwinningothers'consent o, or at leastacquiescencen, theirpos-sessionof exerciseof power.In short, throughcoercionor consent.

The SLideology s oneroute,and a majorone, to establishingonsent.Therearefour immediatelydentifiable roponents f SLideology,all of whicharepartof the "dominantbloc": the educationalsystem,the newsmedia,theentertainmentndustry,andwhathasbeengenerally eferred o as corporateAmerica.At the endof thisarticle,I argue or adding hejudicialsystem othis list.The educational ystemand standard anguage deologyMuchof whatthe American ducational ystem eacheschildrenaboutlan-guageis factually ncorrect;n this it is thorough,consistent,andsuccessfulacrosssocialandeconomicboundaries.Thephenomenonhas been observedby others:

It is a tribute o our educational ystemthat the overwhelmingmajorityof Americans avebeen nstilledwith a rocklike onvictionhat certainin-guisticformsarecorrect,whileothers arewrong.Even those Americanswho are uncertain boutpreciselywhichformsarecorrectareusuallycon-fident hatto findthe answerheyneedonlylook the matterupin therightbook or consult the properauthority.(Burling1973:130)Theseare strongstatements,but theyareeasilyverified.Everyonehas anec-dotesabout anguage rts nstructionromtheirelementarychool education,but stronger vidence s available n a widerangeof textswritten or teach-ers and children.The underlyingmessage s clear in each of the followingexamples.6

(a) A direct inkbetween"nonstandard"anguageandlack of logic andclarity,withblurringof the written/spokenboundaries:Almost any sentenceor sentencefragmentmay be acceptable n casualconversation.n more formalspeaking ndwriting,however,nonstandardgrammars rarelyacceptable.We need to know howto speakand write

Language in Society 23(2) 167


6/36

ROSINA LIPPI-GREEN

in complete, grammatical sentences that convey our thoughts clearly toothers. (Ragno et al. 1987:T22)(b) There is one correct way to speak and write English:[This series of textbooks] focuses on grammar study, listening and speak-ing skills, and correct usage. (Strickland 1983:T21)(c) Overt authoritarianism:Practice saying the following combinations of words. Avoid slurring anysounds, such as whacha for what do you ... Whip is pronounced hwip,not wip . .. pronounce the following troublesome words correctly. Con-sult the dictionary if you are in doubt . .. Twenty-five words often mis-spelled because of faulty pronunciation: busy, which, since, history ...(John et al. 1975:28-9)7

SL ideology is found at work not just in textbooks and language arts instruc-tion classrooms, but also in school administration. In 1987, the Board ofEducation of Hawaii put forth a proposed policy on "StandardEnglish andOral Communication,"which would have outlawed Hawaiian CreoleEnglish(HCE) in the schools. A survey of 986 graduating seniors, conducted by aHonolulu newspaper, indicateshow well many of those students were indoc-trinated in the SL ideology, and serves as an illustration of the ideology'sclose relationship to issues of race and economics.Whereas only 26 percent of the private school students surveyed felt thatHCE use should be allowed in school, 54 percentof the public school stu-dents supportedthis idea . . . Comments ranged from "Pidgin English fos-ters illiteracy," "Pidgin is a lazy way to talk; it promotes backwardthinking," and "CorrectEnglish will get you anywhere"to the polar oppo-sites of "Banning pidgin would violate our freedom of speech," "Pidginis a natural language," and "It's our way to make Hawaii different fromanywhere else in the United States." (Verploegen 1988, as cited by Sato1991:654)Many empirical studies of American students also illustrate this phenom-enon. Using matched-guise testing, Carranza & Ryan 1975 showed thatAfrican-American, Anglo, and Hispanic students all found Spanish-accentedEnglish to be lacking in prestige and inappropriate for a classroom setting;8in Ryan et al. 1977, "Small increments n accentednesswere found to be asso-ciated with gradually less favorable ratings of status, solidarity, and speechcharacteristics"(summarized by Eisenstein 1983:173).Are there no examples of educators with more informed and enlightenedapproaches to diversityin language? teachers who strive to teach children toread and write, and at the same time respect the sanctity of the home lan-

168 Language in Society 23(2)


7/36


guageand social identity? eacherswho questionunderlyingassumptions,andwho do not automatically ontribute o the propagationof the currentpowerdistribution?Of coursethereare.Takefor exampleMaryBergerofColumbiaCollegein Chicago- who, as an Englishteacher,"teaches tan-dardstyleto augment,not replace,dialect" Warren1993:2).The ChicagoTribuneoundMs. Berger'smethodsso remarkablehattheyranan articleon herapproach,andhighlighted er classroompracticeof "[not]scold[ing]her black students . . . when they said 'ax', rather than 'ask' . . . "For themostpart,however, eachersareboundby the standardanguageideology.Forexample,almost exactly15 yearsafterthe controversialKingcasewas decided n AnnArbor,parentsof AfricanAmericanmiddleschoolstudentscomplainedo the schoolboardabouta teacherwho allegedlyhadbeenridiculingBlackstudents or using heirhome language, pecificallyorsayingax insteadof ask (Windsor1993:C1,C3).9Standardanguagedeology s a basicconstruct f our elementary nd sec-ondaryschools' approach o languageand philosophyof education.Theschoolsprovide he firstexposureo SLideology,buttheindoctrinationro-cess does not stop whenthe studentsaredismissed.Themediaand the standard anguage deologyThemedia andbythisis usuallymeantnationalbroadcastingnstitutionshavetaken on the job of defending he "nationalculture" Cormack1993:102-3), whichmeansthe propagationof a homogeneousnation-state, nwhicheveryone mustassimilate rbe marginalized.As partof thisprocess,theprintand broadcastnewsmediaandthe entertainmentndustry ake onthe job of reinforcingSL ideologyon a dailybasis.10Perhaps he mostpervasive epresentativef the standardanguage deol-ogy is the newsmedia.This is accomplished,n part,by meansof language-conscious eporting,which sprescriptive ithout actualbasis.It is sometimesalso overtlydiscriminatory.An excellentexampleof this is the Hawai'ianprint media coverageof GovernorJohn Waihe'e,whose code-switchingbetweenHCEandcreole-accentednglishs madeanissue,and whosegram-matical"errors" recorrected reportedby the NationalPublicRadio show"AllThingsConsidered,"September12, 1990;also verifiedby the show'sreporterWilliamDrummond,p.c.).Moreusually,complaints bout anguageuseare tuckedawayas an after-thought,buttheunderlyingmessages clear: here s a rightanda wrongwayto talk, andit is perfectlyacceptable,evenjudicious,to censorandpunishthose who do not conform:

Residents f Brooklyn,New York,have ongbeenknown andsometimesmocked- for theirheavyaccents.GinnyMostreportson a groupof stu-dents who are tryingto learnto talk right- or should I say correctly.Language in Society 23(2) 169


8/36

ROSINA LIPPI-GREEN

[G.M.]:Somepeoplehave a funnywayof sayingwhat flows undertheBrooklynBridge .. [Student]: wata it'sso ugly"... (CNNHeadlineNews, March12, 1993)Ungrammaticaltreet alkby blackprofessional thletes,andotherblacksin publicprofessions uch as themusic ndustry,has come to be accepted... The dilemma is that it doesn't make much difference for the blackprofessionalathletes,etc., who talkthis waythey'rewealthymen whoaregoingto live well off theirbodilyskills no matter f theycan talk atall, much less correctly . . (Bob Greene's portscolumn,ChicagoTri-bune, December 3, 1979)11WelikeHahn,34, whowasborn in SouthKoreaandwhosepositionsoncontrollinggrowtharemuchlike our own. Unfortunately,we thinkhisheavyaccentand somewhat imitedcontactswould make it difficult forhimto be a councilman. "ForSantaClaraCounty,"San Jose MercuryNews, October18, 1988,as citedin Matsuda1991:1346)[Oprah]is an image. So is Jesse Jackson . . . They can effectively articu-latewithsubjectand verbagreement.And if it had notbeen for Godwhogivesus the wisdom- we have to attribute his to God - to know howtosound,to articulate nd to knowhow to usesubject-verb greements,wewouldn'tbe wherewe are today. (ToniTucker,African-Americanalkshowhost,asaudiencemember n "BlackEnglish,"OprahWinfrey how,November19, 1987)Gov. Clinton,you attendedOxfordUniversityn EnglandandYale LawSchoolin the IvyLeague,two of the finestinstitutionsof learningn theworld. So how comeyou still talk like a hillbilly? MikeRoyko'ssyndi-cated"Opinion" olumn,Ann ArborNews, October11, 1992)12

The mediaclaimthatthe intention s not to makenews, but report t, andthattheydo not intendto serveas an agentof socialchangeor an enforcerof norms. Of course,this lineis crossedrepeatedly y the media,simplybyvirtue of the topicschosenfor reporting. n bringing o the public'satten-tion the boom in accent-reductionchools,andby slanting he toneof theirreports owardanidealizedtandard,hemediabecomecomplicitn thepro-cess of discrimination.The SL ideology is introducedby the schools;it is vigorouslypromotedby the media,and(as is shownin the nextsections) s further nstitutional-izedbythecorporate ector.Thus t is not surprisinghatmany ndividualsdo not recognize he factthat, for spoken anguage,variations systematic,structured,andinherent,andthatthe nationalstandard s an abstraction.Whatis surprising, vendeeplydisturbing,s the way that manyindividu-als - thoughtheyconsider hemselvesdemocratic, ven-handed, ndfree of170 Language in Society 23(2)


9/36

ACCENT AND DISCRIMINATORY PRETEXT IN THE COURTSprejudice - hold tenaciously to a standard ideology which attempts to jus-tify restriction of individuality and rejection of the other.LTF discrimination can be found everywhere in our daily lives. In fact,such behavior is so commonly accepted, so widely perceived as appropriate,that it must be seen as the last widely open backdoor to discrimination.LTF DISCRIMINATION ND THE CIVIL RIGHTS ACTSome types of LTF discrimination have been illegal in the workplace since1964, when Title VII of the Civil Rights Act of 1964 (42 United States Code??2000e-2000e-17 [1982]) was passed into law.'3Title VII provides recourse for workers who are discriminated against onthe basis of race, color, religion, sex, or national origin.'4 However, it wasnot until 1980 that the Equal Employment Opportunity Commission(EEOC), a body created by Title VII, directly addressed trait-based discrim-ination.15In theirGuidelineson discrimination ecauseof nationalorigin,revised on a regular basis, the EEOC currently defines national origin dis-crimination: ". . . broadly as including, but not limited to, the denial ofequal employment opportunity because of an individual's, or his or herancestor's place of origin; or because an individual has the physical, culturalor linguisticcharacteristicsof a national origin group" (FederalRegister 1988,11606.1; emphasis added). The spirit of the law is clear: an employer maynot rejecta job candidate, or fire or refuse to promote an employee, becausethat employee externalizes in some way an allegiance to another culture.In the case of racial discrimination, "It is clearly forbidden by Title VIIto refuse on racial grounds to hire someone because your customers or cli-ents do not like his race" (Matsuda 1991:1376, fn. 169). Similarly, a quali-fied person may not be rejected on the basis of linguistic traits that theemployer or the employer'scustomersfind estheticallyobjectionable. In con-trast to racialdiscrimination,however, an employerhas some latitude in mat-ters of language: 16 "an adverse employment decision may be predicatedupon an individual's accent when - but only when - it interferes materiallywith job performance"(Civil Rights Act of 1964, ?701 et seq., 42 U.S.C.A.?2000e et seq.).'7Title VII is very limited in its scope. Under the law as it currently stands,discriminationon the basis of regional origination is not covered. An accentmust be directly traceable to a specific national origin to be eligible for TitleVII protection.Raj Gupta, attorney counsel to the commissioner of the EEOC, states(p.c.) that some forces within the EEOC would like to see the definition ofLTF-national origin discrimination made more comprehensive. So, in hisexample, a person from Appalachia would have recourse under Title VIILanguage in Society 23(2) 171


10/36

ROSINA LIPPI-GREEN

becausehe features f Appalachianredirectlyraceableo a number f dia-lects in GreatBritain.18The legal processAllegedLTF-national'originiscriminationasesusuallybeginwhenan indi-vidualfiles a complaintwiththe EEOC (or a similaragency on a state orlocal level). Theemployeemaythen file a civil actionin the trialcourts,inwhichhe or she claims that civil liberties,as set out in the federalstatutesknownas the CivilRightsAct of 1964,havebeenviolated. nsome nstances,these cases arebrought o the courtsnot by the individualorgroupof indi-vidualswith the same complaint, but by a privateagency acting for theinjuredparty, suchas the AmericanCivilLibertiesUnion(ACLU), or by agovernment gency,suchas the EEOC. This actionmaybe initiatedat thestatelevel, as manystates haveadoptedcivil liberties egislationpatternedon the federalstatutes.19An individual laimingLTFdiscriminationmustfirstproveaprima aciecase of disparate reatment, n four steps:(a) establishment f identifiablenationalorigin;(b) proof of application or a job for which he or shewasqualified,and for whichthe employerwasseekingapplicants; c) evidencethat the applicantwasrejected n spiteof adequatequalifications; nd (d)evidence that, after such rejection,the job remainedopen, and the em-ployercontinued o seekapplicantswiththe plaintiff'squalifications.Aftera primafacie case has beenestablished, he burdenshiftsto the employerto rebutpresumption f discrimination y articulatingomelegitimate,non-discriminatoryeasonfor the action.If the employerdoes this, the burdenshifts backto theplaintiff,to show thatthepurported eason orthe actionwaspretext orinvidious iscrimination. heplaintiff anshow heemployer'spretextdirectly,by demonstratinghat the employerwas morelikely moti-vatedby discriminatoryeasons;or indirectly, y showing hat theprofferedreason is unworthyof credence Civil RightsAct of 1964, ?701 et seq., 42U.S.C.A. ?2000eet seq.).DISCRIMINATION IN THE WORKPLACEIn an excellentstudy of languageand discriminationn the workplace nGreatBritain,Robertset al. 1992providednumerous xamplesof discrim-inationfocusedon language,and directed owardethnicand racialminori-ties.No suchsystematic ndwell-documentedtudyexistsfor workersn theU.S., although his is an areaof great mportance.The evidenceof discrim-inationprovidedhere s limited o specific nstanceswhichhave found theirway into the legal system.172 Language in Society 23(2)


11/36


TABLE 1. Distribution of 25 LTF discrimination cases in the courts/EEOChearings, by plaintiff's national originCourt Found foraPlaintiff's No. CasesNational Origin Filed Plaintiff Defendant

Asia, Pacific RimPhilippines: Lubitz,Fragante, Carino 3 1 2Vietnam: Tran 1 0 1China: Ang, Hou 2 0 2India: Duddey, Mandhare, Patel 3 1 1Cambodia: Xieng 1 1 0Korea: Park 1 1 0Subtotal 11 4 6Caribbean/West IndiesDominican Republic: Meijia 1 0 1Haiti: Stephen 1 0 1Cuba: Rodriguez 1 0 1Subtotal 3 0 3Central/South AmericaVenezuela: Dercach 1 0 1Bolivia: Ipina 1 0 1Subtotal 2 0 2Eastern EuropeArmenia: Vartivarian 1 0 1Poland: Berke 1 1 0Ukraine: Staruch 1 0 1Subtotal 3 1 2AfricaNigeria: Dabor 1 0 1Liberia: Andrews 1 0 0Ghana: Kpodo 1 0 0Subtotal 3 0 1OtherAfrican-American: Sparks, Edwards 2 2 0

Hawaiian Creole: Kahakua 1 0 1Subtotal 3 2 1Totals 25 7 15aDiscrepancy in some of the totals is due to the fact that one case (Patel) was settled out ofcourt; two others had not yet been decided (Andrews, Kpodo) at the time of this writing.

Table 1 provides a breakdown of 25 LTF-national origin discriminationcases heard in the federal and state courts and by the EEOC since 1972, withexceptions as noted. Furtherexcluded or missing are cases which concernedthe English-Only question (e.g. H. Garcia) and cases in which LTF discrim-ination played a minimal role in the plaintiff's arguments (C. Garcia, Bell,Language in Society 23(2) 173


12/36

ROSINA LIPPI-GREEN

and many others). In some of the cases included, both racial and nationalorigin discrimination were at issue. In most of the cases, accent, languageuse, and communication figured prominently in the testimony, argumenta-tion, discussion, and final opinion.20How widespread is LTF discrimination? The General Accounting Officeof the United States Government(GAO GGD 90-62 Employer Sanctions, 27)conducted a carefully designed statistical study of a stratified random sam-ple of employers nationwide, and reported that 10%oof their sample, or461,000 companies employing millions of persons, openly if naively admitthat they "discriminated on the basis of a person's foreign appearance oraccent" (ibid., 38). In hiring audits, specifically designed to detect discrimi-nation on the basis of accent (telephone inquiries about advertised obs), suchdiscrimination was found to be prevalent (ibid.).21 This type of behaviorwas documented again in Carroll, when an employment agency receptionistwas directed by her manager to screen all persons inquiring over the tele-phone: to those who did not "speak right," the job was closed. The recep-tionist was also told to make notations about the caller's speech and accent(Carroll, 1173).There are a numberof possible reasons for the low numberof documentedcases. Employers who discriminatemay do so in a nonblatant way; the per-sons discriminatedagainst may be so accustomed to this treatment that theyno longer react; if they are aware of the treatment, they may not know thatthey have legal recourse, or how to pursue it; complaints may be handledinternally,and resolved before litigationbecomes necessary.Of course, manypeople discriminatedagainst on the basis of language may not find anythingsurprising or wrong about that fact. This is, after all, not the only societyin the world that promotes a standard language ideology.The bulk of the burden seems to fall, predictably, on the disenfranchisedand the unassimilated. Cutler(1985:1164)claims that the manner of enforce-ment of Title VII "permitsan employerto reject qualifiedapplicantsof a par-ticular national origin as long as he hires more assimilated applicants of thesame origin instead."Once cornered in a courtroom, what do the employers offer by way ofexcuses? The approaches taken by defendants range from the naively andopenly discriminatory to the subtle.

In offering examples of Mr. Dercach's communication problems, Mr.Moser explained that workers would ask Mr. Dercach what he wantedthem to do, and then simply walk away, unable to understand. Mr. Moserrefused to attributesuch incidents to Mr. Dercach'saccent, but offered noother explanation. He said they just couldn't understand him "likenormalpeople with normal language." (Dercach, 899)



13/36


Afterlisteningo thetransmissionescribed y DispatcherMixonasjar-gon, . . . Rodriguezclaims that during[a telephone]conversationSgt.McElligatoldhimto "speakEnglish ikein Queens,NewJersey,not Lit-tle Havana."Sgt. McElligat estified thathe could not recalleverhavingtalked to Rodriguez. Rodriguez,LEXIS)Managerial evelemployeeLindaSincoff told Xieng he was not beingpromotedbecausehe couldnot speak"American."Xieng,AppealCourtOpinion, 5)... the complainant's supervisor had removed her because of concern

aboutthe effect of heraccenton the "image" f the IRS, not any lack ineither communicationor technicalabilities.(Park, EEOCpressreleasedated June8, 1988)... the ability to speak clearly is one of the most importantskills . . . we felt the applicants selected would be better able to work inour office becauseof theircommunication kills. (Fragante1989:598)22

So thecourthasbefore t a plaintiffwhoclaims hat his or herbasiccivillib-erties have been violated, and an employerwho claimsthe rightto makeappropriatebusinessdecisions. How do courts handlethis conflict?Whatfactors, legal and otherwise,play a role in the decision-making rocess?Some of those factors have to do withtechnicalities f the law and stan-dards orevidence:n Vartivarian,orexample, heplaintiffpresented s evi-denceonly doublehearsay:"XsaidY was angrybecausea personwithanaccent[plaintiff]had been hiredbehindhis back."Therewasno direct es-timony, or any way to corroborateVartivarian's laims.In some cases one must assumethat a plaintiffmayclaimLTFdiscrimi-nationwhen n fact none has takenplace.Ortheremaybe clearevidenceofLTFdiscriminationwhichthe courtoverlooksbecause here s, in addition,a bona fide reason o denyemployment. nDercach,the court felt that bla-tant LTFdiscrimination ould not mitigate he factthat the plaintiff, whilehardworkingand knowledgeable,was illiterate.Because the job requiredcloseworkwith a written odebook, andtheability o writemultiple eportson a weeklybasis, the court found for the defendant.The courtshave stated hat "there s nothing mproper boutan employermakingan honestassessmentof the oralcommunicationskillsof a candi-date for a job whensuchskills arereasonably elated o job performance"(Fragante1989:596-7).Matsuda1991callsthis thedoctrinalpuzzle f accentand antidiscriminationaw: TitleVII disallowsdiscrimination n the basisof accentwhen t correlateso nationalorigin,but it allowsemployerso dis-criminate n the basis of job ability.Employers laimthat "accent"mpedesLanguage in Society 23(2) 175


14/36

ROSINA LIPPI-GREEN

communication,ndtherebyposesa validbasisfor rejection;Matsuda oundthat the courts areespeciallyreceptive o this argument 1348 ff.).23Employers urtherpoint out that the decision-makingrocess n businessis often unavoidably ubjectiven nature.The courtshave supported hemin this.It doesnot follow, though, thatethnicdiscriminations the onlyexplana-tion why Plaintiff was not promoted.Otherplausibleexplanationsmayexist. For instance,Nassermaynothavechosen o promotePlaintiffsim-ply becausehe personally idnot like her.Whilemakingallowanceor thiskind of decisionalcriterionwould arguablycall into play the "businessjudgment" ule enunciatedn Williams,he court does not reach ts con-clusion on the basis that it cannot reviewDefendant'sprofferedreason.(Vartivarian, 558)

But howcan the courtsdistinguish n admissiblebusiness udgment,basedon businessnecessityor personalpreference, rom inadmissible onsider-ations, based on race or nationalorigin?Is it simplya matterof presenta-tionof therightarguments ytheemployer?Cutler1985haspointedout thatemployersarefavorablypredisposedo potentialemployeeswho are"like"them,and lessdisposed owardpotentialemployeeswhoare "unlike" hem.Because he courts fail to recognize his fact, andrefuseto rejectthe valid-ity of the personalpreference ationale,"TitleVIIbecomesa statutewhich,at best, coerces ob applicants o assimilateand, at worst, keepsthemjob-less" (1985:1166).I proceed rom thepointwhere he plaintiffandthe defendanthave madetheircases;the courtmust now decidewhoseargumentation etterfulfillsthe requirementset forthby the law. It is possible o tracethe influenceofthe standard deology throughmuch of the court'sdeliberations.STANDARD LANGUAGE IDEOLOGY IN THE COURTROOMThe opinionsput out by the courtsdisplaya rangeof approaches owardcommunication ndaccent.Oneassumes hat the courtsareunbiased,andsometimes hereis evidenceof that.

Accentand nationaloriginareobviously nextricablyntertwinedn manycases. It wouldthereforebe aneasyrefuge n this context or anemployerunlawfullydiscriminating gainstsomeonebased on nationalorigintostate falselythat it was not the person'snationaloriginthat cause theemploymentor promotionproblem,but the candidate'snability o mea-sureupto thecommunicationskillsdemandedbythejob. Weencouragea very searching ook by the [trial]courts at such a claim. (Fragante1989:596)



15/36


Testimonyof both Plaintiff'sand Defendants'witnesseshaveconvincedthe Court hat the Plaintiff'saccentwas a majorfactor n theDefendants'evaluation of his supervisory abilities . . . a trait related to national ori-ginmustbe of animmutable ature n order o comewithinTitle VIIpro-tections ... An accentwouldappear o approach hat sortof immutablecharacteristic . . . (Carino, 1336-7)Plaintiff'saccentdidnot interferemateriallywithhisjob performance, orwould it have interfered materially with his job performance . . . ifhe had been promoted . . . (Xieng, Supreme Court Opinion, 2)But at the sametime, and sometimes n thesame cases, it is clearthat thecourtsarewillingto dependon their own often factually ncorrectunder-standingof language ssues.

Fraganteargues hedistrictcourt erred n considering listener rejudice"as a legitimate, ondiscriminatoryeason or failing o hire.Wefind,how-ever,thatthe district ourtdid not determinethat]Defendants efused ohireFraganteon the basis thatsomelistenerswould "turnoff" a Filipinoaccent.Thedistrictcourtaftertrial notedthat:"Fragante,n fact, has adifficult manner of pronunciation . . . " (Fragante 1989:597)The judge discounted he testimonyof the linguistwho statedthat Ha-waiianCreolepronunciations not incorrect,rather t is one of the manyvarietiesof pronunciationof standardEnglish. The linguist, thejudgestated,wasnot anexpert nspeech.(Matsuda1991:1345-6,ncludingquo-tations fromKahakua1987:22-3,emphasisadded)24Duringthe Vietnamconflict, Mr. Tranworkedas an interpreteror theU.S. forces. That hasmisledhimto believe hat hisEnglish s better hanit is in reality.OccasionallyMr. Tran'sspokenEnglish s readilyunder-stood,whileother imes t is understood nlywithdifficultyandsometimesnot at all (Tran,472).

The judges who wrotethese opinionsare willing to dependon theirownexpertise n mattersof language n a waythey wouldneverpresume o inmattersof geneticsor mechanical ngineering r psychology.In Kahakua,the judgeheardtestimonyof expertwitnesses,and then chose to give cre-dence o thatwitnesswhose estimonymostcloselymatchedhis ownpersonalopinions on mattersof languageuse. In none of these caseswas thereanyattempt o assess hecommunication emandsof thejob in a non-prejudicialway, and intelligibilitywas a matterof opiniononly.How do someplaintiffsmanage o win?Xiengprovidesan exampleof asuccessfulcase.PhannaXieng is a Cambodian-American ho workedfor PeoplesNa-tionalBank of Washington.Mr.Xiengwasrepeatedlydenied a promotionLanguage in Society 23(2) 177


16/36

ROSINA LIPPI-GREEN

although he had an excellent work history, high marks in his reviews, andfor an extended period had been filling in on the very position for which hewas applying. There were documented comments from his superiorsconcern-ing his accent as the primary stumbling block to his promotion. In this case,the court could not overlook the fact that Mr. Xieng could carryout the jobhe claimed he could do, in spite of his accent, precisely because he hadalready been performing well at the job. It might seem that being on theinside - already employed by the defendant - would provide an employeewith a valid LTF discrimination complaint with some strong evidence; butthere are many other cases of denied promotion which were not so success-ful as Xieng.Is it the case, then, that the plaintiff's chances of winning a LTF discrim-ination case depend to the greatest degree on the integrity and objectivity ofthe judge hearingthe trial?Unfortunately, it is not so easy as this. It becomesclear later that, for some areas of employment, even the most open-mindedof courts may be subject to the unwritten laws of the standard languageideology.Education-related casesI consider here four cases in which educators sued their respective schoolsor school systems for racial and/or LTF-national origin discrimination:

(a) Sparks: an African-American who was dismissed from her job as aschool teacher.(b) Hou: a native of China and professor of mathematicswho was refusedpromotion.(c) Edwards: an African-American whose teaching contract was notrenewed.(d) Mandhare: a native of India who was denied reappointment to herposition as a librarian at a K-2 school.Ms. Sparks and Ms. Edwardswon their cases; Ms. Mandhare won at trialcourt but lost on appeal; Dr. Hou lost his case.Academic institutions were meant to be included within the scope of TitleVII; nevertheless, the "trend in many courts has been to exercise minimalscrutiny of college and university employment practices, due, in large part,to the subjective factors on which many academic employment decisions arebased" (Hou, 1546). They will intercede, but seem to do so with consider-able forbearance for the opinions put forth by school administration. Inaddition, the courts have shown reluctanceto reverseadministrativedecisions

(ibid., 1958).This deference for academic decision making was the downfall of Hou.The judge pointed out:178 Language in Society 23(2)


17/36


The issue of accent in a foreign-born person of another race is a conced-edly delicate subject when it becomes part of peer or student evaluations,since many people are prejudiced against those with accents. (Hou, 1547)The judge went on to approve the loophole used by the institution.

We find that comments about Dr. Hou's accent, when made, were directedtoward the legitimate issue of his teaching effectiveness. Teaching effec-tiveness, as the testimony at trial indicated, is an elusive concept ...Teaching effectiveness does, however, include the ability to communicatethe content of a discipline, a quality which should be carefully evaluatedat any college or university. (ibid.)There was never any discussion of appropriate,nonprejudicial assessment ofDr. Hou's communicative competence or intelligibility.The defense dependedexclusively on anecdotal evidence provided by the defendant, and this sat-isfied the court.[The college records showed that] he is at a decided disadvantage in theclassroom because of his natural accent . . . he has a difficult time over-coming this handicap. The obvious grammatical errors on his applicationattest to his communication problems . . . (Hou, 1547).

The question must be, then, why other education cases prevailed where Houcould not. I consider Sparks and Edwards before a discussion of Mandhare.Sparks and Edwardswere built primarilyon racialdiscrimination.In manypages of correspondenceon the matter of Ms. Sparks'sdismissal, the schooladministrator (Mr. Griffin) commented only once on the language issue:"Mrs. Sparks has a language problem. She cannot help the negro dialect, butit is certainlybad for the children to be subjected to it all day" (Sparks, 437).In Edwards, the discussion of language use is limited to general comments:"The plaintiff's contract was not renewed allegedly because of complaintsreceived from parents and students . . . Several complaints concerned stu-dents' alleged inabilityto understandthe plaintiff's 'black accent'" (Edwards,LEXIS).In both these cases, the opinions indicate that the heart of the matter wasracial discrimination.In otherwords, if the accent issuehad never been raisedin Sparks or Edwards, these plaintiffs would still have won. This was for-tunate for the courts, as it relieved them of the trouble of dealing with thematter of language and accent. In discussingthe LTF discriminationportionof Sparks, the court limited its comments to one short footnote: "With nodisposition to be unkind, we question, based on the spelling and composi-tion of the two letters . . . the ability of Mr. Griffin to diagnose a 'language-problem'" (Sparks, 442). The letters written by Mr. Griffin regarding thedismissalof Ms. Sparks,to which the court referred, were in fact poorly writ-Language in Society 23(2) 179


18/36

ROSINA LIPPI-GREEN

ten, and containedmanyspellingand/ortypographicalrrors.Nevertheless,the court is clearlyuncomfortablen chidingan educator in this case, anadministratorwith advanceddegrees) n mattersof languageuse: "withnodisposition o be unkind."More mportantly,he courtneveraddressed hecontent of Mr. Griffin'scomplaint Ms. Sparks's"negrodialect"and itsappropriatenessor the classroom; t addressedonly the superintendent'squalificationso make udgments n thatdialect,givenhispoor letter-writingskills.Would hecourthavethoughtseriouslyaboutthiscriticismf Mr.Griffinhadwrittenelegant,grammaticallyppropriate rose? f he had argued hatMs. Sparks'seachingeffectivenesswascompromised y her languageuse?It seems likelythat the school systemcould have found a line of argumen-tationwhichwould havepleased hecourts; heyfailedto do so in thiscase.Thecourtneatlysidesteppedhe"concededlyelicate ubject" f LTFdis-criminationorEdwards s well:"Thedistrict ourtstated n itsopinion hatit was'apparent'hatthe plaintiffcould be easilyunderstood ndthattherewasno evidence he plaintiffmadegrammatical rrorsrendering erspeechdifficult o understand."nthese wocases,the schoolsweredeservedly un-ishedfor racialdiscrimination;orLTFdiscrimination,heywereslappedonthe wrist.

I returnnow to the Mandhare ase, with whichI began.Earlier t wasestablishedthat Ms. Mandhare's ontract as a school librarianwas notrenewedafterthat first yearbecauseherdutieswere houghtto be compro-misedby herheavy accent,specificallybecauseher"problemswithspeechand grammarmadeit difficult for her to be understoodby studentsandteachers . . . plaintiff would do an excellentjob at a school whereher speech,grammarand story tellingwouldnot be so critical" Mandhare1985:238).The officialpublished ummary f the caseindicates hat Ms. Mandharethen met withtheSuperintendentf Schools,and on the adviceof hersuper-visor requested transfer o Thibodaux uniorHighSchool, as a librarian.25Theschool boardrefused o reappointMs. Mandhare o thisrequested ewposition;testimonyrevealed hat, in theirprivateandpublicdeliberations,Ms. Mandhare'soreignnessand accent werediscussed.The trialcourtwasveryfirm n thiscase:Ms. Mandhare ad beendiscrim-inatedagainst,andmust prevail.However, he school's nitialdecision hattheplaintiffcouldnotteachyoungchildren ecauseof her"heavy ccentandspeech patternsand grammarproblems which]preventedher fromeffec-tivelycommunicatingwithprimary chool students"ibid.)was neverques-tioned. The court took this claim on faith, andinsteadstated:

Defendant'scontention hat its legitimatereasonfor plaintiff's ermina-tion or non-appointmentwas that she had a communicationproblembecauseof her accentwhichpreventedher fromeffectivelycommunicat-180 Language in Society 23(2)


19/36


ing with primary school students is a feigned contention. Plaintiff was notbeing considered for a position which would requiresuch communication.She was to be appointed librarianat a Junior High School, a position forwhich it was established that she was eminently qualified.It is important to rememberthat in this case, as in every other case discussed,no effort was made to make an objective assessment of the communicationskills required for the job, the plaintiff's speech, the quality of her interac-tion with children, or her intelligibility. The administrators found the plain-tiff's accent difficult; they decidednot to reappointherto her job in the gradeschool. This alone would have made them the focus of the court's scrutiny(although not necessarily to the plaintiff's favor). However, they redeemedthemselves in the court's eyes: they praised the plaintiff's industry and skill,and they went out of their way to locate a position in a school where heraccent would neither offend nor inconvenience. The court could then focuson the school board, which refused to give the plaintiff this new job. Thevalidity of the initial firing was never challenged. Thus everyone (except theschool board) was happy: the administrators were left intact as arbiters ofthe SL ideology, and werelionized for their largesse;the court was not forcedto challenge those educators on the factual basis for their decisions aboutappropriate language; and Ms. Mandhare was to be reinstated as a librar-ian, in a junior high school.The question remains: Were Ms. Mandhare's civil rights protected? Wereher best interests really served? Put more controversially, if Ms. Mandharehad been forbidden to ride a public bus, and challenged that restriction,should she then have been pleased to be offered alternate transportation inthe form of a bicycle, a Mercedes-Benz - or another, different but equallyfunctioning, bus?Ms. Mandhare did not really want the transfer to another school in aschool district which had treated her so badly; she wanted back pay, whichshe did not get. Whether or not she would have been satisfied with the newposition was never established, because the trial court decision was reversedby the U.S. Court of Appeals for the Fifth Circuit:

The district court's determinationthat the Board had intentionally discrim-inated against Mandhare is clearly erroneous. The court focused on thewrong issue. It premised its conclusion on the Board's refusal to followLeBlanc's recommendation that Mandharebe transferred to a junior highlibrarian position. That was not the issue as framed by the unamendedpleadings and pre-trial order. Mandhare's action asserted discriminationin the Board's refusal to reemploy her as elementary school librarian, nottheir failure to create and transfer her to a junior high position. (Mand-hare 1986:5)



20/36

ROSINA LIPPI-GREEN

The terrible irony of this reversal should be clear: Ms. Mandhare was origi-nally protesting her dismissal on the basis of LTF-national origin discrimi-nation; the judge in that first case chose not to deal with the delicate issue,but to bypass it completely by focusing on the possibility of a position inanother school. This gave the appeal court an out, which it took. The appealcourt accused the trial court of focusing on the wrong issue; and on thatbasis, it reversed the decision.In the end, both courts were satisfied to let the school administratorsandschool board exclude on the basis of accent. In the analogy previously cited,the first court offered Ms. Mandhare a Mercedes-Benz when all she wantedto do was ride the bus. The appeal court said that the trial court had beenwrong to offer Ms. Mandhare a Mercedes-Benz that did not exist and thatno one was obliged to buy for her; it did not even question why she had beenforced off the bus in the first place, and it certainly did not offer her theopportunity to get back on, or compensate her for her trouble.The appeal court filed the reversal on May 2, 1986, six years after Ms.Mandhare was denied renewal. The failure of the American judicial systemcaused her untold emotional anguish and financial difficulty, and was det-rimental to her health. Today she works as librarian for a private school inher home town of Thibodaux, but she will carrythis experiencewith her forthe rest of her life.Broadcast-relatedasesThe Kahakua and the Staruch cases both have to do with the broadcastmedia, specifically with radio broadcasting. These cases are clearly very dif-ferent from the others presentedhere because they involve decontextualizedcommunication, in which heavierburdens are placed on the speaker.26Nev-ertheless, they provide interesting insight into the court's deliberations onmatters of language.Mr. Kahakua is a native of Hawaii, a bilingual speaker of Hawaiian Cre-ole and English; as a meteorologist with 20 years of experience and consid-erable educational background, he applied for a promotion so that he couldread weather reports on air. Mr. Staruch, a native of Western Ukraine,wanted to read news on the air, for the U.S. Bureau of Information, in hisnative tongue. This time, the plaintiff was penalized for speaking Ukrainianwith a stigmatized regional accent of that language. Both lost their cases.If the courts are deferential to academic institutions in matters of inter-nal administration and language use, they seem to be even more willing todefer to the standards of the broadcast media, even when those standardsinvolve blatant LTF discrimination. The arguments put forward by employ-ers in these cases and accepted by the courts involve the following elements:(a) Refusal to acknowledge accent as an immutable characteristic ofnational origin:182 Language in Society 23(2)


21/36


The court added, "there is no race or physiological reason why Kahakuacould not have used standard English pronunciations. " (Matsuda1991:1345)(b) Allowing direct and non-factual association of negative social valueswith stigmatized linguistic variants:... the agency contended that the appellant's accent was undesir-able . . . found to lack authority, friendliness, clarity and other qualitiesdesired in a broadcasted voice. (Staruch, EEOC Hearing Opinion)[The judge said] The white candidate was selected because he had "betterdiction, better enunciation, better pronunciation, better cadence, betterintonation, better voice clarity, and better understandability." (Matsuda1991:1345, citing from Kahakua).(c) Willingness to allow the media to set its own standards on the basis ofpersonal preferences, even when those preferences necessarily involve LTFdiscrimination:... the judge credited the testimony of speech expertsthat . . . "StandardEnglish should be used by radio broadcasters." (Kahakua)The agency stated that the appellant's voice was not suitable for broadcastpurposes . . . Appellant's voice was described as having a definite West-ern Ukrainian accent. In the United States national network news is broad-cast in "television accent" ratherthan the regionalaccents sometimes heardon local broadcasts . . . (Staruch)(d) Lack of concern with established facts about language structure anduse, or with consistent, non-prejudicial evaluation of language skills:[An external review found] . . . "not persuasive" his pronunciation as"often incorrect," delivery "dull" and "sounding strange to the listener."(Staruch, ibid.)I [expert witness "speech consultant"] urgently recommend he seek pro-fessional help in striving to lessen this handicap . . . Pidgin can be con-trolled. And if an individual is totally committed to improving,professional help on a long-term basis can produce results. (Kahakua 1989,Excerpts of the Record, 31, as cited by Matsuda 1991:1366, originalemphasis)

The Staruch decision has to do with the limited scope of Title VII: the EEOCcommissioners who heard the case accepted the argument that Ukrainianspeakers who had evaluated Staruch'sspeech did not like his regional accent.The courts clearly have bought the argument that, in broadcast media,LTF discrimination is nothing more than good business practice; i.e., main-Language in Society 23(2) 183


22/36

ROSINA LIPPI-GREEN

stream language use is a bona fide occupational qualification. Kahakua'sattorney, RichardHearn, has put this more succinctly:The employer did notwant Kahakua on the radio because Kahakua did not sound White ("AllThings Considered," September 12, 1990).Of course, the behavior of the courts follows logically if one accepts thepremise that media appropriately embody the SL ideology, so that theyshould be entrustedwith both the preservationand propagationof that stan-dard, and the exclusion and disempowermentof those who do not subscribeto it.ACCENT AND COMMUNICATIONEmployers presentto the courts a model of communication in the workplacewhich has three main points:

(a) Good communication skills are necessary for job X.(b) Accent Y impedes communication.(c) The applicant speaks with accent Y.(d) Conclusion: The applicantdoes not possess a basic skill necessary forjob X.A first criticism of this model must address the overly simplistic charac-terization of communication, in which the listener is relieved of any respon-sibility in the communicative act, and the full burden is put directly on thespeaker. Herbert Clark has developed a cognitive model of the communica-tive act (Clark& Wilkes-Gibbs 1986, Clark& Schaefer 1989) which is basedon a principle of mutual responsibility, in which participants in a conversa-tion collaborate in the establishmentof new information. This involves com-plicated processes of repair, expansion, and replacementin iterative fashionuntil both parties are satisfied: "Many purposes in conversation . . . changemoment by moment as the two people toleratemore or less uncertaintyabout

the listener's understandingof the speaker's references. The heavier burdenusually falls on the listener, since she is in the best position to assess her owncomprehension" (Clark & Wilkes-Gibbs 1986:34). This contrasts markedlywith the employer'sversion of communication, in which the speaker (the per-son with the accent) carries the majority of responsibility in the communi-cative act.The whole concept of units of conversation in which two partners worktoward mutual comprehension assumes a certain state of mind on the partof the participants.Work in accommodation theory suggests that a complexinterplay of linguistic and psychological factors will establish the predispo-sition to understand. Thus Thakeraret al. 1982 conducted a series of empir-ical tests to examine accommodation behavior. They were not workingdirectly with "accented" speech, but their findings are generally typical of184 Language in Society 23(2)


23/36

ACCENT AND DISCRIMINATORY PRETEXT IN THE COURTSsuch studies,whichverifysomethingknown ntuitively: istenersand speak-ers will work harderto find a communicativemiddle ground and fostermutual ntelligibilitywhen they are motivated,sociallyand psychologically,to do so. Conversely,when the speakerperceives hat the act of accommo-dating or assimilating inguisticallymay bring more disadvantages hanadvantages, n in-group erms, he or she may divergeeven farther rom thelanguageof the listener.27Robertset al. 1992 (RDJ) point to the largersocial context of languagecomprehensibilityn the workplace,and demonstrate hownativespeakers'assumption hatthey have the rightto dominateand control,and the waythat this is reinforcedby the worker's ack of abilityto negotiatethe rightto be heard,affect the detailedprocesses f routine nteractions nd theirout-comes" 1992:35).All this workpointsto two crucialconceptsnot includedin the employer'smodel of communicationn the workplace.First,Linguis-tic competenceon the part of the employee,taken alone, is insufficient orsuccessfulcommunication.Degreeof accentedness,whether rom LI inter-ference,or a sociallyorgeographicallymarkedanguagevariety,cannotpre-dict the level of an individual's communicative competence. In fact,communicativeompetence an oftenbe so highas to compensateorstrongLi interference. DJ providean excellent xampleof this, in whichan Asianfactoryworker riesto negotiatewith his supervisoro obtainworkfor hisson. The supervisors at firstunwilling o help, but the workernegotiatespastthe supervisor'seluctance. ntheircommentary n theexchange,RDJ(1992:40-1)point out that,in spiteof strong nterferenceromthe native an-guage, the workershows severalpositive qualities.

(a) He is sensitive o context, usingan appropriate iscourseconventionto set the scene.(b) He is focused, and able to keeprelevant opics"on stage."(c) He is able to compensate or and repaircommunicativedifficulties:"Forexample,when there s a confusionover 'first'and'fast', he reformu-lates . . . "

[Worker]: oy say I not workingon the fast.[Supervisor]: ot workingon the first?[Worker]:On the fast Ramadan you know.(d) He is in touch withculturaldifferences,and is able to negotiatethesupervisorout of "a gatekeeping ole."28A secondcrucialconcept s that theburdenof communication s shared,on every level, by bothparticipants. f one acceptsthat good communica-tion skillsarenecessaryorjob X, without urtherdefinitionof those skills,

one muststillquestion heemployer's laimthataccentYimpedescommu-nication.In fact, it is not necessarilyhe accent which is the problem,butnegative ubjective valuationon the partof the listener.It has been shown,Language in Society 23(2) 185


24/36

ROSINA LIPPI-GREEN

in cases such as Dercach, that lack of goodwill can be as much of an obsta-cle to understanding, if not more.Matsuda (1991:1369 ff.) has pointed out the fact that no consistent, dis-interested, fair proceduresexist to verify these claims, and that developmentof such a protocol is imperative. This would provide an objective way toestablish employment situations in which accent really is more likely to posea valid obstacle. Thus claims made by the employer about the effect of accenton job performancewould be subject to scrutinythat moves beyond the sub-jective and anecdotal. Of course, such measures are important preciselybecause accent, in the general sense that has been used here, can sometimesbe an impediment to communication, even when all parties involved in thecommunicative act are willing, and even eager, to understand.In Matsuda's scheme, the full communicative burden might be placed onthe speaker if (a) the consequences of miscommunication are grave; (b) thejob is primarily oral in nature; (c) the setting is stressful, and time is of theessence;or (d) interaction s contextless,and restricted o one-time exchanges.Of course, this list could, and probably must, be expanded and revised. Forexample, there seems to be no real reason to take together the conditions ofcontext and amount of contact; in fact, one can think of cases in which thecontext is indirect(over the telephone) but not limited to one-time exchanges(a dispatcherspeakingto the same truck driversmany times everyday). Thereare many communicative situations wherethe burdenis not distributedevenlybecausethe power and solidarityfactors between speakersinterfere (e.g. doc-tor/patient interactions); all these variables must be taken into account. Inaddition, the variables of stress and time need further definition andclarification.When all four of her conditions are met (as in the case of a 911 operator),Matsuda suggests that the speaker's accent should then be evaluated in anunbiased, consistent way to determine degree of intelligibility - possibly bymeans of matched-guise testing. This is thought to be one way to ascertainwhether or not the candidate is intelligible to the pool of relevant, nonprej-udiced listeners.Obviously, the constructionof an appropriatematched-guiseprotocol would be a challengingtask, and one that the courts are clearlynei-ther able nor willing to take on at present.In other cases where only one or two of these conditions are met, there isroom and opportunity for goodwill and accommodation, and it is reason-able to expect that the burden be distributedbetween speakerand audience.Here Matsuda draws heavily on legislation such as the Physical DisabilitiesAct, where reasonable accommodation is a major factor.Figure 1 presents a graphic view of a model in which an alternate config-urationmakes clearthe link betweenLTF discriminationand associationwithparticularethnic and national groups. Of course, relative positive and neg-ative evaluationof specificaccentswill vary accordingto age and background186 Language in Society 23(2)


25/36


Homeland

English as FirstLanguage English as Second LanguageSocial& FirstGeographical LanguageMarkers Interference

Speaker'sFilipino Accent FrenchAAVE BritishHispanic I DowneastAppalachian MidwestBroadcast

Listener'sSubjectiveReaction

NEGATIVE POSITIVEListener rejects ' Listeneracceptscommunicative communicativeole

1 0 CommunicativeflBurden VEntirelyon Shared betweenSpeaker Speaker andListener

FIGURE 1: Sharing or rejecting the communicative burden.Language in Society 23(2) 187


26/36

ROSINA LIPPI-GREEN

of the listener, as well as a wholerange of stylisticand discourse actors.However, there is empirical vidence or the generalsplit betweenwesternEuropean ccents,whichare generally eenas positive,and accentsperceivedas Slavic,Asian,or Hispanic.The fact that I was not able to document sin-gle case of a plaintiffwitha French,German, r BritishEnglishaccent s onekind of weak evidence,but there is more.Kalin et al. 1979 conductedan experimentn which studentswereaskedto play the roleof personnel onsultant,matching apedvoicesof applicantswith jobs characterizeds "high"and"low" tatus. The"applicants"pokewith a variety of ethnic accents. For the highest status job, the studentsranked he applicantsnthe followingorder:English,German,SouthAsian,WestIndian. This orderwas exactlyreversed or the lowest statusjob.Manyof these studentswillgo out into the workforce, andwillsomedaybecomeinvolved n the hiringprocess.Theywill continueto confuse theirvalid concern that employeesbe able to communicate ffectivelywith thepoliticaland social complexities f accent.Theywill firstjudgeindividualsnot on how logicallyor clearly heytalkaboutthemselves, heirgoals, andtheirabilities,but insteadon the rhythmsof theirspeech- rhythmswhichare linked o skincolor,economicresources, r homeland.Theywillexcludeand discriminaten the basisof languagebecause heyhave beentaught,byexample,that language s sufficientand appropriate ustificationfor thisbehavior.Theywillcontinue o hearwith anaccent: he accentof the intol-erant, empoweredmainstream.THE LINGUIST'S CONTRIBUTIONIf ideologyis mosteffectivewhen its workingsare leastvisible(Fairclough1989:85) hen the first step must be to make visiblethe link betweentheenforcement f SLideologyand social domination.Theeducational ystemis theobviouspointof departure,but thatsystem s itselfpartof the domi-nant bloc. Giventhe way schools, thebroadcast ndprintmedia, the enter-tainmentndustry,andemployersworktogether o promotean SLideology,theeducationof thepublic s both a lonelyanda difficult ask,butcertainlynot an impossibleone.Beyond education, linguists have hard-wonknowledgeto offer whichwouldbe of some assistancen the difficultquestions aced n matters f lan-guage policy.Thatknowledges often not sought;andif sought,it maybesummarily ejected;but in eithercase, it is often hotlyresented.Neverthe-less,therearegoodreasons o persevere, eyond he fact that the kindof lin-guisticdilettantism emonstrated ere s damagingo ourprofessional ride.Thistypeof behavior ausesrealharm o real ndividuals, nd it deserves urattention.188 Language in Society 23(2)


27/36


In the judicial ystem, heremaybe some essons orlinguists o learn rompsychologistsand psychiatrists,whose contributions o trial law are betterestablished,although heeffectivenessand value of those contributions reoften challenged Faust& Ziskin 1988).Although he overallqualityof psy-chologists' ontributionso legalcases s stillbeingdebated, omeissueshavebeen clarifiedas a resultof thatbodyof testimony.The law nowdefinesandtakes seriouslysuchhumanconditionsas batteredwomansyndrome,clini-cal depression,and post-traumatictresssyndrome.By contrast, although he courtshave called on linguists o address ech-nical mattersof authorship nd identification o be used as evidence,29heyare less interestedn a linguist'sdefinitionof communicativeompetenceorassessmentof intelligibility,as was seeninKahakuaandFragante,becausetheseare areas hey deemwithin heirown powersof reasoning ndexpertise.Xieng providesan interestingllustration f the statusof linguisticsn thecourts.Therewasno expert estimony t allon thepivotalmatter,whichwasthe employer's laim that Mr. Xieng's accent was too strongand impededcommunication.However,a psychiatristwas called, who then arguedandconvinced he courtthat theredid exista "causalrelationshipbetweenthe[employer's]ationalorigindiscriminationndXieng'ssevereemotionaldis-tress and depression" Xieng 1991:A13).

Psychologistsask themselvesa two-partquestionto determine he qual-ity of theirforensiccontributions:a) Can we answerquestionswith reason-able accuracy? b) Can we help the judge and jury reach a more accurateconclusion han wouldotherwisebe possible Faust& Ziskin1988:31)?Thatis, does the subject ie beyondthe knowledgeand experience f the averagelayperson?Can the expertgive informationwithout invading he provinceof the jury by expressinga conclusionas to the ultimate ssue?For most of the casespresentedhere,a list of questionscould have beenpresentedo linguistswhichwould havemet boththese basiccriteria.Ques-tionsabouttheprocessof standardization, ifferencesbetweenspokenandwritten anguagevarieties,culturaldifferences n discoursestyleand struc-turewhichmaycauseprocessing ifficulties, econd anguage cquisition ndaccent,subconscious ocial evaluationof activevariation,andchangeovertime andspacecouldbe answeredwith reasonableaccuracy.We couldpro-vide the judge and the jurywithinformationandknowledgebeyondthatofthe average ayperson.But the issue s this:wecannot make them wantthatinformation,no matterhow factuallycorrector howstronglysupportedbyempiricalevidence.Linguistic ontributionso the legal processare not valuedbecause deol-ogy intervenesn a waythatit does not in mattersof mentalhealth.Judgesmayhaveno personal nvestmentn acceptingevidence inking systematic,long-termphysicalabusewith violentbehavior; heyaremorelikelyto haveLanguage in Society 23(2) 189


28/36

ROSINA LIPPI-GREEN

a strongpersonalreactionwhenaskedto reconsider he assumptions nder-lying the standard anguage deology.Fairclough,whoacknowledgeshissomewhatdepressing tate of affairs,also pointsout (1989:4) hat "resistance nd changeare not only possible,but arecontinuouslyhappening.But the effectivenessof resistanceand therealizationof changedependon people developinga criticalconsciousnessof dominationand its modalities,rather han just experiencing hem."CONCLUSIONSThere are many people who must cope, day by day, with LTFdiscrimina-tion. Someof them have other currencies politicalandeconomicpower,social pre-eminence, rtisticexcellence,academicachievement whichtheycan use to offset the disadvantages f accent, and to disarm he prejudicedlistener.Most listeners,no matterhow overtlynegativeand hostile, wouldbe hardpressed o turnawayand ignoreAnnRichards,Jesse JacksonorEdKoch, CesarChavezor DerekWolcott,ButrosButrosGhalior LiuXiaobo,RigobertaMenchu,BenazirBhuttoor CorazonAquino,if the opportunityfor discussionpresentedtself.30Of course,mostpeoplewho do not speak helanguageof themainstreamdo not have suchextraordinaryesources.There aremanyof them: since1961,over 15millionpersonshave egally mmigratednto the UnitedStates(U.S. INS 1992:11).Manytimesthatnumberof citizens,bornin the U.S.,speakwith a regionalaccentthatis not fashionable,or are nativespeakersof a varietyof Englishwhich s directlyinked o race, ethnicity,or income.In a timewhenmulticulturalismnddiversity reheldupasideals,onemightthink that a standardanguage deologywouldgive wayto a more realisticandtolerantapproacho languageuse.Unfortunately,here s littleevidenceof this. LTFdiscriminations a widespread roblemwhichpermeatesmuchof ourday-to-day xistence. t is thesiteon whichracismand ethnocentrismareinstitutionalized.Someof the discussionaround anguage tandardss so emotional ntonethatparallelsanbedrawnodisagreementsetween cientists nd heologiansover the centuries.In our own time, in the courts, scienceand rationalinquiryhave comeup againstpublicopinionbasedon personalpreferencesand intuition:

. . . the realproblem acedis not legalbutsociological.In the centersofpopulationmen havegoneon assuming ertainbodiesof knowledgeandcertainpointsof view withoutrealizing hattheywerelivingin a differ-ent world from that inhabitedby a considerable ortionof theirfellow-citizens,andtheyhave beenunconsciousof the dangerwhichthreatened



29/36


them at the inevitablemomentwhen the two worldsshould comein con-flict. (The Nation, July 22, 1925:28,citedin Caudill1989:23)This editorialwas writtenat the heightof the Scopes trial, in which funda-mentalistsandempiricists rgued he verydefinitionof truth.It wasa trialsurroundedby sensational ournalism,and followed withgreatinterestbymany people.Scopes, a science teacherwho taught the theoryof evolutionin a statewhich forbadehim to do so, lost his case and was fined one hundreddol-lars. But something else, something perhaps more important, was won.Beforethe trial,onemightgather hat themajorityof American itizenshadnever come in contactwith evolutionary heory. After the trial, many ofthose peoplewere hinkingabouttheirown beliefs, about science,and aboutthe natureof authorityandits relationshipo knowledge.Whatever n indi-vidual'spersonalbeliefs, after the Scopes trial it became ncreasinglydiffi-cult for anyone to dismissout of hand the facts put forth by scientists.Today, morethan 70years ater,evolution s taught n allpublicschoolsandmost privateones.The Scopes trial involvedfree speech, educationalpolicy, and a rangeofsociological ssues.When he topic is discriminationn the basisof language,the stakes are very different. Mandhare,Hou, Xieng, Kahakua,and theother cases like them test an even more basic freedom, he individual's ightto be different:

Theway we talk, whether t is a life choice or an immutable haracteris-tic, is akin to other attributesof the self that the law protects.In privacylaw, due process aw, protectionagainstcrueland unusualpunishment,and freedomfrom inquisition,we say the statecannot intrudeupon thecoreof you, cannottake away yoursacredplacesof the self. A citizen'saccent,Iwouldargue,residesnone of thoseplaces. Matsuda 991:1391-2)It seems hatlinguistics ndLTFdiscrimination aveyetto meettheirScopestrial.

NOTES* I am thankful o the followingpersons or theirencouragement ndfor many nsightfulcomments n draftsof thisarticle: oe Salmons,PamelaMoss, Lesleyand JamesMilroy,JackieMacaulay,DeborahKeller-Cohen, nn RugglesGere,Bill Green,Raj K. Gupta,ArnethaBall,Dennis Baron,and RogerShuy.I also thank Raj Guptaof theEqualEmploymentOpportu-nity Commission or his helpfulnessn supplying riginal ourcematerial. n addition,I mustpoint out the importance f workby MariMatsuda1991andStephenCutler1985; hese arti-clesgaveme thestartI needed o explore he legalsideof this issue.I am grateful o all these

personsfor their help, but I retainsole responsibilityor the contentsof this article.1Courtcase citationsare abbreviated s follows:Mandhare v. W S. Lafargue ElementarySchool, the Lafourche Parish School Board, Parish of Lafourche appear in the text as Mand-



30/36

ROSINA LIPPI-GREEN

hare.Thismaterial riginatesromopinions,briefs,Findings f Fact, and other egaldocumentsassociatedwitheach case.Inthe case of Mandhare,nterviewswiththeplaintiffare alsocited.Completereferences re given at the end of this article.2 Later n this article explore n greaterdetail he reasoning f the courts,and their nter-pretationof matters egardinganguageuse andprescriptionn Mandhare.3 Manybasic concepts n sociolinguisticsavebeenchallenged or their ack of theoreticalcohesivenesswithina moregeneral heoryof sociology.Theconceptswhichhavecome underclosescrutinynclude uchgiantsassocialclass, tatus,prestige, ndgender.As such erms omeinto discussion, will outlinemy workingdefinitions. n the case of the concepts lassandsta-tus, I followsociologists ell& Newby 1971:218f.): classrefers o economic esources, hereasstatus s'reserved or the determinationf what s achievedwitheconomicresources.'Crucial here s the distinction etween pokenand writtenormsof language.Becausehewrittenwordwas developed nd existsto conveydecontextualizednformation ver time andspace, standardizations necessaryand appropriate.The problemat handhas come aboutbecauseof a blurring etween hewritten/spokenoundary;hewritten anguagehasacquireddominancen the mindsof speakers, o that goalsappropriateor the written anguage regen-eralized o speaking,and the writtenword is adoptedas a model for all language.5 The historyof standardizations a long andcomplexone. It has been treated xtensivelyelsewhere see, e.g., the excellent resentationf these ssues n Milroy&Milroy1985andBai-ley 1991).6 Spacedoesnotpermit longdiscussion f thedevelopmentf SL ideologyn theschools,whichwas clearlywell established t the beginning f the century. n 1911, J. ForbesRobert-son addressedheIndianaAssociationof Teachers f English:"There rethreecausesof thispoor English.Theyare gnorance, ffectation,and ndifference . . one of the most mportantpointsto remembern thecorrectarticulation ndpronunciationf words s to give the vow-els their correct ound" Robertson1911:5).7 One suchdictionarymightbetheOxfordEnglishDictionary,withthe followingprescrip-tive definition of accent: "the mode of utterancepeculiarto an individual, locality, ornation. . . This utterance onsistsmainly n a prevailing ualityof tone,or in a peculiar lter-ationof pitch,but may ncludemispronunciationf vowelsorconsonants,misplacingf stress,and misinflection f a sentence.Thelocalityof a speakers generallymarkedby this kindofaccent."I Prestige s a particularly ifficultconcept n sociolinguistics. . Milroy 1989hasarguedthat it is nothingmore han a veiledappeal o socio-economic lassstructure;n fact, in mostsuch discussions, he higher ocio-economic roupsareassumed o havethe mostprestige.9On May 17, 1978,JudgeCharlesJoinerhandeddown a decision, nMartinLutherKingElementary choolChildren . TheMichiganBoardof EducationandtheAnnArborSchoolDistrictBoard,whichdirectedhe SchoolBoard o train eachers n the basisof "existing nowl-edge"regardinganguageuse andvariation;he existenceand structure f BlackEnglishVer-nacular; nd the necessary kills o teach he plaintiffs,who werenativespeakers f BEV,howto read(see Chambers 983for detaileddiscussions f the Kingcase).? In entertainment,inguistic tereotypes ave ongbeen a stock-in-trade. ialectwasusedto drawcharactern Chaucer, nd can be followed o the present ime. Inbroadcast nd filmentertainment,he use of linguistic tereotypesmirrorshe evolution f national earsandobses-sions:Japanese nd German haractersn Disneycartoonsduring he SecondWorldWar,Rus-sianspycharactersnchildren'sartoons nthe 1950sand 1960s,Arabcharactersn theeraofhostilitieswithIranandIraq.Moregeneral tereotypings alsoprevalentntelevisionprogram-ming and movies:situationcomedies BeverlyHillbillies, Love Lucy,Sanfordand Son, Allin theFamily)andanimatedilms JungleBook,Dumbo)provide umerous xamples.The 1993filmFallingDownprovides disturbingxample. nthat film,a middle-class orker,portrayedas beleaguered y inner-cityife, loseshistemperwithanirascible onveniencetoreclerk.Theepisodebeginswhen heprotagonist,D-Fens playedbyMichaelDouglas),asks thepriceof anitem. The following s from the script:

Theproprietor, middle-aged sian,readsa Koreannewspaper . . The Asianhas a heavyaccent . . .



31/36


Asian: Eighdyfie sen.D-Fens: What?Asian: Eighdyfie sen.D-Fens: I can'tunderstand ou ... I'mnot payingeight-five entsfor a stinking oda.I'll give you a quarter. You give me seventy "fie" cents back for thephone . . . What s a fie? There'sa "V" n the word. Fie-vuh.Don'ttheyhave"v's" n China?Asian: Not Chinese,I am Korean.D-Fens: Whatever.What differencedoes that make?Youcomeover here and take mymoney and you don'teven havethe graceto learnto speak my language . .Here,the clerk'saccent and the Korean lerk are portrayed s negative lements f urbanlife." For a more lucid discussionof the issuesraised n this column, see Raspberry 990.12Headlines lone areoften revealing: BlackEnglishs silly" ChicagoSun-Times, uly 10,1979);"Hushmahmouth! Some in Southtry to lose the drawl; accentreduction'becomesabig bidness n Atlanta;searchin'or the lost 'G"' (WallStreetJournal,December13, 1991);"Proper nglish,Yes;but Educationalists,o"(New YorkTimes,September 8, 1989);"TwangyGuyNextDoor Ousts he Professionals"Marketing,anuary 3, 1992);"Lose hatThickAccentto GainCareerGround"WallStreetJournal,January , 1990);"Mostofficials don't talkli'dat thesedays" HonoluluAdvertiser,September 9, 1987,citedin Sato 1991).13Title VII is specific o employmentssues; he legislationand court cases here cannotbeapplied o any other arena, e.g. education.14 Companies mploying ess than 15 workersare not bound by these statutes.1 Discriminations a matterof law:the effect of a statuteor established racticewhichconfersparticular rivileges n a classarbitrarily elected roma largenumberof persons,all of whom stand n the samerela-tion to the privileges ranted nd betweenwhomand thosenot favoredno reasonable is-tinction can be found. (Black's Law Dictionary, 1991:323)

16Thediscussionhere excludes he very crucial"EnglishOnly"controversy, nd the moregeneralssues urroundingilingualism. hose opicshavebeencoveredngreatdepthbyCraw-ford 1992and Baron 1990.17Under ?703(e)of Title VII, an employermay defend his or her actions on the basis ofnationalorigin a)"bydemonstratinghe 'business ecessity' f thedisputed mployment rac-tice- i.e., by showinghepracticeto benecessaryo safeandefficientob performance'Cutler1985:1168,n. 20);or (b) by establishing bonafideoccupational ualificationBFOQ) ibid.).TheBFOQ s the moredifficult ase for theemployer.Thepath akendepends n whichof twodifferent heories f liabilitys used:disparatereatment,nwhichproofof discriminatoryntentis crucial,requires BFOQdefense; or disparatempact, n whichsuchproofis notrequired,theemployermustestablish nlybusiness ecessity: ThePlaintiffmakesouta prima aciecaseby showing hat theemployer'selectiondevicehasa substantially dversempacton hispro-tected group . . . it remainsopen to the Plaintiff to show that 'other . .. selection devices, with-out a similarlyundesirable . . effect,wouldalso serve heemployer'segitimate nterest[s]'"(ibid., 1169).18Of course heproblems f associating pecificregional r socialdialectswithspecific or-eign originswouldbe tremendous.oe Salmons p.c.)hasbroughto myattentionworkbyDil-lard 1992 which outlines he considerable ifficultiesof evenidentifyingany salientfeaturesspecificto AppalachianEnglish but see also Wolfram& Christian1976and Christian1988,whichprovideevidence hat these difficultiescan be overcome).9 TheEEOCreviews omplaints;f theyfinda violationhas takenplace,they maytake onthecase,and file suitfor theemployee gainst heemployer.RajGuptaof theEEOCestimatesthat the EEOCprosecutes 0% of suchcases; n the other30%,they mayor maynot granta

Notice of Right o Sue.Lackof suchNoticedoesnotprohibit heemployee romproceeding;theright o pursue uch mattersn thecourts s sacrosanct.Thusthe Noticeof Right o Sueisprimarily n indication o the employeeof the strengthof the case.Foremployeesof federal



32/36

ROSINA LIPPI-GREEN

government gencies, he EEOCconducts he hearing,which s empowered y TitleVII to heardiscrimination ases; f theyfindfor the plaintiff,they can orderremedies.The federalagen-cies can appealonly to the EEOC.20 Trackingdown anddocumentinghesecases was a matterof manyhours n the Univer-sity of MichiganLawLibrary.Certainly, ases have been excludedby oversight: herearenosummary tatistics eptby the EEOC,and no central ogging ystem or thesecases.Manycasesare not summarizedor publication.Thus no guarantee an be madeof thoroughness f rep-resentation.The search or cases included n this articlewasconcluded n May 1993.21 This GAO studywas conducted n response o a seriesof inquiries rom Congress n theeffect of the 1986 mmigrationaws.Not all the GAO's indingswereclearor interpretable,spe-cially in the matterof specifically ccent-based iscrimination. hereportoutlinesa numberof reasons or this, having o do with samplingand designquestions.22Matsuda1991providesa thoroughoverviewof the Fragante nd Kahakua ases.23Dr. JacquelineMacaulay, n attorneywith a Ph.D. in socialpsychology,dealswith fam-ily, employment, nd civil rightscases; she has pointedout to me (p.c.) that the courtsseemto be functioning n the basis of some "phantomegislature" hichhas mandated hat a cer-tain form of English s "Standard"nd "unaccented."24 It seems hat threedistinctkindsof expertwitnesses estify n these trials: inguists e.g.,Charlene atoof theUniversityf HawaiiestifiednKahakua), peechpathologists,nd"speechconsultants." his ast class s the most troublesome ne, composed f thosewhoteach"accentreduction" lasses,or otherwise avea vested nterest n the officialcommendation f a "stan-dardEnglish." ome udges,especiallyhe judgewho heardKahakua, reveryreceptiveo argu-mentsof this kind.2S Ms.Mandhareellsa verydifferent tory.Ina phone nterview, he ndicatedhather firstyearat theK-2 school was alsothe principal'sirst year, and that he openlyadmittedhe hadpromisedherjob as librarian o someoneelse. He askedher to requesta transfer,whichshedid not wishto do. Afterthis episode,hetoldher n a one-on-onemeeting hatshe hada "veryheavyaccent,"althought hadneverbeenmadean issuepreviously, nd she had hadno com-plaintsfromchildrenor teachers.26 "Inmanycircumstances, s in literary orms, ectures,and radiobroadcasts,writers ndspeakers re distant romtheiraddresseesn place, time, or both. They mightbe assumed oadhere o a weakenedversionof mutualresponsibility . . speakers till monitorwhattheysay . . . It is justthattheydo all thiswithout eedback rom isteners"Clark&Wilkes-Gibbs1986:35-6).The radiobroadcasters,f course,are readingrompreparedexts,and so thedis-tributionof communicative urdendoesnot apply n the wayit does in otherdiscourse.27 This has been statedmore simply and admittedly n an anecdotalway) by personswhodailydependon accommodation. oy Cherian,Commissionerf theEEOC,has commented:"Imyselfspeakwitha foreignaccent.My colleagues ometimes ave o listen o memorecare-fully simply o fully understandwhat I amsaying.Perhaps hat makes or bettercommunica-tion betweenus" (EEOCPR, June 8, 1988).28 Fairclough 1989:47)definesa gatekeepingncounter s follows:encounters uch as a job interviewn whicha "gatekeeper"hogenerallybelongs o thesocietally ominant ultural rouping ontrols nencounter hichdetermines hetherome-onegetsa job, orgetsaccess o someothervaluedobjective. ncontemporary ritain, orexample as nthepreceding assage],t is mainlywhitemiddle-classeoplewhoactasgate-keepersn encounters ithmembers f the various thnic andcultural)minoritiesf Asian,WestIndian,African, etc., origin.29Miron 1990andShuy1993provideoverviews f this type of "forensicinguistics"n thecourtroom;ee alsoLevi1994 ora comprehensiveuide o publicationsnlanguage ndthe aw.30 Ann Richards governor nd nativeof Texas),JesseJackson African-Americaneligiousandpolitical eader),Ed Koch formermayorandnativeof NewYork),CesarChavezMexican-Americanctivist or farmworker ights),DerekWolcottWest ndianpoet,awarded 992Nobel

Prize for Literature), utrosButrosGhali Egyptian ecretary-generalf theUnitedNations),Liu Xiaobo(Chinese tudentactivistanddissident, ailedafterTiananmen),RigobertaMen-chu(GuatemalanMayan ndian,awarded1992Nobel PeacePrize),BenazirBhutto Pakistaniprimeminister),CorazonAquino (president f the Philippines).194 Language in Society 23(2)


33/36


REFERENCESCOURT CASES CITED WITH SOURCES

Note: LEXIS is a computerized legal document search and retrieval service.Andrews 1992. George W. Andrews v. Cartex Corporation. Civil Action No. 91-7109. 1992 U.S.District Court. Source: LEXIS 11468.Ang 1991. Ignatius G. Ang v. The Proctor & Gamble Company. Source: Federal Reporter (2d)932:540; LEXIS 8993; Fair Employment Practices Cases (BNA) 55:1666; Employment Prac-tices Decisions (CCH) 56:40732.Bell 1984. Bell v. Home Life Insurance Company. Source: Federal Supplement 596:1549.Berke 1980. Rozalia Berke v. Ohio Department of Public Welfare. Source: Federal Reporter(2d) 628:980-81.Carino 1981. Donaciano Carino v. Regents of the University of Oklahoma. Source: FederalReporter (2d) 750:815; Fair Employment Practices Cases (BNA) 25:1332.Carroll 1989. Doritt Carroll v. Elliott Personnel Services. Source: Fair Employment PracticesCases (BNA) 51:1173; Employment Practice Decisions (CCH) 52:39508.Casas 1983. Casas v. First American Bank. Source: Fair Employment Practices Cases (BNA)31:1479.Dabor 1991. E.G. Dabor v. Dayton Power & Light Company. Source: LEXIS 2402.Dercach 1987. Anthony D

Documents

Lippi-Green-Accent, Standard Language and