Pragmatics2:3.427 -443 InternationalPragmatics Association


Bambi B. Schiefferin& Racheile charrier Doucet


Languagecan be examined as a rich resource for understanding the ways in which speakersrepresent themselves, how they represent others, and how they are represented by others. In this paper we explore a set of languagebeliefs in conjunction with practices of Kreydl speakeis. We are interested in how metalinguistic telms used by regarding varieties of spoken Kreydl manifest themselvesin debatesregarding which best repreients the language.This is followed by ananalysis of competing in teims of how tt rnit" the languagelooi andwhich soundsare.given "y graphi- representation.We view the pro.., of Jreiting an orthographyfor Kreydl not is u n"utril activity which simply reducesan oral language to witten form, but as an important symbolicvehicle for'representing its speak"ersIn termsof national and international identity. We propose thai conteste-dorthographies beviewed as sites of contestedidentities iather ittun ur neutral academicor fnguistic argumentswithout political, socialor educationalconsequences. We suggestthat the debatesregarding the soundsof Kreydl as well as how those soundsshould be written areabout different representations of its speakers.These different arguments and the

1 we refer to the language in the title of the in the form that is most familiar to an Englishspeaking audience, . However,in the bodyof the text we refer to the language as Kreydl, the way it is referred to by Haitians, and we have chosen to write it and our own transcriptionsusing the official orthography(lPN). we have, of course,maintained the original orthographiesused in publishedsources. Thiscollaborative project grew out of an investigation-the of Kreydllanguage use in City, whichwas then extendedto Port-au-Prince. During course of transcribingaudiotaped family interactiondata with native speakers, the issueof orihographicconventions arose frequently. This in mnjunctionwith nativespeakers'metalinguistic commeniariesabout the participants'speechled us to investigatethe relationshipsbenveen varietiesof speakingand orthograptry.nes'earch methods include theuse of historical,sociolinguistic and ethnographicdata. we would like to thank the manyHaitians whoparticipated in our rproject and freely snireo rheir viewsabout Kreydl and other topicswith us; thespencer Foundation and (Research challenge Fund) for supporting this research;and Michel Degraff,Paul Garrettand John Singler who provideotretpruf comments on earlier versionsof this paper. 428 Bambi B. Schiffilin and Rachelle Charlier Doucet languageideology that underliesthem have historicalsocio-political roots which are played out in familial, educational,and political contexts.Examining the complex and often paradoxicalvalues associated with thesedebates is critical to understandingthe ways in which Haitians evaluate each other and think about themselves. Kreydl, like other creole ,still paysthe price of its origins. By-products of European colonizationof the New World, creolelanguages developed on plantations from the forced contact betweenEuropean mastersand their African slaves.According to most creolists,the basic grammaticalstructure of Kreydl was contributed by the dominated West African slaves,while the lexicalbase came largely from the dominating French colonizers. Like many other creole languages,Kreydl continues to exist in a complex political and social relationshipto a standardlanguage, in this caseFrench, which since 1918 has been the official languageof education and government in . Formerly, sociolinguistsdescribed the relationshipbetween and Kreydl as diglossic:both were consideredvarieties of the samelanguage, used by speakersin different social contextsfor different functions.Haitian French was viewed as the high prestige form and Kreydl as the low prestige form. However, Haiti is better described as a nation predominantly composed of trvo linguistic communities - the minority Kreydl/French bilingual elite (7Vo)and the monolingualKreybl urban and rural masses with varying degreesof linguistic interaction betweenthe two. There are approximately 6 million Kreydl speakersin Haiti and its diasporacommunities located in , , and , and elsewhere.

2. Kreybl speechvarieties and metalinguistics

The metalinguisticterms of a speechcommunities can serve as a starting point for investigatingattitudes toward languagevarieties and the speakersassociated with them. Haitians maintain complicatedattitudes toward both Frenchand Kreydl, many of which become apparent when examiningthe metalinguisticterms used both popularly and 'smooth scientificallyto refer to varieties of spoken Kreydl. Kreydl sw4 Kreydl', Kreydl 'rough 'vulgar rdk Kreydl', Kreydl fransize'FrenchifiedKreydl', and gwo Kreydl Kreydl' are metalinguisticterms central to the debatesabout orthography.2 Kreydl fransize is a term used to refer to the variety of Kreydl spoken by educated urban bilinguals (Vernet 1989:20). They contrast their Krqdl fransize with gwo lOeydl'rough, vulgar Kreydl', which they sayis spokenby uneducatedurban people and peasants.Both terms refer not only to the ,vocabulary and intonational contours of the speechitself, but also to the nonverbalgestures used by speakersof 'good each variety. Uneducated speakers,however, prefer the term bon Kreydl Kreydl' to refer to the variety that they themselvesuse, though they know the other terms. Depending on their own language socializationexperiences and social networks, educated speakers can recognize Kreydl rdk, but are not always able to produce it.

2 SeeLudwig 1989and Schieffelin and Doucet ms for a comparisonof similarmetalinguistic termsin Haiti, ,Cuadaloupe, Mauritius and the Seychelles. Metalinguisticsand orthographicchoice 429

Uneducatedspeakers can recognizeKrqtdl but not speak it spontaneously: fransize, 'smooth Fattier-Thomasdistinguishes between Krqtdl.rwA Kreydl', the sociolect of the bilingual educatedminority, and Krqdl rdk'rough Kreydl', the variety spoken by the monolingual masses(1984: 39). Lofficial also uses the term l*qdl rlk, which he describesas"le niveau ressenticomme le plus rude de la langue" (1979:118) [the level felt to be the roughest of the language].While these metalinguistic terms are mentionedin the sociolinguisticliterature on Kreydl (Valdman 1988,1989a, b, 199L), andare commonly known by Haitians in a variety of social classes,the words rdk, swa, andgwo are not listed in the various Kreydl dictionaries in their metalinguistic usage. It is informativefor later discussionsof the orthography to elaborate on the of theseexpressions because they carry important cultural meanings. 'smooth' Swc (Fr. soie) as a singlelexical item is primarily usedas an adjectival 'hair' modifierof cheve and refers to fine, straight hair. The opposite of.cheve swa in thisreferential sense is chevegrenn'tightly curled, kinky hair' (Fr. crdpu).Men are also saidto be swa'smooth'if they are well-mannered,educated, emotionally even and are pleasingto women (ndgsa a swo'that guy is smooth'),a positive description.^Spa as anadverbial modifier refers to the smoothmanner in which someaction is carried out. For example,one saysof a car which runs smoothly and is problem free, machin sa a macheswa.Yet another example,paleswa'speak smoothly and persuasively',describes speechwhich is flowing, regardlessof the social or educational background of the speaker. Krqdl rdk is not strictly opposedto lheydl swa, though thesetwo terms are often contrasted(Fattier-Thomas 1984; Valdman 1989a,b, 1991).Outside of its referenceto speech,the primary meaning of the word r?k is agricultural and refers to fruit or vegetableswhich are not yet ripe but ready to pick. When referring to persons,rdk is oftenapplied to a child in the senseof appearingolder than he or she actually is. Additionally,r?k refers to strength,particularly in men. Unlike swa,which can modiS 'speak a rangeof actions,rdk is only used to describea manner of speaking:pale rCk 'speak roughly',is the opposite of pale swa smoothly'. Pale rdk also indexes a verbal styleof directness,in contrastto the more indirect style of pale swa. Yet another way to refer to stylesof indirection is the Kreydl expressionpale franse'speakingFrench'. The culturalmeaning of this expressionis somewhatdifferent frompale swa.Pale franse emphasizesspeaking indirectly (in either French or Kreydl), with the implication of tricking someone through the use of beautiful sounds, obfuscatedspeech, or irrelevant arguments.The result may be confusion for the listener.Depending on the speakersand the context,what soundsgood Qtaleswa or palefranse) can be viewed as potentially deceptive. The differencesbetween Kreydl swa and Kreydlrek that are most relevant in the orthographicdebates lie in the degreesof similaritiesto Frenchphonetic characteristics (Fattier-Thomas1984: 40-41). Kreydl swa can be most generally characterizedas retainingthe front-rounded lol, lcel, /ii/ where Kreydl r?k would use the non- roundedvowels lel and /i/. Another marker of.Kreydl svvais the use of postvocalic lrl 430 Bambi B. Schiefetin and Rachelle Charlier Doucet

at the ends of words, absent in the rCk variety.3Also relevant is the prevocalic Haitian lrl and its possiblerealization as a more labialized l'xl in three main contexts:lol, ldl, ltl. Since 1986,with the liberation of the press and increasedaccess of the masses to the media, there has been a shift in values associatedwith these speech varieties. Monolinguals try to imitate the prestigiousform, Krqdl swa. Fattier-Thomas (1987) remarks that some bilinguals, in particular politicians,try to modiff their usual Krqdl swa to sound more like the popular variety Kreydl rdk in an attempt to identiff with the masses. For both categories of speakers, this is met with little success.Whereas previously Krqdl swa and Krqdl fransize were both considered desirable modes of speaking,Kreydl r?k and gwo lheydl are now given value in some progressivecircles where claims of authenticity and rootedness in cultural identity are asserted.As this is happening, the term bon Kreydl is becoming an alternative to l?eydl rdk and gwo Krqdl, becausethe latter two have long standingnegative connotations. Bon l{rqdl is now the term used in Haiti to refer to the genuine Kreydl spoken by the masses. While these metalinguisticnotions are worth consideringon their own, it is useful to investigate their pragmatic implications and links to other symbolic systems. Many Haitians and non-Haitians view Kreydl as a simplified form of French, at best. Others claim it is not a real language at all, but a mixture of languageswithout a g"ammar. It has been characterized as lacking in abstract terms, inadequate for scientific, philosophical or logical thought. For most of its history Kreybl has not been considered adequate for government, schoolingor Western religious services.These ideas about Kreydl have not only had important repercussionsfor the ways in which monolingual Kreydl speakers have been viewed, but have also had important c()nsequencesfor the codification and functions of written Kreydl. As we will see, the question of the nature and meaning of theselanguage varieties in Haiti and who speaks them is at the core of the orthography debates that have been going on in Haiti for more than sixty years,and which, despitethe claims of many linguiststhat the question is closed,provokes reactions among linguistsand educators.The debate on the use of Kreydl itself, and particularly on the use of Kreydl in schools,is broader and is taken up by parents and teachersin addition to writers and media people. The issuesof orthographyand use,we will argue,are connectedthrough broader ideologicaltenets.

3. Writing Kreybl

The first accounts of written Kreydl date back to the late eighteenth century, when Kreybl was used for both official and literary purposes.One of the earliest records of written Kreydl was issuedby a French delegatefrom Bonapart's regime, Sonthonax.It is an abstract (from the French) of the proclamation to the slavesabolishing slavery in 7794.This, and other documentsat the time, were read to the nonliteratepopulation

5 The exceptionis in the Northern dialect,which has this postvocalicfeature throughout (Hyppolite 1949;Orjala 1970). Metalinguistics and orthographic choice 437

anddid not circulateas written documents.The orthographyused in this documentand othersresembles a simplified French spellingin use at the time and is best viewed as inventedad hoc as it had no basis in any systematicconventions for representing the soundsof Kreydl. This approach, representing Kreybl as a diminished or simplified versionof French, persistedwell into the 1930sand reflected many ideas about the natureof Kreydl and its speakers.a Orthographic debates began in Haiti in the late 1920sand have involved many differentpositions (D6jean 1980a;Pressoir 1947).As of 1980eleven proposed spelling systemscould be identified. Nonetheless,despite apparent diversity, three main kinds of proposalscan be identified:(1) Thosewho take a pro-etymologicalor anti-phonemic view;(2) those who support a pro-phonemic approach; (3) those in an intermediary camp, proposing a but with some concessionsto French spelling.s The first widely recognized systemwas proposed in 1940 by a Protestant missionary,Ormonde McConnell,who had done work in adult literacy in rural Haiti. It used33 symbols and was phonemic (Ddjean 1980a: 19-20): oral vowels: aeei o du nasal vowels: aed semi-vowels: ily w u (in the dipthong ui) : bdfghjklmnprstvz digraphs: sh gn ln 1943McConnell was joined by an American literacy expert, Frank l.aubach, who was not familiar with Kreydl, and together they revised McConnell's original spellingsystem. The secondversion, called McConnell-l^aubach,was also a phonemic orthography.It used the same alphabet as McConnell's with the following differences: the ou instead of u to represent the sound lul as in Kreydl dou (Fr. dow); and ch insteadof sh to represent the sound /5/ as in Kreydl chante (Fr. chanter). This spelling systemwas used in the literacy campaignsponsored by the government of Elie Irscot. Thematerials, almost all inspired from Protestantreligious texts, were designedto bring salvationand light to the Haitian rural masses. The Haitian scholarPressoir (1947) strongly criticized the McConnell-l-aubach orthographymainly because of the absenceof the front-roundedvowels lrj,l,lol, lel, lcr-l; the broad use of the "Anglo-Saxon"letters w and y; and the use of the circumflex to mark the nasalizedvowels. As French uses the letter n to indicate nasalization,a Frenchdiacritic was used but with a different application. This systemwas called by someKreydl bwa-nan-nen'Kreydl with a wooden stick on its nose' becauseof the abundanceof circumflexes.The term bwa-nan-nen refers \ I t0 puttinga small wooden pincer, like a clothespin, on the face or ear of someonewho

' \ a Hoffman (1939)reviews the developmentof literaryworls in Haiti.

5 Debatesregarding orthography in Mauritiusand Martiniqueshow similar divisionsbetween thoseadvocating a phonemicbasis and thosearguing for an etymologicalone. 432 BambiB. Schiffilin andRachelle Charlier Doucet

losesat each round of dominoesor cards and is still done today in urban as well as rural areas of Haiti. Kreydl bwa-nan-nenref.ers both to the way the words look in their written form and to how they were to be pronouncedwhen read. Pompilus (1973:25) suggestedthat the appellation bwa-nan-nenwas suggestedby the literacy instructors to help their studentsremember that the circumflexrepresented nasalization. Pressoir claimed (1947) that this new orthography was good for "savages"who spoke a gros crdole'rough creole'. One characteristicgenerally admitted of rural Kreydl is extended nasalization(Lofficial 1979:118). Although systematic,the McConnell-l,aubach orthography was contested by many educated Haitians because of its associationwith and its "American" look - a sensitivepolitical issue,since Americans had occupiedHaiti from 1915-7934(Pressoir 1947).Those Haitians did not want any reminders of this American presence, not even in the . They were not happy about the novel representationof nasalizedvowels through the use of the circumflex,which treated these nasal sounds differently from the ways in which they are represented in French. This made the representationof Kreydl appear strangeand foreign,and in particular, far from French. Arguments were made that this orthography would inhibit learning French, a goal in educationalcircles. Pressoir,who was the leader of the opposition againstMcConnell-Laubach's systembut was himself in favor of a pro-phonemicorthography, introduced a number of changes.He eliminated the circumflexto indicate nasalizationand introduced the letter n in its place. In spite of the fact that the "Anglo-Saxon"letters k, y, and w looked "too American", he nonethelessretained the letter k instead of choosingthe letter c (as in French) to representthe soundM, and also kept the letter y together with the letter i to represent the semi-consonantljl (D6jean 1980a: 185), as in for example, ay)'yesterday'.Instead of the letter w, however,he used the digraph ou to representboth the vocalic sound lul and the semivocalicsound r'd, thus adopting the French orthographicconvention for representingthose sounds. Pressoir also introduced the hyphen to distinguishbetween the nasalizedvowels as in lpdl and (non-nasalized) vowelswhich are followed by the nasalconsonant lnl, as for example inpa-n (Fr. panne 'mechanical breakdown').Pressoir's system used the following alphabetof 30 symbols (D6jean 1980a:183): oral vowels: a e e I o ou digraph: ou nasalvowels: an in on semi-vowels: ily ou u (in the dipthong ui) consonants: b d f gjklmnprstvz digraphs:chgn Pressoir'sorthography was adoptedwith some modifications.It has been used primarily by governmentagencies for more than 30 yearsin adult literacyprograms and wasconsidered quite satisfactory(D6jean 1980a: 182; F6rdre t977:59; Pompilus1985: 163). American missionaries, however, used a modified McConnell-Laubach orthographyfor Bible translationand instruction.In literarycircles, independent writers wishing to write in Kreydl createdtheir own orthographicsystems, more or lessclose to French etymologyand orthography,thus contestingthe Pressoirsystem. There was Metalinguisticsand orthographicchoice 433 n0 consensusabout orthography. In the 1970sthere was increasingsocial and political pressureto changethe socialorder of the nation and changethe languageof instructionand literacyin schools fromFrench to Kreydl. Once Kreydl was officially introduced by law in the schoolsin t979,it was vital that the Haitian government unify Kreydl orthography. To do so, it establishedas the official orthography a systemdeveloped by Haitian linguistsfrom the InstitutP6dagogique National (IPN) in collaboration with French linguists from the Universit6Ren6 Descartes.The official orthography has an alphabet of 32 symbols: oralvowels: a e e i o d ou nasalizedvowels an en on oun semi-vowels: w y u (in the dipthong ui) consonants: b, ch, d, f, g, h, j, k, l, m, n, flg, p, r, s, t, v, z In this orthographythe soundlEl is representedby en. It retainsonly one accent mark,the grave for ldl and ld,l.The letter y is used to represent the sound /j/. For the back-roundedvowel lul the representationis ou. The use of hyphensand apostrophe areoptional, as is the representationof the prevocaliclwl as either the letter w or the letter r before the vowels lol, lol, 16l,lul (for example, gwo or gro). In 1980 this orthography,called dtograf /PN or dtograf oftsyel or more simply dtograf Krqdl, was madeofficial on a trial basisby a communiqu6of the Ministry of Educationon January 30,1980. After four years of experimentation,the governmentwas supposedto take a definitivedecision. The four years have long since passed,but no official document asyet has either endorsedor discreditedthe official orthography.The systemis still contestedby some linguistsand provokespassionate reactions among the public for reasonsto be discussedbelow. One of the major goals of the McConnell-l-aubach orthography was that it be easyto learn and use by the monolingualmasses. For Pressoirthe major ideological concernwas that the orthography not look American and function as a bridge to literacyin French. When the dtograf Krqdl was introduced in 1979, the linguistic ideologyregarding Kreydl itself had changedsignificantly, and this affected the attitudes peopleheld toward the orthographyas a representationalsystem. Unlike the periods of adultliteracy campaigns where Kreydl was treatedsimply as a transitionaltool to be usedfor the achievementof other goals,by 1979Kreydl was thought about not only in thecontext of literacy,but more generallyas the languageof instructionas well as an objectof instruction.The imageof Kreydl, both spokenand written, and consequently itsrole in the social and political life of Haitians,had changed.

4. Interpretation of the debates

Whyis the issueof the Kreydl orthographyso important to Haitians?The matter is not simplywhether to write, for example, the Kreydl word for bread lEl with e Qten) or i (pin)or ain @ain) like in French "pain".The underlyingissue is about representations of selfand representationsof the nation. 434 BambiB. Schieffelinand Rachelle Charlier Doucet

Pro-etymologistsargue that Kreydl must stayas closeas possibleto the French orthographic system.The reasoningthat underliesthis position is that French and Kreydl will always coexist in Haiti and thus it is logical to facilitate the learning of French through Kreydl. This idea is usuallyexpressed by the imageof an "orthographic bridge" which will help avoid the confusionwhich might result for Kreydl monolinguals as well as the bilinguals from the use of two totally different systemsfor languages which, they think, are quite similar. Kreydl orthography, they assert,must also reflect Kreydl's origin - the .Those who aligned themselveswith the pro- etymologistposition include mainly literary people and membersof the intelligentsia, primary and secondaryschool teachers, and many parentsfrom the middle classesas well as the lower classeswho were concernedabout the introduction of Kreydl in schools. Pro-phonemicistsinsist that Kreydl must be written in a coherent,systematic, and logical way. A Kreydl orthography must be easy to learn as well as completely independent from the French orthographic system. Since the target groups are monolingual Kreydl speakerswho do not read French, theoretically they can learn any orthography, and the most straightforwardKreydl systemis the easiestfor the not-yet- literate monolingualsbecause the only relevant facts are Kreydl facts.The orthography should represent the soundsof their Kreybl. The pro-phonemicistsare comprisedof those who adopted and used, even if with slight variations, Pressoir's orthographic system and, since 1980, the dtografKreydl. Most who adopt this approach are people involvedin adult literacyand/or children'sliteracy programs in Haiti and in the .Representatives of this camp are the Haitian government;the Catholic and Protestant Churches; linguists including Vernet and Ddjean; and also independent writers. The opinions of this group are not homogeneous,and D6jean, for example,in order to put an end to "useless"discussions, accepted the official orthography,though with reservations(1980 a,b,c). A third position,which advocatesan intermediatesolution, is representedby the Haitian linguists and educators Pompilus (1973) and t official (1979). From their perspectivethe orthographyshould be phonemicbut wheneverpossible should use the sameconventions as the Frenchorthography to representsounds similar in French and Kreydl. Their reasoning is that French and Kreydl coexist in Haiti, and one day the country will be totally bilingual. Their main goal is to include the front-rounded vowels (usually associatedwith the speechof bilinguals)in the phonetic description of Kreydl and to provide a graphic representationfor each of them. Pompilus's reasonsfor this are that many non-educated people also pronounce those vowels in their everyday speech,and eliminating the vowelswill result in the rejection by many people of the orthography itself and of the use of Kreydl for instruction(1973: 30). It is difficult to reconcile Pompilus'sproposal for an intermediate orthography which should integratethe front-roundedvowels with his declarationthat it would be illusory to take the variety of the urban educatedas the standard (cited by D6jean 1980a: 170). This contradictionreveals the tension of creating a bridge betweenthe past ()and the future (when Haiti will be totally bilingual) at a time when the present itself is so full of conflict. Lofficial argues that it is better to prevent Metalingubtics and onhographic choice 435 potentialproblems since it is likely that with the influenceof schoolsand the media the occurrenceof the front-roundedvowels will increase(1979: 118).Thus it is better to anticipate the difficulties that could arise from two totally different systems and integrate the front-rounded vowels now. The Haitian linguist F6rdre advocates somethingquite similar with his proposalof an "ethno-orthography"(1974:23), that is, an orthographyculturally and sociallyacceptable - an orthography which must include the front-roundedvowels, which are not just marginal sounds(Ibid: 50).

5. Discussionof the arguments

The statusof Kreydl and the orthographicdebates are closelyrelated to questionsof representationat both the nationaland internationallevels. Orthographic issues can be includedwithin a broader framework of languageideologies - the cultural beliefs that underlielanguage practices, choices and attitudes of a people. Most of the discussions that have taken place in Haiti from 1930to 1990can be traced to ideologiesabout the inherentsuperiority of the French languageand are connectedto ideologiesabout the superiorityand refinement of French culture based on the achievementsof the great Frenchwriters of two past centuries:the Century of Reason (17th century) and the Century of the Enlightenment (18th century). Ideas of the clarity, exactness,, rationality,natural order, and richnessof the French languageas contributing to the greatnessof French civilization (Grillo 1989;Swiggers 1990) were echoedand defended by many Haitian writers. The notion that the French languageis superior to all others was transmitted to the French colonies and survived in the minds and practicesof both the ex-colonizers and ex-colonized.Schools in Haiti, which until recently remained exclusively in the handsof the French clergy, played an important role in keeping this idea strong. After its independence Haiti tried to reject anything that was reminiscent of France. Norktheless the new nation kept the French language and - also copied French administrativeand organizational structuresin order to survive.oSince that time, the newruling classeshave maintained two contradictory positions in their relationship to Europe,particularly to France: hatred and fear at one pole, admiration and emulation at the oppositepole. The same ambivalenceis reflected in Haitians'attitudes toward French and Kreydl, and it is in this light that the importance given to the issue of Kreybl orthographycan best be understood. If we look carefully,we will see that the debate is not purely about how to write Kreydl, i.e., how to represent graphically the sounds

6 Hoffm"n (1934) explainsthat the Haitians were awareof the paradoxinherent in their consciousimitation of thosewho had causedthem so muchsuffering, the French."[Theyl ... attempted to resolveit by assertingthat they had beenoppressed in mlonial timesnot by the French,but by the colons,a disreputableminority of adventurersrecruited from the dregsof Frenchsociety.." (p.59). These werenot the sameFrench of the metropolewho epitomizedculture and civilization.We thank Paul Garrettfor pointing this out. 436 Bambi B. Schiffilin and Rachelle Charlier Doucet of Kreydl. It is about the conceptionof Kreydl itself as a "language"and as an element of Haitian national identity. It is about how Haitians situate themselvesthrough languagesat the national and international levels.It is about the notions of.Ayisyenite 'Haitianness', of authenticity, nationalism and legitimacy. The battle over the orthographycannot be understoodif we do not situateit in its socialcontext. The dual linguistic systemis a manifestation of the dual social systemexisting in Haiti. From the perspectiveof the different orthographicpositions, all of which have social,political, cultural and symbolicimplications, two issuesare takenup: the questionof the "sounds" and the questionof the "look" of Kreydl .

5.I. The sounds of lbeydl

The first disagreement is about the sounds that the alphabet should represent, particularly about the existenceor non-existencein Kreydl of four front-rounded vowels, /ti/ as in French tu; lol as in French peu; lcpl as in French boeuf; I I as in French un. The existenceor non-existenceof the front-roundedvowels is viewed by many (includingD6jean, Lofficial) as the "dividingline" betweenthe educatedminority and the masses,between rural and urban. But to this is linked a more profound question of representationand legitimacy.Which variety of the languageshould be standardizedand codified? This technicalquestion has its counterpart in the socio- political arena.It hasbeen answeredby anotherquestion: Which varietyconstitutes the "real", "authentic"Kreydl? Consequently,who is the real Haitian and whose interests must be taken into accountand served?Not surprisingto anyone,these questions refer '' to the strugglesfor power that have gone on between Noirs and MulAtres 'Mulattos' since colonial times and the strugglesfor upward social mobility by the masses. 'exclusive Whether the front-rounded vowels are the apanage domain' of a minority of educated bilingualsas D6jean argues,or whether they are also used by monolingualsas F6rdre, Pompilusand l-official assert,what interestsus in the debate is the role of prestige marker ascribedto them by the population. D6jean himself documentsmany casesof hypercorrectionin the speechof monolingualKreydl speakers (1980a:124-125). Fattier-Thomas also notes the useof hypercorrections(related to the front-rounded vowels) in commercialadvertisements to establisha kind of complicity with the public (1984:41). This is to saythat for both bilingualsand monolingualsthe front-roundedvowels have a highlymarked and symbolicvalue.' For the educatedand non-educated urban dwellers, the front- rounded vowels are associatedwith the nonrounded vowels which are consideredtheir antithesis,and function as prestige markers,the front-roundedones being the prestigiousforms. For example,consider the following pairs:

7 Bourdieu's ideas (1975, lg7g, lg82) concerningprestige language, and symbolic capital illuminatethis argument. Metalinguisticsand onhographicchoice 437

Kreydl French English opposition front-rounded front unrounded front-rounded lnl- lil dui din du iz rice lol - lel bleu ble bleu blue lel - lel peu pD peur fear lindi lundi Monday These pairs are not minimal pairs. They are variants which are found in social dialectsand they function as social markers.The use of the front-rounded seriesis associatedwith educatedclasses, good manners,and harmonioussounds, whereas the secondseries is associatedwith popular usage,rough manners,strident and even vulgar sounds.Metalinguistic terms are used to qualify the secondseries, the nonrounded vowels,and reflect the low esteem associatedwith these sounds.To pronounce a nonroundedvowel when a roundedvowel is expectedis to make a mistake,and when 'sour/acidic onemakes this error, Haitians will say that one has a bouch su or bouch si mouth'or, more elegantlysaid, bouch surette'puckeredmouth'. This is related to the ideathat these nonroundedvowels have intrinsicallydisagreeable sounds, and these 'sour/acidic'. arelabled .tu or si Other meaningsof si or su in Kreydl refer to the taste of lemonsand other acidic or unripe fruits, and milk or other kinds of food that can turn sour. People can also have a disagreablesour smell, expressedas nCgsa a santi si 'that man smells sweaty'.These examplesshow the connection of the words salsi with the idea that something which has been agreeablehas turned disagreeable,sour, bad. The negative attitudes toward the nonrounded vowels also explain the hypercorrectionsoften made by non-educatedspeakers when they try to affect a certain degreeof educationand good mannersby avoidingthe marked soundseven in cases wherethere are no alternates.The expressionsbouchsa and bouch si as well asbhk sulsiand dj\l sulsi, of which bdk si and djdl si are the rCk variety and have the pejorative 'disgusting senseof trap', are used to describewhat speakersdo and how they sound whenthey hypercorrectthese vowels in their speech.Those ideas are internalized,it mustbe stressed,by membersof both endsof the socialladder who react with equal vigor against the official decision to eliminate the front-rounded vowels and to generalizethe use of the sulsi sounds.They do not want their children to acquire "bad habits"(i.e*, bouch su), but rather to speak elegantly and as closely to French as possible(Jean-Charles 1987; Zlphir 1990).It is reasonableto suggestthat an important assumptionunderlying the rounded-nonroundedvowel question is that Kreydl is a deformationof French and that French is the model to imitate, Both the linguistswho designedthe official orthographyand the governmentwhich implementedit consider thefront-rounded vowels as "marginalsounds". No graphic representationof them is includedin the alphabet because,says Vernet for example, "ll ne faut pas apprendred I'enfantdes sons qu'il neprononce pas quotidiennementdans sa languematemelle" (1980: 43-44).[One need not teach a child soundsthat he does not pronouncein his mother tongueeveryday.] Ddjean has expresseda similar idea: "Choisir un sysftmegraphique qui intdgreles voyellesafieieures anondies, c'est augmentersa dfficult€ d'apprentissage pour I'immensemajoitd des Haitiens" (1980a: 172-173).[To choose a graphic system 438 BambiB. Schieffelinand Rachelle Charlier Doucet

that includes the front rounded vowels is to increasethe difficulty of its learning for the majority of Haitians.] This implies that the front-rounded vowels sounds are difficult, (if not impossible),to pronouncefor monolinguals.Thus, those sounds should not figure in the alphabet in order not to embarrassand marginalize the majority of speakers. By advocating the use of the front-rounded vowels, D6jean expresseshis socio-political militantism: anyone who proposes to include them is "rdactionnaireet 4litiste" 'reactionary and elitist' and aims at promulgating a class and cultural imperialism(1980a: 175). To a lesserextent, the sameprestige issue holds for the use of certain nasalized forms. For many words, there is a choice between a less nasalizedform and a more nasalizedform; for example agrondm'agronomist' (educatedvariety) versusagronndm (non-educatedvariety). The double n marks the pronunciation as extendednasalization which has rural connotations."Excessive nasalization" is considereda characteristicof rural speechand is felt to be vulgar (rak) [nfficial 1979: 118).The ti diksyonnelcrqdl- franse (Bentolila 1976) chose the nasalized forms as "characteristic" of Kreydl, as indicated by the spelling of the word "diksyonn|".The dictionary has been controversial in Haiti and largely rejected. A set of implicit oppositionsand contradictionsemerges from the debatesabout the sociolectsand standardizationof Kreydl. The educatedare characterizedas being urban, of the higher social classes,elegant, civilized and speaking Kreydl swa, all of which is desirable for both bilinguals and monolinguals. The uneducated are characterizedas rural (habitan) from the lower classes,speaking Krqdl rdk, bouch sulsi, all of which is undesirable (for both bilinguals and urban monolinguals)but desirable for many nationalistsand progressives(linguists, intellectuals and politicians). The sound systemleads directly into the core of the debate about social classes, legitimacy and authenticity.If we push the idea of authenticityto its limits, the following question emerges:What soundsare those of the real, authenticKreydl? What is the real, authentic Kreydl? Thus, who is the real, authentic Haitian? The dominated "Africanized" massesor the dominant "Frenchified"elites? Is there a "pure" Kreydl? We see that the ideologicalbasis of such questionscan lead to a vicious circle. While question of the soundsof Kreydl lead us to issueslocated primarily at the national level, questionsabout the "look" of Kreydl can be situated at an international level and involve the relationship of Haiti with other countries. Here the focus is on issuesof nationality,independence and autonomy.

5.2. The look of Kreydl

The pro-etymologists center their argumentsaround the issue of the roots of Kreydl. According to them, Kreydl, derived from French, must reflect its origin, and shouldbe easily.classifiedby its appearanceinto a world languagefamily, that of the (Archer 1987).An etymologicorthographywill be helpful for learning French later, sinceFrench will alwaysbe spokenin Haiti. Haiti must not forget its membership 'French in the francophonie speakingcommunity'. To adopt an "Anglo-Saxonlook" Metalinguisticsand orthogaphicchoice 439

representedby the "non-[.atin"letters w, k, and y is to deny that membership(Archer 1987).According to the most fervent pro-etymologists,the use of the three above mentionedletters gives Kreydl a weird aspect,and even worse,will prevent the easy learningof French for both bilinguals and monolinguals.The most zealous pro- etymologists,who also present themselvesas zealousnationalists, think that the Anglo- Saxon orthography has been imposed upon the Haitian people by imperialist powers.Thisidea, expressedin t947, pointed to McConnell and l-aubach as agents of the United States, the first imperialist power. As later echoed by Archer (1987), the purposeof the United Stateswas clear: to eliminate French in order to introduce English,a task that would be accomplishedthrough the anglicization of Kreydl. The United Statesalso wanted to substituteProtestant religion for the Catholic faith and Vodoureligions. Their interestin literacyis only basedon their desireto sell Biblesand convertHaitians to their own religion. Still, according to Archer, the aim of the French, the secondimperialist power, in this endeavour is not very clear, but it is still worthwhile to raise the question: why the suddeninterest of the Western powers in the vernaculars,in the as well as in other parts of the world? Why this insistencein using the vernaculars in education?Archer calls for vigilance and the retention of French and an "etymological Kreydl system".Not everyonewho is pro-etymologicalexpresses such extreme ideas as fucher, but there is clearly a defensive reaction or a discomfort about foreign intervention in matters that Haitians consider strictly domestic. The same rejection (D6jean 1980a) or mistrust (lnfficial 1979) is also found in the other camps. Among the most contestedletters is the k, which not only representsthe danger of United States imperialism but has even been claimed to represent the threat of communism.8Deiean (1980a) reports that the literacy activistshave been accusedof beingdisguised communist agentsby a governmentofficial under F. Duvalier's regime. Here is revealeda fear of a sudden"wake up" of the masses.Dej€an also reports that a militant communist accusedthe same orthography of being "bourgeois,r€actionaire et macoute"(1980a: 55). The underlying assumption of the pro-etymologistsis that the orthographic systemof the language must be linked to a culture, to a tradition. The most extreme assumptionis that Kreydl needsFrench to rely on and that French must be given credit for its contribution. On the contrary, answersthe opposite pro-phonemic group, no nationality,religion, culture or political allegiancecan be ascribedto single letters like w, k, y. An orthography is just an arbitrary and conventional system, and from a linguisticpoint of view a writing systemshould be neutral (D6jean 1980a).Thus, the pro-phonemicgroup advocatesan orthographytotally new and independent for Kreydl. From their perspective,the choice of a phonemic systemis scientificallyjustified; it is the more rational and the simpler way of writing Kreydl, which should avoid the errors

8 It is interestingto note that the letter k is ascribedanother value in the developmentof Papiamentu,the creole languagespoken in Aruba, Bonnaireand Curacao,to mark "African ancestryn (Winer1990:26 fn l), while in the United Statesit is usedby blackpower movements and other groups asa markerof socialprotest against undemocratic practices, as in "Amerika'. 440 BambiB. Schieffelinand Rachelle Charlier Doucet and aberrationsof the French orthography.It isjust a conventionalsystem, and the use of w, k, y has nothing to do with Anglo-Saxonor Protestant,Russian or communism. The underlyingassumption is that Kreydl is a languagethat can standby itself and does not need someoneelse's tradition to rely on. Consequently,the pro-phonemicgroup arguesthat the teachingof Kreydl must be independentfrom the teachingof French. For the militants of that group who are the most involved in literacy (D6jean, Vernet), the target is the monolingualmasses, not the educatedelites. It is interestingto note that the two most extremecamps, the pro-phonemicand the pro-etymologists,advocate the authority and neutrality of "science"for their arguments.But we have seen that there are no neutral positions,only ideologicalstances.

6. Conclusions

These arguments reveal the complexity and ambivalenceof cultural definitions of 'Haitianess'. Ayisyenite As Haiti is still in the process of integrating its dual African/European heritage, there are numerous arenas where cultural duality is continuallynegotiated. Religion (voudou and westernreligion) is one important locus of this negotiationprocess, and language(French and Kreybl) is another.The debates about languagealso illustratethe extent to which an issuecan be politicized - in this caseorthography - where intoleranceis the dominant note. With few exceptions, throughout the 19th century, the elites have defined themselves as "colored French" and have spoken of Haiti as the "fille ainte de la France" [the oldest daughterof France] or as France Noire " France".Although acknowledgingtheir African roots, the elites also claimed their affiliation with L^atin culture. The Indigenist School and Haitian Ethnology School affirmed that Haitian culture was Afro-l-atin but with a predominanceof African elements.African elements are associatedwith blackness,vodou, the masses.Kreydl, the only element that the educatedclaim to sharewith the masses,is the enduringsymbol of Haitian identity.But we have seen an ambivalencein the valuesattached to Kreydl: the rdk variety is used for nationalistdiscourse on authenticityand pride, but at the same time is associated with negativeconnotations about thesesame masses. This ambivalenceis reflectedin the orthographyquarrel itself aswell as in the wish on the part of the pro-etymological camp for a Kreydl orthographic"bridge" between the past (French) and the future (when French and Kreydl will co-existpeacefully) but with no solid foundationsin an ever changingpresent. It is noteworthy that Haiti is not the only country where languageissues and orthography issuesin particular provoke emotional reactions.Examples abound of orthographic quarrels in industrializedsocieties as well as in the so-calledthird and fourth worlds.eThe recent argumentsregarding the implementationof the reform of the French orthography are a good example of how a country can stick to its

9 Regardingother orthographicbattles, see for example,Bernabe et al. 1983;Coulmas 190; Fodor and Hegdge1982-l9X); G.E.R.E.C. 1982; Graves 1991; Jaffe 1990;Ludwig 1989. Metalinguisticsand orthographicchoice 441 orthographicicons as symbols of its identity. When a languageis codified and an orthographyis officially adopted,this is usuallyinterpreted to mean that there is one correctway to spell and write the language,and that all others are simply wrong. To designand implement an orthography is neither a simple nor neutral endeavor. It establishesnorms of pronunciation as well as norms for writing. The main reason for reaction is that the elaboration of an orthography implies the choice and standardizationof one dialect over the others. And when a variety through its officialization is given the status of a standard, the users of the other varieties sometimesreact with a surprisingvirulence because they feel that their languagevariety is not being represented. The latest dtografKreydl is implicitly positioningthe Kreydl spoken by the masses asthe standard.This has created resistanceboth to the adoption of the orthography andto the use of Kreydl as a medium of instruction in school.The double resistance comesfrom both the massesand the educatedelite minority. The massessee the officializationof written and spoken Kreydl in school as limiting their accessto French andconsequently social and economic mobility. The elites, who already know Kreydl, don'tsee the point of teachingit, in any form, in school.They also hold the view that kqdl rCk and gwo Kreydl are directed toward the lowest common denominator, bringingthem down insteadof elevatingthe masses. As we have seen, the debate around the orthography takes on issues of legitimacyand authenticity. Who is meant to be represented by this otograf Krqdl, whosespeech, values, cultural identity? Which version of Kreydl can be said to be genuinelyHaitian? Is it the Kreydl swa or Kreydlfransize of the educated minority, or theKrqdl rCk, the gwo Kreydl, or the bon Kreydl, the "real Kreydl" of the masses?So weraise this as a rhetorical questionas a way of questioningthe question:what is the realKreydl? This can only be answeredafter addressingthe question, who is the real Haitian? Any linguisticpolicy that would be exclusivelybased on "purely linguisticfacts" takesthe risk of going the wrong way, since languageis not only an instrument of communication:language also carriessymbolic values that condition social,political, andeconomic spheres. The main questionto raise,in the Haitian case,is how to give thecurrently marginalized variety of Kreydl spokenby the majority of the population, aswell as that majority itself,the effectivemeans to reversethe presentsituation. Such meansare not limited to the linguisticsphere, but necessarilyoverlap and will have consequencesfor all aspectsof sociallife.

References fucher,M. T. (1987)La crdologiehaitienne. Port-au-Prince: Ir Natal.

Bentolifa,A- et al. (1976)Ti diksyonnilcreydl-franse. : Editions Caraibesflatier. 442 Bambi B. Schiefelin and Rachelle Charlier Doucet

Bourdieu, P. (1975) "Ir f6tichisme de la langue."Actes de Ia rechercheen social sciences 4:2-32.

Bourdieu,P. (1979)La distinction.l*s editionsde Minuit: Paris.

Bourdieu,P. (1982)Ce queparler veutdire. Paris: Librarie Artheme Fayard. 'l:nguage Coulmas,F. (1990) adaptationin Meiji Japan.'In B. Weinstein(ed.), Language poliq and political developmenr.Nonvood NJ: Ablex, 69-t6.

Ddjean, Y. (1975a) Dilemme en Haiti: Frangab en pdil ou piril ftangais? New York: lrs Editions Connaissanced'Haiti.

Ddjean,Y. (1975b)'Franse se danje.'Sil23-24:32-39.

Ddjean, Y. (1980a)Comment dcire le criole d'Haiti Qu6bec:Collectif Paroles.

D€jean,Y. (198Ob)'Ann kasekoub dtografla." Sd/8 (a8-49):a-5.

Ddjean,Y. (1980c)'lv Dejan ap reponnkeksyon S0l sou nouv dtografkreydl la.'S?/ 8 (48-a9):25-35.

Fattier-Thomas,D. (1984)'De la varidtdrOk I la vari€t6swa." Conjonction L6l-162:39-5I.

Fattier-Thomas,D. ('1987)"Portraits de bilinguesfrancophones haitiens." Conjonction 176:97-125.

Fdr0re, G. L(1974) 'Haitian Creole sound-system,form-classes, texts.n Ph.D. Dissertation.Dept of ,University of Pennsylvania.

F6rdre,G. A. (1977)" in Haiti.' CaibbeanQuarterly 23 (I):50-60.

Fodor,I. and C. Hag0ge(eds.) (1982-1990) Langtage reform: History and funre.5 volumes.Buske Verlag Hamburg.

Garrett, Paul Brian (1990)'Spoken identity." Senior Essay, Dept of Sociology,Yale University.

G.E.R.E.C.(1982) Chane culurelle crdole. Schoelcher, Martinique Universitaire Antilles Guyane. 'Orthographies Graves, W. (1991) and identities:The politics of writing in an American Indian mmmunity.'Paper presentedat the AES meetings,Charleston, South Carolina.

Grillo, R. D. (1939) Dominant languages:L.anguage and hierarchyin Britain atndFrance. Cambridge: CambridgeUniversity Press.

Hoffman,L-F. (1984)"Francophilia and cultural nationalism." In C. Fosterand A Valdman(eds.), Haiti today and tomonow. New York: University Pressof America.

Hoffman, L-F. (1989) Hai:ti: Couleurs,croyances, cr6ole. Montrdal: Editions du CIDIHCA.

Hyppolite,M.P. (1949)Les oigines de vaiationsdu criole haltien.Port-au-Prince. Imprimerie de I'Etat. '[:nguage, Jaffe,A (1990) identity and resistancein Corsica."Ph.D.dissertation, Indiana University. Metalinguisticsand orthographic choice 443

Jean-Charles,H. L. (1987)'Attitudes of teachersand parents toward French and Creolein Haiti.' Ph.D. Dissertation,School of Education,Stanford University. lnfficial,F. (1979)Criole, frangais: une fausse querelle. Montreal: Collectif Paroles.

Ludwig,R., (ed.) (1989)Les uioles ftangaisentre l'oral et l'ecrit.Tubingen: Gunter Narr Verlag.

Orjala,P. (1970) "A dialect surveyof Haitian Creole."Ph.D. dissertation.The Hartford Seminary Foundation.

Pompilus,P. (19?3)'De I'orthographedu cr6ole."Conjonction l2O: 15-34.

Pompilus,P. (1985)Le problime linguistiquehattien. Port-au-Prince: Imprimerie Deschamps.

Pressoir,C. F. (1947)Ddbat sur le criole et le folklore.Port-au-Prince: Imprimerie de I'Etat.

Pressoir,C. F (1954) Methodede franEais oral. Port-au-Prince.

Schieffelin,B. B. and R.D. Doucet (ms.) nThe nrealnHaitian creole: Metalinguistics,ideology and orlhographicchoice."

Swigers,P. (1990)"Ideolory of the clarityof French.'In J. E. Josephand T. J. Taylor (eds.),Ideologies of language.New York: Routledge, ll2-130.

Valdman,A. (1988)'Diglossia and languageconflict in Haiti."I.J.S.L. 7I:67-80.

Valdman,A (l9S9a) "Aspectssociolinguistiques de l'elaborationd'une norme dcrite pour le cr€ole haitien."In R. Ludwig (ed.), 43-63.

Valdman,A 1989b) "Vers la d6minorisationdes cr6oles."In Actes du symposiumorganisd par I'AssociationInternationale de LinguistiqueAppliqude et la CommissionInteruniversitaire Suisse de LinguistiqueAppliqu€e. Neuchatel: 187 -2M.

Valdman,A (1991)'Decreolization or dialectcontact in Haiti." In F. Byrne and T. Huebner (eds.), Developmentand structuresof Creolelanguages. Philadelphia: John Benjamins,75-88.

Vernet,P. (1980)Techniques d'dciture du crdolehaitien. Port-au-Prince.

Vernet,P. (1989) 'Alphab€tisation en Haiti." In Hommagesau Dr. hadel Pompilus.Centre de LinguistiqueAppliqu6e, Port-au-Prince: Imprimerie Deschamps.

Winer,L. (1990) 'Orthographic standardisationfor Trinidad and Tobago.'Language problems and languageplanning 1a(3): 237-28.

Ttphir,F. (1990) 'I-anguagechoice, language use and languageattitudes of the Haitian bilingual community.'Ph.D. Dissertation. Indiana University.