interviews form the foundation of this analysis (Schieffelin 1986, 1990). ...... Bendix, Edward (1994) The grammaticalization of responsibility and evidence: ...
-243. Pragmatics 5:2.225 International Pragmatics Association
CREATING EYTDENCE:MAKING SENSE OF WRITTEN WORDS TN BOSAVT1 Bambi B. Schieffelin
Sincetheir earliestcontactwith Europeans,the Kaluli people who live at the base of Mt. Bosaviin the Southern Highlands Province of Papua New Guinea have viewedbooksaspowerfuland authoritativesourcesof informationthat white people useto shapeand control the behaviorof others.In a narrative told to Steve Feld andmyselfin 1990about governmentcontact in the early 1940s,an educatedKaluli mantold us about his father who had been selectedby white patrol officers as the first local counsellor.As he put it, "my father was given the black shirt with a red stripe,the belt, knife, stick and a book, that book, people thought that if you kill, the blood of a dead person will go inside in the book, and the white man will know straightaway and come and shoot you with a gun; that .fear, everywhere so, everyonegot frightened when my father got this." Thisbook,whichwaskept by the counselloraspart of his responsibilitiesand takenout only during infrequentgovernmentpatrols made by white officers,listed the namesof villagers.While Kaluli people did not sharean understandingof why their nameswere written down by governmentpeople, they did not miss the fact thatthisbook and its meaningswere createdand owned by white people,who used it as an instrumentof control, authority,and information.These early censusand record keeping actMties, part of pacification efforts, were used to track and documentKaluli people in order to discouragetheir periodic relocation to new villagesites,their solutionto minimalizingthe depletionof local resources.This was oneof the earliestexperiencesfor Kaluli people of what books could do, and what peopledid with books. In the mid 1960stwo other changesoccurredthat would introduceadditional ideasaboutbooksand literacymore generally- the first, which I would like to think wasrelativelybenignthough not inconsequential, was the arrival of anthropologists, first E. L. Schieffelin,followed by myself and Steve Feld, whose visitations of differentdurationswould continueinto the present.The secondchangewhich has had far reaching consequencesfor Kaluli social and ceremonial life, was the
' I would like to thank the National Foundation and the Wenner-Gren Foundation for fundingfieldwork in 1984 and 1990. Thanks also go to Steve Feld and Elinor Ochs for their always helpful comments. Finally, I would like to thank Sue Gal and Kit Woolard for their cogent written comments,as well as the other members of the Center for Transcultural Studies Working Group on language and the participants in the School for Americn Research Seminar on language Ideologiesfor asking tough questions. This essayis dedicated to the late Kulu Fuale; his patience and assistance in helping me understandthe Ikluli language for over fifteen years was enriched by his unique linguistic curiosity.
establishmentof a fundamentalistmissionand air strip which was managedby a Papuan National until Australian missionariesarrived in the early 1970s. This essayfocuseson new communicativepracticesthat emergedas a result of interactionsbetween Kaluli people and Australian fundamentalistmissionaries beginning in the early 1970s.To do this I draw on ethnographicand linguistic fieldwork begun in 7967 that has continued intermittently into the present. Taperecorded,transcribedanalysesof socialinteractionover time and acrossa wide range of situationsand activities,in additionto participantobservationand informal interviewsform the foundation of this analysis(Schieffelin1986,1990).Additional analyseshavealsoinformed my and ethnomusicological ethnographic,sociolinguistic assertions(Feld 1988, 1990; Feld & Schieffelin 1982;E.L. Schieffelin f976). As Kaluli people were introducedto new forms and sourcesof knowledge about their own and the outside world, their ideas about truth, knowledge,and authority were challengedand changed,affectingtheir communicativepracticesas well as their social structures. In such contact situations,new communicative practicesexpressthe interestsof both the missionizedand the missionizers.Kaluli people were active contributorsto the linguisticand socialreorganizationof their own society,as evidencedin the emergenceof severalgenresnew to the area. In suchsituationsof socialchange,new languagesocializationactivitiesoften develop. Particularlyrelevant for this essayare literacy lessonsand sermons,both of which constitute important activitiesfor languagesocialization- socialization through the use of languageand socializationto use language- which continues throughout the life cycle (Ochs & Schieffelin 1984).From a Kaluli perspective, lessonsand sermonsshare interpretive frames and participant structuresas they originatefrom the samesource,the Christianmission.Combiningspokenforms and written materials,both genresprovide a new discursivespacein which Kaluli people rethink their past, one consistingof particular social practicesand beliefs - and distancethemselvesfrom it. To accomplishthis distancing,severaltechniquesare deployed.In the literacy lessonand in the written text that it draws on, the terms mo:lu, tamina'before', 'a long time ago, before' are systematicallyopposd to o:go: 'today, now' to create narrativesabout how thingsusedto be, in contrastto how thingsare and shouldbe. Part of this contrast invokes the source of the difference: What "our fathers" believed is contrastedwith what "this book showsus really well" and what "new words which really tell us." New facts,suchas thosedrawn from health lessonsand bible storiesare used to reviseboundariesor createnew ones.For example,Kaluli people are reminded both in the written text and in the oral presentationsthat "before we didn't know," "we didn't understand,"in contrastto "we now know," "*e hear it really well." In classroominteraction,studentsare asked to register their agreementwith these claims. The practices of the past and the presentfuture are also coded by their assignmentto gender roles; women'sbeliefs and activitiesare connectedwith the past, what was done before the missionwas established,while men's beliefs and actions are seen as forward looking, progessive,taking up the new ideas. Not surprisingis the evaluationassociatedwith each of thesepositionings:Women are negativelyevaluatedwhile men are viewedpositively.Theseconceptsare promoted in literacy materialsand fortified through oral presentationand lessonsby extensive linguistic means, including an innovation in the evidential systemwhich further
Making sense of written words in Bosavi
underscores an increasedremotenessof the not very distant past.
Literacyand social change Thisessaydrawson a body of ethnographicwork that views literacy practicesand activities as historicallycontingent,ideologicallygroundedand culturally organized (Besnier1995;Collins1991;Duranti & Ochs 1986;Gewertz& Errington 1991;Guss 1986;Heath 1983;Kulick & Stroud 1990;McKenzie 7987;Street 7984;Street & Besnier1994).This work demonstratesthat societies"take up" or organizeliteracy practices in culturallyvariablewaysdependingon who is interestedin literacy,and howliteracyis viewed.As with other genresand activities,those invoMng literacy practicesare constitutedthrough specific interactionalroles, arrangements,and which use particularforms of languageto enact socialrelationshipsand sequences negotiatesocial identities.In situationswhere literacy is introduced as part of Christianmissionization,literacy activities are often shaped by competing culturalframeworks.Theseframeworksare encodedin the ways epistemologicaland in whichinformationis presented,knowledgeis talked about, and analogiesare drawn;they are also apparent in the connectionsthat can not be made. Forms of resistence to literacypractices reflect not only linguisticideology,but social and historicalforces more broadly. Thus it is also useful to situate this work within currentethnohistorical accountsof competinglanguageideologiesthat have played a role shaping colonial and missionary encounters (Cohn 1985; Comaroff & Comaroff1991;Fabian 1986;Mignolo 1992;Rafael 1988).
Truth and evidencefor it Kaluli people,like many Papua New Guinean societiesare concernedwith the sourceand truth of what they and others know. They havewell elaboratedideas of how truth is constituted,proven and linguisticallymarked. In the so-called"old" days,before missionaries,government patrols and anthropologists,sources of knowledge and proofs of truth were relativelystable - what the "fathers"said was whatwasknownand believedby mature membersof society.There was little reason to doubt the truthfulnessof what had always been said about the natural and worlds that Kaluli people inhabited.Through direct experience,the supernatural word and faceto face interaction,culturalknowledgewasorally represented, spoken andauthorityand responsibilitycould be argued,and often were. Throughthe mid-1970sand into the 1980sthe missionariesintroduced new factsaboutthe world, ones they claimed to be either scientificor religious.These factswere soon representedthrough new words,genres,registersand conventions for speaking.Literacy in the vernacularand Tok Pisin was introduced.The Bosavi missionprimary school used English as the languageof instruction, but Kaluli peopledid not know either English or Tok Pisin. The few who successfully completed sixyearsof primary schoolwere draftedimmediatelyinto missionservice: Somewere trained as pastors,store assistantsor medical assistants.All became importantplayersin the changingsocialorder. With the exceptionof one individual whosenarrative opened this essay,before the early 1980s no one had been
Banbi B. Schieffetin
educatedbeyond grade six. workedhard to establishtheir authorityand The fundamentalist missionaries took the view that conversionshould move rapidly. Their evaluationsof local arguments,and sermons.By cultural practiceswere echoedin Kaluli conversations, 1984 Kaluli people had given up traditionaiceremoniesand many "traditional" practices.With Christianityand its new material resources,the socialorganization of Kaluli societybegan to change:What had been a small scaleegalitariansociety beganto reshapeitself into a looselystratifiedsocietywith new roles (obs) such as pastor,deacon,aid post orderly,and trade storemanager.Eventually,Christiansand non-Christians took up residencein different parts of the village.Whatever power had been granted formerly to older men was now taken by those who had gained knowledgeof how the mission,the missionstore, and the missionaryworked. These socialchangeshad linguisticconsequences. Young men who acquired Tok Pisin extendedtheir interactionalrangeand could work as interpretersfor the governmentpatrol ofticers.The few young men who becameliterate in Tok Pisin and Kaluli, and were part of the missioneffort, collaboratedwith the Australian and translatedtextsfrom Tok Pisin(MrpelaTestamen) missicrnary or English(health booklets and literacy primers) into new varietiesof the Kaluli language- varieties that were constructedby a powerful non-nativespeakerin conjunctionwith native speakerswho wanted to acquirepower. Of the four mutually intelligibledialectsin the Bosavi area, one had been randomlyselectedearlier by a missionfield linguist to be used for the orthography(Rule 1966)and many featuresof dialect variation exist in the literacy materials, in addition to many spelling inconsistenciesand syntacticsimplificationsand errors. As a result of these collaborations,a new medium was created - booklets printed in a new variety of the vernacular.Written with the authority of the mission, they introduced severalnew typesof evidenceinto Kaluli life; the first is the written Kaluli word. Simple but dramaticnarrativesurged socialchange.The motif here, and in sermons,was consistent:The past versus the presentffuture,articulated through examplesof Kaluli ways of life which were depictedas backwards,wrong, and deriving from talse beliefs, in contrast with European ways of doing things (from building houses,to health practicesto hair styles)which were presentedas new, right and good. Simpleline drawingsby the missionariesservedas illustrations. Narrativestook place in recognizablelocal places,and charactershad Kaluli names. To make these narratives more believable,the missionaryillustrated the more advancedbclokletswith black and white photographsof local people engagedin the activity described in the narrative. So in reading the texts, Kaluli could see themselvesimaged, participatingin the new practicesbeing promoted. These graphic and photographic imageswere used as and became evidentials,that is, another new sourceof evidencefor authorityand truth. Combinedwith print, they became a source of knowledgethat could be seen,referred to and reported on. Book-letsintroducednew informationin new formats.They becamethe sourceof that information, and evidencetor it. Those with accessto these new sourcesof knowledge and truth, or those who could claim to understandthem, became authoritiesthemselves.
Making sense of written words in Bosavi
Evidentials From a linguisticperspectiveas well as a social perspective,Kaluli people have alwaysbeenconcernedwith evidence.Their languageprovidesthem with a range of evidentials, morphologicaland lexicalmeansusedby speakersto formally mark the sourceor evidencefor the basis of their assertions,their attitudes towards knowledge, and the responsibilityassumedin making a claim. Evidential particles, metalinguistic verbs(e.g.,verbsof saying)and other sensoryverbsindicatingsources of knowledge(e.g., hearing, seeing),and reported speech are just some of the meansby which speakers establish the "truth" of their assertions and take responsibility for them. For example, in Kaluli discourse,ranging from casual conversation to more tbrmal arguments,speakersindicate through morphological or lexicalmeans whetherwhat they are sayingderivesfrom direct experience,visual, verbalor sonicinformation,speechreported to them or re-reportedto them, commonknowledge,or inferencemade from other secondaryevidence.
Selectedevidentials in Kaluli
speaker'sassertion is basedon visiblefuisual evidence that can be shared by 'this addressee. Magu we ntogago:lo:b banana is bad I see'; Do:wo: ha:na:nigabo:lo:b my father is about to go I see'.
speaker's assertion is based on deduction or inference from something sensed aurally or through other senses, but without attribution of particular source. To o:dowayo.'nr 'there's talk around I'm hearing' No: nun o:dowayo:nt 'therc is the smell of cooked meat I'm smelling' indicatcs direction in which an event being talked about is taking place.
used in intcrrogative lorms to indicate doubt regarding accuracy of information, can elicit confirmation from addressee. Dimia:iba:le? 'I wonder if he will give it?'. Also used when wondering aloud to indicate uncertainty and possiblyseek an opinion from addressee,Ha:na:no:wa:le? nto:hn:na:no:wa:le?'I wonder will I go? Will I not go?'. 'it's really true;' indicates certainty of assertions, really/truly/only. Hedele 'it's Ho:nde really water and not anything else.'
-ntala: - -bala: negative after inference indicating disappointment. Ne alima:no:mala: 'I will not lie down' (seeing there is still more work to be done). -ntalo: - -balo: affirmative emphasisafter question or when answer is opposite to what is 'is 't's truly expected. Aoleya:le? it his real brother?' aolentalo.'/ his real brtrther!' but aolennla.'/ it's not his real brother!2 -lo:do:
emphasiswith disappointment/sadness. Ha : na :no : lo :do:' alasyou are going away'.
2 The form -ntala.' has two distinct meanings which are disambiguated by stress/pitch differencesin the language. These are not indicated in the orthography.
Bambi B. Schieffelin *-lo:do: a:la:bo: $e now know from this source, we did not know before' (used when referring to information from written sources) *hia
'here', visibleftisual evidence, used to extended use from the Tok Pisin indicate meaninq similar to -to:b.
Evidence based on verbal sources use a range of forms, such as -do: Immediate repeat of direct 'here'. Speaker C ->B wedo:! 'here' quoting of someone else. For example, Speaker A ->B we! (he/she said). The majority of evidentials for reporting speech, however, are formed with a:la:mala:la: sanra'say like that.' Context and pronouns clariff number and person. Such forms include: a:la: a:la: a:la: a:la: a:la: a:la: a:la:
siyo: sa:labeka: siyo:laleka sio:lo:bo:ka: sili sa:la:ingab sili sa:la:ingo: salan
used for speaker self report or to report what another has said 'some one else recently said' (not used for lst person) 3rd hand reported speech 4th hand reported speech someone (sing/pl) is saying (duration) someone (sing/pl) was saying (duration) generally said/one says (habitual)
Other sensory verbs are also used as evidentials,and the appropriate noun disambiguatesor adds 'smell,' goloma'touch,' emphasis when needed. For example,dabunta'hear,' but it can also mean bo:ba 'see,' asuluma'think, feel, know, understand,experience.' Emphatic markers, both lexical and morphological are used extensivelywith evidentials. For example, emphasiswhen close to addressee- lst time or lst repeat emphasis - 2nd repeat emphasiswhen calling out to addressee 'really', lery' mada 'trug' hede 'really truly' hedele made hedele 'really very truly' -ka: -a: -o
In contrast to the more narrow linguisticview of evidentialswhich tends to focus on categories of "truth" (Jakobson 1951), a broader social interactional perspective displays their multifunctionality (Silverstein 1985). Bybee describes evidentialsas "markersthat indicatesomethingabout the sourceof the information in the proposition (1985:184)."Willett has pointed out that the notional boundaries of evidentialityare still unclear,but as a semanticdomain, evidentials"participate in the expressionof the speaker'sattitude toward the situation his/her utterance describes (1988: 52)." Bendix suggeststhat it is not enough to analyze the epistemologicalcategoriesof evidentials,but one must view them as important resourcesused by speakersto manipulateclaimsof responsibilityand evidencein strategicinteraction (1994: 243). The social and historicalcontext of the topic of talk, as well as the socialrelationshipthat holdsbetweeninterlocutorscan affect the choice of evidential marker, which, as Fox and Clifford point out, is sensitiveto differencesin claims to authority (1991). Emphatic markers, affect markers and evidentials often co-occur in the same word or same utterance, and must be consideredtogether. Analysesof evidentialsin discoursehighlight the importance speakers attach to establishing their authority with their aridience, while
Making sense of witten words in Bosavi
acknowledging the dialogicalnature of the production of meaning.Evidentialsare usedto convey affective and propositional meanings,and the same evidential markersmayserveboth functions.Havilandsuggests "propositions... live in a moral whichincludesnot only what participantstake as true, or what they agree universe, to think,but also agreementsabout how to think and feel about what they agree upon"(1989:61). Personsnot only exchangeclaims about the world, but their affectivestancestowardssuch claims.Giv6n (1982)has pointed out that speakers and hearershave an implicit contract to mark degrees of certainty in their propositions.Propositionsthat are to be taken for granted and viewed as unchallengable by the hearer require no evidentiaryjustification by the speaker. Propositions that are assertedwith relative confidenceand are open to challenge by the hearerrequire evidentiaryjustification(ibid: 24). Furthermore,in situations of conflict,what may be contestedis not the claim itself, but how someoneknows it. Thusthe use of an evidentialis telling, and its choiceis critical. Drawingon linguisticexamples,Chafe and Nichols suggestthat an analysis of evidentials revealsa "natural epistemology"(1986:vii). I would like to suggest thatwhenevidentialsare examinedin the contextof their use in socialinteraction, such analysisreveals a "cultural epistemology." Everyday talk offers excellent opportunities to examinehow individualspersuade,argue and make claims using evidentials. Activitiesof talk in situationsof rapid socialchangeoften take up the topicof competingepistemologies, eachdifferentlyvalorized.Suchculturalsituations mayresultin linguisticinnovations,expressedthrough evidentialsas well as other linguistic means.I sharea view with other linguisticanthropologists(e.g.,Haviland 1989;Hill & Irvine 1993,Lucy 1993)that the encodingof knowledge,authority and truth is a linguisticas well as a social phenomenon;the two must be viewed as interdependent.
Thehealthlesson My analysisof Kaluli literacy lessons,which introduce "scientific facts", shows innovationsin morphological torms expressingepistemic stance, as well as in rhetoricaland event structures.These linguisticchangesare a notable responseto missionization and underscore Kaluli concern with the sources and nature of knowledge and truth. The particularevent I draw from is a transcribedaudio-tapedliteracy lesson thattook placein 1984at the Bosavimissionschool.It is part of a larger project on Kaluli languageuse and social changewith Steve Feld that draws on materials collectedsince 7967.In this new speechevent, as in other innovated genres,all levelsof languagehave been affected- the phonology,morphology,lexicon,syntax, pragmaticsand of course,the cultural assumptionsthat organizespeech semantics, activities. In spite of the fact that literacy instruction events draw on models of instructionimported from Western classrooms,there is clear evidence of local languageideology throughout. We will see how at a particular point in Kaluli history,written texts were granted authority as Kaluli people constructedlinguistic meansfor entitling texts and making them authenticand authoritative sourcesof factualknowledge,even when there was no basisin fact for doing so. The participantsin this speechevent are the instructor,Kulu Fuale, one of
the few Kaluli Christianstrained to teachvernacularliteracy,and 24 teenagedmale studentsin the 5th gradeclass.Kulu, who was fluent in Tok Pisinand spokea little English called this a "health lesson."The lessonlasted45 minutes;most of the time was spent focusedon a booklet about malaria. Two printed texts that have been trartslatedinto the vernacularare used in this event. The first, a booklet about malaria,is basedon an English script, and is widely used by missionariesin PapuaNew Guinea.The Kaluli versionwas created and translated by Keith Briggs, the missionaryin charge of the Bosavi station, collaborativelywith Kulu Fuale. However, during the course of the lesson Kulu repeatedly assertsthat "Briggswrote it." Kaluli people do not take credit for the production of these materials.The remainderof the lessondrew on a second written text, a selectionof versesfrom the New Testament translatedfrom Tok Pisin into Kaluli by Kulu. Both are presented as containing truths previously unklown. This essayfocuseson the tirst part of the health lesson. I base my analysis on my transcription of the entire 45 minute long audiotapedevent, during which time Kulu readsfrom and talks about both written texts,writes on the board, talks about other topics,and elicits classresponses.The transcription and the printed texts must be consideredtogether becauseof two significant relationships:between the written texts and the oral text (word-word relationship), and betweenthe oral presentationand the world that is represented, or misrepresented, through it. Selectionsfrom the transcriptwill be usedto illustrate how evidenceis marked in a varietv of wavs.
General participant and event structure The introduction of Christianchurch servicesand adult literacyclassesexposedthe Kaluli people to a radicallydifferent presentationaland participationstructure. Previously,in most Kaluli speechsituations,no singlespeakercontrolled the floor, speakers self-selectedand many voices,some quite loud, made simultaneous contributions to whatever topics were being entertained. In contrast, Christian speakingeventscan be characterisedas thosein which a singlespeakercontrolsthe floor and, in addition, has all the relevant and correct information. There are no interuptions; group responseis elicited,coordinatedand in unison,in responseto questionsthat seekone answer.Question/Answersequences which are usedin local sermonsare similar to those found throughoutlessons.They are unlike any other discoursesequenceI have recordedin over 150hours of transcribedKaluli speech. They resemblethe Western-styleclassroomQ/A sequencewith one correct answer, which is searchedfor until it is reached.Furthermore,Christiandiscussionis orderly and voices are never raised.Literacy lessonsare similar in this regard. In addition, they shareframing devicesand the participantstructurefound in local sermons.This is not surprisinggiven the strong influence of and connectionbetween Christian activitiesand schooling,includingvernacularliteracy classes.For example,in this literacy lesson,Kulu beginsthe classwith the directivethat all will sing a song that
Making sense of witten words in Bosavi
usesa popularhymn melody to which he has set new words.3 KFI 7
gisalowo: mo:la:bi will sing a song
da:da:sen ( s ingsne ) o: b a :g a n a l a b od: a :d a :s e d n a :d a :sen I lrcar hear heur birds sirtgirtg
one two (classsingssongin unison) otrc fwo
madao:m thankyou ##
After singingthe first line of the song,Kulu uses'okay' (4) as a discourseboundary marker,and countsin English,'one, two' so that the studentswill all sing together. This introducedstyle of singingdeparts from Kaluli song style which has a very differentaestheticstructure(Feld 1988).After the studentssing,Kulu thanks them. The useof mada o:m 'thank you' which is viewed by Kaluli people as a missionintroduced conceptand expressionalso marks this as a Christian actMty. All of the studentshave copiesof the booklet on malaria and its prevention calledHamulee wala.fbo:lo:'Hamule got sick' and standingin the front of the class the topic (9) "whatmosquitoesdo" and directseveryone'sattention Kuluannounces to the book,the sourceof the information.
Inconsistenciesin orthography are due to the preservation of the dialect features of speakers. Inconsistencies in the written texts are presentedas they appeai in the original materials, There are inconsistencies of spelling, as well as grammatical crrors throughout the written literary materials, andthesesometimescause difficulty in reading. There are discussionsin the lesson of problems with the \riting" (spelling), but those discussionsare beyond the scope of this paper. The transcription preserves the breath grouping of the speaker, and the numbers used throughoutindicatethose breath groups. However, in some lines, (153), breath groups are indicated asin the original transcript using the transcription convention -. Also used is s/c for self corrections byspeakers. Selections from the transcript are used to illustrate particular points, but the sequential numbersof the complete transcript are retained to make this analysis usable with other analyses of lhe sametranscript. A break in sequence is indicated by # #. There are no interruptions or overlaps in this speechevent. Boldface is used to indicate the phenomena of interest.
Bantbi B. Schieffelin
Kulu is holding the booklet 9
## o:go: kiso:wa:lo:dimidabo:ko:lo: kiso:wa:lo:dimidabo:ko:lo: agelema:niki today what tlrc mosquitoesdo, wlnt tlrc mosqttitoesdo tlnt's what we are readirtgabout
1 0 -1 1 we bo: ba,Ha mu l e w e look at this , Hamtile here (re the booklet) 12
taminamiyo:kiso:wa:yo: a:la:dimidabo:lo:do: a:la:bo:niyo:mo:asulanko:sega before,what mosqttitoesdo xwenow know we did not know thesethings but
mo:lu nili doima:yo:ko:segao:go:dinafa asulabbuko: wema:walasalab beforeottr fatlrcrs (erg), but now wereally know this book (erg) sftowslinstructs
a:la:fo:ko:lo: niliyo: buko: wena ba:da:sa:ga: therefore when we look in this book
tif s/ctaminamiyo:niyo:mo:asulo:ko:segano niyo:nulu alifo:alifo:labamiyo: kiso:wa:lo:nanogdiabo:we aungabo:lo:do: a:la:bo: laler slc before we did not kttow bttt wlrcn we are sleepingat night the work mosquitoesdo is like this *we now know
buko: wenamilo: to salab we da:da:sa:ga:asuluma:niki listening to what the words in this book say makesus knowlunderstand ##
Kulu (9) sets up a rhetorical fiamework of contrast that organizesmuch of the lesson:what was believedin the past as opposedto what is now known. In creating the contrasthe usesan innovativeevidentialconstructionJo:do: a:la:bo:.This form, usedrepeatedlythroughoutthe text, (for example,15) hasthe meaning'knownfrom this source/not known before,' and marks information that is new, true and only known from the written word. This evidentialdoes not appear in any traditional speechgenresnor is it usedin other relativelyrecentforms of Kaluli discourse,such as translation situationsinvolvingTok Pisin. It is an innovationby Kaluli speakers to mark new information and its new source.Kulu further elaboratesthis theme of contrastingthe past (13 and also 15),'what our fathers(knew) befbre'with the state of knowing in the present,'but now we really know,' and makesthe sourcelexically explicit, buko: wema: walasalab'thisbook shows/instructs.' Note that 'his book' is marked with an ergative/instrumental casemarker, IT is the agent (or instrument) which instructsand shows,(wala sa:lqb'showand speak')and it is by looking at the book (14) and listeningto what the wordsin this book say(16), that understanding or knowledge is obtained.This mode of achievingunderstandingis different from the ways Kaluli usuallylearn, which is by listeningto what many others say,arguing with them, watching them, and being instructedwhile participating (Schieffelin 1990). Source or evidenceis made explicit, as is the sensorymode in which it is offered. The book has been granted an authoritative voice and becomes an authoritativesource.This new evidentialmarker is only used in speaking.It never
Making sense of witten words in Bosavi
appearsin written form. In fact, no evidentialmarkers appear in secular written vernacular texts,a point I will return to. In additionto this innovativeevidentialwhich we will see more of below, other evidentialforms are used to indicate different sourcesof information. In directingthe class'sattentionto the booklet,Kulu (KF) usesthe evidentialmarker /o;b'visuallyevident'to guide studentsto the top of the page,the place that they are to begin readingthe text aloud as a group (27-28).He counts 'one two' in Englishto get them into vocal synchronyin the sameway as when he instructedthe groupto singin the openingof the class(5).
KF] 21 28
## wa:laa:no:o:bo:lo:bo:?taminamilo:o:deyo:agela:bione two what is lt (obv) ort tlrc top (of the page)? read the firstpart, one two ha:gaa:no:agela:bi what'stmdenteatlu read
Studentsslowly read aloud, syllable by syllable, a short text in Kaluli about preventing sicknessin Bosavi.
Markingthe end of the group readingwith the boundarymarker'okay' (29) which is usedin Christianspeechevents,Kulu asksa number of questionswhich do not relateto the readingthat hasjust been done. Instead,they relate to the cover of the literacybooklet which he is holding up and displayingto the class.The cover is simple:The words"Bosavi"and "Malaria"aretyped in small letters at the top of the page.In the center of the page is a black and white drawing of a mosquito (side view),whichis encircled.Placedbelow the drawingand filling the bottom half of the pageis the title written in large bolded capital letters, HAMULE E WAf AF BO:LO:.It is apparentfrom the responseof the studentsthat the referent about whichKulu is speakingis not at all clear.
a: no:o: ba ? wlnt is it?
student]walaf sickness 31 KFI 32
kalu wi o:ba? what is the person'snarne?
Hamule Hamule a:no: Hamule it's Hamule
e o:ba: walaf bo:lo:lo:tro: ] front wlwt did he get sick (obv)?
walafo: o:ba: walaf bo:lo:lo:bo:? sickness,from wlnt did lrc get sick (obv) Z
walato: o:b walaf bo:lo:lo:bo:? sickness,from wlnt did lrc get sick(obv) ?
student] malalia 38 malaia
wa:la buko: a:no: bo:ba wa:la wa:lamilo:a:no: bo:ba look on the front of the book, on the front, look on the front [drawing of mosquito]
student] kiso: 40 mosEtito
a:no: wiyo: o:b salaba?(pointingto title) Hamule e walaf bo:lo: what does the name say? Hamule he got sick
a: no:pik saw e o :b o :l o :b o :? tltispicture, what is lt (obv) I
kiso: mosquito 4Jt tf tf
In lines 30-34Kulu establishesthat what he wantsthe group to fbcus on is the title of the booklet.Once that is established, he attemptsto elicit a particularresponse to his question about the sourceof Hamule's sickness.His attempt to get the class to view the drawing of the mosquitoabovethe title as visualevidenceof the source is accomplishedover severalturns. Using the evidentialmarker Jo:b, Kulu asksthe classthree times (35-37)about Hamule'ssickness, what causedit, what it was.What is visually obvious is the drawing of the mosquitoand after receivingan answer to his third question,Kulu asksthe classto look at the picture on the cover of the booklet (39) for the answer.His assurnptionis that the visual evidenceis obvious. When only one studentanswersklso.''Mosquito,'Kulu explicitlyrefersto the picture (a2). Only then does he get the desiredgroup response.Throughout his talk about the sourceof the sickness, malaria(25-31,42andelswhere),he usesthe evidential
Makingsense of writtenwordsin Bosavi 231 f.ormJo:b to indicatethat the information is visuallyavailable.4 In contrast to this pattern of evidential choice, when Kulu refers to information,or wantsthe classto focus on information that is in the written texts, he shiftsto evidentialsthat mark verbal evidence.In other words, the print and the book are classifiedas speakingsubjects.Printed words do not have the same evidentialstatusas graphic representationsor as somethingvisually evident. In additionKulu marks thesespeakingsubjectswith ergativecasemarking, and uses 'say like that' and Soma'spealdsay'to provide verbsof speakingsuch as a:la:ma evidencefor what is in the text, as well as to give authority to the text, verbal authority.He extendsthis authority to himself at the same time.
## kalu nowo: walaf bo:lo:wamiyo:kalu nowo: walaf madaa:la:sa:ga: no ami dimianka:a:la: a:la:sa:ga: o: walafdo:a:na diya:sa:ga: ba:labamiyo: salabka:wema: reallyafter one man getssick, anotlter man getssicl