This article was downloaded by: [University of Stellenbosch]On: 16 February 2015, At: 11:10Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: MortimerHouse, 37-41 Mortimer Street, London W1T 3JH, UK

Southern African Linguistics and Applied LanguageStudiesPublication details, including instructions for authors and subscription information:http://www.tandfonline.com/loi/rall20

Laying foundations for academic languagecompetence: the effects of storybook reading onZulu language, literacy and discourse developmentDanisile Ntuli & Elizabeth J PretoriusPublished online: 12 Nov 2009.

To cite this article: Danisile Ntuli & Elizabeth J Pretorius (2005) Laying foundations for academic language competence:the effects of storybook reading on Zulu language, literacy and discourse development, Southern African Linguistics andApplied Language Studies, 23:1, 91-109, DOI: 10.2989/16073610509486376

To link to this article: http://dx.doi.org/10.2989/16073610509486376

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) containedin the publications on our platform. However, Taylor & Francis, our agents, and our licensors make norepresentations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose ofthe Content. Any opinions and views expressed in this publication are the opinions and views of the authors,and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be reliedupon and should be independently verified with primary sources of information. Taylor and Francis shallnot be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and otherliabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to orarising out of the use of the Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematicreproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in anyform to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

Copyright © 2005 NISC Pty LtdSOUTHERN AAFRICAN LLINGUISTICSAND AAPPLIED LLANGUAGE SSTUDIES

EISSN 1727–9461

Southern African Linguistics and Applied Language Studies 2005, 23(1): 91–109Printed in South Africa — All rights reserved

Laying foundations for academic language competence: theeffects of storybook reading on Zulu language, literacy and

discourse development

Danisile Ntuli1 and Elizabeth J Pretorius2

1 Department of African Languages, University of South Africa, PO Box 392, Pretoria 0003,South Africa

2 Department of Linguistics, University of South Africa, PO Box 392, Pretoria 0003, South AfricaCorresponding author, e-mail: pretoej@unisa.ac.za

Abstract: Success at school depends heavily on language and literacy skills. Research indicatesthat pre-school children whose parents read storybooks to them have a linguistic and literacy headstart over other children when they start school. In contrast, learners who come to school with fewliteracy skills are at a disadvantage. A Family Literacy project was started in 2000 in disadvantagedareas of rural KwaZulu-Natal. One component of the programme promotes storybook reading inZulu to pre-school children in Grade R. Various aspects of the children’s language and emergentliteracy skills were assessed longitudinally. Children in Grade 1 who were not in the project werealso assessed, and their performance compared to those of the children in Grade R. The resultsshowed that, on the whole, the Grade R children who were in the Family Literacy Project scoredbetter on the literacy tests and showed stronger language and discourse development than theGrade 1 learners, even though the pre-schoolers were on average a year younger than the Grade1 learners. These findings suggest that reading storybooks to pre-school children has beneficialeffects on their language, literacy and discourse development.

IntroductionWhy is it that some children seem to adapt toschool and learn to read and write with relativeease while others struggle with the writtenword? Why do some children seem to find iteasy to follow what their teachers say and theyunderstand what their teachers expect of them,while others find the classroom a strange andconfusing experience? Decades of researchhave given researchers and educationistsgreater insights into the myriad of factors thatinteract in complex ways to facilitate orhandicap the acquisition of literacy andconsequent progress in formal schooling.These factors include individual differencesbetween children, differences in homebackgrounds, the impact of cultural, politicaland socioeconomic variables, the nature andamount of linguistic input that children receive,variations in social interaction and discoursepatterns to which children are exposed, andthe effects of styles of school management,instructional methods and classroompractices.

There is often a mismatch between whatthe public perceives to be important in thelearning context and what researchers havefound to be important. For example, manyparents in South Africa think that it is theresponsibility of the school to teach theirchildren to read and to introduce them to theworld of books, yet there is a body of researchfindings that indicate that waiting for children toacquire literacy when they start school mayalready be too late (e.g. Snow et al., 1991; Buset al., 1995; Neuman, 1999). Research in thefield of literacy development reveals thatchildren’s knowledge about language andliteracy before they start school has an impacton their subsequent literacy accomplishments,not only in the early stages of learning to read,but also for later reading achievement (e.g.Wells, 1985; Dickinson & Smith, 1994; Vivas,1996; Jordan et al., 2000). Children’s earlyknowledge about reading and writing thatdevelops during their pre-school days isreferred to as ‘emergent literacy’ (e.g. Sulzby &

Dow

nloa

ded

by [

Uni

vers

ity o

f St

elle

nbos

ch]

at 1

1:10

16

Febr

uary

201

5

Ntuli and Pretorius92

Teale, 1991; Graves et al., 1998). It includesattitudes, expectations and skills related towritten language and an increasing awarenessof literacy behaviours during the pre-schoolyears. Most of this early knowledge is nottaught explicitly but is acquired unconsciouslythrough the social and verbal interactionpatterns that children engage in with adults intheir immediate contexts.

The purpose of this article is to examine theeffects that storybook reading in Zulu had onthe language and emergent literacy of pre-school Zulu children. The communities fromwhich these children came are poor rural areaswhere there are high levels of illiteracy, povertyand unemployment.

The foundations of academiclanguage competenceThe formal language of schooling is referred toas academic language competence and it is afactor strongly associated with schoolachievement (e.g. Cummins, 2000). This typeof language and literacy competence isacquired primarily through exposure to writtenlanguage, or to oral forms of discourse that areembedded in a culture of literacy, such asclassroom discourse and the discourse relatedto the teaching and learning of specific contentsubjects. As Corson (1997: 684) explains,‘(th)is is a kind of discourse where learners cantalk repeatedly about knowledge gained fromtexts, using an acquired metalanguage setagainst a meaning system used to interpret andextend understanding’.

Let us now consider more closely thelanguage and literacy development that takesplace during the pre-school and early schoolyears, and the way in which this paves the wayfor academic language competence, which inturn mediates successful learning at school.

Due to differences in the context in whichlanguage is acquired and developed and thefunctions that it serves, a distinction is oftendrawn between two kinds of languageproficiency that are relevant to the learningcontext, namely Basic InterpersonalCommunicative Skills (BICS) and CognitiveAcademic Language Proficiency (CALP)(Cummins, 1991; 2000). These two types ofproficiency are associated with oral and writtenmodes of language use respectively. Analysesof spoken and written discourse reveal features

characteristic of one or the other mode. Forexample, BICS is used in everydaycommunicative encounters and is described asbeing more context-embedded, in the sensethat it contains many deictic or indexical items,the meaning of which can be recovered fromthe interactional context. Oral discourse alsomakes much use of prosody (pitch, stress,intonation and tone in the case of Africanlanguages) and paralinguistic features(gestures, expressions) to convey meaning.Because oral discourse is essentially dialogic,meaning can be negotiated in the course ofinteraction. In other words, oral modes ofdiscourse are characterised by contextualisedparticipant interaction.

A CALP type of proficiency, on the otherhand, involves the use of a more context-reduced language associated with writtenlanguage and with the more formal aspects ofclassroom and lecture-type language usewhich are typical of the learning context. Thisdoes not mean that written language is context-free, since no language use is context-free;rather, that the locus of meaning is typically‘built into’ the text to a larger extent than is thecase in oral discourse. For example, in writtendiscourse, referents are usually identified andcontextual information is lexicalised, whereasthe shared context of spoken discourse makessuch features redundant and henceunnecessary to state explicitly. Consider, forexample, the following two stretches ofdiscourse concerning a car accident:(a)‘Well, this car suddenly came along and hit

us bang, right here. I swerved like this but itwas too late and we went straight into thatpole over there’.

(b)‘My friend, Sarah, and I were travelling homewhen a car suddenly appeared from the rightand hit us in the driver’s door. I swervedsharply to the left but it was too late and wewent straight into the yield sign on thepavement’.

In (a) the oral discourse does not pose aproblem for the listener, since the latter isprivileged to share the same interactionalcontext as the speaker. The meanings ofindexical items such as this car, us/I/we, righthere, like this, that pole over there are clarifiedby gestures, intonational patterns and theimmediacy of the context at the time ofspeaking. In order to communicate this event in

Dow

nloa

ded

by [

Uni

vers

ity o

f St

elle

nbos

ch]

at 1

1:10

16

Febr

uary

201

5

Southern African Linguistics and Applied Language Studies 2005, 23: 91–109 93

writing (b) to someone displaced in time andspace from the context (e.g. in a letter to aparent two days later), the context needs to belexicalised (right here becomes in the driver’sdoor; like this becomes sharply to the left; thatpole over there becomes the yield sign on thepavement) and referents are identified (e.g. weis My friend, Sarah, and I — the writer of theletter).

One of the characteristics, inter alia, ofacademic language proficiency is the ability toshift the locus of meaning more explicitly to thelanguage itself, to build the context into thediscourse (in spoken language) or text (inwritten language). Exposure to storybookreading and the ‘book language’ of writtenstories provides children with opportunities toacquire this more ‘context-reduced’ type oflanguage use.

The role of storybook reading inlanguage and literacy developmentin the early yearsDuring the pre-school phase, children masterbasic grammatical structures and oraldiscourse skills in their primary language fromthe language input they receive around them.By the time they start school, most childrenhave already acquired fairly complex orallanguage abilities (e.g.Torrance & Olson, 1985;Garton & Pratt, 1989). Oral language skills formthe basis on which early reading is built. Whenlearning to read, children bring their knowledgeof oral language to bear on written language.Research indicates that children who startschool with poor language skills and poorunderstanding of the communicative processoften have problems learning to read (e.g.Mace-Matluck et al., 1989). However, althoughoral language skills form the basis on whichearly reading is built, oral language proficiencydoes not predict school achievement or readingskill in the long run (Saville-Troike, 1984; Wells,1986; Snow & Dickinson, 1991). What kinds ofknowledge and skills facilitate learning to readand write, and how do children cross the bridgefrom oral language to literate modes ofdiscourse? In effect, what factors do predictschool achievement in the long run? One factorthat regularly emerges as playing a central rolein the development of language and literacyskills is storybook reading. It is this factor whichis of concern in this study.

Research throughout the world has shownthat children who have storybooks read tothem during the pre-school years have a headstart over other children when they startschool, and they achieve and maintaineducational success more easily than childrenwho do not start off with this advantage (Wells,1986; Hiebert, 1993; Dombey & Moustafa,1998). Why is it that exposure to storybookreading contributes so meaningfully to theearly development of literacy accomplish-ments? A closer examination of the kinds ofadvantages that storybook reading confers onchildren reveals that, through having storiesread to them on a regular basis, children showadvanced language skills and gains inconceptual and vocabulary knowledge (e.g.Krashen, 1993; Neuman, 1999; Buchorn-Stoll,2002). Not only do they extend theirknowledge of the world through storybooks,but they also acquire strong emergent literacyskills and develop high expectations aboutprint. For example, they know how to handlebooks, how to turn pages, they acquire asense of narrative structure (also referred to asstory schema) and their visual literacydevelops (they can construct meaning frompictures). Children exposed to storybooks tendto have longer attention spans; they can sitthrough a storybook reading session withoutfidgetting (e.g. Feitelstein et al., 1986). Payingattention to the task at hand is a skill thatstands children in good stead when they startformal school and have to attend to what theirteachers say. Children exposed to storybooksfrom an early age also learn to read and writerelatively easily (Harris & Sipay, 1975; Butler,1979; Holdaway, 1979; Elley, 1989). In theInternational Studies in EducationalAchievement (IEA) study of reading literacyamongst 8- and 14-year old children in 32countries around the world, it was found that inevery country, better readers had engaged inliteracy activities before school, such asstorybook reading, and had access tostorybooks in their homes (Elley, 1994: 147).

In sum, exposure to storybook reading froman early age lays the foundations for thedevelopment of academic languageproficiency. In contrast, learners who come toschool without any or with very few emergentliteracy skills and who are unfamiliar with ‘booklanguage’ and the way that books work are at a

Dow

nloa

ded

by [

Uni

vers

ity o

f St

elle

nbos

ch]

at 1

1:10

16

Febr

uary

201

5

Ntuli and Pretorius94

disadvantage when they start school. Theyusually take longer to learn to read and write,they are unfamiliar with the formal language ofschooling and they often have difficulty settlinginto the classroom routine.

In South Africa, many children fromsocioeconomically disadvantaged areas comefrom homes where there are few or no books,and many parents have low literacy levels.Many of these children also attend pre-schoolcentres where the focus is on keeping thechildren out of harm’s way and attending tobasic physical needs. Although the Departmentof Social Development provides a frameworkfor the standards for pre-schools and crèches(Guidelines for Day Care), there are noprerequisite qualifications for people runningthe pre-schools. In addition it is difficult for pre-schools to obtain government financial support.As a result many caregivers at such centreshave little or no training in early childdevelopment (ECD) and minimal educationalfacilities are provided. The intellectualstimulation of these children is often not aprimary concern for many caregivers. Althoughteaching children the letters of the alphabet andnumber counting are favoured activities in suchcentres, they are typically done in a formal,strictly sequenced and rote choral manner. Bythe time these children enter Grade 1 they havehad little or no exposure to the literate practicesand rich language experiences taken forgranted in more privileged areas, where thereare better qualified caregivers and moreplentiful educational resources.

Once they enter the formal schoolingsystem, these learners have few opportunitiesto extend and enrich the scope of theirlanguage and literacy experiences, due largelyto the lack of reading materials in poorly-resourced schools. Furthermore, the way inwhich African languages have been taught inthe past and in fact are still being taught inSouth Africa has often been criticised on thegrounds that it does not develop academiclanguage competence (e.g. Macdonald, 1990;Smyth, 2002). There tends to be a strong biastowards a structural grammar-based approachwith very little attention given to developingacademic language competencies in theAfrican languages. Learners are not exposed toextended discourse, and extensive reading inthe African languages has not been part of

classroom practice. As a result, learners do notdevelop foundational linguistic, cognitive andconceptual skills in their home languages.Smyth (2002) argues that in order for Africanlanguages to become vehicles for meaningfullearning, the teaching of these home languagesmust articulate with the rest of the curriculumso that they can contribute to the building ofliteracy skills. In this way they can preparelearners intellectually for the later demands ofswitching over to English as the language oflearning and teaching (LoLT):

Research conducted in SA ... points to thefact that, for the majority of school children,academic language proficiency in theirhome language is not sufficiently wellestablished for them to be able to transferideas to another language except at themost mundane level. Once these learnersare forced to learn through a languagewhich is not their home language, theirconcept and language development suffersin many ways. This is exacerbated byfactors such as the dominance of thetransmission mode of teaching and socialand affective factors such as low self-esteem in learners and the low status oftheir home languages (Smyth, 2002: 109).

As indicated above, Smyth (2002) suggeststhat mother-tongue African language coursesneed to run ahead of the curriculum so thatlanguage, concepts and cognitive skills areestablished in the home language before beingdealt with in the LoLT. These courses shouldassist learners in moving from using orallanguage and the forms and structuresassociated with it to dealing with more complexwritten language.

From the above it is clear that morechallenging input is required to extend the range,complexity and richness of the children’slanguage and literate experiences. Many childrenfrom disadvantaged areas have little exposure torichness of input in the pre-schools they attend,and even when they enter the formal schoolingsystem, the kind of linguistic and conceptual inputthat is needed to develop academic languagecompetence is missing. Even the rich African oralculture of storytelling is falling into disuse,especially in urban areas. Machet (2002: 5)describes this situation as follows:

Parents are too busy and too tired in theevenings to spend time telling young

Dow

nloa

ded

by [

Uni

vers

ity o

f St

elle

nbos

ch]

at 1

1:10

16

Febr

uary

201

5

Southern African Linguistics and Applied Language Studies 2005, 23: 91–109 95

children stories ... Also, many parents feelthat their oral culture of storytelling doesnot have value today in today’s highlytechnological world. Thus children aredeprived of any form of storytelling. Thishas a serious effect as children start schoolwithout any story schema (i.e. theconventional way in which a story isstructured within a culture).

Storytelling sessions stimulate theimagination and provide children with a richsource of oral language input. From stories thatare told to them or from storybooks read tothem, children also unconsciously acquire thediscourse skills of how to tell a story, the notionof a story schema (i.e. the ‘form’ of a story) andhow stories function in their particular culture.Some children’s television programmes, suchas Takalani Sesame and Yo TV, haveintroduced stimulating storytelling sessions thatcan help bridge the gap for many children.However, there are still many children who donot have access to television, especially inhigh-poverty areas and remote rural areas.

Developing storytelling competenceThe ability to tell a story is part of languagecompetence, often referred to as discoursecompetence. Let us now briefly consider someaspects of the development of storytellingcompetence. Much of this research has beendone in English and European languages, andso may reflect a particular cultural-linguisticperspective.

From the age of two children start ‘tellingstories’. There are two main types of storiesthat children learn to recount, namely personalevent narratives (i.e. telling about a specificevent that happened) and fictional stories (i.e. afictional story derived from an oral or book storyor story the child made up) (Allen et al., 1994).Children use their knowledge of the world andtheir experiences to convey content knowledge(i.e. what happened) as well as linguisticstructure knowledge to recount the story (i.e.how information is organised and conveyed).Allen et al. (1994) argue that personal eventnarratives and fictional stories are related yetdistinct narrative genres. Their research showsthat the underlying structures for personalevents and stories follow different developmentpaths. A simple story, for example, may reflecta series of events that follow one another in a

coordinate way (and ... and then ... and ...),whereas a more complex story contains anelaborated plot with embedded and/orinteractive episodes.

From a linguistic point of view, at an earlystage English children first tend to describeevents in their order of occurrence, with asentence for each event or a series of clausesjoined by additive conjunctives such as and orand then. Children then proceed to a stagewhere more complex sentence structures areused, with the main clause appearing at thebeginning of the complex sentence. Laterchildren learn to use subordinate clauses in firstposition (Romaine, 1984). Initially additive andtemporal relations between discourse units areused, but with increasing age, causal andadversative relations are used. From the age ofabout eight onwards children mark relationsbetween clauses more explicitly through theuse of conjunctives. Although thisdevelopmental sequence is well documented inEnglish (e.g. McCabe & Peterson 1985; Raban,1988; Cox et al., 1990), there has been verylittle research in this domain in the Africanlanguages.

The norms of discourse and storytellingdiffer not only culturally but also socially withincultures, in terms of class or communitydifferences. The research by Allen et al. (1994)showed that children with more advancedproductive language abilities produced storieswith more distinctive narrative structurescharacteristic of the personal event andfictional narrative genres (1994: 169). In herlongitudinal study, Heath (1983) showed howchildren are helped in their transition fromhome to school forms of discourse viaexposure to storytelling or storybook reading,as well as by the form, function and content thatstories are assumed to have by differentcommunities. The style of school or classroomdiscourse is closely related to the norms ofwritten language, and the acquisition of thenorms of written language, in turn, starts withexposure to storybook reading.

Book-based activities not only affectchildren’s productive language and literacydevelopment, but also their receptive languagedevelopment. For example, in assessing theimpact of literate aspects of homeenvironments on comprehension strategies,Reeder and Shapiro’s (1993) study found that

Dow

nloa

ded

by [

Uni

vers

ity o

f St

elle

nbos

ch]

at 1

1:10

16

Febr

uary

201

5

Ntuli and Pretorius96

children who had been more stronglysocialised into literate behaviours tended tolocate meaning within linguistic units ratherthan the surrounding context. They showed astronger bias towards linguistically-dependentstrategies for comprehending speech actswhen input was distorted, whereas the otherchildren relied more on the context to helpthem comprehend speech acts. Theresearchers conclude that early literateexperiences bring about a shift incomprehension strategies where the locus ofmeaning shifts increasingly from context-based to linguistic-based comprehensionstrategies.

There is a large body of literature thatprovides evidence of the overt as well as moresubtle ways in which child-oriented book-basedactivities affect children’s language anddiscourse development, their backgroundknowledge, their knowledge of how storiesfunction as well as the way in which theirliteracy develops. Attitudes towards readingcan begin in infancy and can be enhancedthroughout childhood. Parents are usually thefirst and primary influence, with caregivers andteachers later extending what parents havebegun. When stories are read to children on aregular basis, not only does their love for booksand reading grow but their confidence andability to narrate stories increase. Childrenexposed to more literate modes of discoursedevelop language competence characterisedby more decontextualised or disembeddeddiscourse, whereas children who are onlyexposed to oral modes of discourse developlanguage competence characterised bycontextualised participant interaction.

These are issues that have been wellresearched in studies on English language anddiscourse acquisition and literacy development,but there is very little local research on thesetopics in African languages. In this article welook at the effects of storybook reading on theZulu language, literacy and discoursedevelopment of a group of pre-school childrenin a crèche in rural KwaZulu-Natal. In order tocontextualise the study, an overview is firstgiven of the Family Literacy Project of whichthis study was a component. This is followed bymethodological details concerning theassessment of language and literacydevelopment and the results of the study. The

article concludes with a discussion of thefindings and the implications that follow fromthe findings.

MethodologyBefore moving on to the methodological details,a sketch is first given of the larger project ofwhich this study was a part. Thereafter, theparticipants, procedures and analyticframework used to investigate the Zulu literacyand language development of the children aredescribed, and the preliminary resultspresented. The analysis of the data includesformal aspects of linguistic development as wellas features of discourse and emergent literacydevelopment. Some aspects of this larger studyhave already been written up (Pretorius, 2003),where the focus was primarily on emergentliteracy skills. In contrast, the present articlefocusses attention on the language andstorytelling abilities of two groups of children,one of which was exposed to storybook readingin Zulu.

Context of research: the FamilyLiteracy ProjectIn 2000 a Family Literacy Project (FLP) wasstarted in disadvantaged areas of ruralKwaZulu-Natal1. There are several componentsthat make up the FLP; for example, it providesadult literacy classes, it trains and developsliteracy facilitators, it encourages family literacythrough its adult literacy classes, it promotesliterate activities through its child-to-childgroups, it makes books for babies available,and it puts out a community newsletter(Labuschagne, 2001; Labuschagne, 2002).

One of the aims of the programme is topromote storybook reading in Zulu to pre-school children. Caregivers (parents, oldersiblings and pre-school teachers) at threedisadvantaged crèches that fell within the ambitof the FLP were encouraged to read storybooksto their children on a regular basis, and start-uplibraries of age-appropriate storybooks in Zuluwere put in the sites. In 2001 the children atthese three crèches were assessed todetermine whether the book-based activitieswere having an effect on their language andemergent literacy skills. The three caregivers ateach of the crèches were undergoing earlychild development (ECD) training. They wereencouraged to adopt a more child-centred

Dow

nloa

ded

by [

Uni

vers

ity o

f St

elle

nbos

ch]

at 1

1:10

16

Febr

uary

201

5

Southern African Linguistics and Applied Language Studies 2005, 23: 91–109 97

approach at their crèches, to incorporate morestimulating literacy activities into their dailyroutines, and to read storybooks to the childrenin their care. Some of the mothers andgrandmothers of children at each of the pre-schools attended adult literacy classes andwere made aware of the importance of readingstorybooks to their children. One of their‘homework’ literacy activities included readingstorybooks to their children on a regular basis.How effective were these storybook readingsessions? The methodological aspects of thestudy that sought an answer to this researchproblem are now addressed.

ParticipantsThe children in the storybook readingprogramme were young Zulu children whoattended three different crèches, namely Malindi(n = 32), Mbandi (n = 26) and Fundisa (n = 46)(real names have been changed to protect theiridentity). The crèches accommodate childrenfrom the ages of two to seven years. Forlogistical reasons, not all the children at thesecrèches were assessed. In total, 26 childrenbetween the ages of five and seven from allthree crèches were assessed twice during thecourse of 2001 — first in April and then again inNovember. These pre-school children willhenceforth be referred to as the Grade Rs.

Grade 1 learners selected from the nearbyprimary schools were also assessed in terms ofthe same language and literacy criteria used forthe Grade Rs. The Grade 1 learners constituteda baseline against which the emergent literacyskills of the pre-school children were compared.The criterion for the selection of these Grade 1children was that, although they may haveattended pre-school before they started school,they had not been exposed to any storybookreading. It was therefore felt that the emergentliteracy skills that they brought with them at thestart of school would be representative of theschool entry-level literacy skills characteristic ofchildren in the broader community. The sameassessment procedures were followed with theGrade 1 children, although they were onlytested once, in February 2001. It is important tonote that both the Grade R and the Grade 1children came from high-poverty areas andattended disadvantaged schools.

The linguistic data of children from theMbandi crèche and the corresponding baseline

Grade 1 learners from the nearby primaryschool were subjected to an in-depth analysis.In this article only the data from these twogroups of children will be presented.

ProceduresVarious aspects of the Grade R children’slanguage and emergent literacy skills wereassessed at the start of the programme, early inthe year (these formed the pre-tests). Thesame children were assessed again eightmonths later (these formed the post-tests). Dueto the length of the intervening time, there wereunlikely to be memory effects, so the sametests that were used in the pre-tests were usedin the post-tests (cf Pretorius, 2003, for furtherdetails about the nature of these emergentliteracy assessments).

All the children were tested on an individualbasis in Zulu by mother-tongue speakers ofZulu. Two of the facilitators for the adult literacyclasses were trained to assess all the children.It was also their responsibility to transcribe thetape-recorded protocols in Zulu2.

Assessment measuresAlthough a range of tasks was designed toestablish what skills and knowledge the GradeR learners had acquired in terms of emergentliteracy behaviours, only those measures thatrelate directly to the children’s language anddiscourse development are described here.The relevant assessment measures includedthe following components:• Story recall: this task assessed the child’s

ability to retell a story that had been read tothe children as a group. Performance in thistask reflects not only a child’s comprehensionof the story but also his/her familiarity withnarrative schemata. The story, called uBolekile, is about a boywho unsuccessfully asks various groups ofchildren for different things (e.g. a wheel,sweets and porridge). At the end hisgrandfather says Bolekile’s requests wereturned down because he forgot to ask forthings politely. Although different languages encoderequests in different ways, linguistically andpragmatically (e.g. with gestures and bodylanguage), politeness is a universalphenomenon. Children are socialised froman early age to be polite and, since the

Dow

nloa

ded

by [

Uni

vers

ity o

f St

elle

nbos

ch]

at 1

1:10

16

Febr

uary

201

5

Ntuli and Pretorius98

pictures in the book depicted black children inrural settings, it was thought that the storywould provide a familiar frame of reference. The main events in the story were identifiedand the main items in each child’s recallcompared to this template. This task wastape-recorded. In order to contextualise thestory for the storyreading session, each childwas asked to read the story of uBolekile to ablack doll called Thabo, to cheer him up sincehe was not feeling well. To avoid testingmemory during the story recall, the child wasgiven the book to page through and to showThabo during the ‘reading’ and recall task.

• Free storytelling: the child was asked to talkabout a personal experience, for example,his/her first days at the pre-school. This taskwas tape-recorded and provided data onpersonal event stories.

• Book behaviour: the facilitator observed thechildren’s familiarity with storybooks, whetherthey held the book correctly (the right wayup), turned the pages in the appropriatedirection, could identify a page number and aword on a page, etc. These behaviours werenoted on an observation sheet and apercentage score was computed for eachchild.

• Language development: the data in the storyrecall and free storytelling tasks providedinformation on the child’s languagedevelopment. For each story protocol thefollowing aspects were noted:• Mean length of utterance. Beause the

African languages are richly agglutinatinglanguages, the morpheme was used as thebasic unit for determining utterance length.Thus, a word such as ngahamba consistsof three morphemes: nga+hamb+a.

• Counts of nouns and verbs. The number ofverbs and nouns (repetitions not included)that the child used in each protocol wascounted to get an idea of the child’sconceptual ‘richness’ of language use.

• The use of past and present tenses wasnoted in the protocols, as well as specificsyntactic structures.

The data were captured on computer, usingthe statistical package SPSS.

ResultsThe main question that is addressed in thisarticle is: did storybook reading have an effect

on the language, literacy and discoursedevelopment of the children? In order to arriveat answers to this question we will first considerthe data from a quantitative perspective, interms of morphosyntactic features and specificemergent literacy behaviours. Thereafter wewill examine the data from a qualitativeperspective, in terms of narrative and discoursefeatures.

Language and literacy developmentA set of five measures was used here to assessthe language and literacy skills of the learners:• Each child’s mean length of utterance was

measured in terms of the number ofmorphemes reflected in the utterances in thetwo recorded protocols (the personal eventnarrative of the first day at school, as well asthe recall of the story read to the group as awhole). These are given as raw scores.

• The verb frequency and noun frequencymeasures reflect the average number ofdifferent verbs or nouns used in theprotocols. Thus, if a child said ngangihambaand later used ngahamba, the verb -hambawould be counted once. These verb andnoun frequencies are also given as rawscores.

• For the story recall measure, a template wasmade of the main events in the story and thenumber of main events recalled by the childwas measured against this template andconverted to a percentage score.

• A set of features was also compiled for thecategory book behaviour (e.g. child holds thebook the right way up, child turns pages inright direction, can identify words frompictures, etc.) and a percentage wascomputed for each child, based on observat-ions of their behaviour when handling books.

These five measures of language andliteracy development are shown in Table 1. Thesecond column in the table reflects the pre-testand post-test changes in the performance ofthe Grade R children on these five measures,while the last column reflects the school entrylevel performance of the Grade 1 children onthe same measures. It should be noted that themean length of utterance measure wascomputed on the basis of the number ofchildren in each group who actually producedutterances. For example, in the Group R pre-test for story recall, only five children recalled

Dow

nloa

ded

by [

Uni

vers

ity o

f St

elle

nbos

ch]

at 1

1:10

16

Febr

uary

201

5

Southern African Linguistics and Applied Language Studies 2005, 23: 91–109 99

something of the story while they all (n = 6)produced story recalls in the post-test; incontrast, only three of the six Grade 1 childrenproduced story recalls.

As can be seen from Table 1, the Grade Rchildren showed improvements in all themeasures from the pre-tests in April 2001 to thepost-tests seven months later in November2001. In the recall protocols, the childrenproduced longer recalls, their utterances wereon average longer and their vocabulary wasmore varied. They also remembered far moreof the main events of the story that had beenread to them, and observations of their bookbehaviour also showed greater familiarity withstorybooks.

At the beginning of the year the children’slack of familiarity with storybook reading wasnoticeable. Their classroom routine did notinclude sitting on the floor around an adult tofollow the pictures and story in a book. Therewere several children who were very shy andgave minimal responses or did not say anythingwhen asked to retell the story. The presence ofthe tape recorder might have inhibited them inperforming this task, but even when thefacilitator encouraged them to page through thebook and asked them questions, they were stillreluctant to say anything. Reluctance toperform a task could be taken to indicateinability to perform the task. It is rare forchildren who are accustomed to storybookreadings and encouraged to talk about thestory afterwards to display this kind ofresponse. By November, they were far more atease with storybook-reading activities and their

recalls reflected longer and more accuraterenditions of the original story.

Similarly, it was interesting to notequalitative changes in the Grade R children’sbook behaviour. Initially it was clear fromclassroom observations that the children hadhad very little experience with books; they heldthe books clumsily, turned pages awkwardly,and were unsure which things on the pagessignified meaning. Some children did not showmuch interest in the books, and some were noteven aware that they were looking at the booksupside down. By November there was amarked change in their handling of books.There was greater spontaneity when they saton the floor and paged through books, whichwere held the right way up. Some of thechildren sat in pairs and commented on thepictures or ‘read’ the story aloud to themselves.They also put books back in the bookshelfwhen they had finished reading them.

In contrast, although the Grade 1 childrenhad been at school for two months and were ayear older, their recall protocols were shorter,they showed slightly less variation invocabulary use, and they revealed poor bookbehaviours and story recalls. As can be seenfrom Table 1, by the end of their pre-schoolyear, the Grade R children had outperformedthe Grade 1 learners in all five of thequantitative measures, even though they werea year younger than their school-going peers.

Discourse developmentIn this section we look more closely at possiblechanges that occurred in the discourse of the

Table 1: Differences between Grade Rs (in programme) and Grade 1s (not in programme)

Grade R Grade 1pre-test post-test (baseline)

Mean age 5.8 6.8Mean length of utterance

Personal event 15.5 30.5 20.7Story recall 73 95.5 89

Mean verb frequencyPersonal event 2.7 4.7 2.7Story recall 6 6.8 6

Mean noun frequencyPersonal event 2.5 3.7 3.5Story recall 4.8 6 4.3

Story recall (%) 50 61 25Book behaviour (%) 40 63 20

Dow

nloa

ded

by [

Uni

vers

ity o

f St

elle

nbos

ch]

at 1

1:10

16

Febr

uary

201

5

Ntuli and Pretorius100

children, as reflected in the recorded protocols,with regard to their personal event telling andtheir recounting of the fictional story (retold withpicture stimuli).

Free story recall: personal event-tellingIn order to examine their personal event-tellingabilities, the children were asked to tell abouttheir first day at the crèche (Table 2). The mostobvious change between the pre-test and post-test protocols was the length of the stories. Twoof the six Grade R children did not say anythingin the pre-tests. All the children were shy andthe kinds of activities they typically did at thecrèche did not require them to engage inextended discourse such as telling a story.They seldom recounted more than two events(as seen in Bonga’s pre-test example), and theevents were usually presented in the order ofoccurrence. Thandeka was the only child whowas quite talkative. She used the past tenseand simple sentences with six different verbs torecount the events and although there wassome confusion as to the sequence of events

(Ngabhala, ngahamba was followed by furtheractivities), she mentioned at least nine events.She also provided closure at the end of herstory — kuphela.

By November, the personal event narrativeswere longer, with at least five to six eventsrecalled. The event sequences reflected realorder of occurrence. In her post-test story, forexample, Thandeka shows use of a widerrange of grammatical structures. Whereas inthe pre-test she said sabukela izincwadi, ‘welooked through books’, in the post-test she saiduMiss wasifundisa ngezincwadi zale khreshi,‘Miss taught us by means of the books from thiscrèche’. The noun izincwadi, ‘books’, was nowused as an adverb ngezincwadi, ‘by means ofbooks’. She also used the possessiveconstruction zale khreshi to refer to the crèche(in the pre-test nothing was mentioned of thecrèche).

In contrast, the Grade 1 event narrativeswere similar to the early Grade R ones at pre-test time, consisting mainly of two-eventprotocols, despite the age differences between

Table 2: Grade R pre- and post- personal event narratives

Grade R — My first day at the crèchePre-testsBongaNgangihamba nosisi. Ngafunda

[I was walking with my sister. I read (past).]

Noxolo–

She does not say anything.

ThandekaNgalethwa ubaba nomama. Ngadla irayisinobhontshisi nophuthu nemifino. Ngabhalangahamba. Ngasika izithombe, ngasika amakhanda,sabukela izincwadi, sabhala amagama ethu, sasikaamagama ethu — kuphela.[I was brought by mom and dad. I ate rice and beansand phuthu (pap) and spinach. I wrote (and) left. I cutout pictures, I cut out heads, we looked throughbooks, we wrote our names, we cut out our names —that’s all (past).]

Post-tests

Ngangihamba noMama, ngabuye ngahambangedwa. Ngadla isinkwa nesobho. Ngabuya ngedwa.[I was walking with my mother, I also walked alone. Iate bread and gravy. I came back home alone (past).]

Ngafika noSindi. Safunda amagama, sasikaizithombe. Sadlala amapholi, sabuyela emakhaya.[I arrived with Sindi. We learnt words, we cut outpictures. We played amapholi, (and) went back home(past).]

Ngangihamba noMazuzu kaGogo MaNgcobo.Safunda noMiss, ses’ Thani. Safunda izincwadi.UMiss wasifundisa ngezincwadi zale khreshi. Sathisesiqeda sahamba nezingane zale khreshi.

[I was walking with Mazuzu of Gogo MaNgcobo. Weread with Miss, Sister Thani. We read books. Misstaught us by means of the books from this crèche. Oncompletion, we left with children from this crèche.]

Dow

nloa

ded

by [

Uni

vers

ity o

f St

elle

nbos

ch]

at 1

1:10

16

Febr

uary

201

5

Southern African Linguistics and Applied Language Studies 2005, 23: 91–109 101

the children. Some samples of Grade 1 eventnarratives are shown in Table 3.

On reflecting on these stories, there seemto be three basic episodes that frame this typeof personal event narrative, namely:identification of the person who accompaniedthe child to school, description of a significantevent (or events) that happened at school, andsome kind of closure at the end, such asreturning home. While the second episode wasinvariably included in the recall, the first andthird seemed to be optional. The inclusion of allthree basic episodes resulted in a morecoherent recall. It is noteworthy that, while theGrade Rs usually included all three episodes intheir post-test recalls, the Grade 1 recalls weresparser, with the second episode flanked byeither a first or third episode, thus creating theeffect of less coherent, less ‘rounded off’recalls.

Affect is also an integral part of storytelling,since what we tell in personal stories concernsevents or states that have significance for us.Children’s narratives reflect the representationof things that matter to the children. It is thusinteresting to note the differences in the contentof the events depicted in their little stories. Theevents that the Grade R children described intheir post-test recalls reflected several literateactivities (safunda amagama, sasika izithombe,‘we learnt words, we cut out pictures’; safundaizincwadi, ‘we read books’) whereas this wasnot the case with the Grade 1s. In fact, littlelearning seems to have taken place at school,and some of them seem to have had ratherunhappy experiences on their first day atschool! In contrast, Thandeka’s post-test recall

(Grade R, above in Table 2) in particularemphasises the book-based activities, more sothan her pre-test, suggesting that theseactivities featured quite prominently in the pre-school children’s routine at their crèche.

Storybook recall: recounting a fictionalnarrativeThe story that was read to the childrenconcerned a boy called Bolekile, whounsuccessfully asked different children forthings and at the end his grandfather remindedhim to say ‘please’ when making requests. Toavoid memory effects, the children were giventhe book when asked to recall the story, on thepretext that they were reading the story to thedoll, Thabo, who was sick.

The Grade Rs were noticeably moreconfident when retelling the story in the post-tests than they had been in the pre-tests (Table4). This suggested greater familiarity withstorybooks and talking about stories. On thewhole, the recalls were longer in the post-tests,and they also contained greater elaboration ofevents.

Bonga was the shyest child in the Grade Rgroup and also the one whose progress duringthe year was the slowest. It is interesting tonote that Bonga’s mother did not attend theadult family literacy classes, so the onlystorybook reading to which he was exposedwas that of the teacher reading storybooksduring pre-school time. He said nothing in thepre-test story recall, but in the post-test,although unable to recall the story accurately orin detail, he did proffer some information. Heincluded the negative form of the copula

Table 3: Grade 1 personal event narratives

Grade 1 — My first day at schoolSanelisiweNgangihamba noSindi noNdumo noMpume. Sanikwa izitifiketi.[I was walking with Sindi and Ndumo and Mpume. We were given certificates (past).]

SimoWayengishaya uMawi, wangilamulela uMondli, wase engiyeka. Ngahamba.[Mawi hit me, (and) Mondli helped me, and she let go of me. I left (past).]

SisekeloKwakumnandi, akekho owangishaya.[It was nice, no-one hit me (past).]

Dow

nloa

ded

by [

Uni

vers

ity o

f St

elle

nbos

ch]

at 1

1:10

16

Febr

uary

201

5

Ntuli and Pretorius102

Table 4: Grade R pre- and post-test recalls of fictional story

Grade R Pre-testsBongaPages through the book tentatively, but does not sayanything.

NkululekoUyaboleka induku, uyala, uboleka umdokwe, uyala,uboleka uswidi, uyala. [He is borrowing a stick, he refuses, he is borrowingporridge, he refuses, he is borrowing a sweet, herefuses (present).]

NoxoloSengimkhohliwe ukuthi ubani lo. Abavumi ukumnikainduku lo. Uthi uBolekile, “Ngicela ningipheumdokwe”. Abavumi, bathi nabo bawuphiwe. UthiuBolekile “Ngicela isondo”. Umfana akavumi, uthinaye uliphiwe ubaba. Uthi uBolekile, “ningipheuswidi”. Abavumi, bathi nabo bawuthengile.UBolekile nangu useyisono lana. Uthi umkhulukuBolekile, “Uhlulwe yinto eyodwa”. UBolekile lauseyisono. [I have forgotten who this is. They are refusing to givehim a stick. Bolekile says, “Could you please give mesoft porridge?” They are refusing, they say that it wasalso given to them. Bolekile says, “Can I please havethe wheel?” The boy refuses, he says that he too wasgiven it by his father. Bolekile says “Can you pleasegive me a sweet?” They refuse, they say that they toohave bought it. Here is Bolekile looking worried now.Grandfather says to Bolekile “You were unable tomention one thing”. Bolekile is looking worried now.]

ThobekaNangu uBolekile usehleli lana. Usethi uBolekile,“Ngicela ningiboleka induku”. Bathi “Hhayi, nathisizithole ehlathini”. Wathi uBolekile “Ngisacelaungiboleke isondo”. Bese wathi “Hhayi, namingilitholile”. Bese wathi uBolekile “Awungipheuswidi”. Wathi “Hhayi, nami ngiwuthenge esitolo”.Bese uBolekile useyisono. Bese umkhuluuseyakhuluma, bese umkhulu nomama, umkhulu

Post-tests

UBolekile wayeboleka isondo. Umkhulu wathiakasenalutho, akasenaswidi, akasenanduku.Iyaphela.[Bolekile borrowed a wheel. Grandfather said that hedidn’t have a wheel, he didn’t have a sweet, he didn’thave a stick any more. It’s the end (of the book).]

Uthi uBolekile, “Ngiboleke izinduku”. “Ngeke,sizithole ehlathini”. “Ngiphe umdoko”. “Ngeke,siwuthole kumama”. “Ngiboleke isondo”.“Ngeke, ngilithole kubaba”. “Anginaswidi,anginamdoko, anginasondo, anginanduku”. Uthiumkhulu uBolekile ukhohlwe igama elibalulekileelithi, “ngiyacela”.[Bolekile says “Lend me sticks”. “No way, we gotthem from the forest”. “Give me soft porridge”. “Noway, we got it from Mother”. “Lend me the wheel”.“No way, I got it from Father”. “I do not have a sweet,I do not have porridge, I do not have a wheel, I do nothave a stick”. Grandfather says Bolekile forgot theimportant word which says “please” (present — he isnarrating as if the actions are happening now).]

UNobuhle wayeboleka izinto. Wathi umkhulu iyodwainto ekade kufanele ayisho. UNobuhle uthiAnginanduku, anginamdokwe, anginaswidi. Namiawungiphe uswidi. Wathi angimboleke isondo. “Namingilinikwe ubaba”. “Awungiphe umdoko”. “Nathisiwuphiwe umama”. Wathi “Aningiboleke izinduku”.Bathi “Nathi sizicoshe ehlathini”.

[Nobuhle used to borrow things. Grandfather saidthere is one thing that he was supposed to have said(past). Nobuhle says “I do not have a stick, I do nothave soft porridge, I do not have a sweet”. “Pleasewould you also give me a sweet”. He said that Ishould lend her the wheel. “I was also given it by myfather”. “Give me porridge”. “Mom also gave it to us”.He said “Lend (please) me ( some) sticks”. They said“We also picked them up in the forest” (past).]

Lona uBolekile. Wayethanda ukutsheleka.Wayetsheleke izinduku zabafana. Bathi abafana“Hhayi, sizithole ehlathini”. Uboleka umdokowamantombazana. Athi amantombazana “Hhayi,siwuphiwe umama”. Waboleka isondo labafana.Bathi “Hhayi, siliphiwe ubaba”. Waboleka uswidikumama. “Hhayi, ngilithenge esitolo”. Usethi,“Angisenanduku, angisensamdoko, angisenasondo,

Dow

nloa

ded

by [

Uni

vers

ity o

f St

elle

nbos

ch]

at 1

1:10

16

Febr

uary

201

5

Southern African Linguistics and Applied Language Studies 2005, 23: 91–109 103

predicates, e.g. akasenalutho, akasenaswidi.In contrast, Sanelisiwe in Grade 1 was unableto recall the story at all, despite the fact that hewas given the book to page through.

Besides the fact that Nkululeko’s pre-testrecall is shorter and far less detailed than hispost-test recall, his pre-test discourse is also farmore context-embedded. For example, hemakes no distinction between the referents foru- in uboleka (Bolekile) and uyala (the othercharacters in the story). Given the fact that heis looking at the pictures while he retells thestory, Nkululeko assumes that the listenershares his context and therefore knows towhom he is referring. In fact, he starts his storywith the Class 1 subject prefix for ‘he/she’without first stating who the person referent is inUyaboleka.

In the post-test the discourse is moredisembedded from its context. AlthoughNkululeko is still paging through the book as henarrates the story, he identifies the personreferent explicitly at the beginning of the storyUthi uBolekile. Furthermore, Nkululekoattempts to create a dialogue between Bolekileand the people with whom he interacts in thestory. Later, when the grandfather is introducedinto the story, he is explicitly identified, Uthiumkhulu ... There is no digression in the storyand far more information is given than in thepre-test. Grammatically, there are morecomplex constructions, e.g. the imperative withsubject prefix: ngiboleke, ngiphe, ‘lend me, giveme’, whereas in the pre-test only the indicativemood present tense is used. The emphatic

exclamation Ngeke!, ‘No way!’, is used with thepast indicative, e.g. ...Ngeke, siwutholekumama, ‘No way, we got it from Mother’. Arelative construction, elibalulekile or ‘important’,is used to qualify the noun igama, ‘word’, e.g. ...ukhohlwe igama elibalulekile elithi...

Noxolo’s post-test narration also showstraces of some changes. Whereas in the pre-test she started her recall by saying that shehad forgotten the name of the story character,in the post-test she explicitly identifies theperson (albeit a wrong name, i.e. Nobuhleinstead of Bolekile). She first frames the storyas if to contextualise it (UNobuhle wayebolekaizinto. Wathi umkhulu iyodwa into ekadekufanele ayisho).

Similarly, Thobeka’s post-test recall alsoshows features of a more disembeddeddiscourse. She too frames her story recall withan introductory thematic sentence that helps tocontextualise the story: Lona uBolekile.Wayethanda ukutsheleka, ‘This is Bolekile. Heliked to borrow things’. The ensuing dialoguebetween characters in the story also explicitlyidentifies who is speaking (bathi abafana ... athiamantombazana ... Athi ubaba ... (‘the boyssaid ... the girls said ... Father says’). In theprocess of narrating the story, Thobeka waspaging through the book and attempted tocreate a ‘dialogue’ with Thabo (thedoll/listener). This is a typical of the way inwhich adults mediate a story to children duringstorybook reading. This is indicated by the useof phrases such as Lo uBolekile, ‘This isBolekile’ and Nakhu la usebambelele esilevini,

usekhomba umama. UBolekile useyisono.

[Here is Bolekile sitting here. Now Bolekile says“Please would you (pl) lend me a stick”. They say“No, we also got it in the forest”. Bolekile said “I’d stillplease like you (sg) to lend me a wheel”. Then hesaid “No, I also found it’. Then Bolekile said “Pleasegive me a sweet”. He said “No, I also bought it at theshop”. And then Bolekile is worried. And thenGrandfather is talking now. And then Grandfatherand Mother. Grandfather is accusing Mother now.Bolekile is worried now].

angisenaswidi”. Uthi ubaba “Ukuthi ukhohlwe yigamaelilodwa, Bolekile”. Nakhu la usebambelele esilevini— wamtshela-ke ukuthi ukhohlwe yini: “Ukhohlweukuthi ‘Ngisacela ungiphe’”.[This is Bolekile. He liked to borrow. He borrowed theboys’ sticks. The boys said “No, we found them in theforest”. He borrows the girls’ porridge. The girls said“No, Mother gave it to us”. He borrowed the boys’wheel. They said “No, Father gave it to us”. Heborrowed a sweet from Mother. “No, I bought it at theshop”. Now he says “I no longer have a stick, I nolonger have any porridge, I no longer have anywheel, I no longer have a sweet”. Father says “It’sjust that you have forgotten one word, Bolekile”. Herehe is now, supporting his chin — then he told himwhat he had forgotten: “You forgot to say ‘Pleasewould you give me’”.]

Dow

nloa

ded

by [

Uni

vers

ity o

f St

elle

nbos

ch]

at 1

1:10

16

Febr

uary

201

5

Ntuli and Pretorius104

‘Here he is now, supporting his chin’. LikeNkululeko, Thobeka rounds off her narration atthe end with reference to the implied moral ofthe story — asking for things in a polite manner.

Although the Grade 1s were a year older,their story-retelling skills were not as good asthe Grade R children. Some samples of theGrade 1 recalls of the story are shown in Table5. Two of the Grade 1 learners did not sayanything, suggesting unfamiliarity withstorybook reading and retelling stories. One ofthe children did not want to engage in the taskand said that he had forgotten the story(Angisayazi, sengiyikhohliwe, ‘I don’t know itanymore, I have forgotten it’). Because it iseasier to remember things that are understood,the tendency to forget a story is often anindication that the child did not understand itvery well in the first place.

Although two of the learners made anattempt to say something, their rendition of thestory was not totally accurate and they did notalways identify the relevant characters. For

example, Simo’s narration starts somewhatunexpectedly (‘Thabo is now exclaiming’),without orienting the listener by creating acontext. Some details about the story areinaccurate, for example the sequence of eventsis wrong; at a more mundane level, he refers tothe main character as Thabo, not Bolekile, andin the story Bolekile does not speak to hisfather but to his grandfather. Furthermore, inSimo’s story Thabo keeps asking for items fromvarious characters but Simo does not indicatewhat their responses were.

Although Nomfundo’s recall is creative inthat she reconstructs the story differently, herrecall is in effect inaccurate and lackscoherence. In the story it is Bolekile whounsuccessfully keeps asking for different itemsfrom various people. Instead, Nomfundomentions about five different namelessindividuals who are borrowing things. In herrecall, while paging through the book, sheseems to treat each page separately, anddescribes the boy depicted in the illustrations

Table 5: Grade 1 recall of fictional story

Grade 1: Recall of storySanelisiwePages through the book but does not say anything.

SisekeloAngisayazi, sengiyikhohliwe. [I don’t know it anymore, I have forgotten it.]

SimoUThabo useyababaza, UThabo usekhuluma nobaba wakhe. UThabo usethi “Mama, awungiphe uswidi”. UThabousethi “Mntwana, awungiboleke isondo”. UThabo uthi “Ngane, awungiphe umdoko”. UThabo uthi “Awungibolekeinduku”. UThabo usekhathazekile.[Thabo is now exclaiming. Thabo is now talking to his father. Now Thabo says “Mom, please give me a sweet”.Now Thabo says “Child, please lend me a wheel”. Thabo says “Child, please give me porridge”. Thabo says“Please lend me a stick”. Thabo is worried now (present).]

NomfundoLona ugqoke izimpahla zabantu, uzitshelekile. Nangu futhi omunye otsheleka izimpahla zabantu. Nanguowatsheleka izimpahla zabantu, ehlala ngazo ekhaya. Nangu futhi omunye, uselambile. Usethi akamtshelekeleziya zinto. Wathi “Ngizitsheleke kumama”. Nangu omunye useyabekezela — akanaswidi, akanalutho. Usecelaumdoko. Wathi “Hhayi, ngiwuthathe kumama”. Nangu umkhulu. Usekhamisela ingane yakhe. Akayiboni nalootsheleka izimpahla zabantu.[This one is wearing people’s clothes, he borrowed them. Here is another one who borrows people’s clothes.Here is someone who borrowed people’s clothes/stuff and (at the same time) keeps them at home. And here issomeone else, he is hungry now. So he says he should lend him those things over there. He says “I borrowedthem from Mother”. Here is another one, he is patient/long-suffering — he doesn’t have a sweet, he has nothing.Now he asks for porridge. He says “No, I got it from Mother”. Here is Grandfather. Now he looks at his child. Hedoesn’t see him with this one/person who borrows people’s clothes (present).]

Dow

nloa

ded

by [

Uni

vers

ity o

f St

elle

nbos

ch]

at 1

1:10

16

Febr

uary

201

5

Southern African Linguistics and Applied Language Studies 2005, 23: 91–109 105

on each page as a separate individual ratherthan as the protagonist, Bolekile. She thus failsto draw out the narrative thread in thesequence of pictures. Her narrative account iscontext-embedded in that she uses deictic orindexical items whose meanings need to berecovered from the interactional context, e.g.use of the pronouns lona, ‘this one’, leziyazinto, ‘those things’. Characters in the story arenot clearly identified; instead she refers to themas nangu omunye/nangu futhi omunye, ‘here isanother one/here is yet another’. Most of thesentences start with the demonstrative copulanangu ‘here is’, which assumes sharedparticipation in the discourse.

In sum, the Grade Rs outperformed theGrade 1 learners in all the language andliteracy measures, even though they were ayear younger than their school-going peers.

DiscussionThe research question that informed this studywas: Did storybook reading have an effect onthe language, literacy and discoursedevelopment of the children? On the whole, theGrade R children not only showedimprovements in all the language and literacymeasures after eight months of pre-schoolattendance in which they were regularlyexposed to storybook activities, but they alsoconsistently outperformed their older peers inthese measures. In the post-test protocols,some of the Grade R children were alsoshowing early features of a more context-disembedded style of discourse than theirGrade 1 peers. The Grade 1 learners were ayear older than the pre-school children and hadalready been at school for two months whenthey were assessed, but they had not been in astorybook programme before they startedschool. Their recall protocols were shorter andless accurate than those of the Grade Rs, andtheir language use was not as rich and varied.

In other words, the Grade R children wouldsubsequently start school with strongeremergent literacy skills than their Grade 1counterparts had done the year before. Theyhad stronger notions of story schemas and ofwhat literacy entailed, and they were morefamiliar with book-based activities. Theseemergent literacy skills have been shown toprovide a sound foundation on which laterliteracy skills are built. As Mace-Matluck et al.

(1989: 205) point out: ‘Children who are wellprepared at entry to take advantage of whatschool has to offer make progress ... Childrenless prepared often get off to a slow start (andthey) lag behind their more advantaged peersas they progress in school’.

The power of storybook reading derivesfrom several factors. Storybook readingprovides opportunities for exposure tovocabulary and linguistic structures notfrequently encountered in other types ofinteraction (see Snow & Dickinson, 1991). TheGrade R children certainly showed richer andmore varied language use than the Grade 1s.Storybook reading also provides opportunitiesfor adults and children to engage in extendeddiscourse on a topic. This effect was stronglyevident in the data — by the end of the year theGrade R children were producing samples ofquite extended discourse; this was notmarkedly evident in the Grade 1 protocols.

As indicated at the beginning of the article,oral language ability facilitates entry into theworld of reading and writing. However, orallanguage per se does not predict schoolsuccess in the long run. Some children mayshow good conversational skills when they goto school but research suggests that some orallanguage skills ‘... are relatively irrelevant inexplaining individual differences in literacyaccomplishments, whereas others are crucialprecursors to aspects of literacy achievement’(Snow & Dickinson, 1991: 186). One suchprecursor is decontextualised or disembeddedoral language ability. This is language usewhich does not rely strongly on an interactiveconversational partner and does not assumeshared knowledge with audience; instead,context becomes lexicalised and the locus ofmeaning resides more strongly in the languageitself rather than in the interactional context.The kinds of tasks that help to develop this kindof language ability are activities that requireextended discourse such as storybook reading,telling stories, giving descriptions of things orevents, explaining ideas and planning futureevents. Activities that require short responsesto known questions (for example, chanting theletters of the alphabet or numbers in choralunison) do not develop extended discourse, yetit is often such activities that figure prominentlyin pre-schools. Some of the post-test storyrecalls of the Grade Rs show features of

Dow

nloa

ded

by [

Uni

vers

ity o

f St

elle

nbos

ch]

at 1

1:10

16

Febr

uary

201

5

Ntuli and Pretorius106

disembedded language where the context islexicalised and referents explicitly identified.The Grade 1s who had not been exposed toopportunities for extended discourse viastorybook reading did not produce languagewith these features.

The Vygotskian theory of child development(e.g. Vygotsky, 1986) emphasises the centralityof language in learning. If children do notdevelop appropriate language competencies,especially academic language competencies,then their ability to acquire new concepts and toengage in higher order and more abstract waysof thinking is jeopardised. The development ofacademic language competence can already

start in the pre-school years with exposure tostorybook reading. The development andmaintenance of language competence in thehome language is particularly important, for itcan facilitate the transfer of skills to otherlanguages.

ConclusionAll the children in this study come fromdisadvantaged communities and they attendschools that are poorly-resourced. Even thoughpoverty and low literacy levels tend to go hand-in-hand, children from high-poverty schools donot inevitably have to be doomed to a fate oflow literacy accomplishment. To illustrate this

Figure 1: Exterior and interior views of the Grade R pre-school. Achieving a lot with a little

Dow

nloa

ded

by [

Uni

vers

ity o

f St

elle

nbos

ch]

at 1

1:10

16

Febr

uary

201

5

Southern African Linguistics and Applied Language Studies 2005, 23: 91–109 107

point, Figure 1 shows the modest crèche whichthe Grade R children in this study attended.Despite its unprepossessing exterior, itprovides a print-rich environment inside. Thecaregiver at this small pre-school exposes thechildren to book-based activities and several ofthe mothers of children who attend the crècheread little storybooks to their children. Contraryto popular opinion, it is not so much abundanceof resources (of which there is not much at thisschool) as engagement of learners inmeaningful print-based activities that layssound foundations for the development ofacademic language competence.

Caregivers, educationists and appliedlinguists should not underestimate the powerfuleffects that a seemingly mild activity such asstorybook reading has on language and literacydevelopment. Introducing storybooks andshared book-based activities into pre-schoolsand primary schools does not require fiscalextravagance nor does it require extensive in-service teacher training.

Not only does storybook reading lay thelinguistic and literate foundations for thedevelopment of academic languagecompetencies, it also helps to improve the

confidence and self-esteem of learners.Through the shared experience of engagingwith storybooks in a meaningful yet fun way,they also learn to associate books with readingfor pleasure. Furthermore, early and regularexposure to a variety of storybooks in thechildren’s home languages will help to increasethe status of African languages and validatethese languages as viable vehicles of literateactivities.

Acknowledgements — The authors would like tothank Ms Snoeks Desmond, the facilitators PhumzileNgcobo and Nonzuzo Mbanjwa, and all the learners,for the opportunity to become involved in the FamilyLiteracy Project. Thanks are also extended to theorganisation that generously funded the project andultimately made this research possible. Finally,appreciation is expressed to the anonymousreviewers who provided helpful comments on thearticle.

Notes1 Ms Snoeks Desmond developed the project

and has been the co-ordinator since itsinception in 2000.

2 The first author of the article analysed theZulu data.

Allen MS, Kertoy MK, Sherblom JC & PettitJM. 1994. Children’s narrative productions:a comparison of personal event andfictional stories. Applied Psycholinguistics15: 149–176.

Buchorn-Stoll B. 2002. The influence ofstorybook reading on languagedevelopment. Language Matters 33: 25–48.

Bus A, Van Ijzendoorn M & Pellegrini A.1995. Joint reading makes for success inlearning to read: a meta-analysis ofintergenerational transmission of literacy.Review of Educational Research 65: 1–21.

Butler G. 1979. Guy Butler and Chris Mann ona new book of South African verse inEnglish. English in Africa 6(1): 1–11.

Corson D. 1997. The learning and use ofacademic English words. LanguageLearning 47(4): 671–718.

Cox BE, Shanahan T & Sulzby E. 1990. Goodand poor elementary readers’ use ofcohesion in writing. Reading Research

Quarterly 25: 47–56.Cummins J. 1991. Conversational and

academic language proficiency in bilingualcontexts. AILA Review 8: 75–89.

Cummins J. 2000. Language, Power andPedagogy: Bilingual Children in theCrossfire. Clevedon: Multilingual Matters.

Department of Social Development.Guidelines for Day care. Pretoria.

Dickinson DK & Smith MW. 1994. Long-termeffects of preschool teachers’ book-readings on low-income children’svocabulary and story comprehension.Reading Research Quarterly 29: 104–122.

Dombey H & Moustafa M. 1998. Whole to PartPhonics: How Children Learn to Read andSpell. London: Centre for Language inPrimary Education.

Elley W. 1989. Vocabulary acquisition fromlistening to stories. Reading ResearchQuarterly 24: 174–187.

Elley WB (ed). 1994. The IEA Study of Reading

References

Dow

nloa

ded

by [

Uni

vers

ity o

f St

elle

nbos

ch]

at 1

1:10

16

Febr

uary

201

5

Ntuli and Pretorius108

Literacy: Achievement and Instruction inThirty-two School Systems. New York:Pergamon.

Feitelson D, Kita B & Goldstein Z. 1986.Effects of listening to series stories on firstgraders’ comprehension and use oflanguage. Research in the Teaching ofEnglish 20: 339–357.

Garton A & Pratt C. 1989. Learning to beLiterate: the Development of Spoken andWritten Language. London: Basil Blackwell.

Graves MF, Juel C & Graves BB. 1998.Teaching Reading in the 21st Century.Boston: Allyn and Bacon.

Harris AJ & Sipay ER. 1975. How to IncreaseReading Ability: a Guide to Developmentaland Remedial Method. New York: DMackay Co.

Heath SB. 1983. Ways with Words.Cambridge: Cambridge University Press.

Hiebert EH. 1993. Young children’s literacyexperiences in home and school. In:Yussen SR & Smith MC (eds) Readingacross the Life Span. Berlin: Springer-Verlag. pp 33–55.

Holdaway D. 1979. The Foundations ofLiteracy. Sydney: Ashton Scholastic.

Jordan GE, Snow CE & Porche MV. 2000.Project EASE: the effect of a family literacyproject on kindergarten students’ earlyliteracy skills. Reading Research Quarterly35: 524–546.

Krashen S. 1993. The Power of Reading.Englewood, Colorado: Libraries Unlimited.

Labuschagne S. 2001. Family LiteracyProject: evaluation report, October.

Labuschagne S. 2002. Family LiteracyProject: evaluation report, October.

Macdonald CA. 1990. Crossing the Thresholdinto Standard Three. Pretoria: HSRC.

Mace-Matluck BJ, Hoover WA & Calfee RC.1989. Teaching reading to bilingualchildren: a longitudinal study of teachingand learning in the early grades. Journal ofthe National Association for BilingualEducation (NABE) 13(3): 187–216.

Machet MP. 2002. Addressing problems ofliteracy in disadvantaged communities.Language Matters 33: 1–24.

McCabe A & Peterson C. 1985. A naturalisticstudy of the production of causalconnectives by children. Journal of ChildLanguage 12: 145–159.

Neuman SB. 1999. Books make a difference: astudy of access to literacy. ReadingResearch Quarterly 34: 286–311.

Pretorius EJ. 2003. Differences in emergentliteracy skills between children in the familyliteracy programme and children not in theprogramme. Report, KwaZulu-Natal 2001.

Progress in International Reading LiteracyStudy (PIRLS). Available athttp://timss.bc.edu/PIRLS2001.html

Raban B. 1988. Speaking and writing: youngchildren’s use of connectives. ChildLanguage Teaching and Therapy 4: 13–25.

Reeder K & Shapiro J. 1993. Relationshipsbetween early literate experience andknowledge and children’s linguisticpragmatic strategies. Journal of Pragmatics19: 1–22.

Romaine S. 1984. The Language of Childrenand Adolescents. London: Basil Blackwell.

Saville-Troike M. 1984. What really matters insecond language learning for academicpurposes? TESOL Quarterly 18: 199–219.

Smyth A. 2002. Testing the Foundations: anExploration of Cognitive Academic LanguageDevelopment in an African Home-languageCourse. PhD thesis, University of theWitwatersrand, Johannesburg.

Snow C, Barnes WS, Chandler J, GoodmanIF & Hemphill L. 1991. UnfulfilledExpectations: Home and School Influenceson Literacy. Cambridge, MA: HarvardUniversity Press.

Snow CE & Dickinson DK. 1991. Skills thataren’t basic in a new conception of literacy.In: Jennings EM & Purves AC (eds) LiterateSystems and Individual Lives. New York:State University of New York Press. pp179–218.

Sulzby E & Teale W. 1991. Emergent literacy.In: Barr R, Kamil ML, Mosenthal P &Pearson PD (eds) Handbook of ReadingResearch, Vol. II. New York: Longman. pp727–758.

Torrance N & Olson DR. 1985. Oral andliterate competencies in the early schoolyears. In: Olson DR, Torrance N & HildyardA (eds) Literacy, Language and Learning.Cambridge: Cambridge University Press.pp 256–284.

Vivas E. 1996. Effects of story reading onlanguage. Language Learning 46: 189–216.

Vygotsky LS. 1986. Thought and Language.

Dow

nloa

ded

by [

Uni

vers

ity o

f St

elle

nbos

ch]

at 1

1:10

16

Febr

uary

201

5

Southern African Linguistics and Applied Language Studies 2005, 23: 91–109 109

Cambridge, Mass.: The MIT Press.Wells G. 1985. Preschool literacy-related

activities and success in school. In: OlsonD, Torrance N & Hildyard A (eds) Literacy,Language and Learning. Cambridge:Cambridge University Press. pp 229–255.

Wells G. 1986. The Meaning Makers: ChildrenLearning Language and Using Language toLearn. London: Hodder & Stoughton.

Dow

nloa

ded

by [

Uni

vers

ity o

f St

elle

nbos

ch]

at 1

1:10

16

Febr

uary

201

5